age related macular-degeneration__2nd_edition

Age-RelAted MAculAR degeneRAtion

Edited by

Jennifer I. Lim, M.D.University of Illinois School of Medicine, Department of Ophthalmology

Eye and Ear Infirmary, UIC Eye Center Chicago, Illinois, USA

Age-RelAted MAculAR degeneRAtionSecond edition

Informa Healthcare USA, Inc.

52 Vanderbilt AvenueNew York, NY 10017

q 2008 by Informa Healthcare USA, Inc.

Informa Healthcare is an Informa business

No claim to original U.S. Government worksPrinted in the United States of America on acid-free paper

10 9 8 7 6 5 4 3 2 1

International Standard Book Number-10: 0-8493-7214-3 (Hardcover)International Standard Book Number-13: 978-0-8493-7214-8 (Hardcover)

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with

permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publishreliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or

for the consequence of their use.

No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or othermeans, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage

or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400.

CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have beengranted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only foridentification and explanation without intent to infringe.

Visit the Informa Web site atwww.informa.com

and the Informa Healthcare Web site atwww.informahealthcare.com

Library of Congress Cataloging-in-Publication Data

Age-related macular degeneration / edited by Jennifer I. Lim.– 2nd ed.p. ; cm.

Includes bibliographical references and index.ISBN-13: 978-0-8493-7214-8 (hardcover : alk. paper)ISBN-10: 0-8493-7214-3 (hardcover : alk. paper)1. Retinal degeneration–Age factors.I. Lim, Jennifer I., 1962-[DNLM: 1. Macular Degeneration. WW 270 A26491 2007]

RE661.D3A322 2007617.7’35–dc22

2007023317

I wish to dedicate this book to my students, my family, andespecially my daughter, Bernadette.

Foreword to the Second Edition

Five years ago, Jennifer Lim, MD, and her expertcolleagues published the first edition of Age-RelatedMacular Degeneration, a state-of-the-art summary ofknowledge about this frequently blinding disease. Inthe Foreword to that edition, I noted that Dr. Lim had“fashioned this valuable compendium of the waythings are—for now!” It is relevant to ask, therefore,if the state of the art and level of knowledge haveprogressed in the interval since then. Why indeed theyhave, and, in the vernacular of the day, they haveevolved “big time!” The most important development,of course, has been the clinical proof that agents aimedat inhibiting vascular endothelial growth factor notonly preserve visual acuity in the neovascular (wet)form of age-related macular degeneration, but alsoimprove visual functioning in a substantial percentageof patients. This class of compounds is, therefore, trulyrevolutionary and of major benefit to patients with wetage-related macular degeneration.

There have been numerous additional develop-ments in both the understanding and treatment of age-related macular degeneration since the first edition,and they are very well described in this new book.Some of them are briefly mentioned subsequently inthis Foreword, but none compares with the monu-mental impact of the anti–vascular endothelial growthfactor approach to therapy of wet macular degener-ation. The discoverers of key knowledge related tovascular endothelial growth factor and its antagon-ism—Drs. Harold Dvorak, Judah Folkman, NapoleonFerrara, and others—deserve enormous credit andgratitude for their ingenuity and scientific contri-butions. These achievements have directly lead, overa decades-long evolutionary process of basic andapplied research, to the initial, successful pharma-cologic treatment techniques of today.

In the history of therapy for other blinding retinaldiseases, have there been other clinical advancesthat compare favorably with the tremendous impactof the anti–vascular endothelial growth factorapproach to age-related macular degeneration? Inmy personal experience, photocoagulation, andparticularly laser photocoagulation, had analogoustherapeutic effects when fully developed in the 1960sand 1970s. With the advent of photocoagulation,another irreversibly and commonly blinding disorder,diabetic retinopathy, suddenly and dramatically came

under control, as proved by appropriate clinical trials.The invention and development of automated parsplana vitrectomy in the 1970s and 1980s representedanother quantum leap forward. This innovativesurgery has restored vision to innumerable patientswho were blind from a variety of retinal diseases,including proliferative diabetic retinopathy. Suchexciting therapeutic advances come infrequently inmodern medicine—often they are decades apart.At their outset, they require brilliant insights followedby painstaking, time-consuming, and expensiveclinical trials for adequate proof of both efficacy andsafety. When successful, such activity is clearly worth-while, as measured by enormous improvement in bothpersonal and public health.

Have the beneficial results of age-relatedmacular degeneration therapy yet reached theirasymptote? Far from it! Although the frustratingstate of therapeutic affairs at the time of the firstedition has been substantially ameliorated in theinterim, there is much about age-related maculardegeneration that is yet to be understood and yet tobe accomplished. Even the relatively “established”worlds of photocoagulation and vitrectomy arecharacterized by useful, ongoing refinements. Analo-gous events will undoubtedly characterize theevolution of the anti–vascular endothelial growthfactor approach. We will certainly see a “dramaticflourishing of new hypotheses, experiments, andclinical procedures,” as predicted in the Foreword tothe first edition. New delivery systems and schedules,for example, will undoubtedly enhance the inconve-nient and invasive intravitreal treatment regimens thatare now utilized. Moreover, new combinations oftherapeutic agents (both pharmacologic and physical)will be proposed and evaluated. Many will inevitablyflounder (as have some other forms of initially promis-ing treatments for age-related macular degeneration,such as irradiation, submacular surgery, and photo-coagulation of drusen), but some will indeed succeed.

We have just embarked on a new era in thetreatment of wet age-related macular degeneration.But what of dry age-related macular degeneration(and its major variants, including geographicatrophy), for which prophylaxis, such as that withantioxidants, is in its formative phases and forwhich restoration of lost vision does not yet exist?

And what of wet age-related macular degenerationand an understanding of its etiology, pathogenesis,and prophylaxis? We remain in the infancy of thesesubjects. Fortunately, numerous chapters in this secondedition provide up-to-the-minute summaries ofrelevant knowledge. For example, allele associations,such as those related to complement factor H andothers, and single gene mutations, such as those invol-ving ABCR, ELOVl4, VMD2, TIMP3, peripherin/RDS,and Fibulin3/EFEMP1, represent discoveries that arelargely new since the appearance of the first edition.They are well described herein, and clinicians mustknowabout them.Cliniciansmust, of course, also knowabout advances in fundus imaging. Much of thevaluable, pragmatic information on optical coherencetomography, for example, has appeared since thefirst edition and is well described in this book.Additional new information on avant-garde subjects,such as artificial vision, retinal prostheses, retinalpigment epithelial transplantation, and new surgicaltechniques, are also reported by Dr. Lim and hercollaborators. Thus, this second edition plays apivotal role in educating those individuals concernedwith the diagnosis and treatment of age-relatedmacular degeneration.

Finally, what of visual rehabilitation and itsnewer techniques? Clinicians must be well informedabout developments that impact patients with lowvision. The relevant chapter in this edition has prac-tical, current information.

Despite the extraordinary advances of the lastfive years, and despite the superlative therapeuticarmamentarium that we now possess, too manypatients still lose their vision to both dry and wetforms of age-related macular degeneration. Theirwell-being must remain at the forefront of ourconsciousness.

If progress over the last five years can be used asa predictor for the future, we can assume that ongoingclinical care will continually evolve—until the passageof time (and hopefully another edition) permitsidentification of those ideas that can safely bediscarded and those that herald better vision for ourpatients.

Morton F. GoldbergThe Joseph Green Professor of Ophthalmology

Former Director, the Wilmer Eye InstituteJohns Hopkins University School of Medicine

Baltimore, Maryland, U.S.A.

vi FOREWORD TO THE SECOND EDITION

Foreword from the First Edition

Age-related macular degeneration (AMD) has becomea scourge of modern, developed societies. In suchgroups, where improved living conditions andmedical care extend human longevity, degenerationof bodily tissues slowly but relentlessly occurs as thelife span increases. Sooner or later, the ‘‘warranty’’ onsuch tissues expires, and so do critically importantcells that, in the case of the macula, would haveallowed normal visual function if they had survived.Those cells occupy a tiny area having a diameter ofonly about 2 to 3 mm in human eyes. When the cellslose their function or die and disappear, sharp centralvisual acuity fails, and lifestyle is compromised—oftenseverely. The ability to read, drive, recognize faces, orwatch television can be impaired or lost. This groupof diseases—AMD—has become the leading cause ofvisual impairment in those countries where increas-ingly large numbers of individuals live to a so-called‘‘ripe old age.’’ Most of these senior citizens hadanticipated, with pleasure, the opportunity to enjoytheir mature and less frenetic years, but too many ofthese individuals, ravaged physically and emotionallywith AMD, frequently and understandably complainthat the golden years are not quite so golden. This isthe human and emotional side of AMD, a group ofdisorders now under intense scientific and clinicalscrutiny, as ably summarized herein by Dr. JenniferLim and her expert group of coauthors.

The chapters in this book are devoted to patho-physiology, clinical features, diagnostic tests, currentand experimental therapies, rehabilitation, andresearch. They represent what we know today. Theyalso tell us explicitly or by inference what we need toknow tomorrow. In effect, they are cross-sectionalanalyses of the present state of knowledge, analogousto photos in an album, for example. Here, in this book,we have comprehensive, definitive, analytic reviews ofthe current state of macular affairs. Such albums andbooks are often informative and beautiful, but they bestrealize their inherent potential, as does this book, bywhetting our appetite for more information, both fortoday as well as for tomorrow. For example, what arethe precise etiology and pathophysiology ofAMD? Will they change? What are the best diagnostictests for different forms of AMD? (Parenthetically, it ishistorically noteworthy to realize that fluoresceinangiography remains the definitive test for diagnosing

the presence of choroidal neovascularization andrelated phenomena in AMD, despite having beendeveloped almost half a century ago.) What are thebest therapies of today and how might we improvethem in the future? At present, we think primarily ofthermal laser photocoagulation and photodynamictherapy. How can they be enhanced? What roles, ifany, will other techniques play? Will they includelow-power transpupillary thermal or x-irradiation,antiangiogenic drugs, genetic manipulation, orsurgery? Will combinations of these or even newermodalities be demonstrated to be both safe and effec-tive? Will wide-scale population-based preventivemeasures, including antioxidants, for example,be more important than therapeutic interventionex post facto?

Clairvoyance is an imperfect attribute, but thelargely palliative and incompletely successful treat-ments of today are quite frustrating. There is acompelling mandate for intense and sustained effortsto improve both treatment andprophylaxis. The crystalball for AMD suggests that the immediate future willbe characterized by refinements in today’s favoredinterventions, especially photodynamic therapy, butno one can really hope or believe that the therapeuticstatus quo will be preserved. Substantial change isa certainty. Physicians and patients appropriatelydemand more. The intermediate and long-rangefuture will probably include a large number of defini-tive clinical trials devoted to fascinating newpharmacological agents, many of which are now inthe evaluative pipeline, butmany ofwhich have not yeteven been conceived. Classes of drugs will includeantiangiogenic or angiostatic steroids with glucocorti-coid and nonglucocorticoid qualities, as well as diverseagents to bind and inactivate cytokines and chemo-kines at different points in the angiogenic andvasculogenic cascades. Many will involve blockage ofthe actions of vascular endothelial growth factor(VEGF). Ingenious surgical approaches will alsocome, and some will then go, as more and more newapproaches of this nature undergo clinical evaluationand gain either widespread acceptance or rejection.

Today’s requirements for ‘‘evidence-based’’medical decisions invoke Darwinian selectionprocesses for numerous known, as well as currentlyunknown, diagnostic and therapeutic approaches to

AMD. Outstandingly good techniques, such as fluor-escein angiography, will persist—at least for theforeseeable future. Less desirable ones, such as subfo-veal thermal photocoagulation, for example, will besupplanted by something better, such as photo-dynamic therapy—at least for the moment. Theaccretion of scientific and clinical knowledge isusually an extremely slow process, but that is notnecessarily bad because new ideas and techniquesare afforded ample opportunity for dispassionateevalution. Sudden breakthroughs, on the other hand,intellectual or technical epiphanies, are infrequent.When they do occur—such as angiography, photocoa-gulation, or intravitreal surgery—they abruptly createquantum leaps characterized by dramatic flourishingof new hypotheses, experiments, and clinicalprocedures. The world of AMD would benefit fromsuch giant steps (such as a new class of drugs or a newphysical modality or type of equipment), but, becausethey are unpredictable in their origin and timing, weare presently faced with the less spectacular, butimportant, responsibilities of initiating and sustainingmore prosaic, but potentially useful research efforts.

Hopefully, in the future more emphasis will beplaced on preventive approaches. Modification ofrelevant risk factors for AMD may prove to be muchmore effective, from the perspective of the publichealth, than therapeutic attempts aimed at a diseasethat has already achieved a threshold for progres-sive degeneration and visual impairment. Thus far,epidemiological studies have largely been inconclusiveand occasionally contradictory, and we nowknow of only one clear-cut modifiable risk factor,namely, cigarette smoking (and possibly systemichypertension).

The influences of race and heredity remaintantalizing, and it will be important to understand

why some races are protected from severe visualloss in AMD and why others are not. Moreover, themajor influence of heredity is inescapable, but we nowknow only that this influence is complex, and it may beeven more complex than anticipated because of amultiplicity of unknown contributory environmentaland other genetic factors. We do know the genesresponsible for a previously enigmatic group ofjuvenile forms of inherited macular degeneration,such as the eponymously interesting diseases namedfor Best, Stargardt, Doyne, and Sorsby, but thereappears to be no universally accepted or substantiverelationship between any of these single-gene, rareMendelian traits and the far more common AMD,which has no clear-cut Mendelian transmissionpattern, but currently affects millions of agingindividuals.

The march of time related to scientific progress isceaseless, and this is certainly true of research relatedto AMD. Darwinian selection of the best new ideaswill inevitably emerge, allowing an evolutionaryapproach to enhanced understanding and improvedtreatment or prophylaxis. Should we be fortunateenough to witness a bona fide revolution or break-through in ideas related to AMD, such an advance islikely to emanate from those scientists and cliniciansmeeting Louis Pasteur’s observation that ‘‘chancefavors the prepared mind.’’ It is toward that goal—the creation of the prepared mind—that Dr. Lim hasfashioned this valuable compendium of the waythings are—for now!!

Morton F. GoldbergDirector and William Holland Wilmer

Professor of OphthalmologyThe Wilmer Eye Institute


viii FOREWORD FROM THE FIRST EDITION

Preface

Research has yielded major discoveries about theetiology, pathophysiology, and treatment of age-related macular degeneration over the last decade.Indeed, the desire to summarize and synthesize thisnew information for clinicians and scientists involvedin age-related macular degeneration patient care andresearch resulted in the first edition of this book. Sincethen five years have flown by, and the pace of basicscience and translational research in age-relatedmacular degeneration has accelerated. The resultantnovel discoveries have improved and will continue toimprove the daily lives of our patients. These noveltherapies offer not only sight saving, less destructiveforms of treatment for exudative age-related maculardegeneration, but also treatments that can improvevisual acuity. In addition, preventive treatments arebeing developed for non-exudative age-relatedmacular degeneration. The goal of this second editionis to inform the reader about the latest informationavailable on the pathophysiology, diagnosis, manage-ment, and treatment of age-related maculardegeneration.

A significant amount of new information ispresented throughout this second edition. I askedretinal experts to first summarize the establishedinformation and then to present the most noveldevelopments in their field. The first section of thisbook includes the pathophysiology, epidemiology,and genetics of age-related macular degeneration.Updated light and electron microscopic findings ofage-related macular degeneration are presented tofacilitate the understanding of its ultrastructuralpathophysiology. Such an understanding is useful indirecting future areas of research towards a cure. At thetime of the first edition, the genetics of age-relatedmacular degeneration was largely unknown, and therole of the immune systemwas mostly a theoretical one.Since then, the key role of the immune system on thepathophysiology of age-related macular degenerationhas been shown by such findings as complement factorH and HTRA1 gene associations with exudative age-related macular degeneration.

The second section focuses on the clinicaldiagnosis and treatment of age-related maculardegeneration. The clinical findings seen in the non-exudative and exudative forms are discussed.Additional color photos have been added and are

shown within each chapter (instead of the colorinsert used in the first edition). The natural historydata of untreated age-related macular degenerationis retained and contrasted with the outcomes fromtreatment trials.

The third section on imaging includes newlyadded chapters on fundus autofluorescence andquantitative imaging techniques. The imaging modal-ities are discussed with attention to their usefulness inplanning treatment and assessing treatment responsesof age-related macular degeneration patients.

The next sections present in-depth informationon current and experimental forms of treatment fornon-exudative and exudative forms of age-relatedmacular degeneration. The presentation of the treat-ment options includes a discussion of the mechanism ofaction, the clinical treatment technique, the targetedpatient population, the expected outcomes, and abalanced discussion of both positive and negativeaspects of each treatment.

The following section of this book focuses on visualrehabilitation and active areas of basic science researchthat may lead to other forms of treatment in the nearfuture. It is a reality that despite the recent progress intreatments, some patients still lose visual acuity. Forthese patients, visual rehabilitation remains extremelyimportant. An updated discussion of the available lowvision devices and the psychosocial aspects of visualloss are included to help counsel patients with age-related macular degeneration and visual loss. Theprogress in the areas of retinal prostheses and retinalpigment cell transplantation are presented. Theseareas of research may one day lead to future treatmentsthat help to overcome visual loss and damage. Progressin these areas renews our hope for the future gener-ations afflicted with age-related macular degeneration.Promising new therapies will need to undergo clinicaltrials to evaluate clinical efficacy. The last section ofthis book therefore presents the essentials of clinicaltrial design.

As in the first edition, no single manageablevolume can compile and analyze all of the existingknowledge concerning age-related macular degener-ation. I have attempted to distill the most clinicallysalient and exciting research information from thevast body of knowledge for inclusion in this secondedition. If this book can once again serve as a first-hand

resource for researchers and clinicians in the area ofage-related macular degeneration, then my goal hasbeen achieved. It is my hope that the informationpresented herein continues to incite inquiry and igniteresearch that may unearth those enigmatic answers toquestions about the etiology of and cure for age-relatedmacular degeneration.

I wish to thank my outstanding contributorswithout whom this book would not be possible. Their

eagerness to collaborate, their scholarship, and theirexpertise made my job as editor of this book extremelyenjoyable, educational, and satisfying. I wish to thankmy administrative assistants, Francine and Annel, fortheir invaluable secretarial assistance. I wish to thankmy editors at Informa Healthcare for their assistance incompiling this book.

Jennifer I. Lim

x PREFACE

Contents

Foreword to the Second Edition Morton F. Goldberg vForeword from the First Edition Morton F. Goldberg viiPreface ixContributors xiii

PART I. PATHOPHYSIOLOGY AND EPIDEMIOLOGY OF AGE-RELATED MACULARDEGENERATION

1. Histopathology of Age-Related Macular Degeneration 1Shin J. Kang and Hans E. Grossniklaus

2. Immunology of Age-Related Macular Degeneration 11Karl G. Csaky and Scott W. Cousins

3. Genetics of Age-Related Macular Degeneration 35Jennifer R. Chao, Amani A. Fawzi, and Jennifer I. Lim

4. Risk Factors for Age-Related Macular Degeneration andChoroidal Neovascularization 47Kah-Guan Au Eong, Bakthavatsalu Maheshwar, Stephen Beatty, and Julia A. Haller

5. Choroidal Neovascularization 87Frances E. Kane and Peter A. Campochiaro

PART II. CLINICAL FEATURES OF AGE-RELATED MACULAR DEGENERATION6. Non-exudative Age-related Macular Degeneration 97

Neelakshi Bhagat and Christina J. Flaxel

7. Geographic Atrophy 111Sharon D. Solomon and Janet S. Sunness

8. Exudative (Neovascular) Age-Related Macular Degeneration 125Jennifer I. Lim and Jerry W. Tsong

PART III. IMAGING TECHNIQUES FOR THE CLINICAL EVALUATION OF AGE-RELATEDMACULAR DEGENERATION

9. Indocyanine Green Angiography in Age-Related Macular Degeneration 159Scott C. N. Oliver, Antonio P. Ciardella, Daniela C. A. C. Ferrara,Jason S. Slakter, and Lawrence A. Yannuzzi

10. Optical Coherence Tomography in the Evaluation and Management of Age-RelatedMacular Degeneration 177David Eichenbaum and Elias Reichel

11. Quantitative Retinal Imaging 185Daniel D. Esmaili, Roya H. Ghafouri, Usha Chakravarthy, and Jennifer I. Lim

12. Fundus Autofluorescence in Age-Related Macular Degeneration 191Rishi P. Singh, Jeffrey Y. Chung, and Peter K. Kaiser

PART IV. MEDICAL TREATMENT FOR AGE-RELATED MACULAR DEGENERATION13. Laser Photocoagulation for Choroidal Neovascularization 203

Catherine Cukras and Stuart L. Fine

14. Photocoagulation of AMD-Associated CNV Feeder Vessels: An OptimizedApproach 207Robert W. Flower

15. Photodynamic Therapy 223ATul Jain, Darius M. Moshfeghi, and Mark S. Blumenkranz

16. Radiation Treatment in Age-Related Macular Degeneration 233Christina J. Flaxel and Paul T. Finger

17. Anti-VEGF Drugs and Clinical Trials 247Todd R. Klesert, Jennifer I. Lim, and Phillip J. Rosenfeld

18. Laser Prophylaxis for Age-Related Macular Degeneration 257Jason Hsu and Allen C. Ho

PART V. SURGICAL TREATMENT FOR AGE-RELATED MACULAR DEGENERATION19. Macular Translocation 273

Kah-Guan Au Eong, Dante J. Pieramici, Gildo Y. Fujii, Bakthavatsalu Maheshwar,and Eugene de Juan, Jr.

20. Age-Related Macular Degeneration: Use of Adjuncts in Surgery and NovelSurgical Approaches 295Richard Scartozzi and Lawrence P. Chong

PART VI. VISUAL REHABILITATION21. Clinical Considerations for Visual Rehabilitation 303

Susan A. Primo

22. Retinal Prostheses: A Possible Treatment for End-Stage Age-Related MacularDegeneration 319Thomas M. O’Hearn, Michael Javaheri, Kah-Guan Au Eong, James D. Weiland,and Mark S. Humayun

23. Retinal Pigment Epithelial Cell Transplantation and Macular Reconstruction forAge-Related Macular Degeneration 329Lucian V. Del Priore, Henry J. Kaplan, and Tongalp H. Tezel

PART VII. CLINICAL TRIAL DESIGN24. Clinical Research Trials 349

A. Frances Walonker and Kenneth R. Diddie

Index 355

xii CONTENTS

Contributors

Kah-Guan Au Eong Department of Ophthalmology and Visual Sciences, AlexandraHospital, Department of Ophthalmology, Yong Loo Lin School of Medicine, NationalUniversity of Singapore, The Eye Institute, National Healthcare Group, Jurong MedicalCenter, Singapore Eye Research Institute, and Department of Ophthalmology, Tan TockSeng Hospital, Singapore

Stephen Beatty Department of Ophthalmology, Waterford Regional Hospitaland Department of Chemical and Life Sciences, Waterford Institute of Technology,Waterford, Ireland

Neelakshi Bhagat The Institute of Ophthalmology and Visual Science, New JerseyMedical School, Newark, New Jersey, U.S.A.

Mark S. Blumenkranz Vitreoretinal Surgery, Department of Ophthalmology, StanfordUniversity Medical Center, Stanford, California, U.S.A.

Peter A. Campochiaro Departments of Ophthalmology andNeuroscience, Johns HopkinsUniversity School of Medicine, Baltimore, Maryland, U.S.A.

Usha Chakravarthy The Queen’s University of Belfast and Royal Hospitals, Belfast,Northern Ireland

Jennifer R. Chao Doheny Eye Institute and Department of Ophthalmology, Keck Schoolof Medicine, University of Southern California, Los Angeles, California, U.S.A.

Lawrence P. Chong Doheny Retina Institute of the Doheny Eye Institute, Keck Schoolof Medicine, University of Southern California, Los Angeles, California, U.S.A.

Jeffrey Y. Chung Cole Eye Institute, Cleveland Clinic Foundation, Cleveland, Ohio,U.S.A.

Antonio P. Ciardella Department of Ophthalmology, Denver Health Hospital Authority,Denver, Colorado, U.S.A.

Scott W. Cousins Department of Ophthalmology, Duke University Medical Center,Durham, North Carolina, U.S.A.

Karl G. Csaky Department of Ophthalmology, Duke University Medical Center, Durham,North Carolina, U.S.A.

Catherine Cukras Department of Ophthalmology, Scheie Eye Institute, Universityof Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

Eugene de Juan, Jr. Beckman Vision Center, Department of Ophthalmology, Universityof California, San Francisco, California, U.S.A.

Lucian V. Del Priore Department of Ophthalmology, Columbia University, New York,New York, U.S.A.

Kenneth R. Diddie Retinal Consultants of Southern California, Westlake Village,California, U.S.A.

David Eichenbaum New England Eye Center, Tufts University School of Medicine,Boston, Massachusetts, U.S.A.

Daniel D. Esmaili Doheny Eye Institute and Department of Ophthalmology, Keck Schoolof Medicine, University of Southern California, Los Angeles, California, U.S.A.

Amani A. Fawzi Doheny Eye Institute and Department of Ophthalmology, Keck Schoolof Medicine, University of Southern California, Los Angeles, California, U.S.A.

Daniela C. A. C. Ferrara The LuEsther T. Mertz Retinal Research Department, ManhattanEye, Ear, and Throat Hospital, New York, New York, U.S.A.

Stuart L. Fine Department of Ophthalmology, Scheie Eye Institute, Universityof Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

Paul T. Finger New York University School of Medicine, The New York Eye CancerCenter, New York, New York, U.S.A.

Christina J. Flaxel Casey Eye Institute, Oregon Health & Science University, Portland,Oregon, U.S.A.

Robert W. Flower Department of Ophthalmology, University of Maryland School ofMedicine, Baltimore, Maryland and Department of Ophthalmology, New York UniversitySchool of Medicine and the Macula Foundation, Manhattan Eye, Ear, and Throat Hospital,New York, New York, U.S.A.

Gildo Y. Fujii Vitreous and Retina Department, State University of Londrina, Londrina,Parana, Brazil

Roya H. Ghafouri Department of Ophthalmology, Boston University Medical Center,Boston University School of Medicine, Boston, Massachusetts, U.S.A.

Hans E. Grossniklaus Department of Ophthalmology, Emory University School ofMedicine, Atlanta, Georgia, U.S.A.

Julia A. Haller The Wilmer Ophthalmological Institute, Johns Hopkins UniversitySchool of Medicine, Johns Hopkins Hospital, Baltimore, Maryland, U.S.A.

Allen C. Ho Retina Service, Wills Eye Hospital, Philadelphia, Pennsylvania, U.S.A.

Jason Hsu Retina Service, Wills Eye Hospital, Philadelphia, Pennsylvania, U.S.A.

Mark S. Humayun Doheny Retina Institute, Doheny Eye Institute, Departmentof Ophthalmology, Keck School of Medicine, University of Southern California,Los Angeles, California, U.S.A.

ATul Jain Department of Ophthalmology, Stanford University Medical Center, Stanford,California, U.S.A.

Michael Javaheri Doheny Eye Institute and Department of Ophthalmology, Keck Schoolof Medicine, University of Southern California, Los Angeles, California, U.S.A.

Peter K. Kaiser Digital Optical Coherence Tomography Reading Center, Cleveland,Ohio, U.S.A.

Frances E. Kane Alimera Sciences, Inc., Alpharetta, Georgia, U.S.A.

Shin J. Kang L.F. Montgomery Ophthalmic Pathology Laboratory, Emory Eye Center,Emory University School of Medicine, Atlanta, Georgia, U.S.A.

Henry J. Kaplan Department of Ophthalmology and Visual Sciences, Universityof Louisville, Louisville, Kentucky, U.S.A.

Todd R. Klesert Doheny Eye Institute, University of Southern California, Los Angeles,California, U.S.A.

Jennifer I. Lim University of Illinois School of Medicine, Department of Ophthalmology,Eye and Ear Infirmary, UIC Eye Center, Chicago, Illinois, U.S.A.

xiv CONTRIBUTORS

Bakthavatsalu Maheshwar Department of Ophthalmology and Visual Sciences,Alexandra Hospital and Jurong Medical Center, Singapore

Darius M. Moshfeghi Adult and Pediatric Vitreoretinal Surgery, Stanford UniversityMedical Center, Stanford, California, U.S.A.

Thomas M. O’Hearn Doheny Eye Institute and Department of Ophthalmology,Keck School of Medicine, University of Southern California, Los Angeles,California, U.S.A.

Scott C. N. Oliver Department of Ophthalmology, Rocky Mountain Lions Eye Institute,University of Colorado School of Medicine, Aurora, Colorado, U.S.A.

Dante J. Pieramici California Retina Research Foundation and California RetinaConsultants, Santa Barbara, California, U.S.A., and Doheny Eye Institute and Departmentof Ophthalmology, Keck School of Medicine, University of Southern California,Los Angeles, California, U.S.A.

Susan A. Primo Department of Ophthalmology, Emory University School of Medicine,Atlanta, Georgia, U.S.A.

Elias Reichel New England Eye Center, Tufts University School of Medicine, Boston,Massachusetts, U.S.A.

Phillip J. Rosenfeld Bascom Palmer Eye Institute, Miami, Florida, U.S.A.

Richard Scartozzi Doheny Retina Institute of the Doheny Eye Institute, Keck Schoolof Medicine, University of Southern California, Los Angeles, California, U.S.A.

Rishi P. Singh Cole Eye Institute, Cleveland Clinic Foundation, Cleveland, Ohio, U.S.A.

Jason S. Slakter The LuEsther T. Mertz Retinal Research Department, Manhattan Eye,Ear, and Throat Hospital, New York, New York, U.S.A.

Sharon D. Solomon Retina Division, Wilmer Eye Institute, Johns Hopkins UniversitySchool of Medicine, Baltimore, Maryland, U.S.A.

Janet S. Sunness The Richard E. Hoover Services for Low Vision and Blindness,Greater Baltimore Medical Center, Baltimore, Maryland, U.S.A.

Tongalp H. Tezel Department of Ophthalmology and Visual Sciences, Universityof Louisville, Louisville, Kentucky, U.S.A.

Jerry W. Tsong Doheny Eye Institute and Department of Ophthalmology, Keck Schoolof Medicine, University of Southern California, Los Angeles, California, U.S.A.

A. Frances Walonker Doheny Eye Institute and Department of Ophthalmology, KeckSchool of Medicine, University of Southern California, Los Angeles, California, U.S.A.

James D. Weiland Doheny Retina Institute, Doheny Eye Institute, Departmentof Ophthalmology, Keck School of Medicine, University of Southern California,Los Angeles, California, U.S.A.

Lawrence A. Yannuzzi The LuEsther T. Mertz Retinal Research Department, ManhattanEye, Ear, and Throat Hospital, New York, New York, U.S.A.

CONTRIBUTORS xv

Part I: Pathophysiology and Epidemiology ofAge-Related Macular Degeneration

1

Histopathology of Age-Related Macular DegenerationShin J. KangL.F. Montgomery Ophthalmic Pathology Laboratory, Emory Eye Center, Emory University School of

Medicine, Atlanta, Georgia, U.S.A.

Hans E. GrossniklausDepartment of Ophthalmology, Emory University School of Medicine, Atlanta, Georgia, U.S.A.

INTRODUCTION

Pathologic changes in age-related macular degener-ation (AMD) occur in the various structures in theposterior pole, such as the outer retina, the retinalpigment epithelium (RPE), Bruch’s membrane andthe choriocapillaries (1,2). Early lesions of AMD arelocated either between the RPE and its basementmembrane [e.g., basal laminar deposits (BlamD)] orbetween the basement membrane of the RPE and theremainder of Bruch’s membrane [e.g., basal lineardeposits (BlinD)] (2–5). Focal and diffuse depositionbetween the RPE and Bruch’s membrane is calleddrusen. Alterations of RPE such as hypopigmentation,depigmentation or atrophy as well as attenuation ofphotoreceptor cells are also observed. This formof macular degeneration is known as dry AMD(non-exudative AMD), whereas choroidal neovascu-larization (CNV) is the main feature of wet AMD(exudative AMD), which ultimately results in a disci-form scar in end stage AMD.

HISTOPATHOLOGY OF NON-EXUDATIVE (DRY) AMD

Changes of Bruch’s MembraneBruch’smembrane increases in thicknesswith age (6,7).The pathologic changes with AMD first appear in theinner collagenous zone, and generally extend into thecentral elastic zone and outer collagenous zone, andthe intercapillary connective tissue during later stagesof the disease (8). Drusen and BlinD contribute to adiffuse thickening of the inner aspect of Bruch’s mem-brane (1,6,9–14). With change of pH of the collagenousfibers and the deposition of calcium salts in the elastictissue, Bruch’s membrane shows increased basophilia.Accumulation of lipid substance from the RPE alsoresults in Sudanophilia (Fig. 1A) (12,14–16). Thickeningand hyalinization of Bruch’s membrane in the macular

area has also been found in the outer collagenous zone(5,17), presumably is due to the accumulation ofcellular waste products (12,18).

Ultrastructural examination of Bruch’s mem-brane in elderly humans typically shows focal areasof wide-spaced banded collagen, membrane-boundedbodies, tube-like structures of degenerated collagenfibers, electron dense granular material surrounded bya double membrane, and electron lucent droplets(3,6,14). These findings may be accompanied by anincrease in native collagen within the central elasticlayer (type IV collagen), the inner and outer collage-nous zone (type I and III collagen), and in theintercapillary connective tissue (6,19). Focal thinningand disruption of Bruch’s membrane is also foundassociated with an increased cellular activity (e.g.,macrophage-derived hematopoietic cells, leukocytes)on both sides of the membrane (Fig. 1A). The closerelationship between inflammatory cell componentand breaks in Bruch’s membrane suggests that thesecells might be involved in the focal destruction ofBruch’s membrane (Fig. 1B) (5,18).

Spraul and coworker showed that the degreeof calcification as well as the number of fragmenta-tions in Bruch’s membrane correlated with thepresence of non-exudative and exudative AMD(Fig. 1C) (10). Eyes with exudative AMDdemonstrateda higher degree of calcification and fragmentation ofBruch’smembrane in themacular area compared to theextramacular regions than eyes with non-exudativeAMD. A correlation was also found between thedegree of calcification, ranging from focal patches tolong continuous areas, and the number of breaks inBruch’s membrane (10).

Changes of Retinal Pigment EpitheliumRPE cells with AMD have cytoplasmic “lipofuscin”granules, as the result of incompletely digested

photoreceptor outer segments. Accumulation of lipo-fuscin granules increases in the cytoplasm of RPE.Eyes with early AMD show a decreased number anddensity of RPE cells in the macula, resulting in RPEmottling (20). These changes include pleomorphism,enlargement, depigmentation, hypertrophy, hyper-plasia, and atrophy of the RPE cells (1,9).

Another clinical finding called non-geographicRPE atrophy is related to moderate RPE hypopigmen-tation and atrophy in areas overlying diffuse BlamDand BlinD (Fig. 2) (9). Hypopigmentation, attenuationor atrophy of the RPEmay also be accompanied by softdrusen, RPE detachment and geographic atrophy(2,9,21,22). Lipoidal degeneration of individual RPEcells which are characterized by foamy cytoplasmmaybe found in eyes with nodular drusen.

Changes of ChoriocapillarisThe choriocapillaris in eyes with AMD is usuallythinned and sclerosed with a thickening of the inter-capillary septae (23). Capillaries between hyalinizedpillars of Bruch’s membrane are occasionally morewidely spaced than in age-matched control eyes (14).The choroidal arteries are usually shrunken and showreplacement of the muscular media by fibrillar fibrous

Figure 2 Histopathology of retinal pigment epithelium (RPE)and Bruch’s membrane in an eye with non-exudative age-related

macular degeneration. The RPE cell monolayer (arrows) isdiminished and exhibits hypopigmentation associated with

areas of scattered prominent pigment granules. A thick layer ofbasal laminar deposit (asterisks) is located between the plasma

membrane and basal lamina of the RPE. The remaining Bruch’smembrane is also thickened (arrowheads).

(A) (B)

(C)

Figure 1 (A) Macrophage derived inflammatory cells (arrows) are

present at the outer aspect of Bruch’s membrane (arrowheads). (B)Transmission electron microscopy shows macrophages and multi-

nucleated giant cells (asterisks) digesting basal laminar depositsoverlying Bruch’s membrane. (C) Focal disruption of Bruch’s

membrane (arrows) with ingrowth of new vessels (asterisks) inthe space between the inner aspect (arrowheads) and the

remainder of Bruch’s membrane in an eye with choroidalneovascularization.

2 KANG AND GROSSNIKLAUS

tissue with retention of wide vascular lumens.Occasionally, remains of occluded vessels withcollapsed fibrous walls may be present (2). However,it is unclear if these observed changes in the chorioca-pillaris in AMD are secondary to changes in theoverlying RPE, or are primary changes directly fromthe disease (24,25).

Changes of Neurosensory RetinaAging changes of the neurosensory retina occur inMuller cells and axons of ganglion cells includinghypertrophy, lipid accumulation or decrease andreplacement by connective tissue (26). While rodsgradually disappear with aging even withoutevidence of overt RPE disease, cones only begin todegenerate by advanced stages of non-exudativeAMD (27,28). Red–green cones seem to be moreresistant than blue cones to aging and may alsoincrease in size in AMD (4,28,29). The greatest photo-receptor cell loss is located in the parafovea (1.58–108)and may finally result in disappearance of all photo-receptors in the presence of geographic atrophy ordisciform degeneration (4,28).

Basal DepositsAccumulation of waste material between the RPE andBruch’s membrane (Fig. 3A) is termed “basal deposit”,one of the earliest pathologic features of AMD. Greenand Enger have defined two distinct types of basaldeposit(s); BlamD and BlinD (1–3,9,12,30).

Basal Laminar DepositBlamD is composed of granular material with muchwide-spaced collagen located between the plasma andbasement membranes of the RPE. BlamD stains lightblue with Masson’s trichrome (Fig. 3A) and magentawith periodic acid-Schiff staining (Fig. 3A, inset).Electron microscopic examination shows that BlamDis composed of long-spacing collagen with a periodi-city of 120 nm, membrane-bounded vacuoles andminor deposits of granular electron-dense material(Fig. 3B) (1). Studies have shown that BlamD iscomposed of collagen (type IV), laminin, glyco-proteins, glycosaminoglycans (chondroitin-, heparin-sulfate), carbohydrates (N-acetylgalactosamine),cholesterol (unesterified, esterified), and apolipopro-teins B and E (31–33).

Basal Linear DepositBlinD is located external to the RPE basementmembrane (e.g., in the inner collagenous zone ofBruch’smembrane; Fig. 3A, inset). Electronmicroscopyshows that BlinD is primarily composed of an electrondense, lipid-rich material with coated and non-coatedvesicles and granules that result in diffuse thickening ofthe inner aspect of Bruch’smembrane (Fig. 3B, inset topleft). BlinD may represent an extension or progressionof BlamD and is found in association with soft drusenand small detachments of the RPE. BlinD appears to bea more specific marker than BlamD for AMD, particu-larly for progression to late stage disease, whereas theamount of BlamD seems to be a more reliable indicator

(A)

BlamD

Brunch's Membrane

pm

Choriocapillaris

BlamD

RPE

(B)

bm

BlamD

BlindDBrunch's Membrane

Bm

RPE

Figure 3 (A) A prominent layer of basal laminar deposit (BlamD; asterisks) and basal linear deposits(BlinD; arrowheads) is located between the retinal pigment epithelium and Bruch’s membrane. Artifactual

spaces are present between the inner collagenous zone and the remaining layers of Bruch’s membrane(arrows). (B) BlamD are composed of wide-spaced collagen (insets, arrows), electron dense material and

membrane-bounded vacuoles. They are located between the plasmamembrane and the basal lamina of theRPE. Ultrastructure of the BlinD shows abundant coated vesicles and electron dense granules. Abbrevi-

ations: BlamD, basal laminar deposits; BlinD, basal linear deposits; bm, basal lamina; pm, plasmamembrane; RPE, retinal pigment epithelium.

1: HISTOPATHOLOGY OF AGE-RELATED MACULAR DEGENERATION 3

of the degree of RPE atrophy and photoreceptordegeneration (2,5,13).

DrusenDrusen are important features of AMD, which can beophthalmoscopically observed as small yellowishwhite lesions located deep to the retina in theposterior pole.

Nodular (Hard) DrusenNodular (hard) drusen are smooth surfaced, dome-shaped structures between the RPE and Bruch’smembrane (Fig. 4). They consist of hyaline materialand stain positively with periodic acid-Schiff (1).Nodular drusen often contain multiple globular calci-fications, mucopolysaccarides, and lipids (34). Thelatter supports the possibility of lipoidal degenerationof individual RPE cells (21,35,36). Ultrastructurally,nodular drusen are composed of finely granular oramorphous material, which is the same electrondensity as the basement membrane of the RPE. Vari-able numbers of pale and bristle-coated vesicles,tubular structures, curly membranes and occasionallyabnormal collagen may also be found within thesedrusen (21,31,37). The RPE overlying the drusen isoften attenuated and hypopigmented, while the cellslocated at the lateral border demonstrate a hyperpig-mented and hypertrophic appearance (38).

Drusen are primarily located in the inner collage-nous zone of Bruch’s membrane, but may extend tothe outer collagenous zone and to the intercapillarypillars if discontinuities of the central elastic layeroccur (14,39).

Immunohistochemical studies have shown thatdrusen are composed of acute phase proteins (e.g.,vitronectin, a1-antichymotrypsin, C-reactive protein,

amyloid P component, and fibrinogen), complementcomponents (e.g., C3C5 and C5b-9 complex), comp-lement inhibitors (e.g., clusterin), apolipoproteins(B, E), tissue metalloproteinase inhibitor 3, crystalline,serum albumin, fibronectin, mucopolysaccarides (e.g.,sialomucin), lipids (e.g., cerebroside), mannose,sialic acid, N-acetylglucosamine, b-galactose andimmunoreactive factors like immunoglobulin G,immunoglobulin light chains, Factor X and othercomponents, termed drusen-associated molecules(DRAMS) (34,40–43).

Soft DrusenCleavage in BlamD and BlinD may occur with theformation of a localized detachment (soft drusen). Softdrusen may become confluent with diameters largerthan 63 mm, and are then termed “large drusen.” Softdrusen formation may result in a diffuse thickening ofthe inner aspect of Bruch’s membrane with separationof the overlying RPE basement membrane from theremaining Bruch’s membrane (Fig. 5) (9,21).

At least three types of soft drusen can be differ-entiated by light microscopic examination: (i) alocalized detachment of the RPE with BlamD in eyeswith diffuse BlamD, (ii) a localized detachment of RPEby BlinD in eyes with diffuse BlamD and BlinD, or(iii) a localized detachment due to the localizedaccumulation of BlinD in eyes with diffuse BlamDbut in absence of diffuse BlinD (9). All subtypes mayappear as large drusen with sloping edges. Thehydrophobic space between these types of softdrusen and Bruch’s membrane is a potential spacefor CNV (10). Soft drusen seem to be often empty or to

Figure 4 Photomicrograph shows a nodular druse with loss of

the overlying retinal pigment epithelium.

Figure 5 Photomicrograph shows soft drusen formation

(asterisks) consisting of lightly staining proteinaceous material

between the basement membrane of the retinal pigment epi-thelium and inner aspect of Bruch’s membrane (arrows). The

overlying retinal pigment epithelium is partially lost orhypertrophic.


contain pale staining amorphous membranous orfibrillar material (44). The overlying RPE may beattenuated, diminished or atrophic. In late stages,geographic atrophy may occur (1).

Electron microscopy shows that soft drusen arecomposed of double-layered coiled membranes withamorphous material and calcification (18). BlamDoverlying the soft drusen has been found in manyeyes with AMD (21).

Diffuse DrusenDiffuse drusen is a diffuse thickening of the inneraspect of Bruch’s membrane (21,23). This term alsoincludes basal laminar (cuticular) drusen, which arecharacterized by an internal nodularity (1,30). Electronmicroscopy shows that diffuse drusen have revealedthe presence of vesicles, electron-dense particles, andfibrils between the thickened basement membrane ofthe RPE and the inner collagenous layer of Bruch’smembrane (21,23).

Geographic AtrophyGeographic atrophy, which is characterized by theareas of well demarcated atrophy of RPE, representsthe classic clinical picture of end-stage non-exudativeAMD. Although drusen are apparently central directfactors for initiation of RPE cell loss, they may disap-pear over time, especially when geographic atrophyoccurs (2).

Histological studies have shown that the lossof RPE is usually accompanied by a gradual

degeneration of the outer layers of the neurosensoryretina (photoreceptors, outer nuclear layer, externallimiting membrane), marked atrophy and sclerosis ofthe choriocapillaris, without breaks in Bruch’smembrane (Fig. 6) (2,21,22,45). Areas of geographicatrophy also are commonly characterized by residualpigmented material and a closely related monolayer ofmacrophages, which develop between the basementmembrane of the RPE and the inner collagenous layerof Bruch’s membrane (22). Occasionally accom-panying the macrophages are other cell types likemelanocytes, fibroblasts and detached RPE cells inthe subretinal space (22). The edges adjacent to areasof geographic atrophy, also termed junctional zones,are usually hyperpigmented and characterized by thepresence of hypertrophic RPE cells and multinu-cleated giant cells which contain RPE-derivedpigment in association with secondary lysosomes(22,31).

HISTOPATHOLOGY OF EXUDATIVE (WET) AMD

Choroidal NeovascularizationThe hallmark of exudative (wet type) AMD is thedevelopment of CNV. CNV represents new bloodvessel formation typically from the choroid (20).

Such changes in Bruch’s membrane as calcifica-tion and focal breaks correlate with the presence ofexudative AMD (10). Decreased thickness and disrup-tion of the elastic lamina of Bruch’s membrane in themacula may also be a prerequisite for invasion ofCNV into the space underneath the RPE (46). Vascularchannels supplied by the choroid begin as acapillary-like structure and evolve into arterioles andvenules (1,20,23,47,48). Most of the vessels arise fromthe choroid, although a retinal vessel contribution hasbeen observed in about 6% of CNV in AMD (1). Thesechoroidal vessels traverse the defects in the Bruch’smembrane and grow into the plane between the RPEand Bruch’s membrane (sub-RPE CNV: type 1growth pattern), between the retina and RPE (sub-retinal CNV: type 2 growth pattern), or in thecombination of both patterns (combined growthpattern) (48,49). The latter appears to arise from thetype 1 growth pattern.

Subretinal Pigment Epithelium CNV (Type 1Growth Pattern)In type 1 pattern, CNV originates with multipleingrowth sites, ranging from 1 to 12, from the chor-iocapillaris (Fig. 7). After breaking through Bruch’smembrane, CNV tufts extend laterally and merge in ahorizontal fashion under the RPE. This is facilitated bya natural cleavage plane in the space between BlamD

Figure 6 Photomicrograph shows a section of an eye with

geographic atrophy of the retinal pigment epithelium (RPE).

The photoreceptor cell layer is atrophic and the RPE is largelyabsent (arrowheads). A thin fibrotic scar (asterisks) associated

with mononuclear inflammatory cells is covering the inner aspectof Bruch’s membrane. Bruch’s membrane is focally disrupted

(arrows).


and Bruch’s membrane that has accumulated lipidswith aging (Fig. 8) (1,2,44,48–53). The CNV growthrecapitulates the embryologic development of thechoriocapillaris, presumably in an attempt to providenutrients and oxygen to ischemic RPE and photo-receptors. The relationship between the CNV andBlamD is similar to that between the choriocapillariesand Bruch’s membrane.

Patients with type 1 CNV have relatively intactretina and few visual symptoms. This growth patternlikely corresponds to the “occult” type of angiographicappearance of CNV (49). Secondary changes can benoted in the surrounding retina such as serous orhemorrhagic detachment of the RPE and overlying

retina, RPE tears, and lipid exudation (20,54).In histopathologic studies of surgically excised CNV,type 1 membranes are firmly attached to the overlyingnative RPE as well as the underlying Bruch’smembrane. Therefore, it is difficult to surgicallyremove type 1 membrane without damaging thesurrounding tissue (48).

Subretinal CNV (Type 2 Growth Pattern)The type 2 (subretinal) growth pattern demonstratessingle or few ingrowth sites with a focal defect inBruch’s membrane (Fig. 9). There is a reflected layerof RPE on the outer surface of the CNVand little or noRPE on its inner surface. Since there is no support fromthe RPE, the overlying outer layers of retina becomeatrophic. Angiographically, type II CNV membranesleak under the RPE and in the outer retina. Thisgrowth pattern correlates with the “classic” angio-graphic appearance (49,55). In the study of surgicallyexcised CNV, there is a reflected layer of RPE lined onthe outer surface of type 2 CNV by a monolayer ofinverted proliferating RPE cells and the native RPE(Figs. 10 and 11) (48). The overlying photoreceptorsare atrophic.

Combined Growth Pattern CNVThere aremany theoretical variations leading to a com-bined pattern of CNV growth. A progression from thetype 1 to the type 2 growth pattern as well as temporaldevelopment of the type 2 growth prior to the type 1growthhavebeendiscussed (Fig. 10) (49). Thesegrowthpatterns correspond to angiographic “minimallyclassic” and “predominantly classic” appearances.

Figure 7 Photomicrographs of an eye with exudative age-

related macular degeneration. A choroidal neovascularmembrane (asterisk) with prominent vessels (arrowheads)

grows between Bruch’s membrane (arrows) and the overlyingretinal pigment epithelium.

Figure 8 Separation of the basal laminar deposits (BlamD,arrowheads) and the remainder of Bruch’s membrane (arrows).

The space between the BlamD and Bruch’s membrane acts as anatural cleavage plane facilitating vessel ingrowth (asterisk).

Figure 9 Choroidal neovascular membrane with a type 2

growth pattern (arrowheads) between the retinal pigment epi-thelium (RPE) and the outer segments of the photoreceptor cell

layer. A reflected layer of RPE (asterisk) and atrophy of photo-receptors is present.


Histopathology of CNVThe cellular and extracellular components of CNVinclude RPE, vascular endothelium, fibrocytes, macro-phages, photoreceptors, erythrocytes, lymphocytes,myofibroblasts, collagen, fibrin, and BlamD (48,56).These components are similar regardless of the under-lying disease including AMD, ocular histoplasmosissyndrome, myopia, idiopathic, and pattern dystrophy.The only exception is BlamD, which is seen almostexclusively in AMD. These findings suggest that CNVrepresents a nonspecific wound repair response to aspecific stimulus, similar to fibrovascular granulationtissue proliferation (48,54,56,57).

Disciform ScarDisciform scar represents the end-stage of the exuda-tive form of AMD. Disciform scars are usuallyvascularized, but predominantly composed of fibroticscar tissue (Fig. 11). The vascular supply is providedfrom the choroid (96%), retina (2.5%) or both (0.6%)(1,54). A disciform scar is generally associated with theloss of neural tissue. Photoreceptor loss increases asthe diameter and thickness of the disciform scarincreases. In a morphometric analysis, eyes with disci-form scars due to AMD showed severe reduction inthe number of outer nuclear layer cells, but goodpreservation of cells in the inner nuclear layer andganglion cell layer (58). Despite massive photoreceptorloss in exudative AMD, ganglion cell neurons areknown to survive in relatively large numbers (59).

SUMMARY POINTS

& Early lesions of AMD are located either betweenthe RPE and its basement membrane (e.g., BlamD)or between the basementmembrane of the RPE andthe remainder of Bruch’s membrane (e.g., BlinD).

& Focal and diffuse deposits between the RPE andBruch’s membrane are called drusen.

& Pathologic changes with AMD first appear in theinner collagenous zone and generally extend intothe central elastic zone and outer collagenous zone,and the intercapillary connective tissue duringlater stages of the disease.

& RPE cells with AMDhave cytoplasmic “lipofuscin”granules due to incompletely digested photo-receptor outer segments.

& Although rods gradually disappearwith age, conesbegin to degenerate only with advanced stages ofnon-exudative AMD.

& Immunohistochemical studies have shown thatdrusen are composed of acute phase proteins,complement components, complement inhibitors,apolipoproteins, tissue metalloproteinase inhi-bitor 3, crystalline, serum albumin, fibronectin,mucopolysaccarides, lipids, mannose, sialic acid,N-acetylglucosamine, b-galactose and immuno-reactive factors like IgG, immunoglobulin lightchains, Factor X, and other components, termedDRAMS.

& CNV has two patterns: subretinal associated with“classic CNV” and sub-RPE associated with“occult” CNV.

REFERENCES

1. Green WR, Enger C. Age-related macular degenerationhistopathologic studies: the 1992 Lorenz E. ZimmermanLecture. Ophthalmology 1993; 100:1519–39.

Figure 10 Photomicrographs demonstrates a combined growth

pattern of a choroidal neovascularization with a reflected layer ofretinal pigment epithelium (arrows). A new vessel (asterisk)

extends through a break in the basal laminar deposits (inset,arrowheads).

Figure 11 Late stage of age-related macular degeneration with

the formation of a disciform scar between Bruch’s membrane

(black arrows) and the photoreceptor outer segments. Prominentvessels (white arrows) and a reflected layer of the retinal pigment

epithelium (arrowheads) are present in the scar.


2. Sarks SH. Ageing and degeneration in macular region: aclinicopathological study. Br J Ophthalmol 1976; 60:324–41.

3. Loffler KU, Lee WR. Basal linear deposits in the humanmacula. Graefes Arch Clin Exp Ophthalmol 1986;224:493–501.

4. Sarks JP, Sarks SH, Killingsworth MC. Evolution ofgeographic atrophy of the retinal pigment epithelium. Eye1988; 2:552–77.

5. van der Schaft TL, de BruijnWC,Mooy CM, et al. Histologicfeatures of the early stages of age-related macular degener-ation: a statistical analysis. Ophthalmology 1992; 99:278–86.

6. Hogan M, Alvarado J. Studies on the human macula: IV.Aging changes in Bruch’s membrane. Arch Ophthalmol1967; 77:410–20.

7. Ramrattan RS, van der Schaft TL, Mooy CM, et al. Morpho-metric analysis of Bruch’s membrane, the choriocapillarisand the choroid in aging. Invest Ophthalmol Vis Sci 1994;35:2857–64.

8. HoganMJ, Alvarado J, Weddell JE. Histology of the HumanEye. Philadelphia, PA: Saunders, 1971:344.

9. Bressler NM, Silva JC, Bressler SB, et al. Clinicopathologiccorrelation of drusen and retinal pigment abnormalities inage-related macular degeneration. Retina 1994; 14:130–42.

10. Spraul CW, Grossniklaus HE. Characteristics of drusen andBruch’s membrane in post-mortem eyes with age-relatedmacular degeneration. Arch Ophthalmol 1997; 115:267–73.

11. Grindle CFJ, Marshall J. Ageing changes in Bruch’smembrane and their functional implications. TransOphthalmol Soc UK 1978; 98:172–5.

12. Feeney-Burns L, Ellersieck M. Age-related changes in theultrastructure of Bruch’s membrane. Am J Ophthalmol1985; 100:686–97.

13. Curcio CA, Millican CL. Basal linear deposit and largedrusen are specific for early age-related maculopathy.Arch Ophthalmol 1999; 117:329–39.

14. Sarks SH, Arnold JJ, Killingsworth MC, et al. Early drusenformation in the normal and aging eye and their relation toage-related maculopathy: a clinicopathological study. BrJ Ophthalmol 1999; 83:358–68.

15. Spencer WH. Macular disease; pathogenesis: lightmicroscopy (symposium). Trans Am Acad OphthalmolOtolaryngol 1965; 69:662–7.

16. Holz FG, Sheraidah G, Pauleikhoff D, et al. Analysisof lipid deposits extracted from human macular andperipheral Bruch’s membrane. Arch Ophthalmol 1994;112:402–6.

17. Killingsworth MC. Age-related components of Bruch’smembrane in the human eye. Graefes Arch Clin ExpOphthalmol 1987; 225:406–12.

18. Killingsworth MC, Sarks JP, Sarks SH. Macrophages relatedto Bruch’s membrane in age-related macular degeneration.Eye 1990; 4:613–21.

19. Das A, Frank RN, Zhang NL, et al. Ultrastructural local-ization of extracellular matrix components in the humanretinal vessels and Bruch’s membrane. Arch Ophthalmol1990; 108:421–9.

20. Green WR. Histopathology of age-related macular degener-ation. Mol Vis 1999; 5:27–36.

21. Green WR, McDonnell PH, Yeo JH. Pathologic featuresof senile macular degeneration. Ophthalmology 1985;92:615–27.

22. Penfold PL, Killingsworth MC, Sarks SH. Senile maculardegeneration. Invest Ophthalmol Vis Sci 1986; 27:364–71.

23. Green WR, Key SN. Senile macular degeneration: a histo-pathologic study. Trans Am Ophthalmol Soc 1977;75:180–254.

24. Tso MOM, Friedman E. The retinal pigment epithelium: I.Comparative histology. Arch Ophthalmol 1967; 78:641–9.

25. Delaney WV, Oates RP. Senile macular degeneration: apreliminary study. Ann Ophthalmol 1982; 14:21–4.

26. Sharma RK, Ehinger BEJ. Development and structure of theretina. In: Kaufman PL, Alm A, eds. Adler’s Physiology ofthe Eye. 10th ed. Mosby: St. Louis, 2003:319–47.

27. Curcio CA, Millican CL, Allen KA, et al. Aging of thehuman photoreceptor mosaic: evidence for selectivevulnerability of rods in the central retina. InvestOphthalmol Vis Sci 1993; 34:3278–96.

28. Curcio CA, Medeiros NE, Millican LC. Photoreceptor lossin age-related macular degeneration. Invest OphthalmolVis Sci 1996; 37:1236–49.

29. Eisner A, Klien ML, Zilis JD, et al. Visual function andthe subsequent development of exudative age-relatedmacular degeneration. Invest Ophthalmol Vis Sci 1992;33:3091–102.

30. van der Schaft TL, de Bruijn WC, Mooy CM, et al. Is basallaminar deposit unique for age-related macular degener-ation? Arch Ophthalmol 1991; 109:420–5.

31. Kliffen M, Van der Schaft TL, Mooy CM, et al. Morphologicchanges in age-relatedmaculopathy.Microsc Res Tech 1997;36:106–22.

32. van der Schaft TL, Mooy CM, de Bruijn WC, et al. Immu-nohistochemical light and electron microscopy of basallaminar deposit. Graefes Arch Clin Exp Ophthalmol 1994;232:40–6.

33. Malek G, Li C-M, Guidry C, et al. Apolipoprotein B incholesterol-containing drusen and basal deposits of humaneyes with age-related maculopathy. Am J Pathol 2003;162:413–25.

34. Farkas TG, Sylvester V, Archer D, et al. The histochemistryof drusen. Am J Ophthalmol 1971; 71:1206–15.

35. El Baba F, Green WR, Fleischmann J, et al. Clinicopatho-logic correlation of lipidization and detachment of theretinal pigment epithelium. Am J Ophthalmol 1986;101:576–83.

36. Fine BS. Lipoidal degeneration of the retinal pigmentepithelium. Am J Ophthalmol 1981; 91:469–73.

37. Hogan MJ. Role of the retinal pigment epithelium inmacular disease. Trans Am Acad Ophthalmol Otolaryngol1972; 76:64–80.

38. Burns RP, Feeney-Burns L. Clinico-morphologiccorrelations of drusen of Bruch’s membrane. Trans AmOphthalmol Soc 1980; 78:206–25.

39. Farkas TG, Sylvester V, Archer D. The ultrastructure ofdrusen. Am J Ophthalmol 1971; 71:1196–205.

40. Hageman G, Mullins R, Russel S, et al. Vibronectin is aconstituent of ocular drusen and the vitronectin gene isexpressed in human retinal pigment epithelial cells. FASEBJ 1999; 13:477–84.

41. Mullins RF, Russel SR, Anderson DH, et al. Drusen associ-ated with aging and age-related degeneration containproteins common to extracellular deposits associated withatherosclerosis, elastosis, amyloidosis, and dense depositdisease. FASEB J 2000; 14:835–46.

42. Crabb JW,Miyagi M, Gu X, et al. Drusen proteome analysis:an approach to the etiology of age-related macular degener-ation. Proc Natl Acad Sci USA 2002; 99:14682–7.

43. Anderson DH, Johnson LV, Schneider BL, et al. Age-relatedmaculopathy: a model of drusen biogenesis. InvestOphthalmol Vis Sci 1999; 40:S922.

44. Sarks SH. Drusen and their relationship to senile maculardegeneration. Aust J Ophthalmol 1980; 8:117–30.

45. Bressler NM, Bressler B, Fine SL. Age-related maculardegeneration. Surv Ophthalmol 1988; 32:375–413.


46. Chong NHV, Keonin J, Luthert PJ, et al. Decreased thicknessand integrity of the macular elastic layer of Bruch’smembrane correspond to the distribution of lesions associ-ated with age-related macular degeneration. Am J Pathol2005; 16:241–51.

47. Schneider S, Greven CM, Green WR. Photocoagulation ofwell-defined choroidal neovascularization in age-relatedmacular degeneration: clinicopathologic correlation.Retina 1998; 18:242–50.

48. Grossniklaus HE, Gass JDM. Clinicopathologic correlationof surgically excised type 1 and type 2 submacular choroidalneovascular membranes. Am J Ophthalmol 1998; 126:59–69.

49. Grossniklaus HE, GreenWR. Choroidal neovascularization.Am J Ophthalmol 2004; 137:496–503.

50. Gass JDM. Biomicroscopic and histopathologic consider-ation regarding the feasibility of surgical excision ofsubfoveal neovascular memebranes. Am J Ophthalmol1994; 118:285–98.

51. Gass JDM. Stereoscopic Atlas of Macular Diseases:Diagrams and Treatment. 4th ed., Vol. 1. Mosby: St. Louis,1997:26–37.

52. Gass JDM. Pathogenesis of disciform detachment of theneuroepithelium: III. Senile disciform degeneration. AmJ Ophthalmol 1967; 63:617–44.

53. Sarks SH.Newvessel formation beneath the retinal pigmentepithelium in senile eyes. Br J Ophthalmol 1973; 57:951–65.

54. Ambati J, Ambati BK, Yoo SH, et al. Age-related maculardegeneration: etiology, pathogenesis, and therapeuticstrategies. Surv Ophthalmol 2003; 48:257–93.

55. LaFaut BA, Bartz-Schmidt KU, van den Broecke C, et al.Clinicopathologic correlation in exudative age-relatedmacular degeneration: histological differentiation betweenclassic and occult neovascularization. Br J Ophthalmol2000; 84:239–43.

56. Grossniklaus HE, Martinez JA, Brown VB, et al. Immuno-histochemical and histochemical properties of surgicallyexcised subretinal neovascular membranes in age-relatedmacular degeneration. Am J Ophthalmol 1992; 114:464–72.

57. Frank RN, Amin RH, Eliott D, et al. Basic fibroblast growthfactor and vascular endothelial growth factor are present inepiretinal and choroidal neovascular membranes. AmJ Ophthalmol 1996; 122:393–403.

58. Kim SY, Sadda S, Pearlman J, et al. Morphometric analysisof the macula in eyes with disciform age-related maculardegeneration. Retina 2002; 22:471–7.

59. Medeiros NE, Curcio CA. Preservation of ganglion celllayer neurons in age-related macular degeneration. InvestOphthalmol Vis Sci 2001; 42:795–803.


2

Immunology of Age-Related Macular DegenerationKarl G. Csaky and Scott W. CousinsDepartment of Ophthalmology, Duke University Medical Center, Durham, North Carolina, U.S.A.

INTRODUCTION

Traditionally, immune and inflammatory mechanismsof disease pathogenesis were applied only todisorders characterized by acute onset and pro-gression associated with obvious clinical signs ofinflammation. Recently, however, it has become clearthat many chronic degenerative diseases associatedwith aging demonstrate important immune andinflammatory components. Indeed, over the last yearan explosion of genetic findings have linked comp-lement dysregulation and age-related maculardegeneration (AMD).

This chapter attempts to achieve three goals.First, a brief overview is provided of the biology ofthe low-grade inflammatory mechanisms relevant tochronic degenerative diseases of aging, excluding themechanisms associated with acute severe inflam-mation. Innate immunity, antigen-specific immunity,and amplification systems are differentiated. Second,the immunology of AMD is discussed in the context ofcomplement activation and in particular relationshipsto two other age-related degenerative diseases withimmunologic features, including atherosclerosis andrenal glomerular diseases. Since these disorders shareepidemiologic, genetic, and physiological associationswith AMD, the approach attempts to delineate thescope of the subject based on analysis of other age-related degenerative diseases, and to highlight areasof potential importance to future AMD research.Finally, this chapter introduces the paradigm of“response to injury” as a model for AMD patho-genesis. This paradigm proposes that immunemechanisms, including the complement system, notonly participate in the initiation of injury, but alsosignificantly contribute to abnormal reparativeresponses resulting in disease pathogenesis andcomplications. The response to injury paradigm,emerging as a central hypothesis in the pathogenesisof atherosclerosis and glomerular diseases, provides aconnection between immunologic mechanisms ofdisease and the biology of tissue injury and repair inchronic degenerative disorders.

OVERVIEW OF BIOLOGY OF IMMUNOLOGYRELEVANT TO AMD

Innate vs. Antigen-Specific ImmunityIn general, an immune response is a sequence ofcellular and molecular events designed to rid the hostof an offending stimulus, which usually represents apathogenic organism, toxic substance, cellular debris,neoplastic cell, or other similar signal. Two broadcategories of immune responses have been recognized:innate and antigen-specific immunity (1–3).

Innate ImmunityInnate immunity (also called “natural” immunity) isa pattern recognition response by certain cells of theimmune system, typically macrophages and neutro-phils, to identify broad groups of offensive stimuli,especially infectious agents, toxins or cellular debrisfrom injury (4–6). Additionally, many stimuli of innateimmunity can directly interact with parenchymal cellsof tissues [i.e., the retinal pigmented epithelium (RPE)]to initiate a response. Innate immunity is triggered bya preprogrammed, antigen-independent cellularresponse, determined by the preexistence of receptorsfor a category of stimuli, leading to generation ofbiochemical mediators which recruit additionalinflammatory cells. These cells remove the offendingstimulus in a nonspecific manner via phagocytosis orenzymatic degradation. The key concept is that thestimuli of innate immunity interact with receptors onmonocytes, neutrophils, or parenchymal cells thathave been genetically predetermined by evolution torecognize and respond to conserved molecularpatterns or “motifs” on different triggering stimuli.These motifs often include specific amino acidsequences, certain lipoproteins, certain phospholipids,or other specific molecular patterns. Different stimulioften trigger the same stereotyped program. Thus, thereceptors of innate immunity are identical among allindividuals within a species in the same way thatreceptors for neurotransmitters or hormones aregenetically identical within a species.

The classic example of the innate immuneresponse is the immune response to acute infection.For example, in endophthalmitis, bacterial-derivedtoxins or host cell debris stimulate the recruitment ofneutrophils and monocytes, leading to the productionof inflammatory mediators and phagocytosis ofthe bacteria. Bacterial toxins can also directly activatereceptors on retinal neurons, leading to injury. Thetriggering mechanisms and subsequent effectorresponses to bacteria such as staphylococcus arenearly identical to those of other organisms, determinedby nonspecific receptors recognizing families of relatedtoxins or molecules in the environment.

Antigen-Specific ImmunityAntigen-specific immunity (also called “adaptive” or“acquired” immunity) is an acquired host response,generated in reaction to exposure to a specific “anti-genic” molecule and is not a genetically pre-determined response to a broad category of stimuli(1–3). The response is initially triggered by the “recog-nition” of a unique foreign antigenic substance asdistinguished from “self” by cells of the immunesystem (and not by nonimmune parenchymal cells).Recognition is followed by subsequent “processing”of the unique antigen by specialized cells of theimmune system. The response results in uniqueantigen-specific immunologic effector cells (T and Blymphocytes) and unique antigen-specific solubleeffector molecules (antibodies) whose aim is toremove the specific stimulating antigenic substancefrom the organism, and to ignore the presence of otherirrelevant antigenic stimuli. The key concept is that anantigen (usually) represents an alien, completelyforeign substance against which specific cells ofimmune system must generate, de novo, a specificreceptor, which, in turn, must recognize a uniquemolecular structure in the antigen for which no preex-isting gene was present. Thus, the antigen-specificimmune system has evolved away for an individual’sB and T lymphocytes to continually generate newantigen receptor genes through recombination,rearrangement, and mutation of the germline geneticstructure to create a “repertoire” of novel antigenreceptor molecules that vary tremendously in spec-trum of recognition among individuals withina species.

The classic example of acquired immunity is theimmune response to a mutated virus. Viruses (such asadenovirus found in epidemic keratoconjunctivitis)are continuously evolving or mutating new antigenicstructures. The susceptible host could not havepossibly evolved receptors for recognition to thesenew viral mutations. However, these new mutationsdo serve as “antigens” which stimulate an adaptive

antigen-specific immune response by the host tothe virus. The antigen-specific response recognizesthe virus in question and not other organisms (suchas the polio virus).

Amplification Mechanisms for BothForms of ImmunityAlthough innate or antigen-specific immunity maydirectly induce injury or inflammation, in mostcases, these effectors initiate a process that must beamplified in order to produce overt clinical mani-festations. Molecules generated within tissues whichamplify immunity are termed “mediators”, andseveral categories of molecules qualify including:(i) cytokines (growth factors, angiogenic factors,others), (ii) oxidants (free radicals, reactive nitrogen),(iii) plasma-derived enzyme systems (complement,kinins, and fibrin), (iv) vasoactive amines (histamineand serotonin), (v) lipid mediators [prostaglandins(PGs), leukotrienes, other eicosanoids, and plateletactivating factors], and (vi) neutrophil-derivedgranule products. Since principally complement, cyto-kines, and oxidants seem to be relevant to manydegenerative diseases of aging and AMD, these arediscussed below.

ComplementComponents and fragments of the complementcascade, accounting for approximately 5% of plasmaprotein concentration and over 30 different proteinmolecules, represent important endogenous ampli-fiers of innate and antigen-specific immunity as wellas mediators of injury responses (7–9). All comp-lement factors are synthesized by the liver andreleased into blood. However, some specific factorscan also be synthesized locally within tissues,including within cornea, sclera, and retina. Uponactivation, the various proteins of the complementsystem interact in a sequential cascade to producedifferent fragments and products capable of affectinga variety of functions. Three pathways have beenidentified to activate the complement cascade: clas-sical pathway, alternative pathway, and the lectinpathway (Fig. 1).

Antigen-specific immunity typically activatescomplement via the classical pathway with antigen/antibody (immune) complexes, especially thoseformed by IgM, IgG1, and IgG3 (7–9). Innate immunitytypically activates complement via the alternativepathway using certain chemical moieties on the cellwall of microorganisms [e.g., lipopolysaccharides(LPS)] or activated surfaces (e.g., implanted medicaldevices) (10). However, some innate stimuli, such asDNA, RNA, insoluble deposits of abnormal proteins(e.g., amyloid P), or apoptotic cells can also trigger the

12 CSAKY AND COUSINS

classical pathway (10–13). Recently, a new innateactivational pathway, the lectin pathway, has beenidentified (14). This pathway utilizes mannose-binding lectin (MBL) to recognize sugar moieties,such as mannose and N-acetylglucosamine, on cellsurfaces. While MBL does not normally recognizethe body’s own tissue, oxidant injury, as can occur inAMD, may alter surface protein expression and glyco-sylation causing MBL deposition and complementactivation (15–18). Recently, photooxidative productsof A2E, a bis-retinoid pigment that may accumulate inthe RPE in AMD, have been shown to activate C3 intoC3b and C3a (19). The activation of complement is alsoregulated by inhibitors, such as decay acceleratingfactor, factor H and others which serve to block,resist, or modulate the induction of various activationpathways (7–9). As will be discussed below, the role ofcomplement factor H (CFH) in particular may havecritical relevance for AMD.

Each activation pathways results in the gener-ation of the same complement byproducts whichamplify injury or inflammation by at least threemechanisms: (i) a specific fragment of the third com-ponent, C3b, can coat antigenic or pathogenic surfacesin order to enhance phagocytosis by macrophages orneutrophils; (ii) activation of terminal complementcomponents C5–C9, called the membrane attackcomplex (MAC), forms pores or leaky patches in cellmembranes leading to activation of the cell, entranceof extracellular chemicals, loss of cytoplasm or lysis ofthe cell; and (iii) generation of small pro-inflammatory

polypeptides, called anaphylatoxins (C3a, C4a, andC5a), can induce many inflammatory mediators andlead to the recruitment of inflammatory cells.

In addition, individual complement components(especially C3) can be produced locally by cells withintissue sites rather than derived from the blood (8).C3 and other complement proteins can be cleaved intobiologically activated fragments by various enzymesystems, in the absence of the entire cascade, toactivate certain specific cellular functions. Further,complement activation inhibitors can be produced bycells within tissues, including the RPE, serving as localprotective mechanism against complement-mediatedinjury (20,21). Recently, several components of thecomplement system have been identified withinBruch’s membrane and drusen indicating a potentialrole for complement in AMD (22).

CytokinesCytokine is a generic term for any soluble polypeptidemediator (i.e., protein) synthesized and released bycells for the purposes of intercellular signaling andcommunication. Cytokines can be released to signalneighboring cells at the site (paracrine action), tostimulate a receptor on its own surface (autocrineaction) or in some cases, released into the blood toact upon a distant site (hormonal action). Traditionally,investigators have used terms like “growth factors,”“angiogenic factors,” “interleukins,” “lymphokines,”“interferons,” “monokines,” “chemokines,” etc. to sub-divide cytokines into families with related activities,

ClassicalPathway

AlternativePathway

LetinPathway

(Immune Complexes (IgM, IgG),DNA, RNA, Apoptotic Cell,Memebrane Blebs, Amyloid P, etc)

(Activating Surfaces,Biomaterials, LPS)

(+)C1q C3b

C3

C3a

C3

C5EnhancedPhagocytosis

C5bC5b-9MAC

C6, C7, C8, C9

+C4

+C2

+C3b

+C4

+C2

+C3b

C3a

(+)

(Cell Surface Carbohydrates,Oxidative Stress)

Mannose Binding Lectin (MBL)

(+)

Figure 1 Schematic of the components and fragments of the complement cascade indicating three

primary sources of activation via the classical, alternative, or lectin pathway. Abbreviations: LPS,lippopolysacccharide; MBL, mannose-binding lectin; MAC, membrane attack complex.

2: IMMUNOLOGY OF AGE-RELATED MACULAR DEGENERATION 13

sources and targets. Nevertheless, research has demon-strated that although some cytokines are cell-typespecific, most cytokines have such multiplicity andredundancy of source, function and target that exces-sive focus on specific terminology is not particularlyconceptually useful for the clinician. RPE as well ascells of the immune system can produce manydifferent cytokines relevant to AMD such as monocytechemoattractant protein-1 (MCP-1) and vascular endo-thelial growth factor (VEGF).

OxidantsUnder certain conditions, oxygen-containing mole-cules can accept an electron from various substratesto become highly reactive products with the potentialto damage cellular molecules and inhibit functionalproperties in pathogens or host cells. Four of the mostimportant oxidants are singlet oxygen, superoxideanion, hydrogen peroxide and the hydroxyl radical.In addition, various nitrogen oxides, certain metal ionsand other molecules can become reactive oxidants orparticipate in oxidizing reactions.

Oxidants are continuously generated as aconsequence of normal noninflammatory cellular bio-chemical processes, including electron transportduring mitochondrial respiration, autooxidation ofcatecholamines, cellular interactions with environ-mental light or radiation, or PG metabolism withincell membranes. During immune responses, however,oxidants are typically produced by neutrophils andmacrophages by various enzyme-dependent oxidasesystems (23). Some of these enzymes are bound to theinner cell membrane (e.g., NADPH oxidase) andcatalyze the intracellular transfer of electrons fromspecific substrates (like NADPH) to oxygen orhydrogen peroxide to form highly chemically reactivecompounds meant to destroy internalized, phagocy-tosed pathogens (24). Other oxidases, likemyeloperoxidase, can be secreted extracellularly orreleased into phagocytic vesicles to catalyze oxidantreactions between hydrogen peroxide and chloride toform extremely toxic products that are highly dama-ging to bacteria, cell surfaces, and extracellular matrixmolecules (25). Finally, several important oxidantreactions involve the formation of reactive nitrogenspecies (5).

Oxidants can interact with several cellulartargets to cause injury. Among the most importantare damage to proteins (i.e., enzymes, receptors) bycross-linking of sulfhydryl groups or other chemicalmodifications, damage to the cell membrane by lipidperoxidation of fatty acids in the phospholipidbilayers, depletion of ATP by loss of integrity of theinner membrane of the mitochondria, and breaks orcross-links in DNA due to chemical alterations of

nucleotides (1,26). Not surprisingly, nature hasdeveloped many protective antioxidant systemsincluding soluble intracellular antioxidants (i.e., gluta-thione or vitamin C), cell membrane-bound lipidsoluble antioxidants (i.e., vitamin E) and extracellularantioxidants (1,26).

In the retina, oxidation induced lipid per-oxidation and protein damage in RPE andphotoreceptors have been proposed as major injurystimuli (27–30). Relevant sources of oxidants in AMDmight include both noninflammatory biochemicalsources (e.g., light interactions between photo-receptors and RPE, lysosomal metabolism in RPE,PG biosynthesis, oxidants in cigarette smoke) andinnate immunity (e.g., macrophage release ofmyeloperoxidase).

Cells of the Immune ResponseBoth innate and antigen-specific immune system useleukocytes as cellular mediators to effect and amplifythe response (i.e., immune effectors). In general, leuko-cyte subsets include lymphocytes (T cells, B cells),monocytes [macrophages, microglia, dendritic cells(DC)] and granulocytes (neutrophils, eosinophils andbasophils). A complete overview is beyond the scopeof this chapter, especially since no evidence exists thatall of these cellular effectors participate in AMD. Thus,this section will focus only upon leukocyte subsetspotentially relevant to AMD, including monocytes,basophils/mast cells and B lymphocytes/antibodies.

Monocytes and MacrophagesThe monocyte (the circulating cell) and the macro-phage (the tissue-infiltrating equivalent) are importanteffectors in all forms of immunity and inflammation(4). Monocytes are relatively large cells (12–20 mm insuspension, but up to 40 mm in tissues) and trafficthrough many normal sites. Most normal tissueshave at least two identifiable macrophage populations:tissue resident macrophages and blood-derivedmacrophages. Although many exceptions exist, ingeneral, tissue-resident macrophages represent mono-cytes that migrated into a tissue weeks or monthspreviously, or even during embryologic developmentof the tissue, thereby acquiring tissue-specific proper-ties and specific cellular markers. In many tissues,resident macrophages have been given tissue-specificnames (e.g., microglia in the brain and retina, Kupffercells in the liver, alveolar macrophages in the lung,etc.) (31–33). In contrast, blood-derived macrophagesusually represent monocytes that have recentlymigrated from the blood into a fully developedtissue site, usually within a few days, still maintainingmany generic properties of the circulating cell.


Macrophages serve three primary functions: asscavengers to clear cell debris and pathogens withouttissue damage, as antigen presenting cells (APCs) forT lymphocytes, and as inflammatory effector cells.Conceptually, macrophages exist in different levels orstages of metabolic and functional activity, eachrepresenting different “programs” of gene activationand synthesis of mediators. Three different stagesare often described: (i) scavenging or immaturemacrophages, (ii) “primed” macrophages, and (iii)“activated” macrophages. Activated macrophagesoften undergo a morphologic change in size andhistologic features into a cell called an epithelioidcell. Epithelioid cells can fuse into multinucleatedgiant cells. Only upon full activation are macro-phages most efficient at synthesis and release ofmediators to amplify inflammation and to kill patho-gens. Typical activational stimuli include bacterialtoxins (such as LPS), antibody-coated pathogens,complement-coated debris or certain cytokines(Fig. 2) (34–36).

A fourth category of macrophage, often called“reparative” or “stimulated,” is used by someauthorities to refer to macrophages with partial orintermediate level of activation (37–40). Reparativemacrophages can mediate chronic injury in theabsence of inflammatory cell infiltration or wide-spread tissue destruction. For example, reparativemacrophages contribute to physiologic processessuch as fibrosis, wound repair, extracellular matrix

turnover and angiogenesis (41–49). Reparativemacrophages play important roles in the pathogenesisof atherosclerosis, glomerulosclerosis, osteoarthritis,keloid formation, pulmonary fibrosis and other non-inflammatory disorders, indicating that the “repair”process is not always beneficial to delicate tissues withprecise structure-function requirements. In eyes withAMD, choroidal macrophages and occasionally chor-oidal epithelioid cells have been observed underlyingareas of drusen, geographic atrophy and choroidalneovascularization (CNV) (50–54). Also, cell culturedata suggest that blood monocytes from patients withAMD can become partially activated into reparativemacrophages by growth factors and debris released byoxidant-injured RPE (55).

Dendritic CellsDC are terminally differentiated bone-marrow derivedcirculating mononuclear cells distinct from the macro-phage–monocyte lineage and comprise approximately0.1% to 1% of blood mononuclear cells (56). However,in tissue sites, DC become large (15–30 mm) withcytoplasmic veils which form extensions two to threetimes the diameter of the cell, resembling the dendriticstructure of neurons. In many non-lymphoid andlymphoid organs, DC become a system of APCs.These sites recruit DC by defined migration pathways,and in each site, DC share features of structure andfunction. DC function as accessory cells which playan important role in processing and presentation ofantigens to T cells, and their distinctive role is toinitiate responses in naive lymphocytes. Thus, DCserve as the most potent leukocytes for activating Tcell dependent immune responses. However, DC donot seem to serve as phagocytic scavengers noreffectors of repair or inflammation. Both the retinaand the choroid contain high density of DC (57,58).

Basophils and Mast CellsBasophils are the blood-borne equivalent of the tissuebound mast cell. Mast cells exist in two majorsubtypes, connective tissue versus mucosal types,both of which can release preformed granules andsynthesize certain mediators de novo (59,60). Connec-tive tissue mast cells contain abundant granules withhistamine and heparin, and synthesize PGD2 uponstimulation. In contrast, mucosal mast cells require Tcell cytokine help for granule formation, and thereforenormally contain low levels of histamine. Also,mucosal mast cells synthesize mostly leukotrienesafter stimulation. Importantly, the granule type andfunctional activity can be altered by the tissue location,but the regulation of these important differences is notwell understood. Basophils and mast cells differfrom other granulocytes in several important ways.

Stimulated T Cells BacterialToxins

LPSOthers

"Activated"Macrophage

"Reparative"Macrophage

"Primed"Macrophage

1

3

2"Resting"Monocyte

IFN-γOthers

Wound RepairAngiogensis

Mild Inflammation

BelbsOthers

Various InnateStimuli

ScavengingPhagocytosis

TumorcidalBacteriocidal

Delayed Hypersensitivity

Oxidants

Eicosanoids

Cytokines

Oxidants

Eicosanoids

Cytokines

Figure 2 Overview of macrophage biology indicating process to“primed” macrophage (step 1) by interferon-g and subsequent

activation through the exposure to lipopolysaccharide (step 2).Alternatively, via scavenging and phagocytosis (step 3), macro-

phages can become “reparative” resulting in local tissuerearrangement. Abbreviation: LPS, lipopolysacccharide.


The granule contents are different from those ofpolymorphonuclear neutrophils or eosinophils andmast cells express high-affinity Fc receptors for IgE.They act as the major effector cells in IgE-mediatedimmune-triggered inflammatory reactions, especiallyallergy or immediate hypersensitivity. Mast cells alsoparticipate in the induction of cell-mediated immu-nity, wound healing, and other functions not directlyrelated to IgE-mediated degranulation (61,62). Otherstimuli, such as complement or certain cytokines, mayalso trigger degranulation (63). Mast cells are alsocapable of inducing cell injury or death through theirrelease of TNF-a. For example, mast cells have beenassociated with neuronal degeneration and death inthiamine deficiency and toxic metabolic diseases.Recent reports have demonstrated the presence ofmast cells in atherosclerotic lesions and the co-localiza-tion of mast cells with the angiogenic protein, platelet-derived endothelial growth factor (63–69).

Mast cells are widely distributed in the connec-tive tissue and are frequently found in close proximityto blood vessels and are in present in abundance in thechoroid (57,70). Mast cells may play important roles inthe pathogenesis of AMD since they have an ability toinduce angiogenesis and are mediators of cell injury.Mast cells have also been shown to accumulate at sitesof angiogenesis and have been demonstrated to bepresent around Bruch’s membrane during both theearly and late stages of CNV in AMD (51). Mast cellscan interact with endothelial cells and induce theirproliferation through the release of heparin, metallo-proteinases (MMPs) and VEGF (71–73). Interestingly,oral tranilast, an antiallergic drug which inhibits therelease of chemical mediators frommast cells has beenshown to suppress laser induced CNV in the rat (74).

T LymphocytesLymphocytes are small (10–20 mm) cells with largedense nuclei also derived from stem cell precursorswithin the bone marrow (3,75,76). However, unlikeother leukocytes, lymphocytes require subsequentmaturation in peripheral lymphoid organs. Originallycharacterized and differentiated based upon a series ofingenious but esoteric laboratory tests, lymphocytescan now be subdivided based upon detection ofspecific cell surface proteins (i.e., surface markers).These “markers” are in turn related to functional andmolecular activity of individual subsets. Three broadcategories of lymphocytes have been determined: Bcells, T cells and non-T, non-B lymphocytes.

Thymus-derived lymphocytes (or Tcells) exist inseveral subsets (77,78). Helper Tcells function to assistin antigen processing for antigen-specific immunitywithin lymph nodes, especially in helping B cells toproduce antibody and effector T cells to becomesensitized. Effector T lymphocyte subsets function as

effector cells to mediate antigen-specific inflammationand immune responses. Effector T cells can be distin-guished into two main types. CD8 T cells (often calledcytotoxic T lymphocytes) serve as effector cells forkilling tumors or virally infected host cells viarelease of cytotoxic cytokines or specialized poreforming molecules. It is possible, but unlikely thatthese cells play a major role in AMD.

CD4 Tcells (often called delayed hypersensitivityT cells) effect responses by the release of specificcytokines such as interferon-g and TNF-b. They func-tion by homing into a tissue, recognizing antigen andAPC, becoming fully activated and releasing cytokinesand mediators which then amplify the reaction.Occasionally, CD4 T cells can also become activatedin an antigen-independent manner, called bystanderactivation (79–81), a process which may explain thepresence of T lymphocytes identified in CNV speci-mens surgically excised from AMD eyes.

B Lymphocytes and AntibodyB-lymphocytes mature in the bone marrow, and areresponsible for the production of antibodies.Antibodies [or immunoglobulins (Igs)] are solubleantigen-specific effector molecules of antigen-specificimmunity (3,75,76). After appropriate antigenic stimu-lation with T cell help, B cells secrete IgM antibodies,and later other isotypes, into the efferent lymph fluiddraining into the venous circulation. Antibodies thenmediate a variety of immune effector activities bybinding to antigen in the blood or in tissues.

Antibodies serve as effectors of tissue-specificimmune responses by four main mechanisms. Intra-vascular circulating antibodies can bind antigen in theblood, thereby form circulating immune complexes(ICs). Then the entire complex of antigen plusantibody can deposit into tissues. Alternatively, circu-lating B cells can infiltrate into a tissue and secreteantibody locally to form an IC. Third, antibody canbind to an effector cell (especially mast cell, macro-phage, or neutrophil) by the Fc portion of the moleculeto produce a combined antibody and cellular effectormechanism. It is unlikely that any of thesemechanisms play a major role in AMD.

However, one possible antibody-dependentmechanism relevant to AMD is the capacity forcirculating antibodies, usually of the IgG subclassespreviously formed in lymph nodes or in other tissuesites, to passively leak into a tissue with fenestratedcapillaries (like the choriocapillaris). Then, theseantibodies form an IC with antigens trapped in theextracellular matrix, molecules expressed on thesurface of cells or even antigens sequestered insidethe cell to initiate one of several types of effectorresponses described below (Fig. 3) (3,75,76,82–85).


Immune Complexes with Extracellular-BoundAntigensWhen free antibody passively leaks from the seruminto a tissue, it can combine with tissue-bound anti-gens (i.e., antigen trapped in the extracellular matrix).These “in situ” or locally formed complexes some-times activate the complement pathway to producecomplement fragments called anaphylatoxins. Thismechanism should be differentiated from depositionof circulating ICs which are preformed in the blood.Typically, the histology is dominated by neutrophilsand monocytes, but at low level of activation minimalcellular infiltration may be observed. Many types ofglomerulonephritis and vasculitis are thought torepresent this mechanism.

Immune Complexes with Cell-Surface AntigenIf an antigen is associated with the external surface ofthe plasma membrane, antibody binding might acti-vate the terminal complement cascade to induce cellinjury via formation of specialized pore-like structurescalled the MAC. Hemolytic anemia of the newborndue to Rh incompatibility is the classic example ofthis process. Hashimoto’s thyroiditis, nephritis ofGoodpasture’s syndrome, and autoimmune throm-bocytopenia are other examples.

Immune Complex with Intracellular Antigen:A Novel MechanismCirculating antibodies can cause tissue injury bymechanisms different from complement activation,using pathogenic mechanisms not yet clearly eluci-dated (84,85). For example, some autoantibodies insystemic lupus erythematosus appear to be interna-lized by renal cells independent of antigen binding,but then combine with intracellular nuclear or

ribosomal antigens to alter cellular metabolism andsignaling pathways. This novel pathway of intra-cellular antibody/antigen complex formationmay cause some cases of nephritis in the absenceof complement activation. This pathway hasalso been implicated in paraneoplastic syndromes,especially cancer associated retinopathy (CAR),in which autoantibodies to intracellular photo-receptor-associated antigens may mediate rod orcone degeneration (86).

Mechanisms for the Activation of the ImmuneResponses in Degenerative DiseasesActivation of Innate ImmunityCellular Injury as a Trigger of Innate ImmunityNot only can immune responses cause cellular injury,but cellular responses to nonimmune injury are alsocommon initiators of innate immunity (3,75,76,87–89).Injury can be defined as tissue exposure to anyphysical and/or biochemical stimulus that alterspreexisting homeostasis to produce a physiologicalcellular response. In addition to injury stimuliproduced by the immune effector and amplificationsystems described above, nonimmune injuriousstimuli include physical injury (heat, light, mechan-ical) or biochemical stimulation (hypoxia, pH change,oxidants, chemical mediators, cytokines) (89). Typicalcellular reactions to injury include a wide spectrum ofresponses, including changes in intracellular metab-olism, plasma membrane alterations, cytokineproduction, and gene upregulation, morphologicalchanges, cellular migration, proliferation, or evendeath. Some of these cellular responses, in turn, canresult in the recruitment and activation of macro-phages or activation of amplification systems,especially if they include upregulation of cell adhesionmolecules, production of macrophage chemotacticfactors or release of activational stimuli.

Two important injury responses relevant to AMDthat commonly activate innate immunity includevascular injury and extracellular deposit accumulation(89,90). Vascular injury induced by physical stimuli(i.e., mechanical stretch of capillaries or arterioles byhydrostatic expansion induced by hypertension orthermal injury from laser) or biochemical stimuli(i.e., hormones associated with hypertension andaging) can upregulate cell adhesion molecules andchemotactic factors that lead to macrophage recruit-ment into various vascularized tissues. Extracellulardeposit accumulation can also contribute to activationof innate immunity by serving as a substrate forscavenging and phagocytosis, especially if thedeposits are chemically modified by oxidation orother processes (see atherosclerosis below).

RPE

BM

CC

IC Formationin Bruch'sMembrane

Antibody Effectors in ARMD

IC on Cell Surfacewith MAC Activation

Intracellular IC

Figure 3 Possible antibody effects in age-related maculardegeneration (AMD) with subsequent immune complex (IC)

formation at variation locations in the subretinal space, on orwithin the retinal pigment epithelium (RPE). Abbreviation: MAC,

membrane attack complex.


Infection as a Trigger of Innate ImmunityInfection can also activate innate immunity, usually bythe release of toxic molecules (i.e., endotoxins,exotoxins, cell wall components) that directly interactwith receptors on macrophages, on neutrophils or, insome cases, on parenchymal cells. Active infection isdifferentiated from harmless colonization by thepresence of invasion and replication of the infectiousagent (91). However, active infections do not alwaystrigger innate immunity, illustrated by some retinalparasite infections in which inflammation occurs onlywhen the parasite dies.

Recently, the idea has emerged that certain kindsof chronic infections might cause (or at least contributeto) degenerative diseases that are not considered to betruly inflammatory (88–91). One of the most dramaticexamples is peptic ulcer disease, recently recognizedto be caused by infection of the gastric subepithelialmucosa with a gram-positive bacterium called Helico-bacter pylori (92). Accordingly, ulcer disease is nowtreated by antibiotics and not with diet or surgery.Recently, chronic bacterial or viral infection of vascularendothelial cells has been suggested as an etiology forcoronary artery atherosclerosis, and infection withan unusual agent called a prion has been shown as acause of certain neurodegenerative diseases. Therelevance to AMD is discussed below.

Activation of Normal and AberrantAntigen-Specific ImmunityActivation of Antigen-Specific ImmunityIt is often expressed as the idea of the “immuneresponse arc.” This idea proposes that interactionbetween antigen and the antigen-specific immunesystem at a peripheral site (such as the skin) can

conceptually be subdivided into three phases: afferent(at the site), processing (within the immune system),and effector (at the original site completing the arc)(Fig. 4) (3,75,76). Antigen within the skin or anyother site is recognized by the afferent phase of theimmune response, which conveys the antigenic infor-mation to the lymph node in one of two forms. APCs,typically DC, can take up antigen (almost always inthe form of a protein) at a site, digest the antigen intofragments and carry the digested fragments to thelymph node to interact with T cells (77,78,93). Alter-natively, the natural, intact antigen can directly flowinto the node via lymphatics where it interacts with Bcells (3,75,76).

In the lymph node, processing of the antigenicsignal occurs where antigen, APC, T cells and B cellsinteract to activate the immune response. For tissueswithout draining lymph nodes (such as the retina andchoroid), the spleen is often a major site of processing.Immunologic processing has been the topic of exten-sive research and the details are too complex to discussin this brief review. Processing results in release ofimmune effectors (antibodies, B cells and T cells) intoefferent lymphatics and venous circulation whichconveys the intent of the immune system back to theoriginal site where an effector response occurs (i.e., ICformation or delayed hypersensitivity reaction).Compared to that of the skin, the immune responsearc of the retina and choroid express many similaritiesas well as important differences (i.e., immune privi-lege, anatomy), which are discussed in recent reviews(94,95).

Aberrant Activation of Antigen-SpecificImmunityThe inappropriate activation of antigen-specificimmunity may play a role in the pathogenesis ofchronic degenerative diseases. Autoimmunity is theactivation of antigen-specific immunity to normal selfantigens, and two different mechanisms of autoimmu-nity may be relevant to AMD: molecular mimicry anddesequestration. Additionally, immune responsesdirected at “neo-antigens” or foreign antigens inap-propriately trapped within normal tissues may alsoplay a role in AMD.

Molecular mimicry is the immunologic cross-reaction between antigenic regions (epitopes) of anunrelated foreign molecule and self-antigens withsimilar structures (96). For example, immune systemexposure to foreign antigens, such as those presentwithin yeast, viruses, or bacteria, can induce anappropriate afferent, processing, and effectorimmune response to the organism. However, antimi-crobial antibodies or effector lymphocytes generatedto the organism can inappropriately cross-react withsimilar antigenic regions of a self-antigen. A dynamic

Immune Response Arc

AfferentDendriticCell

APC

Efferent

Lymphnode

TissueSite

T

Lymphatic

CirculationCTL

B

TCTL

TDH

TDH

Plasma Cell

Figure 4 The immune response arc indicating cross-talkbetween the tissue site, where antigen recognition and effector

processes take place, and the lymph node, the site of antigenprocessing. Abbreviation: APC, antigen presenting cell.


process would then be initiated, causing tissue injuryby an autoimmune response that would induceadditional lymphocyte responses directed at otherself-antigens. Thus, the process would not requirethe ongoing replication of a pathogen or the continu-ous presence of the inciting antigen. Molecularmimicry against antigens from a wide range of organ-isms, including Streptococcus, yeast, E. Coli and variousviruses, has been shown to be a potential mechanismfor anti-retinal autoimmunity (97).

A second mechanism for aberrant autoimmu-nity is desequestration (98–100). For most self-antigens, the immune system is actively “tolerized”to the antigen by various mechanisms, preventing theactivation of antigen-specific immune effectorresponses even when the self antigen is fullyexposed to the immune system. For some otherantigens, however, the immune system relies onsequestration of the antigen within cellular compart-ments that are isolated from APCs and effectormechanisms. If the sequestered molecules areallowed to escape their protective isolation, they canbecome recognized as foreign, thereby initiating anautoimmune reaction. For example, certain nuclear orribosomal-associated enzymes are apparently seques-tered, and if organelles become extruded into alocation with exposure to DC or macrophages, animmune response can be triggered against theseantigens (99). Accordingly, some RPE and retina-associated peptides appear to be sequestered fromthe immune system and could potentially serve asantigens if RPE injury or death leads to their releaseinto the choroid (94,100).

Another mechanism for aberrant activation ofantigen-specific immunity is the formation of neo-antigens secondary to chemical modification ofnormal self proteins trapped or deposited withintissues (101). For example, oxidation or acetylation ofpeptides in apolipoproteins trapped within athero-sclerotic plaques can induce new antigenic propertiesresulting in specific T cell and antibodies immunizedto the modified protein.

A final mechanism for aberrant antigen-specificimmunity is antigen trapping (102).Antigen trapping isthe immunologic reaction to circulating foreign anti-gens inappropriately trapped within the extracellularmatrix of a normal tissue site containing fenestratedcapillaries. Typically occurring after invasive infectionor iatrogenically administered drugs, this mechanismmaybevery important in glomerulardiseases (102) andhas been postulated to induce ocular inflammation(103,104). Physical size and charge of the antigen areimportant. For example, antigen trapping within thechoriocapillaris may contribute to ocular histoplas-mosis syndrome (OHS) (104).

EXAMPLES OF IMMUNE AND INFLAMMATORYMECHANISMS OF NONOCULARDEGENERATIVE DISEASES

Immune Mechanisms in AtherosclerosisMyocardial infarction due to thrombosis of athero-sclerotic coronary arteries is the major cause of deathin western countries, and epidemiologic studiessuggest a possible association with AMD (105,106).The pathology of atherosclerosis suggests a spectrumof changes whose pathogenesis may be relevant to theunderstanding of AMD (107,108). The fatty streak,representing the earliest phase of atherosclerosis, ischaracterized by lipid deposition and macrophageinfiltration within the vessel wall (101,108,109).Some investigators have suggested similarities inpathogenesis between fatty streak formation andearly AMD (110). The fatty streak can progress intothe fibrous plaque, characterized by the proliferationof smooth muscle cells, increasing inflammation, andformation of connective tissue with neovasculariza-tion within the vessel wall. The fibrous plaquepredisposes to the complications of atherosclerosissuch as thrombosis, dissection or plaque ulceration(101,108,109). The pathogenesis of the fibrous plaquemay share similarity with mechanisms for the latecomplications of AMD, including formation of CNVand disciform scars (Fig. 5).

Many mechanisms contribute to the patho-genesis of atherosclerosis, including geneticpredisposition and physiologic risk factors like highblood cholesterol, smoking, diabetes, and hyperten-sion. However, most authorities now believe thatchronic low grade inflammation, induced by a widevariety of injury stimuli, followed by a fibroprolifera-tive (wound healing) response within the vessel wallis central to the pathogenesis of atherosclerosis. Thus,various immune mechanisms implicated in athero-sclerosis might be relevant to AMD.

Innate MechanismsInjury and AtherosclerosisThe response to injury hypothesis for the initiation andprogression of atherosclerosis has been supported bynumerous investigators who cite many different parti-cipating injury stimuli (101,108,109). For example,hemodynamic injury by blood flow turbulence candirectly injure endothelial cells at bifurcations ofmajor vessels (113). Biochemical injury secondary toexposure to polypetide mediators associated withhypertension (i.e., angiotensin II or endothelin-1) canstimulate the endothelial and smooth muscleresponses. Oxidized low density lipoprotein (LDL)cholesterol particles in the blood, advanced glycosyla-tion end products in diabetes or toxic chemicals


secondary to smoking are other potential sources ofinjury (114).

Macrophages in AtherosclerosisBlood-derived macrophages are major contributors tothe pathogenesis of atherosclerosis (101,108,115). In thefatty streak phase of atherosclerosis, lipids accumulatein the subendothelial vascular wall at sites of vascularinjury. Injury results in the oxidation of lipids orendothelial production of specific macrophage chemo-tactic signals, like macrophage chemotactic protein-1,recruiting circulating monocytes to sites of endothelialinjury. There, they migrate into the subendothelialextracellular matrix to scavenge the extracellularlipid-rich deposits (i.e., scavenging macrophages).Macrophages may also contribute to the solubilizationof lipid deposits by the release of apolipoprotein E(ApoE), whichmay facilitate uptake and scavenging oflipids. Genetic polymorphisms of ApoE have beenassociated with variations in the severity of athero-sclerosis and AMD (116).

Foam cells and macrophages are very numerousin fibrous plaques, and probably play a major role inlesion progression. Although overly simplistic, experi-mental data suggest that scavenging macrophagescan become activated into reparative “foam” cells bynumerous stimuli, including phagocytosis of oxidizedlipoproteins (115,116). Reparative macrophagessecrete amplifying mediators, including plateletderived growth factor (PDGF), VEGF, matrix MMPsor others which contribute to fibrosis, smoothmuscle proliferation, or vascularization of the plaque(117–120).

Infectious Etiology of AtherosclerosisAlthough numerous risk factors are associated withthe initiation and progression of atherosclerosis, aninfectious etiology has been suggested by recent data.Many patients with atherosclerosis exhibit signs ofmild systemic inflammation, especially elevatedserum C-reactive protein (CRP) and erythrocyte sedi-mentation rate (121). Statistical evidence has beengenerated to suggest that infection with various infec-tious agents, especially Chlamydia pneumoniae orcytomegalovirus (CMV), might initiate vascularinjury and explain the systemic inflammatory signs(122–125).

Numerous epidemiologic studies have revealeda statistical correlation between atherosclerosis andserologic evidence of infection with C. pneumoniae(122). Follow-up studies have demonstrated thepresence of C. pneumoniae by histochemical methodswithin atherosclerotic plaques and organisms havebeen cultured from the lesions (125). Additionally,pilot studies using appropriate antibiotic therapyhave demonstrated a beneficial effect in patientswith severe atherosclerosis (123,124). Severalproposed mechanisms for the role of C. pneumoniaein atherosclerosis may be relevant to AMD. Chronicinfection of vascular endothelial cells may upregulatecell surface molecules that recruit macrophages oralter responses to injury. For instance, C. pneumoniaeendothelial infection can enhance endotoxin bindingto LDL particles which might induce various inflam-matory cascades at the site of uptake (126).Additionally, chlamydial heat shock proteins (HSPs)can directly stimulate macrophages and other cellular

(B) (A)

BLD

AMDPlaque

Figure 5 Micrographs of an atheromatous plaque (left) and a choroidal neovascular membrane (right)indicating similar histologic components of intrastromal neovascularization (arrows) and macrophages

(left—B) and (right—asterisk). Abbreviations: AMD, age-related macular degeneration; BLD, basal laminardeposits. Source: From Refs. 111, 112.


amplification systems (127). Also, antigen-specificimmune responses directed against chlamydial HSPsmay cross-react with host cellular HSP including thoseexpressed in the retina (128).

Similar, but less extensive data have been gener-ated to support a role of CMV infection inatherosclerosis (129–131). CMV infects 60% to 70% ofadults in the U.S.A. Several studies have linkedserologic evidence of prior CMV infection to athero-sclerosis. Although the association is mild, studieshave elucidated possible mechanisms for this associ-ation such as enhanced scavenging of LDL particles byvirally infected endothelial cells.

Antigen-Specific ImmunityThe potential importance of antigen-specific immunemechanisms in atherosclerosis is illustrated by theobservation of accelerated atherosclerosis in hearttransplant patients who experience vascular injuryassociated with mild, chronic allograft rejection (93).In normal patients with atherosclerosis, T lympho-cytes are numerous in fibrous plaques and a role forlymphocyte-mediated antigen specific immunity hasbeen proposed for progression of atheroscleroticfibrous plaques (101). Experimental data suggest thatoxidized lipoproteins can become neo-antigens toactivate an immune response arc (132). Scavengingmacrophages may become APCs at the site, serving torestimulate recruited T cells thereby activating theeffector phase of the immune response. Immuneresponses to bacterial or viral antigens, especiallychlamydial HSPs, trapped in tissues after occult infec-tion may also stimulate antigen-specific immunity, orautoimmunity by cross-reactive molecular mimicry(133). Alternatively, T cells may be recruited byinnate responses and become activated by antigen-independent bystander mechanisms. Interestingly,vaccination against oxidized LDL produces antibodieswhich seem to prevent or reduce formation of athero-sclerotic plaques (19), similar to that observed inAlzheimer’s disease (AD) (see below).

Nonspecific Amplification CascadesComplement Activation in AtherosclerosisIn atherosclerotic lesions, several complement com-ponents and inhibitory proteins have been detectedincluding MAC complexes (134–136). Cholesterol isalso a potent activator of the complement systemin vitro. Alternatively, MAC complex concentrationhas been shown to induce macrophage chemotacticfactor production in smooth muscle cells and studieshave shown MAC deposition in the arterial wall priorto monocyte infiltration and foam cell formation.Interestingly, in addition to its cytotoxic function,limited complement activation and deposition of thecomplement precursor protein C1q on apoptotic cells,

cell debris and cell membrane blebs can enhancephagocytosis by C1q—receptor bearing macrophagesand may play a role in tissue repair.

Oxidants and Cytokines in AtherosclerosisOxidation is considered to be a major injury stimulusin the initiation and progression of atherosclerosis. Therole of oxidized lipoproteins in circulating LDL choles-terol as an initiating injury stimulus as well asoxidation of lipid deposits within vessel walls as anamplifier of foam cell activation has been discussedabove (114,115,137). Numerous cytokines, especiallyPDGF and transforming growth factor-b have alsobeen implicated as major mediators of atherosclerosisprogression (117–120).

Immune Mechanisms in Glomerular DiseasesGlomerular diseases account for 70% of chronic renalfailure in the U.S.A. Many glomerular diseases areprimarily mediated by inflammatory mechanisms,and are usually classified as glomerulonephritis.Other glomerular diseases are mediated by a mixtureof degenerative and inflammatory mechanisms, andthese are often classified as glomerulosclerosis(138,139). Genetic and systemic health factors contrib-ute to the pathogenesis of both groups (138–141).

The glomerulus shares some anatomic simi-larities with the outer retina and inner choroid, sothat analysis of the mechanism of deposit formationand extracellular matrix changes of glomerulardisorders might be informative in terms of AMD(138). For instance, both the glomerulus and innerchoroid/outer retina can be described as containingcapillary lobules with endothelium on one side of anextracellular matrix and epithelium on the other. In theglomerulus, endothelial cells (conceptually corre-sponding to the choriocapillaris) cover the internalsurface of an extracellular matrix, whose externalsurface is covered by an epithelial layer (the podo-cyte). External to the podocyte is Bowman’s capsule(conceptually corresponding to the subretinal space).Smooth muscle cells located internally to the endo-thelium, called mesangial cells, are responsible forregulating contractility and maintaining the glomer-ular matrix. These cells may share analogies withchoroidal pericytes underlying and surroundingthe choriocapillaris.

Innate Immunity in Glomerular DiseasesChronic InjuryAs is the case for atherosclerosis, a response to injuryhypothesis has been substantiated for glomerulo-sclerosis due to aging, hypertension or diabetes(138–145). Glomerulosclerosis is characterized by pro-gressive thickening of the glomerular extracellularmatrix ultimately associated with loss of glomerular


capillaries and epithelial cells. If enough glomeruli areinvolved, renal impairment occurs. In some ways,glomerulosclerosis resembles geographic atrophy inAMD (Fig. 6).

The response to injury hypothesis has beenthoroughly evaluated for renal hypertension, a majorcause of glomerulosclerosis (141–145,148). The hemo-dynamic injury hypothesis proposes that glomerularcapillary hypertension causes excessive flow throughthe glomerulus or hydraulic stretching of the capillarywall to activate injury responses in glomerular cells.The humoral hypothesis proposes that hypertension-associated hormones or cytokines associated with lowgrade systemic inflammation induced by hypertensivevascular injury, activate cellular injury responses. Ineither case, the injured endothelium, podocytes andmesangial cells demonstrate abnormal productionand turnover of collagen and other matrix mole-cules, leading to collagenous thickening of the matrixwith degeneration of the glomerulus (148–150).Genetic background and gender can influence therate of progression. Since hypertension is a riskfactor associated with AMD and glomerular disease,hypertension-associated inflammation may also injurethe choriocapillaris endothelium or RPE in ananalogous fashion.

Macrophage-Mediated InjuryMacrophages contribute significantly to glomerulardamage in renal diseases (151–161). Not surprisingly,infiltration with activated inflammatory macro-phages is a significant histologic feature in inflam-matory glomerulonephritis caused by antigen-specificimmune mechanisms (i.e., IC disease or allograft

rejection) (161), and blockade of macrophage infiltra-tion or function ameliorates glomerular damage (155).Perhaps of more relevance to AMD is the contributionof reparative macrophages to glomerulosclerosis.Recruitment of blood-derived reparative macrophagesdevelops early in the course of glomerulosclerosis inproportion to the severity of the injury (151,152).Various innate injury stimuli, including renal hyperten-sion, hyperlipidemia, and glomerular capillaryendothelial injury by oxidized LDL, can upregulatemacrophage chemotactic factors and adhesionmolecules in the capillaries to induce macrophagerecruitment (156–158). Experimental data suggest thatreparative macrophages release mediators that inducemesangial cell proliferation, amplify the accumulationof extracellular matrix and might induce killing ofendothelial cells.

Antigen-Specific Immunity in Glomerular DiseasesAntigen-specific immunity contributes significantly toinflammatory glomerular disorders. Lymphocyte-mediated immunity clearly contributes to glo-merulonephritis, especially in renal allograft rejection(161). However, the relevance of this mechanism toAMD is probably minimal. Many forms of chronicglomerulonephritis are caused by antibody-dependentmechanisms, and some of these disorders are charac-terized by subendothelial or subepithelial depositformation (102,162–164). Direct deposition of circu-lating antibodies targeted at antigens uniformlyexpressed within the glomerular matrix is a well-defined but rare form of glomerulonephritis, especiallyin Goodpasture’s syndrome. Deposition of preformedcirculating antigen/antibody complexes in the bloodhas been proposed as another major mechanism inmany types of glomerulonephritis associated withdeposit formation. Nevertheless, it is unlikely thatdeposition of either anti-basement membrane anti-bodies or circulating ICs plays an important rolein AMD.

However, another interpretation of the clinicaland experimental data is that some forms of glomer-ulonephritis may actually represent antigen trappedor “planted” within the glomerular matrix, followedby the subsequent formation of in situ ICs. Thisalternative explanation is probably especially relevantto glomerulonephritis associated with subepithelialdeposits rather than subendothelial deposits (since itis unlikely that large ICs would be able to filterthrough the matrix). For example, glomerulonephritisthat occurs 10 to 20 days after streptococcal pharyn-gitis or streptococcal skin infections is characterizedby subepithelial deposits [similar to homogenousbasal laminar deposits (BLD)]. These do not stain forICs (165).

ELM

RPE RPE

BLD

AMD

M

M

GS

Figure 6 Electron micrographs from glomerulosclerosis andgeographic atrophy from age-related macular degeneration

(AMD) showing appearance of excessive extracellular materialand cellular loss. In glomerulosclerosis (GS) there is accumu-

lation of glomerular extracellular material (asterisks) and loss ofcellular structure (M) while in AMD there is accumulation of basal

laminar deposits (BLD) and loss of retinal pigment epithelium(RPE) cells under the external limiting membrane. Abbreviation:

ELM, external limiting membrane. Source: From Refs. 146, 147.


Nonspecific Amplification Cascadesin Glomerular DiseasesComplement deposition plays a major primary role inmany glomerular diseases associated with deposits,especially those mediated by antigen-specific ICs. Inthese disorders, various fragments of the complementcascade, including C3, C5, and others are usuallyidentified within extracellular deposits in associationwith Ig and acute cellular inflammation (166–168).

Complement seems to participate as a secondaryamplification mechanism in some glomerulardiseases. Type II membranoproliferative glomerulone-phritis (or dense deposit disease), is especially relevantto AMD since these patients also develop drusen-likechanges in the retina (168–171). Clinically, the retinademonstrates whitish drusen-like changes, and someeyes develop CNV. Histologically, the subretinaldeposits appear to be localized between the RPE andits basement membrane (similar to BLD) (Fig. 7). Theglomerular deposits are characterized as electrondense linear deposits within the glomerular extracellu-lar matrix, occasionally demonstrating dome-shapedsubepithelial “humps” under the podocyte. Comp-lement 3 is present within the deposits, but thepresence of other complement proteins, Igs, andfibronectin is highly variable. Systemic comple-ment is usually normal. The source of complement(i.e., locally synthesis or blood-derived) as well asthe mechanisms for activation (typical cascades vs.enzymatic cleavage) remain unknown. Finally,oxidants have been implicated as important mediatorsand amplifiers in progression of renal disease (173).

EVIDENCE FOR IMMUNE AND INFLAMMATORYMECHANISMS IN AMD

Direct Evidence for Innate or Antigen-SpecificImmune EffectorsDirect evidence for the role of immune mechanism inAMD is scant. The best data suggest an important rolefor macrophage-mediated innate immunity (22,50–58).Investigators have observed that choroidal macro-phages appear to be important in the pathogenesis ofboth early and late AMD. However, macrophageinvolvement is clearly different than their partici-pation in overt inflammatory disorders characterizedby widespread cellular infiltration.

In early AMD, macrophages have been detectedalong the choriocapillaris-side of Bruch’s membraneunderlying areas of thick deposits. Processes fromchoroidal monocytes have been noted to insert intoBruch’s membrane deposits, presumably for thepurpose of scavenging debris. The identity of thesecells is uncertain, but they seem to lack typicalphagocytic vacuoles and express human leukocyteantigen DR, suggesting that the cells may representDC or nonactivated macrophages (22).

In late AMD, macrophages and giant cells havebeen observed around choroidal neovascularmembranes (CNVM) and are numerous in excisedCNVM, suggesting a role in promoting choroidalangiogenesis (50,53,174). Also, macrophages arepresent underlying zones of geographic atrophy,suggesting a role in RPE or endothelial death (52).These observations imply a potential pathogenic rolefor cytokines, chemical mediators, MMPs, mitogens orangiogenic factors released by macrophages from thechoroid. In support of this concept, numerous investi-gators have demonstrated that macrophage-derivedcytokines (especially TNF-a) induce major functionaland morphological changes in RPE cells (175–179).Further, macrophage involvement may be underesti-mated in AMD. Choroidal macrophages are oftendifficult to detect by routine histopathology in nonin-flammatory disorders because they typically acquiremuch flattened profiles. Finally, evidence from severalrecent clinical trials has shown a benefit from intra-vitreal corticosteroid therapy in the treatment of CNVin AMD patients (180,181). Corticosteroids are potentmodulators of macrophage function and these studiessuggest that more research should explore the thera-peutic potential of nonspecific anti-inflammatorytherapy in AMD.

Evidence for antigen-specific immunity has notbeen described in AMD. The possible contribution ofantibody-dependent mechanisms is suggested byrecent understanding of the mechanism for CAR (seenext section below). In AMD, one group has identifiedIgG and MAC association with RPE overlying drusen

RPE

Choroid

Figure 7 Electron micrograph of dense deposit disease of the

retina demonstrating subretinal deposit (box) located betweenthe retinal pigment epithelium (RPE) and its basement

membrane. Source: From Ref. 172.


(182). However, another study has identified onlyantibody light chains within drusen, but not thepresence of associated heavy chain to indicate anintact Ig molecule (22). Lymphocytes, especially Tcells, have been identified within some CNV (53).It remains unknown if these cells are recruited aspart of bystander activation or are responding toantigen-specific immunity. However, bystanderrecruitment of T cells occurs in many other forms ofpathological neovascularization and wound healing.

Nonspecific amplification mechanisms may alsoplay a role in AMD. Recently, several groups haveidentified complement components in drusen (22,182).Fragments of C5 and the MAC were identified in mostspecimens, and C3was present in some. The activationpathway remains unknown. The RPE express specificand nonspecific complement inhibitors such as decayaccelerating factor and vitronectin to suggest intrinsicdefense mechanisms to prevent against complement-mediated injury (183).

Ocular Immune and Inflammatory DisordersResulting in Atrophic RetinalDegeneration or CNV

Ocular Histoplasmosis SyndromeOHS may represent a condition to suggest a role forinfection-triggered immunity as a mechanism for RPEinjury and CNV formation. The syndrome ispresumed to be induced by the inhalation of livehistoplasmosis capsulatum, which infects the lungand hilar lymph nodes (184). In some patients, sys-temic dissemination of the organism occurs, includinginto the choroid, but the organism is rapidly recog-nized and killed by the immune response. Accordingto data from a primate model, the acute phase ofimmune response can induce clinically detectable,small multifocal creamy lesions in the deep retinaand choroid caused by localized choriocapillarisinflammation mediated by CD4 T cells (presumablydelayed hypersensitivity) (185,186). However, manyother areas of active choroidal inflammation are clini-cally not apparent. Ultimately the overlying RPEbecome detectable as atrophic “histo spots.” Chronicpersistent low grade inflammation apparently triggersCNV formation (187,188). Additionally, many otherforms of chronic chorioretinitis are also associatedwith RPE atrophy and CNV formation, and some ofthese may represent occult infection of the choroid orRPE with virus or other infectious agents (189).

The role of infection in AMD remains entirelyspeculative. Although it is unlikely that histoplas-mosis contributes to AMD, trapping of antigensrelated to other common organisms conceivablycould occur. Based on analogies to the role of infection

in fibrous plaque progression, investigation of possiblecontributions from choroidal endothelial infectionwith chlamydia or CMV might be informative.Finally, as new unusual infectious agents, such asprions, are being discovered, the potential role ofretinal or RPE infection in AMD should at leastbe considered.

Complement Activation in AMDCFH is a single polypeptide chain plasma glycoproteinof 155 kDa size that is found in the plasma at aconcentration of 110–615 mg/mL as well as in multipletissues (190). The structure of CFH is composed of20 repetitive units of 60 amino acids so-called shortconsensus repeats (SCRs) (Fig. 10). CFH binds princi-pally to C3b and accelerates the decay of the alternativepathway D3-convertase and participates as a cofactorfor the factor-I–mediated proteolytic inactivation ofC3b (190). Interestingly, it is the binding of CFH tosialic residues on the cell surface which is critical toits ability to inhibit C3b. CFH also binds to othermultiple residues within various bacteria. While theprimary site of synthesis is the liver, multiple extra-hepatic sites of synthesis have been demonstratedincluding within lymphocytes, fibroblasts, endothelialcells, neurons, and glial cells (192). Recently CFH hasalso been shown to be produced by the RPE/choroidcomplex (193) and abundant CFH is present in bothchoroid and outer retina of patients with AMD(193,194). The function of the protein is to preventinadvertent complement activation in all tissues.

Multiple studies have confirmed the associationof mutations with the CFH gene and an increased riskof AMD (194–202). Of the many mutations, a tyrosineto histidine amino acid shift at residue 402 has beenthe most consistent finding. This amino acid shiftoccurs within SCR 7, a position which importantbecause of binding both to heparin residues and CRPand also to various bacterial components specificallythose from Streptococcus pyogenes, Borrelia burgdorferi,Borrelia afzellii, and Candida albincans (Fig. 8) (203,204).This is interesting because of the recent demonstrationof C. pneumoniae remnants found in CNV pathologyspecimens suggesting a link between acquired infec-tion, interaction with CFH and CNV formation (205).

The idea that altered activity of CFH may allowunbridled activation of the complement therebyleading to various stages of AMD has been supportedby laboratory experiments. Complement 3 depositionhas been shown in a laser induced CNV model and itsdepletion in knockout animals prevents CNV fromforming (206). In addition, MAC deposition was alsodemonstrated and its inhibition also prevented experi-mental CNV (206) while the absence of receptors forC3a and C5a also reduced CNV formation. The role ofother inflammatory mediators which associate with


CFH is also now being explored. For example, recently,high levels of CRP have been demonstrated in thechoroid of patients with a homozygous 402 CFHmutation (207). New information suggests that thealternative pathway for complement activation maybe the most critical method for CNV formation (207).The possibility of inadvertent activation of comp-lement in the study of an AMD microenvironmentwas shown in-culture with photoactivation of A2E, aportion of lipofuscin that accumulated in RPE cells,can activate C3 (19). Bioactive fragments of C3a andC3b have been demonstrated in drusen of patientswith AMD (208).

IMMUNE MECHANISMS IN AMD: FINALQUESTIONS AND FUTURE DIRECTIONS

Is the Response to Injury HypothesisApplicable to ARMD?As discussed above, the response to injury hypothesishas become one of the central paradigms for thepathogenesis of atherosclerosis, AD, and glomerulo-sclerosis. The response to injury paradigm proposesthat pathological features of degenerative diseases canbe explained by exaggerated or abnormal cellularreparative responses induced by exposure to chronic,recurrent injurious stimuli. Both genetic and physio-logic factors can contribute to injury or repair. Thischapter has focussed on the physiologic role of innateimmunity, antigen-specific immunity and immuneamplification systems as potential triggers of injuryand as modulators of abnormal repair.

In terms of AMD, response to injury is implicit inpathogenic models that propose a role for variousinjurious stimuli, such as oxidants, lipofuscin cytotox-icty and other factors. Injury stimuli relevant to othersystemic diseases associated with AMD have not beencarefully evaluated, including hyperlipidemia,

oxidized lipoproteins, hormonal changes associatedwith aging or hypertension (209). Presumably, photo-receptors, RPE, choricapillaris endothelium and/orchoroidal pericytes may all be relevant targets.However, to exploit the full power of the response toinjury paradigm, AMD investigators must moreprecisely delineate the relevant cellular responses toinjury in order to explain the specific pathologicalchanges in AMD. Cellular repair responses are mani-fested by a wide spectrum, ranging from transientmetabolic changes to cell death (210,211). The appro-priate cellular response must be matched to a specificpathological change. For example, analysis ofprogrammed cell death in response to lethal injury isrelevant to the understanding of geographic atrophyof the RPE (212). However, it is unlikely that analysisof cell death will explain the formation of sub RPEdeposits, recruitment of macrophages or CNV forma-tion. RPE can react to nonlethal injury by manyresponses relevant to deposit formation, includingby extruding patches of cell membranes and cytosol(i.e., blebs) (211), by altering the synthesis of collagen,matrix MMPs and other matrix molecules, byincreasing production of chemotactic signals or angio-genic factors, and many other responses (213). Theseother specific responses need to be correlated withspecific injury stimuli.

Recent studies of RPE injury responses mayserve as an example how immunity can inducedeposits or promote abnormal repair. RPE can beinjured by myeloperoxidase-mediated lipid per-oxidation of the cell membrane, which represents aphysiologically relevant macrophage-derived oxi-dative stimulus. Such oxidant-injured RPE undergosignificant blebbing of cell membrane (Fig. 9), cytosol,

CRPHeparinC3b

HeparinC3b

Sialic acidHeparin

C3b

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

S. PyogenesC. AlbincansB. Burgdorferi

Figure 8 Schematic of the structure of complement factor H

demonstrating the 20 tandem repeats of short consensus repeats(SCR). The SCR 7 is highlighted because of the high prevalence

of mutations at the tyrosine 402 residue within that domainwhich might affect binding to the various structures outlined.

Abbreviations: CRP, C-reactive protein.

RPE

RPE

Blebs

Figure 9 Image of retinal pigment epithelial (RPE) cellsin-culture exhibiting extensive cell membrane blebbing following

sublethal oxidative injury.


and organelles, but without activation of programmedcell death or nuclear fragmentation. However, oxidantinjured cells downregulate another response, matrixMMPs production (Cousins and Csaky, personalcommunication).

Irrespective of the stimulus, accumulation ofblebs can lead to deposit formation which, in turn,can activate an immune response which interfereswith healthy repair. For example, under certain con-ditions, blebs might serve as an innate stimulus forrecruitment and activation of reparative macrophages(see below). In addition to innate immunity, blebbingmight cause desequestration of intracellular antigensto provide a target for antigen-specific immunity orblebs might provide a substrate for nonspecific acti-vation of complement or other amplification systems,as described for atherosclerosis, AD, and glomerulardiseases.

Response to injury may also be relevant toformation of CNV. All blood vessels, including thechoriocapillaris, must continuously repair endothelialand vessel wall damage following injury. Increasingevidence suggests that aging is associated with dysre-gulated vascular repair after injury (113,214,215). Forexample, abnormal and exaggerated repair followingacute vascular injury is a well-defined mechanism foraccelerated restenosis after coronary artery angio-plasty in older patients (196). A similar phenomenonmay exist in the choroid in terms of CNV. Aging miceexposed to laser injury of the choroid develop muchlarger CNV than do younger animals. Investigation ofdifferences between younger and aging individuals interms of activation of immune and reparativeresponses after vascular injury may be an importanttopic for research in AMD (Cousins and Csaky,personal communication).

What Is the Role of Choroidal Monocytes?Although the presence of choroidal monocytes inAMD has been established, their identity and functionremains uncertain. If analogies with atherosclerosisare correct, then these cells are probably scavengingmacrophages recruited to remove lipids and depositswithin Bruch’s membrane. As is the case for athero-sclerosis or AD, the function of scavenging monocytesin AMD can be protective or pathogenic dependingupon the activation status (Fig. 10). Scavenging macro-phages probably can remove sub RPE deposits safelyand assist in healthy repair of Bruch’s membrane.However, activation into reparative macrophagesmay result in the production of mediators that candamage Bruch’s membrane, injure choriocapillarisand promote CNV. Recently, it has been shown thatblood monocytes from some patients with AMD canbecome stimulated into reparative macrophages after

phagocytosis with RPE-derived cell debris andmembrane blebs. Analysis of interaction between subRPE deposits and scavenging macrophages mayaddress this topic.

Does Antigen-Specific Immunity Participatein the Progression of AMD?If the identification of DC in association with drusen isconfirmed, this observation implies an entirelydifferent function for choroidal monocytes andsuggests a role for T cell-mediated antigen-specificimmunity. DC lack scavenging and inflammatoryeffector functions. However, they might sample anti-gens within drusen (perhaps inappropriatelydesequestered or chemically modified proteins), andthen might initiate the afferent phase of the immuneresponse by presenting these antigens to T lympho-cytes within lymphoid tissues.

The participation of antibody-dependentimmune responses in AMD is intriguing but remainsspeculative. Typically, B cells require exposure to thenatural, intact antigen within lymph nodes to becomeactivated, not exposure to antigens that wereprocessed and presented by DC. It is possible, butunlikely, that intact retina-specific antigens in AMDcan diffuse into the choroid, gain access to lymphoidcompartments and trigger a “de novo” retina-specificimmune antibody response. Nevertheless, as in CAR,circulating antibodies, perhaps produced in responseto immunity triggered elsewhere in the body bymolecular mimicry, desequestration, or neoantigenformation, might cross-react with similar antigenstrapped within subretinal deposits or expressedwithin ocular cells. A similar mechanism has beendescribed in atherosclerosis. Finally, investigatorsshould explore the idea that protective immunizationmay improve the clearance of extracellular deposits, asobserved in AD and atherosclerosis.

AD

N

M

MO

BLD

AMD

Figure 10 Electron micrographs from a patient with Alzheimer’s

disease (AD) and age-related macular degeneration (AMD)indicating similar appearance of scavenging of cellular debris

by immune cells. AD shows microglia (M) with cellular processes(asterisks) around extracellular debris (arrows) from a dying

neuronal cell (N) while figure AMD indicates digestion of basal

laminar deposits (BLD) by subretinal macrophage (MO). Abbrevi-ation: MO, subretinal macrophage. Source: From Refs. 52, 191.


Do Inflammatory Amplification Cascades Contributeto Injury or Progression?Ongoing research indicates that various cytokines andgrowth factors are crucial in the development of AMDcomplications. However, the contribution of macro-phages, mast cells or lymphocytes as potential sourcesfor these factors in AMD remains unexplored. Theidentification of terminal complement componentsC5–C9 (i.e., MAC) within drusen and near RPE isintriguing and suggests that complement-mediatedcell injury may play a role in AMD. However, inves-tigators must demonstrate intact MAC in associationwith endothelial or RPE cell membranes as well aslocal activation of these complement fragments.Further, a clear mechanism must be established tolink this injury stimulus to relevant cellular responsesinvolved in deposit formation.

The role of immune and nonimmune derivedoxidants as potential injury stimuli and amplifiers ofinjury responses was briefly described above andreviewed elsewhere. Evidence to demonstrate anage-related loss of protective antioxidants in AMDpatients is controversial, but is currently being eval-uated by several groups (216).

Can Anti-inflammatory Therapy Play a Rolein the Treatment of AMD?Recently, intravitreal corticosteroids were found to bepartially effective in improving vision and decreasingexudation due to CNV, suggesting the possibility thatanti-inflammatory therapy might be effective in AMDtreatment. Clinical medicine is on the verge of arevolution in anti-inflammatory therapy based ondrugs and other therapeutics developed from knowl-edge of the molecular basis of effector mechanismsand amplification systems described above. Perhapssome of these new approaches may be relevant to thetreatment of AMD.

One anti-inflammatory approach might be toblock the upregulation of amplification systemsdiscussed above. For instance, various complementinhibitors are in development, especially inhibitors ofC3 activation and the MAC formation (217). Thepotential role for vitronectin as an inhibitor of MACwas mentioned above (183). The role of specific anti-oxidant agents, rather than generic antioxidantcocktails, must also be better evaluated. Relativelyhigh doses of the lipophilic antioxidant vitamin E,which inserts into the plasma membrane to quenchcell membrane lipid peroxidation, has been shown todiminish complications of myocardial infarctionand stoke, in part by diminishing secondary inflam-mation-mediated oxidant injury (218). However,recent research suggests that dosing and bioavail-ability will be important issues for the eye. For

example, exogenous supplementation with solubleantioxidants, such as glutathione, may be inadequatebecause the compound is not taken up by RPE (218).Effective treatment may require the use of agents thatupregulate intracellular synthesis. Biosynthesis of PGsby immune or parenchymal cells also results in thegeneration of oxidants. Accordingly, the use ofnonsteroidal anti-inflammatory agents slow the pro-gression of other chronic neurodegenerative disorders,although they have not been evaluated in AMD (219).

Another anti-inflammatory approach is to blockmast cell or macrophage recruitment to the choroid,or inhibit their local activation. In this regard, blockadeof endothelial cell adhesion molecule expression toprevent the recruitment of macrophages or otherleukocytes to injured sites, is an active area of research(220). Pentoxyphylline has been shown to diminishmacrophage adhesiveness and cell activation inarthritis, suggesting a rationale for use in AMD (221).The mast cell inhibitor, tranilast, was observed to beeffective in experimental CNV (74). These approachesmight not only target macrophage-derived cytokines,like TNF-a, which can injure RPE or endothelium, butalso to RPE-derived cytokines, likeMCP-1 which serveto activate macrophages. Finally, should an infectiousetiology be determined, specific anti-infective agentsfor chlamydia or CMV might be considered.

SUMMARY POINTS

Biology of the Immune Response in AMD

& Innate immunity& Activation by retinal or choroidal injury

or infection& Antigen specific immunity

& Normal activation by foreign antigens& Aberrant activation in AMD by molecular

mimicry, antigen desequestration, neoantigenformation, or antigen trapping

& Amplification mechanisms& Complement, cytokines, oxidants, others

& Immune cells& Monocytes/macrophages, DC, mast cells,

lymphocytes& Innate immunity, antigen-specific immunity, and

amplification cascades contribute to pathogenesisof atherosclerosis, AD, and glomerular diseases

& Innate immunity, antigen-specific immunity, andamplification cascades may contribute to AMD

REFERENCES

1. Halliwell B, Gutteridge JMC. Free Radicals in Biology andMedicine. Oxford: Clarendon Press, 1999.


2. Oppenheim JJ, Feldman M. Cytokine Reference: ACompendium of Cytokines and Other Mediators of HostDefense. London: Academic Press, 2000.

3. Male D, Cooke A, Owen M, Trowsdale J, Champion B.Advanced Immunology. London: Mosby, 1996.

4. Gordon S. Macrophages and the Immune Response.Philadelphia, PA: Lippencott-Raven Publishers, 1999.

5. Moilanen W, Whittle B, Moncada S. Nitric oxide as afactor in inflammation. In: Gallin JI, Synderman R, eds.Inflammation: Basic Principles and Clinical Correlates. 3rded. Philadelphia, PA: Lippincott Williams & Wilkins,1999:787–800.

6. Hamrick TS, Havell EA, Horton JR, Orndorff PE. Host andbacterial factors involved in the innate ability of mousemacrophages to eliminate internalized unopsonizedEscherichia coli. Infect Immun 2000; 68:125–32.

7. Cooper NR. Biology of complement. In: Gallin JI,Synderman R, eds. Inflammation: Basic Principles andClinical Correlates. 3rd ed. Philadelphia, PA: LippincottWilliams & Wilkins, 1999:281–315.

8. Prodinger WM, Wurzner R, Erdei A, Dietrich MP.Complement. In: Paul WE, ed. Fundamental Immunology.4th ed. Philadelphia, PA: Lippencott-Raven Publishers,1999:967–95.

9. Gasque P, Dean YD, McGreal EP, VanBeek J, Morgan BP.Complement components of the innate immune system inhealth and disease in the CNS. Immunopharmacology2000; 49:171–86.

10. Gewurz H, Ying SC, Jiang H, Lint TF. Nonimmuneactivation of the classical complement pathway. BehringInst Mitt 1993; 93:138–47.

11. Preissner KT, Seiffert D. Role of vitronectin and its receptorsin haemostasis and vascular remodeling. Thromb Res 1998;89:1–21.

12. Hogasen K, Mollnes TE, Harboe M. Heparin-bindingproperties of vitronectin are linked to complex formationas illustrated by in vitro polymerization and bindingto the terminal complement complex. J Biol Chem 1992;267:23076–82.

13. Sorensen IJ, Nielsen EH, Andersen O, Danielsen B,Svehag SE. Binding of complement proteins C1q andC4bp to serum amyloid P component (SAP) in solidcontra liquid phase. Scand J Immunol 1996; 44:401–7.

14. Turner MW. Mannose-binding lectin: the pluripotentmolecule of the innate immune system. Immunol Today1996; 17:532–40.

15. Ogawa S, Clauss M, Kuwabara K, et al. Hypoxiainduces endothelial cell synthesis of membrane-associatedproteins. Proc Natl Acad Sci USA 1991; 88:9897–901.

16. Weinhouse GL, Belloni PN, Farber HW. Effect of hypoxiaon endothelial cell surface glycoprotein expression: modu-lation of glycoprotein IIIa and other specific surfaceglycoproteins. Exp Cell Res 1993; 208:465–78.

17. Collard CD, Lekowski R, Jordan JE, Agah A, Stahl GL.Complement activation following oxidative stress. MolImmunol 1999; 36:941–8.

18. Collard CD, Vakeva A, Morrissey MA, et al. Complementactivation after oxidative stress: role of the lectin comp-lement pathway. Am J Pathol 2000; 156:1549–56.

19. Zhou J, Jang YP, Kim SR, Sparrow JR. Complementactivation by photooxidation products of A2E, a lipofuscinconstituent of the retinal pigment epithelium. Proc NatlAcad Sci USA 2006; 103(44):6182–7.

20. Bardenstein DS, Cheyer C, Okada N, Morgan BP,Medof ME. Cell surface regulators of complement, 5I2antigen, and CD59, in the rat eye and adnexal tissues.Invest Ophthalmol Vis Sci 1999; 40:519–24.

21. Lass JH, Walter EI, Burris TE, et al. Expression of twomolecular forms of the complement decay-acceleratingfactor in the eye and lacrimal gland. Invest OphthalmolVis Sci 1990; 31:1136–48.

22. Mullins RF, Russell SR, Anderson DH, Hageman GS.Drusen associated with aging and age-related maculardegeneration contain proteins common to extracellulardeposits associated with atherosclerosis, elastosis, amyloi-dosis, and dense deposit disease. Faseb J 2000; 14:835–46.

23. Khodr B, Khalil Z. Modulation of inflammation by reactiveoxygen species: implications for aging and tissue repair.Free Radic Biol Med 2001; 30:1–8.

24. Leto TS. Respiratory burst oxidase. In: Gallin JI,Synderman R, eds. Inflammation: Basic Principles andClinical Correlates. 3rd ed. Philadelphia, PA: LippincottWilliams & Wilkins, 1999:769–86.

25. Heinecke JW. Mechanisms of oxidative damage by myelo-peroxidase in atherosclerosis and other inflammatorydisorders. J Lab Clin Med 1999; 133:321–5.

26. Knight JA. Free radicals: their history and current status inaging and disease. Ann Clin Lab Sci 1998; 28:331–46.

27. Winkler BS, Boulton ME, Gottsch JD, Sternberg P.Oxidative damage and age-related macular degeneration.Mol Vis 1999; 5:32.

28. Rozanowska M, Jarvis-Evans J, Korytowski W,Boulton ME, Burke JM, Sarna T. Blue light-inducedreactivity of retinal age pigment. In vitro generation ofoxygen-reactive species. J Biol Chem 1995; 270:18825–30.

29. Gottsch JD, Bynoe LA, Harlan JB, Rencs EV, Green WR.Light-induced deposits in Bruch’s membrane of proto-porphyric mice. Arch Ophthalmol 1993; 111:126–9.

30. Boulton M, Dontsov A, Jarvis-Evans J, Ostrovsky M,Svistunenko D. Lipofuscin is a photoinducible freeradical generator. J Photochem Photobiol B 1993; 19:201–4.

31. Wozniak W. Origin and the functional role of microglia.Folia Morphol (Warsz) 1998; 57:277–85.

32. Naito M, Umeda S, Yamamoto T, et al. Development,differentiation, and phenotypic heterogeneity of murinetissue macrophages. J Leukoc Biol 1996; 59:133–8.

33. Faust N, Huber MC, Sippel AE, Bonifer C. Differentmacrophage populations develop from embryonic/fetaland adult hematopoietic tissues. Exp Hematol 1997;25:432–44.

34. Jiang Y, Beller DI, Frendl G, Graves DT. Monocytechemoattractant protein-1 regulates adhesion moleculeexpression and cytokine production in human monocytes.J Immunol 1992; 148:2423–8.

35. Schumann RR, Latz E. Lipopolysaccharide-bindingprotein. Chem Immunol 2000; 74:42–60.

36. Schlegel RA, Krahling S, Callahan MK, Williamson P.CD14 is a component of multiple recognition systemsused by macrophages to phagocytose apoptotic lympho-cytes. Cell Death Differ 1999; 6:583–92.

37. Hammerstrom J. Human macrophage differentiationin vivo and in vitro. A comparison of human peritonealmacrophages and monocytes. Acta Pathol MicrobiolScand [C] 1979; 87C:113–20.

38. Takahashi K, Naito M, Takeya M. Development andheterogeneity of macrophages and their related cellsthrough their differentiation pathways. Pathol Int 1996;46:473–85.

39. Blackwell JM, Searle S. Genetic regulation of macrophageactivation: understanding the function of Nramp1(ZIty/Lsh/Bcg). Immunol Lett 1999; 65:73–80.

40. Rutherford MS, Witsell A, Schook LB. Mechanisms gener-ating functionally heterogeneous macrophages: chaosrevisited. J Leukoc Biol 1993; 53:602–18.


41. Everson MP, Chandler DB. Changes in distribution,morphology, and tumor necrosis factor-alpha secretionof alveolar macrophage subpopulations during thedevelopment of bleomycin-induced pulmonary fibrosis.Am J Pathol 1992; 140:503–12.

42. Chettibi S, Ferguson MJ. Wound repari: an overview. In:Gallin JI, Synderman R, eds. Inflammation: BasicPrinciples and Clinical Correlates. 3rd ed. Philadelphia,PA: Lippincott Williams & Wilkins, 1999:865–81.

43. Arenberg DA, Strieter RM. Angiogenesis. In: Gallin JI,Synderman R, eds. Inflammation: Basic Principles andClinical Correlates. 3rd ed. Philadelphia, PA: LippincottWilliams & Wilkins, 1999:851–64.

44. Postlewaite AE, KangAH. Fibroblasts andmatrix proteins.In: Gallin JI, Synderman R, eds. Inflammation: BasicPrinciples and Clinical Correlates. 3rd ed. Philadelphia,PA: Lippincott Williams & Wilkins, 1999:227–57.

45. Jackson JR, Seed MP, Kircher CH, Willoughby DA,Winkler JD. The codependence of angiogenesis andchronic inflammation. Faseb J 1997; 11:457–65.

46. Polverini PJ. How the extracellular matrix and macro-phages contribute to angiogenesis-dependent diseases.Eur J Cancer 1996; 32A:2430–7.

47. Laskin DL, Laskin JD. Macrophages, inflammatorymediators, and lung injury. Methods 1996; 10:61–70.

48. Hauser CJ. Regional macrophage activation after injuryand the compartmentalization of inflammation in trauma.New Horiz 1996; 4:235–51.

49. Raines EW, Ross R. Is overamplification of the normalmacrophage defensive role critical to lesion development?Ann N YAcad Sci 1997; 811:76–85 (discussion 85–77).

50. Penfold PL, Killingsworth MC, Sarks SH. Senile maculardegeneration: the involvement of immunocompetent cells.Graefes Arch Clin Exp Ophthalmol 1985; 223:69–76.

51. Penfold P, Killingsworth M, Sarks S. An ultrastructuralstudy of the role of leucocytes and fibroblasts in thebreakdown of Bruch’s membrane. Aust J Ophthalmol1984; 12:23–31.

52. Killingsworth MC, Sarks JP, Sarks SH. Macrophagesrelated to Bruch’s membrane in age-related maculardegeneration. Eye 1990; 4(Pt 4):613–21.

53. Lopez PF, Grossniklaus HE, Lambert HM, et al. Pathologicfeatures of surgically excised subretinal neovascularmembranes in age-related macular degeneration. AmJ Ophthalmol 1991; 112:647–56.

54. Oh H, Takagi H, Takagi C, et al. The potential angiogenicrole of macrophages in the formation of choroidal neovas-cular membranes. Invest Ophthalmol Vis Sci 1999;40:1891–8.

55. Cousins SW, Espinosa-Heidmann DG, Csaky KG. Mono-cyte activation in patients with age-related maculardegeneration: a biomarker of risk for choroidal neovascu-larization? Arch Ophthalmol 2004; 122:1013–8.

56. Steinman RM. Dendritic Cells. Philadelphia, PA: Lippen-cott-Raven Publishers, 1999.

57. McMenamin PG. The distribution of immune cells in theuveal tract of the normal eye. Eye 1997; 11(Pt 2):183–93.

58. Forrester JV, Liversidge J, Dick A, et al. What determinesthe site of inflammation in uveitis and chorioretinitis? Eye1997; 11(Pt 2):162–6.

59. Nilsson G, Costa JJ, Metcalfe DD.Mast cells and basophils.In: Gallin JI, Synderman R, eds. Inflammation: BasicPrinciples and Clinical Correlates. 3rd ed. Philadelphia,PA: Lippincott Williams & Wilkins, 1999:97.

60. Dines KC, Powell HC. Mast cell interactions with thenervous system: relationship to mechanisms of disease.J Neuropathol Exp Neurol 1997; 56:627–40.

61. Meininger CJ. Mast cells and tumor-associated angio-genesis. Chem Immunol 1995; 62:239–57.

62. Hagiwara K, Khaskhely NM, Uezato H, Nonaka S. Mastcell “densities” in vascular proliferations: a preliminarystudy of pyogenic granuloma, portwine stain, cavernoushemangioma, cherry angioma, Kaposi’s sarcoma, andmalignant hemangioendothelioma. J Dermatol 1999;26:577–86.

63. Costa JJ, Galli SJ. Mast cells and basophils. In: Rich R,Flesher TA, Schwartz BD, Shearer WT, Strober W, eds.Clinical Immunology: Principles and Practice. Vol. 1. St.Louis: Mosby, 1996:408–30.

64. Kovanen PT. Role of mast cells in atherosclerosis. ChemImmunol 1995; 62:132–70.

65. Ignatescu MC, Gharehbaghi-Schnell E, Hassan A, et al.Expression of the angiogenic protein, platelet-derivedendothelial cell growth factor, in coronary atheroscleroticplaques: in vivo correlation of lesional microvessel densityand constrictive vascular remodeling. ArteriosclerThromb Vasc Biol 1999; 19:2340–7.

66. Kaartinen M, van der Wal AC, van der Loos CM, et al.Mast cell infiltration in acute coronary syndromes: impli-cations for plaque rupture. J Am Coll Cardiol 1998;32:606–12.

67. Boesiger J, Tsai M, Maurer M, et al. Mast cells can secretevascular permeability factor/vascular endothelial cellgrowth factor and exhibit enhanced release after immu-noglobulin E-dependent upregulation of fc epsilonreceptor I expression. J Exp Med 1998; 188:1135–45.

68. Kanbe N, Tanaka A, Kanbe M, Itakura A, Kurosawa M,Matsuda H. Human mast cells produce matrix metallo-proteinase 9. Eur J Immunol 1999; 29:2645–9.

69. Johnson JL, Jackson CL, Angelini GD, George SJ.Activation of matrix-degrading metalloproteinases bymast cell proteases in atherosclerotic plaques. ArteriosclerThromb Vasc Biol 1998; 18:1707–15.

70. May CA. Mast cell heterogeneity in the human uvea.Histochem Cell Biol 1999; 112:381–6.

71. Tonnesen MG, Feng X, Clark RA. Angiogenesis in woundhealing. J Investig Dermatol Symp Proc 2000; 5:40–6.

72. Azizkhan RG, Azizkhan JC, Zetter BR, Folkman J. Mastcell heparin stimulates migration of capillary endothelialcells in vitro. J Exp Med 1980; 152:931–44.

73. Tharp MD. The interaction between mast cells and endo-thelial cells. J Invest Dermatol 1989; 93:107S–12.

74. Takehana Y, Kurokawa T, Kitamura T, et al. Suppression oflaser-induced choroidal neovascularization by oral trani-last in the rat. Invest Ophthalmol Vis Sci 1999; 40:459–66.

75. Janeway CA, Tavers P, Walport M. Immunobiology.London: Academic Press, 1999.

76. Roitt IM. Roitt’s Essential Immunology. Oxford: BlackwellScience Ltd, 1999.

77. Seder RA, Mosmann TM. Differentiation of effector pheno-types of CD4C and CD8C cells. Philadelphia, PA:Lippencott-Raven Publishers, 1999.

78. Benoist C, Mathis D. T-lymphocyte Differentiation andBiology. Philadelphia, PA: Lippencott-Raven Publishers,1999.

79. Dunn DE, Jin JP, Lancki DW, Fitch FW. An alternativepathway of induction of lymphokine production by Tlymphocyte clones. J Immunol 1989; 142:3847–56.

80. Lee KP, Harlan DM, June CH. Role of costimulation in thehost response to infection. In: Gallin JI, Synderman R, eds.Inflammation: Basic Principles and Clinical Correlates. 3rded. Philadelphia, PA: Lippincott Williams & Wilkins,1999:191–206.


81. Augustin AA, Julius MH, Cosenza H. Antigen-specificstimulation and trans-stimulation of T cells in long-termculture. Eur J Immunol 1979; 9:665–70.

82. Clark MR. IgG effector mechanisms. Chem Immunol 1997;65:88–110.

83. Dwyer JM. Immunoglobulins in autoimmunity: historyand mechanisms of action. Clin Exp Rheumatol 1996;14(Suppl. 15):S3–7.

84. Reichlin M. Cellular dysfunction induced by penetrationof autoantibodies into living cells: cellular damage anddysfunction mediated by antibodies to dsDNA and ribo-somal P proteins. J Autoimmun 1998; 11:557–61.

85. Shoenfeld Y, Alarcon-Segovia D, Buskila D, Abu-Shakra M,LorberM, Sherer Y, Berden J,Meroni PL, Valesini G,Koike T,et al. Frontiers of SLE: review of the 5th InternationalCongress of Systemic Lupus Erythematosus, Cancun,Mexico, April 20–25, 1998. Semin Arthritis Rheum 1999;29:112–30.

86. AdamusG,MachnickiM, ElerdingH, SugdenB, Blocker YS,Fox DA. Antibodies to recoverin induce apoptosis of photo-receptor and bipolar cells in vivo. J Autoimmun 1998;11:523–33.

87. Rosenberg HF, Gallin JI. Inflammation. Philadelphia, PA:Lippencott-Raven Publishers, 1999.

88. Descotes J, Choquet-Kastylevsky G, Van Ganse E. Vial Tresponses of the immune system to injury. Toxicol Pathol2000; 28:479–81.

89. Cotran RS, Kumar V, Collins T, Robbins SL. RobbinsPathologic Basis of Disease. Philadelphia, PA: WB Saun-ders Co., 1999.

90. Silverstein RL. The vascular endothelium. In: Gallin JI,Synderman R, eds. Inflammation: Basic Principles andClinical Correlates. 3rd ed. Philadelphia, PA: LippincottWilliams & Wilkins, 1999:207–25.

91. Mims CA, Nash A, Stephen J. Mims Pathogenesis ofInfectious Diseases. London: Academic Press, 2001.

92. Blaser MJ, Smith PD. Persistent mucosal colonizationby Helicobacter pylori and the induction of inflammation.In: Gallin JI, Synderman R, eds. Inflammation: BasicPrinciples and Clinical Correlates. 3rd ed. Philadelphia,PA: Lippincott Williams & Wilkins, 1999:1107–16.

93. Weiss A. T-lymphocyte Activation. Philadelphia, PA:Lippencott-Raven Publishers, 1999.

94. Gregerson DS. Immune privilege in the retina. OculImmunol Inflamm 1998; 6:257–67.

95. Cousins SW, Dix RD. Immunology of the eye. In:Keane RW, HIckey WF, eds. Immunology of the NervousSystem. New York: Oxford University Press, 1997:668–702.

96. Shevach EM. Organ-Specific Autoimmunity. Philadelphia,PA: Lippencott-Raven Publishers, 1999.

97. Singh VK, Kalra HK, Yamaki K, Abe T, Donoso LA,Shinohara T. Molecular mimicry between a uveitopatho-genic site of S-antigen and viral peptides. Inductionof experimental autoimmune uveitis in Lewis rats.J Immunol 1990; 144:1282–7.

98. Levine JS, Koh JS. The role of apoptosis in autoimmunity:immunogen, antigen, and accelerant. Semin Nephrol 1999;19:34–47.

99. Berden JH, van Bruggen MC. Nucleosomes and thepathogenesis of lupus nephritis. Kidney Blood Press Res1997; 20:198–200.

100. Gregerson DS, Torseth JW, McPherson SW, Roberts JP,Shinohara T, Zack DJ. Retinal expression of a neo-selfantigen, beta-galactosidase, is not tolerogenic and createsa target for autoimmune uveoretinitis. J Immunol 1999;163:1073–80.

101. Ross R. Atherogenesis. In: Gallin JI, Synderman R, eds.Inflammation: Basic Principles and Clinical Correlates. 3rded. Philadelphia, PA: Lippincott Williams &Wilkins, 1999.

102. Adler S, Couser W. Immunologic mechanisms of renaldisease. Am J Med Sci 1985; 289:55–60.

103. Dick AD. Immune mechanisms of uveitis: insights intodisease pathogenesis and treatment. Int Ophthalmol Clin2000; 40:1–18.

104. Smith RE. Commentary on histoplasmosis. Ocul ImmunolInflamm 1997; 5:69–70.

105. Klein R, Klein BE, Jensen SC. The relation of cardiovas-cular disease and its risk factors to the 5-year incidence ofage-related maculopathy: the Beaver Dam Eye Study.Ophthalmology 1997; 104:1804–12.

106. Hyman L, Schachat AP, He Q, Leske MC. Hypertension,cardiovascular disease, and age-related macular degener-ation. Age-Related Macular Degeneration Risk FactorsStudy Group. Arch Ophthalmol 2000; 118:351–8.

107. Vingerling JR, Dielemans I, Bots ML, et al. Age-relatedmacular degeneration is associated with atherosclerosis.The Rotterdam Study. Am J Epidemiol 1995; 142:404–9.

108. Ross R. Atherosclerosis—an inflammatory disease. N EnglJ Med 1999; 340:115–26.

109. Masuda J, Ross R. Atherogenesis during low levelhypercholesterolemia in the nonhuman primate. I. Fattystreak formation. Arteriosclerosis 1990; 10:164–77.

110. Curcio CA,Millican CL, Bailey T, Kruth HS. Accumulationof cholesterol with age in human Bruch’s membrane.Invest Ophthalmol Vis Sci 2001; 42:265–74.

111. Jeziorska M, Woolley DE. Local neovascularization andcellular composition within vulnerable regions of athero-sclerotic plaques of human carotid arteries. J Pathol 1999;188:189–96.

112. Green WR, Enger C. Age-related macular degenerationhistopathologic studies. The 1992 Lorenz E. ZimmermanLecture. Ophthalmology 1993; 100:1519–35.

113. Hariri RJ, Alonso DR, Hajjar DP, Coletti D, Weksler ME.Aging and arteriosclerosis. I. Development of myointimalhyperplasia after endothelial injury. J Exp Med 1986;164:1171–8.

114. Napoli C. Low density lipoprotein oxidation and ather-ogenesis: from experimental models to clinical studies. GItal Cardiol 1997; 27:1302–14.

115. Nagornev VA, Maltseva SV. The phenotype of macro-phages which are not transformed into foam cells inatherogenesis. Atherosclerosis 1996; 121:245–51.

116. Klaver CC, Kliffen M, van Duijn CM, et al. Geneticassociation of apolipoprotein E with age-related maculardegeneration. Am J Hum Genet 1998; 63:200–6.

117. Ross R,Masuda J, Raines EW, et al. Localization of PDGF-Bprotein in macrophages in all phases of atherogenesis.Science 1990; 248:1009–12.

118. Rajavashisth TB, Xu XP, Jovinge S, Meisel S, Xu XO,Chai NN, Fishbein MC, Kaul S, Cercek B, Sharifi B, et al.Membrane type 1 matrix metalloproteinase expression inhuman atherosclerotic plaques: evidence for activation byproinflammatory mediators. Circulation 1999; 99:3103–9.

119. George SJ. Tissue inhibitors of metalloproteinases andmetalloproteinases in atherosclerosis. Curr Opin Lipidol1998; 9:413–23.

120. Clinton SK, Underwood R, Hayes L, Sherman ML,Kufe DW, Libby P. Macrophage colony-stimulating factorgene expression in vascular cells and in experimental andhuman atherosclerosis. Am J Pathol 1992; 140:301–16.

121. de Boer OJ, van der Wal AC, Becker AE. Atherosclerosis,inflammation, and infection. J Pathol 2000; 190:237–43.


122. Saikku P. Epidemiologic association of Chlamydia pneumo-niae and atherosclerosis: the initial serologic observationand more. J Infect Dis 2000; 181(Suppl. 3):S411–3.

123. Taylor-Robinson D, Thomas BJ. Chlamydia pneumoniaein atherosclerotic tissue. J Infect Dis 2000; 181(Suppl.3):S437–40.

124. Meier CR. Antibiotics in the prevention and treatmentof coronary heart disease. J Infect Dis 2000; 181(Suppl.3):S558–62.

125. Leinonen M. Chlamydia pneumoniae and other risk factorsfor atherosclerosis. J Infect Dis 2000; 181(Suppl. 3):S414–6.

126. Kol A, Lichtman AH, Finberg RW, Libby P, Kurt-Jones EA.Cutting edge: heat shock protein (HSP) 60 activates theinnate immune response: CD14 is an essential receptor forHSP60 activation of mononuclear cells. J Immunol 2000;164:13–7.

127. Kol A, Bourcier T, Lichtman AH, Libby P. Chlamydial andhuman heat shock protein 60s activate human vascularendothelium, smooth muscle cells, and macrophages.J Clin Invest 1999; 103:571–7.

128. Tezel G, Wax MB. The mechanisms of HSP27 antibody-mediated apoptosis in retinal neuronal cells. J Neurosci2000; 20:3552–62.

129. Leinonen M, Saikku P. Infections and atherosclerosis.Scand Cardiovasc J 2000; 34:12–20.

130. High KP. Atherosclerosis and infection due to Chlamydiapneumoniae or cytomegalovirus: weighing the evidence.Clin Infect Dis 1999; 28:746–9.

131. Epstein SE, Zhou YF, Zhu J. Potential role of cytomegalo-virus in the pathogenesis of restenosis and atherosclerosis.Am Heart J 1999; 138:S476–8.

132. Silverman GJ, Shaw PX, Luo L, et al. Neo-self antigensand the expansion of B-1 cells: lessons from athero-sclerosis-prone mice. Curr Top Microbiol Immunol 2000;252:189–200.

133. Wick G, Perschinka H, Xu Q. Autoimmunity and athero-sclerosis. Am Heart J 1999; 138:S444–9.

134. Seifert PS, Hugo F, Hansson GK, Bhakdi S. Prelesionalcomplement activation in experimental atherosclerosis.Terminal C5b-9 complement deposition coincides withcholesterol accumulation in the aortic intima of hyperch-olesterolemic rabbits. Lab Invest 1989; 60:747–54.

135. Benzaquen LR, Nicholson-Weller A, Halperin JA. Terminalcomplement proteins C5b-9 release basic fibroblast growthfactor and platelet-derived growth factor from endothelialcells. J Exp Med 1994; 179:985–92.

136. Torzewski M, Torzewski J, Bowyer DE, et al. Immunohis-tochemical colocalization of the terminal complex ofhuman complement and smooth muscle cell alpha-actinin early atherosclerotic lesions. Arterioscler Thromb VascBiol 1997; 17:2448–52.

137. Hazen SL. Oxidation and atherosclerosis. Free Radic BiolMed 2000; 28:1683–4.

138. Brenner BM, Rector FC. Brenner and Rector’s the Kidney.6th ed. Philadelphia, PA: Saunders Co., 2000.

139. Makker SP. Mediators of immune glomerular injury. AmJ Nephrol 1993; 13:324–36.

140. Olson JL, Heptinstall RH. Nonimmunologic mechanismsof glomerular injury. Lab Invest 1988; 59:564–78.

141. JohnsonRJ. The glomerular response to injury: progressionor resolution? Kidney Int 1994; 45:1769–82.

142. Neuringer JR, Brenner BM. Glomerular hypertension:cause and consequence of renal injury. J HypertensSuppl 1992; 10:S91–7.

143. Suzuki D. Metalloproteinases in the pathogenesis ofdiabetic nephropathy. Nephron 1998; 80:125–33.

144. Anderson S, Vora JP. Current concepts of renalhemodynamics in diabetes. J Diabetes Complications1995; 9:304–7.

145. O’Bryan GT, Hostetter TH. The renal hemodynamic basisof diabetic nephropathy. Semin Nephrol 1997; 17:93–100.

146. Tisher CC, Hostetter TH. Diabetic nephropathy. In:Tisher CC, Brenner BM, eds. Renal Pathology with Clinicaland Functional Correlations. Philadelphia, PA: J.B. Lippin-cott, 1994:1309–34.

147. Sarks JP, Sarks SH, Killingsworth MC. Evolution ofgeographic atrophy of the retinal pigment epithelium.Eye 1988; 2:552–77.

148. Luft FC, Mervaala E, Muller DN, Gross V, Schmidt F,Park JK, Schmitz C, Lippoldt A, Breu V, Dechend R,et al. Hypertension-induced end-organ damage: a newtransgenic approach to an old problem. Hypertension1999; 33:212–8.

149. Peten EP, Garcia-Perez A, Terada Y, et al. Age-relatedchanges in alpha 1- and alpha 2-chain type IV collagenmRNAs in adult mouse glomeruli: competitive PCR. AmJ Physiol 1992; 263:F951–7.

150. Ungar A, Castellani S, Di Serio C, et al. Changes in renalautacoids and hemodynamics associated with aging andisolated systolic hypertension. Prostaglandins Other LipidMediat 2000; 62:117–33.

151. Yang N, Wu LL, Nikolic-Paterson DJ, et al. Local macro-phage andmyofibroblast proliferation in progressive renalinjury in the rat remnant kidney. Nephrol Dial Transplant1998; 13:1967–74.

152. Suto TS, Fine LG, Shimizu F, Kitamura M. In vivo transferof engineered macrophages into the glomerulus: endogen-ous TGF-beta-mediated defense against macrophage-induced glomerular cell activation. J Immunol 1997;159:2476–83.

153. Pawluczyk IZ, Harris KP. Macrophages promoteprosclerotic responses in cultured rat mesangial cells: amechanism for the initiation of glomerulosclerosis.J Am Soc Nephrol 1997; 8:1525–36.

154. Pawluczyk IZ, Harris KP. Cholesterol feeding activatesmacrophages to upregulate rat mesangial cell fibronectinproduction. Nephrol Dial Transplant 2000; 15:161–6.

155. D’SouzaMJ,OettingerCW, ShahA,TippingPG,HuangXR,Milton GV. Macrophage depletion by albumin microen-capsulated clodronate: attenuation of cytokine release inmacrophage-dependent glomerulonephritis. Drug DevInd Pharm 1999; 25:591–6.

156. Kamanna VS, Pai R, Ha H, Kirschenbaum MA, Roh DD.Oxidized low-density lipoprotein stimulates monocyteadhesion to glomerular endothelial cells. Kidney Int1999; 55:2192–202.

157. Hattori M, Nikolic-Paterson DJ, Miyazaki K, et al.Mechanisms of glomerular macrophage infiltration inlipid-induced renal injury. Kidney Int Suppl 1999;71:S47–50.

158. Duffield JS, Erwig LP, Wei X, Liew FY, Rees AJ, Savill JS.Activated macrophages direct apoptosis and suppressmitosis of mesangial cells. J Immunol 2000; 164:2110–9.

159. Kitamura M. Adoptive transfer of nuclear factor-kappaB-inactive macrophages to the glomerulus. Kidney Int 2000;57:709–16.

160. Lan HY, Yang N, Brown FG, et al. Macrophage migrationinhibitory factor expression in human renal allograftrejection. Transplantation 1998; 66:1465–71.

161. Ponticelli C. Progression of renal damage in chronicrejection. Kidney Int Suppl 2000; 75:S62–70.


162. BennettWM, Fassett RG,Walker RG, FairleyKF, d’ApiceAJ,Kincaid-Smith P. Mesangiocapillary glomerulonephritistype II (dense-deposit disease): clinical features of pro-gressive disease. Am J Kidney Dis 1989; 13:469–76.

163. Joh K, Aizawa S, Matsuyama N, et al. Morphologicvariations of dense deposit disease: light and electronmicroscopic, immunohistochemical and clinical findingsin 10 patients. Acta Pathol Jpn 1993; 43:552–65.

164. Nangaku M, Couser WG. Mechanisms of immune-depositformation and the mediation of immune renal injury. ClinExp Nephrol 2005; 9:183–91.

165. Nordstrand A, Norgren M, Holm SE. Pathogenicmechanism of acute post-streptococcal glomerulone-phritis. Scand J Infect Dis 1999; 31:523–37.

166. Roos A, Sato T, Maier H, van Kooten C, Daha MR.Induction of renal cell apoptosis by antibodies and comp-lement. Exp Nephrol 2001; 9:65–70.

167. al-NawabMD, Jones NF, Davies DR. Glomerular epithelialcell endocytosis of immune deposits in human lupusnephritis. Nephrol Dial Transplant 1991; 6:316–23.

168. Jansen JH, Hogasen K, Mollnes TE. Extensive complementactivation in hereditary porcine membranoproliferativeglomerulonephritis type II (porcine dense depositdisease). Am J Pathol 1993; 143:1356–65.

169. LeysA,VanrenterghemY,VanDammeB, SnyersB, PirsonY,Leys M. Fundus changes in membranoproliferativeglomerulonephritis type II. A fluorescein angiographicstudy of 23 patients. Graefes Arch Clin Exp Ophthalmol1991; 229:406–10.

170. Leys A, Michielsen B, Leys M, Vanrenterghem Y,Missotten L, Van Damme B. Subretinal neovascularmembranes associated with chronic membranoprolifera-tive glomerulonephritis type II. Graefes Arch Clin ExpOphthalmol 1990; 228:499–504.

171. Michielsen B, Leys A, Van Damme B, Missotten L. Funduschanges in chronic membranoproliferative glomerulone-phritis type II. Doc Ophthalmol 1990; 76:219–29.

172. Duvall-Young J, MacDonald MK, McKechnie NM. Funduschanges in (type II) mesangiocapillary glomerulonephritissimulating drusen: a histopathological report. BrJ Ophthalmol 1989; 73:297–302.

173. Baud L, Fouqueray B, Philippe C, Ardaillou R. Reactiveoxygen species as glomerular autacoids. J Am Soc Nephrol1992; 2:S132–8.

174. Grossniklaus HE, Cingle KA, Yoon YD, Ketkar N,L’Hernault N, Brown S. Correlation of histologic 2-dimen-sional reconstruction and confocal scanning lasermicroscopic imaging of choroidal neovascularization ineyes with age-related maculopathy. Arch Ophthalmol2000; 118:625–9.

175. Osusky R, Malik P, Aurora Y, Ryan SJ. Monocyte-macro-phage differentiation induced by coculture of retinalpigment epithelium cells with monocytes. OphthalmicRes 1997; 29:124–9.

176. Kurtz RM, Elner VM, Bian ZM, Strieter RM, Kunkel SL,Elner SG. Dexamethasone and cyclosporin A modulationof human retinal pigment epithelial cell monocyte chemo-tactic protein-1 and interleukin-8. Invest Ophthalmol VisSci 1997; 38:436–45.

177. Platts KE, Benson MT, Rennie IG, Sharrard RM, Rees RC.Cytokine modulation of adhesion molecule expression onhuman retinal pigment epithelial cells. Invest OphthalmolVis Sci 1995; 36:2262–9.

178. Nagineni CN, Kutty RK, Detrick B, Hooks JJ. Inflammatorycytokines induce intercellular adhesion molecule-1

(ICAM-1) mRNA synthesis and protein secretion byhuman retinal pigment epithelial cell cultures. Cytokine1996; 8:622–30.

179. Osusky R, Soriano D, Ye J, Ryan SJ. Cytokine effect onfibronectin release by retinal pigment epithelial cells. CurrEye Res 1994; 13:569–74.

180. Danis RP, Ciulla TA, Pratt LM, Anliker W. Intravitrealtriamcinolone acetonide in exudative age-related maculardegeneration. Retina 2000; 20:244–50.

181. Challa JK, Gillies MC, Penfold PL, Gyory JF, Hunyor AB,Billson FA. Exudative macular degeneration and intra-vitreal triamcinolone: 18 month follow up. Aust N ZJ Ophthalmol 1998; 26:277–81.

182. Johnson LV, Ozaki S, Staples MK, Erickson PA,Anderson DH. A potential role for immune complexpathogenesis in drusen formation. Exp Eye Res 2000;70:441–9.

183. Hageman GS, Mullins RF, Russell SR, Johnson LV,Anderson DH. Vitronectin is a constituent of oculardrusen and the vitronectin gene is expressed in humanretinal pigmented epithelial cells. Faseb J 1999; 13:477–84.

184. Anderson A, Taylor C, Azen S, et al. Immunopathology ofacute experimental histoplasmic choroiditis in theprimate. Invest Ophthalmol Vis Sci 1987; 28:1195–9.

185. Anderson A, Clifford W, Palvolgyi I, Rife L, Taylor C,Smith RE. Immunopathology of chronic experimentalhistoplasmic choroiditis in the primate. InvestOphthalmol Vis Sci 1992; 33:1637–41.

186. Palvolgyi I, Anderson A, Rife L, Taylor C, Smith RE.Immunopathology of reactivation of experimental ocularhistoplasmosis. Exp Eye Res 1993; 57:169–75.

187. Smith RE, Dunn S, Jester JV. Natural history of experi-mental histoplasmic choroiditis in the primate. II.Histopathologic features. Invest Ophthalmol Vis Sci1984; 25:810–9.

188. Jester JV, Smith RE. Subretinal neovascularization afterexperimental ocular histoplasmosis in a subhumanprimate. Am J Ophthalmol 1985; 100:252–8.

189. Smith RE. The uvea: questions of pathogenesis and treat-ment. Eye 1997; 11(Pt 2):145–7.

190. Rodriguez de Cordoba S, Esparza-Gordillo J, Goicoecheade Jorge E, et al. The human complement factor H:functional roles, genetic variations and disease associ-ations. Mol Immunol 2004; 41(4):355–67.

191. Hachimi KH, Foncin JF. Do microglial cells phagocyte thebeta/A4-amyloid senile plaque core of Alzheimer disease?C R Acad Sci III 1994; 317:445–51.

192. FrieseMA,Hellwage J, Jokiranta TS, et al. FHL-1/reconectinand factor H: two human complement regulators which areencoded by the same gene are differently expressed andregulated. Mol Immunol 1999; 36(13–14):809–18.

193. Mandal MN, Ayyagari R. Complement factor H: spatialand temporal expression and localization in the eye. InvestOphthalmol Vis Sci 2006; 47(9):4091–7.

194. Hageman GS, Anderson DH, Johnson LV, et al. A commonhaplotype in the complement regulatory gene factor H(HF1/CFH) predisposes individuals to age-relatedmacular degeneration. Proc Natl Acad Sci USA 2005;102(20):7227–32.

195. Conley YP, Jakobsdottir J, Mah T, et al. CFH, ELOVL4,PLEKHA1 and LOC387715 genes and susceptibility toage-related maculopathy: AREDS and CHS cohorts andmeta-analyses. Hum Mol Genet 2006; 15(21):3206–18.

196. Postel EA, Agarwal A, Caldwell J, et al. Complementfactor H increases risk for atrophic age-related maculardegeneration. Ophthalmology 2006; 113(9):1504–7.


197. Seddon JM, George S, Rosner B, Klein ML. CFH genevariant, Y402H, and smoking, body mass index, environ-mental associations with advanced age-related maculardegeneration. Hum Hered 2006; 61(3):157–65.

198. Sepp T, Khan JC, Thurlby DA, et al. Complement factor Hvariant Y402H is a major risk determinant for geographicatrophy and choroidal neovascularization in smokersand nonsmokers. Invest Ophthalmol Vis Sci 2006;47(2):536–40.

199. Klein RJ, Zeiss C, Chew EY, et al. Complement factor Hpolymorphism in age-related macular degeneration.Science 2005; 308(5720):385–9.

200. Haines JL, Hauser MA, Schmidt S, et al. Complementfactor H variant increases the risk of age-related maculardegeneration. Science 2005; 308(5720):419–21.

201. Edwards AO, Ritter R, III, Abel KJ, et al. Complementfactor H polymorphism and age-related macular degener-ation. Science 2005; 308(5720):421–4.

202. Shastry BS. Further support for the common variantsin complement factor H (Y402H) and LOC387715(A69S) genes as major risk factors for the exudative age-related macular degeneration. Ophthalmologica 2006;220(5):291–5.

203. Zipfel PF, Skerka C. Complement factor H and relatedproteins: an expanding family of complement-regulatoryproteins? Immunol Today 1994; 15(3):121–6.

204. Zipfel PF, Skerka C, Hellwage J, et al. Factor H familyproteins: on complement, microbes and human diseases.Biochem Soc Trans 2002; 30(Pt 6):971–8.

205. Kalayoglu MV, Bula D, Arroyo J, et al. Identification ofChlamydia pneumoniae within human choroidal neovas-cular membranes secondary to age-related maculardegeneration. Graefes Arch Clin Exp Ophthalmol 2005;243(11):1080–90.

206. Bora PS, Sohn JH, Cruz JM, et al. Role of complementand complement membrane attack complex in laser-induced choroidal neovascularization. J Immunol 2005;174(1):491–7.

207. Johnson PT, Betts KE, Radeke MJ, et al. Individualshomozygous for the age-related macular degenerationrisk-conferring variant of complement factor H haveelevated levels of CRP in the choroid. Proc Natl Acad SciUSA 2006; 103(46):17456–61.

208. Nozaki M, Raisler BJ, Sakurai E, et al. Drusen comp-lement components C3a and C5a promote choroidalneovascularization. Proc Natl Acad Sci USA 2006;103(7):2328–33.

209. Kain HL, Reuter U. Release of lysosomal protease fromretinal pigment epithelium andfibroblasts duringmechan-ical stresses. Graefes Arch Clin Exp Ophthalmol 1995;233:236–43.

210. Malorni W, Donelli G. Cell death. General featuresand morphological aspects. Ann N Y Acad Sci 1992;663:218–33.

211. MalorniW, Iosi F, Mirabelli F, Bellomo G. Cytoskeleton as atarget in menadione-induced oxidative stress in culturedmammalian cells: alterations underlying surface blebformation. Chem Biol Interact 1991; 80:217–36.

212. Adler R, Curcio C, Hicks D, Price D, Wong F. Cell death inage-related macular degeneration. Mol Vis 1999; 5:31.

213. Campochiaro PA, Soloway P, Ryan SJ, Miller JW.The pathogenesis of choroidal neovascularization inpatients with age-related macular degeneration. Mol Vis1999; 5:34.

214. Bilato C, Crow MT. Atherosclerosis and the vascularbiology of aging. Aging (Milano) 1996; 8:221–34.

215. Spagnoli LG, Sambuy Y, Palmieri G, Mauriello A. Age-related modulation of vascular smooth muscle cellsproliferation following arterial wall damage. Artery 1985;13:187–98.

216. Cai J, Nelson KC, Wu M, Sternberg P, Jr., Jones DP.Oxidative damage and protection of the RPE. Prog RetinEye Res 2000; 19:205–21.

217. Makrides SC. Therapeutic inhibition of the complementsystem. Pharmacol Rev 1998; 50:59–87.

218. Rimm EB, Stampfer MJ. Antioxidants for vascular disease.Med Clin North Am 2000; 84:239–49.

219. Hull M, Lieb K, Fiebich BL. Anti-inflammatory drugs: ahope for Alzheimer’s disease? Expert Opin Investig Drugs2000; 9:671–83.

220. Simon LS, Yocum D. New and future drug therapiesfor rheumatoid arthritis. Rheumatology (Oxford) 2000;39(Suppl. 1):36–42.

221. Graninger W, Wenisch C. Pentoxifylline in severe inflam-matory response syndrome. J Cardiovasc Pharmacol 1995;25(Suppl. 2):S134–8.


3

Genetics of Age-Related Macular DegenerationJennifer R. Chao and Amani A. FawziDoheny Eye Institute and Department of Ophthalmology, Keck School of Medicine,

University of Southern California, Los Angeles, California, U.S.A.

Jennifer I. LimUniversity of Illinois School of Medicine, Department of Ophthalmology, Eye and Ear Infirmary,

UIC Eye Center, Chicago, Illinois, U.S.A.

INTRODUCTION

Age-related macular degeneration (AMD), likeAlzheimer’s disease and atherosclerosis, is a late-onset degenerative disease. The multifactorial natureof these diseases has made the search for absolutegenetic contributions challenging. However, recentadvances in the study of genetic associations withAMD have provided evidence that there maybe strong genetic contributions to this disease.

The goal of recent genetic analysis in AMD is toidentify mutations and polymorphic variants thataffect the lifetime risk of developing the disease.The three main methods of finding genes contri-buting to AMD are candidate gene screening,linkage mapping, and case-association studies (1).Candidate gene screening involves evaluating genesresponsible for phenotypically similar diseases andgenes involved in pathways thought to contribute tothe pathophysiology of AMD. Linkage analysissearches for chromosomal regions that cosegregatewith the AMD disease trait by evaluating the segre-gation of chromosomal regions in families withAMD. Finally, case–control association studies findgenetic variants of genes that are associated withAMD by evaluating differences in frequencybetween those variants in persons with AMD andtheir matched controls. All of the above methodshave resulted in our current understanding of thegenetics of AMD.

EARLY SUSPICIONS: TWIN AND FAMILIALAGGREGATION STUDIES

The search for a genetic etiology of AMD was initiallysparked by twin studies that found familial clusteringof AMD cases. The first report of genetically testedmonozygotic twins affected by AMD showed highconcordance in both degree of disease severity andonset of vision loss (2). The concordance amongmonozygotic twins may have been explained bysimilarities in environmental factors; however, the

twin studies provided a strong rationale to pursuefurther study of the potential genetic etiology of AMD.Klein and coworkers reported on nine twin pairs(seven confirmed monozygotic) examined between1984 and 1993 (3). Of the nine twin pairs, whoseenvironmental factors such as diet, geographic back-ground, and medical history were similar, eight pairsdemonstrated similar fundus appearances and inci-dence of visual impairment. In 1995, Meyers reporteda statistically different concordance of AMD betweenmonozygotic twin pairs (100%, 25 of 25) and dizygotictwin pairs (42%, 5 of 12), further emphasizing theimportance of a genetic etiology (4). Two otherstudies confirmed this finding, demonstrating a signi-ficantly higher concordance of AMD in monozygoticversus dizygotic twin pairs (5,6). A Scandinaviaustudy demonstrated a significantly greater concor-dance of AMD (pZ0.0279) among monozygotic twinpairs (90%) as opposed to that of twin/spouse pairs(70.2%) (7).

Most recently, Seddon and colleagues evaluated840 elderly male twins (210 monozygotic and 181dizygotic), out of which 509 were diagnosed withmaculopathy and 106 had evidence of severe disease(8). They reported heritability estimates of 46% forthe overall five-step grade assignment (based on theClinical Age-Related Eye Disease Study), 0.67% forintermediate and advanced disease (grades 3–5), and0.71% for advanced disease only (grades 4 and 5). Ithas been suggested from these data that advanceddisease may have higher heritability. The heritabilityestimates of the Seddon report are similar to thatdescribed in an earlier twin study by Hammond andcolleagues of 45% (5). The latter study found thatthe most heritable phenotypes were soft drusenR125 mm (57%) and hard drusen R20 mm (81%),although it should be noted that none of the studyparticipants demonstrated lesions consistent withadvanced AMD.

Other groups have studied the concordance ratesamong persons with AMD and their non-twin

siblings, offspring, and spouses (9–14). Klaver et al.examined the first-degree relatives (siblings andoffspring) of 87 persons with late AMD and 135control subjects (9). They reported that the lifetimerisk estimate of late AMD for first-degree relatives ofpatients was 50% [95% confidence interval (CI)Z26–73], while that of non-affected controls was 12% (95%CIZ2.6–6.8). The risk for first-degree relatives wassignificantly higher among relatives of affected indi-viduals (p!0.001). These data confirmed the findingsreported by Seddon et al., who also found that theprevalence of AMD was significantly higher amongfirst-degree relatives (mostly siblings) of patients withexudative AMD when compared with those of unaf-fected controls (26.9% and 11.6% respectively).Together, these studies served to initiate and under-score the strong role of genetics and heritability in theetiology of AMD.

HEREDITARY RETINAL DYSTROPHIES:CANDIDATE GENES

Genes examined for their role in AMD have includedthose associated with phenotypically similar diseases.The genes responsible for monogenic hereditarymacular dystrophies such as Stargardt disease, Star-gardt-like macular dystrophy (STGD3), autosomaldominant macular dystrophy (adMD), Sorsby fundusdystrophy, Best macular dystrophy, Butterflydystrophy, and Doyne honeycomb retinal dystrophy(malattia leventinese) have been well characterized(15–20). Several of the genes implicated in thesediseases have been considered candidate genesfor AMD.

ABCRABCR (also ABCA4 or STGD1) is a gene that encodes aphotoreceptor-specific ATP-binding cassette trans-porter of retinaldehyde. ABCR is defective inautosomal recessive Stargardt disease, autosomalrecessive cone–rod dystrophy, and autosomal recessiveretinitis pigmentosa (16). Abnormal function of thetransporter, caused by mutations in the ABCR gene, ischaracterized by accumulation of a major lipofuscinfluorophore (A2-E) in the retinal pigment epithelium(RPE), making this gene attractive as a candidate genefor AMD (21). An early study identified heterozygousmutations of ABCR in 16% of AMD patients (22).Subsequently, two specific sequence changes in theABCR gene, G1961E and D2177N, were found topredict a three- and fivefold increased risk of AMDrespectively (23). Further research indicated that in acohort of families, the AMD-affected relatives of Star-gardt disease patients weremore likely to be carriers ofthe pathogenic Stargardt alleles (24). Sixteen specific

ABCR mutations were found to cause to functionalabnormalities of the transporter protein, includingATP-binding and ATPase activities (24). Additionally,it is believed that ABCR gene variants may be associ-atedwithAMDinat least six families (25,26).One studyof a group of unrelated multiplex cases of exudativeAMD reported finding six heterozygous missensechanges in the ABCR gene. Using familial segregationanalysis, Souied et al. were able to associate two of thecodon changes with familial AMD (25).

In contrast, other studies did not find anassociation of specific ABCR allelic variants to AMD(27–30). The allelic variants, G1961E and D2177N,from the initial report by Allikmets et al. were laterevaluated in individuals of Somali ancestry, and theallelic frequencies were not significantly differentbetween those with AMD and controls (31). Studiesevaluating other allelic variations of the ABCR genein participants of Japanese, Chinese, and Germanorigin have also reported no significant differencebetween allelic variants in participants (32–34).

The disparate findings encountered in thesestudies can be difficult to reconcile. A possibleconsideration is the unique prevalence of ABCR poly-morphisms in each study population. For example, themost common ABCR allele associated with Stargardtdisease in patients of European origin was found to bequite common in normal controls of Somali origin(31,35). Moreover, there is a large spectrum in allelicvariations of the ABCR gene in the populations as awhole, making the differentiation between disease-causingmutations and nonpathogenic polymorphismsdifficult.

ELOVL4A five-base pair deletion in the gene located onchromosome 6q14, ELOVL4, has been reported to beclosely associated with two forms of maculardystrophy, STGD3 and adMD, in two families (17).The clinical findings of STGD3 and adMD are similarto the atrophic form of AMD. The normal geneproduct, ELOVL4, is thought to be a retinal photo-receptor-specific protein that functions in thebiosynthesis of very long-chain fatty acids. A studyexamining ELOVL4 polymorphisms in unrelated indi-viduals with predominantly atrophic AMD revealedeight variants in the coding region; however, none ofthem were significantly associated with AMD suscep-tibility (36). Interestingly, a later case–control study ofpredominantly exudative AMD in familial casesobserved that a variant of the ELOVL4 gene previouslydescribed by Ayyagari et al., Met299Val, was signi-ficantly associated with AMD (37). The discrepancy inthese findings may be due to differences in the type ofAMD examined (atrophic versus exudative) or in the

36 CHAO ET AL.

sampling of study participants (sporadic versusfamilial).

Other Genes: VMD2, TIMP3, Peripherin/RDS,Fibulin 3/EFEMP1A variety of genes responsible for phenotypicallysimilar, monogenic macular dystrophies have hadless promising associations with AMD (15,19,38–48).Mutations in VMD2 (Best macular dystrophy), TIMP3(Sorsby fundus dystrophy), peripherin/RDS (butterflydystrophy), and Fibulin 3/EFEMP1 (Doyne honeycombretinal dystrophy or malattia leventinese) have beenstudied and were not been found to be significantlyassociated with AMD (38,39,42,43,45–48).

PATHOGENESIS OF AMD: CANDIDATE GENES

Multiple studies support the hypothesis that drusenare products of inflammatory responses to RPE injuryand are composed of proteins similar to deposits seenin diseases where inflammatory and oxidative damageplay a significant role (49,50). Analysis of the molecu-lar components of drusen has revealed evidence oflocalized inflammation and oxidative injury (49–54).Protein components found in drusen includecomplement factors, apolipoproteins B and E,immunoglobulins, MHC class II antigens, humanleukocyte antigen (HLA) DR, cholesterol esters, phos-pholipids, and carboxyethyl pyrrole protein adducts(53–56). Systemic inflammatory markers, such asC-reactive protein and interleukin-6, have beenshown to be independent risk factors for AMD andprogression of the disease (57,58). Drusen observed inthe disease, membranoproliferative glomerulone-phritis type II, believed to result from a complement-mediated immune system dysfunction, are immuno-histochemically similar to drusen found in AMD (59).Finally, there is a distinct similarity between proteinscontained in drusen in AMD and extracellulardeposits seen in atherosclerosis (49). Thus, multiplegenes derived from inflammatory and oxidativepathways, RPE basement membrane proteins, andextracellular deposits of atherosclerotic disease,amyloidosis, and Alzheimer’s disease have beenconsidered candidate genes in the pathogenesisof AMD.

Genes with Possible Association to AMDExtracellular Matrix: Fibulin, CST3, and MMP-9Due to the association of a Fibulin 3 mutation toheritable drusen and the significant role played inbasement membrane structure by the Fibulin familyof proteins, Fibulins 1–6 were evaluated for theirassociation with AMD (47,60,61). While allelic vari-ations in the Fibulin 1–4 genes could not conclusively

be associated with AMD (47), a missense mutation inFibulin 5 was noted to be present in 1.7% of partici-pants with AMD and absent in controls (47). Fibulin 5is thought to connect cellular surface receptors andextracellular elastic fibers, and thus play a key role inthe link between the RPE and Bruch’s membrane (60).

An allelic variation in exon 104 of Fibulin 6 (orHEMICENTIN-1) results in a non-conserved aminoacid substitution, Gln5345Arg in a large AMD familycohort, which segregates exclusively with thepresumptive disease haplotype (61). However,multiple subsequent studies have not been able toconfirm this finding (37,41,47,62–64). In a few studies,the allelic variant was not detected in any of theparticipants with AMD or in controls (41,62,64). TheGln5345Arg variant was found in 2 of 402 patientsand in 1 of 263 controls in a study by Stone et al., andthere was no significant association between theallelic variation and AMD (47). Additionally, in thestudy population where the Tyr402His variant ofcomplement factor H (CFH) was found to be signi-ficantly associated with AMD, the Gln5345Argvariant of HEMICENTIN-1 did not demonstrateallelic association to AMD in the discovery sample(63). Nevertheless, it is possible that the associationof this allelic variant of Fibulin 6 (HEMICENTIN-1)and AMD is unique to the family in which it wasoriginally reported, but other allelic variations inHEMICENTIN-1 have to be explored for significantassociations in a broader population of affectedindividuals (61).

Two other genes thought to play a role in thefunctioning of the RPE and extracellular matrix com-ponents are CST3 and MMP-9 (65,66). The CST3 geneencodes for cystatin C, a cysteine protease inhibitorthat regulates the activity of cathepsin S, a proteasewith regulatory functions in the RPE (67). One study ofGerman AMD patients revealed an increased suscep-tibility to the disease in individuals homozygous forthe recessive allele, CST B (66). The second gene,MMP-9, encodes the matrix metalloproteinase-9protein, and was found to have a polymorphic allelein its promoter region that was significantly associatedto exudative AMD in an Italian population (65).

Inflammation: CX3CR1, TLR4, and HLAGenes encoding inflammatory factors have beenstudied, including polymorphisms in the CX3CR1, toll-like receptor 4 (TLR4), and HLA genes (41,62,68–71) .Two single-nucleotide polymorphisms (SNPs), V249Iand T280M, in the CX3CR1gene, which encodes achemokine receptor expressed in the eye, were screenedand found to have a significantly higher prevalenceamong persons with AMD when compared withcontrols (70). Additional analysis of ocular tissue withevidence of advanced AMD revealed an even higher

3: GENETICS OF AGE-RELATED MACULAR DEGENERATION 37

prevalence of the T280M allele compared with thosewith a clinical AMD diagnosis.

TLR4was examined as a possible candidate genesince it is located in a chromosomal region with stronglinkage to AMD, 9q32–33 (41,62,69). Additionally, thegene product is thought to function as a key mediatorin pro-inflammatory signaling pathways, regulation ofcholesterol efflux, and the phagocytosis of photo-receptor outer segments by the RPE (71). Two allelicvariants were screened, D299G and T399I, in 667unrelated AMD patients and 439 controls. The studydemonstrated an increased risk of AMD in D299Gallele carriers. Interestingly, the authors examinedthe effects of the TLR4 allelic variant in combinationwith the ABCA1 R219K and the APOE-34 alleles, andthey reported a fourfold increased risk of AMD incarriers who exhibit the D299G TLR4 and R219KABCA1 alleles but not the APOE-34 allele (71). Thislatter finding supports a polygenic etiology of AMD.

Principal allele groups of HLA genes, includingHLA class I-A, -B, -Cw and Class II DRB1 and DQB1,were examined for their relationship to AMD (68).HLA antigens are expressed in eyes, and HLA DR hasbeen located immunohistochemically in both hardand soft drusen (49). The principal allele groupswere first screened in a cohort of 100 AMD casesand 92 controls. Alleles with p!0.1 on initial typingwere then screened in an additional 100 AMD casesand 100 controls. Logistic regression for all possiblepairwise HLA combinations was performed, alongwith Bonferroni corrections. The results werea positive correlation of allele Cw*0701 with AMD,whereas the B*4001 and DRB1*1301 alleles werenegatively associated (68).

Lipid Metabolism: APOE and PON1APOE encodes apolipoprotein E, a protein that plays acentral role in lipid transport and distribution in theperipheral and central nervous system (72). It has beenfound in soft drusen and basal laminar deposits,making it a good candidate gene for AMD (49,73).The gene has three alleles (32, 33, and 35), each codingdifferent protein isoforms, with the 33 allele beingmost common. Several case–control studies conductedin The Netherlands, Italy, France, United States,Australia, and Iceland have reported that the 34allele may confer a protective effect against AMD(74–79). However, there appeared to be no significantassociation of the 34 allele with AMD in other case–control studies conducted in Japan, Hong Kong, andthe United States (37,80–83). Separately, the APOEallele, 32, has been suggested to confer an increasedrisk of developing AMD (74,76,78); however, otherstudies have found no significant association(37,80,82,83). The disparate results of these studiesmay be a result of the variable baseline distribution

of the 32, 33, and 35 alleles in different ethnic popu-lations (84). Additionally, the studies differed greatlyin the severity and type of AMD examined, whichvaried from early stages of the disease to advancedatrophic or exudative AMD.

One recent study conducted in the United Statesexamined the combined effect of APOE genotypes andsmoking history (85). The study was based on thepremise of the authors’ earlier work that the 34 allelereduces the risk of AMDwhile the 32 allele increases it(78). The more recent analysis suggested that amongparticipants with exudative AMD (nZ260), smokingconferred the greatest risk in 32 allele carrierswith oddsratio (OR) of 1.9 for 34 carriers (pZ0.11), 2.2 for 33homozygotes (pZ0.007), and 4.6 for 32 carriers(pZ0.001) when compared with non-smoking 33homozygote controls (85). They conclude thatsmoking likely poses a greater risk factor in 32 allelecarriers compared with other APOE alleles’ carriers.

PON1 encodes paraoxonase, a calcium-depen-dent glycoprotein that prevents low-densitylipoprotein oxidation. It contains two polymorphicsites, Gly192Arg (A/B) and Leu54Met (L/M), whichgive rise to different protein products of varyingenzymatic activities. Ikeda et al. reported a higherfrequency of the BB and LL genotypes in participantswith exudative AMD when compared with controls(52.8% vs. 35% with pZ0.0127 and 91.7% vs. 77.1%with pZ0.009 respectively) in unrelated Japaneseparticipants (72 exudative AMD and 140 age- andgender-matched controls) (86). Later studies in thepopulations of Anglo-Celtic and Northern Irishdescent did not find a significant association ofallelic variation to either exudative or atrophic AMD(87,88). The association of the PON1 alleles andexudative AMD may therefore be population specific.

Other Genes: LPR6, VEGF, VLDLR, ACE, MnSOD,and EPHX1Several candidate genes have been studied in bothfamily-based and case–control cohorts. Low-densitylipoprotein receptor-related protein 6 and vascularendothelial growth factor showed linkage and allelicassociation in both family-based and case–control datasets (89). In the same study, the gene encodingvery low-density lipoprotein receptor (VLDLR) did not demon-strate significant linkage, but the family-based resultwas nominally significant and case–control resultswere significant (89). The ambiguous VLDLR associ-ation results echo those previously reported by Conleyet al., where VLDLRwas significant only for the allele-based test but not the linkage analysis (37).

Angiotensin-converting enzyme (ACE) wasthought to be a good candidate gene for neovascularAMD because an Alu polymorphism had been asso-ciated with proliferative diabetic retinopathy. Hamdi

38 CHAO ET AL.

et al. examined the association of the Alu polymor-phism in patients with neovascular/wet AMD (nZ86),atrophic AMD (nZ87), and age-matched controls (nZ189). Individuals carrying the Alu element insertion(AluC/C) in the genewere 4.5 timesmore frequent inthe control population than in the dry/atrophic AMDpatients (OR 5, pZ0.004), while the frequency did notdiffer significantly from the neovascular/wet AMDpopulation (OR 1.4, pZ0.4). The Alu polymorphismin the ACE gene was therefore believed to conferprotection against dry AMD (90). However, two latermultiple candidate gene studies did not find a signi-ficant association between the Alu polymorphism inthe ACE gene (DCP1) and either atrophic or neovas-cular AMD (37,89).

Other studies have sought to evaluate the role ofoxidative damage in AMD (88,91). The genetic poly-morphisms of four genes, cytochrome P-450 (CYP)1A1, glutathione-S-transferase (GPX1), microsomalepoxide hydrolase (EPHX1), and manganese super-oxide dismutase (MnSOD), were evaluated in 102Japanese participants with exudative AMD and in200 controls (91). The results suggested a strongassociation between a valine/alanine polymorphismof the MnSOD gene and exudative AMD. A weakerassociation to an exon-3 polymorphism of the EPHX1gene was also noted. In contrast, a subsequent candi-date gene analysis of patients with exudative AMD inat least one eye found no significant association withany of the genes evaluated in the earlier study byKimura et al., including MnSOD and multiple CYPgenes (including CYP1A1, CYP1A2, CYP2E1, andCYP2D6), EPHX1, and GPX1 (88).

Genes Not Associated with AMDIMPG2IMPG2 is a gene encoding the retinal interphotore-ceptor matrix proteoglycan IMP200, which is thoughtto be integral to the interaction of RPE and photo-receptors, specifically regulating the turnover ofphotoreceptor outer segments. Kuehn et al. screened92 individuals with AMD and 92 controls andreported three coding and one intronic polymorphismin IMPG2. However, none of the allelic variants werepresent at a significantly different frequency in theAMD versus control participants (92).

GPR75Rhodopsin is a G-protein coupled receptor, and whenit became known that the gene GPR75 encodedanother G-protein coupled receptor expressed inthe retina, it was thought to be a possible candidategene for AMD. However, in a screening of 535 AMDand 252 control cases, only six allelic variants werefound once in single AMD patients (93). These raremutations were deemed unlikely to be significantly

associated with AMD pathology in the majority ofaffected patients.

LAMC1, LAMC2, and LAMB3The LAMC1, LAMC2, and LAMB3 genes were selectedas positional and functional candidate genes. They arelocated in a region on chromosome 1q25–31 that hasbeen strongly linked to AMD (41,62,69,82,94–98). Thegenes code for laminins, which are extracellular matrixproteins located in the basal lamina of the RPE,Bruch’s membrane, and choriocapillaris (64). A totalof 69 sequence variants, 25 in coding regions, weredetected in the three laminin genes. However, nonewere found to be at a significantly higher frequency inthe AMD population when compared with thecontrols (64). In a separate study, polymorphisms inLAMC1 and LAMC2 were also not significantlydifferent between the affected individuals and controlcases (37).

Multicandidate Gene ScreeningSeveral large candidate genetic screening studies havesearched for genes with significant associations toAMD (37,88,89). Esfandiary and colleagues examinedgenes involved in the detoxification of reactive oxygenspecies, including CYP1A1, CYP1A2, CYP2E1,CYP2D6, EPHX1, MnSOD, AhR, NAT2, CAT, GPX1,PON1, and ADPRT1. Their study population wascomprised of 94 persons with exudative AMD and95 controls from Northern Ireland (88). The studyscreened a number of SNPs for 12 genes, but none ofthem revealed a significant association with AMD (88).Conley et al. examined a second category of genesinvolved in fatty acid biosynthesis and inflammatorypathways (37). They reported a significant associationfor allelic variants of CFH and ELOVL4 as describedearlier; however, no association was noted for othergenes, including GLRX2, OCLM, PRELP, RGS16,TGFb2, ApoH, and ITGB4. This study also did notfind an association with ACE and APOE, andthese genes are described elsewhere in detail.Finally, Haines and colleagues conducted a largescreening study of family-based and case–controldata sets, and evaluated several genes, of which a-2macroglobulin (A2M), creatine kinase (CKB), ACE(DCP1), interleukin-1a (IL1A), and microsomalglutathione-S-transferase 1 (MGST1) were found tohave no significant association with AMD (89).

LINKAGE MAPPING

Linkage mapping has provided a wealth of possiblegenetic associations to AMD. The multifactorial natureof the disease is reflected in the number and variety ofchromosomal associations detected. Genetic markerscovering several human chromosomes have been


tested for segregation in multiple combined subsets ofAMD families. Genetic loci purportedly linked toAMD include 1q25–31, 2q14.3, 2q31.2–2q32.3, 2p21,3p13, 4q32, 4p16, 5p, 5q34, 6q25.3, 6q14, 8, 9p24,9q31, 9q33, 10q26, 12q13, 12q23, 14q13, 15q21, 16p12,17q25, 18p11, 19p, 20q13, 22q, and X (41,62,69,95–100).Once a chromosomal region has been identified, finermapping narrows the search for possible candidategenes. Two chromosomal loci that are most consist-ently associated with AMD are discussed here.

One of the first disease loci mapped by linkageanalysis to AMD was a 9-cM region of chromosome1q25–31 (gene symbol, ARMD1) (101). The associationwas demonstrated in a family who demonstrated apredominantly dry phenotype of AMD. While thedisease segregated as an autosomal dominant trait inthis family, two individuals were identified as havingthe disease allele but not the phenotype, i.e., the allelewas non-penetrant. At this time, the Stargardt diseasegene (ABCR) was known to be located near this newdisease locus at chromosome 1p21, but linkageanalysis excluded it as an AMD disease locus in thisfamily. Since then, this region of chromosome 1 hasbeen the most commonly found site to segregate withAMD, both dry and exudative types, in genome-widescans involving large numbers (34–530) of families(41,62,69,82,94–98). Interestingly, one study of 70families did not demonstrate linkage to chromosome1q (102). Nevertheless, genes located in this regionwere viewed as possible candidate genes for AMD,including CFH, HEMICENTIN-1 (or Fibulin 6),LAMC1, LAMC2, and LAMB3 (37,61,64). These genesare discussed in detail elsewhere in this chapter.

Another locus consistently associated with AMDwas found during an early full genome-wide scan of225 families with both wet and dry forms of AMD,revealing a strong linkage to chromosome 10 (99).Further evidence from independent studies ofdifferent family cohorts narrowed the region tochromosome 10q26 (69,96), and follow-up studiesconfirmed this finding, including a recent meta-analysis of genome scans (41,94,98,102). Three genesin this locus have since been implicated in AMD,including PLEKA1, LOC387715, and PRSS11 (103).

RECENT ALLELE ASSOCIATION STUDIES

Complement Factor HThree research groups, each working from distinctcohorts, reported that a common allelic variant of theCFH gene was found at a significantly higherfrequency in affected individuals when comparedwith controls (63,104,105). The AMD participants, allAmericans of European origin, exhibited a range ofclinical findings, including extensive drusen,

geographic atrophy, and neovascular complications(63,105). The studies utilized SNPs and haplotypeblocks to test for associations among AMD cases,and they independently found a strong signal at aCFH SNP (rs1061170). Thus, they were able to map aspecific chromosomal location to the disease mani-fested in their study populations (106). The CFHpolymorphism is located in exon 9, and the allelicvariant results in the replacement of tyrosine withhistidine at amino acid 402 (Tyr402His). The tyrosineto histidine substitution is located within a region ofthe CFH protein (SCR7) that contains overlappingbinding sites for C-reactive protein, heparin, and Mprotein (107). This substitution is thought to alter thelevel of inflammation in the outer retina. An in-depthdiscussion of the role of CFH can be found inChapter 2.

The initial studies report that persons hetero-zygous (carrying a single copy) for the histidineallele in the Tyr402His polymorphism have a 2.45- to4.6-fold increased risk of AMD, while individualshomozygous for the histidine allele have a 5.57- to7.4-fold increased likelihood compared with thosewho do not carry the allele. The attributable risk ofAMD due to the histidine allele is estimated to beapproximately 50% in their study populations(63,104,105).

Further case–control association studies invol-ving individuals of European descent (American,Icelandic, and French) manifesting a wide clinicalrange of AMD, including drusen only, geographicatrophy, and/or neovascular, were subsequentlypublished. They independently confirmed thefinding that the Tyr402His variant was significantlyassociated with AMD (37,103,108–111). An incidencestudy of the Rotterdam population in The Nether-lands indicated that the presence of two histidinealleles (homozygous) increased the risk of developingAMD by 12.5 times, and that smoking, in combinationwith being homozygous for the allele, increased therisk 34-fold (smoking alone increased the risk by3.3 times) (112). The particularly high risk for AMDin smokers who are homozygous for the Tyr402Hisallele was confirmed by another study involvingparticipants from England (113). A prospective studyconfirmed the association between the Tyr402Hisvariant and an increased risk of AMD, reportingan estimated population-attributable risk for CFHTyr402His of 25% (114).

In contrast to studies involving persons ofEuropean descent, reports of AMD in persons ofother ethnicities describe a more tenuous associationwith the Tyr402His polymorphism. A case–controlstudy of Japanese individuals with exudative AMDreported that affected patients were at no greaterfrequency of having the histidine allele in the

40 CHAO ET AL.

Tyr402His polymorphism than unaffected individuals(115). Another recent study reported wideethnic variations in the frequencies of the Tyr402Hisallele in control populations: African Americans0.35G0.04, Caucasians 0.34G0.03, Somalis 0.34G0.03,Hispanics 0.17G0.03, and Japanese 0.07G0.02 (116).Because the frequency of the Tyr402His poly-morphism is not proportionate to the frequencyof AMD in their respective populations, it has beensuggested that the Tyr402His polymorphism maynot play as integral a role in AMD of some ethnicgroups as in those of European descent (116). Finally, astudy evaluating the Tyr402His polymorphism inLatinos suggests that the allele is not a major riskfactor for AMD in this population (117). However, itis important to note that in contrast to the original CFHstudies, the large majority of affected individuals inthe latter study demonstrated only early AMD. Theremay be some association between the Tyr402His alleleand the severity of AMD. Postel et al. recently reportedthat the polymorphism is associated with an increasedrisk of developing grades 3–5 AMD, but not grades 1and 2 (118). Together, these studies indicate that therelative importance of the CFH polymorphism inAMD is in part dictated by both the particular ethnicpopulation in question and the severity of AMDexhibited in the population.

Factor B and Complement Component 2Given the significant association of the CFH poly-morphism with AMD, Gold et al. screened forpolymorphisms in two other regulatory genes inthe same pathway, factor B (BF) and complementcomponent 2 (C2) (119). They report a statisticallysignificant common risk haplotype (H1) and twoprotective haplotypes, the L9H variant of BF andthe E318D variant of C2, as well as a variant in intron10 of C2 and the R32Q variant of BF. The latter twocombination of haplotypes confer a significantlyreduced risk of AMD, with an OR of 0.45 and 0.36respectively (119).

Chromosome 10q26: PLEKHA1, LOC387715,PRSS11, and HTRA1Jakobsdottir and colleagues had previously identifieda strong association of chromosome 10q26 and AMD,and they conducted a follow-up study in order toidentify candidate genes in that region (103). Threeoverlying genes, PLEKHA1, LOC387715, and PRSS11,and their respective non-synonymous SNPs wereidentified. Genotyping yielded a highly significantassociation between PLEKHA1/LOC387715 andAMD, with the SNPs in PLEKHA1 being more highlyassociated to AMD than those of LOC387715.

PLEKHA1 encodes the protein TAPP1, which is anactivator of lymphocytes, and PLEKHA1 transcriptsare expressed in the central macula. They report thatthe association of either a single or double copy of thehigh-risk allele in the PLEKHA1/LOC387715 locusaccounts for an OR of 5.0 and an attributable risk of57% in their study population (103). Additionally, thestudy notes a weaker association of the GRK5/RGS10locus with AMD. All of these associations wereindependent of the association of AMD with theTyr402His allele of CFH.

Another study reported a significant associationof the polymorphism Ala69Ser at LOC387715 in twocase–control cohorts of German descent (120). Thispolymorphism was associated with AMD, indepen-dent of the Tyr402His CFH polymorphism. In fact, thecontribution of the two genetic alleles were additive.A third study confirmed the role of the Ala69Serpolymorphism of the LOC287715 gene as anothermajor AMD-susceptibility allele (121). The adjustedpopulation-attributable risk percentage estimatesreported in their study were 36% for LOC387715 and43% for CFH, with a significantly higher risk of AMDwhen coupled with cigarette smoking.

Most recently, Yang and associates found that aSNP in the promoter region of the HTRA1 gene,rs11200638, conferred a population-attributable riskof 49.3% in a Caucasian cohort of persons with AMDin Utah (122). They demonstrated HRTA1 expressionin drusen from eyes of patients with AMD by labelingwith HTRA1 antibody. Additionally, they reportelevated expression of HRTA1 mRNA and protein inthe RPE and lymphocytes of AMD patients. HRTA1appears to regulate the degradation of extracellularmatrix proteoglycans and facilitates the access of otherdegrading matrix enzymes, such as matrix metallo-proteinases and collagenases. This study noted anallele dosage effect, where persons homozygous forthe allele have an increased risk [OR 7.29 (3.18, 16.74)]over those who are heterozygous [OR 1.83 (1.25, 2.68)].An estimated population-attributable risk from ajoint model with the CFH Tyr402His allele (i.e., a riskallele at either locus) is 71.4%. DeWan and colleaguesconcurrently reported an association of the identicalSNP from the HTRA1 promoter in a Chinese popu-lation with wet AMD, thus confirming the significantrole of this allelic variation in AMD populations ofvarious ethnicities (123).

CONCLUSION

The recent discovery of specific allelic variants ofmajor disease genes has come after many years ofsearching for the genetic etiology of AMD. The earlytwin and familial aggregation studies stronglysuggested that the disease was heritable, and this


was soon followed by linkage analyses implicatinglarge regions of chromosomes and candidate genescreenings describing possible culprit disease genes.Only recently, case-association studies, in conjunctionwith the completion of the human genome project,have enabled the identification of SNPs in majordisease genes, including CFH, BF/C2, PLEKHA1,LOC387715, and HTRA1.

Despite the discovery of major disease loci, ourunderstanding of the fundamental nature of AMDetiology is still lacking. Current evidence still suggeststhe disease to be multifactorial, shaped by multiplegenes as well as environmental influences. Forexample, several studies have demonstrated an addi-tive effect in the population risk of having more thanone allelic variant from a major disease locus.Additionally, there appears to be an increased risk ofdeveloping AMD when both the disease allele andcertain environmental factors, such as cigarettesmoking, are present. Moreover, the varying effectsof the major disease loci on AMD development acrossdifferent ethnic groups underscore the multifactorialnature of the disease.

Future progress in studying the etiology of AMDnot only includes the discovery of other major diseaseloci throughout the genome, but also the proteinproducts of identified allelic variants. SNPs thatresult in coding changes need to be studied for theireffect on protein function. An understanding of thisstep downstream from genetic coding is fundamentalto providing important confirmation of the signi-ficance of these genetic variations. The identificationof genetic allelic variants has opened a window intothe study of the pathophysiologic mechanisms ofAMD disease development and also diseaseintervention.

SUMMARY POINTS

& Twin studies and familial aggregation studiesprovided the earliest evidence for the role ofheritability in AMD.

& Candidate genes were derived from phenotypi-cally similar diseases and from genes involved inpathways thought to contribute to the pathophy-siology of AMD.

& Linkage mapping, which searches for chromo-somal regions that cosegregate with the AMDdisease trait, and multicandidate gene screeninghave implicated multiple genetic loci in almostevery chromosome, including 1q25–31, 2q14.3,2q31.2–2q32.3, 2p21, 3p13, 4q32, 4p16, 5p, 5q34,6q25.3, 6q14, 8, 9p24, 9q31, 9q33, 10q26, 12q13,12q23, 14q13, 15q21, 16p12, 17q25, 18p11, 19p,20q13, 22q, and X.

& Recent case-association studies have identifiedallelic variations of several genes thought to bemajor risk loci for AMD. These are CFH, BF/C2,PLEKHA1, LOC387715, and HTRA1.

& The etiology of AMD is multifactorial, involvingthe presence of multiple disease loci as well asenvironmental factors.

REFERENCES

1. Daiger SP. Genetics. Was the human genome project worththe effort? Science 2005; 308(5720):362–4.

2. Meyers SM, Zachary AA. Monozygotic twins with age-related macular degeneration. Arch Ophthalmol 1988;106(5):651–3.

3. Klein ML, Mauldin WM, Stoumbos VD. Heredity and age-related macular degeneration. Observations in monozy-gotic twins. Arch Ophthalmol 1994; 112(7):932–7.

4. Meyers SM, Greene T, Gutman FA. A twin study of age-related macular degeneration. Am J Ophthalmol 1995;120(6):757–66.

5. Hammond CJ, Webster AR, Snieder H, Bird AC,Gilbert CE, Spector TD. Genetic influence on early age-related maculopathy: a twin study. Ophthalmology 2002;109(4):730–6.

6. Grizzard SW, Arnett D, Haag SL. Twin study of age-related macular degeneration. Ophthalmic Epidemiol2003; 10(5):315–22.

7. Gottfredsdottir MS, Sverrisson T, Musch DC, Stefansson E.Age related macular degeneration in monozygotic twinsand their spouses in Iceland. Acta Ophthalmol Scand 1999;77(4):422–5.

8. Seddon JM, Cote J, Page WF, Aggen SH, Neale MC. TheU.S. twin study of age-related macular degeneration:relative roles of genetic and environmental influences.Arch Ophthalmol 2005; 123(3):321–7.

9. Klaver CC,Wolfs RC, Assink JJ, van Duijn CM, Hofman A,de Jong PT. Genetic risk of age-related maculopathy.Population-based familial aggregation study. ArchOphthalmol 1998; 116(12):1646–51.

10. Klein BE, Klein R, Lee KE, Moore EL, Danforth L. Risk ofincident age-related eye diseases in people with anaffected sibling: the Beaver Dam Eye Study. AmJ Epidemiol 2001; 154(3):207–11.

11. Seddon JM, Ajani UA, Mitchell BD. Familial aggregationof age-related maculopathy. Am J Ophthalmol 1997;123(2):199–206.

12. Silvestri G, Johnston PB, Hughes AE. Is genetic predis-position an important risk factor in age-related maculardegeneration? Eye 1994; 8(Pt 5):564–8.

13. Heiba IM, Elston RC, Klein BE, Klein R. Sibling corre-lations and segregation analysis of age-relatedmaculopathy: the Beaver Dam Eye Study. Genet Epide-miol 1994; 11(1):51–67.

14. Piguet B, Wells JA, Palmvang IB, Wormald R,Chisholm IH, Bird AC. Age-related Bruch’s membranechange: a clinical study of the relative role of heredity andenvironment. Br J Ophthalmol 1993; 77(7):400–3.

15. Weber BH, Vogt G, Pruett RC, Stohr H, Felbor U.Mutations in the tissue inhibitor of metalloproteinases-3(TIMP3) in patients with Sorsby’s fundus dystrophy. NatGenet 1994; 8(4):352–6.

16. Allikmets R. A photoreceptor cell-specific ATP-bindingtransporter gene (ABCR) is mutated in recessive Stargardtmacular dystrophy. Nat Genet 1997; 17(1):122.

42 CHAO ET AL.

17. Zhang K, Kniazeva M, Han M, et al. A 5-bp deletion inELOVL4 is associated with two related forms of autosomaldominant macular dystrophy. Nat Genet 2001; 27(1):89–93.

18. Petrukhin K, Koisti MJ, Bakall B, et al. Identification of thegene responsible for best macular dystrophy. Nat Genet1998; 19(3):241–7.

19. Stone EM, Lotery AJ, Munier FL, et al. A single EFEMP1mutation associated with both Malattia Leventinese andDoyne honeycomb retinal dystrophy. Nat Genet 1999;22(2):199–202.

20. Gregory CY, Evans K, Wijesuriya SD, et al. The generesponsible for autosomal dominant Doyne’s honeycombretinal dystrophy (DHRD) maps to chromosome 2p16.Hum Mol Genet 1996; 5(7):1055–9.

21. Weng J, Mata NL, Azarian SM, Tzekov RT, Birch DG,Travis GH. Insights into the function of rim protein inphotoreceptors and etiology of stargardt’s disease fromthe phenotype in abcr knockout mice. Cell 1999;98(1):13–23.

22. Allikmets R, Shroyer NF, Singh N, et al. Mutation of theStargardt disease gene (ABCR) in age-related maculardegeneration. Science 1997; 277(5333):1805–7.

23. Allikmets R. Further evidence for an association of ABCRalleles with age-related macular degeneration. The Inter-national ABCR Screening Consortium. Am J Hum Genet2000; 67(2):487–91.

24. Shroyer NF, Lewis RA, Yatsenko AN, Wensel TG,Lupski JR. Cosegregation and functional analysis ofmutant ABCR (ABCA4) alleles in families that manifestboth Stargardt disease and age-related macular degener-ation. Hum Mol Genet 2001; 10(23):2671–8.

25. Souied EH, Ducroq D, Rozet JM, et al. ABCR gene analysisin familial exudative age-related macular degeneration.Invest Ophthalmol Vis Sci 2000; 41(1):244–7.

26. Bernstein PS, LeppertM, SinghN, et al. Genotype-phenotypeanalysis ofABCR variants inmacular degeneration probandsand siblings. Invest Ophthalmol Vis Sci 2002; 43(2):466–73.

27. De La Paz MA, Guy VK, Abou-Donia S, et al. Analysis ofthe Stargardt disease gene (ABCR) in age-related maculardegeneration. Ophthalmology 1999; 106(8):1531–6.

28. Kuroiwa S, Kojima H, Kikuchi T, Yoshimura N. ATPbinding cassette transporter retina genotypes and agerelated macular degeneration: an analysis on exudativenon-familial Japanese patients. Br J Ophthalmol 1999;83(5):613–5.

29. Stone EM, Webster AR, Vandenburgh K, et al. Allelicvariation in ABCR associated with Stargardt disease butnot age-related macular degeneration. Nat Genet 1998;20(4):328–9.

30. Schmidt S, Postel EA, Agarwal A, et al. Detailed analysis ofallelic variation in the ABCA4 gene in age-related maculo-pathy. Invest Ophthalmol Vis Sci 2003; 44(7):2868–75.

31. Guymer RH, Heon E, Lotery AJ, et al. Variation of codons1961 and 2177 of the Stargardt disease gene is not associ-ated with age-related macular degeneration. ArchOphthalmol 2001; 119(5):745–51.

32. Baum L, Chan WM, Li WY, Lam DS, Wang PB, Pang CP.ABCA4 sequence variants in Chinese patients with age-related macular degeneration or Stargardt’s disease.Ophthalmologica 2003; 217(2):111–4.

33. Fuse N, Suzuki T, Wada Y, et al. Molecular genetic analysisof ABCR gene in Japanese dry form age-related maculardegeneration. Jpn J Ophthalmol 2000; 44(3):245–9.

34. Rivera A, White K, Stohr H, et al. A comprehensive surveyof sequence variation in the ABCA4 (ABCR) gene inStargardt disease and age-related macular degeneration.Am J Hum Genet 2000; 67(4):800–13.

35. Lewis RA, Shroyer NF, Singh N, et al. Genotype/Pheno-type analysis of a photoreceptor-specific ATP-bindingcassette transporter gene, ABCR, in Stargardt disease.Am J Hum Genet 1999; 64(2):422–34.

36. Ayyagari R, Zhang K, Hutchinson A, et al. Evaluation ofthe ELOVL4 gene in patients with age-related maculardegeneration. Ophthalmic Genet 2001; 22(4):233–9.

37. Conley YP, Thalamuthu A, Jakobsdottir J, et al. Candidategene analysis suggests a role for fatty acid biosynthesisand regulation of the complement system in the etiologyof age-related maculopathy. Hum Mol Genet 2005;14(14):1991–2002.

38. Akimoto A, Akimoto M, Kuroiwa S, Kikuchi T,Yoshimura N. Lack of association of mutations of thebestrophin gene with age-related macular degenerationin non-familial Japanese patients. Graefes Arch Clin ExpOphthalmol 2001; 239(1):66–8.

39. Allikmets R, Seddon JM, Bernstein PS, et al. Evaluation ofthe best disease gene in patients with age-related maculardegeneration and other maculopathies. Hum Genet 1999;104(6):449–53.

40. Guymer RH, McNeil R, Cain M, et al. Analysis of theArg345Trp disease-associated allele of the EFEMP1 genein individuals with early onset drusen or familial age-related macular degeneration. Clin ExperimentOphthalmol 2002; 30(6):419–23.

41. Iyengar SK, Song D, Klein BE, et al. Dissection of genome-wide-scan data in extended families reveals a major locusand oligogenic susceptibility for age-related maculardegeneration. Am J Hum Genet 2004; 74(1):20–39.

42. Kramer F, White K, Pauleikhoff D, et al. Mutations in theVMD2 gene are associated with juvenile-onset vitelliformmacular dystrophy (Best disease) and adult vitelliformmacular dystrophy but not age-related macular degener-ation. Eur J Hum Genet 2000; 8(4):286–92.

43. Lotery AJ, Munier FL, Fishman GA, et al. Allelic variationin the VMD2 gene in best disease and age-related maculardegeneration. Invest Ophthalmol Vis Sci 2000;41(6):1291–6.

44. Sauer CG, White K, Kellner U, et al. EFEMP1 is notassociated with sporadic early onset drusen. OphthalmicGenet 2001; 22(1):27–34.

45. Seddon JM, Afshari MA, Sharma S, et al. Assessment ofmutations in the Best macular dystrophy (VMD2) gene inpatients with adult-onset foveomacular vitelliformdystrophy, age-related maculopathy, and bull’s-eye macu-lopathy. Ophthalmology 2001; 108(11):2060–7.

46. ShastryBS, TreseMT.Evaluation of the peripherin/RDSgeneas a candidate gene in families with age-related maculardegeneration. Ophthalmologica 1999; 213(3):165–70.

47. Stone EM, Braun TA, Russell SR, et al. Missense variationsin the fibulin 5 gene and age-related macular degeneration.N Engl J Med 2004; 351(4):346–53.

48. Li Z, Clarke MP, Barker MD, McKie N. TIMP3mutation inSorsby’s fundus dystrophy: molecular insights. ExpertRev Mol Med 2005; 7(24):1–15.

49. Mullins RF, Russell SR, Anderson DH, Hageman GS.Drusen associated with aging and age-related maculardegeneration contain proteins common to extracellulardeposits associated with atherosclerosis, elastosis, amyloi-dosis, anddensedeposit disease. Faseb J 2000; 14(7):835–46.

50. Donoso LA, Kim D, Frost A, Callahan A, Hageman G. Therole of inflammation in the pathogenesis of age-relatedmacular degeneration. Surv Ophthalmol 2006; 51(2):137–52.

51. Killingsworth MC, Sarks JP, Sarks SH. Macrophagesrelated to Bruch’s membrane in age-related maculardegeneration. Eye 1990; 4(Pt 4):613–21.


52. Gurne DH, Tso MO, Edward DP, Ripps H. Antiretinalantibodies in serum of patients with age-related maculardegeneration. Ophthalmology 1991; 98(5):602–7.

53. Hageman GS, Luthert PJ, Victor Chong NH, Johnson LV,Anderson DH, Mullins RF. An integrated hypothesis thatconsiders drusen as biomarkers of immune-mediatedprocesses at the RPE-Bruch’s membrane interface inaging and age-related macular degeneration. Prog RetinEye Res 2001; 20(6):705–32.

54. Crabb JW, Miyagi M, Gu X, et al. Drusen proteomeanalysis: an approach to the etiology of age-relatedmacular degeneration. Proc Natl Acad Sci USA 2002;99(23):14682–7.

55. Zarbin MA. Current concepts in the pathogenesis of age-related macular degeneration. Arch Ophthalmol 2004;122(4):598–614.

56. Johnson LV, Ozaki S, Staples MK, Erickson PA,Anderson DH. A potential role for immune complexpathogenesis in drusen formation. Exp Eye Res 2000;70(4):441–9.

57. Seddon JM, Gensler G, Milton RC, Klein ML, Rifai N.Association between C-reactive protein and age-relatedmacular degeneration. JAMA 2004; 291(6):704–10.

58. Seddon JM, George S, Rosner B, Rifai N. Progression ofage-related macular degeneration: prospective assessmentof C-reactive protein, interleukin 6, and other cardiovas-cular biomarkers. Arch Ophthalmol 2005; 123(6):774–82.

59. Mullins RF, Aptsiauri N, Hageman GS. Structure andcomposition of drusen associated with glomerulone-phritis: implications for the role of complementactivation in drusen biogenesis. Eye 2001; 15(Pt 3):390–5.

60. Johnson LV, Anderson DH. Age-related macular degener-ation and the extracellular matrix. N Engl J Med 2004;351(4):320–2.

61. Schultz DW, Klein ML, Humpert AJ, et al. Analysis of theARMD1 locus: evidence that a mutation inHEMICENTIN-1 is associated with age-related macular degeneration in alarge family. Hum Mol Genet 2003; 12(24):3315–23.

62. Abecasis GR, Yashar BM, Zhao Y, et al. Age-relatedmacular degeneration: a high-resolution genome scan forsusceptibility loci in a population enriched for late-stagedisease. Am J Hum Genet 2004; 74(3):482–94.

63. Edwards AO, Ritter R, III, Abel KJ, Manning A,Panhuysen C, Farrer LA. Complement factor H poly-morphism and age-related macular degeneration.Science 2005; 308(5720):421–4.

64. Hayashi M,Merriam JE, Klaver CC, et al. Evaluation of theARMD1 locus on 1q25–31 in patients with age-relatedmaculopathy: genetic variation in laminin genes and inexon 104 of HEMICENTIN-1. Ophthalmic Genet 2004;25(2):111–9.

65. Fiotti N, Pedio M, Battaglia Parodi M, et al. MMP-9microsatellite polymorphism and susceptibility to exuda-tive form of age-related macular degeneration. Genet Med2005; 7(4):272–7.

66. Zurdel J, Finckh U, Menzer G, Nitsch RM, Richard G.CST3 genotype associated with exudative age relatedmacular degeneration. Br J Ophthalmol 2002; 86(2):214–9.

67. Rakoczy PE, Mann K, Cavaney DM, Robertson T,Papadimitreou J, Constable IJ. Detection and possiblefunctions of a cysteine protease involved in digestion ofrod outer segments by retinal pigment epithelial cells.Invest Ophthalmol Vis Sci 1994; 35(12):4100–8.

68. Goverdhan SV, Howell MW, Mullins RF, et al. AssociationofHLA class I and class II polymorphisms with age-relatedmacular degeneration. Invest Ophthalmol Vis Sci 2005;46(5):1726–34.

69. Majewski J, Schultz DW, Weleber RG, et al. Age-relatedmacular degeneration—a genome scan in extendedfamilies. Am J Hum Genet 2003; 73(3):540–50.

70. Tuo J, Smith BC, Bojanowski CM, et al. The involvement ofsequence variation and expression of CX3CR1 in thepathogenesis of age-related macular degeneration. FasebJ 2004; 18(11):1297–9.

71. Zareparsi S, Buraczynska M, Branham KE, et al. Toll-likereceptor 4 variant D299G is associated with susceptibilityto age-related macular degeneration. Hum Mol Genet2005; 14(11):1449–55.

72. Ignatius MJ, Gebicke-Harter PJ, Skene JH, et al. Expressionof apolipoprotein E during nerve degeneration and regen-eration. Proc Natl Acad Sci USA 1986; 83(4):1125–9.

73. Anderson DH, Ozaki S, Nealon M, et al. Local cellularsources of apolipoprotein E in the human retina andretinal pigmented epithelium: implications for theprocess of drusen formation. Am J Ophthalmol 2001;131(6):767–81.

74. Baird PN, Guida E, Chu DT, Vu HT, Guymer RH. Theepsilon2 and epsilon4 alleles of the apolipoprotein geneare associated with age-related macular degeneration.Invest Ophthalmol Vis Sci 2004; 45(5):1311–5.

75. Klaver CC, Kliffen M, van Duijn CM, et al. Geneticassociation of apolipoprotein E with age-related maculardegeneration. Am J Hum Genet 1998; 63(1):200–6.

76. Simonelli F, Margaglione M, Testa F, et al. ApolipoproteinE polymorphisms in age-related macular degeneration inan Italian population. Ophthalmic Res 2001; 33(6):325–8.

77. Souied EH, Benlian P, Amouyel P, et al. The epsilon4 alleleof the apolipoprotein E gene as a potential protectivefactor for exudative age-related macular degeneration.Am J Ophthalmol 1998; 125(3):353–9.

78. Schmidt S, Klaver C, Saunders A, et al. A pooled case-control study of the apolipoprotein E (APOE) gene in age-related maculopathy. Ophthalmic Genet 2002; 23(4):209–23.

79. Zareparsi S, Reddick AC, Branham KE, et al. Associationof apolipoprotein E alleles with susceptibility to age-related macular degeneration in a large cohort from asingle center. Invest Ophthalmol Vis Sci 2004;45(5):1306–10.

80. Gotoh N, Kuroiwa S, Kikuchi T, et al. Apolipoprotein Epolymorphisms in Japanese patients with polypoidal chor-oidal vasculopathy and exudative age-related maculardegeneration. Am J Ophthalmol 2004; 138(4):567–73.

81. Pang CP, Baum L, Chan WM, Lau TC, Poon PM, Lam DS.The apolipoprotein E epsilon4 allele is unlikely to be amajor risk factor of age-related macular degeneration inChinese. Ophthalmologica 2000; 214(4):289–91.

82. Schultz DW, Klein ML, Humpert A, et al. Lack of anassociation of apolipoprotein E gene polymorphismswith familial age-related macular degeneration. ArchOphthalmol 2003; 121(5):679–83.

83. Wong TY, Shankar A, Klein R, et al. Apolipoprotein E geneand early age-related maculopathy: the AtherosclerosisRisk in Communities Study. Ophthalmology 2006;113(2):255–9.

84. Corbo RM, Scacchi R. Apolipoprotein E (APOE) alleledistribution in the world. Is APOE*4 a ’thrifty’ allele?Ann Hum Genet 1999; 63(Pt 4):301–10.

85. Schmidt S, Haines JL, Postel EA, et al. Joint effects ofsmoking history and APOE genotypes in age-relatedmacular degeneration. Mol Vis 2005; 11:941–9.

86. Ikeda T, Obayashi H, Hasegawa G, et al. Paraoxonase genepolymorphisms and plasma oxidized low-density

44 CHAO ET AL.

lipoprotein level as possible risk factors for exudative age-related macular degeneration. Am J Ophthalmol 2001;132(2):191–5.

87. Baird PN, Chu D, Guida E, Vu HT, Guymer R. Associationof theM55L andQ192R paraoxonase gene polymorphismswith age-related macular degeneration. Am J Ophthalmol2004; 138(4):665–6.

88. Esfandiary H, Chakravarthy U, Patterson C, Young I,Hughes AE. Association study of detoxification genes inage related macular degeneration. Br J Ophthalmol 2005;89(4):470–4.

89. Haines JL, Schnetz-Boutaud N, Schmidt S, et al. Functionalcandidate genes in age-related macular degeneration:significant association with VEGF, VLDLR, and LRP6.Invest Ophthalmol Vis Sci 2006; 47(1):329–35.

90. Hamdi HK, Reznik J, Castellon R, et al. Alu D.N.A.polymorphism in ACE gene is protective for age-relatedmacular degeneration. Biochem Biophys Res Commun2002; 295(3):668–72.

91. Kimura K, Isashiki Y, Sonoda S, Kakiuchi-Matsumoto T,Ohba N. Genetic association of manganese superoxidedismutase with exudative age-related macular degener-ation. Am J Ophthalmol 2000; 130(6):769–73.

92. Kuehn MH, Stone EM, Hageman GS. Organization of thehuman IMPG2 gene and its evaluation as a candidate genein age-related macular degeneration and other retinaldegenerative disorders. Invest Ophthalmol Vis Sci 2001;42(13):3123–9.

93. Sauer CG, White K, Stohr H, et al. Evaluation of the Gprotein coupled receptor-75 (GPR75) in age relatedmacular degeneration. Br J Ophthalmol 2001;85(8):969–75.

94. Fisher SA, Abecasis GR, Yashar BM, et al. Meta-analysis ofgenome scans of age-related macular degeneration. HumMol Genet 2005; 14(15):2257–64.

95. Santangelo SL, Yen CH, Haddad S, Fagerness J, Huang C,Seddon JM. A discordant sib-pair linkage analysis of age-related macular degeneration. Ophthalmic Genet 2005;26(2):61–7.

96. Seddon JM, Santangelo SL, Book K, Chong S, Cote J. Agenomewide scan for age-related macular degenerationprovides evidence for linkage to several chromosomalregions. Am J Hum Genet 2003; 73(4):780–90.

97. Weeks DE, Conley YP, Tsai HJ, et al. Age-related maculo-pathy: an expanded genome-wide scan with evidence ofsusceptibility loci within the 1q31 and 17q25 regions. AmJ Ophthalmol 2001; 132(5):682–92.

98. Weeks DE, Conley YP, Tsai HJ, et al. Age-related maculo-pathy: a genomewide scan with continued evidence ofsusceptibility loci within the 1q31, 10q26, and 17q25regions. Am J Hum Genet 2004; 75(2):174–89.

99. Weeks DE, Conley YP, Mah TS, et al. A full genome scanfor age-related maculopathy. Hum Mol Genet 2000;9(9):1329–49.

100. Schmidt S, Scott WK, Postel EA, et al. Ordered subsetlinkage analysis supports a susceptibility locus for age-related macular degeneration on chromosome 16p12.BMC Genet 2004; 5:18.

101. Klein ML, Schultz DW, Edwards A, et al. Age-relatedmacular degeneration. Clinical features in a large familyand linkage to chromosome 1q. Arch Ophthalmol 1998;116(8):1082–8.

102. Kenealy SJ, Schmidt S, Agarwal A, et al. Linkage analysisfor age-related macular degeneration supports a gene onchromosome 10q26. Mol Vis 2004; 10:57–61.

103. Jakobsdottir J, Conley YP, Weeks DE, Mah TS, Ferrell RE,Gorin MB. Susceptibility genes for age-related maculo-pathy on chromosome 10q26. Am J Hum Genet 2005;77(3):389–407.

104. Haines JL, Hauser MA, Schmidt S, et al. Complementfactor H variant increases the risk of age-related maculardegeneration. Science 2005; 308(5720):419–21.

105. Klein RJ, Zeiss C, Chew EY, et al. Complement factor Hpolymorphism in age-related macular degeneration.Science 2005; 308(5720):385–9.

106. Clark AG. The role of haplotypes in candidate genestudies. Genet Epidemiol 2004; 27(4):321–33.

107. Giannakis E, Jokiranta TS, Male DA, et al. A common sitewithin factor H SCR 7 responsible for binding heparin,C-reactive protein and streptococcal M protein. EurJ Immunol 2003; 33(4):962–9.

108. Hageman GS, Anderson DH, Johnson LV, et al. A commonhaplotype in the complement regulatory gene factor H(HF1/CFH) predisposes individuals to age-relatedmacular degeneration. Proc Natl Acad Sci USA 2005;102(20):7227–32.

109. Magnusson KP, Duan S, Sigurdsson H, et al. CFH Y402Hconfers similar risk of soft drusen and both forms ofadvanced AMD. PLoS Med 2006; 3(1):e5.

110. Souied EH, Leveziel N, Richard F, et al. Y402H complementfactor H polymorphism associated with exudative age-related macular degeneration in the French population.Mol Vis 2005; 11:1135–40.

111. Zareparsi S, Branham KE, Li M, et al. Strong association ofthe Y402H variant in complement factor H at 1q32 withsusceptibility to age-related macular degeneration. AmJ Hum Genet 2005; 77(1):149–53.

112. Despriet DD, Klaver CC, Witteman JC, et al. Complementfactor H polymorphism, complement activators, and riskof age-related macular degeneration. JAMA 2006;296(3):301–9.

113. Sepp T, Khan JC, Thurlby DA, et al. Complement factor Hvariant Y402H is a major risk determinant for geographicatrophy and choroidal neovascularization in smokers andnonsmokers. Invest Ophthalmol Vis Sci 2006; 47(2):536–40.

114. Schaumberg DA, Christen WG, Kozlowski P, Miller DT,Ridker PM, Zee RY. A prospective assessment of the Y402Hvariant in complement factor H, genetic variants inC-reactive protein, and risk of age-relatedmacular degener-ation. Invest Ophthalmol Vis Sci 2006; 47(6):2336–40.

115. Gotoh N, Yamada R, Hiratani H, et al. No associationbetween complement factor H gene polymorphism andexudative age-related macular degeneration in Japanese.Hum Genet 2006; 120(1):139–43.

116. Grassi MA, Fingert JH, Scheetz TE, et al. Ethnic variation inAMD-associated complement factor H polymorphismp.Tyr402His. Hum Mutat 2006; 27(9):921–5.

117. Tedeschi-BlokN, Buckley J, Varma R, Triche TJ, Hinton DR.Population-based study of early age-related maculardegeneration: role of the complement factor H Y402Hpolymorphism in bilateral, but not unilateral disease.Ophthalmology 2007; 114(1):99–103.

118. Postel EA, Agarwal A, Caldwell J, et al. Complementfactor H increases risk for atrophic age-related maculardegeneration. Ophthalmology 2006; 113(9):1504–7.

119. Gold B, Merriam JE, Zernant J, et al. Variation in factor B(BF) and complement component 2 (C2) genes is associ-ated with age-related macular degeneration. Nat Genet2006; 38(4):458–62.

120. Rivera A, Fisher SA, Fritsche LG, et al. HypotheticalLOC387715 is a second major susceptibility gene for age-


related macular degeneration, contributing independentlyof complement factor H to disease risk. Hum Mol Genet2005; 14(21):3227–36.

121. Schmidt S, Hauser MA, Scott WK, et al. Cigarette smokingstrongly modifies the association of LOC387715 and age-related macular degeneration. Am J Hum Genet 2006;78(5):852–64.

122. Yang Z, Camp NJ, Sun H, et al. A variant of the HRTA1gene increases susceptibility to age-related maculardegeneration. Science 2006; 314(5801):992–3.

123. DeWan A, Liu M, Hartman S, et al. HTRA1 promoterpolymorphism in wet age-related macular degeneration.Science 2006; 314(5801):989–92.

46 CHAO ET AL.

4

Risk Factors for Age-Related Macular Degenerationand Choroidal NeovascularizationKah-Guan Au EongDepartment of Ophthalmology and Visual Sciences, Alexandra Hospital, Department of Ophthalmology,

Yong Loo Lin School of Medicine, National University of Singapore, The Eye Institute,

National Healthcare Group, Jurong Medical Center, Singapore Eye Research Institute, and Department

of Ophthalmology, Tan Tock Seng Hospital, Singapore

Bakthavatsalu MaheshwarDepartment of Ophthalmology and Visual Sciences, Alexandra Hospital and Jurong Medical Center,

Singapore

Stephen BeattyDepartment of Ophthalmology, Waterford Regional Hospital and Department of Chemical and Life

Sciences, Waterford Institute of Technology, Waterford, Ireland

Julia A. HallerThe Wilmer Ophthalmological Institute, Johns Hopkins University School of Medicine,

Johns Hopkins Hospital, Baltimore, Maryland, U.S.A.

INTRODUCTION

Age-related macular degeneration (AMD), the mostfrequent cause of blindness among individuals R55years in developed countries, is a major public healthproblem (1–5). Using estimated rates from a meta-analysis of recent population-based studies in theUnited States, Australia, and Europe, and the 2000U.S. census data, it has been estimated that theoverall prevalence of late AMD (neovascular AMDand/or geographic atrophy) in the U.S. populationR40 years is 1.47% [95% confidence interval (CI),1.38–1.55] (1). This translates to 1.75 million citizenshaving the most severe forms of the disease. Theprevalence of AMD increases dramatically with agesuch that in white women R80 years, more than 15%have neovascular AMD and/or geographic atrophy.More than 7 million individuals have drusenmeasuring 125 mm or larger and are, therefore, atsubstantial risk of developing late AMD. Owing tothe progressive increase in the life expectancy and theproportion of elderly persons in the population, it isestimated that the number of persons having lateAMD will increase by 50% to 2.95 million in 2020 (1).The increasing impact of AMD, coupled with thelimited therapy available for its treatment, has ledmany investigators to search for factors that could bemodified to prevent the onset or alter the naturalcourse and prognosis of AMD. The identification andmodification of risk factors has the potential for greater

public health impact on the morbidity from AMDthan the few treatment modalities at hand.

EPIDEMIOLOGIC STUDIES ON RISKFACTORS FOR AMD

Despite the high prevalence and public health import-ance of AMD, its pathogenesis remains unknown. Thetypes of epidemiologic studies that have exploredAMD risk factors are case–control, cross-sectional,and prospective cohort studies. Case–control studies[e.g., the Eye Disease Case–Control Study (6,7)] havecompared the frequency of possible risk factors amongindividuals with AMD to a cohort of control patientswithout the disease. Cross-sectional studies [e.g., theFramingham Eye Study (3) and the National Healthand Nutrition Examination Survey I (NHANES-I) (8)]have correlated eye examination data with sociodemo-graphic, medical, and other variables collected as partof larger studies. Prospective cohort studies [e.g., thePhysicians’ Health Study (9)] collect data in a group ofsubjects over time. Tables 1–3 show some case–control,cross-sectional, and prospective cohort studies thathave explored risk factors for AMD.

PROBLEMS AND LIMITATIONS OF EPIDEMIOLOGICSTUDIES ON RISK FACTORS FOR AMD

There may be different causative factors thatdamage the macula and result in common clinical

(Text continues on page 51.)

Table1

SomeCase–ControlStudiesthatHaveInvestigatedtheRiskFactorsofAMD

Author(s)(year)

Place/nameofstudy

Design

Study

population

Methodofdiagnosis

Riskfactors

studied

Maltzmanetal.(1979)(10)

JerseyCity,NewJersey

Case–control

30cases

Clinicalexamination

Variouspersonaland

environmentalfactors

30controls

DelaneyandOates(11)

Syracuse,NewYork

Case–control

50cases

Clinicalexaminationandfundus

photography

Variouspersonaland


50controls

Hymanetal.(1983)(12)

Baltimore,Maryland

Case–control

162cases

Fundusphotography

Variouspersonaland


175controls

Weiteretal.(1985)(13)

Boston,MassachusettsandFort

Myers,Florida

Case–control

650cases

Fundusphotography

Iriscolorandfundus

pigmentation

363controls

Blumenkranzetal.(1986)(14)

Miami,Florida

Case–control

26cases

Fundusphotography

Variouspersonaland


23controls

Tsangetal.(1992)(15)

Sydney,Australia

Case–control

80cases

Fundusphotography

Variouspersonaland


86controls

EyeDiseaseCase–ControlStudy

Group(1992,1993)(6,7),Seddon

etal.(1994)(16)

Baltimore,Maryland;Boston,

Massachusetts;Chicago,Illinois;

Milwaukee,Wisconsin;andNew

York,NewYork/EyeDisease

Case–ControlStudy

Case–control

421cases

Fundusphotography

Variouspersonaland


615controls

Holzetal.(1994)(17)

London,England

Case–control

101cases

Clinicalexaminationandfundus

photographyforcases

Iriscolor

102controls

Clinicalexaminationforcontrols

Mares-Perlmanetal.(1995)(18)

BeaverDam,Wisconsin/Beaver

DamEyeStudy

Nestedcase–controlwithina

population-basedcohort

167cases

Fundusphotography

Antioxidants

167controls

Vingerlingetal.(1995)(19)

Rotterdam,

TheNetherlands/

RotterdamStudy



59femalecases

Fundusphotography

Reproductiveandrelated

factors

295femalecontrols

Darzinsetal.(1997)(20)

Newcastle,Australia

Case–control

409cases

Fundusphotography

Sunlightexposure

286controls

Tamakoshietal.(1997)(21)

Kantodistrict,Japan

Case–control

52malecases

Fundusphotographyforcases

Cigarettesmoking

82malecontrols

Clinicalexaminationforcontrols

Chaineetal.(1998)(22)

France/FRANCE-DMLAStudy

Case–control

1844cases

Fundusphotography

Variouspersonaland


1844controls

Beldaetal.(1998)(23)

Valencia,Spain

Case–control

25cases

15controls

Clinicalexamination

SerumvitaminEandzinc,

sunlightexposure

Hymanetal.(2000)(24)

NewYork,NewYork/Age-related

MacularDegenerationRisk

FactorsStudy

Case–control

182neovascularAMD

cases

Fundusphotography

Systemichypertension,

cardiovasculardisease,

andcholesterolintake

227nonneovascular

AMDcases

235controls

Kalayogluetal.(2003)(25)

PaloAlto,California

Case–control

25cases

Clinicalexamination

Chlamydiapneumoniae

infection

18controls

Seddonetal.(26)

Boston,Massachusettsand

Portland,Oregon

Case–control

747caseswithvarying

degreesofAMD

Fundusphotography

C-reactiveprotein

183controls

Khanetal.(2006)(27)

CountiesofNorfolk,Suffolk,

Cambridgeshire,and

Buckinghamshire,United

Kingdom

Case–control

435cases

Fundusphotography

Smoking

280controls

McGwinetal.(2006)(28)

FourcountiesinNorthCarolina,

California,Maryland,and

Pennsylvania/Cardiovascular

HealthStudy



390cases

Fundusphotography

Useofcholesterol-lowering

medications

2365controls

Abbreviation:AMD,age-relatedmaculardegeneration.

48 AU EONG ET AL.

Table2

SomeCross-SectionalStudiesthatHaveInvestigatedtheRiskFactorsofAge-RelatedMacularDegeneration

Author(s)(year)

Place/nameofstudy

Design

Studypopulationa

Methodofdiagnosis

Riskfactors

studied

Kahnetal.(1977)(29,30),

Leibowitzetal.(1980)(3),

Sperdutoetal.(1980,

1981,1986)(31–33)

Framingham,

Massachusetts/

FraminghamEyeStudy

Population-basedcross-

sectional

2631survivorsofthe

FraminghamHeartStudy

cohort,meanageZ65.3

years

Clinicalexamination

Variouspersonaland


Martinezetal.(1982)(34)

Gisborne,NewZealand


sectional

481participantsagedR65

years

Clinicalexamination

Ageandsex

KleinandKlein(1982)(35),

Goldbergetal.(1988)(8),

Liuetal.(1989)(36),

Obisesanetal.(1998)(37)

UnitedStates/National

Health

andNutritional

ExaminationSurveyI


sectional


years

Clinicalexamination

Variouspersonaland


Gibsonetal.(1986)(38)

MeltonMowbray,

England/MeltonMowbray

EyeStudy


sectional


years

Fundusphotography

Variouspersonaland


Westetal.(1989)(39),

Bressleretal.(1989)(40),

Tayloretal.(1990,1992)

(41,42)

SomersetCounty,Maryland

andlowerDorchester

County,Maryland/

Chesapeake

Bay

WatermenStudy

Occupationalcross-sectional

782malewatermenaged

R30years

Fundusphotography

Ageandsunlightexposure

Vinding(1989,1990,1992)

(43–45)

Copenhagen,Denmark


sectional

924survivorsfromthe

CopenhagenCityHeart

Studyaged60–79years

Fundusphotography

Variouspersonaland


Westetal.(1994)(46)

Baltimore,Marylandand

Washington,DC/Baltimore

LongitudinalStudyof

Aging

Cross-sectional

916participantsofthe

BaltimoreLongitudinal

StudyofAgingagedR40

years

Fundusphotography

Antioxidants

Kleinetal.(1992,1993,

1994)(47–52),

Cruickshanks

etal.(1993,

2001)(53,54),Heibaetal.

(1994)(55),Mares-

Perlmanetal.(1995)(56)

BeaverDam,Wisconsin/

BeaverDamEye

Study


sectional

4926participantsaged

43–84years

Fundusphotography

Variouspersonaland


Schachatetal.(1995)(57)

Barbados,WestIndies/

BarbadosEyeStudy


sectional

3444participants

aged40–84years

Fundusphotography

Variouspersonaland


Vingerlingetal.(1995)

(58,59),Ikrametal.(2005)

(60)

Rotterdam,The

Netherlands/Rotterdam

Study


sectional

6251participantsaged

55–98years

Fundusphotography

Variouspersonaland


(Continued)

4: RISK FACTORS FOR AMD AND CHOROIDAL NEOVASCULARIZATION 49

Table2

SomeCross-SectionalStudiesthatHaveInvestigatedtheRiskFactorsofAge-RelatedMacularDegeneration(Continued)

Author(s)(year)

Place/nameofstudy

Design

Studypopulationa

Methodofdiagnosis

Riskfactors

studied

Mitchelletal.(1995,1996,

1998,1999,2002)(61–65),

Atteboetal.(1996)(4),

Smith

etal.(1997,1998,

1999,2000)(66–70),

Wangetal.(1998,1999)

(71,72)

BlueMountainsregion,

Sydney,Australia/Blue

MountainsEyeStudy


sectional


years

Fundusphotography

Variouspersonaland


Hirvelaetal.(1996)(73)

OuluCounty,Northern

Finland


sectional


years

Fundusphotography

Variouspersonaland


Delcourtetal.(1998,1999)

(74–76),Defayetal.

(2004)(77)

Sete,France/Pathologies

OculairesLieesal’Age

Study


sectional


years

Fundusphotography

Variouspersonaland


Friedmanetal.(1999)(78)

EastBaltimore,Maryland/

BaltimoreEyeStudy


sectional


years

Fundusphotography

Ageandrace

Kleinetal.(1999)(79)

ForsythCounty,North

Carolina;cityofJackson,

Mississippi;Minneapolis,

Minnesota;and

WashingtonCounty,

Maryland/Atherosclerosis

RiskinCommunitiesStudy


sectional

11,532participantsaged

48–72years

Fundusphotography

Variouspersonaland


Kleinetal.(1995,1999)

(80,81)

UnitedStates/National

Health

andNutritional

ExaminationSurveyIII

Complex,multistagearea

probabilitysampledesign

(certaingroups,e.g.,

AmericansR60years,

MexicanAmericans,and

non-Hispanicblackswere

sampledatahigher

probabilitythanother

persons)


years

Fundusphotography

Varioussociodemographic,

ocular,medical,and


McCartyetal.(2001)(82)

Victoria,Australia/Visual

ImpairmentProject


sectional


years

Fundusphotography

Variouspersonaland


Miyazakietal.(83)

FukuokaCity,Kyushu,

Japan/HisayamaStudy


sectional


years

Fundusphotography

Variouspersonaland


Fraser-Belletal.(2005,

2006)(84,85)

LosAngeles,California/Los

AngelesLatinoEye

Study


sectional


years

Fundusphotography

Variouspersonaland


Krishnaiahetal.(2005)(86)

StateofAndhraPradesh,

India/AndhraPradeshEye

DiseaseStudy


sectional

3723participantsaged40–

102years

Fundusphotography

Variouspersonaland


aThesamplesizemayvaryslightlyamongthedifferentreports.

50 AU EONG ET AL.

manifestations that we recognize as AMD. Theanalysis of risk factors for AMD is inherently difficultbecause many of them are closely interrelated, e.g.,race, ocular pigmentation, and sunlight exposure, orsocioeconomic status, smoking, and nutrition. Study-ing risk factors such as sunlight exposure includechallenges in measurement of acute and chronic life-time exposure and the effect of potential confoundingfactors such as sun sensitivity and sun-avoidancebehavior. In addition, the difficulties in establishing acausal link between a chronic disease and a potentialrisk factor are magnified for a condition such asAMD because it manifests itself late in life. Additionalproblems in this circumstance include a long lead time,a possible recall bias, and survivor cohort effects.

Despite the extensive past and ongoing researchon AMD worldwide, there is currently no universallyaccepted definition of AMD. Different definitions ofearly and late signs of AMD have been used in variousstudies, making direct comparison of the resultsdifficult or impossible (Table 4) (99). The problem isfurther compounded by differences in methodologyused in the various studies. A wide range of differentdiagnostic tools has been used in different clinical andepidemiologic studies (99). For example, NHANES-I,a population-based study of a sample of the nonin-stitutionalized U.S. population, relied solely on clinicalexaminations by multiple independent examinerswith varying levels of experience, and standardizationof the diagnosis of AMD was uncertain (8,35). Fundusphotographs of a subset of the study population werereviewed and discrepancies in the macular gradings

were disclosed (100). The Framingham Eye Study,which has provided the most frequently cited preva-lence data on AMD to date, was based mainly onclinical examination and fundus photography wasperformed only on a small subset of the study popu-lation (29). More recent studies have used fundusphotography to detect and grade AMD but thedetails were not always standardized among thestudies (40,47,57,61,74).

In an effort to standardize disease definitionand study methodology, the International Age-relatedMaculopathy Epidemiological Study Group publishedin 1995 an international classification and gradingsystem for AMD in the hope of producing a commondetection and classification system for epidemiologicstudies (99). It defined age-related maculopathy (ARM)to include two alternate late lesions (neovascularmaculo-pathy and geographic atrophy), termed AMD or lateARM, and early lesions (soft or large drusen andretinal pigmentary abnormalities), termed early ARM(Table 5). In this definition, visual acuity is not acriterion for the presence or absence of ARM. Thisnew terminology, however, has not been universallyaccepted. In this chapter, we will use the more conven-tional definition of AMD to include the entire spectrumof the disease (i.e., equivalent toARM in the new termi-nology). Neovascular AMD and geographic atrophywill be collectively termed late AMD (equivalent tolate ARM) and the early lesions of AMD will betermed early AMD (equivalent to early ARM).

It is possible that the factors associated with earlyAMD may be different from those associated with

Table 3 Some Prospective Cohort Studies that Have Investigated the Risk Factors of Age-Related Macular Degeneration

Author(s) (year)Place/nameof study Design

Studypopulationa

Method ofdiagnosis

Risk factorsstudied

Moss et al. (1996) (87),

Klein et al. (1997,

1998) (88–90)

Beaver Dam,

Wisconsin/Beaver

Dam Eye Study

Population-based

prospective

cohort

3684 participants

aged 43–86 years

Fundus photography Alcohol consumption,

cardiovascular

disease risk factors

Seddon et al. (91) 11 states in the United

States/Nurses’

Health Study

Prospective

cohort

31,843 female

registered nurses

agedR50 years

Diagnosis by treating

ophthalmologists

or optometrists

Cigarette smoking

Christen et al. (1996,

1999) (9,92), Ajani

et al. (1999) (93),

Schaumberg et al.

(2001) (94)

United States/

Physicians’ Health

Study

Prospective

cohort

21,157 male

physicians aged

40–84 years at

baseline


ophthalmologists

or optometrists

Cigarette smoking,

antioxidant vitamin

supplements,

alcohol consumption

Cho et al. (2000, 2001,

2004) (95–97)

United States/Nurses’

Health Study and

Health Professionals

Follow-Up Study

Prospective

cohort

32,764 female

registered nurses

and 29,488 male

health professionals

agedR50 years


ophthalmologists

or optometrists

Intake of alcohol, zinc,

fruits, vegetables,

vitamins, and

carotenoids

Klein et al. (2006) (98) Six communities in

United States/Multi-

Ethnic Study of

Atherosclerosis

Prospective

cohort

6166 participants

aged 45–85 years

Fundus photography Race/ethnicity

a The sample size may vary slightly among the different reports.


progression toneovascularAMDorgeographic atrophy.In addition, althoughneovascularAMDandgeographicatrophy are termed collectively as late AMD (or lateARM), they may have different causes (99). For these

reasons, it may be important to pay attention to thedifferent stages of AMD and to separate the two mani-festations of late AMD in epidemiologic studies, as hasbeen done in several recent studies (24,66).

Table 4 Definitions of and Age Limits in AMD (ARM) Used in Population-Based Studies

1. Framingham Eye Study (3)

An eye was diagnosed as having senile macular degeneration if its visual acuity was 20/30 or worse and the ophthalmologist designated the

etiology of changes in the macula or posterior pole as senile

Age limits: 52–85 years

2. National Health and Nutrition Eye Study I (8)

Age-related macular degeneration: Loss of macular reflex, pigment dispersion and clumping, and drusen associated with visual acuity

of 20/25 or worse believed to be due to this disease

Age-related disciformmacular degeneration: Choroidal hemorrhage and connective-tissue proliferation betweenRPE andBruch’smembrane

causing an elevation of the foveal retina (this condition should be differentiated from disciform degenerations of other causes, e.g.,

Histoplasmosis, toxoplasmosis, angioid streaks, and high myopia)

Age-related circinate macular degeneration: Perimacular accumulation of lipoid material within the retina


3. Gisborne Study (34)

Senile macular degeneration. When the visual acuity in the affected eye was 6/9 (20/30) or worse and senile macular degeneration was

identified as the probable cause of this visual loss

Age limits: R65 years

4. Copenhagen Study (43)

AMD. Best-corrected visual (Snellen) acuity (including pinhole improvement) of 6/9 or less, explained by age-related morphologic changes of

the macula

Atrophic (dry) changes. Disarrangement of the pigment epithelium (atrophy/clustering) and/or a small cluster of small

drusen and/or medium drusen and/or large drusen and/or pronounced senile macular choroidal atrophy/sclerosis without general fundus

involvement

Exudative (wet) changes. Elevation of the neurosensory retina and/or the pigment epithelium and/or hemorrhages, and/or hard exudates

and/or fibrovascular tissue

Age-related macular changes without visual impairment (AMCW) is defined as similar morphological lesions but without visual deterioration


5. Chesapeake Bay Study (40)

No specific overall definition

Geographic atrophy. An area of well-demarcated atrophy of the RPE in which the overlying retina appeared thin

Exudative changes. Choroidal neovascularization, detachments of the RPE, and disciform scarring

Grading of AMD in four grades:

Grade 4: Geographic atrophy of the RPE or exudative changes

Grade 3: Grade 4 or eyes with large or confluent drusen or eyes with focal hyperpigmentation of the RPE

Grade 2: Grade 4 or 3 or eyes with many small drusen (R20) within 1500 mm of the foveal center

Grade 1: Grade 4, 3, or 2 or eyes with at least five small drusen within 1500 mmof the foveal center or at least 10 small drusen between 1500

and 3000 mm from the foveal center

No visual acuity included


6. Beaver Dam Eye Study (47)

Early ARMwas defined as the absence of signs of late ARMas defined below and as the presence of soft indistinct or reticular drusen or by the

presence of any drusen type except hard indistinct, with RPE degeneration or increased retinal pigment in the macular area. Late ARMwas

defined as the presence of signs of exudative AMD or geographic atrophy

The grade assigned for the participant was that of the more severely involved eye



7. Rotterdam Study (58)

All ARMchangeshad to bewithin a radius of 3000 mmof the foveola. No definition of early ARM, but separate prevalence figures for drusen and

RPE hyperpigmentations or hypopigmentations attributable to age-related causes

Late ARM (is similar to AMD). The presence of atrophic AMD (well-demarcated area of RPE atrophy with visible choroidal vessels) and/or

neovascular AMD (serous and/or hemorrhagic RPE detachment, and/or subretinal neovascular membrane and/or hemorrhage, and/or

periretinal fibrous scar) attributable to age-related causes. In a participant, the most severely involved eye was taken for the analysis



Abbreviations: AMD, age-related macular degeneration; ARM, age-related maculopathy; RPE, retinal pigment epithelial.Source: From Ref. 99.

52 AU EONG ET AL.

Some studies have evaluated huge numbers ofvariables for possible associations with ocular find-ings. For example, the Framingham Eye Studycorrelated its ophthalmic diagnoses with almost allof 667 variables from the Framingham Heart Study(29). Because of the very large number of variablesevaluated, it is possible that some of the associationsfound may be due to chance alone (101). Similarly,while it is plausible that risk factors may be differentfor the various manifestations of AMD [e.g., drusen,increased retinal pigment, retinal pigment epithelial(RPE) depigmentation, geographic atrophy, andneovascular AMD], simultaneously conductingmultiple comparisons within individual studiesincreases the likelihood of chance findings (102). Infact, one in 20 variables should have a positiveassociation (for pZ0.05) by chance alone (103), andthis probably contributes partly to the inconsistentresults between studies. To provide compellingevidence of a real association between AMD andpotential risk factors, repeated findings of the samerisk factors in well-designed studies conducted indifferent populations are necessary.

While results from epidemiologic studies mayidentify risk factors for AMD, proof that modifyinga particular established risk factor can influence thecourse of the disease can emerge only from random-ized prospective clinical trials.

RISK FACTORS OF AMD

A number of risk factors for AMD have been incrimi-nated from various epidemiologic studies, suggestingthat the condition is multifactorial in etiology (Table 6).These risk factors may be broadly classified intopersonal or environmental factors, and the personalfactors may be further subdivided into sociodemo-graphic, ocular, and systemic factors.

SOCIODEMOGRAPHIC FACTORS

AgeAge is the strongest risk factor associated with AMD.The prevalence, incidence, and progression of all formsofAMD rise steeplywith advancing age (47,88). There isa consistent finding across multiple population-basedstudies of an increase in prevalence of late AMD withage, from near absence at age 50 years to about 2%prevalence at age 70, and about 6% at age80 (47,80,86,104). In the Framingham Eye Study, theprevalence of any AMD (defined as degenerativechanges of the macula with visual acuity of 20/30 orworse) was 1.6% for persons 52 to 64 years, 11.0% forpersons 65 to 74 years, and 27.9% for persons 75 to 85years (30). Although closely linked to the aging process,AMD is not universal and is not inevitable withincreasing age.

Table 5 Definitions of ARM

ARM is a disorder of the macular area of the retina, most often clinically apparent after 50 years of age, characterized by any of the following

primary items, without indication that they are secondary to another disorder (e.g., ocular trauma, retinal detachment, high myopia,

chorioretinal infective or inflammatory process, choroidal dystrophy, etc.):

† Drusen that are discrete whitish-yellow spots external to the neuroretina or the RPE. They may be soft and confluent, often with indistinct

borders

Soft distinct drusen have uniform density with sharp edges

Soft indistinct drusen have decreasing density from center outwards with fuzzy edges

Hard drusen, usually present in eyes with as well as those without ARM, do not of themselves characterize the disorder

† Areas of increased pigment or hyperpigmentation (in the outer retina or choroid) associated with drusen

† Areas of depigmentation or hypopigmentation of the RPE, most often more sharply demarcated than drusen, without any visibility of

choroidal vessels associated with drusen

† Late stages of ARM, also called age-related macular degeneration

AMD is a later stage of ARM and includes both “dry” and “wet” AMD

Dry AMD is also called geographic atrophy and is characterized by:

† Any sharply delineated roughly round or oval area of hypopigmentation or depigmentation or apparent absence of the RPE in which

choroidal vessels are more visible than in surrounding areas that must be at least 175 mm in diameter on the color slide (using a 308

or 358 camera)

Wet AMD is also called “neovascular” AMD, “disciform” AMD, or “exudative” AMD and is characterized by any of the following:

† RPE detachment(s), which may be associated with neurosensory retinal detachment, associated with other forms of ARM

† Neovascular membrane(s) that may be subretinal or sub-RPE

† Scar/glial tissue or fibrin-like deposits that may be epiretinal (with exclusion of idiopathic macular puckers), intraretinal, subretinal, or sub-

pigment epithelial

† Subretinal hemorrhages that may be nearly black, bright red, or whitish-yellow and that are not related to other retinal vascular disease.

Hemorrhages in the retina (retinal hemorrhages) or breaking through it into the vitreous (vitreous hemorrhages) may also be present

Hard exudates (lipids) within the macular area related to any of the above and not to other retinal vascular disease

Abbreviations: ARM, age-related maculopathy; RPE, retinal pigment epithelial.Source: From Ref. 99.


GenderGender has not been consistently found to be a riskfactor for AMD. Sex was not associated with AMD in astudy in Gisborne, New Zealand (34), the NHANES-I(8), the Copenhagen Study (43), the Rotterdam Study(58), a Finnish population-based study (73), and theAndhra Pradesh Eye Study in South India (86).Frequency estimates for drusen and the high-riskfeatures of AMD among the black participants inthe Barbados Eye Study were similar for men andwomen (57).

In the Blue Mountains Eye Study, there wasconsistent, although not statistically significant, sexdifferences in prevalence for most lesions of AMD,with women having higher rates for late AMD and softindistinct drusen than men, but not retinal pigmentaryabnormalities, which were slightly more frequentin men (61). In addition, a significantly higher rate ofbilateral involvement in women than men was foundfor neovascular AMD [odds ratio (OR), 7.7; 95% CI,1.3–46.7] in the Blue Mountains Eye Study (71). For allother lesions of AMD, nonsignificant increased ORswere found for bilateral involvement in women(OR, 2.4; 95% CI, 0.6–10.0 for geographic atrophyand OR, 1.6; 95% CI, 0.7–3.5 for early AMD). In theBeaver Dam Eye Study, exudative AMD was more

frequent in womenR75 years compared with men inthe same age group (6.7% vs. 2.6%, pZ0.02) (47). Inaddition, in an incidence study, after adjusting forage, the incidence of early AMD was 2.2 times (95%CI, 1.6–3.2) as likely in women R75 years comparedwith men this age (88).

Smith and associates pooled data from theRotterdam Study (58), the Beaver Dam Eye Study(47), and the Blue Mountains Eye Study (61) todetermine if females have a higher age-specific AMDprevalence than males (105). These three recent large-scale population-based studies used almost identicaldiagnostic techniques and criteria for AMD, and thepublished data are presented in identical form for agegroups 55 to 85 years. The overall pooled data show asignificant but modest increase in AMD prevalenceamong females compared with males, with OR of 1.15(95% CI, 1.10–1.21) adjusting for 10-year agecategories. Age stratum-specific pooled ORs (95% CI)show an increase in risk, rising from 0.62 (0.35–1.10)for ages 55 to 64 years to 1.04 (0.87–1.26) for ages 65–74years, and 1.29 (1.20–1.38) for ages 75 to 84 years.

The Melton Mowbray Eye Study (38) and theFramingham Eye Study (3,106) also found a higherprevalence of AMD among women. In NHANES-III,after controlling for age, white women (OR, 1.32; 95%CI, 1.10–1.61) and black women (OR, 1.39; 95% CI,1.00–1.92) had statistically significant higher odds ofhaving soft drusen (defined as drusen O63 mm) thandid men of the same race/ethnicity group, respect-ively (80). White women (OR, 1.24; 95% CI, 1.02–1.51)and black women (OR, 1.47; 95% CI, 1.06–2.03) werealso more likely to have early AMD present than whiteand black men, respectively (80).

The Los Angeles Latino Eye Study, a population-based, cross-sectional study of Latinos (primarilyMexican American) aged 40 years and older, foundthat compared with Latino women, Latino men wereat an approximately twofold increased risk of any(OR, 1.78; 95% CI, 1.47–2.16) or early (OR, 1.80; 95% CI,1.47–2.19) AMD (84). Menwere also more likely to havelateAMDthanwomen (OR, 1.6; 95%CI, 0.7–3.5), but thiswas not statistically significant. The reason for theincreased risk of AMD in men is not known. Somestudies have shown that several reported risk factorsfor AMD such as smoking, alcohol consumption,and cardiovascular disease tend to have a higher preva-lence in men than women. Indeed, Latino men weremore likely to smoke (21% vs. 9%, p!0.0001), drinkalcohol regularly (22% vs. 3%, p!0.0001), and hadelevated diastolic blood pressure (23% vs. 16%,p!0.0001) thanLatinowomen.However, after adjustingfor smoking, alcohol intake and elevated diastolicblood pressure, men were still more likely than womento have early AMD (OR, 1.91; 95% CI, 1.56–2.34).

Table 6 Risk Factors for Age-Related Macular Degeneration

Established risk factors

Age

Race/ethnicity

Heredity

Smoking

Possible risk factors

Sex

Socioeconomic status

Iris color

Macular pigment optical density

Cataract and its surgery

Refractive error

Cup/disc ratio

Cardiovascular disease

Hypertension and blood pressure

Serum lipid levels and dietary fat intake

Body mass index, waist circumference, and waist–hip ratio

Hematologic factors

Chlamydia pneumoniae infection

Reproductive and related factors

Dermal elastotic degeneration

Antioxidant enzymes

Sunlight exposure

Micronutrients

Dietary fish intake

Alcohol consumption

Factors probably not associated with AMD

Diabetes and hyperglycemia

Abbreviation: AMD, age-related macular degeneration.

54 AU EONG ET AL.

Race/EthnicityDifferences in genetic susceptibility probably explainpart of the disparities in the prevalence of AMD indifferent races. The low numbers of black participantsin the Macular Photocoagulation Study (MPS) trialsfor AMD suggested that the condition is less prevalentin black than in white populations (107). As of July 1,1991, only 1 (0.08%) out of 1319 patients enrolled inthe MPS trials for AMD was black while 1314 werewhite and 4 were listed as “other” (107).

Several studies have suggested that AMD ismore prevalent among whites than blacks (57,78,108,109). Gregor and Joffe, comparing 377 white patientsfrom London, England, with 864 age- and sex-matchedblack South Africans, found that drusen and pigmentepithelial changes were twice as common in whites asin black Africans (18.3% vs. 9.3%, p!0.001 and 11.4%vs. 4.6%, p!0.001, respectively) (108). They alsoobserved that disciform degeneration was present in3.5% of white patients compared with 0.1% of SouthAfrican patients (p!0.001).

In the Baltimore Eye Survey, a cross-sectionalpopulation-based study of black and white residentsof East Baltimore in Maryland, all AMD-related blind-ness were found in whites (78,109). Drusen (O63 mm)were identified in about 20% of individuals in bothblacks and whites, but large drusen (O125 mm) weremore common among older whites (15% for whites vs.9% for blacks over 70 years old) (78). Retinal pigmen-tary abnormalities were also more common amongolder whites (7.9% for whites vs. 0.4% for blacks over70 years old) (78). The prevalence ratio (white:black)was 10.7 for geographic atrophy, 8.8 for neovascularAMD, and 10.1 for all late AMD (geographic atrophyplus neovascular AMD) (78).

In the Barbados Eye Study (57), a population-based study in a large population of persons primarilyof African descent, age-related macular changesoccurred at a lower frequency than in the predomi-nantly white populations of the Maryland WatermenStudy (40) and the Beaver Dam Eye Study (47). Thefindings of at least one small (!63 mm) drusen waspresent in 66.2% of the Barbados Eye Study partici-pants, which is lower than that of 86% of MarylandWatermen Study participants and 94% of the BeaverDam Eye Study participants. The frequency of at leastone large drusen of 1.1% in the Barbados Eye Studywas also lower compared with these other studies,which had rates of 9% and 20% for the MarylandWatermen Study and Beaver Dam Eye Study, respect-ively. Neovascular AMD was found in 0.6% in theBarbados Eye Study. This was similar to the MarylandWatermen Study but lower than the 1.2% found in theBeaver Dam Eye Study. One caveat in the interpre-tation of the Barbados Eye Study, which is based on 308stereoscopic fundus photographic grading, is that

because the gradability of the fundus photographsdecreased significantly with increasing age, predomi-nantly as a result of an increasing incidence andseverity of media opacities, and the participantsexcluded from the data analyses tended to the older,the frequencies presented in the Barbados Eye Studymay underestimate the true frequency of AMD in thispopulation (57).

In NHANES-III, after adjusting for age, thefrequency of early AMD was similar in non-Hispanicwhites comparedwith that of non-Hispanic blacks andMexican Americans (80). Although the frequencies ofsoft drusen appear similar among the racial/ethnicgroups, retinal pigmentary abnormalities and signs oflate AMD are more frequent in non-Hispanic whitesthan in non-Hispanic blacks and Mexican Americans.For increased retinal pigment and RPE depigmenta-tion, the ORs (95% CI) comparing non-Hispanic blackswith non-Hispanic whites were 0.47 (0.31–0.72) and0.59 (0.33–1.04), respectively, and for comparingMexican Americans with non-Hispanic whites, theywere 0.41 (0.21–0.81) and 0.72 (0.44–1.19), respectively.For late AMD, the OR (95% CI) for non-Hispanicblacks compared with non-Hispanic whites was 0.34(0.10–1.18) and for Mexican Americans compared withnon-Hispanic whites, it was 0.25 (0.07–0.90). Interest-ingly, before 60 years of age, Mexican Americans(OR, 1.53; 95% CI, 1.00–2.35) and non-Hispanicblacks (OR, 1.59; 95% CI, 0.86–2.95) had a greaterchance of having any AMD than non-Hispanicwhites; thereafter, Mexican Americans (OR, 0.63; 95%CI, 0.44–0.90) and non-Hispanic blacks (OR, 0.50; 95%CI, 0.37–0.68) had a lesser chance than non-Hispanicwhites (81). Other Hispanics, Asians, and nativeAmericans were included in NHANES-III but werenot reported due to inadequate sample sizes.

Klein et al. studied theprevalence of a large cohortof black and white participants in the AtherosclerosisRisk In Communities Study and found that the overallprevalence of anyAMDwas lower in blacks (3.7%) thanwhites (5.6%) (79). After controlling for age and sex, theOR for any AMD in blacks compared with whites was0.73 (95% CI, 0.58–0.91). The prevalence of most of thecomponent lesions that define early AMD was alsolower in blacks than whitesR60 years of age.

Klein et al. recently reported the prevalence ofAMD in four racial/ethnic groups (white, black,Hispanic, and Chinese) that participated in theMulti-Ethnic Study of Atherosclerosis (98). Thisprospective cohort study examined 6166 45- to85-year-old subjects selected from six U.S. commu-nities. The study found the prevalences of any AMDwere 2.4%, 4.2%, 4.6% and 5.4% for blacks, Hispanics,Chinese, and whites, respectively (p!0.001 for anydifferences among groups). Estimated prevalences oflate AMD were 0.3%, 0.2%, 1.0%, and 0.6% for blacks,


Hispanics, Chinese, and whites, respectively. Thefrequency of neovascular AMD was highest inChinese (age- and gender-adjusted OR, 4.30; 95% CI,1.3–14.27) compared with whites. Differences inage, gender, pupil size, body mass index (BMI),smoking, alcohol drinking history, diabetes, andhypertension status did not explain the differences ofAMD prevalences among the racial/ethnic groups.

Klein and Klein, using data from NHANES-I,found no difference between whites and blacks in thepercentage of patientswithAMD (35). Another analysisof the same data came to the same conclusion (8).

It is unclear whether the degree of funduspigmentation affects the ability to detect lesions suchas hyperpigmentation and hypopigmentation of theRPE, and soft drusen that characterize AMD. It isplausible that variations in normal fundus pigmenta-tion may lead to errors in detecting subtle early AMDlesions, resulting in apparent differences among theethnic groups.

Overall, current evidence suggests that earlyAMD is common among blacks and Hispanics butless common than among non-Hispanic whites.However, late AMD is less frequent in these groupscomparedwith non-Hispanicwhites. Racial differencesin AMD support a potential genetic component to thiscondition.

HeredityAnalysis of heredity in the disease process of AMDis limited by the fact that the disorder is associatedwith aging, frequently causing its most significantphenotypic manifestations in the later years of life.As a result, usually only one generation in the appro-priate age range is available for study. The parentsof the proband are often deceased, and the childrenare often too young to manifest the disease. Becauseinformation from several generations of families ofmultiple affected individuals is often lacking, geneticanalysis is limited.

Clinical experience indicates that AMD demon-strates familial clustering, suggesting that hereditymay be an important factor in the etiology ofthis condition although the exact role and relativecontribution of genetics in the pathogenesis isunknown (55,110–113). It is believed that this geneticpredisposition, in the presence of appropriateenvironmental influences, causes the aging maculato manifest AMD.

Although Hutchinson and Tay observed afamilial occurrence of AMD as early as 1875 (114),the association between heredity and AMD has notbeen well studied until recently. Bradley in 1966commented on his patients with AMD that “nearlyevery patient I have seen has had other members of thefamily similarly afflicted” (115). In 1973, Gass reported

a positive family history of loss of central vision in 10%to 20% of his patients with AMD (116).

Hyman et al. reported a statistically significantassociation between AMD and a family history ofthe disease either in the parents and siblings (OR,2.9; 95% CI, 1.5–5.5) (12). A significantly higher cor-relation of number of drusen between siblings thanbetween spouses was found by Piguet et al. (110).The lack of concordance between spouses who haveshared a common environment for at least 20 yearssuggests that environmental factors may not play akey role in the etiology of AMD (110). Seddon et al.found the overall prevalence of AMD was higheramong first-degree relatives of cases than amongrelatives of controls (OR, 2.4; 95% CI, 1.2–4.7) (117).They also found that familial aggregation of AMDwasassociated with the type of AMD in the proband, i.e.,dry AMD (large or extensive macular drusen, RPEabnormalities, and geographic atrophy) versusexudative AMD [RPE detachment or chor-oidal neovascularization (CNV)]. Relatives ofprobands with exudative disease were significantlymore likely to have AMD than were relatives ofcontrol probands after adjusting for age and sex(OR, 3.1; 95% CI, 1.5–6.7). On the other hand,relatives of probands with dry AMD were slightlymore likely to have AMD than were relatives ofcontrol probands (OR, 1.5; 95% CI, 0.6–3.7), but thisdifference was not statistically significant. In anotherstudy, the OR of siblings for AMD of patientscompared with siblings of controls was 25.2 (95%CI, 3.4–519.0) (118).

In the Blue Mountains Eye Study, subjects withsigns of AMD (4.5%) were more likely to report a first-degree family history of AMD than among subjectswithout AMD (2.3%) (67). The highest rate wasreported by subjects with late AMD (6.9%), particu-larly those with neovascular AMD (8.2%). Afteradjusting for age, sex, and current smoking, a clearincrease in risk associated with family history, from noAMD [OR, 1.0 (index)] to early AMD (OR, 2.17; 95%CI, 1.04–4.55), late AMD (geographic atrophy orneovascular AMD) (OR, 3.92; 95% CI, 1.34–11.46),and neovascular AMD (OR, 4.30; 95% CI, 1.37–13.45)was observed (67).

Klaver et al. examined the siblings and children ofprobands derived from the population-basedRotterdam Study (119). First-degree relatives of 87patients with late AMD (geographic atrophy orneovascular AMD) were compared with those of 135controls without AMD. For siblings, the prevalence ofearly AMD was 9.5% for siblings of patients versus2.9% for siblings of controls (pZ0.04, age and sexadjusted), and for late AMD was 13.4% versus 2.2%(pZ0.001, age and sex adjusted). For offspring, theprevalence of early AMD was 6.3% for offspring of

56 AU EONG ET AL.

patients versus 1.9% for offspring of controls (pZ0.05,age and sex adjusted), and late AMD was presentin only 1.4% of offspring of patients (pZ0.20, age andsex adjusted). The prevalence of early (OR, 4.8; 95% CI,1.8–12.2) and late (OR, 19.8; 95% CI, 3.1–126.0) AMDwas significantly higher in first-degree relatives ofpatients with late AMD than in relatives of controls.The lifetime absolute risk estimate of developing earlyAMD was 48% (95% CI, 31–65%) for relatives ofpatients versus 23% (95% CI, 10–37%) for relativesof controls (pZ0.001), yielding a risk ratio of 2.1 (95%CI, 1.4–3.1). The lifetime risk estimate of late AMDwas50% (95% CI, 26–73%) for relatives of patients versus12% (95%CI, 2–16%) for relatives of controls (p!0.001),yielding a risk ratio of 4.2 (95%CI, 2.6–6.8). The authorscalculated that the population-attributable risk relatedto genetic factors was 23% (119).

No association, however, was found betweenfamily history and AMD in the small population-based Melton Mowbray Eye Study (38). It should bepointed out that in studies in which the family historydata were ascertained by interview alone, the datashould be interpreted with caution since reportedhistories of ocular disease are unreliable (120).

Three reports of single pairs of monozygotictwins (121–123) and two larger series, with 9 (112)and 50 pairs of identical twins (124), described a highconcordance of early and late AMD in the twins.Gottfredsdottir et al. examined the concordance ofAMD in 100 monozygotic twins (50 pairs) and 47spouses (124). The average duration of marriage forthe twin/spouse pairwas 30 years (range, 26–50 years).The concordance of AMD was 90% in monozygotictwin pairs which significantly exceeded that of 70% fortwin/spouse pairs (pZ0.0279). In the nine twin pairsthat were concordant, fundus appearance and visualimpairment were similar. Although the environmentalinfluences are probably more similar for identicaltwins than for dizygotic twins, other siblings, orunrelated individuals, the strikingly similar incidenceof age-related macular changes in these identical twinssuggests that a substantial genetic component mayexist in some patients with AMD.

Although AMD runs in families, the phenotypicappearance of the macula within families with thedisorder tends to be quite variable and representativeof the wide range of findings typically associated withAMD, i.e., both neovascular AMD and geographicatrophy, and early signs of AMD may be present indifferent individuals within the families (125). Indeed,neovascular and nonneovascular AMDwere observedamong different individuals in four of eight families inthe study, suggesting that geographic atrophy may bepart of the same disease process as neovascular AMD.On the other hand, the distinctly different phenotypesof the two forms of late AMD may also indicate

different origins. It is currently unknown whygeographic atrophy develops in some instances andneovascular AMD in others, even within thesame family.

Socioeconomic StatusIn NHANES-I, a significant negative trend (p!0.03) ofdecreased prevalence of AMD was found withincreasing levels of education (8). Compared withthe least educated group, persons who attended highschool have a reduced prevalence of AMD (OR, 0.64;95% CI, 0.44–0.92) as do persons who have someeducation beyond high school (OR, 0.71; 95% CI,0.44–1.15). The Eye Disease Case–Control Studyfound that persons with higher levels of educationhad a slightly reduced risk of neovascular AMD, butthe association did not remain statistically significantafter multiple regression modeling (7).

The Beaver Dam Eye Study found no relation ofincome, educational level, or marital status to AMD(48). No association between social class and AMDwas found in the Melton Mowbray Eye Study (38).Two case–control studies found no associationbetween AMD and occupations (10,12).

OCULAR FACTORS

Macular Pigment Optical DensityRecently, there is heightened interest in the potentialrole of macular pigment in protecting against AMD(126). The yellow macular pigment, which charac-terizes the retinas of primates including man, wasshown in 1985 to be composed of two chromatographi-cally separable components, namely lutein andzeaxanthin (127). Of note, lutein and zeaxanthin areentirely of dietary origin.

Although the exact role of the macular pigmentremains uncertain, several functions have beenhypothesized. These include limiting the effects oflight scatter and chromatic aberration on visualperformance (128,129), reducing the damaging photo-oxidative effects of blue light through its pre-receptorialabsorption (130,131), and protecting against theadverse effects of reactive oxygen intermediatesthrough its antioxidant properties (132). There is agrowing body of evidence that oxidative damageplays a role in the pathogenesis of AMD (133–136).Consequently, some have suggested that the absorp-tion characteristics and antioxidant properties ofmacular pigment confer protection against AMD(132,137).

In brief, the evidence that macular pigmentoptical density confers protection against AMD restson a biologically plausible rationale and the fact thatseveral risk factors for this condition are themselvesassociated with a relative lack of the pigment. Any


beneficial effect of macular pigment must reside in itsability to protect against chronic and cumulativedamage. In other words, macular pigment levels inyoung and middle age are likely to determine theprotection, if any, that this pigment confersagainst AMD.

For example, some studies have found thatmacular pigment optical density declines withincreasing age in normal eyes (138,139), althoughsome have not (140,141). In addition, it has beenfound to be significantly different between males andfemales. In one study, macular pigment optical densityfor males was 38% higher than for females (142). Giventhe putative protective role of macular pigment (132),this finding may explain the higher prevalenceof AMD in females found in some studies (seeabove). Likewise, a strong inverse relationshipbetween smoking and macular pigment opticaldensity has been shown by Hammond et al., and thismay explain how smoking increases the risk of AMD(see below) (143). Interestingly, the average levels ofmacular pigment have been reported as 32% lower ineyes with AMD than in normal age-matched controleyes in subjects not consuming high-dose luteinsupplements (pZ0.001) (138).

Although discussed under the heading of ocularrisk factors, macular pigment optical density is inher-ently related to nutrition since it can be altered bydietary modification or supplementation (144–147).Consumption of certain fruits and vegetables willincrease the dietary intake of lutein and zeaxanthin(148). Hammond et al. reported that an averageincrease of approximately 20% in human macularpigment optical density was obtained after fourweeks of a diet enriched in corn and spinach (145).The Eye Disease Case–Control Study reported that ahigh dietary intake of macular pigments from leafygreen vegetables was associated with a reduced risk ofneovascular AMD (see below) (16). In one report, twosubjects who took a daily dose of 30 mg of lutein for140 days had mean increases in the macular pigmentoptical density of 39% and 21% in their eyes, respect-ively (146). The authors estimated that this increasein macular pigment resulted in a 30% to 40% reductionin blue light reaching the photoreceptors, Bruch’smembrane, and the RPE. Because human macularpigment can be augmented with dietary modificationand nutritional supplementation, the protectiveeffect of macular pigment, if proven, has potentialtherapeutic implications.

Nevertheless, evidence that dietary modifi-cations or supplementation with lutein and/orzeaxanthin can prevent, delay, or modify the courseof AMD is still lacking. Ultimately, a well-designedrandomized controlled trial with a long follow-upsuch as the second phase of the National Eye

Institute’s Age-Related Eye Disease Study (AREDS)will be required to test such a hypothesis.

Cataract and Its SurgerySince cataract and AMD are the most frequent causesof visual impairment in older individuals and theirprevalence is strongly age related (149), a possibleassociation between the two conditions has longbeen debated. There are potential risk factorscommon to both conditions, such as antioxidantintake (150), cigarette smoking (151), and sunlightexposure (41,42,152–154).

The association between cataract and AMDhas not been found consistently. In the small popu-lation-based study in Melton Mowbray (38) anda case–control study by Tsang et al. (15), no statisticallysignificant association was found between cataractand AMD. Sperduto and Siegel found no associationbetween cataract and AMD when the various age-related lens changes were pooled in the FraminghamEye Study and they concluded that cataract and AMDare unrelated and developed entirely independently(31). However, when they reexamined the same data tostudy specific types of cataracts, they found a positiveassociation between AMD and cortical changes and anegative association between AMD and nuclearsclerosis (32). The Andhra Pradesh Eye DiseaseStudy, a population-based study involving 3723participants aged 40 to 102 years in southern India,also found cortical cataract, but not nuclear sclerotic orposterior subcapsular cataract, to be significantlyassociated with an increased prevalence of AMD(adjusted OR, 2.87; 95% CI, 1.57–5.27) (86). In contrast,Klein et al. found a positive association between earlyor any AMD and nuclear sclerosis but no relationshipof cortical cataract or of posterior subcapsular cataractto early or late AMD in the Beaver Dam Eye Study (49).In addition, there was no relationship of nuclear orcortical cataract to the incidence and progression ofAMD (89).

An analysis of the data from NHANES-I by Liuet al. found that the ORs of having AMD in eyes withlens opacity without visual impairment and cataractwhen compared with eyes with no lens opacity were1.80 (95% CI, 1.40–2.30) and 1.14 (95% CI, 0.84–1.55),respectively (36). The authors postulated that theweak association between cataract and AMD mayreflect the difficulty of visualizing the ocular fundusin eyes with dense cataract. Other theories include thepossibility that retardation of transmission of light tothe retina by cataract decreases the extent of lightdamage, and that different kinds of cataracts mayhave differing pathogeneses and for some types, nocommon factors may be shared with AMD (36). TheFRANCE-DMLA Study Group, comparing 1844cases of AMD with a similar number of age- and

58 AU EONG ET AL.

sex-matched controls, found that persons with lensopacities had an increased risk of AMD (OR, 1.69;95% CI, 1.45–1.97) (22).

Several authors have noted deterioration ofAMD following cataract surgery (155–159). In onestudy, Pollack et al. evaluated 47 patients withbilateral, symmetric, early AMD who underwentextracapsular cataract extraction with intraocularlens implantation in one eye (157). They found thatprogression to neovascular AMD occurred moreoften in the operated eyes (19.1%) compared withthe fellow eyes (4.3%). This concurs with a histologicstudy that suggested a higher prevalence of disciformmacular degeneration in pseudophakic eyes than inage-matched phakic eyes (160). Interestingly, Pollacket al. found that progression to neovascular AMDoccurred significantly more often in men than inwomen (p!0.05) (157).

In the Beaver Dam Eye Study, eyes that hadundergone cataract surgery before baseline, comparedwith eyes that were phakic at baseline, were morelikely to have progression of AMD (OR, 2.71; 95% CI,1.69–4.35) and to develop signs of late AMD (OR, 2.80;95% CI, 1.03–7.63) after controlling for age (89). Theserelationships remained after controlling for other riskfactors in multivariate analyses. The FRANCE-DMLAStudy Group found an increased risk of AMD inpersons with a history of previous cataract surgerycompared with those with no lens opacities or cataractsurgery (OR, 1.68; 95% CI, 1.45–1.95) (22). Similarly,prior cataract surgery was significantly associatedwith an increased prevalence of AMD in the AndhraPradesh Eye Disease Study (adjusted OR, 3.79; 95% CI,2.1–6.78) (86). Liu et al. found that data fromNHANES-I suggest the OR of having AMD in eyeswith aphakia compared with eyes with no lens opacitywas 2.00 (CI, 1.44–2.78) (36). They suggested that anincrease in light transmittance following cataractsurgery may reinitiate and dramatically accelerateprogression to more advanced AMD. It is also possiblethat the association is a result of easier visualizationand detection of AMD lesions after cataract surgery(89). It has also been hypothesized that inflammatorychanges that may occur in eyes following cataractsurgery may be related to the development of lateAMD (160).

In the Blue Mountains Eye Study, a higherprevalence of late AMD in eyes with past cataractsurgery (6.3%) than in phakic eyes (1.3%) wasobserved. However, the association was primarily aneffect of age because the OR for late AMD reduced to1.3 (95% CI, 0.6–2.6) and became nonsignificant afteradjusting for age and sex, and to 1.2 (95% CI, 0.5–2.9),after multivariate adjustment (72). Similarly, a higherprevalence of early AMD was found in eyes with ahistory of cataract surgery (7.1%) than in phakic eyes

(4.4%), with a multivariate-adjusted OR of 0.7 (95% CI,0.4–0.9), which suggests a protective effect for cataractsurgery (72). The Rotterdam Study also did not findany association between cataract surgery and AMDprevalence (161).

It is unclear why the results vary among thestudies. It is possible that these variations in findingsmay have resulted from differences in the studypopulation and/or from differences in methodologyand case definitions.

Iris ColorIris color is a hereditary factor that may be associatedwith AMD (13). However, this association has not beenconsistently found in studies. A number of studieshave reported an increased risk of AMD in peoplewith blue or light iris color compared with those withdarker iris pigmentation (12,13,17,22,62) and onestudy documented worse AMD in subjects with lightiris color (162). Others, however, have found noassociation between iris color and AMD (7,14,15,38,39,44,89). The Beaver Dam Eye Study did not find anyrelationship between iris color and the incidenceand progression of AMD (89). One histologic studyfound no significant correlation between iris color andmacular aging (160). Data from NHANES-III showedthat blue iris color was negatively associated withsoft drusen in non-Hispanic whites (OR, 0.69; 95%CI, 0.55–0.88) but not in Mexican Americans (OR,0.35; 95% CI, 0.05–2.72) (80). The reasons for thesedisparities are not clear.

Case–control studies by Hyman et al. (12) andWeiter et al. (13) demonstrated a positive associationbetween light iris color and AMD. In Hyman et al.’sseries, only 9.2% of 162 cases had brown iridescompared with 26.4% of 174 controls (pZ0.0002) (12).Blue or lightly pigmented irides were associatedwith ahigher risk of AMD, the degree of association beinggreater for men (OR, 8.3; 95% CI, 2.3–29.7) than forwomen (OR, 2.4; 95% CI, 1.1–5.0) (12). Weiter et al.found that 76% of 650 patients with AMD had lightirides compared with 40% of 363 controls (pZ0.0001)(13). In addition, patients with AMDand light iris colorwere found to be significantly younger (mean age,73.6G7.3 years) than those with dark iris color (meanage, 78.3G5.8 years; pZ0.0008) (13). The FRANCE-DMLA Study Group found that persons with lightiris color (blue, green, and gray) had increased risk ofAMD compared with those with dark iris color (OR,1.22; 95%CI, 1.05–1.42) (22). This concurs with the BlueMountains Eye Study which found that blue iris colorwas significantly associated with an increased riskfor both early AMD (OR, 1.5; 95% CI, 1.1–1.9) andlate AMD (OR, 1.7; 95% CI, 1.0–2.9) (62).

Holz et al. found that lighter present iriscolor, but not initial iris color during youth, was


associated with an increased risk of AMD (17). Theycalculated that a history of decreasing iris color wasassociated with a 5.55-fold (95% CI, 2.03–15.91)increase in risk of AMD (pZ0.0001). Some studieshave shown that declines in the melanin content ofthe iris and RPE occur with age (163,164). The BeaverDam Eye Study showed higher prevalences of blue orgray iris color with increased age, but no relationshipwas found between iris color and the incidence orprogression of AMD in the study (89).

The mechanism by which iris pigmentationmight influence AMD is uncertain, but a plausibleexplanation is that the lower risk for AMDamong subjects with darker iris color may be due tothe fact that these individuals have more tissuemelanin, including the choroid. Indeed, funduspigmentation was found to correspond closely to irispigmentation both clinically and by objective histo-logic microdensitometric techniques (13). Thisincreased pigmentation may provide some protectionto the retina from exposure to sunlight, reducing directphotooxidative damage and thus reducing the risk ofAMD (see below). This is consistent with the obser-vation in some studies that AMD is more prevalentamong whites than among the more pigmented races(57,78,109).

Refractive ErrorSeveral case–control studies have found an associationbetween AMD and refractive error, with hyperopiceyes at greater risk of AMD (10–12,22). Hyman et al.found that statistically significant differences in meanrefractive error were present between female casesand controls (pZ0.009), but not between male casesand controls (pZ0.16) (12). Female cases had a morepositive refractive error (meanZ1.8 diopters) thanfemale controls (meanZ1.1 diopters). The FRANCE-DMLA Study Group found the ORs for AMD inhyperopes and myopes, compared with emmetropes,were 1.33 (95% CI, 1.11–1.59) and 0.99 (95% CI,0.78–1.25) (22). The Eye Disease Case–Control Studyfound that persons with hyperopia had a slightlyhigher risk of neovascular AMD, but the associationdid not remain statistically significant after multi-variate modeling (7). One caveat in the interpretationof findings in these case–control studies is that becausethe controls were recruited from ophthalmologicclinics, the control groups may be enriched in theproportion of myopes compared with the generalpopulation. In fact, in the case–control study by theFRANCE-DMLA Study Group, the authors stated that“the majority of the control group was seen forrefractive problems” (22).

Data from NHANES-I showed that the ORs(95% CI) of AMD in hyperopes and myopes,

compared with emmetropes, were 1.61 (1.15–2.25)and 1.33 (0.69–2.57), respectively (8). This differsfrom the Beaver Dam Eye Study, which showed aprotective effect of borderline significance of hyper-opia at baseline on the incidence of early AMD, butno relationship to the incidence of late AMD or tothe progression of AMD (89).

Cup/Disc RatioThe Eye Disease Case–Control Study found that eyeswith large horizontal and vertical cup/disc ratioswere at reduced risk for neovascular AMD (7). Thehorizontal cup/disc ratio persisted as statisticallysignificant after multivariate modeling, adjusting forknown and potential confounding factors. Thisfinding is consistent with the association betweenAMD and hyperopia.

SYSTEMIC FACTORS

Cardiovascular Disease and Its Risk FactorsA number of documented risk factors for cardio-vascular disease such as age, hypertension,hypercholesterolemia, diabetes, smoking, and dietaryintake of fats, alcohol, and antioxidants have beenassociated with AMD in some studies (165). Thisraises the possibility that the causal pathways forcardiovascular disease and AMD may share similarrisk factors. Results from studies, however, have notbeen consistent.

Cardiovascular DiseaseA number of studies have suggested an associationbetween AMD and various clinical manifestations ofcardiovascular disease. In a case–control study,Hyman et al. found AMD to be positively associatedwith a history of three cardiovascular conditions (12).These conditions are arteriosclerosis, circulatoryproblems, and stroke and/or transient ischemicattacks, with ORs (95% CI) of 2.3 (1.9–2.7), 2.0(1.1–3.5), and 2.9 (1.3–6.9), respectively (12). TheFRANCE-DMLA Study Group found an increasedrisk of AMD in persons with a history of coronaryartery disease (OR, 1.31; 95% CI, 1.02–1.68) (22). InNHANES-I, a positive association between AMD andcerebrovascular disease was found, but positiveassociations with other vascular diseases did notreach statistical significance (8).

The Rotterdam Study found that atheroscleroticplaques in the carotid bifurcation, as assessed ultra-sonographically, were associated with a 4.5 timesincreased prevalence OR (95% CI, 1.9–10.7) of eithergeographic atrophy or neovascular AMD (59). Thosewith plaques in the common carotid artery or withlower extremity arterial disease (as measured by theratio of the systolic blood pressure level of the ankle

60 AU EONG ET AL.

to systolic blood pressure of the arm) had the sameincreased prevalence OR of 2.5 (95% CI, 1.4–4.5). Fromthese observations, the authors suggested that athero-sclerosis may be involved in the etiology of AMD.However, other cardiovascular disease risk factorssuch as hypertension, systolic blood pressure, totalcholesterol, and high-density lipoprotein (HDL)cholesterol were not associated with AMD in thesame study (59). Diastolic blood pressure was margin-ally higher in AMD cases than in those withoutAMD, but this did not reach statistical significance (59).In subjects participating in the Atherosclerosis Risk InCommunities Study, presence of carotid artery plaquewas significantly associated with RPE depigmentation(OR, 1.77; 95%CI, 1.18–2.65) (79). Focal retinal arteriolarnarrowing was also associated with RPE depigmen-tation (OR, 1.79; 95% CI, 1.07–2.98) in the same study.In a Finnishpopulation-based study, a significant corre-lation between the severity of retinal arteriolar sclerosisand AMD (pZ0.0034) was found (73).

Several case–control studies, including the EyeDisease Case–Control Study, found that persons whoreport a history of cardiovascular disease did not havea significantly increased risk of AMD (7,10,15). TheBeaver Dam Study (50), the Atherosclerosis Risk InCommunities Study (79), and the Blue Mountains EyeStudy also found no statistically significant relation-ship between a history of stroke or cardiovasculardisease with early or late AMD.

Hypertension and Blood PressureTwo large population-based studies showed a smalland consistent significant association between AMDand systemic hypertension (8,29,33). Kahn et al., usingthe Framingham Heart and Eye Studies data, found apositive association between the presence of AMD andhigher levels of diastolic blood pressure measuredmany years before the eye examination (29). Diastolicblood pressure was also associated with AMD in asmall Israeli study (166). Also using data from theFramingham Heart and Eye Studies, Sperduto andHiller found the age- and sex-adjusted relative risk(RR) for any AMD was 1.18 (95% CI, 1.01–1.37) forpersons diagnosed with hypertension 25 years beforethe eye examination and 1.04 (95% CI, 0.96–1.23) forpersons with hypertension at the time of the eyeexamination, when compared with those withouthypertension (33). In addition, an increase in the ORof AMD with longer duration of systemic hyperten-sion was documented. The NHANES-I showed thatsystolic blood pressure and hypertension were associ-ated with AMD (8). Persons with a history ofhypertension were 1.36 times (95% CI, 1.00–1.85)more likely to have AMD compared with personswithout such a history. In addition, the prevalence ofAMD increased with increasing levels of systolic blood

pressure although the test for trend was only margin-ally significant (p!0.08). However, elevated diastolicblood pressure was not associated with AMD.

The Beaver Dam Eye Study found elevatedsystolic blood pressure to be significantly related tothe presence of RPE depigmentation in females(OR, 1.07; 95% CI, 1.00–1.14) but not in males (50).Pulse pressure was also related to the presence of RPEdepigmentation (OR, 1.10; 95%CI, 1.01–1.19), increasedretinal pigment (OR, 1.07; 95% CI, 1.00–1.15), andpigmentary abnormalities (OR, 1.08; 95% CI,1.01–1.15) in females but not in males (50). However,hypertension or diastolic blood pressure was notrelated to any sign of early or late AMD in either sex.In an incidence study, after controlling for age andsex, both higher systolic blood pressure (OR per10 mmHg, 1.16; 95% CI, 1.05–1.27) and uncontrolled“treated” hypertension (OR, 1.98; 95% CI, 1.00–3.94)were related to the incidence of RPE depigmentation,but not development of neovascular AMD (90). Higherpulse pressure was significantly associated withincreased incidence of RPE depigmentation (OR per10 mmHg, 1.27; 95% CI, 1.14–1.42) and neovascularAMD (OR per 10 mmHg, 1.29; 95% CI, 1.02–1.65) aftercontrolling for age and sex.

Systemic hypertension was found to be a signi-ficant risk factor for AMD by the FRANCE-DMLAStudy Group (22). Another recent case–control studyby the Age-Related Macular Degeneration RiskFactors Study Group analyzed risk factors separatelyfor neovascular and nonneovascular AMD to addressthe possibility that the two forms of AMD havedifferent risk factors (24). The group showed thatneovascular AMD, but not nonneovascular AMD, isassociated with moderate to severe hypertension (24).Neovascular AMD was found to be positively associ-ated with diastolic blood pressure greater than95 mmHg (OR, 4.4; 95% CI, 1.4–14.2), self-reporteduse of antihypertensive medications more potentthan diuretics (OR, 2.1; 95% CI, 1.2–3.0), physician-reported history of hypertension (OR, 1.8; 95%CI, 1.2–3.0), and physician-reported use of any anti-hypertensive medications (OR, 2.5; 95% CI, 1.5–4.2).The findings in this study suggest that neovascularAMD and hypertension may have a similar systemicprocess. In addition, it supports the hypothesis thatneovascular and nonneovascular AMD may arisethrough different pathogenetic mechanisms.

No association between hypertension and AMDwas found in several population-based cross-sectionalstudies including the Rotterdam Study (59), BlueMountains Eye Study (66), Atherosclerosis Risk InCommunities Study (79), and Andhra Pradesh EyeDisease Study (86), or in several case–control studies(7,10,12,15). In the Eye Disease Case–Control Study, nosignificant association was found with hypertension


and AMD, but a trend for an increased risk associatedwith higher systolic blood pressure was seen (7).

Serum Lipid Levels and Dietary Fat IntakeSome evidence suggests that dietary fat intake,particularly intake of saturated fat and cholesterol, isassociated with an increased risk for atherosclerosis(167). It is biologically plausible that higher dietarysaturated fat intake increases the risk of AMD bypromoting atherosclerosis.

The Eye Disease Case–Control Study found thatpersons with midrange (4.889–6.748 mmol/L) andhigh (R6.749 mmol/L) total cholesterol levels com-pared with those with low levels (%4.888 mmol/L)had ORs for neovascular AMD of 2.2 (95% CI, 1.3–3.4)and 4.1 (95% CI, 2.3–7.3), respectively, after controllingfor other factors (7). A slight but not statisticallysignificant increased risk in neovascular AMD wasseen with increasing levels of serum triglycerides inthe same study (7).

In the Beaver Dam Eye Study, after controllingfor age, total serum cholesterol was inversely relatedto early AMD in women (OR, 0.89; 95% CI, 0.80–0.98),whereas the total cholesterol/HDL cholesterol ratiowas inversely related (OR, 0.89; 95% CI, 0.84–0.96) andHDL cholesterol was positively related to early AMDin men (50). The Cardiovascular Health Study alsoshowed a small but significant inverse associationbetween total serum cholesterol and early AMD(168). Pooled data from three cross-sectional studies(Blue Mountains Eye Study, Beaver Dam Eye Study,and Rotterdam Study) found that total serum choles-terol was associated inversely with incidentneovascular AMD (OR, 0.94 per 10 mg/dL; 95% CI,0.88–0.99) (169). The reasons for these associations arenot clear although some authors have suggestedselective survival as a possible explanation. Becausepersons with higher cholesterol levels or lower HDLcholesterol levels are at higher risk of cardiovasculardeaths than persons with normal levels of cholesterol,a positive relationship may have been obscured. Inter-estingly, the pooled data also showed that total serumcholesterol was associated directly with incidentgeographic atrophy (OR, 1.08 per 10 mg/dL; 95% CI,1.00–1.15) (169), and this association cannot beexplained by selective survival.

The Age-Related Macular Degeneration RiskFactors Study Group found neovascular AMD, butnot nonneovascular AMD, to be positively associatedwith HDL level (OR, 2.3; 95% CI, 1.1–4.7) and dietarycholesterol level (OR, 2.2; 95% CI, 1.0–4.8) (24).

In the Beaver Dam Eye Study, persons withintake of saturated fat and cholesterol in the highestcomparedwith the lowest quintile had ORs of 1.8 (95%CI, 1.2–2.7) and 1.6 (95% CI, 1.1–2.4) for early AMD,respectively, after adjusting for age and beer intake

(56). However, no significant association betweenthese intakes was found with late AMD (56). Thefindings in this study concurs with the BlueMountainsEye Study, which found that total and saturated fatintake were associated with a borderline significantincrease in risk for early AMD [ORs (95% CI) forhighest compared with lowest quintiles of intake,1.60 (0.94–2.73) and 1.50 (0.91–2.48), respectively],but not for late AMD (68). A significant association(p for trend Z0.05) for increasing prevalence of earlyAMDwith increasing monounsaturated fat intake wasobserved. Cholesterol intake was associated with aborderline significant increased risk for late AMD[OR (95% CI) for highest compared with lowestquintiles of intake, 2.71 (0.93–7.96); p for trendZ0.04].

The Rotterdam Study (59), Blue Mountains EyeStudy (66), and Atherosclerosis Risk In CommunitiesStudy (79) did not find any association betweenserum cholesterol and HDL cholesterol with AMD.No significant association between AMD and serumcholesterol was also found in the Framingham EyeStudy (29), NHANES-I (8), and several small studies(15,170,171). No difference in the levels of plasmacholesterol and fatty acids was found between 65cases of neovascular AMD and control pairs in astudy by Sanders et al. (172).

Several studies have evaluated the relationship oflipid-lowering agents and AMD and found conflictingresults. The Beaver Dam Eye Study (173), Blue Moun-tains Eye Study (174), Rotterdam Study (175), and arecent case–control study using data from the Cardio-vascular Health Study (28) found no associationbetween the use of a lipid-lowering agent and therisk of developing AMD. There was, however, asuggestion that use of 3-hydroxy-3-methylglutarylcoenzyme A reductase inhibitors or statins mightincrease the risk of AMD (OR, 1.40; 95% CI, 0.99–1.98)after controlling for age, sex, and race (28). Conversely,two cross-sectional studies (176,177) and one nestedcase–control study (178) reported that individuals withAMD were less likely to have used statins.

Diabetes and HyperglycemiaThe majority of studies that have investigated therelationship between diabetes and/or hyperglycemiaand AMD have found no significant association(7,10,12,15,29,73,79,179).

One small study by Vidaurri et al. observed anassociation between drusen and serum glucose infemales but not in males (166). In the Beaver DamEye Study, diabetes was not associated with earlyAMD (51). In persons R75 years, those with diabeteshad a higher frequency of neovascular AMD (9.4%)than those without (4.7%) but both groups had similarfrequencies of geographic atrophy. The RR of neovas-cularAMD indiabeticmen comparedwith nondiabetic

62 AU EONG ET AL.

menR75 yearswas 10.2 (95%CI, 2.4–43.7); for females,it was 1.1 (95% CI, 0.4–3.0). The authors suggested thatthe relationship of neovascular AMD in older men, butnot women, might be the result of chance. In the samestudy, no relationshipwas found between glycosylatedhemoglobin and any signs of AMD in nondiabeticpersons (51). The Blue Mountains Eye Study foundgeographic atrophy to be significantly associated withdiabetes (OR, 4.0; 95% CI, 1.6–10.3), but no associationwas found with either neovascular AMD (OR, 1.2; 95%CI, 0.4–3.5) or earlyAMD (OR, 1.0; 95%CI, 0.5–1.8) (63).There was also no association found between impairedfasting glucose and AMD (63). The AtherosclerosisRisk In Communities Study (79) did not find anyassociation between AMD with diabetes.

Overall, there is scant evidence in the literature tosuggest a real relationship between diabetes and/orhyperglycemia and AMD.

BMI, Waist Conference, and Waist Hip RatioIn the Blue Mountains Eye Study, having a BMI[defined as body weight in kilograms divided byheight in meters squared (kg/m2)] either lower orhigher than the accepted normal range (20–25) wasassociated with a significantly increased risk of earlyAMD (66). Low BMI (OR, 1.92; 95% CI, 1.16–3.18)conferred an increased risk for early AMD almostequal to that of obesity (OR, 1.78; 95% CI, 1.19–2.68).Although the ORs were similar for association withlate AMD, they did not reach statistical significance.This finding is similar to that of the Physicians’ HealthStudy, which also found a J- or U-shaped associationbetween BMI and the incidence of visually significantAMD, with the highest incidence among obese menwith a BMIR30 and a somewhat less elevated inci-dence among the leanest men with a BMI!22 (94).This association is difficult to explain in terms ofcardiovascular risk. A Finnish population-basedstudy found that a high BMI was associated with anincreased risk of AMD in men but not in women (73).On the other hand, the Beaver Dam Eye Study foundthat BMI was associated with increased frequency ofRPE degeneration, increased retinal pigment, andincreased presence of pigmentary abnormalities inwomen but not in men (50). No association betweenBMI and AMD was found in the Atherosclerosis RiskIn Communities Study (79) or the Andhra Pradesh EyeDisease Study (86).

Seddon et al. found that persons with higherBMI had increased risk for progression to advancedAMD. The RR was 2.35 (95% CI, 1.27–4.34) forBMIR30, and 2.32 (95% CI, 1.32–4.07) for BMI 25 to29 relative to the lowest category (BMI!25) aftercontrolling for other factors (pZ0.007 for trend)(180). In addition, the authors also found that higherwaist circumference was associated with a twofold

increased risk for progression of AMD (RR for thehighest tertile compared with the lowest, 2.04; 95% CI,1.12–3.72), with a significant trend for increasing riskwith a greater waist circumference (pZ0.02). Higherwaist–hip ratio also increased the risk for progressionof AMD (RR, 1.84; 95% CI, 1.07–3.15) for the highesttertile compared with lowest (pZ0.02).

Hematologic Factors and OtherCardiovascular BiomarkersThe Beaver Dam Eye Study found that, after control-ling for age, sex, diabetes, and smoking history,neovascular AMD was associated with higherhematocrit values (OR, 1.09; 95% CI, 1.00–1.19) andhigher leukocyte count (OR, 1.10; 95% CI, 1.00–1.19) inpeopleR65 years (50). Blumenkranz et al. also found ahigher leukocyte count in caseswith neovascular AMDcompared with controls (14). No association betweenhematocrit and AMD was found in NHANES-I (8).

The BlueMountains Eye Study found that plasmafibrinogen level was associated with late but not earlyAMD (66). The Eye Disease Case–Control Study founda nonsignificant increased risk of neovascular AMDwith increasing plasma fibrinogen levels (7).

A number of inflammatory biomarkers which areknown to be associated with cardiovascular diseasehave now been found to be independently associatedwith the progression of AMD (181). These includeC-reactive protein (26,181) and interleukin 6 (181).

Chlamydia pneumoniae InfectionChronic inflammatory events have recently been ident-ified as plausible causes of atherosclerosis and muchinterest has been focused on infections by Chlamydiapneumoniae. C. pneumoniae can multiply in varioushost cells including macrophages and endothelialcells. Like a parasite, the obligate intracellular prokar-yote consumes energy that is needed by the host cells,and in the end, destroys them and then infects nearbycells. Thus, the hallmark of chlamydial disease ispersistent infection and chronic inflammation.

There is strong evidence indicating a closeinteraction between C. pneumoniae and systemicvascular diseases, including the direct detection ofC. pneumoniae (182,183) and the heat shock proteinsof C. pneumoniae (184) in the plaques of coronary andcarotid arteries.

Recent sero-epidemiologic data suggest thatC. pneumoniae infection is associated with AMD.Case–control studies have shown that patients withAMD were more likely to have higher levels of anti-C. pneumoniae antibodies compared with patients withAMD (25,111). Although the significance of theincreased titers of specific IgG and IgA antibodiesagainst C. pneumoniae is not fully understood, higherIgG and IgA antibodies titers may indicate an


exposure to greater amounts of C. pneumoniae andrecurrent or chronic infections.

Remarkably, C. pneumoniae has been detected infour out of nine AMD CNV by immunohistochemistryand two out of nine AMD CNV by polymerase chainreaction (185). In contrast, none of 22 non-AMD speci-mens including 5 non-AMD CNV showed evidencefor C. pneumoniae. These data indicate that a pathogencapable of inducing chronic inflammation can bedetected in some AMD CNV, and support the theorythat infection may contribute to the pathogenesisof AMD.

The Cardiovascular Health and Age-RelatedMaculopathy Study in Australia recently showedthat the rate of progression of AMD over a seven-year period was increased in those with higher titers ofanti-C. pneumoniae antibodies, after controlling for age,smoking, family history of AMD, and history ofcardiovascular diseases (186). Subjects in the twoupper tertiles of antibody titer were at a significantlygreater risk of AMD progression than those in thelowest tertile. A twofold increased risk of AMDprogression for subjects in the middle tertile ofantibody titers was consistent for three differentdefinitions of AMD progression. In the upper tertileof antibody titers, the risk of progression was 2.07(95% CI, 0.92–4.69), 2.58 (95% CI, 1.24–5.41), and 3.05(95% CI, 1.46–6.37) using different definitions of AMDprogression.

Cigarette SmokingThis will be discussed under environmental factors(see below).

Reproductive and Related FactorsThe relationship of cardiovascular disease to AMD hasgenerated some interest in the effect of estrogen-related variables on the risk of AMD in women. TheEye Disease Case–Control Study found that use ofpostmenopausal exogenous estrogen was negativelyassociated with neovascular AMD (7). Current andformer users of estrogen had ORs of 0.3 (95% CI,0.1–0.8) and 0.6 (95% CI, 0.3–0.98) for neovascularAMD, respectively, when compared with womenwho never used estrogen. This is compatible withfindings from a nested case–control study within theRotterdam Study which suggest that early artificialmenopause increases the risk of late AMD (atrophic orneovascular AMD) (19). Women with early meno-pause after unilateral or bilateral oophorectomieshad an increased risk of late AMD compared withwomen who had their menopause at 45 years or later.No significant excess risk was found for early spon-taneous menopause and early hysterectomy. In theBlue Mountains Eye Study, a significant protectiveassociation for early AMD was found with increased

years from menarche to menopause (OR, 0.97; 95% CI,0.95–0.99) (105). Other female-specific factorsincluding late menarche, history of hormone replace-ment therapy, and early menopause were notsignificantly associated with early or late AMD (105).

No significant relationship, however, was foundin the Beaver Dam Eye Study between years ofestrogen therapy and neovascular AMD, geographicatrophy, or early AMD (187). It should be noted thatbecause the number of cases of late AMD in the BeaverDam Eye Study was small, the power to detect a realassociation is limited. Similarly, the PathologiesOculaires Liees a l’Age (POLA) Study did not findany association of hormone replacement therapy,hysterectomy, or oophorectomy with soft drusen,pigmentary abnormalities, or late AMD (77).

Womenwho have ever been pregnant (parityR1)had increased OR of 2.2 (95% CI, 1.3–3.9) comparedwithwomenwhohavenever beenpregnant (parityZ0)in the Eye Disease Case–Control Study (7). On theother hand, the Beaver Dam Eye Study documentedthat the number of past pregnancies was significantlyinversely related to soft drusen, with OR of 0.94 perpregnancy (95% CI, 0.90–0.98) (187). The relationshipwith the number of pregnancies to any AMD was ofborderline significance, the OR being 0.96 per preg-nancy (95% CI, 0.92–1.01). The number of pregnancieswas not significantly related to neovascular AMD orgeographic atrophy. Past use of birth control pills, ageof menarche, or the number of years of menstruationhad not significant effect on AMD in the Beaver DamEye Study (187).

Dermal Elastotic DegenerationIn a small case–control study, Blumenkranz et al.found a correlation between the degree of dermalelastic degeneration in sun-protected skin with thedevelopment of neovascular AMD (14). However,there was no significant difference in outdoor sunexposure as estimated by patients. In fact, casesadmitted to fewer average hour outdoors weeklythan controls. The authors suggested that patientswith neovascular AMD may have a generalizedsystemic disorder characterized by abnormal suscep-tibility of elastic fibers to photic or other as yetunrecognized degenerative stimuli.

Antioxidant EnzymesRecently, the POLA Study, a large-scale population-based cross-sectional study in Southern France, foundthat higher levels of plasma glutathione peroxidasewere significantly associated with a ninefold increasein late AMD prevalence, but not with prevalence ofearly AMD (75). Plasma glutathione peroxidase there-fore appears to be one of the strongest indicators oflate AMD ever found, but the biologic meaning of

64 AU EONG ET AL.

this finding remains to be elucidated. The authorssuggest that oxidative stress may lead to the inductionof antioxidant enzymes, and therefore high concen-trations of antioxidant enzymes may be indicators ofoxidative stress. In the same study, levels of erythro-cyte superoxide dismutase activity were notassociated with either early or late AMD.

ENVIRONMENTAL FACTORS

Cigarette SmokingOf the environmental influences, smoking has mostconsistently been associated with increased risks ofAMD and is the strongest environmental risk factorfor all forms of AMD (7,12,21,27,45,52,64,65,76,82,85,91,92,104,188–190). A group of authors have estimatedthat 28,000 cases of AMD causing visual loss worsethan 20/60 in people R75 years in the UnitedKingdom may be attributable to smoking (104).Another group estimated that 53,900 U.K. residentsolder than 69 years have visual impairment because ofAMD attributable to smoking of whom 17,800 areblind (191).

Paetkau et al. noted in their case series of 114patients with at least one eye blind from AMDthat the mean age at the onset of blindness in thefirst eye was 64 years in current smokers comparedwith 71 years in the group that had never smoked(189). However, because there was no control group,confounding factors such as increased mortality in thesmoking group cannot be excluded. In a Japanesecase–control study, compared with male nonsmokers,the age-adjusted OR of developing neovascular AMDwas 2.97 (95% CI, 1.00–8.84) for male current smokersand 2.09 (95% CI, 0.71–6.13) for male former smokers(21). In addition, smoking habit-related variablessuch as use of extra filter, smoke inhalation level, ageat starting smoking, duration of smoking, and theBrinkman index, defined as the numbers of cigarettesmoked per day times smoking years, were found tobe significantly related to an increased risk of neovas-cular AMD (21).

The Beaver Dam Eye Study found that therelative OR for neovascular AMD in men andwomen who were current smokers compared withthose who were former smokers or who neversmoked were 3.29 (95% CI, 1.03–10.50) and 2.50 (95%CI, 1.01–6.20), respectively (52). However, there was nosignificant relation between smoking status andgeographic atrophy. In addition, smoking status,pack-years smoked, and current exposure to passivesmoking were not associated with signs of early AMD,except for a higher frequency of increased retinalpigment in men who were former smokers comparedwith those who had never smoked (52).

The Blue Mountains Eye Study found currentcigarette smoking to be significantly associated withboth early and late AMD, after adjusting for the effectsof age and sex (65). TheORof early and lateAMDwhencomparing current smokers with those who neversmoked was 1.89 (95% CI, 1.25–2.84) and 4.46 (95%CI, 2.20–9.03), respectively. A history of having eversmokedwas significant for lateAMD (OR, 1.83; 95%CI,1.07–3.13) but not early AMD (65). In addition, passivesmoking among subjects who never themselvessmoked, but lived with a smoking spouse, incurred amoderate but not statistically significant increase inthe risk of late AMD (OR, 1.42; 95% CI, 0.62–3.26). Inthe Genetic Factors in AMD Study, passive smokingexposure was associated with an increased risk of lateAMD (OR, 1.87; 95% CI, 1.03–3.40) (27).

The Blue Mountains Eye Study also disclosedage-standardized five-year incidence rates of earlyAMD at 10.6%, 8.2%, and 9.3%, respectively, amongbaseline current, past, or never smokers (64). Themean age for cases with incident late AMD was67 years for baseline current smokers, 73 years forpast smokers, and 77 years for those who had neversmoked (pZ0.02). After adjusting for age, currentsmokers, compared with never smokers, had anincreased risk of incident geographic atrophy (age-adjusted RR, 3.6; 95% CI, 1.1–11.3) and any late AMDlesions (RR, 2.5; 95% CI, 1.0–6.2).

In the POLA Study, after adjustment for age andsex, current (OR, 3.6; 95% CI, 1.1–12.4) and formersmokers (OR, 3.2; 95% CI, 1.3–7.7) had an increasedprevalence of late AMD when compared withnonsmokers (76). The risk of late AMD increasedwith increasing number of pack-years, with up to a5.2-fold increase in risk among participants (currentand former smokers combined) who smoked 40 pack-years or more (OR 1.9, 95% CI 0.6–6.4 for 1–19 pack-years; OR 3.0, 95% CI 0.9–9.5 for 20–39 pack-years; andOR 5.2, 95% CI 2.0–13.6 for 40 pack-years and more).In addition, the risk of late AMD remained increaseduntil 20 years after cessation of smoking. Anothertwo studies from the United Kingdom also foundthat the risk in those who had stopped smoking forover 20 years was comparable to nonsmokers (27,104).

The Los Angeles Latino Eye Study also disclosedthat having ever smoked was associated with a higherrisk of having late AMD (OR, 2.4; 95% CI, 1.03–5.4)(85). The strength of association is confirmed in apooled analysis of data from three cross-sectionalstudies (Blue Mountains Eye Study, Beaver Dam EyeStudy, Rotterdam Study), totaling 12,468 participants,in which current smokers had a significant three- tofour-fold increased age-adjusted risk of AMDcompared with never smokers (192). A latter analysisof pooled data from the same three studies also foundcurrent smoking to be associated with an increased


incidence of late AMD (OR relative to nonsmokers,2.35; 95% CI, 1.30–4.27) (169).

Two large prospective cohort studies evaluatedthe relationship between smoking and AMD (91,92).In the Nurses’ Health Study with 12 years of follow-up, women who currently smoked R25 cigarettesper day had a RR of AMD of 2.4 (95% CI, 1.4–4.0)compared with women who never smoked (91). Riskof AMD also increased with an increasing number ofpack-years smoked (p for trend!0.001). Past smokersof this amount also had a RR of 2.0 (95% CI, 1.2–3.4)comparedwith womenwho never smoked. Comparedwith current smokers, little reduction in risk wasfound even after quitting smoking for 15 or moreyears. In the Physicians’ Health Study, men whowere current smokers of R20 cigarettes per day hada RR of AMD of 2.5 (95% CI, 1.6–3.8) compared withmen who never smoked (92). Men who were pastsmokers had a modest elevation in RR of AMD of1.3 (95% CI, 1.0–1.7).

Some investigators have suggested that the effectof cigarette smoking on the development of AMDmaybe related to its effect on antioxidants in the body (21).Studies have shown that smokers have much lowerplasma levels of b-carotene than do nonsmokers(172,193). Stryker et al. found that men and womenwho smoked one pack per day had 72% (95% CI,58–89) and 79% (95% CI, 64–99) of the plasmab-carotene levels of nonsmokers, respectively, afteraccounting for dietary carotene and other variables(193). Another study also disclosed that smokers hadlower plasma concentrations of total carotenoids,a-carotene, and b-carotene than nonsmokers (172). Inaddition, smokers have significantly lower macularpigment optical density compared with nonsmoking-matched controls (143). The macular pigment opticaldensity and smoking frequency are inversely relatedin a dose–response relationship.

In an experimental study using mice, exposure tocigarette smoke or the smoke-related potent oxidanthydroquinone results in the formation of sub-RPEdeposits, thickening of Bruch’s membrane, andaccumulation of deposits within Bruch’s membrane(194). In another study, nicotine has been found toincrease the size and severity of experimental CNV ina mouse model, with older mice being more affectedthan younger mice (195). Interestingly, the effects ofnicotine on the CNV lesions were reversed withconcurrent subconjunctival administration of hexam-ethonium, a nonspecific nicotinic receptor antagonistwhich could counteract the effects of nicotine.

Despite the strong association between smokingand AMD, the awareness of blindness as anothersmoking-related condition is low. In a recent cross-sectional survey of 358 adult patients (both smokersand nonsmokers) attending a district general hospital

in the United Kingdom, only 9.5% of patients believedthat smoking was definitely or probably a cause ofblindness, compared with 92.2% for lung cancer,87.6% for heart disease, and 70.6% for stroke (196).Although there was a disparity in the knowledge ofthese smoking-related conditions, about half of thesmokers stated that they would definitely or probablyquit smoking if they developed early signs of blind-ness and the other three conditions, with no significantdifferences in the proportions for these four conditions.Increasing the awareness of the link between smokingand blindness may therefore be an effective additionalapproach to encouraging smoking cessation.

A small number of studies (10,14,15,22,73),including the Framingham Eye Study (29) andNHANES-III (80), did not find an association betweensmoking and AMD. In fact, one study by West et al.even showed smoking to be protective (39). However,when this decreased risk of AMD associated withsmoking was further investigated, no clear dose–response relationship was demonstrated. In the largecase–control study by the FRANCE-DMLA StudyGroup, a past history of smoking, but not currentsmoking status, was associated with an increased riskof AMD after univariate analysis (22). After multi-variate adjustment, both factors were not significantlyassociated with AMD.

In summary, data from several large population-based studies (52,65,76,85,190), case–control studies(7,12,21), and two large prospective cohort studies(91,92) provide convincing evidence that cigarettesmoking is a risk factor for AMD. The strongest riskis for current smokers, suggesting that there may bepotential benefits of targeting antismoking patienteducation, especially for those who are currentsmokers and have signs of early AMD (65). Thebenefit of stopping smoking is seen after 10 yearswith reductions in risk although the risks do notreturn to that of never smokers until 20 years afterstopping smoking (27,104).

Sunlight ExposureIt is well established that ultraviolet (UV) and visibleradiation has the potential to damage the retina andRPE (197,198). Fortunately, the human retina isprotected from short-wavelength radiation, which isparticular damaging, by the cornea which absorbsbelow 295 nm and the lens which absorbs stronglybelow 400 nm (199). The human retina is thereforeonly exposed to the “visible component” of the electro-magnetic spectrum from 400 to 760 nm and someshorter wavelength infrared. This part of the electro-magnetic spectrum may result in chronic or acutetissue damage when it is absorbed by any one ofa number of photosensitisers or chromophores,

66 AU EONG ET AL.

e.g., the visual pigments, melanin, melanopsin, lipo-fuscin, flavins, and flavoproteins (199).

There are some similarities between long-termchanges seen in laboratory animals exposed to shorterwavelength visible light and changes seen in patientswith AMD (133,200–205). It is theorized that light maylead to the generation of reactive oxygen species inthe outer retina and/or choroid (133), perhaps byphotoactivation of protoporhyrin (206). The activatedforms of oxygen may, in turn, cause lipid peroxidationof the photoreceptor outer segment membranes,leading to the development of AMD.

Tso and Woodford have shown that shortexposure of intense visible light can produce atrophyat the photoreceptor level in nonhuman primates(207), but these animals did not develop histopatho-logic changes of drusen, diffuse thickening of Bruch’smembrane, or CNV seen in clinicopathologic studiesof AMD (208). In addition, the short intense lightexposure used in animal studies is different from thetypical chronic exposure to light that occurs in peoplein their lifetime. The only animal model for light-induced deposits in Bruch’s membrane is that ofGottsch et al. who have proposed that photosensitiza-tion of choriocapillary endothelium with blood-bornephotosensitizers, such as photoporphyrin IX, is amechanism for the histopathologic features seen inAMD (206,209).

The epidemiologic evidence of an associationbetween light exposure and AMD is lacking, withonly a few clinical studies showing a positive associ-ation between sun exposure and late AMD. A smallSpanish case–control study found a higher sunexposure index in AMD cases compared with controls(23). In the Chesapeake Bay Watermen Study, anassociation between late AMD (geographic atrophyor disciform scarring) and ocular exposure in theprevious 20 years to blue or visible light (OR, 1.36;95% CI, 1.00–1.85) was found in phakic men (41).However, no positive association was seen for earlyAMD (large drusen or RPE abnormalities) (41). Inaddition, there was no association between UV-A orUV-B exposure and any degree of AMD in the samepopulation (39,41).

The Beaver Dam Eye Study found that leisuretime outdoors in summer was significantly associatedwith the presence of neovascular AMD when bothmen and women were analyzed together (OR, 2.26;95% CI, 1.06–4.81) (53). Time spent outdoors insummer was significantly associated with the preva-lence of increased retinal pigment in men (OR, 1.44;95% CI, 1.01–2.04) but not in women (OR, 0.93; 95%CI, 0.63–1.38). Use of sunglasses and hats with brimswas inversely associated with the prevalence of softindistinct drusen in men (OR, 0.61; 95% CI, 0.38–0.98)but not in women (OR, 0.99; 95% CI, 0.69–1.45). The

association between light exposure and AMD is notconsistent across the study, since an association wasfound in men only and involves only a specific subsetof light exposure (time spent outdoors in summer butnot in winter) and a specific subset of early AMD (53).

Another analysis from the Beaver Dam EyeStudy to investigate the relation of sunlight exposureand indicators of sun sensitivity with the five-yearincidence of early AMD showed that leisure timespent outdoors while person were teenagers (aged13–19 years) and in their 30s (aged 30–39 years) wassignificantly associated with the risk of early AMD(OR, 2.09; 95% CI, 1.19–3.65) (54). However, there wereno association between estimated ambient UV-Bexposure or markers of sun sensitivity and the inci-dence of early AMD.

A number of case–control studies, including theEye Disease Case–Control Study (7), failed to show anassociation between sunlight exposure and AMD(12,20). An Australian case–control study in factshowed that control subjects had greater medianannual ocular sun exposure (865 hours) than cases(723 hours) (pO0.0001) (20). Despite the analysis stra-tified by sun sensitivity, sun exposure was greater incontrol subjects than in cases with AMD (20).

Margrain and colleagues have suggested that theequivocal findings reported in epidemiologic studiesare quite unremarkable because firstly, the absence of arelationship of AMD with UV exposure simplyconfirms that the adult lens absorbs almost all radi-ation below 400 nm (199). Secondly, they suggestedthat the assumption that it is lifetime exposure tosunlight that is the relevant variable is probablyincorrect. Instead, they suggested that the phototoxi-city of blue light increases with age and is likely to beparticularly great for those with lipofuscin “hot spots.”

In summary, there is currently no convincing datato support strategies to reduce light exposure to the eyefor the prevention of AMD. It would be premature torecommend thewidespread use of blue-blocking intra-ocular lens during cataract surgery in the elderlybecause although there is considerable circumstantialevidence for such ameasure, there is no direct evidencethat environmental light causes retinal damage (199).However, there are now compelling reasons for under-taking a large-scale clinical trail to evaluate theprophylactic effects of blue light filtration in AMD. Inaddition, since there is little, if any, risk to a personwearing sunglasses, and UV light exposure has beenassociated with the presence of cataract (153), it isreasonable to suggest that individuals wear sunglassesfor comfort and to reduce exposure of UV light toocular structures. It must be emphasized, however,that there is no published data to indicate whetherthe wearing of sunglasses is of any benefit inpreventing any eye disease, including AMD.


Nutritional Factors

MicronutrientsThe potential role of nutritional supplement to reducethe incidence or severity of AMD has received a greatdeal of attention (132,150). The lack of an effectivetreatment for the majority of cases of AMD, coupledwith the public’s perception that over-the-counternutritional supplements are relatively harmless,creates the potential for widespread use of thesesupplements in the absence of demonstrated effective-ness (210). Because of a possible, but as yet unproven,benefits of antioxidant vitamins in cancer, cardiovas-cular, and other chronic diseases, vitamin supplementusage in the United States has increased steadily inrecent years. It is estimated that more than half of theadult population in the United States uses dietarysupplements, including supplements of antioxidantvitamins, at a cost of approximately $12 billionannually (102).

Although epidemiologic studies provide supportfor a protective role of nutritional antioxidants in theprevention of AMD, results of prospective randomizedclinical trials are necessary before firm conclusions canbe drawn about the balance of benefits and risks ofnutritional supplements for the prevention of AMD.In fact, use of nutritional supplement has been shownto have deleterious effects in some nonophthalmicmedical trials. The Alpha-Tocopherol, Beta-Carotene(ATBC) Cancer Prevention Study found a higherincidence of lung cancer among men who receivedb-carotene than among those who did not (change inincidence, 18%; 95% CI, 3–36%) (211). There were alsomore deaths due to lung cancer, ischemic heart disease,and ischemic and hemorrhagic stroke among recipi-ents of b-carotene, with an increased overall mortalityof 8% (95% CI, 1–16%). Those randomized to vitamin Esupplementation had higher rates of hemorrhagicstroke, but there was no overall difference in mortalityrates or cancer incidence (211). In the Carotene andRetinol Efficacy Trial, participants who were givenb-carotene and vitamin A supplements had a 28%(95% CI, 4–57%) increased incidence of lung cancerand a 17% (95% CI, 3–33%) higher mortality comparedwith those who were not (212).

AntioxidantsSome have suggested that supplementation withantioxidants and a variety of trace mineralsnecessary for the proper functioning of some keyenzyme systems may reduce the risk of AMD(133,213,214). Photochemical damage from light caninduce the production of activated forms of oxygen,which in turn can cause lipid peroxidation of thephotoreceptor outer segment membranes. Antioxi-dants, such as vitamin C, vitamin E, b-carotene, and

glutathione, and antioxidant enzymes, such asselenium-dependent glutathione peroxidase, intheory could act as singlet oxygen and free radicalscavengers and thereby prevent cellular damage(215). There is considerable interest in determining iffree radicals contribute to the pathogenesis of AMDand if high levels of these antioxidants may protectagainst AMD. This hypothesis is supported by findingsof disruption of retinal photoreceptors in nonhumanprimates with deficiencies of vitamins A and E (216)and a protective effect of vitamin C in reducing the lossof rhodopsin and photoreceptor cell nuclei in ratsexposed to photic injury (217).

Many studies have used serum levels of micro-nutrients to investigate the relationship of thesemicronutrients and AMD. Unfortunately, high andlow levels are defined differently for most studies.Most have defined the high and low categories onthe basis of percentile categories, i.e., those individualswith serum concentrations above a given percentilewere categorized as high and those below a givenpercentile were categorized as low.

Blumenkranz et al. reported in their small case–control study that the serum levels of vitamins A, C,and E were not different in cases of neovascular AMDand in controls (14). In another case–control study,serum levels of vitamin E in cases and controls weresimilar but serum selenium was significantly lower incases compared with controls (pZ0.02) (15). The EyeDisease Case–Control Study found that persons withcarotenoid scores in the medium and high percentilegroups, compared with those in the low group, hadmarkedly reduced levels of risk of neovascular AMD,with levels of risk reduced to one-half (OR, 0.5; 95%CI, 0.4–0.8) and one-third (OR, 0.3; 95% CI, 0.2–0.6),respectively (6). Similarly, except for lycopene, higherlevels of individual carotenoids (lutein/zeaxanthin,b-carotene, a-carotene, or cryptoxanthin) were associ-ated with statistically significant reductions in risk ofneovascular AMD. In addition, there was a pro-gressive decrease in risk of neovascular AMD withincreasing levels of the carotenoids and increasinglevels of the antioxidant index. However, no statisti-cally significant overall association was seen withneovascular AMD and serum levels of vitamin C,Vitamin E, and selenium in the study (6).

West et al. examined the relationship betweenplasma levels of retinol, ascorbic acid, a-tocopherol,and b-carotene in 630 participants of the BaltimoreLongitudinal Study on Aging (46). They found afavorable association between plasma antioxidantsand AMD. Their data suggest that only a-tocopherolwas significantly associated with a protectiveeffect (OR for middle vs. lowest quartiles 0.50, 95%CI 0.32–0.79; OR for highest vs. lowest quartiles 0.43,95%CI 0.25–0.73). This is consistent with findings from

68 AU EONG ET AL.

a small Spanish case–control study (23). There was asuggestion of a protective effect with ascorbic acid andb-carotene in the Baltimore Longitudinal Study onAging, but their effects were not statistically significant(46). No protective effect was noted for retinol. Forlate AMD (neovascular AMD or geographic atrophy),no significant protective effect was observed forany plasma micronutrient. An antioxidant index con-structed of ascorbic acid, a-tocopherol, and b-carotene,controlled for age and sex, suggested that high valueswere protective for AMD compared with low values.

It is now generally recognized that plasmaa-tocopherol level should be expressed in terms ofits concentration within lipids or lipoproteins(218–220). For this reason, the POLA Study correlatedocular findings with both plasma a-tocopherol andlipid-standardized a-tocopherol levels (74). The studyfound a weak negative association between late AMDand plasma a-tocopherol level which was not statisti-cally significant (pZ0.07) but this relationship wasstrengthened when a-tocopherol–lipid ratio insteadof plasma level was used (pZ0.003). After adjustingfor confounding factors, the ORs (95% CI) for lateAMD in persons with a-tocopherol–lipid ratio in thehighest and middle quintiles, compared with thosewith ratio in the lowest quintile, were 0.18 (0.05–0.67)and 0.46 (0.22–0.95), respectively. The ORs (95% CI) forany sign of early AMD in persons with a-tocopherol–lipid ratio in the highest and middle quintiles,compared with those with ratio in the lowestquintile, were 0.72 (0.53–0.98) and 0.78 (0.61–1.00),respectively. No association was found with plasmaretinol and ascorbic acid levels or with red blood cellglutathione values (74).

Data from NHANES-I, collected between 1971and 1972, suggest that the frequency of consumptionof fruits and vegetables characterized as rich invitamin A is inversely related to the prevalence ofAMD, after adjustment for medical and demographicfactors (8). This concurs with the Nurses’ HealthStudy and the Health Professionals Follow-UpStudy which showed that fruit intake was inverselyassociated with the risk of neovascular AMD (95).Participants from the two studies who consumedthree or more servings per day of fruits had apooled multivariate RR of 0.64 (95% CI, 0.44–0.93)compared with those who consumed less than 1.5servings per day. The Eye Disease Case–ControlStudy evaluated the relationship of dietary intake ofcarotenoids, and vitamins A, C, and E, with neovas-cular AMD (16). Those in the highest quintile ofcarotenoid intake, after adjusting for other riskfactors of AMD, had an OR of 0.57 (95% CI, 0.35–0.92)for neovascular AMD compared with those in thelowest quintile. Among the specific carotenoids, thestrongest association with a reduced risk for

neovascular AMD was found with lutein and zeax-anthin, which are primarily obtained from dark green,leafy vegetables. Intake of vitamin C was associatedwith a small but nonsignificant reduction in risk ofneovascularAMD.No reduction in riskwas foundwithintake of vitamin A or E.

The Rotterdam Study, using a 170-item semi-quantitative food frequency questionnaire, found asignificant inverse association for intake of vitamin Eand incident AMD (221). After adjustment, a onestandard deviation increase in intake of vitamin Ewas associated with a reduced risk of AMD of 8%(95% CI, 0–16%). The risk of AMD by quartile ofnutrient intake also indicated a dose–responserelationship between vitamin E and reduced risk ofAMD (p value for trend Z0.04). The authors in theRotterdam Study also estimated the impact of thecombined dietary intake of the four nutrients(b-carotene, vitamins C and E, and zinc) that werestudied in AREDS (see below) (222). It should,however, be pointed out that the intake of thesenutrients in the Rotterdam Study was considerablylower than the high-dose supplements used inAREDS. An above-median intake of the four nutrientscomparedwith a below-median intake of at least one ofthese nutrients, was associated with a reduced risk ofAMD [hazard ratio (HR), 0.65; 95% CI, 0.46–0.92]adjusted for all potential confounders. In personswith a below-median intake of all four nutrients, therisk of AMD was increased but not significantly so(HR, 1.20; 95% CI, 0.92–1.56).

The Blue Mountains Eye Study, using a validated145-item semiquantitative food frequency question-naire, found no significant associations between earlyor late AMD and dietary intakes of carotene, vitaminA, or vitamin C, from combined diet and supplement,after adjusting for age, sex, current smoking, andAMD family history (69). There were no statisticallysignificant trends for decreasing AMD prevalencewith increasing intake of any antioxidant. Consump-tion of supplements was also not significantlyassociated with either early (OR, 1.0; 95% CI, 0.7–1.4)or late (OR, 1.2; 95% CI, 0.6–2.3) AMD. In addition, anested case–control study within the Blue MountainsEye Study did not find any association between AMDor serum a-tocopherol or b-carotene (70). Similarly, nosignificant associations between the intake of vitaminC or E, or carotenoids from the diet or supplementsand the prevalence of early or late AMD wereobserved in the Beaver Dam Eye Study (223).However, in a nested case–control study within theBeaver Dam Eye Study population-based cohort, lowlevels of serum lycopene, but not other carotenoids(a-carotene, b-carotene, b-cryptoxanthin, or lutein andzeaxanthin), was related to an increased likelihoodof AMD (OR, 2.2; 95% CI, 1.1–4.5) (18).


The association between self-selection for anti-oxidant vitamin supplement use and incidence ofAMD was examined among 21,120 participants inthe Physicians’ Health Study I who did not have adiagnosis of AMD at baseline (9). A total of 279incident cases of AMD with vision loss to 20/30 orworse were confirmed during an average follow-up of12.5 years. Compared to nonusers of vitamin supple-ments, persons who reported taking vitamin Esupplements at baseline had a nonsignificant 13%reduced risk of AMD (RR, 0.87; 95% CI, 0.53–1.43),after adjusting for other risk factors. Users of multi-vitamins had a nonsignificant 10% reduced risk ofAMD (RR, 0.90; 95% CI, 0.68–1.19). No reduced riskof AMD was observed for users of vitamin C supple-ments (RR, 1.03; 95% CI, 0.71–1.50).

ZincZinc has received attention because of its high concen-tration in ocular tissues, particularly the sensoryretina, RPE, and choroid (224) and its role as a cofactorfor numerous metalloenzymes, including retinoldehydrogenase and catalase (225). In addition, thereare some reports of zinc deficiency in the elderly, thepopulation subgroup at greatest risk of AMD (226).Data from NHANES-III suggest that persons agedR71 years, together with young children aged 1 to 3years and adolescent females aged 12 to 19 years, wereat the greatest risk of inadequate zinc intakes (227). Ithas been hypothesized that zinc deficiency in elderlypersons may cause the loss of zinc-dependent coen-zymes in the RPE, resulting in the development orworsening of AMD (228).

Newsome et al. conducted a prospective, ran-domized, double-blind, placebo-controlled trial thatinvestigated the effects of oral zinc administration onthe visual acuity outcome in 151 subjects with early tolate AMD (229). They showed that eyes in zinc-treatedgrouphad significantly less visual loss than the placebogroup after a follow-up of 12 to 24 months. In addition,there was less accumulation of drusen in the zinc-treated group compared with the placebo group.However, in another double-masked, randomized,placebo-controlled trial, oral zinc supplementationdid not have any short-term effect on the course ofAMD in patients who have neovascular AMD in oneeye (230).

The Beaver Dam Eye Study found that people inthe highest quintile, comparedwith those in the lowestquintile, for intake of zinc from foods had lower risk ofearly AMD (OR, 0.6; 95% CI, 0.4–1.0) (223). This isconsistent with the Rotterdam Study which showed asignificant inverse association between zinc intake andincident AMD (221). After adjustment, a one standarddeviation increase in intake of zinc was associatedwith a reduced risk of incident AMD of 9% (95% CI,

2–17%). The risk of AMD by quartile of zinc intakealso showed a dose–response relationship betweenzinc intake and reduced risk of AMD (p value fortrend Z0.06). A lower serum level of zinc was foundin AMD cases compared with controls in a smallSpanish case–control study (23). However, zincintake was unrelated to late AMD in the same study.The Eye Disease Case–Control Study did not find anysignificant relationships between serum zinc levels orzinc supplementation and risk of neovascularAMD(7).This concurs with findings from the Blue MountainsEye Study (69). Two large prospective studies, theNurses’ Health Survey, and the Health ProfessionalsFollow-up Study, also concluded that moderate zincintake, either in food or in supplements, was notassociated with a reduced risk of AMD (96).

Randomized Trials of AntioxidantVitamins and AMDThe most reliable, and only direct, method of testingthe potential protective effects of nutritional supple-ments is to conduct randomized clinical trials. A smallprospective randomized clinical trial showed that aspecific 14-component antioxidant-mineral capsule(Ocuguardw, Twin Lab, Inc., Ronkonkoma, NewYork, U.S.A.) taken twice daily stabilized but didnot improve dry AMD over one-and-a-half years(231,232). Several large-scale randomized clinicaltrials, including AREDS (210,222), the Physicians’Health Study II (233), the Vitamin E, Cataract, andAge-related macular degeneration Trial (VECAT)(234,235), the Women’s Health Study (236), and theWomen’s Antioxidant Cardiovascular Study (237),have been designed to address the issue of antioxidantvitamins and AMD (Table 7). Results of these majortrials should provide the strongest evidence to supportor to refute an association of antioxidant intake withAMD. Of these trials, AREDS (222), sponsored by theNational Eye Institute (National Institutes of Health,Bethesda, Maryland, U.S.A.), and VECAT (235) havebeen completed.

The AREDS is an 11-center double-maskedclinical trial that randomly assigned participants toreceive oral total daily supplementation of (i) anti-oxidants (vitamin C, 500 mg; vitamin E, 400 IU; andb-carotene, 15 mg); (ii) zinc (zinc, 80 mg as zinc oxide,and copper, 2 mg as cupric oxide to prevent potentialanemia); (iii) antioxidants plus zinc; or (iv) placebo(222). Participants from aged 55 to 80 years wereenrolled from November 1992 through January 1998and followed-up until April 2001. Enrolled partici-pants in the AREDS AMD trial had extensive[drusen area R125 mm diameter circle (about 1/150disc area)] small (!63 mm) drusen, intermediate(63–124 mm) drusen, large (R125 mm) drusen, noncen-tral geographic atrophy, or pigment abnormalities in

70 AU EONG ET AL.

one or both eyes, or advanced AMD or vision lossdue to AMD in one eye. At least one eye had a best-corrected visual acuity of 20/32 or better [the studyeye(s)].

The average follow-up of the 3640 enrolledstudy participants in the AREDS AMD trial was 6.3years, with 2.4% lost to follow-up. Compared withpatients receiving placebo, patients randomized tosupplementation with antioxidants plus zinc hada statistically significant odds reduction for thedevelopment of advanced AMD (OR, 0.72; 99% CI,0.52–0.98). Advanced AMD was defined as photocoa-gulation or other treatment for CNV, or photographicdocumentation of any of the following: geographicatrophy involving the center of the macula, nondruse-noid RPE detachment, serous or hemorrhagicretinal detachment, hemorrhage under the retina orRPE, and/or subretinal fibrosis. The ORs for zinc aloneand antioxidants alone are 0.75 (99% CI, 0.55–1.03) and0.80 (99% CI, 0.59–1.09), respectively. The study foundthat participants with extensive small drusen, non-extensive intermediate size drusen, or pigmentabnormalities had only a 1.3% five-year probability ofprogression to advanced AMD. There was no evidenceof any treatment benefit in delaying the progression ofthese patients to more severe drusen pathology. When

these 1063 participants were excluded and analysisperformed for the rest of the participants who hadmore severe age-related macular features {extensive[drusen area R360 mm diameter circle (about 1/16disc area) if soft indistinct drusen are present ordrusen area R656 mm diameter circle (about 1/5 discarea) if soft indistinct drusen are absent] intermediatedrusen, largedrusen, or noncentral geographic atrophyin one or both eyes, or advanced AMD or vision loss[best-corrected visual acuity !20/32] due to AMD inone eye} andwho are at the highest risk for progressionto advanced AMD, the odds reduction estimatesincreased (antioxidants plus zinc: OR 0.66, 99% CI0.47–0.91; zinc: OR 0.71; 99%CI 0.52–0.99; antioxidants:OR 0.76; 99% CI 0.55–1.05). Estimates of RRs derivedfrom theORs suggested risk reductions for those takingantioxidants plus zinc, zinc alone, and antioxidantsalone of 25%, 21%, and 17%, respectively. Both anti-oxidants plus zinc and zinc significantly reduced theOR of developing advanced AMD in this higher riskgroup. However, the only statistically significantreduction in rates of at least moderate vision loss[defined as decrease in best-corrected visual acuityscore from baseline of R15 letters in a study eye(equivalent to a doubling or more of the initial visualangle, e.g., 20/20 to 20/40 or worse, or 20/50 to 20/100

Table 7 Some Large-Scale Randomized Trials Addressing the Balance of Risks and Benefits of Antioxidant Vitamins for Age-Related Macular Degeneration

Name of randomized trial Details of trial Remarks

Age-Related Eye Disease

Study (210,228)

A multicenter prospective, double-blind, randomized clinical trial

evaluating the role of antioxidant micronutrients (b-carotene,

vitamins E and C, and/or zinc) in AMD and cataract. Patients

with early AMD to advanced unilateral AMD were randomized

to receive antioxidant vitamins, zinc, combination therapy, or

placebo. Four thousand, seven hundred and fifty-seven

individuals aged 55–80 years at baseline were enrolled.

Morbidity and mortality associated with the supplements were

monitored. Endpoints include doubling of visual angle and

morphologic progression of AMD

Sponsored by the National Eye Institute

of the National Institutes of Health.

Trial was completed in 2001

Physicians’ Health Study II

(PHS II) (233)

A randomized, double-blind, placebo-controlled trial enrolling

15,000 willing and eligible physiciansR55 years. It will testalternate day b-carotene, alternate day vitamin E, daily vitamin

C, and a daily multivitamin, in the prevention of AMD as well as

cataract, total and prostate cancer, and cardiovascular disease

PHS II is sponsored by BASF AG.

Approximately half of the PHS II

cohort comprises participants of the

PHS I cohort which was sponsored by

the National Institutes of Health

Vitamin E, Cataract, and

Age-Related Macular

Degeneration Trial

(234,235)

A four-year randomized, placebo-controlled, double-masked trial

of vitamin E on the rate of progression of cataract and AMD in

1193 elderly Australian volunteers

Sponsored by the National Health and

Medical Research Council of

Australia and other sources

Women’s Health Study

(102,236)

A randomized, double-blind, placebo-controlled trial of vitamin E

and low-dose aspirin in the prevention of cancer and

cardiovascular disease among 39,876 apparently healthy,

postmenopausal U.S. female health professionals

Has been funded by the National Eye

Institute to extend its investigation to

include AMD and cataract

Women’s Antioxidant

Cardiovascular Study

(102,237)

A randomized, double-blind, placebo-controlled, secondary

prevention trial to test antioxidant vitamins (b-carotene, vitamin

C, vitamin E), and a combination of folate, vitamin B6, and

vitamin B12, among 8171 female health professionals, aged 40

or older, who are at high risk for cardiovascular disease

Has been funded by the National Eye

Institute to extend its investigation to

include AMD and cataract

Abbreviations: AMD, age-related macular degeneration; PHS, Physicians’ Health Study.


or worse)] occurred in persons randomized to receiveantioxidants plus zinc (OR, 0.73; 99% CI, 0.54–0.99) inthis same group. The estimated 27% odds reduction ofat least moderate vision loss for the combination arm(antioxidants plus zinc)may be the combined benefit ofthe zinc component (odds reduction of 17%) and theantioxidant component (odds reduction of 15%). Therewas no statistically significant serious adverse effectassociated with any of the formulations.

The study recommended that persons older than55 years should have dilated eye examinations todetermine their risk of developing advanced AMD.Those with extensive intermediate size drusen, atleast one large druse, noncentral geographic atrophyin one or both eyes, or advanced AMD or vision lossdue to AMD in one eye, and without contraindicationssuch as smoking, should consider taking a supplementof antioxidants plus zinc to reduce their risk ofprogression to advanced AMD and vision loss.Because results from two other randomized clinicaltrials suggested increased risk of mortality amongsmokers supplementing with b-carotene (211,212),persons who smoke cigarettes should probably avoidtaking b-carotene, and they might choose to supple-ment with only some of the study ingredients.

It has been estimated that 8 million personsaged R55 years in the United States have monocularor binocular intermediate AMD, or monocularadvanced AMD as defined in AREDS (238). Theyare considered to be at high risk for advancedAMD. Of these people, 1.3 million are expected todevelop advanced AMD if left untreated. It is thoughtthat if all of those at risk of advanced AMD receivedsupplements such as those used in AREDS, more than300,000 (95% CI, 158,000–487,000) of them wouldavoid advanced AMD and any associated visionloss during the next five years.

The VECAT is a prospective randomizedplacebo-controlled clinical trial in Australia involving1193 healthy volunteers aged 55 to 80 years (235). Oneof the major arms of the trial looked at vitamin Esupplementation and incidence and progression ofAMD. Participants were randomized to receiveeither 500 IU natural vitamin E (335 mg D-a toco-pherol) in a soybean oil suspension encapsulated ingelatin or a matched placebo capsule and werefollowed-up for four years. No protective or deleter-ious effect of the daily dietary supplementation wasfound on the incidence or progression of AMD.Secondary analyses of visual acuity and visual func-tion also failed to show an intervention effect.

The lack of a protective effect of vitamin Esupplementation in VECAT could mean that vitaminE does not have an important role in protecting againstAMD (235). However, it is possible that the follow-upof four years in this study was too short and vitamin E

may need to be taken for a longtime to have an effect.The lowered risk of AMD linked with high intakes orblood levels of antioxidants in some observationalstudies could reflect a lifelong pattern of eating (239).There may be a longtime lag between the time ofdamage and appearance of clinical signs of AMD.Another possibility is that the baseline antioxidantstatus of the trial participants was too high for supple-mentation to be effective (239). The plasma vitamin Elevels were near the top of the reference range andover 25% of participants had been taking supple-mentary vitamin E prior to the trial. Lastly, the trialwas originally set up with statistical power to detect a15% reduction in cataract. Although the authors statedthat the sample size may have been adequate to detecta 50% reduction in the incidence of AMD, it may havebeen unrealistic to expect vitamin E to have such ahuge effect.

The ATBC Cancer Prevention Study, which tookplace in Finland between 1984 and 1993, was originallydesigned to investigate the efficacy of a-tocopheroland b-carotene in the prevention of lung cancer inover 29,000 smoking males aged 50 to 69 years (211).The participants were randomly assigned to a-toco-pherol (50 mg/day), b-carotene (20 mg/day), both ofthese, or placebo. An end-of-trial ophthalmic exami-nation on a random sample of 941 participants agedR65 years from 2 of the 14 study areas was performedto investigate if the five- to eight-year intervention wasassociated with a difference in the AMD prevalence(240). Although no ophthalmic examination wasperformed at baseline, an equal spread of AMDamong the different treatment groups is assumeddue to randomization. The study found more casesof AMD in the a-tocopherol group (32%; 75/237),b-carotene group (29%; 68/234), and combined anti-oxidant group (28%; 73/257) than in the placebo group(25%; 53/213). However, neither antioxidant was sig-nificantly associated with an increased risk of AMD ina logistic regression analysis controlling for possiblerisk factors.

Dietary Carotenoids Lutein and ZeaxanthinThese have been discussed under the subheadingmacular pigment optical density as one of the ocularrisk factors of AMD (see above).

Dietary Fish IntakeA high proportion of polyunsaturated u-3 fatty acids,particularly docosahexaenoic acid, is present in thehuman retina and macula (213,241). Docosahexaenoicacid appears to play an important role in the normalfunctioning of the retina and is found predominantlyin oily fish and offal (172). Increased consumption offish and fish oils containing u-3 fatty acids has been

72 AU EONG ET AL.

associated with a protective effect against athero-sclerosis in several studies (242–244).

The Blue Mountains Eye Study found that morefrequent consumption of fish appeared to protectagainst late AMD but not early AMD, after adjustingfor age, sex, and smoking (68). The protective effect offish intake for late AMD commenced at a relatively lowfrequency of consumption (OR for intake 1–3 times/month vs. intake !1 time/month, 0.23; 95% CI, 0.08–0.63) and overall had an OR of 0.5. A borderlineprotective effect for consumption of polyunsaturatedfat was also observed (OR for intake in highest vs.lowest quintile, 0.40; 95% CI, 0.14–1.18). This concurswith the finding in the Beaver Dam Eye Study thatincreased consumption of margarine, which containshigher ratios of polyunsaturated to saturated fattyacids, was associated with a reduction in risk forearly AMD (OR for intakes in highest vs. lowestquintile, 0.5; 95% CI, 0.4–0.8) (56). However, intake ofseafood, a marker of intake of u-3 fatty acids, wasunrelated to early or late AMD (56). Sanders et al. alsofound no association between AMD and theproportion of polyunsaturated fatty acids in theplasma and erythrocyte phospholipids in a case–control study (172).

The relation of other dietary fat intake and AMDhas been dealt with under cardiovascular disease riskfactors (see above).

Alcohol ConsumptionObisesan et al. used data from NHANES-I to investi-gate the relationship of alcohol consumption andAMDand found that persons who consumed 12 or fewerdrinks of alcohol per year appear to be less likely todevelop AMD when compared with nondrinkers(4% vs. 7%, respectively), although this was not statisti-cally significant (37). Beer consumption alone did nothave a significant effect on the development of AMD(OR, 0.72; 95% CI, 0.45–1.12). After adjusting for theeffect of age, gender, income, history of congestiveheart failure, and hypertension, wine consumptionshowed a statistically significant negative associationwith AMD (OR, 0.81; 95% CI, 0.67–0.99). In the EyeDisease Case–Control Study, higher alcohol intake wasalso found to be related to a reduced risk of neovas-cular AMD (245). The Andhra Pradesh Eye DiseaseStudy also found a lower prevalence of AMD in lightalcohol drinkers comparedwith nondrinkers (adjustedOR, 0.38; 95% CI, 0.19–0.7) (86). Considering that AMDmay share similar pathologic processes with cardio-vascular diseases (73), the findings that moderate wineconsumption is associatedwith decreasedOR of devel-oping AMD are consistent with reports of a protectiveeffect of moderate alcohol intake for coronary arterydisease and stroke (246).

In the Beaver Dam Eye Study, beer consumptionwas found to be associated with increased prevalenceof retinal pigment and neovascular AMD (247). In anincidence study, beer consumption was found to bepositively associated with the incidence of soft indis-tinct drusen, increased drusen area, and confluence ofsoft drusen (87). People who reported being heavydrinkers at baseline were more likely to develop lateAMD (RR, 6.94; 95% CI, 1.85–26.1) after 10 years thanpeople who reported never having been heavy drin-kers (248). The Blue Mountains Eye Study found noassociation between alcohol consumption and theprevalence of early or late AMD or large drusen,although there was a significant positive associationbetween consumption of distilled spirits and earlyAMD (249).

Prospective data from 111,238 women and menin the Nurses’ Health Study and the Health Pro-fessionals Follow-Up Study do not support aprotective effect of moderate alcohol consumption onthe risk of AMD (97). No substantial associationbetween total alcohol intake and incidence of AMDwas found from the 697,498 person-years of follow-upin women and 229,180 person-years of follow-up inmen. After controlling for age, smoking, and other riskfactors, the pooled RRs (95% CI) for AMD comparedwith nondrinkers were 1.0 (0.7–1.2) for drinkers whoconsumed 0.1–4.9 g/day of alcohol, 0.9 (0.6–1.4) for5–14.9 g/day, 1.1 (0.7–1.7) for 15–29.9 g/day, and 1.3(0.9–1.8) for 30 g/day or more. However, there was amodest increased risk of early AMD and geographicatrophy in women who consumed 30 g/day or moreof alcohol (RR, 2.04; 95% CI, 1.22–3.42). There was noassociation between alcohol intake and neovascularAMD in either sex, but it should be pointed out thatthe number of neovascular AMD was small inthe study.

Prospective data of 21,041 male physicians withan average follow-up of 12.5 years in the Physicians’Health Study also indicate that alcohol intake is notappreciably associated with the risk of AMD (93). Theoverall RR of any AMD among men who reportedbaseline alcohol consumption of R1 drink/weekcompared with those drinking !1 drink/week was0.97 (95% CI, 0.78–1.21) after multivariate adjustment.Similarly, the RR of AMD with visual loss and neo-vascular AMD were 0.99 (95% CI, 0.75–1.31) and 0.87(95% CI, 0.51–1.51), respectively, after multivariateadjustment. For AMD with vision loss, the RRs (95%CI) for those reporting!1 drink/week, 1 drink/week,2 to 4 drinks/week, 5 to 6 drinks/week, and R1drink/day at baseline were 1.0 (referent), 0.75(0.47–1.21), 1.0 (0.69–1.45), 1.20 (0.81–1.78), and 1.19(0.87–1.61), respectively. Several other smaller studiesalso found no association between the history ofalcohol consumption and AMD (10,15,45).


DEVELOPMENT OF CNV IN AMD

AMD is a bilateral condition that tends to be fairlysymmetric in its presentation and clinical course(250,251). A study of the symmetry of disciform scarsfound a significant correlation between eyes in termsof the final scar size, and large macular scars weremore frequent in the second eye if the first eye had alarge scar (250). In the Blue Mountains Eye Study, 40%of the neovascular AMD cases were bilateral (71).Once one eye is affected, there is a significant risk forinvolvement in the fellow eye. Although peripheralvision is almost always retained in late AMD, bilateralcentral scotomas result in decreased mobility andimpaired reading ability, and dramatically impact onoccupational and recreational activities.

It has been demonstrated that choroidal neovas-cular lesions of AMD account for the vast majority ofsevere visual loss from this condition (252). The 79%and 90% of eyes legally blind due to AMD in theFramingham Eye Study (3) and a large case–controlstudy (12), respectively, had CNV. Thus, patients atrisk of bilateral CNV are at the greatest risk of severevisual loss. Because the treatment of CNV is mosteffective when it is new and has not caused irreversiblescarring and photoreceptor damage, it is important toidentify high-risk patients and educate them about the

importance of daily self-monitoring of the centralvisual field for each eye.

Risk of CNV in AMDA number of studies have reported the natural courseof patients with bilateral drusen with good visualacuity (Table 8) (116,254–256) while others haveassessed the risk of developing CNV in the felloweye in patients with age-related CNV in one eye(Table 9) (116,257–266). Variation in the reported riskamong the studies is probably due partly to variationin the clinical features of the macula (e.g., drusen sizeand confluence, presence of focal hyperpigmentation,and/or RPE depigmentation) (253).

Lanchoney et al. (267), using the follow-upstudies of Smiddy and Fine (254) and Holz et al.(255), predicted that the proportion of patients withbilateral soft drusen developing CNV in either one orboth eyes would be 12.4% within 10 years, but this riskvaried from 8.6% to 15.9%, depending on sex and ageof the patient. In their model, the rate of developmentof CNV in the first eye was reduced after five years to75% of the initial rate observed in follow-up studiesand to 50% of the initial rate after 10 years (267).

Gass reported that of 91 patients who were seeninitially with loss of vision due to disciform maculardetachment or degeneration in one eye, neovascular

Table 8 Risk of Developing Choroidal Neovascularization in Age-Related Macular Degeneration Patients with Bilateral Drusen andGood Bilateral Visual Acuity

StudyNumber of

eyes/patients

Mean age(range inyears) Initial visual acuity

Mean follow-up(range in years) Results

Gass (1973)

(116)

98/49 61 (29–81) 20/20 OU in 21 patients

(43%)

4.9 9 (18%) of 49 patients developed

central visual loss in one eye

because of CNVs20/25 to 20/40 OU in 18

patients (37%)

Smiddy and Fine

(1984) (254)

142/71 58 (16–78) 20/50 or better in 132 (93%)

of eyes studied

4.3 (0.5–8.6) 8 eyes (9.9%) of 7 patients developed

CNVs over 4.3 years (14.5%

cumulative risk)

7 eyes (8.5%) of 6 patients developed

severe visual loss (12.7% 5-year

cumulative risk)

Holz et al. (1994)

(255)

126 patients 68 “Good” 3 17 (13.5%) of 126 patients developed

new lesionsa

Cumulative incidence of new lesions

among patientsR 65 years old

was:

8.55% @1 year

16.37% @ 2 years

23.52% @ 3 years

Bressler et al.

(1995) (256)

483 patients NA NA 5 1 (0.2%) of 483 patients developed

CNV

1 (0.2%) of 483 patients developed

peripapillary CNVs

None developed geographic atrophy

a Classic or occult CNVs, RPE detachmentGCNVs, or geographic atrophy extending to the fovea.Abbreviations: NA, information not available; OU, both eyes; CNV, choroidal neovascularization.

74 AU EONG ET AL.

Table9

RiskofDevelopingChoroidalNeovascularizationintheFellowEyeofAge-RelatedMacularDegenerationPatientswithChoroidalNeovascularizationinOneEye

Study

Numberofpatients

Meanage

(rangeinyears)

Initialvisualacuity

Meanfollow-up(range)

Results

Gass(1973)(116)

91patients

67(49–82)

20/20in30patients

(33%)

4years

31eyes(34%)lostcentralvisionbecause

ofCNVsduringfollow-up

20/40orbetterinallbut

7patients(92%)

TeetersandBird(1973)(257)

42patients

NA

NA

21eyes(50%)followed-upfor12

months(7–19months)

Nochange


months(4–19months)

Increaseddrusenandpigmentation

3eyes(7%)followed-upfor9,

16,and21months

Allthreeeyesdevelopedavascular

disciform

appearancea


and24months

Botheyesdevelopedneovascular

disciform

appearance

Overall,5(12%)of42eyesdeveloped

“avascular”andneovascular

complications

Gragoudasetal.(1976)(259)

36patients

NA

“Good”

22months(12–48months)

13(36%)of36patientsdeveloped

disciform

macularlesions

Gregoretal.(1977)(260)

104patients

NA

NA

Upto5years

12–15%/yeardevelopedCNVs

Resultswere:

9/104(9.8%)@

1year

18/74(19%)@

2years

17/53(30%)@

3years

11/23(48%)@

4years

5/11(45%)@

5years

Strahlmanetal.(1983)(261)

84patients

68(47–91)

NA

27months(6–95months)

UsingKaplan–Meiertechnique,theriskof

developingexudativemaculopathyin

felloweyewasestimatedtobe3–7%

yearly

6/84(7%)developedCNVs

2/84(2%)developedpigmentepithelial

detachment

1/84(1%)developedgeographicatrophy

over18months(range5–36months)

Bressleretal.(1990)(269)

127patientswithextrafoveal

CNVsinoneeye

NA

NA

5years

10%ofeyeswithnolargedrusenorRPE

hyperpigmentationcomparedwith58%

ofeyeswithbothlargedrusenand

hyperpigmentationdevelopedCNVsin

thefelloweyewithin5years

MacularPhotocoagulation

StudyGroup(1997)(266)

670patientswithjuxtafovealor

subfovealCNVsinoneeye

NA

20/400orbetter

NA

Estimated5-yearincidenceratesranged

from7%forthesubgroupwithonerisk

factorto87%forthesubgroupwithall

fourriskfactorsb

ThepresenceofoccultCNVsinthefirst

eyeaffectedhadnoinfluenceonthe

typeofCNVsinthefelloweye

aSerousRPEdetachmentofRPEandretinawithoutevidenceofCNVs.

bFiveormoredrusen,large(O

63mmindiameter)focalhyperpigmentation,systemichypertension.

Abbreviations:NA,informationnotavailable;RPE,retinalpigmentepithelial;CNV,choroidalneovascularization.


lesions developed in the second eye in 31 patients(34%) over an average follow-up of four years (116).Chandra et al. reported that among 36 patients withunilateral disciform lesions, bilateral involvementoccurred in 13 (36%) after an average follow-up of 22months (258). Gregor et al. followed 104 patients aged60 to 69 years who initially had a disciform maculardegeneration in one eye for between one and five years(260). From their data, they estimated that the annualincidence of developing a disciform lesion in thefellow eye to be 12% per year in the first five years.Strahlman et al. reported that among 84 patients withunilateral exudative AMD, 9 (11%) developed bilateralinvolvement after a mean follow-up of 27 months(261). Baun et al. studied 45 patients with unilateralneovascular AMD for four years and documentedCNV in the fellow eye in 14 (31%) patients (262).Sandberg et al. found an average of 8.8% of patientswith unilateral neovascular AMD develop CNV in thefellow eye each year in their prospective series of 127patients with 4.5 years of follow-up (263).

The MPS Group examined the data of fellow eyesof study participants in the MPS randomized trial forargon laser photocoagulation for extrafoveal CNVsecondary to AMD (265) and the randomized trials oflaser photocoagulation for new juxtafoveal CNV, newsubfoveal CNV, or recurrent subfoveal CNV secondaryto AMD (266). In the extrafoveal CNV trial, 128

participants had a fellow eye that was initially free ofCNV at baseline (265). During five years of follow-up,choroidal neovascular lesions associated with AMDwere observed in 33 (26%) of the 128 fellow eyes. In theother three MPS trials, among 670 patients with noclassic or occult CNV in the fellow eye at the time ofenrollment, CNV developed in 236 (35%) within fiveyears (266). The cumulative incidence rates of CNV inthe fellow eye for this group of patients were estimatedto be 10%, 28%, and 42% at one, three, and five years,respectively (Fig. 1).

The AREDS also evaluated the incidence ofneovascular AMD among participants of the random-ized trial (268). Neovascular AMD was defined in thestudy as photocoagulation for CNV, or photographicevidence of any of the following: nondrusenoid RPEdetachment, serous or hemorrhagic retinal detach-ment, hemorrhage under the retina or the RPE, andsubretinal fibrosis. Of individuals with early or inter-mediate AMD at baseline with a median follow-up of6.3 years, 788 were at risk of developing advancedAMD in one eye (the fellow eye had advanced AMD)and 2506 were at risk in both eyes. Of the 2506participants in the bilateral drusen group, 256 (10%)developed neovascular AMD in at least one eyeduring the course of the study. Of the 788 participantsin the unilateral advanced AMD group, 278 (35%)developed neovascular AMD during the study.

20

24

Follow-up period, mo

18126

04

11

28

40

48

61

6872

81

87

53

54 60

44

42 48

25

38

4947

32

23

19

30

30

38

36

34

28

1513

31.5

3300

7 7 7 7 757300

0

10

20

30

40

50

60

70

80

90

1000 (n=35)1 (n=105)

3 (n=105)4 (n=45)

2 (n=142)

12

1614

0

22

272320

13

23

Eye

sw

ithch

oroi

daln

eova

scul

ariz

atio

n,%

Figure 1 Incidence of choroidal neovascularization by number of risk factors present including hypertension, R5 drusen, R1 largedrusen (greatest linear dimension O63 mm), and focal hyperpigmentation. Source: From Ref. 266.

76 AU EONG ET AL.

Risk Factors for Progression to CNVThe MPS Group evaluated selected risk factors fordevelopment of CNV in the fellow eye of patients inthe randomized trials of laser photocoagulation fornew juxtafoveal CNV, new subfoveal CNV, or recur-rent subfoveal CNV secondary to AMD (266). A trendfor increased incidence with age (pZ0.06) wasobserved. No strong association was found betweenfemale sex, higher frequency of aspirin usage, cigarettesmoking, and hyperopia with an increased riskof CNV.

Certain drusen and RPE abnormalities within1500 mm of the foveal center present in the fellow eyeand patient characteristics at baseline were identifiedas risk factors for the development of CNV in theseeyes (266,269). Specific risk factors include the presenceof five or more drusen (RR, 2.1; 95% CI, 1.3–3.5), focalhyperpigmentation (RR, 2.0; 95% CI, 1.4–2.9), definitesystemic hypertension (systolic pressureR140 mmHg,diastolic pressureR90 mmHg, or use of antihyperten-sive medications) (RR, 1.7; 95% CI, 1.2–2.4), and oneor more large drusen (greater 63 mm in greatest lineardimension) (RR, 1.5; 95% CI, 1.0–2.2). The risk of CNVdeveloping within five years after presenting withCNV in the first eye ranged from 7% if none of theserisk factors was present to 87% if all four risk factorswere present (Fig. 1).

Multivariate analysis of the risk factors for pro-gression to CNV in AREDS participants yieldedtwo risk factors (268). In persons at risk of advancedAMD in both eyes, while controlling for age, gender,and AREDS treatment group, white race (OR,white vs. black, 6.77; 95% CI, 1.24–36.9) and smokingO10 pack-years (OR, O10 vs. %10 pack-years, 1.55;95% CI, 1.15–2.09) were independently associated withincident neovascular AMD.

CONCLUSION

In summary, many risk factors for AMD have beenidentified from case–control, cross-sectional, andprospective cohort studies. Risk factors such asincreasing age, gender, or family history of thedisease cannot be modified. One important modifiablerisk factor is cigarette smoking (91). Dietary habits arealso modifiable, and findings from AREDS suggestthat persons with extensive intermediate size drusen,at least one large druse, noncentral geographic atrophyin one or both eyes, or advanced AMD or vision lossdue to AMD in one eye, and without contraindicationssuch as cigarette smoking, should consider taking asupplement of antioxidants plus zinc to reduce theirrisk of progression to advanced AMD and vision loss.Since sunglasses may protect against cataract forma-tion, are inexpensive, and are not associated with any

major side effects, it may be reasonable to wearsunglasses to reduce UV and other light exposure toocular structures. The challenge for researchers is tomore firmly establish modifiable risk factors and toconduct large-scale prospective intervention trials onthese factors so that preventive measures and bettertreatments can be developed.

SUMMARY POINTS

& Importance of identifying risk factors for AMD. Theidentification and modification of risk factors forAMD has the potential for greater public healthimpact on the morbidity from the disease than thefew treatment modalities currently available

& Studies on risk factors for AMD. Case–control, cross-sectional, and prospective cohort studies can iden-tify risk factors for AMD. Repeated findings ofthe same risk factors in well-designed studiesconducted in different populations are necessaryto provide compelling evidence of a real associ-ation between AMD and potential risk factors.However, only randomized prospective clinicaltrials can prove that modifying a particularestablished risk factor can influence the courseof AMD

& Classification of risk factors. Risk factors for AMDmay be broadly classified into personal or environ-mental factors (e.g., smoking, sunlight exposure, andnutritional factors including micronutrients,dietary fish intake, and alcohol consumption).Personal factors may be further subdividedinto sociodemographic (e.g., age, sex, race/ethnicity,heredity, and socioeconomic status), ocular (e.g.,iris color, macular pigment optical density, cataractand its surgery, refractive error, and cup/discratio), and systemic factors (e.g., cardiovasculardisease and its risk factors, reproductive andrelated factors, dermal elastotic degeneration, andantioxidant enzymes)

& Established risk factors. Age, race/ethnicity,heredity, and smoking

& Possible risk factors. Sex, socioeconomic status, iriscolor, macular pigment optical density, cataractand its surgery, refractive error, cup/disc ratio,cardiovascular disease, hypertension and bloodpressure, serum lipid levels and dietary fat intake,body mass index, hematologic factors, Chlamydiapneumoniae infection, reproductive and relatedfactors, dermal elastotic degeneration, antioxidantenzymes, sunlight exposure, micronutrients,dietary fish intake, and alcohol consumption

& Factors probably not associated with AMD.Diabetes and hyperglycemia

& Risk factors for progression to choroidal neovasculari-zation. Presence of five or more drusen, focal


hyperpigmentation, systemic hypertension, one ormore large drusen (O63 mm in greatest lineardimension), white race, and smoking

& Current opinion on modifying risk factors. A numberof well-established factors such as increasing ageand a family history of the disease unfortunatelycannot be modified. One modifiable well-estab-lished risk factor is cigarette smoking. There maybe potential benefits of antismoking patient edu-cation for primary and secondary prevention ofAMD. The Age-Related Eye Disease Studysuggested that persons older than 55 years withextensive intermediate size drusen, at least onelarge druse, noncentral geographic atrophy in oneor both eyes, or advanced AMD or vision loss dueto AMD in one eye, and without contraindicationssuch as cigarette smoking, should consider taking asupplement of antioxidants plus zinc to reducetheir risk of progression to advanced AMD andvision loss. Although sunlight exposure has notbeen established as a risk factor for AMD, it may bereasonable to wear sunglasses to reduce ultravioletand other light exposure to ocular structures sincesunglasses may protect against cataract formation,are inexpensive, and are not associated with anymajor side effects

REFERENCES

1. Friedman DS, O’Colmain BJ, Munoz B, et al. Prevalence ofage-related macular degeneration in the United States.Arch Ophthalmol 2004; 122(4):564–72.

2. Krumpaszky HG, Ludtke R, Mickler A, Klauss V,Selbmann HK. Blindness incidence in Germany: a popu-lation-based study from Wurttemberg–Hohenzollern.Ophthalmologica 1999; 213(3):176–82.

3. Leibowitz HM, Krueger DE, Maunder LR, et al. TheFramingham Eye Study Monograph: an ophthalmologicaland epidemiological study of cataract, glaucoma, diabeticretinopathy, macular degeneration, and visual acuity in ageneral population of 2631 adults, 1973–1975. SurvOphthalmol 1980; 24:335–610.

4. Attebo K,Mitchell P, SmithW. Visual acuity and the causesof visual loss in Australia: the Blue Mountains Eye Study.Ophthalmology 1996; 103(3):357–64.

5. Hawkins BS, Bird AC, Klein R, West SK. Epidemiology ofage-related macular degeneration. Mol Vis 1999; 5:26(http://www.molvis.org/molvis/v5/p26).

6. Eye Disease Case-Control Study Group. Antioxidantstatus and neovascular age-related macular degeneration.Arch Ophthalmol 1993; 111(1):104–9.

7. The Eye Disease Case-Control Study Group. Risk factorsfor neovascular age-related macular degeneration. ArchOphthalmol 1992; 110(12):1701–8.

8. Goldberg J, Flowerdew G, Smith E, Brody JA, Tso MOM.Factors associated with age-related macular degeneration.An analysis of data from the first National Health andNutritional Examination Survey. Am J Epidemiol 1988;128(4):700–10.

9. ChristenWG, Ajani UA, Glynn RJ, et al. Prospective cohortstudy of antioxidant vitamin supplement use and therisk of age-related maculopathy. Am J Epidemiol 1999;149(5):476–84.

10. Maltzman BA, Mulvihill MN, Greenbaum A. Senilemacular degeneration and risk factors: a case-controlstudy. Ann Ophthalmol 1979; 11(8):1197–201.

11. Delaney WV, Oates RP. Senile macular degeneration: apreliminary study. Ann Ophthalmol 1982; 14(1):21–4.

12. Hyman LG, Lilienfeld AM, Ferris FLI, Fine SL. Senilemacular degeneration: a case-control study. Am J Epide-miol 1983; 118(2):213–27.

13. Weiter JJ, Delori FC, Wing GL, Fitch KA. Relationship ofsenile macular degeneration to ocular pigmentation. AmJ Ophthalmol 1985; 99(2):185–7.

14. Blumenkranz MS, Russell SR, Roeby MG, Kott-Blumenkranz R, Penneys N. Risk factors in age-relatedmaculopathy complicated by choroidal neovasculariza-tion. Ophthalmology 1986; 96(5):552–8.

15. Tsang NCK, Penfold PL, Snitch PJ, Billson F. Serum levelsof antioxidants and age-related macular degeneration.Doc Ophthalmol 1992; 81(4):387–400.

16. Seddon JM, Ajani UA, Sperduto RD, et al. Dietary caroten-oids, vitamins A, C, and E, and advanced age-relatedmacular degeneration. JAMA 1994; 272(18):1413–20.

17. Holz FG, Piguet B, Minassian DC, Bird AC, Weale RA.Decreasing stromal iris pigmentation as a risk factor forage-related macular degeneration. Am J Ophthalmol 1994;117(1):19–23.

18. Mares-Perlman JA, Brady WE, Klein R, et al. Serumantioxidants and age-related macular degeneration in apopulation-based case-control study. Arch Ophthalmol1995; 113(12):1518–23.

19. Vingerling JR, Dielemans I, Witteman JCM, Hofman A,Grobbee DE, de Jong PTVM. Macular degeneration andearly menopause: a case-control study. BMJ 1995;310(6994):1570–1.

20. Darzins P, Mitchell P, Heller RF. Sun exposure and age-related macular degeneration: an Australian case-controlstudy. Ophthalmology 1997; 104(5):770–6.

21. Tamakoshi A, Yuzawa M, Matsui M, et al. Smoking andneovascular form of age-related macular degeneration inlater middle aged males: findings from a case-controlstudy in Japan. Br J Ophthalmol 1997; 81(10):901–4.

22. Chaine G, Hullo A, Sahel J, et al. Case-control study of therisk factors for age-related macular degeneration. BrJ Ophthalmol 1998; 82(9):996–1002.

23. Belda JI, Roma J, Vilela C, et al. Serum vitamin E levelsnegatively correlate with severity of age-related maculardegeneration. Mech Ageing Dev 1999; 107(2):159–64.

24. Hyman L, Schachat AP, He Q, Leske C, Age-RelatedMacular Degeneration Risk Factors Study Group. Hyper-tension, cardiovascular disease, and age-related maculardegeneration. Arch Ophthalmol 2000; 118(3):351–8.

25. KalayogluMV, Galvan C,Mahdi OS, Byrne GI, Mansour S.Serological association between Chlamydia pneumoniaeinfection and age-related macular degeneration. ArchOphthalmol 2003; 121(4):478–82.

26. Seddon JM, Gensler G, Milton RC, Klein ML, Rifai N.Association between C-reactive protein and age-relatedmacular degeneration. JAMA 2004; 291(6):704–10.

27. Khan JC, Thurlby DA, Shahid H, et al. Smoking and agerelated macular degeneration: the number of pack yearsof cigarette smoking is a major determinant of risk forboth geographic atrophy and choroidal neovasculariza-tion. Br J Ophthalmol 2006; 90(1):75–80.

78 AU EONG ET AL.

28. McGwin G, Modjarrad K, Hall AT, Xie A, Owsley C.3-hydroxy-3-methylglutaryl coenzyme A reductase inhibi-tors and the presence of age-related macular degenerationin the Cardiovascular Health Study. Arch Ophthalmol2006; 124(1):33–7.

29. Kahn HA, Leibowitz HM, Ganley JP, et al. TheFramingham Eye Study: II. Association of ophthalmicpathology with single variables previously measured inthe Framingham Heart Study. Am J Epidemiol 1977;106(1):33–41.

30. Kahn HA, Leibowitz HM, Ganley JP, et al. TheFramingham Eye Study: I. Outline and major prevalencefindings. Am J Epidemiol 1977; 106(1):17–32.

31. Sperduto RD, Seigel D. Senile lens and senile macularchanges in a population-based sample. Am J Ophthalmol1980; 90(1):86–91.

32. Sperduto RD, Hiller R, Seigel D. Lens opacities and senilemaculopathy. Arch Ophthalmol 1981; 99(6):1004–8.

33. Sperduto RD, Hiller R. Systemic hypertension and age-related maculopathy in the Framingham Study. ArchOphthalmol 1986; 104(2):216–9.

34. Martinez GS, Campbell AJ, Reinken J, Allan BC.Prevalence of ocular disease in a population study ofsubjects 65 years old and older. Am J Ophthalmol 1982;94(2):181–9.

35. Klein BE, Klein R. Cataracts and macular degeneration inolder Americans. Arch Ophthalmol 1982; 100(4):571–3.

36. Liu IY,White L, LaCroix AZ. The association of age-relatedmacular degeneration and lens opacities in the aged. AmJ Public Health 1989; 79(6):765–9.

37. Obisesan TO, Hirsch R, Kosoko O, Carlson L, Parrott M.Moderate wine consumption is associated with decreasedodds of developing age-related macular degeneration inNHANES-1. J Am Geriatr Soc 1998; 46(1):1–7.

38. Gibson JM, Shaw DE, Rosenthal AR. Senile cataractand senile macular degeneration: an investigation intopossible risk factors. Trans Ophthalmol Soc UK 1986;105(Pt 4):463–8.

39. West SK, Rosenthal FS, Bressler NM, et al. Exposure tosunlight and other risk factors for age-related maculardegeneration. Arch Ophthalmol 1989; 107(6):875–9.

40. Bressler NM, Bressler SB, West SK, Fine SL, Taylor HR.The grading and prevalence of macular degeneration inChesapeake Bay watermen. Arch Ophthalmol 1989;107(6):847–52.

41. Taylor HR, West SK, Munoz B, Rosenthal FS, Bressler SB,Bressler NM. The long-term effects of visible light on theeye. Arch Ophthalmol 1992; 110(1):99–104.

42. Taylor HR, Munoz B, West S, Bressler NM, Bressler SB,Rosenthal FS. Visible light and risk of age-related maculardegeneration. Trans Am Ophthalmol Soc 1990; 88:163–73.

43. Vinding T. Age-related macular degeneration: macularchanges, prevalence and sex ratio. An epidemiologicalstudy of 1000 aged individuals. Acta Ophthalmol(Copenh) 1989; 67(6):609–16.

44. Vinding T. Pigmentation of the eye and hair in relation toage-related macular degeneration: an epidemiologicalstudy of 1000 aged individuals. Acta Ophthalmol(Copenh) 1990; 68(1):53–8.

45. Vinding T, Appleyard M, Nyboe J, Jensen G. Risk factoranalysis for atrophic and exudative age-related maculardegeneration: an epidemiological study of 1000 agedindividuals. Acta Ophthalmol (Copenh) 1992; 70(1):66–72.

46. West S, Vitale S, Hallfrisch J, et al. Are antioxidants orsupplements protective for age-related macular degener-ation? Arch Ophthalmol 1994; 112(2):222–7.

47. Klein R, Klein BEK, Linton LKP. Prevalence of age-relatedmaculopathy: the Beaver Dam Eye Study. Ophthalmology1992; 99(6):933–43.

48. Klein R, Klein BEK, Jensen SC, Moss SE, Cruickshanks KJ.The relation of socioeconomic factors to cataract, maculo-pathy, and impaired vision: the Beaver Dam Eye Study.Ophthalmology 1994; 101(12):1969–79.

49. Klein R, Klein BEK, Wang Q, Moss SE. Is age-relatedmaculopathy associated with cataracts? Arch Ophthalmol1994; 112(2):191–6.

50. Klein R, Klein BEK, Franke T. The relationship of cardio-vascular disease and its risk factors to age-relatedmaculopathy: the Beaver Dam Eye Study. Ophthalmology1993; 100(3):406–14.

51. Klein R, Klein BEK, Moss SE. Diabetes, hyperglycemia,and age-related maculopathy: the Beaver Dam Eye Study.Ophthalmology 1992; 99(10):1527–34.

52. Klein R, Klein BE, Linton KLP, Demets DL. The BeaverDam Eye Study: the relation of age-relatedmaculopathy tosmoking. Am J Epidemiol 1993; 137(2):190–200.

53. Cruickshanks KJ, Klein R, Klein BEK. Sunlight and age-related macular degeneration: the Beaver Dam Eye Study.Arch Ophthalmol 1993; 111(4):514–8.

54. Cruickshanks KJ, Klein R, Klein BEK, Nondahl DM.Sunlight and the 5-year incidence of early age-relatedmaculopathy: the Beaver Dam Eye Study. Arch Ophthal-mol 2001; 119(2):246–50.

55. Heiba IM, Elston RC, Klein BEK, Klein R. Sibling correlat-ions and segregation analysis of age-related maculopathy:the Beaver Dam Eye Study. Genet Epidemiol 1994;11(1):51–67.

56. Mares-Perlman JA, Brady WE, Klein R, VandenLangen-berg GM, Klein BEK, Palta M. Dietary fat and age-relatedmaculopathy. Arch Ophthalmol 1995; 113(6):743–8.

57. Schachat AP, Hyman L, Leske C, Connell AMS, Wu SY,Barbados Eye Study Group. Features of age-relatedmacular degeneration in a black population. ArchOphthalmol 1995; 113(6):728–35.

58. Vingerling JR, Dielemans I, Hofman A, et al. The preva-lence of age-related maculopathy in the Rotterdam study.Ophthalmology 1995; 102(2):205–10.

59. Vingerling JR, Dielemans I, Bots ML, Hofman A,Grobbee DE, de Jong PTVM. Age-related macular degener-ation is associated with atherosclerosis: the RotterdamStudy. Am J Epidemiol 1995; 142(4):404–9.

60. Ikram MK, van Leeuwen R, Vingerling JR, Hofman A, deJong PTVM. Retinal vessel diameters and the risk ofincident age-related macular disease: the RotterdamStudy. Ophthalmology 2005; 112(4):548–52.

61. Mitchell P, Smith W, Attebo K, Wang JJ. Prevalence of age-related maculopathy in Australia: the Blue Mountains EyeStudy. Ophthalmology 1995; 102(10):1450–60.

62. Mitchell P, Smith W, Wang JJ. Iris color, skin sun sensi-tivity, and age-related maculopathy: the Blue MountainsEye Study. Ophthalmology 1998; 105(8):1359–63.

63. Mitchell P, Wang JJ. Diabetes, fasting blood glucose andage-related maculopathy: the Blue Mountains Eye Study.Aust NZ J Ophthalmol 1999; 27(3-4):197–9.

64. Mitchell P, Wang JJ, Smith W, Leeder SR. Smoking andthe 5-year incidence of age-related maculopathy: theBlue Mountains Eye Study. Arch Ophthalmol 2002;120(10):1357–63.

65. Smith W, Mitchell P, Leeder SR. Smoking and age-relatedmaculopathy: the Blue Mountain Eye Study. ArchOphthalmol 1996; 114(12):1518–23.


66. Smith W, Mitchell P, Leeder SR, Wang JJ. Plasma fibri-nogen levels, other cardiovascular risk factors, and age-related maculopathy: the Blue Mountains Eye Study. ArchOphthalmol 1998; 116(5):583–7.

67. Smith W, Mitchell P. Family history and age-relatedmaculopathy: the Blue Mountains Eye Study. Aust NZJ Ophthalmol 1998; 26(3):203–6.

68. Smith W, Mitchell P, Leeder SR. Dietary fat and fish intakeand age-related maculopathy. The Blue Mountains EyeStudy. Arch Ophthalmol 2000; 118(3):401–4.

69. Smith W, Mitchell P, Webb K, Leeder SR. Dietary antiox-idants and age-related maculopathy: the Blue MountainsEye Study. Ophthalmology 1999; 106(4):761–7.

70. Smith W, Mitchell P, Rochester C. Serum beta carotene,alpha tocopherol, and age-related maculopathy: theBlue Mountains Eye Study. Am J Ophthalmol 1997;124(6):838–40.

71. Wang JJ, Mitchell P, Smith W, Cumming RG. Bilateralinvolvement by age related maculopathy lesions in apopulation. Br J Ophthalmol 1998; 82(7):743–7.

72. Wang JJ, Mitchell P, Cumming RG, Lim R. Cataract andage-related maculopathy: the Blue Mountains Eye Study.Ophthalmic Epidemiol 1999; 6(4):317–26.

73. Hirvela H, Luukinen H, Laara ESL, Laatikainen L. Riskfactors of age-related maculopathy in a population 70years of age or older. Ophthalmology 1996; 103(6):871–7.

74. Delcourt C, Cristol J-P, Tessier F, et al. Age-related maculardegeneration and antioxidant status in the POLA study.Arch Ophthalmol 1999; 117(10):1384–90.

75. Delcourt C, Cristol J-P, Leger CL, Descomps B, Papoz L,POLA Study Group. Associations of antioxidant enzymeswith cataract and age-related macular degeneration.Ophthalmology 1999; 106(2):215–22.

76. Delcourt C, Diaz J-L, Ponton-Sanchez A, Papoz L, POLAStudy Group. Smoking and age-related macular degener-ation: the POLA Study. Arch Ophthalmol 1998;116(8):1031–5.

77. Defay R, Pinchinat S, Lumbroso S, Sutan C, Delcourt C,POLA Study Group. Sex steroids and age-related maculardegeneration in older French women: the POLA Study.Ann Epidemiol 2004; 14(3):202–8.

78. Friedman DS, Katz J, Bressler NM, Rahmani B, Tielsch JM.Racial differences in the prevalence of age-related maculardegeneration. Ophthalmology 1999; 106(6):1049–55.

79. Klein R, Clegg L, Cooper LS, et al. Prevalence of age-related maculopathy in the atherosclerosis risk in commu-nities study. Arch Ophthalmol 1999; 117(9):1203–10.

80. Klein R, Klein BEK, Jensen SC, Mares-Perlman JA,Cruickshanks KJ, Palta M. Age-related maculopathy in amultiracial United States population: the National Healthand Nutrition Examination Survey III. Ophthalmology1999; 106(6):1056–65.

81. Klein R, Rowland ML, Harris MI. Racial/ethnicdifferences in age-related maculopathy: Third NationalHealth and Nutrition Examination Survey. Ophthal-mology 1995; 102(3):371–81.

82. McCarty CA, Mukesh BN, Fu CL, Mitchell P, Wang JJ,Taylor HR. Risk factors for age-related maculopathy: theVisual Impairment Project. Arch Ophthalmol 2001;119(10):1455–62.

83. Miyazaki M, Kiyohara Y, Yoshida A, Iida M, Nose Y,Ishibashi T. The 5-year incidence and risk factors forage-related maculopathy in a general Japanese popu-lation: the Hisayama Study. Invest Ophthalmol Vis Sci2005; 46(6):1907–10.

84. Fraser-Bell S, Donofrio J, Wu J, et al. Sociodemographicfactors and age-related macular degeneration in Latinos:the Los Angeles Latino Eye Study. Am J Ophthalmol 2005;139(1):30–8.

85. Fraser-Bell S, Wu J, Klein R, Azen SP, Varma R, LosAngeles Latino Eye Study Group. Smoking, alcoholintake, estrogen use, and age-related macular degener-ation in Latinos: the Los Angeles Latino Eye Study. AmJ Ophthalmol 2006; 141(1):79–87.

86. Krishnaiah S, Das T, Nirmalan PK, et al. Risk factors forage-related macular degeneration: findings from theAndhra Pradesh Eye Disease Study in South India.Invest Ophthalmol Vis Sci 2005; 46(12):4442–9.

87. Moss SE, Klein R, Klein BEK, Jensen SC, Meuer SM.Alcohol consumption and the 5-year incidence of age-relatedmaculopathy: the Beaver Dam Eye Study. Ophthal-mology 1998; 105(5):789–94.

88. Klein R, Klein BEK, Jensen SC, Meuer SM. The five-yearincidence and progression of age-relatedmaculopathy: theBeaver Dam Eye Study. Ophthalmology 1997; 104(1):7–21.

89. Klein R, Klein BEK, Jensen SC, Cruickshanks KJ. Therelationship of ocular factors to the incidence and pro-gression of age-related maculopathy. Arch Ophthalmol1998; 116(4):506–13.

90. Klein R, Klein BEK, Jensen SC. The relation of cardiovas-cular disease and its risk factors to the 5-year incidence ofage-related maculopathy: the Beaver Dam Eye Study.Ophthalmology 1997; 104(11):1804–12.

91. Seddon JM, Willett WC, Speizer FE, Hackinson SE. Aprospective study of cigarette smoking and age-relatedmacular disease in women. JAMA 1996; 276(14):1141–6.

92. Christen WG, Glynn RJ, Manson JE, Ajani UA, Buring JE.A prospective study of cigarette smoking and risk ofage-related macular degeneration in men. JAMA 1996;276(14):1147–51.

93. Ajani UA, Christen WG, Manson JE, et al. A prospectivestudy of alcohol consumption and the risk of age-relatedmacular degeneration. Ann Epidemiol 1999; 9(3):172–7.

94. Schaumberg DA, Christen WG, Hankinson SE, Glynn RJ.Body mass index and the incidence of visually significantage-related maculopathy in men. Arch Ophthalmol 2001;119(9):1259–65.

95. Cho E, Seddon JM, Rosner B, Willett WC, Hankinson SE.Prospective study of intake of fruits, vegetables, vitamins,and carotenoids and risk of age-relatedmaculopathy. ArchOphthalmol 2004; 122(6):883–92.

96. Cho E, Stampfer MJ, Seddon JM, et al. Prospective study ofzinc intake and the risk of age-related macular degener-ation. Ann Epidemiol 2001; 11(5):328–36.

97. Cho E, Hankinson SE, Willett WC, et al. Prospective studyof alcohol consumption and the risk of age-relatedmaculardegeneration. Arch Ophthalmol 2000; 118(5):681–8.

98. Klein R, Klein BEK, Knudtson MD, et al. Prevalence ofage-related macular degeneration in 4 racial/ethnicgroups in the Multi-ethnic Study of Atherosclerosis.Ophthalmology 2006; 113(3):373–80.

99. The International ARM Epidemiological Study Group.An international classification and grading system forage-related maculopathy and age-related maculardegeneration. Surv Ophthalmol 1995; 39(5):367–74.

100. Boldt HC, Bressler SB, Fine SL, Bressler NM. Age-relatedmacular degeneration. Curr Opin Ophthalmol 1990;1(3):247–57.

101. Ferris FLI. Senile macular degeneration: review of epide-miologic features. Am J Epidemiol 1983; 118(2):132–51.

80 AU EONG ET AL.

102. Christen WG. Randomized trials of antioxidant vitaminsand eye disease. Comp Ophthalmol Update 2000;1(1):55–61.

103. Bressler NM, Bressler SB, Fine SL. Age-related maculardegeneration. Surv Ophthalmol 1988; 32(6):375–413.

104. Evans JR, Fletcher AE, Wormald R. 28,000 cases ofage related macular degeneration causing visual loss inpeople aged 75 years and above in the United Kingdommay be attributable to smoking. Br J Ophthalmol 2005;89(5):550–3.

105. Smith W, Mitchell P, Wang JJ. Gender, oestrogen, hormonereplacement and age-relatedmacular degeneration: resultsfrom the BlueMountains Eye Study.AustNZ JOphthalmol1997; 25(Suppl. 1):S13–5.

106. Kini MM, Leibowitz HM, Colton T, Nickerson RJ, Ganley J,Dawber TR. Prevalence of senile cataract diabetic retino-pathy, senile macular degeneration, and open-angleglaucoma in the FraminghamEye Study. Am JOphthalmol1978; 85(1):28–34.

107. Jampol LM, Tielsch J. Race, macular degeneration, and theMacular Photocoagulation Study. Arch Ophthalmol 1992;110(12):1699–700.

108. Gregor Z, Joffe L. Senile macular changes in the blackAfrican. Br J Ophthalmol 1978; 62(8):547–50.

109. Sommer A, Tielsch JM, Katz J, et al. Racial differences inthe cause-specific prevalence of blindness in East Balti-more. N Engl J Med 1991; 325(20):1412–7.

110. Piguet B, Wells JA, Palmvang IB, Wormald R,Chisholm IH, Bird AC. Age-related Bruch’s membranechange: a clinical study of the relative role of heredity andenvironment. Br J Ophthalmol 1993; 77(7):400–3.

111. Ishida O, Oku H, Ikeda T, Nishimura M, Kawagoe K,Nakamura K. Is Chlamydia pneumoniae infection a riskfactor for age related macular degeneration? Br J Ophthal-mol 2003; 87(5):523–4.

112. Klein ML, Mauldin WM, Stoumbos VD. Heredity and age-related macular degeneration: observations in monozy-gotic twins. Arch Ophthalmol 1994; 112(7):932–7.

113. Yoshida A, Yoshida M, Yoshida S, Shiose S, Hiroishi G,Ishibashi T. Familial cases with age-related maculardegeneration. Jpn J Ophthalmol 2000; 44(3):290–5.

114. Hutchinson J, Tay W. Symmetrical central chorio-retinaldisease in senile persons. R Lond Ophthalmol Hosp Rep1875; 8:231–44.

115. Bradley AE. Dystrophy of the macula. The XIX ProctorLecture. Am J Ophthalmol 1966; 61(1):1–24.

116. Gass JDM. Drusen and disciformmacular detachment anddegeneration. Arch Ophthalmol 1973; 90(9):206–17.

117. Seddon JM, Ajani UA, Mitchell BD. Familial aggregationof age-related maculopathy. Am J Ophthalmol 1997;123(2):199–206.

118. Silvestri G, Johnston PB, Hughes AE. Is genetic predis-position an important risk factor in age-related maculardegeneration? Eye 1994; 8(Pt 5):564–8.

119. Klaver CCW, Wolfs RCW, Assink JJM, Duijn CM,Hofman A, de Jong PT. Genetic risk of age-related maculo-pathy: population-based familial aggregation study. ArchOphthalmol 1998; 116(12):1646–51.

120. Linton KLP, Klein BEK, Klein R. The validity of self-reported and surrogate-reported cataract and age-relatedmacular degeneration in the Beaver Dam Eye Study. AmJ Epidemiol 1991; 134(12):1438–46.

121. MelroseMA,Magargal LE, Lucier AC. Identical twinswithsubretinal neovascularization complicating senile maculardegeneration. Ophthalmic Surg 1985; 16(10):648–51.

122. Meyers SM, Zachary AA. Monozygotic twins with age-related macular degeneration. Arch Ophthalmol 1988;106(5):651–3.

123. Dosso AA, Bovet J. Monozygotic twin brothers with age-related macular degeneration. Ophthalmologica 1992;205(1):24–8.

124. Gottfredsdottir MS, Sverrisson T, Musch DC, Stefansson E.Age related macular degeneration in monozygotic twinsand their spouses in Iceland. Acta Ophthalmol Scand 1999;77(4):422–5.

125. De La Paz MA, Pericak-Vance MA, Haines JL, Seddon JM.Phenotypic heterogeneity in families with age-relatedmacular degeneration. Am J Ophthalmol 1997;124(3):331–43.

126. Beatty S, Boulton M, Henson D, Koh H-H, Murray IJ.Macular pigment and age-related macular degeneration.Br J Ophthalmol 1999; 83(7):867–77.

127. Bone RA, Landrum JT, Tarsis SL. Preliminary identifica-tion of the human macular pigment. Vision Res 1985;25(11):1531–5.

128. Nussbaum JJ, Pruett RC, Delori FC. Historic perspectives:macular yellow pigment: the first 200 years. Retina 1981;1(4):296–310.

129. Reading VM, Weale RA. Macular pigment and chromaticaberration. J Opt Soc Am 1974; 64(2):231–4.

130. Kirschfeld K. Carotenoid pigments: their possible rolein protecting against photooxidation in eyes and photo-receptor cells. Proc R Soc Lond B Biol Sci 1982;216(1202):71–85.

131. Bone RA, Landrum JT. Macular pigment in Henle fibermembranes: a model for Haidinger’s brushes. Vision Res1984; 24(2):103–8.

132. Snodderly DM. Evidence for protection against age-related macular degeneration by carotenoids and antiox-idant vitamins. Am J Clin Nutr 1995; 62(Suppl.):1448S–61.

133. Young RW. Solar radiation and age-related maculardegeneration. Surv Ophthalmol 1988; 32(4):252–69.

134. De La Paz MA, Anderson RE. Regional and age-dependent variation in susceptibility of the human retinato lipid peroxidation. Invest Ophthalmol Vis Sci 1992;33(13):3497–9.

135. Winkler BS, Boulton ME, Gottsch JD, Sternberg P. Oxi-dative damage and age-relatedmacular degeneration. MolVis 1999; 5:32 (http://www.molvis.org/molvis/v5/p32).

136. Cai J, Nelson KC, Wu M, Sternberg P, Jones DP. Oxidativedamage and protection of the RPE. Prog Retin Eye Res2000; 19(2):205–21.

137. Landrum JT, Bone RA, KilburnMD. The macular pigment:a possible role in protection from age-related maculardegeneration. Adv Pharmacol 1997; 38:537–56.

138. Bernstein PS, Zhao D-Y, Wintch SW, Ermakov IV,McClane RW, Gellermann W. Resonance Raman measure-ment of macular carotenoids in normal subjects and inage-related macular degeneration patients. Ophthal-mology 2002; 109(10):1780–7.

139. Beatty S, Murray IJ, Henson DB, Carden D, Koh H-H,Boulton ME. Macular pigment and risk for age-relatedmacular degeneration in subjects from a NorthernEuropean population. Invest Ophthalmol Vis Sci 2001;42(2):439–46.

140. Ciulla TA, Hammond BR. Macular pigment density andaging, assessed in the normal elderly and those withcataracts and age-related macular degeneration. AmJ Ophthalmol 2004; 138(4):582–7.

141. Berendschot TTJM, van Norren D. On the age dependencyof the macular pigment optical density. Exp Eye Res 2005;81(5):602–9.


142. Hammond BR, Curran-Celentano J, Judd S, et al. Sexdifferences in macular pigment optical density: relationto plasma carotenoid concentrations and dietary patterns.Vision Res 1996; 36(13):2001–12.

143. Hammond BR, Wooten BR, Snodderly DM. Cigarettesmoking and retinal carotenoids: implications for age-related macular degeneration. Vision Res 1996;36(18):3003–9.

144. Koh H-H, Murray IJ, Nolan D, Carden D, Feather J,Beatty S. Plasma and macular responses to luteinsupplement in subjects with and without age-relatedmaculopathy: a pilot study. Exp Eye Res 2004; 79(1):21–7.

145. Hammond BR, Johnson EJ, Russell RM, et al. Dietarymodification of human macular pigment density. InvestOphthalmol Vis Sci 1997; 38(9):1795–801.

146. Landrum JT, Bone RA, Joa H, Kilburn MD, Moore LL,Sprague KE. A one year study of the macular pigment:the effect of 140 days of a lutein supplement. Exp Eye Res1997; 65(1):57–62.

147. Bone RA, Landrum JT, Guerra LH, Ruiz CA. Lutein andzeaxanthin dietary supplements raise macular pigmentdensity and serum concentrations of these carotenoids inhumans. J Nutr 2003; 133(6):992–8.

148. Sommerburg O, Keunen JEE, Bird AC, van Kuijk FJ. Fruitsand vegetables that are sources for lutein and zeaxanthin:the macular pigment in human eyes. Br J Ophthalmol1998; 82(8):907–10.

149. Klaver CCW, Wolfs RCW, Vingerling JR, Hofman A, deJong PTVM. Age-specific prevalence and causes of blind-ness and visual impairment in an older population: theRotterdam Study. Arch Ophthalmol 1998; 116(5):653–8.

150. Jacques PF. The potential preventive effects of vitamins forcataract and age-related macular degeneration. Int J VitamNutr Res 1999; 69(3):198–205.

151. Christen WG, Manson JE, Seddon JM, et al. A prospectivestudy of cigarette smoking and risk of cataract in men.JAMA 1992; 268(8):989–93.

152. Hiller R, Sperduto RD, Ederer F. Epidemiologic associ-ations with nuclear, cortical, and posterior subcapsularcataracts. Am J Epidemiol 1986; 124(6):916–25.

153. Taylor HR, West SK, Rosenthal FS, et al. Effect of ultra-violet radiation on cataract formation. N Engl J Med 1988;319(22):1429–33.

154. Delcourt C, Carriere I, Ponton-Sanchez A, et al. Lightexposure and the risk of cortical, nuclear, and posteriorsubcapsular cataracts. Arch Ophthalmol 2000;118(3):385–92.

155. Oliver M. Posterior pole changes after cataract extractionin elderly subjects. Am J Ophthalmol 1966; 62(6):1145–8.

156. Blair CJ, Ferguson J. Exacerbation of senile maculardegeneration following cataract extraction. Am J Ophthal-mol 1979; 87(1):77–83.

157. Pollack A, Marcovich A, Bukelman A, Oliver M. Age-related macular degeneration after extracapsular cataractextraction with intraocular lens implantation. Ophthal-mology 1996; 103(10):1546–54.

158. Pollack A, Marcovich A, Bukelman A, Zalish M, Oliver M.Development of exudative age-related macular degener-ation after cataract surgery. Eye 1997; 4(Pt 4):523–30.

159. Pollack A, Bukelman A, Zalish M, Leiba H, Oliver M. Thecourse of age-related macular degeneration followingbilateral cataract surgery. Ophthalmic Surg Lasers 1998;29(4):286–94.

160. van der Schaft TL, Mooy CM, de Bruijn WC, Mulder PGH,Pameyer JH, de Jong PTVM. Increased prevalence of

disciform macular degeneration after cataract extractionwith implantation of an intraocular lens. Br J Ophthalmol1994; 78(6):441–5.

161. Vingerling JR, Klaver CCW, Hofman A, de Jong PTVM.Cataract extraction and age-related macular degeneration:the Rotterdam Study. Invest Ophthalmol Vis Sci 1997;38(Suppl.):S472 (Abstract).

162. Sandberg MA, Gaudio AR, Miller S, Weiner A. Irispigmentation and extent of disease in patients withneovascular age-related macular degeneration. InvestOphthalmol Vis Sci 1994; 35(6):2734–40.

163. Feeney-Burns L, Hilderbrand ES, Eldridge S. Aginghuman RPE: morphometric analysis of macular, equa-torial, and peripheral cells. Invest Ophthalmol Vis Sci1984; 25(2):195–200.

164. Bito LZ, Matheny A, Cruickshanks KJ, Nondahl DM,Carino OB. Eye color changes past early childhood: theLouisville Twin Study. Arch Ophthalmol 1997;115(5):659–63.

165. Snow KK, Seddon JM. Do age-related macular degen-eration and cardiovascular disease share commonantecedents? Ophthalmic Epidemiol 1999; 6(2):125–43.

166. Vidaurri JS, Pe’er J, Halfon S-T, Halperin G, Zauberman H.Association between drusen and some of the risk factorsfor coronary artery disease. Ophthalmologica 1984;188(4):243–7.

167. Willett WC. Diet and health: what should we eat? Science1994; 264(5158):532–7.

168. Klein R, Klein BEK, Marino EK, et al. Early age-relatedmaculopathy in the Cardiovascular Health Study.Ophthalmology 2003; 110(1):25–33.

169. Tomany SC, Wang JJ, van Leeuwen R, et al. Risk factorsfor incident age-related macular degeneration: pooledfindings from 3 continents. Ophthalmology 2004;111(7):1280–7.

170. Landolfo V, Albini L, DeSimone S. Senile macular degener-ation and alteration of the metabolism of the lipids.Ophthalmologica 1978; 177(5):248–53.

171. Albrink MJ, Fasanella RM. Serum lipids in patients withsenile macular degeneration. Am J Ophthalmol 1963;55(4):709–13.

172. Sanders TAB, Haines AP, Wormald R, Wright LA,Obeid O. Essential fatty acids, plasma cholesterol, andfat-soluble vitamins in subjects with age-related maculo-pathy and matched control subjects. Am J Clin Nutr 1993;57(3):428–33.

173. Klein R, Klein BEK, Jensen SC, et al. Medication useand the 5-year incidence of early age-relatedmaculopathy:the Beaver Dam Eye Study. Arch Ophthalmol 2001;119(9):1354–9.

174. Mitchell P, Wang JJ, Foran S, Smith W. Five-year incidenceof age-related maculopathy lesions: the Blue MountainsEye Study. Ophthalmology 2002; 109(6):1092–7.

175. van Leeuwen R, Klaver CCW, Vingerling JR, Hofman A,de Jong PTVM. The risk and natural course of age-relatedmaculopathy: follow-up at 6 1/2 years in the RotterdamStudy. Arch Ophthalmol 2003; 121(4):519–26.

176. Hall NF, Gale CR, Syddall H, Phillips DI, Martyn CN. Riskof macular degeneration in users of statins: cross sectionalstudy. BMJ 2001; 323(7309):375–6.

177. McCarty CA, Mukesh BN, Guymer RH, Baird PN,Taylor HR. Cholesterol-lowering medications reduce therisk of age-related maculopathy progression. Med J Aust2001; 175(6):340.

178. McGwin G, Owsley C, Curcio CA, Crain RJ. The asso-ciation between statin use and age related maculopathy. BrJ Ophthalmol 2003; 87(9):1121–5.

82 AU EONG ET AL.

179. Pauleikhoff D, Wormald RP, Wright L, Wessing A,Bird AC. Macular disease in an elderly population. GerJ Ophthalmol 1992; 1(1):12–5.

180. Seddon JM, Cote J, Rosner B. Progression of age-relatedmacular degeneration: association with body mass index,waist circumference, and waist–hip ratio. Arch Ophthal-mol 2003; 121(6):785–92.

181. Seddon JM, George S, Rosner B, Rifai N. Progression ofage-related macular degeneration: prospective assessmentof C-reactive protein, interleukin 6, and other cardiovas-cular biomarkers. Arch Ophthalmol 2005; 123(6):774–82.

182. Kuo CC, Shor A, Campbell LA, Fukushi H, Patton DL,Grayson JT. Demonstration of Chlamydia pneumoniae inatherosclerotic lesions of coronary arteries. J Infect Dis1993; 167(4):841–9.

183. Campbell LA, O’Brien ER, Cappuccio AL, et al. Detectionof Chlamydia pneumoniae TWAR in human coronary ather-ectomy tissues. J Infect Dis 1995; 172(2):585–8.

184. Kol A, Sukhova GK, Lichtman AH, Libby P. Chlamydialheat shock protein 60 localizes in human atheroma andregulates macrophage tumor necrosis factor-alpha andmatrix metalloproteinase expression. Circulation 1998;98(4):300–7.

185. Kalayoglu MV, Bula D, Arroyo J, Gragoudas ES,D’Amico D, Miller JW. Identification of Chlamydia pneu-moniae within human choroidal neovascular membranessecondary to age-related macular degeneration. GraefesArch Clin Exp Ophthalmol 2005; 243(11):1080–90.

186. Robman L, Mahdi OS, McCarty C, et al. Exposure toChlamydia pneumoniae infection and progression of age-related macular degeneration. Am J Epidemiol 2005;161(11):1013–9.

187. Klein BEK, Klein R, Jensen SC, Ritter LL. Are sexhormones associated with age-related maculopathy inwomen? The Beaver Dam Eye Study. Trans Am Ophthal-mol Soc 1994; 92:289–97.

188. Thornton J, Edwards R, Mitchell P, Harrison RA, Buchan I,Kelly SP. Smoking and age-related macular degeneration:a review of association. Eye 2005; 19(9):935–44.

189. Paetkau ME, Boyd TA, Grace M, Bach-Mills J, Winship B.Senile disciform macular degeneration and smoking. CanJ Ophthalmol 1978; 13(2):67–71.

190. Vingerling JR, Hofman A, Grobbee DE, de Jong PT. Age-related macular degeneration and smoking: the RotterdamStudy. Arch Ophthalmol 1996; 114(10):1193–6.

191. Kelly SP, Thornton J, Lyratzopoulos G, Edwards R,Mitchell P. Smoking and blindness. BMJ 2004;328(7439):537–8.

192. Smith W, Assink J, Klein R, et al. Risk factors for age-related macular degeneration: pooled findings from threecontinents. Ophthalmology 2001; 108(4):697–704.

193. Stryker WS, Kaplan LA, Stein EA, Stampfer MJ, Sober A,Willett WC. The relation of diet, cigarette smoking, andalcohol consumption to plasma beta-carotene and alpha-tocopherol levels. Am J Epidemiol 1988; 127(2):283–96.

194. Espinosa-Heidmann DG, Suner IJ, Catanuto P,Hernandez EP, Marin-Castano ME, Cousins SW. Cigarettesmoke-related oxidants and the development of sub-RPEdeposits in an experimental animal model of dry AMD.Invest Ophthalmol Vis Sci 2006; 47(2):729–37.

195. Suner IJ, Espinosa-Heidmann DG, Marin-Castano ME,Hernandez EP, Pereira-Simon S, Cousins SW. Nicotineincreases size and severity of experimental choroidalneovascularization. Invest Ophthalmol Vis Sci 2004;45(1):311–7.

196. Bidwell G, Sahu A, Edwards R, Harrison RA, Thornton J,Kelly SP. Perceptions of blindness related to smoking: ahospital-based cross-sectional study. Eye 2005; 19(9):945–8.

197. Ham WT, Mueller HA, Ruffolo JJ, Guerry DI, Guerry RK.Action spectrum for retinal injury from near-ultravioletradiation in aphakic monkey. Am J Ophthalmol 1982;93(3):299–306.

198. Noell WK, Walker VS, Kang BS, Berman S. Retinal damageby light in rats. Invest Ophthalmol 1966; 5(5):450–73.

199. Margrain TH, Boulton M, Marshall J, Sliney DH. Do bluelight filters confer protection against age-related maculardegeneration? Prog Retin Eye Res 2004; 23(5):523–31.

200. Tso MOM. Pathogenetic factors of aging macular degener-ation. Ophthalmology 1985; 92(5):628–35.

201. Noell WK. Possible mechanisms of photoreceptordamage by light in mammalian eyes. Vision Res 1980;20(12):1163–71.

202. Lawwill T. Three major pathologic processes caused bylight in the primate retina: a search for mechanisms. TransAm Ophthalmol Soc 1982; 80:517–79.

203. Borges J, Li Z-Y, Tso MOM. Effects of repeated photicexposures on the monkey macula. Arch Ophthalmol 1990;108(5):727–33.

204. Tso MOM. Photic maculopathy in rhesus monkey: a lightand electron microscopic study. Invest Ophthalmol 1973;12(1):17–34.

205. Ewald RA, Ritchey CL. Sun gazing as the cause of foveo-macular retinitis. Am J Ophthalmol 1970; 70(4):491–7.

206. Gottsch JD, Bynoe LA, Harlan JB, Rencs EV, Green WR.Light-induced deposits in Bruch’s membrane of proto-porphyric mice. Arch Ophthalmol 1993; 111(1):126–9.

207. Tso MOM, Woodford BJ. Effect of photic injury on theretinal tissues. Ophthalmology 1983; 90(8):952–63.

208. Green WR, Enger C. Age-related macular degenerationhistopathologic studies. The 1992 Lorenz E Zimmermanlecture. Ophthalmology 1993; 100(10):1519–35.

209. Gottsch JD, Pou S, Bynoe LA, Rosen GM. Hematogenousphotosensitization: a mechanism for the development ofage-related macular degeneration. Invest Ophthalmol VisSci 1990; 31(9):1674–82.

210. Sperduto RD, Ferris FLI, Kurinij N. Do we have a nutri-tional treatment for age-related cataract or maculardegeneration? Arch Ophthalmol 1990; 108(10):1403–5(editorial).

211. The Alpha-Tocopherol BCCPSG. The effect of vitamin Eand beta carotene on the incidence of lung cancer andother cancers in male smokers. N Engl J Med 1994;330(15):1029–35.

212. Omenn GS, Goodman GE, Thornquist MD, et al. Effects ofa combination of beta-carotene and vitamin A on lungcancer and cardiovascular disease. N Engl J Med 1996;334(18):1150–5.

213. Robison WG, Kuwabara T, Bieri JG. The roles of vitamin Eand unsaturated fatty acids in the visual process. Retina1982; 2(4):263–81.

214. Christen WG. Antioxidants and eye disease. Am J Med1994; 97(Suppl. 3A):14S–7.

215. Frei B. Reactive oxygen species and antioxidant vitamins:mechanisms of action. Am J Med 1994; 97(Suppl.3A):5S–13.

216. Hayes KC. Retinal degeneration in monkeys induced bydeficiencies of vitamin E or A. Invest Ophthalmol Vis Sci1974; 13(7):499–510.

217. Organisciak DT, Wang HM, Li Z-Y, Tso MOM. Theprotective effect of ascorbate in retinal light damage ofrats. Invest Ophthalmol Vis Sci 1985; 26(11):1580–8.


218. Horwitt MK, Harvey CC, Dahm CH, Scarey MT. Relation-ship between tocopherol and serum lipid levels fordetermination of nutritional adequacy. Ann NY Acad Sci1972; 203:223–36.

219. Thurnham DI, Davies JA, Crump BJ, Situnayake RD,Davis M. The use of different lipids to express serumtocopherol: lipid ratios for the measurement of vitamin Estatus. Ann Clin Biochem 1986; 23(Pt 5):514–20.

220. Jordan P, Brubacher D, Moser U, Stahelin HB, Gey KF.Vitamin E and vitamin A concentrations in plasmaadjusted for cholesterol and triglycerides by multipleregression. Clin Chem 1995; 41(6 Pt 1):924–7.

221. van Leeuwen R, Boekhoorn S, Vingerling JR, et al. Dietaryintake of antioxidants and risk of age-related maculardegeneration. JAMA 2005; 294(24):3101–7.

222. Age-Related Eye Disease Study Research Group.A randomized, placebo-controlled, clinical trial of high-dose supplementation with vitamins C and E, beta caro-tene, and zinc for age-related macular degeneration andvision loss: AREDS report no. 8. Arch Ophthalmol 2001;119(10):1417–36.

223. Mares-Perlman JA, Klein R, Klein BEK, et al. Associationof zinc and antioxidant nutrients with age-related maculo-pathy. Arch Ophthalmol 1996; 114(8):991–7.

224. Galin MA, Nano HD, Hall T. Ocular zinc concentration.Invest Ophthalmol 1962; 1(1):142–8.

225. Sigel H. Zinc and Its Role in Biology and Nutrition. NewYork: Marcel Dekker, 1983.

226. Silverstone BZ, Landau L, Berson D, Sternbuch J. Zinc andcopper metabolism in patients with senile maculardegeneration. Ann Ophthalmol 1985; 17:419–22.

227. Briefel RR, Bialostosky K, Kennedy-Stephenson J,McDowell MA, Ervin RB, Wright JD. Zinc intake of theU.S. population: findings from the Third National Healthand Nutrition Examination Survey, 1988–1994. J Nutr2000; 130(Suppl. 5S):1367S–73.

228. Age-Related Macular Degeneration Study ResearchGroup. The Age-Related Eye Disease Study: a clinicaltrial of zinc and antioxidants-Age-Related Eye DiseaseStudy report no. 2. J Nutr 2000; 130(Suppl. 5S):1516S–9.

229. Newsome DA, Swartz M, Leone NC, Elston RC, Miller E.Oral zinc in macular degeneration. Arch Ophthalmol 1988;106(2):192–8.

230. Stur M, Tittl M, Reitner A, Meisinger V. Oral zinc and thesecond eye in age-related macular degeneration. InvestOphthalmol Vis Sci 1996; 37(7):1225–35.

231. Richer S. Multicenter ophthalmic and nutritional age-related macular degeneration study—part 1: design,subjects and procedures. J Am Optom Assoc 1997;67(1):12–29.

232. Richer S. Multicenter ophthalmic and nutritional age-related macular degeneration study—part 2: antioxidantintervention and conclusions. J Am Optom Assoc 1996;67(1):30–49.

233. Christen WG, Gaziano JM, Hennekens CH. Design ofPhysicians’ Health Study II—a randomized trial ofbeta-carotene, vitamins E and C, and multivitamins, inprevention of cancer, cardiovascular disease, and eyedisease, and review of results of completed trials. AnnEpidemiol 2000; 10(2):125–34.

234. Garrett SKM, McNeil JJ, Silagy C, et al. Methodology ofthe VECAT study: vitamin E intervention in cataractand age-related maculopathy. Ophthalmic Epidemiol1999; 6(3):195–208.

235. Taylor HR, Tikellis G, Robman LD, McCarty CA,McNeil JJ. Vitamin E supplementation and macular

degeneration: randomized controlled trial. BMJ 2002;325(7354):11.

236. Buring JE, Hennekens CH. The Women’s Health Study:summary of the study design. J Myocardial Ischemia 1992;4(3):27–9.

237. Manson JE, Gaziano JM, Spelsberg A, et al. A secondaryprevention trial of antioxidant vitamins and cardiovas-cular disease in women: rationale, design, and methods.Ann Epidemiol 1995; 5(4):261–9.

238. Bressler NM, Bressler SB, Congdon NG, et al. Potentialpublic health impact of Age-Related Eye Disease Studyresults: AREDS report no. 11. Arch Ophthalmol 2003;121(11):1621–4.

239. Hall NF, Gale CR. Prevention of age related maculardegeneration. BMJ 2002; 325(7354):1–2.

240. Teikari JM, Laatikainen L, Virtamo J, et al. Six-yearsupplementation with alpha-tocopherol and beta-caroteneand age-related maculopathy. Acta Ophthalmol Scand1998; 76(2):224–9.

241. van Kuijk FJ, Buck P. Fatty acid composition of the humanmacula and peripheral retina. Invest Ophthalmol Vis Sci1992; 33(13):3493–6.

242. Kromhout D, Bosschieter EB, de Lezenne Coulander C.The inverse relation between fish consumption and20-year mortality from coronary heart disease. N EnglJ Med 1985; 312(19):1205–9.

243. Sanders TA, Sullivan DR, Reeve J, Thompson GR. Trigly-ceride-lowering effect of marine polyunsaturates inpatients with hypertriglyceridemia. Arteriosclerosis 1985;5(5):459–65.

244. Katan MB. Fish and heart disease. N Engl J Med 1995;332(15):1024–5.

245. Ajani U, Willett W, Miller D, et al. Alcohol consumptionand neovascular age-related macular degeneration. AmJ Epidemiol 1993; 138(8):646 (Abstract).

246. Gaziano JM, Buring JE, Breslow JL, et al. Moderatealcohol intake, increased levels of high-density lipopro-tein and its subfractions, and decreased risk ofmyocardial infarction. N Engl J Med 1993;329(25):1829–34.

247. Ritter LL, Klein R, Klein BEK, Mares-Perlman JA,Jensen SC. Alcohol use and age-related maculopathy inthe Beaver Dam Eye Study. Am J Ophthalmol 1995;120(2):190–6.

248. Klein R, Klein BEK, Tomany SC, Moss SE. Ten-yearincidence of age-related maculopathy and smoking anddrinking: the Beaver Dam Eye Study. Am J Epidemiol2002; 156(7):589–98.

249. Smith W, Mitchell P. Alcohol intake and age-relatedmaculopathy. Am J Ophthalmol 1996; 122(5):743–5.

250. Lavin MJ, Eldem B, Gregor ZJ. Symmetry of disciformscars in bilateral age-related macular degeneration. Br JOphthalmol 1991; 75(3):133–6.

251. Chang B, Yannuzzi LA, Ladas ID, Guyer DR, Slakter JS,Sorenson JA. Choroidal neovascularization in secondeyes of patients with unilateral exudative age-relatedmacular degeneration. Ophthalmology 1995;102(9):1380–6.

252. Ferris FLI, Fine SL, Hyman L. Age-related maculardegeneration and blindness due to neovascular maculo-pathy. Arch Ophthalmol 1984; 102(11):1640–2.

253. Abdelsalam A, Del Priore L, Zarbin MA. Drusen in age-related macular degeneration: pathogenesis, naturalcourse, and laser photocoagulation—induced regression.Surv Ophthalmol 1999; 44(1):1–29.

254. Smiddy WE, Fine SL. Prognosis of patients with bilateralmacular drusen. Ophthalmology 1984; 91(3):271–7.

84 AU EONG ET AL.

255. Holz FG, Wolfensberger TJ, Piguet B, et al. Bilateralmacular drusen in age-related macular degeneration:prognosis and risk factors. Ophthalmology 1994;101(9):1522–8.

256. Bressler NM, Munoz B, Maguire MG, et al. Five-yearincidence and disappearance of drusen and retinalpigment epithelial abnormalities: Waterman Study. ArchOphthalmol 1995; 113(3):301–8.

257. Teeters VW, Bird AC. The development of neovasculariza-tion of senile disciform macular degeneration. AmJ Ophthalmol 1973; 76(1):1–18.

258. Chandra SR, Gragoudas ES, Friedman E, Van Buskirk EM,KleinML. Natural history of disciform degeneration of themacula. Am J Ophthalmol 1974; 78(4):579–82.

259. Gragoudas ES, Chandra SR, Friedman E, Klein ML, VanBuskirk EM. Disciform degeneration of the macula: II.Pathogenesis. Arch Ophthalmol 1976; 94(5):755–7.

260. Gregor Z, Bird AC, Chisholm IH. Senile disciformmaculardegeneration in the second eye. Br J Ophthalmol 1977;1977(2):141–7.

261. Strahlman ER, Fine SL, Hillis A. The second eye of patientswith senile macular degeneration. Arch Ophthalmol 1983;101(8):1191–3.

262. Baun O, Vinding T, Krogh E. Natural course in fellow eyesof patients with unilateral age-related exudative maculo-pathy: a fluorescein angiographic 4-year follow-up of45 patients. Acta Ophthalmol 1993; 71(3):398–401.

263. Sandberg MA, Weiner A, Miller S, Gaudio AR. High-riskcharacteristics of fellow eyes of patients with unilateralneovascular age-related macular degeneration. Ophthal-mology 1998; 105(3):441–7.

264. Roy M, Kaiser-Kupfer M. Second eye involvement in age-related macular degeneration: a four-year prospectivestudy. Eye 1990; 4(Pt 6):813–8.

265. Macular Photocoagulation Study Group. Five-yearfollow-up of fellow eyes of patients with age-relatedmacular degeneration and unilateral extrafoveal chor-oidal neovascularization. Arch Ophthalmol 1993;111(9):1189–99.

266. Macular Photocoagulation Study Group. Risk factors forchoroidal neovascularization in the second eye of patientswith juxtafoveal or subfoveal choroidal neovasculariza-tion secondary to age-related macular degeneration. ArchOphthalmol 1997; 115(6):741–7.

267. Lanchoney DM, Maguire MG, Fine SL. A model of theincidence and consequences of choroidal neovasculari-zation secondary to age-related macular degeneration:comparative effects of current treatment and potentialprophylaxis on visual outcomes in high-risk patients.Arch Ophthalmol 1998; 116(8):1045–52.

268. Clemons TE, Milton RC, Klein R, Seddon JM, Ferris FLI.Age-Related Eye Disease Study Research Group. Riskfactors for the incidence of advanced age-related maculardegeneration in the Age-Related Eye Disease Study(AREDS): AREDS report no. 19. Ophthalmology 2005;112(4):533–9.

269. Bressler SB, Maguire MG, Bressler NM, Fine SL, MacularPhotocoagulation Study Group. Relationship of drusenand abnormalities of the retinal pigment epithelium tothe prognosis of neovascular macular degeneration.Arch Ophthalmol 1990; 108(10):1442–7.


5

Choroidal NeovascularizationFrances E. KaneAlimera Sciences, Inc., Alpharetta, Georgia, U.S.A.

Peter A. CampochiaroDepartments of Ophthalmology and Neuroscience, Johns Hopkins University School of Medicine,


INTRODUCTION

Despite substantial recent progress in treatmentdevelopment, choroidal neovascularization (CNV)remains one of the most challenging problems facedby retina specialists. It is a common cause of severevisual loss in patients with age-related maculardegeneration (AMD) and younger patients with oneof many diseases that affect the choroid–Bruch’smembrane–retinal pigment epithelium (RPE)complex, including but not limited to ocular histoplas-mosis, myopic degeneration, angioid streaks, andmultifocal choroiditis. As our understanding of themolecular pathogenesis of CNV is increasing, newtreatments are being developed which specificallytarget molecules that are involved. Therefore, it iscritical to continue to elucidate the molecularmechanisms involved in CNV.

INFERENCES FROM NEOVASCULARIZATIONELSEWHERE IN THE BODY

Neovascularization (NV) is a critical process duringembryonic development and wound repair and occursin almost all tissues of the body. It is well-tolerated inmost tissues, but not in the eye where normal func-tioning depends upon maintenance of blood–ocularbarriers. NV varies somewhat in different tissuesbecause endothelial cells differ in different parts ofthe body and surrounding cells participate in theneovascular response resulting in tissue-specificaspects [see (1) for review]. One thing that is similarfor several disease processes in many tissues is thatvascular endothelial growth factor-A (VEGF-A) playsa central role as a stimulator of NV.

VASCULAR ENDOTHELIAL GROWTH FACTOR

Increased expression of VEGF-A in the retina is suf-ficient to cause sprouting of new vessels from the deepcapillary bed of the retina, but is not sufficient to induce

NVat the retinal surface typical of that seen in ischemicretinopathies (2,3). Likewise, increased expression ofVEGF-A in RPE cells is not sufficient to cause sproutingof new vessels from the choroid (4). Thus, other factorsare involved in the initiation of new vessel growth inthe retina and choroid, but despite any contribution byother factors, VEGF antagonists strongly suppressischemia-induced retinal NV or CNV in animalmodels (5–11). In order to determine the magnitudeof the contribution of VEGF-A, complete blockade ofVEGF would be needed, which is not possible bypharmacologic means since no drug treatment is100% efficient. Also, many antagonists are not totallyselective making it difficult to know how much inhibi-tory effect can be attributed to blockade of VEGF.Soluble VEGF receptors are very efficient whenexpressed by gene transfer or when given systemicallyso that sustained serum levels are achieved, and theyare relatively specific, although they cannot distinguisheffects of VEGF-A and placental growth factor (PlGF).In a mouse model of laser-induced CNV, systemicadministration of VEGF-trap, a chimeric protein thathas binding domains from VEGF receptors 1 and 2,resulted in 66% inhibition (11) and periocular genetransfer of soluble VEGF receptor 1 resulted in 86%inhibition (12). Intraocular gene transfer of solubleVEGF receptor 1 resulted in 53% inhibition ofischemia-induced retinal NV (13). These data suggestthat VEGF family members account for a large portionof thedriving force for these two types of ocularNVandare key targets for therapeutic intervention.

These predictions have been largely substan-tiated in clinical trials. Intravitreous injection ofpegaptanib, an aptamer that binds only the VEGF165isoform, reduced the percentage of patients withclassic CNV due to AMD who experienced moderateloss of vision (loss of 15 letters or more) over the courseof a year from 45% in the sham injection group to 30%(14). Six percent of patients treated with pegaptanibcompared to 2% in the sham injection group had a

substantial improvement in vision (gain of 15 or moreletters). Compared to sham treatment, increase in sizeof CNV lesions was slowed, but not stopped. Ranibi-zumab is an Fab fragment of an antibody that binds allisoforms of VEGF-A. Monthly intravitreous injectionsof ranibizumab in AMD patients with occult or mini-mally classic subfoveal CNVreduced the percentage ofpatients with moderate loss of vision over the courseof a year from 38% in the sham injection group to 5%,and the percentage of patients who experiencedsubstantial improvement in vision was increasedfrom 4.6% to 34% (15). These data suggest thatantagonism of VEGF-A in AMD patients with CNVcan result in stabilization of vision in the majority ofpatients and substantial improvement in vision inabout one-third of patients. These results confirmthat VEGF-A is a very important target in the treat-ment of neovascular AMD, but suggest severalquestions. Are the superior results with intravitreousinjection of ranibizumab compared to those withpegaptanib due to the inhibition of all VEGF-Aisoforms compared to inhibition of only VEGF165,superior pharmacokinetics, a combination of both, orsome other reason? Are there alternative modes ofdelivery of VEGF antagonists that provide superiorpharmacokinetics compared to repeated intraocularinjections? What is the anatomic basis for the visualimprovement with ranibizumab treatment? Clinicalobservations have suggested that a substantialamount of the initial improvement due to VEGFantagonists is related to reduction in excessivevascular permeability resulting in reduction inretinal thickness and subretinal fluid as visualized byoptical coherence tomography (OCT). A case series ofAMD patients with subfoveal CNV treated with sys-temic infusions of bevacizumab, a full-lengthhumanized monoclonal antibody that binds allisoforms of VEGF-A, showed rapid reduction inretinal thickening and subretinal fluid visualized byOCT and an average improvement in visual acuity of12 letters over the course of 12 weeks (16). Canimprovement be sustained long-term (over severalyears) if the CNV is not eliminated? Can additionalimprovement be achieved by inhibiting PlGF as wellas all isoforms of VEGF-A? Clinical trials investigatingthe efficacy of VEGF-trap should answer this question.Can greater benefit be achieved by combiningantagonism of VEGF with other antiangiogenicagents that work by other mechanisms? Is it possibleto achieve drug-induced involution of CNV andwould that result in greater benefits than simplysuppressing leakage and growth of CNV? Clearlygreat progress has been made, but there is still agreat deal of work to do.

The studies described above have conclu-sively shown that VEGF-A is a critical stimulus for

development of CNV in patients with AMD. Whatabout CNV that occurs in other disease processes, suchas pathologic myopia, ocular histoplasmosis, angioidstreaks, multifocal choroiditis, and others? Recently,two patients with subfoveal CNV due to pathologicmyopia were treated with four or five infusions ofbevacizumab, resulting in resolution of retinal thicken-ing assessed by OCT, elimination of leakage andreduction in size of CNV lesions assessed by fluor-escein angiography, and improvement in visual acuity(17). These patients have not had evidence of leakageor other signs of recurrent CNV for over nine months.This finding suggests that VEGF is an importantstimulus for CNV growth and maintenance in patho-logic myopia and it appears that in that diseaseprocess as opposed to AMD, blockage of VEGF canresult in complete involution of CNV. Additionalstudies are needed to determine if this is in fact thecase and whether antagonism of VEGF providesbenefits in patients with CNV due to all causes.

Other Soluble Proangiogenic FactorsBased upon in vitro assays and in vivo effects in sometissues in addition to VEGFs, some other proteins havebeen demonstrated to have proangiogenic acitivity,including fibroblast growth factors (FGFs) (18),tumor necrosis factor-a (19), insulin-like growthfactor-1 (20,21), and hepatocyte growth factor (22).The FGFs do not participate in CNV (23), but it ispossible that the other factors may; however, despitetheir possible participation, blockade of VEGF has aprofound effect on CNV. It will be interesting todetermine if inhibition of one or more of thesefactors combined with antagonism of VEGF providesadded benefit.

Soluble Antiangiogenic FactorsIn many tissues, including the eye, new vessel growthappears to be regulated by a balance between proan-giogenic and antiangiogenic factors.

Transforming growth factor-b and related familymembers inhibit endothelial cell migration andproliferation in vitro, but have been suggested to beproangiogenic or antiangiogenic in vivo, depending onthe context (24–26). Several purported endogenousinhibitors of angiogenesis have been described,including angiostatin (27), endostatin (28), antith-rombin III (29), platelet factor 4 (30), thrombo-spondin (31), and pigment epithelium-derived factor(PEDF) (32).

Signals from the Extracellular MatrixAlong with soluble proangiogenic and antiangiogenicfactors, extracellular matrix (ECM) molecules also

88 KANE AND CAMPOCHIARO

participate in several ways in the regulation of NV.Acting through integrins on the surface of endothelialcells, ECM molecules may directly stimulate or inhibitendothelial cell processes involved in angiogenesis(33). Soluble angiogenic factors exert some of theireffects through ECM molecules by altering expressionof integrins on endothelial cells. Endothelial cells ofdermal vessels have increased expression of avb3integrin when participating in angiogenesis and avb3antagonists block angiogenesis (34). Integrin avb3 isupregulated in endothelial cells participating in retinalNV and avb3 antagonists suppress retinal NV (35).Integrin a5b1 is upregulated in CNV and a smallmolecule antagonist of a5b1 causes regression ofestablished NV by inducing apoptosis of endothelialcells within the NV (36).

Signals from the ECM are often unmasked oreliminated by proteolysis. Components of the ECMmay bind and sequester soluble factors, preventingthem from activating receptors on endothelial cellsuntil they are released by proteolysis (37–39).Degradation of ECM also liberates fragments withantiangiogenic activity that provide negative feedbackslowing vessel growth, making it more orderly, andeventually helping to turn it off and reestablish quies-cence. Endostatin was the first collagen fragmentdemonstrated to inhibit angiogenesis (28), but sub-sequently several others have been identified (40–46).Interestingly, several of these antiangiogenic peptidesare derived from noncollagenous (NC1) domains ofthe basement membrane collagens IV, XV, and XVIII.The NC1 domains are important for assembly of thesupramolecular structures of the collagens and undernormal circumstances do not interact with cells(47–49). However, after cleavage from native collagensseveral of the NC1 domains bind endothelial cells andinhibit angiogenesis. Endostatin is derived from theNC1 domain of collagen XVIII and restin is a some-what similar antiangiogenic peptide derived from theNC1 domain of collagen XV (40). Collagen IV isunusual in that there are six distinct collagen IVchains that have different tissue distributions (50–54).The NC1 domains of several of the collagen IV chains,including 1, 2, 3, and 6 have antiangiogenic activity,but effects may vary in different organs (41–46). In theeye, the NC1 domain of a2(IV) causes regression ofCNV (55).

Two proteolytic systems that play a prominentrole in angiogenesis are the urokinase type of plasmi-nogen activator (56) and matrix metalloproteinases(57,58), and the relative importance of these systemscould vary in different types of angiogenesis. Tissueinhibitor of metalloproteinases-1 (TIMP-1) has beentouted as an inhibitor of NV (59), but it stimulatesVEGF-induced NV in the retina (60).

Transcription Factors that Participate in NVThe clinical observation that retinal NV almost alwaysoccurs in association with retinal capillary nonperfu-sion led to the hypothesis that retinal ischemia is thedriving force (61–63). This hypothesis is supported byexperimental models in which damage to retinalvessels leads to retinal NV (64–68). Advances in theunderstanding of hypoxia-mediated gene regulationhave suggested potential molecular signals such ashypoxia-inducible factor-1 (HIF-1), involvement ofwhich has been confirmed by experimental studies(69). As a result, many of the molecular signalsinvolved in retinal NV have been defined [for reviewsee (70)].

Hypoxia has not been definitely implicated in theoccurrence of CNV. While there is evidence thatchoroidal blood flow is decreased in patients withAMD, it is not clear whether the decrease is sufficientto cause hypoxia of photoreceptors and RPE (71,72).Furthermore, hypoxia cannot be invoked in patientswith ocular histoplasmosis, myopic degeneration,angioid streaks, or many other diseases in whichyoung people get CNV. However, whether or nothypoxia plays a role, it appears that HIF-1 is involved.Mice in which the hypoxia response element, throughwhich HIF-1 acts, has been deleted from the VEGFpromoter are protected from laser-induced CNV. Thisindicates that HIF-1-induced stimulation of VEGF isnecessary for CNV (73).

The JAK–STAT pathway has been implicated inangiogenesis in some tissues (74). In the retina, Leptin,an adipocyte-derived hormone, stimulates NV bySTAT-3-mediated enhancement of VEGF expression(75). The proinflammatory cytokine IL-6 increasesexpression of VEGF by activating STAT-3 (76). STAT-5 may work downstream of VEGF to enhance endo-thelial survival in the setting of hypoxia and therebypromote angiogenesis (77).

The Pathogenesis of CNVOne thing that patients with CNV share is that they allhave abnormalities of Bruch’s membrane and the RPE.In patients with AMD, pathologic studies havedemonstrated that diffuse thickening of Bruch’smembrane is highly associated with the occurrenceof CNV (78). Large soft drusen and pigmentaryabnormalities are clinical risk factors for CNV (79);soft drusen indicate the presence of diffuse subRPEdeposits and pigmentary changes suggest compro-mise of the RPE. Therefore, there is disorderedmetabolism of ECM in patients with AMD that maycompromise RPE cells leading to cell dropout andproliferation, and CNV. Breaks in Bruch’s membraneand/or other abnormalities of the ECM of RPE cellsoccur in other diseases in which CNVoccurs. Patientswith Sorsby’s fundus dystrophy have a mutation in

5: CHOROIDAL NEOVASCULARIZATION 89

the TIMP-3 gene that results in abnormal processing ofthe protein so that it is deposited along Bruch’smembrane (80). This collection of an ectopic proteinalong Bruch’s membrane is associated with RPE andphotoreceptor degeneration and a high incidence ofCNV (81,82).

Why would abnormal ECM along the basalsurface of RPE cells result in cell compromise andCNV? Like most epithelial cells, the phenotype andbehavior of RPE cells is regulated in part by interactionwith its ECM. Cultured RPE cells display alterations inmorphology and gene expression when grown ondifferent ECMs (83). Presentation of some ECMmolecules such as vitronectin or thrombospondin tothe apical or basal surface of RPE cells results in smallincreases in FGF-2 and large increases in VEGF in themedia of the cells (84). Therefore, alterations in theECM of RPE cells can cause them to increase pro-duction of proteins with angiogenic activity.

Defects in Bruch’s membrane contribute to CNV.In wild-type mice, laser-induced rupture of Bruch’smembrane results in CNV (23). In rho/VEGF or rho/FGF2 transgenic mice, rupture of Bruch’s membraneresulted in very large areas of CNV, much larger thanthose in wild-type mice (85). Low-intensity laser,which ruptured photoreceptor cells but did notrupture Bruch’s membrane, resulted in CNV in rho/FGF2 mice, but not rho/VEGF or wild-type mice.These experiments demonstrate that choroidalvessels are capable of responding to excess VEGF orextracellular FGF2 when there is a concomitantrupture of Bruch’s membrane. This suggests thatBruch’s membrane constitutes a mechanical and bio-chemical barrier to CNV. Increased expression ofVEGF or FGF2 is unable to cause a breech in thebarrier. In the case of FGF2, sequestration is likely tobe an important control mechanism, because low-intensity laser that ruptures photoreceptor cells andreleases FGF2, but does not rupture Bruch’smembrane, results in CNV. This is not the case forVEGF, which stimulates CNV only when the Bruch’smembrane barrier has been disrupted by anothermeans.

The importance of the Bruch’s membrane barrierfor prevention of CNV may help to explain difficultiesin modeling CNV. Laser-induced rupture of Bruch’smembrane, first established in primates and lateradapted to rodents, has been widely used (23,86,87).All other models of CNV, whether they involveimplantation of sustain release polymers or genetransfer, have a component of surgical damage toBruch’s membrane (88,89). Therefore, some sort ofcompromise of Bruch’s membrane must accompanyincreased levels of angiogenic factors in order togenerate CNV.

Laser-induced rupture of CNV in mice (23) hasprovided a particularly valuable tool, because it can beused in genetically engineered mice to explore the roleof individual gene products. Using this strategy, Ozakiet al. (90) demonstrated that mice with targeteddeletion of FGF2 develop CNV similar to that inwild-type mice indicating that FGF2 is not necessaryfor the development of CNV after rupture of Bruch’smembrane. This approach was also used to demon-strate that nitric oxide (NO) is proangiogenic in boththe retina and the choroid, but different isoforms ofNO synthetase play a role (91). For retinal NV, eNOSplays an important role, while for CNV, nNOS isimportant. This suggests that nitric oxide synthaseinhibitors may be useful in patients at risk for CNV.

Pharmacologic Treatments for CNVRanibizumab, a potent antagonist of VEGF-A, is thefirst treatment to improve visual acuity in a substantialproportion of patients with neovascular AMD. It islikely that VEGF antagonists will remain the basis oftreatment, but improvements may be made in mode ofdelivery. Orally active VEGF antagonists are being tobe tested. New agents will be added if they canprovide additional benefits to treatment with VEGFantagonists alone.

An appealing approach is to reduce leakage andstop growth of CNV with intraocular injections of aVEGF antagonist and then maintain stability with aless invasive approach. Topically active drugs wouldbe ideal. Amfenac, 2-amino-3-benzoylbenzeneaceticacid, is an inhibitor of cyclooxygenase-1 (COX-1) andCOX-2 that strongly suppresses pain (92). Nepafenac,the amide analog of amfenac, has unusually highocular penetration and acts as a prodrug that signi-ficantly inhibits prostaglandin synthesis in the retina/choroid by 55% for four hours after topical adminis-tration and blocks ocular inflammation (93,94).Topically administered Nepafenac also reducesexpression of VEGF and inhibits the development ofischemia-induced retinal NV and CNV due to ruptureof Bruch’s membrane (95). This is consistent withrecent studies that have demonstrated that increasedCOX activity enhances and COX inhibitors reduceVEGF expression in several other tissues (96–98).

Agents that block NV by mechanisms distinctfrom those of VEGF antagonists are good candidatesfor combination therapy. Polyamine analogs blockpolyamine metabolism, which is required by all prolif-erating cells including endothelial cells participatingin NV. Intravitreous or periocular injections of poly-amine analogs induced regression of established CNVby inducing apoptosis in endothelial cells partici-pating in CNV (99). Over a seven-day treatmentperiod, the regression was not complete and could


not be increased beyond 40% by increasing the dose ofpolyamine analogs or by combining them withDL-alpha-difluoromethylornithine, an inhibitor ofpolyamine biosynthesis. To provide perspective onthis effect, intravitreous injection of adenoviralvectors expressing PEDF caused a similar amount ofregression in the same model over a 10-day period(100) and over a seven-day period combretastatin-A-4-phosphate, a vascular targeting agent, caused 66%regression of CNV (101). Intraocular injections ofpolyamine analogs cause apoptosis of some retinalneurons, but after periocular injections only endo-thelial cells participating in CNV are affected andretinal function assessed by electroretinograms(ERGs) remains normal. Therefore, periocular injec-tion of polyamine analogs deserves further study.

VEGF promotes the survival of endothelial cellsin newly formed vessels (102). Over time endothelialcells in new vessels become less dependent uponVEGF for survival, because they obtain new sourcesof survival signals. The ECM is a major source ofsurvival signals and blockade of those signals islikely to enhance the effects of VEGF antagonists.This is particularly true for CNV in which endothelialcells seem to achieve independence in terms ofsurvival from VEGF more rapidly than in retinal NV,possibly because of the exuberant ECM associatedwith CNV. One of the receptors on endothelial cellsthat mediates survival signals from ECM is integrina5b1, and a small molecule that binds a5b1 andprevents its interaction with ECM componentscauses regression of CNV (36). This and other agentsthat target a5b1 are good candidates for combinationtherapy. Some of the endogenous antiangiogenicproteins, such as endostatin or the NC1 domain ofthe a2 chain of collagen IV, block survival signals fromthe ECM. Intraocular or periocular injection of arecombinant fragment of the NC1 domain of the a2chain of collagen IV causes regression of CNV (55) andis another good candidate for combination therapy.Increased expression of endostatin either systemicallyor in the eye by gene transfer not only inhibits severaltypes of ocular NV, but also prevents VEGF-inducedvascular leakage (103,104).

A potential advantage of gene therapy is thatintraocular injection of a vector containing anexpression construct provides a potential means ofsustained local delivery. Intravitreous injection of anadenoviral vector encoding PEDF (AdPEDF) resultedin high levels of PEDF in the eye and stronglysuppressed several types of ocular NV (105) andcaused regression of CNV (100). Gene transfer ofPEDF using adeno-associated viral vectors was also avery effective way to inhibit CNV many weeks later(106). Several studies have suggested that PEDF hasneuroprotective activity (107–112) and it might

contribute to the tropic support of photoreceptorsprovided by RPE cells, because in an in vitro modelof photoreceptor degeneration in which the RPE isremoved from Xenopus eyecups, PEDF protectedphotoreceptors from degeneration and loss of opsinimmunoreactivity (113). Therefore, intraocular PEDFgene transfer may provide a good approach in patientswith AMD, because it could possibly benefit bothneovascular and nonneovascular AMD.

A phase I study investigating the effects of asingle intraocular injection of AdPEDF.11 in patientswith advanced neovascular AMD showed an excellentsafety profile (114). There were no serious adverseevents related to AdPEDF.11 and no dose-limitingtoxicities. Signs of mild, transient intraocular inflam-mation occurred in 25% of patients, but there was nosevere inflammation. Six patients experiencedincreased intraocular pressure that was easilycontrolled by topical medication. At three and sixmonths after injection, 55% and 50%, respectively, ofpatients treated with 106 to 107.5 pu and 94% and 71%of patients treated with 108 to 109.5 pu had no changeor improvement in lesion size from baseline. Themedian increase in lesion size at 6 and 12 monthswas 0.5 and 1.0 disc areas in the low-dose groupcompared to 0 and 0 disc areas in the high-dosegroup. These data suggest the possibility of antiangio-genic activity that may last for several months after asingle intravitreous injection of doses greater than108 pu of AdPEDF.11. This study provides evidencethat adenoviral vector-mediated ocular gene transferis a viable approach for treatment of ocular disordersand that further studies investigating the efficacy ofAdPEDF.11 in patients with neovascular AMD shouldbe performed.

SUMMARY POINTS

& Over the past few years our understanding of themolecular pathogenesis of CNV has improved.

& It is clear that VEGF is a critical stimulus andantagonism of VEGF has led to the first treatmentthat improves vision in a substantial number ofpatients with neovascular AMD.

& As knowledge of the molecular signals thatcontribute to CNV continues to improve,additional treatments will be developed. VEGFantagonists will serve as baseline treatment towhich other therapies are added, although theirmode of delivery is likely to improve.

& Induction of regression followed by noninvasivetreatments that suppress recurrences is an import-ant strategy for prolonged treatment of patientswith neovascular AMD.


ACKNOWLEDGMENTS

Supported by grants from the Macula Vision ResearchFoundation. PAC is the George S. and Dolores DoreEccles Professor of Ophthalmology and Neuroscience.

REFERENCES

1. Campochiaro PA. Ocular versus extraocular neovascular-ization: mirror images or vague resemblances. InvestOphthalmol Vis Sci 2006; 47:462–74.

2. Okamoto N, Tobe T, Hackett SF, et al. Transgenic mice withincreased expression of vascular endothelial growth factorin the retina: a new model of intraretinal and subretinalneovascularization. Am J Pathol 1997; 151:281–91.

3. Tobe T, Okamoto N, Vinores MA, et al. Evolution ofneovascularization in mice with overexpression ofvascular endothelial growth factor in photoreceptors.Invest Ophthalmol Vis Sci 1998; 39:180–8.

4. Oshima Y, Oshima S, NambuH, et al. Increased expressionof VEGF in retinal pigmented epithelial cells is notsufficient to cause choroidal neovascularization. J CellPhysiol 2004; 201:393–400.

5. Aiello LP, Pierce EA, Foley ED, et al. Suppression of retinalneovascularization in vivo by inhibition of vascular endo-thelial growth factor (VEGF) using soluble VEGF-receptorchimeric proteins. Proc Natl Acad Sci USA 1995;92:10457–61.

6. Adamis AP, Shima DT, Tolentino MJ, et al. Inhibition ofvascular endothelial growth factor prevents retinalischemia-associated iris neovascularization. ArchOphthalmol 1996; 114:66–71.

7. Seo M-S, Kwak N, Ozaki H, et al. Dramatic inhibition ofretinal and choroidal neovascularization by oral adminis-tration of a kinase inhibitor. Am J Pathol 1999; 154:1743–53.

8. Ozaki H, Seo M-S, Ozaki K, et al. Blockade of vascularendothelial cell growth factor receptor signaling is suf-ficient to completely prevent retinal neovascularization.Am J Pathol 2000; 156:679–707.

9. Kwak N, Okamoto N, Wood JM, Campochiaro PA. VEGFis an important stimulator in a model of choroidalneovascularization. Invest Ophthalmol Vis Sci 2000;41:3158–64.

10. Kryzstolik MG, Afshari MA, Adamis AP, et al. Preventionof experimental choroidal neovascularization with intra-vitreal anti-vascular endothelial growth factor antibodyfragment. Arch Ophthalmol 2002; 120:338–46.

11. Saishin Y, Saishin Y, Takahashi K, et al. VEGF-TRAPR1R2suppresses choroidal neovascularization and VEGF-induced breakdown of the blood–retinal barrier. J CellPhysiol 2003; 195:241–8.

12. Gehlbach P, Demetriades AM, Yamamoto S, et al. Perio-cular gene transfer of sFlt-1 suppresses ocularneovascularization and VEGF-induced breakdown of theblood–retinal barrier. Hum Gene Ther 2003; 14:129–41.

13. Bainbridge J, Mistry A, Alwis MD, et al. Inhibition ofretinal neovascularization by gene transfer of solubleVEGF receptor sFlt-1. Gene Ther 2002; 9:320–6.

14. Gragoudas ES, Adamis AP, Cunningham ET, Jr.,Feinsod M, Guyer DR. Pegaptanib for neovascular age-related macular degeneration. N Eng J Med 2004;351:2805–16.

15. Rosenfeld P, Brown DM, Heier J, et al. Ranibizumab forneovascular age-related macular degeneration. N EngJ Med 2006; 355:1419–31.

16. Michels S, Rosenfeld PJ, Puliafito CA, Marcus EN,Venkatraman MS. Systemic bevacizumab (Avastin)therapy for neovascular age-related macular degener-ation. Ophthalmology 2005; 112:1035–47.

17. Nguyen QD, Shah SM, Tatlipinar S, Do DV, Van Anden E,Campochiaro PA. Bevacizumab suppresses choroidalneovascularization due to pathologic myopia. Br J Ophth-almol 2005; 89:1368–70.

18. Abraham JA, Whang JL, Tumolo A, et al. Human basicfibroblast growth factor: nucleotide sequence and genomicorganization. EMBO J 1986; 5:2523–8.

19. Leibovich SJ, Polverini PJ, Shepard HM, Wiseman DM,Shively V, Nuseir N. Macrophage-induced angiogenesis ismediated by tumor necrosis factor-alpha. Nature 1987;329:630–2.

20. Grant MB, Mames RN, Fitzgerald C, et al. Insulin-likegrowth factor I as an angiogenic agent. In vivo and in vitrostudies. Ann NYAcad Sci 1993; 692:230–42.

21. Smith LEH, Kopchick JJ, Chen W, et al. Essential role ofgrowth hormone in ischemia-induced retinal neovascular-ization. Science 1997; 276:1706–9.

22. Laterra J, NamM, Rosen E, et al. Scatter factor/hepatocytegrowth factor gene transfer enhances glioms growth andangiogenesis in vivo. Lab Invest 1997; 76:565–77.

23. Tobe T, Ortega S, Luna JD, et al. Targeted disruption of theFGF2 gene does not prevent choroidal neovascularizationin a murine model. Am J Pathol 1998; 153:1641–6.

24. Madri J, Reidy M, Kocher O, Bell L. Endothelial cellbehavior following denudation injury is modulated byTGF-beta and fibronectin. Lab Invest 1989; 60:755–65.

25. Hayasaka K, Oikawa S, Hashizume E, et al. Anti-angio-genic effect of TGFbeta in aqeous humor. Life Sci 1998;63:1089–96.

26. Hasegawa Y, Takanashi S, Kanehira Y, Tsushima T, Imai T,Okumura K. Transforming growth factor-beta 1 levelcorrelates with angiogenesis, tumor progression, andprognosis in patients with nonsmall cell lung carcinoma.Cancer 2001; 91:964–71.

27. O’Reilly MS, Holmgren S, Shing Y, et al. Angiostatin: anovel angiogenesis inhibitor that mediates the suppres-sion of metastases by a Lewis lung carcinoma. Cell 1994;79:315–28.

28. O’Reilly MS, Boehm T, Shing Y, et al. Endostatin: anendogenous inhibitor of angiogenesis and tumor growth.Cell 1997; 88:277–85.

29. O’Reilly MS, Pirie-Sheherd S, Lane WS, Folkman J. Anti-angiogenic activity of the cleaved conformation of theserpin antithrombin. Science 1999; 285:1926–8.

30. Maione TE, Gray GS, Petro J, et al. Inhibition of angio-gensis by recombinant human platelet factor-4 and relatedpeptides. Science 1990; 247:77–9.

31. Good DJ, Polverini PJ, Rastinejad F, et al. A tumorsupressor-dependent inhibitor of angiogenesis is immu-nologically and functionally indistinguishable from afragment of thrombospondin. Proc Natl Acad Sci USA1990; 87:6624–8.

32. Dawson DW, Volpert OV, Gillis P, et al. Pigmentepithelium-derived factor: a potent inhibitor of angio-genesis. Science 1999; 285:245–8.

33. Dike LE, Ingber DE. Integrin-dependent induction of earlygrowth response genes in capillary endothelial cells. J CellSci 1996; 109:2855–63.

34. Brooks P, Clark R, Cheresh D. Requirement of vascularintegrin alpha-v beta-3 for angiogenesis. Science 1994;264:569–71.


35. Luna J, Tobe T, Mousa SA, Reilly TM, Campochiaro PA.Antagonists of integrin alpha-v beta-3 inhibit retinalneovascularization in a murine model. Lab Invest 1996;75:563–73.

36. Umeda N, Kachi S, Akiyama H, et al. Suppression andregression of choroidal neovascularization by systemicadministration of an Alpha5Beta1 integrin antagonist.Mol Pharmacol 2006; 69:1820–8 (epub ahead of print).

37. Vlodavsky I, Folkman J, Sullivan R, et al. Endothelial cell-derived basic fibroblast growth factor: synthesis anddeposition into subendothelial extracellular matrix. ProcNatl Acad Sci USA 1987; 84:2292–6.

38. Vlodavsky I, Korner G, Ishai-Michaeli R, Bashkin P, Bar-Shavit R, Fuks Z. Extracellular matrix-resident growthfactors and enzymes: possible involvement in tumormetastasis and angiogenesis. Cancer Metastasis Rev1990; 9:203–26.

39. Park JE, Keller G-A, Ferrara N. The vascular endothelialgrowth factor (VEGF) isoforms: differential depositioninto the subepithelial extracellular matrix. Mol Biol Cell1993; 4:1317–26.

40. Ramchandran R, Dhanabal M, Volk R, et al. Antiangio-genic activity of restin, NC10 domain of human collagenXV: comparison to endostatin. Biochem Biophys ResCommun 1999; 255:735–9.

41. Colorado PC, Torre A, Kamphaus G, et al. Anti-angiogeniccues from vascular basement membrane collagen. CancerRes 2000; 60:2520–6.

42. Kamphaus GD, Colorado PC, Panka DJ, et al. Canstatin, anovel matrix-derived inhibitor of angiogenesis and tumorgrowth. J Biol Chem 2000; 275:1209–15.

43. Petitclerc C, Boutaud A, Prestayko A, et al. New functionsfor non-collagenous domains of human collagen type IV.Novel integrin ligands inhibiting angiogenesis and tumorgrowth in vivo. J Biol Chem 2000; 275:8051–61.

44. Maeshima Y, Colorado PC, Torre A, et al. Distinctantitumor properties of a type IV collagen domainderived from basement membrane. J Biol Chem 2000;275:21340–8.

45. Maeshima Y, Colorado PC, Kalluri R. Two RGD-indepen-dent alphavbeta3 integrin binding sites on tumstatinregulate distinct anti-tumor properties. J Biol Chem 2000;275:23745–50.

46. Shahan T, Grant D, Tootell M, et al. Oncothanin, a peptidefrom the alpha 3 chain of type IV collagen, modifiesendothelial cell function and inhibits angiogenesis.Connect Tissue Res 2004; 45:151–63.

47. Oberbaumer I, Wiedemann H, Timpl R, Kuhn K. Shapeand assembly of type IV procollagen obtained from cellculture. EMBO J 1982; 1:805–10.

48. Sundaramoorty M, Meiyappan M, Todd P, Hudson BG.Crystal structure of NC1 domains. Structural basis fortype IV collagen assembly in basement membranes.J Biol Chem 2002; 277:31142–53.

49. Ortega N, Werb Z. New functional roles for noncolla-genous domains of basement membrane collagens. J CellSci 2002; 115:4201–14.

50. Shen GQ, Butkowski R, Cheng T, et al. Comparison of non-collagenous type IV collagen subunits in human glomer-ular basement membrane, alveolar basement membrane,and placenta. Connect Tissue Res 1990; 24:289–301.

51. Hudson BG, Reeders ST, Tryggvason K. Type IV collagen:structure, gene organization, and role in human diseases.Molecular basis of Goodpasture and Alport sydromesanddiffuse leiomyomatosis. J Biol Chem1993; 268:26033–6.

52. Miner JH, Sanes JR. Collagen IV. alpha 3, alpha 4, andalpha 5 chains in rodent basal laminae: sequence, distri-bution, association with laminins, and developmentalswitches. J Cell Biol 1994; 127:879–91.

53. Tanaka K, Iyama K, Kitaoka M, et al. Differentialexpression of alpha 1(IV), alpha 2 (IV), alpha 5(IV), andalpha 6 (IV) collagen chains in the basement membrane ofbasal cell carcinoma. Histochem J 1997; 29:563–70.

54. Fleischmajer R, Kuhn K, Sato Y, et al. There is temporaland spatial expression of alpha1(IV), alpha2(IV),alpha5(IV), and alpha6(IV) collagen chains and beta 1integrins during the development of the basal laminain an “in vitro” skin model. J Invest Dermatol 1997;109:527–33.

55. Lima e Silva R, Kachi S, Akiyama H, et al. Recombinantnon-collagenous domain of alpha2(IV) collagen causesinvolution of choroidal neovascularization by inducingapoptosis. J Cell Physiol 2006; 208(1):161–6 (epub aheadof print).

56. Pepper MS, Vassalli J-D, Montesano R, Orci L. Urokinase-type plasminogen activator is induced in migrating capil-lary endothelial cells. J Cell Biol 1987; 105:2535–41.

57. Moscatelli DA, Rifkin DB, Jaffe EA. Production of latentcollagenase by human umbilical vein endothelial cells inresponse to angiogenic preparations. Exp Cell Res 1985;156:379–90.

58. Cornelius LA, Nehring LC, Roby JD, Parks WC,Welgus HG. Human dermal microvascular endothelialcells produce matrix metalloproteinases in response toangiogenic factors and migration. J Invest Dermatol1995; 105:170–6.

59. Johnson MD, Kim H-RC, Chesler L, Tsao-Wu G, Bouck N,Polverini PJ. Inhibition of angiogenesis by tissue inhibitorof metalloproteinase. J Cell Physiol 1994; 160:194–202.

60. Yamada E, Tobe T, Yamada H, et al. TIMP-1 promotesVEGF-induced neovascularization in the retina. HistolHistopath 2001; 16:87–97.

61. Michaelson I. The mode of development of the vascularsystem of the retina with some observations on its signi-ficance for certain retinal diseases. Trans Ophthalmol SocUK 1948; 68:137–80.

62. Ashton N. Retinal vascularization in health and disease.Am J Ophthalmol 1957; 44(4):7–17.

63. Shimizu K, Kobayashi Y, Muraoka K. Midperipheralfundus involvement in diabetic retinopathy. Ophthal-mology 1981; 88:601–12.

64. Miller JW, Stinson W, Folkman J. Regression of experi-mental iris neovascularization with systemic alpha-interferon. Ophthalmology 1993; 100:9–14.

65. Virdi P, Hayreh S. Ocular neovascularization with retinalvascular occlusion. I. Association with retinal vein occlu-sion. Arch Ophthalmol 1980; 100:331–41.

66. Pournaras C, Tsacopoulos M, Strommer K, Gilodi N,Leuenberger PM. Experimental retinal branch vein occlu-sion in miniature pigs induces local tissue hypoxia andvasoproliferative microangiopathy. Ophthalmology 1990;97:1321–8.

67. Penn JS, Tolman BL, Lowery LA. Variable oxygenexposure causes preretinal neovascularization in thenewborn rat. Invest Ophthalmol Vis Sci 1993; 34:576–85.

68. Smith LEH, Wesolowski E, McLellan A, et al. Oxygen-induced retinopathy in the mouse. Invest Ophthalmol VisSci 1994; 35:101–11.

69. Ozaki H, Yu A, Della N, et al. Hypoxia inducible factor-1ais increased in ischemic retina: temporal and spatialcorrelation with VEGF expression. Invest OphthalmolVis Sci 1999; 40:182–9.


70. Campochiaro PA. Retinal and choroidal neovascularization.J Cell Physiol 2000; 184:301–10.

71. Grunwald J, Hariprasad S, DuPont J, et al. Foveolarchoroidal blood flow in age-related macular degeneration.Invest Ophthalmol Vis Sci 1998; 39:385–90.

72. Ross RD, Barofsky JM, et al. Presumed macular choroidalwatershed vascular filling, choroidal neovascularization,and systemic vascular disease in patients with age-relatedmacular degeneration. Am J Ophthalmol 1998; 125:71–80.

73. Vinores SA, Xiao WH, Aslam S, et al. Implication of thehypoxia response element of the VEGF promoter in mousemodels of retinal and choroidal neovascularization, butnot retinal vascular development. J Cell Physiol 2005;206:749–58 (epub ahead of print, 21 Oct 2005).

74. Valdembri D, Serini G, Vacca A, Ribatti D, Bussolino F.In vivo activation of JAK/STAT-3 pathway during angio-genesis induced by GM-CSF. FASEB J 2002; 16:225–7.

75. Suganami E, Takagi H, Ohashi H, et al. Leptin stimulatesischemia-induced retinal neovascularization: possible roleof vascular endothelial growth factor expressed in retinalendothelial cells. Diabetes 2004; 53:2443–8.

76. Huang SP, Wu MS, Shun CT, et al. Interleukin-6 increasesvascular endothelial growth factor and angiogenesis ingastric carcinoma. J Biomed Sci 2004; 11:517–27.

77. Dudley AC, Thomas D, Best J, Jenkins A. AVEGF/JAK2/-STAT5 axis may partially mediate endothelial celltolerance to hypoxia. Biochem J 2005; 390:427–36.

78. Green WR, Enger C. Age-related macular degenerationhistopathologic studies. Ophthalmology 1993; 100:1519–35.

79. Bressler SB, Maguire MG, Bressler NM, Fine SL, Group atMPS. Relationship of drusen and abnormalities of theretinal pigment epithelium to the prognosis of neovascularmacular degeneration. Arch Ophthalmol 1990; 108:1442–7.

80. Weber BHF, Vogt G, Pruett RC, Stohr H, Felbor U.Mutations in the tissue inhibitor of metalloproteinases-3(TIMP3) in patients with Sorsby’s fundus dystrophy. NatGenet 1994; 8:352–6.

81. Sorsby A, Mason MEJ, Gardener N. A fundus dystrophywith unusual features. Br J Ophthalmol 1949; 33:67–97.

82. Hoskin A, Sehmi K, Bird AC. Sorsby’s pseudoinflamma-tory macular dystrophy. Br J Ophthalmol 1981; 65:859–65.

83. Campochiaro PA, Hackett SF. Corneal endothelial cellmatrix promotes expression of differentiated features ofretinal pigmented epithelial cells: implication of lamininand basic fibroblast growth factor as active components.Exp Eye Res 1993; 57:539–47.

84. Mousa SA, Lorelli W, Campochiaro PA. Extracellularmatrix-integrin binding modulates secretion of angiogenicgrowth factors by retinal pigmented epithelial cells. J CellBiochem 1999; 74:135–43.

85. Yamada H, Yamada E, Ando A, et al. FGF2 decreaseshyperoxia-induced cell death in mice. J Am Pathlol 2001;159:1113–20.

86. Miller H, Miller B, Ryan SJ. The role of the retinalpigmented epithelium in the involution of subretinalneovascularization. Invest Ophthalmol Vis Sci 1986;27:1644–52.

87. Dobi ET, Puliafito CA, Destro M. A new model of chor-oidal neovascularization in the rat. Arch Ophthalmol 1989;107:264–9.

88. Spilsbury K, Garrett KS, Shen WY, Constable IJ,Rakoczy PE. Overexpression of vascular endothelialgrowth factor (VEGF) in the retinal pigment epitheliumleads to the development of choroidal neovascularization.Am J Pathol 2000; 157:135–44.

89. Baffi J, Byrnes G, Chan C-C, Csaky KG. Choroidal neovas-cularization in the rat induced by adenovirus mediatedexpression of vascular endothelial growth factor. InvestOphthalmol Vis Sci 2000; 41:3582–9.

90. Ozaki H, Okamoto N, Ortega S, et al. Basic fibroblastgrowth factor is neither necessary nor sufficient for thedevelopment of retinal neovascularization. Am J Pathol1998; 153:757–65.

91. Ando A, Mori K, Yamada H, et al. Nitric oxide isproangiogenic in retina and choroid. J Cell Physiol 2001;191:116–24.

92. Jain AK, Hunley CC, Kuebel J, McMahon FG, Ryan JJ.Analgesic efficacy of amfenac, aspirin and placebo afterextraction of impacted teeth. Pharmacotherapy 1986;6:236–40.

93. Gamache DA, Graff G, Brady MT, Spellman JM, Yanni JM.Nepafenac, a unique nonsteroidal prodrug with potentialutility in the treatment of trauma-induced ocular inflam-mation: I. Assessment of anti-inflammatory efficacy.Inflammation 2000; 24:357–70.

94. Ke T-L, Graff G, Spellman JM, Yanni JM. Nepafenac, aunique nonsteroidal prodrug with potential utility in thetreatment of trauma-induced ocular inflammation: II.In vitro bioactivation and permeation of external ocularbarriers. Inflammation 2000; 24:371–84.

95. Takahashi K, Saishin Y, Saishin Y, et al. Topical nepafenacinhibits ocular neovascularization. Invest Ophthalmol VisSci 2003; 44:409–15.

96. Ben-Av P, Crofford LJ, Wilder RL, Hla T. Induction ofvascular endothelial growth factor expression in synovialfiboblasts by prostaglandin E and interleukin-1: a potentialmechanism for inflammaroty angiogenesis. FEBSLett 1995;372:83–7.

97. Hoper MM, Voelkel NF, Bates TO, et al. Prostaglandinsinduce vascular endothelial growth factor in a humanmonocytic cell line and rat lungs via cAMP. Am J RespirCell Mol Biol 1997; 17:748–56.

98. Majima M, Hayashi I, Muramatsu M, Katada J,Yamashina S, Katori M. Cyclo-oxygenase-2 enhancesbasic fibroblast growth factor-induced angiogensisthrough induction of vascular endothelial growth factorin rat sponge implants. Br J Pharmacol 2000; 130:641–9.

99. Lima e Silva R, Saishin Y, Saishin Y, et al. Suppression andregression of choroidal neovascularization by polyamineanalogs. Invest Ophthalmol Vis Sci 2005; 46:3323–30.

100. Mori K, Gehlbach P, Ando A, McVey D, Wei L,Campochiaro PA. Regression of ocular neovascularizationby increased expression of pigment epithelium-derivedfactor. Invest Ophthalmol Vis Sci 2001; 43:2428–34.

101. Nambu H, Nambu R, Melia M, Campochiaro PA. Combre-tastatin A-4 phosphate suppresses development andinduces regression of choroidal neovascularization.Invest Ophthalmol Vis Sci 2003; 44:3650–5.

102. Alon T, Hemo I, Itin A, Pe’er J, Stone J, Keshet E. Vascularendothelial growth factor acts as a survival factor fornewly formed retinal vessels and has implications forretinopathy of prematurity. Nature Med 1995; 1:1024–8.

103. Mori K, Ando A, Gehlbach P, et al. Inhibition of choroidalneovascularization by intravenous injection of adenoviralvectors expressing secretable endostatin. Am J Pathol2001; 159:313–20.

104. Takahashi K, Saishin Y, Saishin Y, et al. Intraocularexpression of endostatin reduces VEGF-induced retinalvascular permeability, neovascularization, and retinaldetachment. FASEB J 2003; 17:896–8.


105. Mori K, Duh E, Gehlbach P, et al. Pigment epithelium-derived factor inhibits retinal and choroidal neovascular-ization. J Cell Physiol 2001; 188:253–63.

106. Mori K, Gehlbach P, Yamamoto S, et al. AAV-mediatedgene transfer of pigment epithelium-derived factorinhibits choroidal neovascularization. Invest OphthalmolVis Sci 2002; 43:1994–2000.

107. Taniwaki T, Becerra SP, Chader GJ, Schwartz JP. Pigmentepithelium-derived factor is a survival factor for cerebellargranule cells in culture. J Neurochem 1995; 64:2509–17.

108. Araki T, Taniwaki T, Becerra SP, Chader GJ, Schwartz JP.Pigment epithelium-derived factor (PEDF) differentiallyprotects immature but not mature cerebellar granule cellsagainst apoptotic cell death. J Neurosci Res 1998; 53:7–15.

109. DeCoster MA, Schabelman E, Tombran-Tink J, Bazan NG.Neuroprotection by pigment epithelial-derived factoragainst glutamate toxicity in developing primary hippo-campal neurons. J Neurosci Res 1999; 56:604–10.

110. Bilak MM, Corse AM, Bilak SR, Lehar M, Tombran-Tink J,Kuncl RW. Pigment epithelium-derived factor (PEDF)

protects motor neurons from chronic glutamate-mediatedneurodegeneration. J Neuropathol Exp Neurol 1999;58:719–28.

111. Cao W, Tombrin-Tink J, Chen W, Mrazek D, Elias R,McGinnis JF. Pigment epithelium-derived factor protectscultured retinal neurons against hydrogen peroxide-induced cell death. J Neurosci Res 1999; 57:789–800.

112. Houenou LJ, D’Costa AP, Li L, et al. Pigment epitheliumderived factor promotes the survival and differentiation ofdeveloping spinal motor neurons. J Comp Neurol 1999;412:506–14.

113. JablonskiMM, Tombran-Tink J,MrazekDA, IannoacconeA.Pigment epithelium-derived factor supports normal develo-pment of photoreceptor neurons and opsin expression afterretinal pigment epithelium removal. J Neurosci 2000;20:7149–57.

114. Campochiaro PA, Nguyen QD, Shah SM, et al. Adenoviralvector-delivered pigment epithelium-derived factor forneovascular age-related macular degeneration: results ofa phase I clinical trial. Hum Gene Ther 2006; 17:167–76.


Part II: Clinical Features of Age-Related Macular Degeneration

6

Non-exudative Age-related Macular DegenerationNeelakshi BhagatThe Institute of Ophthalmology and Visual Science, New Jersey Medical School,

Newark, New Jersey, U.S.A.

Christina J. FlaxelCasey Eye Institute, Oregon Health & Science University, Portland, Oregon, U.S.A.

INTRODUCTION

Age-related macular degeneration (AMD), also known asage-related maculopathy (ARM), is the leading cause ofblindness in the Western world (1). It is also the leadingcause of irreversible central vision loss in whites over 50years of age in the United States (2). The disease affectsapproximatelyeightmillionpeople in theUnitedStates (3);its advanced formaffectsmore than1.75millionpeople (4).

The prevalence and progression of AMDincreases with age (Table 1), from a prevalence of1.6% in the age group 52 to 64 years to 28% in the agegroup74 to 85 years (5). In the BlueMountain Study, theprevalence of earlyAMDwas reported to increase from1.3% in the age group 49 to 54 years to 28.0% for thoseover 80 years of age; the prevalence of late ARM, on theother hand, increased from 0.1% in the age-group 49 to54 years to 7.1% in the age group 75 to 86 years. Thepopulation older than 65 years is the fastest growingsegment of our society and the prevalence of AMD ispredicted to increase dramatically in the next decade(6,7). There is a pressing need for new therapies toeither prevent AMD or treat the exudative AMD.

CLINICAL FEATURES OF AMD

The clinical hallmarks of non-exudative AMD are softdrusen, localized deposits noted between the base-ment membrane of the retinal pigment epithelium(RPE) and the Bruch’s membrane, associated RPEpigmentary changes, and mild loss in visual acuity(VA) (8). The advanced form of non-exudative AMD,termed geographic atrophy (GA), is characterizedby outer retinal and RPE atrophy with loss ofchoriocapillaris.

The presence of subretinal fluid, subretinalhemorrhage, RPE detachment, a subretinal greenish–greyish membrane, or hard exudates indicateschoroidal neovascularization (CNV), which heraldsthe onset of exudative macular degeneration (9).Fluorescein angiography delineates the exact location

(subfoveal, juxtafoveal, or extrafoveal), the size, andthe pattern of leakage (classic vs. occult). Loss ofcentral vision is usually due to RPE atrophy or GAin non-exudative AMD and due to subretinal fluid orsubretinal hemorrhage in exudative AMD.

Early-stage AMD (or early ARM) is defined asthe presence of soft drusen (R63 mm) alone, RPEdepigmentation alone, or a combination of distinct/indistinct drusen with pigment irregularities. Late-stage AMD (or late ARM) is defined as pure GA(both central and noncentral), signs of exudativemacular degeneration, or a combination of both.

ASSOCIATED FACTORS

Many epidemiologic studies have provided insight intothe various factors that may be associated with AMD(10–17). Hereditary influence, photic injury, nutritionaldeficiency, toxic insult, and systemic factors havebeen implicated in epidemiologic studies (18–20).These factors can be grouped into following categories:(i) demographic characteristics, which include age,sex, race, and eye color (21–23); (ii) systemic diseases,e.g., hypertension, cardiovascular disease, and hyper-cholesterolemia (1,14,18, 24–27); (iii) environmentalinfluences such as smoking, sunlight, nutrition; and(iv) genetic predisposition (14,24,28–30).

Demographic CharacteristicsThe prevalence and progression of all types ofAMD increases with age (1,11,22,31–33). A statisticallysignificant increased incidence of ARM lesions isalso noted with age (P !0.05). Individuals 75 yearsof age or older at baseline have significantly (P!0.01)higher 10-year incidences of the following charac-teristics than people 43 to 54 years of age: largersized drusen (125–249 mm), 26.3% versus 3.3%;R250 mm, 16.2% versus 1.0%; soft indistinct drusen,22.2% versus 2.2%; retinal pigment abnormalities,19.5% versus 0.8%; exudative macular degeneration,

4.1% versus 0%; and pure GA, 3.1% versus 0%. Eyeswith soft indistinct drusen or retinal pigmentaryabnormalities at baseline are more likely to developlate ARM at follow-up than eyes without these lesions(15.1% vs. 0.4% and 20.0% vs. 0.8%, respectively) (3).

AMD is commonly reported to be moreprevalent in women (25,34), although conflictingresults have been noted. The Beaver Dam Study,after adjusting for age, revealed that the incidence ofearly AMD was 2.2% higher in women 75 years of ageand older than in men in this age group (25). On theother hand, the prevalence of early ARMwas higher inmen than in women in each age category in the BlueMountains Eye Study (35). However, the pooled datafrom three different continents (the Beaver DamEye Study, the Blue Mountains Eye Study, and theRotterdam Study) did not note such a difference inAMD prevalence between men and women (32).

AMD is noted to be more common in whites thanin pigmented races (36–38). The prevalence of anyARM in blacks is almost half of that noted in whites;9.1% compared with 18.2% (38). It has beenhypothesized that melanin may function as an anti-oxidant, and protect against lipofuscin accumulationin the RPE cells and development of CNV (39).The association between light-colored irides andAMD has been controversial (40,32). Most of thelarge case-control and population studies have foundno association between iris color and AMD (9,41–44),but a few case-control studies have (45–47).

Systemic DiseasesHypertension and Cardiovascular DiseaseThe Framingham Eye Study (1), the Age-Related EyeDisease Study (AREDS) (42), and the Macular Photo-coagulation Study (26) reported a positive correlationbetweenAMDand hypertension. This associationwas,however, was not seen in the Beaver Dam Eye Study(27), the EyeDisease Case-Control Study (EDCCS) (43),or in the Cardiovascular Health Study (38). A strongassociation has been noted between exudative AMDand moderate to severe hypertension, particularly inpatients on antihypertensive therapy (24,26,34).

In most studies, no justifiable association hasbeen noted between AMD and atherosclerosis(24,25,42) though Hyman and coworkers (45) noted apositive correlation of AMD with stroke, arterio-sclerosis, and ischemic attacks.

HypercholesterolemiaThere exist conflicting data regarding the effect ofhypercholesterolemia on AMD. A positive correlationhas been found between high intake of saturated fatand cholesterol, and AMD (29). A large prospectivestudy of 70,000 individuals (Nurses’ Health Study andthe Health Professionals Follow-up Study) clearlyrevealed that the total fat intake was positively associ-ated with the risk of AMD (46). The EDCCS noted thatpatients with exudative ARMwere more likely to havea higher serum total cholesterol. Some studies,however, have noted a protective effect of serumcholesterol on AMD—the total serum cholesterol hasbeen reported to be inversely related to earlyAMD (25).A positive relationship has also been noted betweenhigh-density lipoprotein levels and AMD (25).

Environmental InfluencesEnvironmental influences such as photic injury,smoking, and nutrition may have an effect on thedevelopment of AMD.

Cumulative exposure to light may cause gradualloss of photoreceptor cells in the macula (48). Photo-oxidative damage by reactive oxygen intermediatesinduced by light may promote the development ofAMD (28,30,49). The retina, particularly the macula,is highly susceptible to oxidative stress due to ahigh polysaturated fatty acid content that is prone tolipid peroxidation (28). There have been, however,conflicting reports regarding the association of ultra-violet or visible light with AMD (50,51). Antioxidantsmay prevent this damage (28).

A low dietary intake or low plasma concen-trations of antioxidants may be associated with AMD(52). High-dose antioxidants, as recommended byAREDS trial, are associated with the decreased riskof progression to the exudative form of AMD (53).

Table 1 Prevalence of Age-Related Macular Degeneration

Epidemiologic studiesAge

(years)

AMD prevalance (%)

Early Late Early or late

1. Chesapeake Bay (9) !50 4.0

50–59 6.0

60–69 13.0

70–79 26.0 4.3

80C 13.6

2. Beaver Dam (9) 43–54 8.4 0.1

55–64 13.8 0.6

65–74 18.0 1.4

75C 29.7 7.1

3. Klein and Klein (27) 45–64 2.3

65–74 9.0

4. Blue Mountains

Study group (36)

49–54 1.3 0.0

55–64 2.6 0.2

65–74 8.5 0.7

75–84 15.5 5.4

85C 28.0 18.5

5. Copenhagen (47) 60–69 4.1

70–80 20.0

6. Framingham (18) 52–64 1.6

65–74 11.0

75–85 27.9

98 BHAGAT AND FLAXEL

In the randomized, placebo-controlled AREDS, sup-plements containing 5 to 13 times the recommendeddaily allowance of beta-carotene (15 mg), vitamins C(500 IU) and E (400 IU), and zinc (20 mg) taken bypatients with early or monocular late AMD resulted ina 25% reduction in the five-year progression to lateAMD (53). This benefit did not extend to patientswithout AMD or with few drusen.

Much attention has been given to the dietaryimportance of carotenoids, lutein and zeaxanthin(54), and omega-3 fatty acids. A high intake ofomega-3 fatty acids and fish is inversely associatedwith the risk of AMD when intake of linoleic acid islow (55), and high levels of docosahexaenoic acid(DHA) have been associated with a 30% reduced riskof AMD (46). Seddon and coworkers have reported theresults of the EDCCS trial that revealed that a highdietary intake of carotenoids, particularly dark-greenleafy vegetables, is associated with a 43% lower risk ofAMD (53). AREDS II, a multicenter prospectiveNational Institutes of Health-sponsored study, willevaluate the effects of DHA and omega-3 fatty acidson the progression of AMD. The effects of lutein andzeaxanthin will also be evaluated in AREDS II.

The combined data from the Blue Mountain EyeStudy, the Beaver Dam Eye Study, and the RotterdamStudy have clearly shown that current smokers have asignificantly higher risk of incident GA and late AMDthan both past smokers and those who never smoked.The mean age at diagnosis of AMD (GA, neovascularAMD, and late AMD) is lower for current smokersthan for past smokers or those who never smoked (56).A statistically significant association has been notedbetween smoking and any one or more types of AMD,with increased risks for current smokers or pastsmokers compared with nonsmokers; the risk ratioor odds ratio has been described to be between 1.06and 4.96 [reviewed in (57,58)]. The mechanism ofinjury with smoking is not well understood. It isplausible that smoking decreases choroidal flowpotentiating macular hypoxia and ischemia promotingAMD (59).

It has also been suggested that a possible small,independent association may exist between highhomocysteine levels and AMD (60). Lycopene, aserum carotenoid, may be associated with AMD (61).

Genetic InfluenceA familial component of AMD has been suggested bytwin concordance (62) and first-degree relative studies(63). A population-based study revealed an overallconcordance of 37% in monozygotic twins versus19% in dizygotic twins for early AMD (64). AMD ismore likely in first-degree relatives than in age-matched controls (65).

The pathogenesis of AMD is not well under-stood. Recent published literature has demonstratedties between inflammation and the pathogenesis ofAMD (66–68). Complement factor H (CFH) genelocated on chromosome 1 is in a region linked toAMD. Three different CFH gene variants have beenidentified that increase the risk of AMD from 2.45 to7.4 times more likely to have AMD (66–68). CFHregulates the alternate complement pathway. Themutated variant gene may prevent inactivation of thealternate complement pathway, possibly leading topersistent inflammation in the retina and choroid.

DRUSEN

Drusen were first described in 1854 by Donders (69).They are deposits of membranous debris, extracellularmaterial (ECM) between the RPE and its basementmembrane (basal laminar drusen) or between theRPE basement layer and the inner collagenous layerof Bruch’s membrane (basal linear drusen) (16,70–73).Drusen lead to secondary thickening of Bruch’smembrane and RPE degeneration. Visual loss inmacular degeneration is the result of photoreceptoratrophy that follows RPE atrophy as a result ofinvolution of choriocapillaris (74).

Drusen form as deposition of membranousmaterial between the plasma membrane and the base-ment membrane of the RPE, and can be found as earlyas the second decade of life. They represent a normalaging change (75,76). Experimental and postmortemhuman studies have shown that the drusen are RPE-derived (77–79). Ishibashi and coworkers describedthe formation of drusen under electron microscopyas follows: (i) evagination or budding of the RPE cellin the subepithelial space; (ii) separation of theevaginated portion from the parent RPE cell;(iii) degeneration and disintegration of these evagi-nated cell components devoid of a nucleus; and(iv) accumulation of granular, vesicular, tubular, andlinear material in the sub-RPE space (15). The etiologyof the evagination is not known.

AMD involves aging changes along withadditional pathologic changes. In both aging andAMD, following changes occur: (i) oxidative stressresulting in RPE injury; (ii) inflammatory response inthe Bruch’s membrane caused by the RPE injury;(iii) production of abnormal ECM by the injured RPEand choroid; and (iv) resultant disturbance in theRPE—Bruch’s membrane homeostasis ultimatelyleading to RPE and choriocapillaris atrophy orgrowth of choroidal neovascular membrane (80).Environment and genetics may have a superimposingeffect that most likely alters a given patient’s sus-ceptibility to the disease (80).

6: NON-EXUDATIVE AGE-RELATED MACULAR DEGENERATION 99

Types of DrusenDifferent types of drusen are noted in the retina:(i) hard, (ii) soft, (iii) crystalline, and (iv) cuticular orbasal laminar.

Hard DrusenHard drusen are discrete, small, yellow, nodularhyaline deposits in the sub-RPE space, between thebasement membrane of RPE and the inner collagenouslayer of Bruch’s membrane (81). These drusen aresmaller than 50 mm in diameter (Fig. 1). Focal densifi-cations of Bruch’s membrane, termed microdrusen,may precede the formation of hard drusen (82).Preclinical drusen appear ultrastructurally as “entrap-ment sites”, with coated membrane-bound bodies thatform adjacent to the inner collagenous layer of Bruch’smembrane (82). These are structurally different frombasal linear deposit.

Hard drusen are common in young people anddo not lead to macular degeneration (70). Small, hard,distinct drusen were found in the macula of 94% of theBeaver Dam Eye Study population (9). These were notnoted to increase in number with age. If present inexcessive number, however, they may predispose toRPE atrophy (83).

Hard drusen act as window defects on fluor-escein angiogram with early hyperfluorecence andfading of fluorescence in late frames (Fig. 2).

Soft DrusenSoft drusen are clinically noted as pale yellow lesionswith poorly defined edges (Fig. 3). They can alsorepresent focal accentuations of basal linear deposits(71). They also represent localized accumulation ofbasal laminar deposits in an eye with diffuse basallaminar deposits (84). They gradually enlarge andmaycoalesce, termed confluent drusen, to form multipleirregular areas of localized RPE detachments. Withtime, soft drusen can become crystalline in nature.Crystalline drusen are discrete calcific refractiledrusen (Fig. 4). These are dehydrated soft drusenthat predispose to GA (Fig. 5) (83,85).

Soft drusen are classified by size into small,medium, and large. A small soft druse is !63 mmwide, intermediate is between 63 and 128 mm andlarge is O128 mm. (The width of the retinal vein offthe optic nerve is 128 mm.) The risk of progressionfrom non-exudative to exudative AMD increases withthe size and the total area of the drusen (86). Clinicaland histological studies have shown that soft drusenprecede macular degeneration (87,88). They lead tosecondary Bruch’s membrane thickening and RPEatrophy with subsequent photoreceptor loss. Thispromotes the development of choroidal neovascularmembrane (13,73,89).

(A) (B)

Figure 2 Fluorescein angiogram: (A) early and (B) late frames of hard drusen in the macula.

Figure 1 Hard drusen.


On fluorescein angiography, soft drusen showearly hypofluorescence or hyperfluorescence with nolate leakage.

Basal Laminar DrusenBasal laminar or cuticular drusen are tiny, whitedeposits found between the plasma membrane ofRPE and its basement membrane (Fig. 6) (70). Suchdrusen are mainly composed of collagen, laminin,membrane-bound vesicles, and fibronectin. Thedeposits tend to accumulate over the thickenedBruch’s membrane, suggesting that they may be alocal response to altered filtration at these sites (81).

Basal laminar drusen are typically very numer-ous, distributed in a bilaterally symmetrical pattern,and are most prominent in the posterior poles. Theseunusual drusen are often seen in association withother typical hard, soft, or semisolid drusen. VA istypically minimally affected despite the large number

of these drusen. They tend to occur in youngerindividuals and in normal eyes and do not predisposeto macular degeneration.

On fluorescein angiography, the basal laminardrusen hyperfluoresce early and give an appearance of“starry night” (Fig. 7) (85).

DISAPPEARANCE OF DRUSEN

Variousreportshavedescribedspontaneousdisappear-ance of drusen (10,90). Bressler and colleagues noted intheir Waterman study that the large drusen disap-peared spontaneously in 35% of 47 individuals in fiveyears of follow-up (87). They have also been reported todisappear after laser photocoagulation (Fig. 8) (94).RPE atrophy is noted as drusen disappear followedby photoreceptor and choriocapillaris loss (91).

It has been hypothesized that disappearance ofdrusen may be linked to a lower risk of CNV. This was

Figure 6 Basal laminar drusen.

Figure 5 Geographic atrophy.

Figure 4 Calcific drusen.

Figure 3 Soft drusen.


the basis of undertaking three multicenter trials toevaluate the effect of light laser ondrusen, the choroidalneovascularization prevention trial (CNVPT) (92),the Prophylactic Treatment of AMD (PTAMD) Study(93), and the Drusen Laser Study (DLS) (95). TheCNVPT study, a multicenter, randomized prospectivestudy of laser treatment versus observation evaluatedthe effect of low-intensity argon laser treatment in eyeswith drusen secondary to AMD. The study includedtwo groups: (i) unilateral group, 120 patients withCNVs in one eye and 10 or more drusen larger than63 mm in diameter in the fellow eye, and (ii) bilateral-drusen group, 156 patients with non-exudative AMDand 10 or more drusen larger than 63 mm in diameterwithin 3000 mm of the foveola and good VA (92,96,97).The rate of CNV formation in the bilateral group

was not found to be statistically different, 4/152 in thelaser-treated group versus 2/156 in the control group,PZ0.42 (92). However, a relatively high rate of CNVdevelopment was seen in the unilateral arm andthe study was suspended in the treated cohort; 16.9%(10/59) versus 3.2% (2/61 eyes) in the laser treatmentand control groups, respectively (PZ0.02) (92).

The PTAMD trial, using subthreshold diodelaser, has shown similar findings of increased risk ofexudative maculopathy in fellow eyes of patientstreated with laser (93). In the phase III PTAMDstudy, at one year, the rate of CNV formation was15.8% for lasered versus 1.4% for observed eyes(PZ0.05). Most of the intergroup differences in CNVevents occurred during the first two years of follow-up. Treated eyes showed a higher rate of VA loss(R3 lines) at three and six months follow-up relativeto observed eyes (8.3% vs. 1% and 11.4% vs. 4%,respectively; PZ0.02, 0.07). After six months, no sig-nificant differences were observed in VA loss betweengroups. Prophylactic subthreshold 810-nm-diode lasertreatment to an eye with multiple large drusen in apatient whose fellow eye has already suffered aneovascular event places the treated eye at higherrisk of developing CNV (96). The bilateral-drusenarms of this trial are still in progress.

The DLS trial has shown that the incidenceof CNV in the unilateral arm was higher in thelaser-treated arm than the control arm, though thiswas not found to be statistically significant. However,the onset of CNV was approximately six monthsearlier in the laser-treated group than in the controlgroup, a statistically significant finding (95). The grouprecommends against prophylactic laser treatmentwhen a neovascular process has already occurred inone eye. However, they were unable to determine the

Figure 7 Fluorescein angiographic late frame demonstrating

“starry night” appearance.

(A) (B)

Figure 8 This patient underwent diode laser photocoagulation in his left eye. (A) Fundus appearanceshows multiple large soft drusen prior to laser. (B) Fundus appearance three years later shows a marked

reduction in the number of drusen in this eye.


role of prophylactic laser in patients with bilateraldrusen and good vision as the event rate is very lowin these eyes and a large number of eyes are needed,they were not able to achieve the necessary recruit-ment goals to answer this question (Fig. 9) (95).

Another method of preventing the developmentof sight-threatening CNV is to deliver a medicationto an at-risk eye to slow or stop angiogenesis. Theanecortave acetate risk reduction trial completedenrollment in January 2006 of 2500 eligible patientsto determine whether anecortave acetate, an angio-static steroid given as a juxtascleral injection everysix months, will reduce the risk of developing CNVand vision loss in eyes that are at high risk of thiscomplication (Fig. 10A–J).

NON-EXUDATIVE AMD

Soft drusen precede macular degeneration (74,87). Themere presence of drusen does not account for signi-ficant loss of vision (72). Soft drusen can lead to RPEatrophy, with resultant overlying photoreceptoratrophy and vision loss. When the vision falls belowor equal to 20/30, the disease process is termed non-exudative or dry macular degeneration. Subretinalfluid, subretinal hemorrhage, RPE detachment, hardexudates, and subretinal fibrosis, all signs of exudativemaculopathy, are absent in dry macular degeneration.GA is an advanced form of dry macular degeneration.This involves RPE atrophy with subjacent choriocapil-laris and small choroidal vessel atrophy. Thiscondition progresses slowly over years and oftenspares the center of the foveal avascular zone untillate in the course of the disease (99).

Non-exudative AMD is the most common form ofAMD, accounting for 80% to 90% of cases overall (18).Bressler and coworkers reported a prevalence of 1.8% ofAMD inmen 50 years of age or older in the Chesapeake

Bay study. Of these, almost 75% had non-exudativemaculopathy (100). It accounts for only 20% of all legalblindness associated with AMD and occurs with GA,an advanced form of dry AMD (101).

Soft drusen and retinal pigmentary changesincrease with age (31,10). In the five-year period of theBeaver Dam Eye Study, people 75 years or older were3.3 to 8.4 times as likely to develop drusen between 63and 250 mm indiameterwhen comparedwith person 43to 54 years of age. Also, persons 75 years of age orover were 16 times more likely to develop confluentdrusen when compared with people 43 to 54 years ofage (31). The incidence of early AMD increases withadvancing age (Table 2). These findings have beennoted in all population-based studies: The BeaverDam Eye Study (31), Blue Mountains–Australian (36),Rotterdam (102), and Colorado–Wisconsin (103)studies of AMD.

Focal hyperpigmentation along with thepresence of greater than five soft, large, and confluentdrusen is associated with the increased risk of pro-gression of RPE atrophy and choroidal atrophy. Theseeyes have a higher incidence of developing CNV(72,73). The mere presence of CNV in one eye increasesthe risk for CNV development in the remaining eye,compared with patients having bilateral drusen. Thepresence of large drusen in both eyes was a strongerrisk factor for progression to advanced AMD than thepresence in only one eye (53).

The five-year risk of eyes with bilateral softdrusen and good VA to develop CNV is 0.2% to18% (10,74,89,91). This risk increases to 7% to 87% ifthe fellow eye has CNV (34,73,87,104). Bressler andcolleagues, in their age-adjusted analysis, showedthat greater than 20 drusen, the presence of softconfluent type, and focal RPE hyperpigmentationwere more often noted in the fellow eyes withunilateral exudative maculopathy than in eyes with

(A) (B) (C)

Figure 9 A 76-year-old woman who was enrolled and treated in the unilateral DLS pilot study. Visual acuity

in the right eye was 20/30-2; confluent soft drusen were present. (A) By four months, drusen had resolvedand at nine months, fleck hemorrhages were evident at the right fovea associated with a laser scar (arrow).

(B,C) Fluorescein angiography showed leakage from this laser scar (arrow). Source: From Ref. 98.


(B)

(D)

(F)

(A)

(E)

(C)

Figure 10 An 81-year-old male who had undergone photodynamic therapy five times for

subfoveal choroidal neovascularization in right eye, OD [3/28/05: (A) Color photo; (B) red-freephoto; (E, G) - fluorescein angiogram, (FA)] was enrolled into the anacortave acetate risk reduction

trial for high risk drusen for the left eye, OS [(C) Color photo; (D) red-free photo; (F, G) - FA].He underwent anacortave acetate injections on 4/7/05 and 9/30/05. His visual acuity OS

remained 20/20 even though a new pigment epithelial detachment was noted on 7/7/05[(H) Color photo; (I & J) - FA]. (Continued )


bilateral drusen (88). Focal hyperpigmentation andconfluence of drusen are associated with an increasedrisk of progression to exudative AMD (11). Focalhyperpigmentation may be associated with subclini-cal subretinal neovascularization that cannot bedetected by fluorescein angiogram (91). It may also

Table 2 The Beaver Dam Eye Study; 5-Year Incidence ofNon-exudative AMD Findings

p-valueO75 yearsof age

43–54 yearsof age

1. Large drusen

(125–249 microns)

!0.05 17.6% 2.1%

2. Large drusen

(O250 microns)!0.05 6.5% 0.2%

3. Soft indistinct drusen !0.05 16.3% 1.8%

4. Retinal pigment

abnormalities

!0.05 12.9% 0.9%

5. Pure geographic

atrophy

!0.05 1.7% 0%

Source: From Ref. 31.Figure 11 High risk drusen showing large confluent drusen withRPE hyperpigmentation.

(H)

(I) (J)

(G)

Figure 10 Continued (Caption on facing page ).


reflect the changes that have already occurred inthe RPE, Bruch’s membrane, and choriocapillaris,which facilitate future development of CNV andmay simply suggest the chronicity of the diseaseprocess (Fig. 11) (91).

The AREDS research study group has describeda simplified clinical scale to define risk categories for afive-year risk of developing advanced AMD (86) ineyes without advanced AMD at baseline, or the riskin the unaffected fellow eye when advanced AMD ispresent in one eye at baseline. It is a five-step scale(0–4) that predicts an approximate five-year risk ofdeveloping advanced AMD in at least one eye asfollows: 0 factor, 0.5%; 1 factor, 3%; 2 factors, 12%;3 factors, 25%; and 4 factors, 50%. The scale sumsretinal risk factors in both eyes. The risk factors are thepresence of one or more large drusen (O125 mm—width of a retinal vein at the disk margin) andpigment stippling; each characteristic gets one pointfor each eye. Advanced AMD in one eye at baseline isgiven two scores. The presence of intermediate drusen(R63–128 mm) in both eyes is given one score.

The AREDS trial also noted that the drusen areawas stronger and a more consistent predictor of pro-gression to advanced AMD than the drusen size.However, for practical clinical purposes, the drusennumber and type was used for calculating theseverity score.

MONITORING NON-EXUDATIVE AMD

Patients with intermediate drusen (O63 mm) or thosewith exudative AMD in fellow eyes are recommendedto take high-dose vitamins as per the AREDS study.

Amsler grid testing is a sensitive indicator ofprogression of the disease process. Straight door andwindow frames may be the ways to check for anymetamorphopsia. Patients are encouraged to seekmedical help if visual distortion, metamorphopsia,loss of central vision, or any new symptoms occur.These herald the growth of choroidal neovascularmembranes. The early detection of the choroidalneovascular membranes may facilitate treatment asdiscussed in other chapters.

SUMMARY POINTS

& Prevalence of AMD increases with age.& Non-exudative AMD is the most common form

of AMD.& Factors associated with AMD include advancing

age, genetic component, nutrition, photic injury,smoking, and systemic hypertension.

& High-risk characteristics of drusen for develop-ment of CNV include: soft type, large size, greaterthan five in number, confluent, and presence ofRPE stippling.

& Disappearance of drusen can occur spontaneouslyor may follow after laser.

& GA is the advanced form of non-exudative AMD.& Monitoring VA and visual symptoms for the pro-

gression to the exudative AMD is of utmostimportance in applying timely treatment (105).

REFERENCES

1. Leibowitz HM, Krueger DE, Maunder LR, et al. TheFramingham Eye Study monograph: an ophthalmologicaland epidemiological study of cataract, glaucoma, diabeticretinopathy, macular degeneration, and visual acuity in ageneral population of 2631 adults, 1973–1975. SurvOphthalmol 1980; 24(Suppl.):335–610.

2. American Academy of Ophthalmology. New Therapiesfor Macular Degeneration, 2005. (www.aao.org/news-room/facts/amd.cfm)

3. Age-Related Eye Disease Study Research Group. Potentialpublic health impact of Age-Related Eye Disease Studyresults: AREDS report no. 11. Arch Ophthalmol 2003;121:1621–4.

4. Eye Diseases Prevalence Research Group. Prevalence ofage-related macular degeneration in the United States.Arch Ophthalmol 2004; 122:564–72.

5. Kahn HA, Leibowitz HM, Ganley JP, et al. TheFramingham Eye Study. I. Outline and major prevalencefindings. Am J Epidemiol 1977; 106:17–32.

6. Thylefors B. A global initiative for the elimination ofavoidable blindness. Am J Ophthalmol 1998; 125:90–3.

7. Evans J, Wormald R. Is the incidence of registrable age-related macular degeneration increasing? Br J Ophthalmol1996; 80:9–14.

8. Ambati J, Ambati BK, Yoo SH, Lanchulev S, Adamis AP.Age-related macular degeneration: etiology, pathogenesis,and therapeutic strategies. Surv Ophthalmol. 2003;48(3):257–93.

9. Klein R, Klein BE, Linton KL. Prevalence of age-relatedmaculopathy. The Beaver Dam Eye Study. Ophthalmology1992; 99:933–43.

10. Bressler NM, Munoz B, Maguire MG, et al. Five-yearincidence and disappearance of drusen and pigmentabnormalities. Waterman study. Arch Ophthalmol 1995;113:301–8.

11. Sarraf D, Gin T, Yu F, Brannon A, Owens SL, Bird AC.Long-term drusen study. Retina 1999; 19:513–9.

12. Gragoudas ES, Chandra SR, Friedman E, Klein ML, vanBuskirk M. Disciform degeneration of the macula. II.Pathogenesis. Arch ophthalmol 1976; 94:755–7.

13. GreenWR,McDonnellPJ,Yeo JH.Pathologic featuresof senilemacular degeneration. Ophthalmology 1985; 92:615–27.

14. Sommerburg O, Kuenen JE, Bird AC, van Kuijk FJ. Fruitsand vegetables that are sources for lutein and zeaxanthine:the macular pigment in human eyes. Br J Ophthalmol1998; 82:907–10.

15. Ishibashi T, Patterson R, Ohnishi Y, Inomata H, Ryan SJ.Formation of drusen in the human eye. Am J Ophthalmol1986; 101:342–53.


16. Abdelsalam A, Del Priore L, Zarbin MA. Drusen in age-related macular degeneration: pathogenesis, naturalcourse, and laser-photociagulation-induced regression.Surv Ophthalmol 1999; 44:1–29.

17. Gregor Z, Bird AC, Chisolm IH. Senile disciform maculardegeneration in the second eye. Br J Ophthalmol 1977;61:141–7.

18. Kahn HA, Leibowitz HM, Ganley JP, et al. TheFramingham Eye Study. II. Association of ophthalmicpathology with single variables previously measured inthe Framingham Heart Study. Am J Epidemmiol 1977;106:33–41.

19. Goldberg J, Flowedew G, Smith E, Brody JA, Tso MO.Factors associated with age related macular degeneration.An analysis of data from the first National Health andNutrition Examination Survey. Am J Epidemiol 1988;128:700–10.

20. Vinding T, Appleyard M, Nyboe J, Jensen G. Risk factoranalysis for atrophic and exudative age related maculardegeneration. An epidemiologic study of 1000 aged indi-viduals. Acta Ophthalmol (Copenh) 1992; 70:66–72.

21. Klein ML, Mauldin WM, Stoumbos VD. Hereditary andage related macular degeneration. Observation in mono-zygotic twins. Arch Ophthalmol 1994; 112:932–7.

22. Klein ML, Schultz DW, Edwards A, et al. Age-relatedmacular degeneration: clinical features in a large familyand linkage to chromosome 1q. Arch Ophthalmol 1998;116:1082–8.

23. Maltzman BA, Mulvihill MN, GreenbaumA. Senile macu-lardegeneration and risk factors; a case-controlled study.Ann Ophthalmol 1979; 11:1197–201.

24. Hyman L, Schachat AP, He Q, Leske MC. Hypertension,cardiovascular disease, and age-related macular degener-ation. Age-related Macular Degeneration risk FactorsStudy Group. Arch Ophthalmol 2000; 118:351–8.

25. Klein R, Klein BE, Jensen SC. The relation of cardiovascu-lardisease and its risk factors to the 5-year incidence ofage-related maculopathy: the Beaver Dam Eye Study.Ophthalmology 1997; 104:1804–12.

26. Macular Photocoagulation Study Group. Laser photocoa-gulation for juxtafoveal choroidal neovascularization.Five-year results from randomized clinical trials. ArchOphthalmol 1994; 112:500–9.

27. Klein BE, Klein R. Cataracts and macular degeneration inolder Americans. Arch Ophthalmol 1982; 100:571–3.

28. Delcourt C, Cristol JP, Tesseir F, et al. Age-related maculardegeneration and antioxidant status in the POLA study.Pathologies Oculaires Liees a l’Age. Arch Ophthalmol1999; 117:1384–90.

29. Mares-Perlman JA, Brady WE, Klein R, Vanden-Lagenberg GM, Klein BE, Palta M. Dietary fat and age-related maculopathy. Arch Ophthalmol 1995; 113:743–8.

30. Tso MO. Pathogenetic factors of aging macular degener-ation. Ophthalmology 1985; 92:628–35.

31. Klein R, Klein BEK, Jensen SC, Meuer SM. The five-year incidence and progression of age-related maculo-pathy: the Beaver Dam Eye Study. Ophthalmology 1997;104:7–21.

32. Smith W, Assink J, Klein R, et al. Risk factors for age-related macular degeneration: pooled findings from threecontinents. Ophthalmology 2001; 108:697–704.

33. Mitchell P, Wang JJ, Foran S, Smith W. Five-year incidenceof age-related maculopathy lesions: the Blue MountainsEye. Ophthalmology 2002; 109:1092–7.

34. Macular Photocoagulation Study Group. Risk factors forchoroidal neovascularization in the second eye of patients

with juxtafoveal or subfoveal choroidal neovasculariza-tion secondary to age-related macular degeneration. ArchOphthalmol 1997; 115:741–7.

35. Mitchell P, Smith W, Attebo K, Wang JJ. Prevalence of age-related maculopathy in Australia. The Blue MountainsStudy. Ophthalmology 1995; 102:1450–60.

36. Gregor Z, Joffe L. Senile macular changes in black African.Br J Ophthalmol 1978; 62:547–50.

37. Friedman DS, Katz J, Bressler NM, et al. Racial differencesin the prevalence of age-related macular degeneration:the Baltimore Eye Survey. Ophthalmology 1999; 106:1049–55.

38. Klein R, Klein BEK, Marino EK, et al. Early age-relatedmaculopathy in the cardiovascular health study. Ophthal-mology 2003; 110(1):25–33.

39. Sundelin SP, Nilsson SE. Lipofuscin-formation in culturedretinal pigment epithelial cells is related to their melanincontent. Free Radic Biol Med 2001; 30:74–81.

40. Mitchell P, Smith W, Attebo K, Wang JJ. Iris color, skinsun sensitivity and age-related maculopathy. The BlueMountains Eye Study. Ophthalmology 1998; 105:1359–63.

41. West SK, Rosenthal FS, Bressler NM, et al. Exposure tosunlight and other risk factors for age-related maculardegeneration. Arch Ophthalmol 1989; 107:875–979.

42. Age-Related Eye Disease Study Research Group. Riskfactors associated with age-related macular degeneration.A case-control study in the age-related eye disease study:age-related eye disease study report number 3. Ophthal-mology 2000; 107:2224–32.

43. The Eye Disease Case-Control Study Group. Risk factorsfor neovascular age related macular degeneration. ArchOphthalmol 1992; 110:1701–8.

44. Vinding T. Age-related macular degeneration. Macularchanges, prevalence and sex ratio. An epidemiologicalstudy of 1000 aged individuals. Acta Ophthalmol(Copenh) 1989; 67:609–16.

45. Hyman LG, Lilienfeld AM, Ferris FL, III, et al. Senilemacular degeneration: a case-control study. AmJ Epidemiol 1983; 118:213–27.

46. Weiter JJ, Delori FC, Wing GL, et al. Relationship of senilemacular degeneration to ocular pigmentation. AmJ Ophthalmol 1985; 99:185–7.

47. Frank RN, Puklin JE, Stock C, et al. Race, iris color, andage-related macular degeneration. Trans Am OphthalmolSoc 2000; 98:109–15.

48. Gartner S, Henkind P. Aging and degneration of thehumanmacula. 1. Outer nuclear layer and photoreceptors.Br J Ophthalmol 1981; 63:23–8.

49. Gottsch JD, Bynoe LA, Harlan JB, et al. Light—induceddeposits in Bruch’s membrane of protoporphyric mice.Arch Ophthalmol 1993; 111:126–9.

50. Cruikshanks KJ, Klein R, Klein BE. Sunlight and age-related macular degeneration. The Beaver Dam EyeStudy. Arch Ophthalmol 1993; 111:514–8.

51. Taylor HR, Munoz B, West S, et al. Visible light and risk ofage-related macular degeneration. Trans Am OphthalmolSoc 1990; 88:163–73.

52. Evans JR. Risk factors for age-related macular degener-ation. Prog Retin Eye Res 2001; 20:227–53.

53. Age-Related Eye Disease Study Research Group. Arandomized, placebo-controlled, clinical trial of high-dose supplementation with vitamins C and E, beta caro-tene, and zinc for age-related macular degeneration andvision loss: AREDS Report No. 8. Arch Ophthalmol 2001;119:1417–36.

54. Seddon JM, Ajani UA, Sperduto RD, Hiller R, Blair N,Burtin TC, Farber MD, Gragoudas ES, Haller J, Miller DT,


et al. Dietary carotenoids, vitamins A, C and E andadvanced age-related macular degeneration. Eye DiseaseCase-Control Study group. JAMA 1994; 272:1413–20.

55. Seddon JM, Rosner B, Sperduto RD, et al. Dietary fat andrisk for advanced age-related macular degeneration. ArchOphthalmol 2001; 119:1191–9.

56. Tomany SC, Mmed JJW, Van Leevuen R, et al. Riskfactors for incident age-related macular degeneration:pooled data from 3 continents. Ophthalmology 2004;111(7):1280–7.

57. Thorton J, Edwards R, Mitchell P, Harrison RA, Buchan I,Kelly SP. Smoking and age-related macular degeneration:a review of association. Eye 2005; 19:935–44.

58. Khan JC, Thurlby DA, Shahid H, et al. Smoking and agerelated macular degeneration: the number of pack years ofcigarette smoking is a major determinant of risk for bothgeographic atrophy and choroidal neovascularisation. BrJ Ophthalmol 2006; 90(1):75–80.

59. Solberg Y, Rosner M, Belkon M. The association betweencigarette smoking and ocular diseases. Surv Ophthalmol1998; 42:535–47.

60. Seddon J, Gensler G, Klein ML, Milton RC. Evaluationof plasma homocysteine and age-related maculardegeenration. Am J of Ophthalmol 2006; 141(1):201–3.

61. Mares-Perlman JA, Brady WE, Klein R, et al. Serumantioxidants and age-related macular degeneration in apopulation-based case-control study. Arch Ophthalmol1995; 113:1518–23.

62. Gottfredsdottir MS, Sverrisson T, Musch DC, Stefansson E.Age Related macular degeneration in monozygotic twinsand their spouces in Iceland. Acta Ophthalmol Scand1999; 77:422–5.

63. Klaver CC, Wolfs RC, Assink JJ, et al. Genetic risk of agerelated maculopathy. Population based familial aggrega-tion study. Arch Ophthalmol 1998; 116:1646–51.

64. Hammond BR, Jr., Webster AR, Sneider H, et al. Geneticinfluence on early age-related maculopathy: a twin study.Ophthalmology 2002; 109:730–6.

65. Seddon JM, Ajani UA, Mitchell BD. Familial aggregationof age-related maculopathy. Am J Ophthalmol 1997;123:199–206.

66. Klein RJ, Zeiss C, Chew EY, et al. Complement factor Hpolymorphism in age-related macular degeneration.Science 2005; 308:385–9.

67. Edwards AO, Ritter R, III, Abel KJ, Manning A,Panhuysen C, Farrer LA. Complement factor H polymor-phism and age-related macular degeneration. Science2005; 308:421–4.

68. Haines JL, Hauser MA, Schmidt S, et al. Complementfactor H variant increases the risk of age-related maculardegeneration. Science 2005; 308:419–21.

69. Donders FC. Beitrage zur pathologischen Anatomie desAuges. Arch Ophthalmol 1854; 1(II):106–18.

70. Spraul CW, Grossniklaus HE. Characteristics of drusenand Bruch’s membrane in postmortem eyes with age-related macular degeneration. Arch Ophthalmol 1997;115:267–73.

71. Green WR, Enger C. Age-related macular degenerationhistopathologic studies. The 1992 Lorenz E. ZimmermanLecture. Ophthalmology 1993; 100:12519–35.

72. Gass JD. Drusen and disciform macular detachment anddegeneration. Trans Am Ophthalmol Soc 1972; 70:409–36.

73. Sarks SH. Council lecture. Drusen and their relationship tosnile macular degeneration. Aust J Ophthalmol 1980;8:117–30.

74. Gass JD. Drusen and disciform macular detachment anddegeneration. Arch Ophthalmol 1973; 90:206–17.

75. Sarks JP, Sarks SH, Killingsworth MC. Evolution ofgeographic atrophy of the retinal pigmnet epithelium.Eye 1988; 2:552–77.

76. Leu ST, Batni S, Radeke MJ, Johnson LV, Anderson DH,Clegg DO. Drusen are cold spots for proteolysis: expres-sion of matrix metalloproteinases and their tissue inhibitorproteins in age-related macular degeneration. Exp Eye Res2002; 74:141–54.

77. Sarks JP, Sarks SH, Killingsworth MC. Evolution of softdrusen in age-related macular degeneration. Eye 1994;8:269–83.

78. Green WR, Key SN. Senile macular degeneration: ahistopathologic study. Trans Am Ophthalmol Soc 1977;75:180–254.

79. CoffeyAJ, Brownstein S. The prevalence ofmacular drusenin postmortem eyes. Am J Ophthalmol 1986; 102: 164–71.

80. Zarbin MA. Current concepts in the pathogenesis of age-related macular degeneration. Arch Ophthalmol 2004;122:598–614.

81. Sarks SH. Ageing and degeneration in the macular region:a clinicopathologic study. Br J Ophthalmol 1976; 60:324–421.

82. Sarks SH, Arnold JJ, Killingsworth MC, Sarks JP. Earlydrusen formation in the normal and ageing eye and theirrelation to age-related maculopathy: a clinicopathologicalstudy. Br J Ophthalmol 1999; 83:358–68.

83. Sarks SH. Drusen patterns predisposing to geographicatrophy of the retinal pigment epithelium. Aust J Ophthal-mol 1982; 10:91–7.

84. Bressler NM, Silva JC, Bressler SB, et al. Clinicopathologiccorrelation of drusen and retinal pigment epithelial abnor-malities in age-related macular degeneration. Retina 1994;14:130–42.

85. Gass JD. Stereoscopic Atlas of Macular Diseases: Diseasesand Treatment. Vol. 1. Louis: CV Mosby, 1987.

86. Age-Related Eye Disease Study Research Group. A simpli-fied severity scale for age-related macular degeneration:AREDS Report No. 18. Arch Ophthalmol 2005; 123:1570–4.

87. Bressler SB, Maguire MG, Bressler NM, Fine SL, TheMacular Photocoagulation Study Group. Relationship ofdrusen and abnormalities of the retinal pigment epi-thelium to the prognosis of neovascular maculardegeneration. Arch Ophthalmol 1990; 108:1442–7.

88. Bressler NM, Bressler SB, Seddon JM, Gragoudas ES,Jacobson LP. Drusen characteristics in patients withexudative vs non-exudative age-related macular degener-ation. Retina 1988; 8:109–14.

89. HolzFG,WolfensbergerTJ, PiguetB, et al. Bilateralmaculardrtusen in age-related macular degeneration. Prognosisand risk factors. Ophthalmology 1994; 101:1522–8.

90. Javornik NB, Hiner CJ, Marsh MJ, Maguire MG,Bressler NM, MPS Group. Changes in drusen and RPEabnormalities in age-related macular degeneration. InvestOphthalmol Vis Sci 1992; 33:1230.

91. Smiddy WE, Fine SL. Prognosis of patients with bilateralmacular drusen. Ophthalmology 1984; 91:271–7.

92. Choroidal Neovascularization Prevention Trial ResearchGroup. Laser treatment in fellow eyes with large drusen:updated findings from a pilot randomized clinical trial.Ophthalmology 2003; 110:971–8.

93. OlkRJ, FribergTR, StickneyKL, et al. Therapeutic benefits ofinfrared (810 nm) diode laser macular grid photocoagula-tion in prophylactic treatment of non-exudative age-relatedmacular degeneration: two-year results of a randomizedpilot study. Ophthalmology 1999; 106:2082–90.


94. Friberg TR,Musch DC, Lim JI, et al. Prophylactic treatmentof age-related macular degeneration report number 1:810-nanometer laser to eyes with drusen. unilaterallyeligible patients. Ophthalmology 2006; 113:612–22.

95. Owens SL, Bunce C, Brannon AJ, et al. Prophylactic lasertreatment hastens choroidal neovascularization in theunilateral age-related maculopathy: final results ofdrusen laser study. Am J Ophthalmol 2006; 141:276–81.

96. Choroidal Neovascularization Trial Research Group. Lasertreatment in eyes with large drusen. Short-term effectsseen in a pilot randomized clinical trial. Ophthalmology1998; 105:11–23.

97. Kaiser RS, Berger JW, Shin DS, Maguire MG, CNVPTStudy Group. Laser burn intensity and the risk forchoroidal neovascularization in the CNVPT Fellow EyeStudy. Invest Ophthalmol Vis Sci 1999; 40(Suppl.):S377(Abstract).

98. Sarah L, Owens MD, Robyn H, et al. Fluorescein angio-graphic abnormalities after prophylactic macularphotocoagulation for high-risk age-related maculopathy.Am J Opthalmol 1999; 127(6):681–7.

99. Sunness JS, Rubin GS, Applegate CA, et al. Visual functionabnormalities and prognosis in eyes with age-related

geographic atrophy of the macula and good visualacuity. Ophthalmogy 1997; 104:1677–91.

100. Bressler NM, Bressler SB, West SK, Fine SL, Taylor HR. Thegrading and prevalence of macular degeneration in Chesa-peake Bay watermen. Arch Ophthalmol 1989; 107:847–52.

101. Ferris FL, Fine SL, Hyman G. Age-related maculardegeneration and blindness due to neovascular maculo-pathy. Arch Ophthalmol 1984; 102:1640–2.

102. Vingerling JR, Dielemans I, Hofman A, et al. The preva-lence of age-related maculopathy in the Rotterdam study.Ophthalmology 1995; 102:205–10.

103. Cruickshanks KJ, Hamman RE, Klein R, Nondahl DM,Shetterly SM. The prevalence of age-related maculopathyby geographic region and ethnicity. The Colorado–Wisconsin Study of age-Related Maculopathy. ArchOphthalmol 1997; 115:242–50.

104. Strahlman ER, Fine SL, Hillis A. The second eye of patientswith senile macular degeneration. Arch ophthalmol 1983;101:1522–8.

105. Mitchell P, Foran S. Age-related eye disease study severityscale and simplified severity scale for age-related maculardegeneration. Arch Ophthalmol 2005; 123(11):1598–9(Comment; Editorial).


7

Geographic AtrophySharon D. SolomonRetina Division, Wilmer Eye Institute, Johns Hopkins University School of Medicine,


Janet S. SunnessThe Richard E. Hoover Services for Low Vision and Blindness, Greater Baltimore Medical Center,


INTRODUCTION

Geographic atrophy (GA) of the retinal pigment epi-thelium (RPE) is the advanced form of non-neovascular age-related macular degeneration(AMD) and is associated with gradual, progressiveloss of central vision. Dense scotomas have beenshown to correspond to the retinal areas affected byGA (1). These scotomas involve the parafoveal andperifoveal retina early in the course of the disease,sparing the foveal center until late in the course of thedisease (2–5). Consequently, GA is responsible forapproximately 20% of the legal blindness secondaryto AMD, compared to choroidal neovascularization(CNV), which tends to involve the foveal center muchearlier in the course of the disease and accounts fornearly 80% of the legal blindness secondary to AMD(6). However, the parafoveal and perifoveal scotomasin the early stages of GA compromise the patient’sability to read and to recognize faces, often despite theretention of good visual acuity, and account for a muchlarger percentage of moderate visual loss in thoseaffected (7). In addition, GA is present binocularly inmost patients. The prevalence of GA in the population75 years of age or older is approximately 3.5%, halfthat of neovascular AMD (8,9), and increases to 22% inthose 90 years of age or older (10). While there aretreatments for CNV, there is presently no definitivetreatment available for GA. As our understanding ofGA grows, it is hoped that medical and surgicalinterventions will be developed to completely halt itsprogression rate and to prevent subsequent moderateand severe visual loss.

CLINICAL FEATURES OF GA

GA is easily recognized clinically, as it appears as awell-demarcated area of decreased retinal thickness,compared to the surrounding retina, with a relativechange in color that allows for increased visualization

of the underlying choroidal vessels. Both the locationand pattern of the atrophy may vary in appearance.Forty percent of eyes with macular GA also haveperipapillary GA, which may become confluent withthe macular atrophy (7). There may be pigmentaryalteration, either hypopigmentation or hyperpigmen-tation, surrounding the macular atrophy. Peripheralreticular degeneration of the pigment epithelium ispresent in about 41% of eyes (7). Drusen, usually amixture of the soft and calcific types, are present inmost eyes until the GA becomes so extensive as toresorb the macular drusen (2). The increased choroidalvessel detail in the area of GA is usually the mosteasily identified fundus change and further reflectsthe extent of RPE attenuation. On fluoresceinangiography, this translates into an area of hyperfluor-escence that corresponds to the ophthalmoscopicborders of the GA, secondary to transmission defectand staining. The intensity of hyperfluorescence fromchoroidal flush may vary depending on the presenceor absence of the underlying choriocapillaris (4).Fluorescein angiography may also aid in distinguish-ing GA from occult CNV, which may otherwise appearclinically indistinguishable.

Hemorrhagemay occur in eyes with GA. Thoughthis may be a reflection of the development of CNVand herald a more precipitous decline in visual func-tion secondary to neovascular maculopathy, often thesmall areas of CNV that develop are transient and maybecome clinically inapparent a few months later (11).Hemorrhages have also been described in GA in thecomplete absence of any CNV (11,12). In general,however, the presence of hemorrhage, especiallywhen associated with a sudden change in vision,warrants an angiographic evaluation for the presenceof CNV (13).

Though there are frequently small areas of retinalsparing within the GA, especially at the center of themacula, foveal localization may still prove challenging

on clinical examination, on the color fundus photo-graph, and on the angiogram. Clinically, the location ofxanthophyll, if visible, is helpful in determining thelocation of the foveal center. On fluorescein angio-graphy, the intense hyperfluorescence associatedwith the GA may obscure the view of the entirefoveal avascular zone, making foveal localization aless certain task. Under such circumstances, the red-free photographs can often be of significant help. Thepresence of xanthophyll may suggest that the foveahas visual function, even if the retina appears atrophicand nonfunctional (14). Testing with devices such asthe scanning laser ophthalmoscope (SLO) may help toascertain the central visual potential that remains (1,5).

HISTOPATHOLOGY AND PATHOGENESIS

Histopathologic examination of eyes with GA demon-strates a loss of RPE cells in the area of atrophy with asecondary loss of overlying photoreceptor cells (15).The choriocapillaris may also be absent, and there isindeed some experimental evidence to suggest thatwhen the RPE is removed or has atrophied, thechoriocapillaris involutes secondarily (16,17). GA isassociated with thickening of Bruch’s membranesecondary to the deposition of basal laminar andbasal linear deposits in the surrounding retina (18).Therefore, histologically, GA has been thought of asthe end stage of the AMD process if CNV does notdevelop (19). GA may also occur following the flat-tening of a retinal pigment epithelial detachment (20).

There is controversy as to whether the loss ofRPE cells, perhaps related to the deposits in andnear Bruch’s membrane, is the primary event in theevolution of GA, or whether this RPE atrophydevelops secondary to choroidal vascular insuffi-ciency. Green and others have argued that thepresence of choroidal vascular insuffiency shouldresult in the subsequent degeneration of all the outerretinal layers (15). This is not seen in eyes with GA.Friedman suggests that choroidal vascular resistancemay predispose to the development of AMD, speci-fically to the development of high-risk drusen andCNV (21). However, a causal association betweenchoroidal vascular resistance and GA has not beenestablished to date. Grunwald has found that fovealchoroidal blood flow is reduced in eyes with earlyatrophic AMD and good visual acuity (22), and thiswork is continuing to better characterize the relation-ship between choroidal blood flow and thedevelopment of advanced AMD.

PREVALENCE AND EPIDEMIOLOGY OF GA

Population-based studies, such as the Beaver Dam EyeStudy and the Rotterdam Study, have examined the

prevalence of GA in the elderly and compared it to theprevalence of CNV in the same groups. The prevalenceof GA is approximately 3.5% for people age 75 andabove in the United States and other developednations, half that of CNV (8,9). The prevalence of GAincreases with age and is actually more common thanCNV in older age groups. In the population overage 90, the prevalence of GA can reach levels of 20%to 35% (10,23). The studies indicate that there is alower prevalence of GA in African-Americans (24).There does not appear to be a gender difference inprevalence across the populations studied.

In the Beaver Dam Eye Study, 8% of eyes withdrusen larger than 250 mm went on to develop GAover a five-year period. Eyes that developed GA allhad pigmentary abnormalities and at least 0.2 MacularPhotocoagulation Study (MPS) disc areas of drusenat baseline (25). Of the eyes with GA, 42% had a visualacuity of 20/200 or worse. This was similar to the 48%of eyes with neovascular AMD that had a comparablelevel of severe visual loss (25).

GA is bilateral in 48% to 65% of cases (7,26).While the rate of bilateral severe vision loss is lowerfrom GA than from CNV, GA is still responsible for afull 20% of the binocular legal blindness secondaryto AMD (6). These statistics for severe visual lossmeasure only the incidence of legal blindness andsignificantly underestimate the disability associatedwith GA. A patient with GA and only moderatelyimpaired visual acuity may not be able to read or torecognize faces because the object being visualizeddoes not “fit” into the spared central island ofvision (5).

Systemic Risk FactorsA number of population-based studies haveattempted to identify possible risk factors for thedevelopment of GA and neovascular AMD. TheBeaver Dam Eye Study did not demonstrate a relation-ship between GA and cholesterol level or alcoholintake (8,27). While current or past smoking was asignificant risk factor for the presence of GA forwomen in the Blue Mountains Eye Study, the sameassociation did not reach statistical significance formen (28). In Sunness’ study, there was a trend forcurrent smokers to have a more rapid progression ofGA than nonsmokers (7). The same study suggestedthat patients who are pseudophakic or aphakic do nothave more rapid progression of GA than their phakiccounterparts (7).

More recently, the Age-Related Eye DiseaseStudy (AREDS), a multicenter study of the naturalhistory of AMD and cataract, reported its findings onpossible risk factors for the development of GA andneovascular AMD. The presence of GA was found tobe associated with increasing age and smoking,

112 SOLOMON AND SUNNESS

confirming the findings of previous population-basedstudies (29). In addition, there appeared to be apositive association between the use of antacids andthe use of thyroid hormones and the presence of GA(29). These two associations have not been previouslyreported and will certainly prompt further investi-gation. Level of education was found to be inverselyproportional to the presence of GA in that personswith more years of formal schooling seemed to be atlower risk for GA (29).

HereditySeveral studies have suggested that genetic factorsmayalso be important in the pathogenesis of AMD. Heredi-tary retinal dystrophies, with clinical manifestationssimilar to AMD,may share potential candidate suscep-tibility genes. For example, a mutation of theRDS/peripherin gene has been shown to be associatedwith Zermatt macular dystrophy, which is a dominant,age-related, progressive macular dystrophy thatresembles GA in its later stages (30). Particular interesthas focused on the ABCR gene which is responsible forautosomal recessive Stargardt macular dystrophy. Onestudy reported that 16% of patients with AMD had amutation in this gene, compared with 13% of Star-gardt’s patients and 0.5% of the general population(31). The mutation was identified primarily in eyesaffected by atrophic AMD, on the continuum betweenearly and advanced disease (32,33). There is somedisagreement with these findings however.

Pedigree studies have included families in whichGA and CNV occur in different members as well astwin studies, where both twins are affected by GA orwhere one twin has GA and the other has an earlieratrophic form of AMD. In one large AMD family,

linkage has been reported to markers in 1q25-q31(34). Recent data also suggest that the apolipoproteinE epsilon 4 allele may be associated with a reducedrisk for the development of AMD (34). Identification ofthose genetic factors that play a role in the patho-genesis of AMD may aid with the recognition of thoseat risk and permit possible lifestyle modifications toprevent or decrease the severity of disease.

A recent exciting finding by several groups ofinvestigators is the association of advanced AMDwithspecific complement factor H polymorphisms. Thisassociation is present for GA alone as well. Theassociation suggests that there might be an inflam-matory basis for advanced AMD (35).

NATURAL HISTORY OF GA

Over the last three decades, several studies havedescribed the progression of GA with respect tovisual acuity loss and actual enlargement of theatrophy in populations of patients. Their observationsform the foundation of our knowledge of the naturalhistory of GA.

GA typically develops in eyes that, at baseline,have drusen or pigmentary alteration. As drusen fade,focal areas of atrophy may develop in theirplace, enlarge, and evolve into GA (2,36,37). Alter-natively, areas of mottled hypopigmentation may alsopredispose to the development of GA (2). The pro-gression goes through a number of stages. Initially,single or multifocal areas of GA may be found in theregion around the fovea. As these areas enlarge andcoalesce over time, they can form a horseshoe ofatrophy that spares the foveal center (Fig. 1). Thishorseshoe of atrophy may close off into a ring of

(A) (B)

Figure 1 Four-year progression in geographic atrophy (GA). (A) There are multifocal areas of GA, along

with drusen and pigmentary alteration. (B) Four years later, the areas of GA have enlarged and coalesced,forming a horseshoe of atrophy surrounding the fovea.

7: GEOGRAPHIC ATROPHY 113

atrophy that still permits foveal preservation. In the latestages of GA, the fovea itself becomes atrophic andnonseeing, from further coalescence of the GA,requiring the patient to use eccentric retinal loci forfixation and seeing (2). GAmay also occur secondary toan RPE detachment. Elman and others have reportedRPEdetachments flattening andgoing on to evolve intoGA in about 20% of cases (20,38–40). Whether the GAwas extrafoveal or foveal depended on the precedinglocation of the RPE detachment. Because visual losstends to be gradual and subtle, and takes place over aperiodof years, patientsmaynot seekmedical attentionuntil the solid central GA is present.

Sunness demonstrated that the median visualacuity tended to be worse in eyes with larger totalatrophic areas, with the most dramatic difference inmedian acuity occurring between eyes with less thanthreeMPSdisc areas of central atrophy (i.e., within fourMPS disc areas of the foveal center) and those withgreater than threeMPS disc areas of central atrophy (7).

GA continues to enlarge over time with a medianrate of enlargement over a two-year period of 1.8 MPSdisc areas (7). The amount of enlargement of totalatrophic area has been shown to increase withincreasing baseline size of atrophyup to approximatelyfiveMPS disc areas of baseline, above which the rate ofenlargement stayed about the same. For eyeswithmorethan 10 MPS disc areas, it is difficult to measureenlargement because the borders of the GA oftenextend past the photographic field. Baseline level ofvisual acuity did not appear to significantly affect thedegree of enlargement of total atrophy (7). The only riskfactor that has been linked to more rapid enlargementof the total atrophic area was a baseline total atrophyarea greater than three MPS disc areas (7). Neither thephakic status of the study eye nor a history of hyperten-sion in the patient was shown to be a risk factor for theenlargement of total atrophy (7). Therewas an apparenttrend for smokers to have a more rapid enlargement ofatrophy (7).

GA is associated with a significant decline invisual acuity over time in many eyes. Overall, 31% ofall study eyes lost three or more lines of visual acuity,doubling the visual angle, by two years of follow-up,and 53% lost three or more lines of visual acuity byfour years of follow-up (7). Rates of severe vision loss,that is a quadrupling of the visual angle or at least sixlines of visual acuity loss, were 13% for all study eyesby two years and 29% by four years (7). There was asignificantly larger rate of moderate and severe visionloss for eyes with a baseline visual acuity of better than20/50. At two years of follow-up, 41% of these eyeswith good acuity at baseline lost three or more lines ofvisual acuity and 21% lost six or more lines of visualacuity (7). Those numbers grew to 70% at four years offollow-up for moderate vision loss and 45% at four

years for severe vision loss (7). Twenty-seven percentof the eyes with visual acuity of 20/50 or better atbaseline had visual acuity of 20/200 or worse at fouryears of follow-up (7). The presence of CNV in thefellow eye did not appear to affect the rate of visualacuity loss in the GA study eye (7). Risk factors thathave thus far been identified for moderate vision lossinclude baseline visual acuity of better than 20/50 andlightest iris color (7). Among eyes with visual acuitybetter than 20/50, the presence of GA within 250 mmof the foveal center was a strong risk factor for athree-line visual acuity loss (7). There was no apparentassociation between phakic status of the study eye,hypertension, or smoking with moderate vision lossdemonstrated in the study by Sunness et al. (7).

During the four-year follow-up period, GAappeared to progress through various stages,including the small, multifocal, horseshoe, ring, andsolid stages described in previous studies (2,3,7). Foreyes that did not have the solid pattern of atrophy atbaseline and which did not develop CNV during thecourse of the study, 61% advanced to a differentconfiguration over the two-year follow-up period (7).However, those eyes that had the same configurationat the two-year follow-up as at baseline still had visualacuity loss, most notably in the ring group where 50%of eyes lost three or more lines of visual acuity (7).

The size and rate of progression of atrophy arehighly correlated between the two eyes of patientswith bilateral GA. This includes the baseline area oftotal atrophy, the baseline area of central atrophy, thepresence of peripapillary atrophy, and the progressionof total atrophy. However, the correlation betweeneyes for baseline acuity, for acuity at two years, andfor two-year change in acuity is significantly smaller,reflecting the importance of the difference in fovealsparing between eyes (Fig. 2) (7).

The two parameters used to describe the pro-gression of GA in the Sunness’ study, namely theenlargement of the atrophic area and visual acuityloss, do not completely gauge the actual impact ofGA on visual function and performance. Maximumreading rate can be significantly affected by encroach-ment of GA on the fovea, even while there may still belittle change in visual acuity (5). Some patients may beable to read single letters on acuity charts but areunable to read words because of the size of thepreserved foveal island (5). The median maximumreading rate decreased from 110 words per minute(wpm) to 51 wpm over a two-year period in patientswith visual acuity better than 20/50, where the normalmedian rate for the reading test used in elderly peoplewithout advanced AMD is 130 wpm. Eighty-threepercent of eyes that lost three or more lines of visualacuity had maximum reading rates less than 50 wpmat two-year follow-up. However, even in the group


that maintained good acuity at two years, one-thirdhad maximum reading rates below 50 wpm (5). Foreyes with visual acuity between 20/80 and 20/200,when the fovea is already involved at baseline, there isevidence to suggest that the maximum reading rate isinversely related to the size of the total atrophic area(41). This may mean that an intervention that couldslow the rate of enlargement of atrophy could have asignificant impact on preserving visual function, evenin the presence of a central scotoma.

DEVELOPMENT OF NEW GA

The relatively high prevalence of bilaterality of GA,reported to be anywhere from 48% to 65% (26) in theliterature, would suggest that patients with GA in oneeye and only drusen or pigmentary change inthe fellow eye are at significant risk for developingGA in the fellow eye. In the Beaver Dam Eye Study,12 patients had GA in one eye at baseline. After fiveyears of follow-up, three of these patients (25%) haddeveloped GA in the fellow eye (25). Patients with GAin only one eye were found to be 2.8 times more likelyto develop advanced AMD in the fellow eye than werepatients with only early changes from age-relatedmaculopathy in either eye at baseline. This is incontrast to patients with neovascular AMD in onlyone eye where the relative risk of developingadvanced AMD in the fellow eye, 1.1, was not signi-ficantly different from the rate at which advancedAMD developed in the fellow eye of those patientswith only early changes from age-relatedmaculopathyat baseline (25). In Sunness’ progression study of GA,two of nine patients (22%) with GA in one eye and

only drusen or pigmentary changes in the fellow eyedeveloped new GA in the fellow eye during the two-year follow-up period (7).

There is limited information available on the rateof development of GA in the eyes of patients who haveonly drusen and pigmentary alteration bilaterally atbaseline. In the Beaver Dam study population, therewas a five-year incidence of new GA of 0.3% (25). Eyeswith only drusen less than 125 mm in linear dimensionat baseline were not observed to go on to develop GA.Of the eyes with drusen between 125 and 250 mm atbaseline, 1% were described as developing GA.In comparison, 8% of eyes with drusen 250 mm orlarger in linear dimension developed GA over a five-year period. Similarly, only those eyes with greaterthan 0.2 MPS disc areas of drusen had a tendencytoward developing GA. All eyes that developed GAhad pigmentary abnormalities at baseline as well (25).In addition to drusen size, there may be some corre-lation between the type of drusen present in eyes withearly age-related maculopathy and the eventualdevelopment of GA. Both calcific drusen (42) andclusters of small, hard drusen have often beenobserved to be present in eyes with GA. Finally,other potential risk factors that have been identifiedin the development of GA include delayed choroidalfilling on fluorescein angiography (43,44) and dimin-ished foveal dark-adapted sensitivity (45).

DEVELOPMENT OF CNV IN EYES WITH GA

Population-based studies have confirmed that theincidence of CNV in an eye with GA depends, in part,

(A) (B)

Figure 2 Bilateral geographic atrophy. (A) This eye had 20/30 visual acuity, and the patient was able toread 80 wpm, using the spared central area. (B) The fellow eye did not have a useable spared region and

had 20/400 visual acuity.


upon the status of the fellow eye. In patients with GAand no CNV in one eye, and CNV in the fellow eye, theeye affected with only GA follows a course that isessentially identical to that of patients with bilateralGAwith respect to foveal preservation, rates of acuityloss, and rates of enlargement of atrophy, so long as itdoes not develop CNV (7). However, when the inci-dence of developing CNV in these eyes with baselineGA is assessed, it is found to be significantly higher ifthe fellow eye has CNV at baseline as opposed to GA.Of the patients enrolled in the extrafoveal MPS withCNV in the study eye, 11 were found to have only GAin the non-study eye at baseline. During the next fiveyears of follow-up, 45% of these eyes went on todevelop CNV (46). More recent findings from theMPS Group’s juxtafoveal and subfoveal CNV trialssupport this incidence. Forty-nine percent of patientswith CNV in the study eye and only GA in the felloweye at baseline went on to develop CNVover the five-year follow-up period (47). A prospective study bySunness et al. in which 31 patients had GA and noCNV in the study eye and CNV in the fellow eyereported a two-year rate of 18% and a four-year rateof 34% for developing CNV in the GA study eye (11).This is in contrast to the results reported by Sunness etal. for patients with bilateral GA at baseline, who had atwo-year rate of developing CNV in one eye of 2% anda four-year rate of 11%. Also, none of the patients withGA in one eye and drusen in the fellow eye developedCNV over the two-year period (11). These data alldemonstrate that there is a higher incidence of CNVin eyes with GA at baseline that have fellow eyeswith CNV.

When CNV does develop in an eye with GA,it seems to have a propensity for areas of preservedretina surrounding the GA or in spared foveal regions.In a study by Schatz and McDonald, 8 of 10 patientswho developed CNV in eyes that had only GApreviously at baseline, developed the CNV at theedge of the atrophy. In the two cases, where the CNVdeveloped over the atrophy, fluorescein angiographywas able to demonstrate evidence of intact chorio-capillaris in those areas (4). Sunness et al. observedthe development of CNVover areas of GA only whenthere were areas of sparing within the atrophy. Other-wise, patients developed CNV in areas that wereadjacent to atrophy (11). Some histologic work like-wise suggests that CNV does not develop where thechoriocapillaris is absent (15).

CNV that is newly developing in eyes withbaseline GA may be difficult to detect by both clinicalexamination and fluorescein angiography. In theabsence of subretinal hemorrhage, it may be difficultto detect subretinal fluid that is shallow and overlyingan area of atrophy. On fluorescein angiography, thehyperfluorescence already present from transmission

defects and staining due to the GA may obscure anynew hyperfluorescence that is secondary to CNV.Because GA does not generally cause an abrupt lossin vision, a patient who presents with subjective andobjective evidence of significant changes in baselinevisual function should undergo evaluation for thepresence of CNV (13). Although GA itself has beenassociated with subretinal hemorrhages withoutevidence of CNV (11,12), the presence of hemorrhageshould certainly prompt further evaluation to detectnewly developing CNV. In some patients, the CNVmay spontaneously involute and have an appearanceidentical to that of GA or may leave small areas offibrosis as remnants of earlier CNV (11).

IMPAIRMENT OF VISUAL FUNCTIONIN EYES WITH GA

Visual acuity alone is an inadequate marker of visualfunction in patients with GA. In addition to centraland paracentral scotoma, eyes with GA have othervisual function abnormalities that may be secondary tochanges in the function of retina that is not yet atrophic(5). Eyes with GA have marked loss of function indim environments and benefit greatly from increasedlighting (5). Aside from delayed and decreased darkadaptation for both rods and cones (5,48–50), eyes withatrophic AMD may also be compromised by reducedcontrast sensitivity (5,51,52). Therefore, despite goodvisual acuity, the patient’s ability to read may besignificantly impaired by a combination of factors.

Central and Paracentral ScotomasMany patients with GA have difficulty in readingbecause of an inability to see a full enough centralfield. Even in the presence of good visual acuity,scotomas near the fovea and involving the fovealcenter compromise visual performance. Patients maycomplain that they can read small newsprint but notlarger news headlines. On clinical examination, it maybe apparent that the foveal center remains intact butwith only a tiny preserved foveal island, which cannotaccommodate the larger headline letters. For thisreason, it is important to take into account that suchpatients may be able to read smaller letters on aneye chart even if they are unable to read the 20/400letter (5).

The impact that GA has on a patient’s lifestyle isnot limited solely to the ability to read. Patients withGA may also describe having great difficulty inrecognizing faces stemming from their inability toassimilate all the features simultaneously (53). Somefind themselves assuming a more reclusive lifestyleafter having repeatedly encountered friends andfamily that they fail to recognize and to greet. More-over, the same small central islands of preserved retina


that impair visual function in the first place alsocomplicate low-vision treatment in these patients. Bymagnifying the object of interest, these low-visiondevices can result in even fewer characters or featuresbeing seen by the patient within the spared area.

Conventional visual field measurement is unreli-able when an eye lacks steady, central fixation, andcan result in plotting scotomas in the wrong locationand of the wrong size (1). The SLO provides directand real-time viewing of stimuli on the retina andpermits the precise correlation of visual function withretinal location. SLO macular perimetry has demon-strated that areas of GA are indeed associated withdense scotomas with surrounding retinal sensitivitythat may be near normal (1). The fixation behaviorsadopted by patients and observed during SLO evalu-ation may explain the inherent variation in visualacuity in eyes with central scotomas from GA.

In order for a patient with scotomas that involvethe foveal center to realize his visual potential, hehas to place the object of regard on functioningretina by adopting an extrafoveal location for fixation,referred to as a preferred retinal locus (PRL). Sunnesset al. found that in a study of eyes with central GAand visual acuities ranging from 20/80 to 20/200, allpatients who were able to adopt an extrafoveallocation for fixation placed their PRL immediatelyadjacent to the area of atrophy. Most patients fixatedwith the scotomas to their right or above fixation invisual field space (41). In another study of GA patientsby Sunness et al., patients reported improvement inthe acuity of their worse-seeing eye when their better-seeing eye worsened somewhat. At baseline, it wasnoted that these patients had not developed eccentricPRLs in the worse-seeing eye so that they placed theobject of regard into their scotoma where it could notbe seen. Over three years of follow-up, these patients,with visual acuities ranging from 20/80 to 20/500,did demonstrate a spontaneous mean improvement of3.2 lines in visual acuity in the worse-seeing eye. Thisimprovement in the worse-seeing eye was concomi-tant with the deterioration of vision in the previouslybetter-seeing eye. At follow-up with SLO macularperimetry, the patients were observed to haveadopted eccentric PRLs, which appeared to accountfor the improvement in the vision of the previouslyworse-seeing eye (54).

Awareness of the presence and location ofscotomas can aid in the effective utilization of theremaining functional retina, lessening the searchingeye movements some patients make when they haveno strategy for moving the object of regard away fromthe scotomas. For example, having the patient fixatesuperior to the area of atrophy on the retina, that is,placing the scotoma above fixation, is a good strategybecause it moves the blind spot out of the most

important part of the visual field. Similarly, fixatingwith the scotoma to the right, that is to the left of theatrophy in a fundus photograph, allows the patient toanchor himself at the beginning of a line while reading(41,55).

With the aid of a fundus photograph, a physiciancanhelp facilitate thepatient’s development of aPRL.Afundus photograph has the same left-to-right orien-tation as visual field space since it has already beenreversed by being viewed from the photographer’sperspective. Therefore, an area of atrophy to the leftof the fovea, or of fixation, corresponds with a scotomato the left of fixation. The fundus photograph isinverted in superior–inferior orientation relative tovisual field space, such that a patient fixating abovean area of atrophy in a fundus photograph has thescotoma above fixation in visual field space. If thefundus photograph indicates the likely location offixation relative to the scotoma, the physician canthen instruct the patient to look toward the scotomain visual field space. This will have the effect of movingthe scotoma farther out of theway. For example, if thereis a scotoma to the right of fixation, as when a patientneglects the last letters on each line of an eye chart,having the patient look farther to the right should allowthe object of regard to come into view.

Difficulties in Dimly Lit EnvironmentsRegardless of their level of visual acuity, most patientswith GA have difficulties with reading and withperforming other visually related tasks in dimly litenvironments. A review of Sunness’ questionnaireresponse found that at least two-thirds of their patientswith GA who still had good enough visual acuity todrive during the day had stopped driving at night (53).

Visual function testing objectively confirms thepresence of reduced function in dim illumination ineyes with GA, as demonstrated by Sunness et al. in astudy of eyes with GA and visual acuity better than20/50. When a control group of elderly patients withocular findings limited to only drusen or pigmentaryalteration, without advanced AMD, had a 1.5-logunit neutral density filter placed over the study eye,the median worsening in acuity was 2.2 lines on theEarly Treatment Diabetic Retinopathy Study (ETDRS)acuity chart. No eye worsened more than five lines (5).For the study group, there was a median worsening inacuity of 4.6 lines on the ETDRS acuity chart when a1.5-log unit neutral density filter was placed over thestudy eye (5). When foveal dark-adapted sensitivitywas measured by gauging the patient’s ability to see asmall red target light in the dark after dark adaptation,eyes with GA and good visual acuity had a mediansensitivity that was 1.2 log units lower than thesensitivity of the control group of elderly eyes withonly early changes from AMD (5).


There is less worsening of visual acuity in dimillumination for eyes that have lost foveal fixation,suggesting that dark-adapted changes may be asensitive marker for foveal changes even before clini-cally apparent atrophy of the fovea develops fromencroachment of surrounding GA (5). These changesin dark-adapted function may also help to predictwhich patients with high-risk drusen and pigmentaryalteration are more likely to eventually develop GA. Ina small prospective study of eyes with drusen,Sunness et al. found that the eyes with the mostreduced foveal dark-adapted sensitivity were thosemost likely to develop advanced AMD, includingGA (45). Steinmetz et al. observed similar outcomes.Eyes with drusen that had associated delayed chor-oidal filling and dark-adaptation abnormalities weremore likely to develop GAwith time (43,44).

In order to maximize the remaining retinal func-tion in these patients, low-vision management of thesepatients should include an evaluation of lightingneeds and appropriate recommendation for thenecessary degree of lighting for reading and othertasks. For example, a GA patient may gain anincreased sense of independence with the use of asmall, handheld penlight to use in a dimly lit restau-rant to read a menu. Sloan demonstrated, in a study ofvisual acuity as a function of chart luminance, thatnormal eyes reach a plateau and then do not improvefurther in visual acuity beyond a certain thresholdluminance. Though GA was not specifically assessed,she found that eyes with AMD in general continued toimprove in acuity with increased luminance for thevalues tested (56). Eyes with GA and some preser-vation of central vision likely follow a similar pattern.

Other Visual Function AbnormalitiesSeveral other abnormalities in visual function mayoccur in eyes affected with GA. Contrast sensitivityhas been found to be reduced in eyes with GA andvisual acuity better than 20/50 compared to eyes ofelderly patients with only drusen and pigmentaryalteration. Specifically, contrast sensitivity is reducedat low spatial frequencies, and is even more markedlyreduced at higher spatial frequencies (5). Despite thepresence of good acuity from preserved foveal islandsin eyes affected with GA, the reading rate may bedramatically decreased secondary to paracentralscotomas. In Sunness et al.’s study of visual functionin eyes with GA and visual acuity better than 20/50,50% of eyes had maximum reading rates less than100 wpm while 17% had maximum reading rates lessthan 50 wpm. In a comparison group of eyes with onlythe earliest manifestations of AMD, the medianmaximum reading rate was found to be 130 wpm,with no eye having a maximum reading rate less

than 100 wpm (5). For this reason, visual acuity aloneis an inadequate measure of a patient’s ability to read.

Patients with small, functional foveal islandsmay have to find an acceptable compromise betweenusing their central fixation and their eccentric PRL tooptimize their visual capacity. While the small fovealregion has good acuity, it by definition has a limitedvisual field extent. Moreover, before the foveal centeris frankly atrophic, it may still be affected by reducedretinal sensitivity, reduced contrast sensitivity, and asubstantial worsening of function in dim illumination.An eccentric, preferred, retinal locus for fixationpositioned outside the area of GA will inherentlyhave a lower visual acuity but may be able to offer alarger area of functional retina less affected by dimillumination and reduced contrast sensitivity. Patientsmay therefore find themselves switching back andforth from foveal to eccentric fixation dependingupon the visual tasks at hand, illumination conditions,and other factors (5,13,57,58).

The combination of variables that can ultimatelyaffect a GA patient’s ability to perform visually relatedtasks can make it difficult to prescribe low-visionmagnification devices that can make the object ofregard too large to be accommodated by the intactcentral region. Evaluation of low-vision requirementsshould always keep these variables in mind. Goodillumination is essential in almost all visuallyrelated tasks.

The Motor Vehicle Administration, along with anumber of researchers, is currently attempting todevelop better ways to evaluate the driving abilityof patients with GA and compromised visual function.Patients with GA and good acuity are often able topass the visual acuity test required for their driver’slicense renewal and continue to drive. Thosewithmorereduced acuity may still be able to secure restrictedlicenses. Most patients with GA tend to limit theirdriving only to areas that they are intimately familiarwith and during daylight hours. One study of AMDpatients that assessed their driving ability with asimulated video-type driving test found that per-formance was poor compared to age-matchedcontrols without evidence of macular degeneration.However, it was observed that these patients hadvery few accidents as they tended to limit theirdriving (59). Specialized driver-training programs forlow-vision patients are becoming increasingly avail-able in an attempt to assess the ability of patients withGA to drive and to aid them in improving theirdriving ability.

CONDITIONS RESEMBLING GA

There are other conditions of the eye that in one stageor other of their progression can resemble GA. Some of


these are other manifestations of AMD and simplyexist on a different part of the continuum from GA.Other conditions would be classified as retinal ormacular degenerations that are not age related.

CNV that has spontaneously involuted canleave an atrophic scar that resembles GA (60,61).Some scars may have small fibrotic areas that areremnants of previous CNV. Other scars appear iden-tical to GA. In such cases, fluorescein angiographymay aid in distinguishing CNV from GA.

Old laser photocoagulation scars may alsoresemble GA. The history, however, should dis-tinguish the two. Again, fluorescein angiographyshould demonstrate areas of hypofluorescence thatcorrespond to the laser scars. Areas of GA generallyshow hyperfluorescence on angiography.

An RPE tear may clinically resemble GA. Onfluorescein angiography, however, the straight-lineborder of hyperfluorescence should be characteristicof a rip. It is unclear whether RPE tears developatrophy in adjacent areas with time (62).

Eyes with pattern dystrophy and vitelliformchanges may develop atrophic changes that progressin a fashion similar to AMD-related GA. Thesepatients may have areas of macular GA, and somemay be accompanied by pigmentary alterationscharacteristic of these conditions and occasionallyby reduced electrooculograms. However, other casesmay be difficult to distinguish from age-related GA.The atrophy spreads in a parafoveal pattern withearly foveal sparing, often resulting in a similardegree of visual compromise (63).

Central areolar choroidal dystrophy is anotherdegenerative, retinal condition that spares the foveaearly in the course of disease. This hereditary conditionis generally autosomal dominant and causes areas ofatrophy in the macular region to develop since earlyadulthood.Unlike age-relatedGA, these lesions tend tohave early atrophy of the choroidal circulation andchoriocapillaris so that involved areas on fluoresceinangiography appear as hypofluorescent (64).

Disorders that cause central, atrophic lesions, andbull’s-eye maculopathies may also mimic age-relatedGA. Stargardt’s disease, cone dystrophy, North Caro-lina macular dystrophy, benign concentric annularmacular dystrophy, and chloroquine, and other toxicmaculopathies, may all have manifestations similarto GA from AMD. The history, including age of onsetof symptoms and prior medication use, may be helpfulin differentiating some of these disorders from GA.Associated clinical findings, such as sensitivity tolight and significant electroretinographic or colorvision abnormalities in cone dystrophy or pisciformflecks and an angiographically dark choroid inStargardt’s disease may also facilitate differentiating

the GA that results from these other entities fromage-related GA.

AUTOFLUORESCENCE IMAGING AND GA

Lipofuscin (LF) accumulates in the lysosomalcompartments of postmitotic eukaryotic cells withage and may represent a biomarker for cellularageing (65). It is assumed that in retinal pigmentepithelial cells, LF is a byproduct of the phagocytosisof membranous discs shed from outer segments ofphotoreceptors and its accumulation may play apathogenic role in retinal disease; thus, its detectionin vivo may help to elucidate its pathophysiologicalsignificance (65). Cross-sectional studies on eyeswith early and late AMD have shown that increasedaccumulations of autofluorescent material are presentmore frequently in eyes with GA than in eyes withdrusen alone or with CNV. Furthermore, severalfindings suggest that this autofluorescence, whichoriginates at the level of the RPE, is derived fromLF accumulations (65). While the retina itself hasa normal background level of autofluorescence, anumber of groups have reported that there is a distinctloss of autofluorescence in areas of GA, which may beconsistent with the absence of RPE cells (65).

Autofluorescence imaging takes advantage ofthe fluorescent properties of LF within RPE cells.In fundus autofluorescence imaging, an argon bluelaser (488 nm) is used for excitation, and a barrierfilter prevents blue light from returning to the imagedetector, such that only light emitted by autofluores-cence above 500 nm is detected. A confocal SLO canthen be used to examine the fundus autofluorescence(66,67). Several studies have described the develop-ment of new and enlarging areas of preexisting atrophyassociated with areas of abnormally high in vivo auto-fluorescence in eyes with GA secondary to AMD.

Fundus autofluorescence was examined with aconfocal SLO by Holz et al. in a study of 57 eyes in 38patients with unifocal or multifocal GA secondary toAMD. The findings were compared to 43 eyes withatrophy secondary to non-AMD etiologies, such asjuvenile macular dystrophy. The investigators foundthat increased autofluorescence outside GA wasobserved in 47 (82.5%) of 57 eyes with GA secondaryto AMD compared to only 4 (9.3%) of 43 eyes withGA secondary to other causes (65). In addition,various patterns of autofluorescence in the presenceof GA associated with AMD were noted. In 36 eyes,76.6%, a continuous band at the margin of the GAwith variable peripheral extension was observed.A diffusely increased autofluorescence that involvedthe entire posterior pole was noted in six eyes (12.8%)with GA secondary to AMD. Small focal spots ofincreased autofluorescence in the junctional zone of


three eyes (6.4%) with GA associated with AMD werenoted. The study also reported that of 19 patientswith bilateral GA, 17 (89.5%) had an identical auto-fluorescent pattern in each eye (65). From their cross-sectional examinations of fundus autofluorescence,the investigators also speculated that it is possiblethat one pattern of autofluorescence may evolve intoanother pattern over time, i.e., focal spots of increasedautofluorescence could coalesce to become a continu-ous band of autofluorescence, thus representing aspectrum of the severity of functional compromiseat the level of the RPE, which might serve as a usefulprognostic indicator of disease progression (65).

Holz and colleagues also performed a small pilotstudy in which the intensity of fundus autofluores-cence as well as the occurrence of new GA and thespread of existing GAwere recorded in three patientswith AMD over a period of three years using aconfocal SLO. Preliminary findings suggested thatareas of increased autofluorescence preceded thedevelopment or enlargement of GA in eyes withAMD (67). GA did not develop in areas with normalbackground fluorescence.

While autofluorescence imaging may have thepotential to yield important clinical informationregarding the pathogenesis and progression ofdisease, various limitations in its application have tobe considered.Media opacity, such as a yellowing lens,may absorb the blue excitation light and preventadequate imaging. In addition, it is important to beaware that absolute levels of autofluorescence cannotbe assessed, only relative levels of autofluorescence.However, as more is learned about the correlationof autofluorescence pattern with progression of GA,autofluorescent imaging will become an importanttool for assessing risk. Potential therapies maypossibly be assessed as to whether they decrease theamount of increased autofluorescence, which mayserve as an earlier marker of treatment benefit thangauging progression of GA clinically by ophthal-moscopy. Autofluorescence imaging is also anexcellent way for imaging areas of GA themselves,which lack autofluorescence and appear black (68,69).This appearance allows for a simpler way of definingthe borders of an area of GA, and is more amenableto automation. However, there must be clinical corre-lation, because drusen and other types of pigmentarychange may also show a lack of autofluorescence (69).

POTENTIAL TREATMENT FOR GA

Because GA can be clinically visualized in manypatients before the development of moderate orsevere vision loss, unlike CNV, there is greaterpotential for medical intervention to preserve visualfunction. While there is currently no definitive

treatment to reverse the progression of GA, there istherapy for retarding disease progression.

The AREDS, a double-masked clinical trial,enrolled 3640 patients, ages 55 to 80 years, who hadclinical evidence of extensive small drusen, intermedi-ate drusen, large drusen, noncentral GA, or advancedAMD in one eye and randomly assigned them toreceive daily oral supplements containing eitherantioxidants (vitamin C, 500 mg; vitamin E, 400 IU;and beta-carotene, 15 mg), zinc (80 mg, as zinc oxideand copper, 2 mg as cupric oxide), antioxidants pluszinc, or placebo. Average follow-up was 6.3 years.Investigators observed that treatment with zinc aloneor in combination with antioxidants significantlyreduced the risk of progression to advanced AMD inindividuals with intermediate drusen, large drusen,noncentral GA, and advanced AMD (70). The riskreduction for those taking antioxidants plus zinc wasmost favorable at 25% (70). However, the data for GAalone were not statistically significant, perhapsbecause of inadequate numbers. In addition, the twomeasures related to GA that are reported had a trendtoward showing opposite findings. The risk of devel-oping central GA had a trend toward being lowered byantioxidants, zinc, or the combination. However, thedevelopment of any new GA (360 mm or greater indiameter) in an eye without GA at baseline showed atrend toward a reduced risk with antioxidants but anincreased risk for zinc and the combination (70). Thesewere small effects, and a larger clinical trial is necess-ary to be able to detect statistically and clinicallysignificant effects on GA. Many retina specialists dorecommend the AREDS supplements, because of theoverall lessening of risk of advanced AMD.

One approach that may hold promise for patientswith GA in the future is to somehow supply trophicfactors to delay or prevent RPE and photoreceptor celldeath. A study is about to begin at National Instituteof Health (NIH), using encapsulated cell technology,to determine if ciliary neurotrophic growth factor,made by encapsulated genetically engineered RPEcells placed intraocularly, can improve photoreceptorhealth. Studies have been performed to transplantfetal RPE to try to provide presumed humoral trophicfactors, but attempts at transplantation, as reportedfor example in one case by Weisz et al., have beencomplicated by rejection of the transplanted cells (71).

Currently, patients with GA and visual compro-mise can be offered rehabilitation in terms of low-vision intervention and new strategies for maximizingtheir utilization of remaining, healthy retina throughthe development of preferred retinal loci (PRLs). Morecases of GA will continue to be seen in ensuing yearsas its prevalence grows in an ageing population. It ishoped that as more is learned about GA in the future,we can offer more to the patient with respect to the


management and eventually prevention of this formof AMD.

SUMMARY POINTS

& The prevalence of GA increases with age, beinghalf as common as CNV at age 75, and morecommon than CNV in older age groups.

& GA continues to enlarge over time with a medianrate of enlargement over a two-year period of 1.8MPS disc areas.

& Scotomas from GA, the advanced form of non-neovascular AMD, involve the parafoveal andperifoveal retina early in the course of thedisease, sparing the foveal center until late inthe course of the disease.

& These parafoveal and perifoveal scotomas compro-mise the ability to read and to recognize faces, oftendespite the retention of good visual acuity,accounting for a large percentage of moderatevisual loss in those affected.

& Hemorrhage may occur in eyes with GA in theabsence of CNV. Small areas of CNV that can beassociated with hemorrhage may be transient,becoming clinically inapparent, or appearing asincreased atrophy, a few months later.

& There is a higher incidence of CNV in eyes with GAat baseline that have fellow eyes with CNV.

& GA is bilateral in more than half of the people withthis condition. The size and rate of progression ofatrophy are highly correlated between the two eyesof patients with bilateral GA, but the acuities maydiffer due to central sparing.

& Among eyes with GAwith visual acuity better than20/50, there is a 40% rate of three-line visual loss attwo years.

& Maximum reading rate can be significantly affectedby encroachment of GA on the fovea, even whilethere may still be little change in visual acuity.

& Eyes with GA have marked loss of vision in dimenvironments and benefit greatly from increasedlighting.

& Oral supplementation with antioxidants and zinc,per the AREDS protocol, may slow the progressionof GA and delay loss of vision, although theAREDS study did not have enough power toshow a significant effect on GA itself.

& The development of a PRL can aid in the effectiveutilization of the remaining functional retina.

ACKNOWLEDGMENTS

Supported in part by NIH R01 EY 08552 (JSS), NIH R03EY14148 (JSS), the James S. Adams RPB SpecialScholar Award (JSS), and the RPB Physician ScientistAward (JSS).

REFERENCES

1. Sunness JS, Schuchard R, Shen N, Rubin GS, Dagnelie G,Haselwood DM. Landmark-driven fundus perimetry usingthe scanning laser ophthalmoscope (SLO). InvestOphthalmol Vis Sci 1995; 36:1863–74.

2. Sarks JP, Sarks SH, Killingsworth MC. Evolution ofgeographic atrophy of the retinal pigment epithelium. Eye1988; 2:552–77.

3. Maguire P, Vine AP. Geographic atrophy of the retinalpigment epithelium. Am J Ophthalmol 1986; 102:621–5.

4. Schatz H, McDonald HR. Atrophic macular degeneration:rate of spread of geographic atrophy and visual loss.Ophthalmology 1989; 96:1541–51.

5. Sunness JS, Rubin GS, Applegate CA, et al. Visual functionabnormalities and prognosis in eyes with age-relatedgeographic atrophy of the macula and good acuity.Ophthalmology 1997; 104:1677–91.

6. Ferris FL, III, Fine SL, Hyman L. Age-related maculardegeneration and blindness due to neovascular maculo-pathy. Arch Ophthalmol 1984; 102:1640–2.

7. Sunness JS, Gonzalez-Baron J, Applegate CA, et al. Enlarge-ment of atrophy and visual acuity loss in the geographicatrophy form of age-related macular degeneration.Ophthalmology 1999; 106:1768–79.

8. Klein R, Klein BEK, Franke T. The relationship of cardio-vascular disease and its risk factors to age-relatedmaculopathy: the Beaver Dam Eye Study. Ophthalmology1993; 100:406–14.

9. Vingerling JR, Dielemans I, Hofman A, et al. The prevalenceof age-related maculopathy in the Rotterdam Study.Ophthalmology 1995; 102:205–10.

10. Quillen D, Blankenship G, Gardner T. Aged eyes: ocularfindings in individuals 90 years of age and older. InvestOphthalmol Vis Sci 1996; 47:S111.

11. Sunness JS, Gonzalez-Baron J, Bressler NM, Hawkins B,Applegate CA. The development of choroidal neovascular-ization in eyes with the geographic atrophy form ofage-related macular degeneration. Ophthalmology 1999;106:910–9.

12. Nasrallah F, Jalkh AE, Trempe CL, McMeel JW,Schepens CL. Subretinal hemorrhage in atrophic age-related macular degeneration. Am J Ophthalmol 1988;107:38–41.

13. Sunness JS, Bressler NM, Maguire MG. Scanning laserophthalmoscopic analysis of the pattern of visual loss inage-related geographic atrophy of the macula. AmJ Ophthalmol 1995; 119:143–51.

14. Sunness JS, Bressler NM, Tian Y, Alexander J,Applegate CA. Measuring geographic atrophy in advancedage-related macular degeneration. Invest Ophthalmol VisSci 1999; 40:1761–9.

15. Green WR, Key SN, III. Senile macular degeneration: ahistopathologic study. Trans Am Ophthalmol Soc 1977;75:180–254.

16. Korte GE, Reppucci V, Henkind P. Retinal pigment epi-thelium destruction causes choriocapillary atrophy. InvestOphthalmol Vis Sci 1984; 25:1135–45.

17. Leonard DS, Zhang XG, Panozzo G, Sugino IK, Zarbin MA.Clinicopathologic correlation of localized retinal pigmentepithelial debridement. Invest Ophthalmol Vis Sci 1997;38:1094–109.

18. Green WR, Enger C. Age-related macular degenerationhistopathologic studies: the 1992 Lorenz E. Zimmermanlecture. Ophthalmology 1993; 100:1519–35.


19. Sarks SH. Changes in the region of the choriocapillaris inageing and degeneration. XXIII Concilium Ophthalmolo-gicum, Kyoto. Amsterdam-Oxford: Excerpta Medica, 1979.

20. ElmanMJ, Fine SL, Murphy RP, Patz A, Auer C. The naturalhistory of serous retinal pigment epithelium detachment inpatients with age-related macular degeneration. Ophthal-mology 1986; 93:224–30.

21. Friedman E, Krupsky S, Lane AM, et al. Ocular blood flowvelocity in age-related macular degeneration. Ophthal-mology 1995; 102:640–6.

22. Grunwald JE, Hariprasad SM, DuPont J, et al. Foveolarchoroidal blood flow in age-related macular degeneration.Invest Ophthalmol Vis Sci 1998; 39:385–90.

23. Hirvela H, Luukinen H, Laara E, Sc L, Laatikainen L. Riskfactors of age-related maculopathy in a population 70 yearsof age or older. Ophthalmology 1996; 103:871–7.

24. Friedman DS, Katz J, Bressler NM, Rahmani B, Tielsch JM.Racial differences in the prevalence of age-related maculardegeneration: the Baltimore Eye Survey. Ophthalmology1999; 106:1049–55.

25. Klein R, Klein BE, Jensen SC, Meuer SM. The five-yearincidence and progression of age-related maculopathy: theBeaver Dam Eye Study. Ophthalmology 1997; 104:7–21.

26. Porter JW, Thallemer JM. Geographic atrophy of the retinalpigment epithelium: diagnosis and vision rehabilitation.J Am Opt Assoc 1981; 52:503–8.

27. Ritter LL, Klein R, Klein BE, Mares-Perlman JA, Jensen SC.Alcohol use and age-related maculopathy in the BeaverDam Eye Study. Am J Ophthalmol 1995; 120:190–6.

28. Smith W, Mitchell P, Leeder SR. Smoking and age-relatedmaculopathy: the Blue Mountain Eye Study. ArchOphthalmol 1996; 114:1518–23.

29. Age-Related Eye Disease Study Research Group. Riskfactors associated with age-related macular degeneration:a case-control study in the Age-Related Eye Disease Study:AREDS report no. 3. Ophthalmology 2000; 107:2224–32.

30. Piguet B,HeonE,Munier FL, et al. Full characterization of themaculopathy associated with an Arg-12-Trp mutation in theRDS/peripherin gene. Ophthalmic Genet 1996; 17:175–86.

31. Allikmets R, Shroyer NF, Singh N, et al. Mutation of theStargardt disease gene (ABCR) in age-related maculardegeneration. Science 1997; 277:1805–7.

32. Pennisi E. Human genetics: gene found for the fadingeyesight of old age. Science 1997; 277:1765–6.

33. Allikmets and International ABCR screening Consortium.Further evidence for an association of the ABCR alleles withage-related macular degeneration. Am J Hum Genet 2000;67:487–91.

34. Yates JR, Moore AT. Genetic susceptibility to age-relatedmacular degeneration. J Med Genet 2000; 37(2):83–7.

35. Sepp T, Khan JC, Thurlby DA, et al. Complement factor Hvariant Y402H is a major risk determinant for geographicatrophy and choroidal neovascularization in smokers andnonsmokers. Invest Ophthalmol Vis Sci 2006; 47:536–40.

36. Gass JDM. Drusen and disciform macular detachment anddegeneration. Arch Ophthalmol 1973; 90:206–17.

37. Peli E, Lahav M. Drusen measurements from fundusphotographs using computer image analysis. Ophthal-mology 1986; 93:1575–80.

38. Braunstein RA, Gass JDM. Serous detachments of theretinal pigment epithelium in patients with senile maculardisease. Am J Ophthalmol 1979; 88:652–60.

39. Casswell AG, Kohen D, Bird AC. Retinal pigment epithelialdetachments in the elderly: classification and outcome. Br JOphthalmol 1985; 69:397–403.

40. Meredith TA, Braley RE, Aaberg TM. Natural history ofserous detachments of the retinal pigment epithelium. AmJ Ophthalmol 1979; 88:643–51.

41. Sunness JS, Applegate CA, Haselwood D, Rubin GS.Fixation patterns and reading rates in eyes with centralscotomas from advanced atrophic age-related maculardegeneration and Stargardt disease. Ophthalmology 1996;103:1458–66.

42. Sunness JS, Bressler NM, Applegate CA. Ophthalmoscopicfeatures associated with geographic atrophy from age-related macular degeneration. Invest Ophthalmol Vis Sci1999; 40:S314 (Abstract).

43. Steinmetz RL, Walter D, Fitzke FW, Bird AC. Prolongeddark adaptation in patients with age-related maculardegeneration. Invest Ophthalmol Vis Sci 1991; 32:S711.

44. Steinmetz RL, Haimovici R, Jubb C, Fitzke FW, Bird AC.Symptomatic abnormalities of dark adaptation in patientswith age-related Bruch’s membrane change. Br JOphthalmol 1993; 77:549–54.

45. Sunness JS, Massof RW, Johnson MA, Bressler NM,Bressler SB, Fine SL. Diminished foveal sensitivity maypredict the development of advanced age-related maculardegeneration. Ophthalmology 1989; 96:375–81.

46. Macular Photocoagulation Study Group. Five-year follow-up of fellow eyes of patients with age-related maculardegeneration and unilateral extrafoveal choroidal neovas-cularization. Arch Ophthalmol 1993; 111:1189–99.

47. Macular Photocoagulation Study Group. Risk factors forchoroidal neovascularization in the second eye of patientswith juxtafoveal or subfoveal choroidal neovascularizationsecondary to age-related macular degeneration. ArchOphthalmol 1997; 115:741–7.

48. Brown B, Kitchin JL. Dark adaptation and the acuity/lumi-nance response in senile macular degeneration (SMD). AmJ Optom Physiol Opt 1983; 60:645–50.

49. Brown B, Tobin C, Roche N, Wolanowski A. Cone adap-tation in age-relatedmaculopathy. Am J Optom Physiol Opt1986; 63:450–4.

50. Sunness JS, Massof RW, Johnson MA, Finkelstein D,Fine SL. Peripheral retinal function in age-related maculardegeneration. Arch Ophthalmol 1985; 103:811–6.

51. Brown B, Lovie-Kitchin J. Contrast sensitivity in central andparacentral retina in age-related maculopathy. Clin ExpOptom 1987; 70:145–8.

52. Midena E, Degli Angeli C, Blarzino MC, Valenti M,Segato T. Macular function impairment in eyes with earlyage-related macular degeneration. Invest Ophthalmol VisSci 1997; 38:469–77.

53. ApplegateCA,Sunness JS,HaselwoodDM.Visual symptomsassociatedwithgeographic atrophy fromage-relatedmaculardegeneration. Invest Ophthalmol Vis Sci 1996; 37:S112.

54. Sunness JS, Applegate CA, Gonzalez-Baron J. Improvementof visual acuity over time in patients with bilateral geo-graphic atrophy from age-related macular degeneration.Retina 2000; 20:162–9.

55. Guez JE, Le Gargasson JF, Rigaudiere F, O’Regan JK. Is therea systematic location for the pseudo-fovea in patients withcentral scotoma? Vis Res 1993; 9:1271–9.

56. Sloan L. Variation of acuity with luminance in oculardiseases and anomalies. Doc Ophthalmol 1969; 26:384–93.

57. Schuchard RA, Raasch TW. Retinal locus for fixation:pericentral fixation targets. Clin Vis Sci 1992; 7:511–20.

58. Lei H, Schuchard RA. Using two preferred retinal loci fordifferent lighting conditions in patients with centralscotomas. Invest Ophthalmol Vis Sci 1997; 38:1812–8.


59. Szlyk JP, Pizzimenti CE, Fishman GA, et al. A comparisonof driving in older subjects with and without age-relatedmacular degeneration. Arch Ophthalmol 1995; 113:1033–40.

60. Bressler NM, Frost LA, Bressler SB, Murphy RP, Fine SL.Natural course of poorly defined choroidal neovasculariza-tion in macular degeneration. Arch Ophthalmol 1988;106:1537–42.

61. Jalkh AE, Nasrallah FP, Marinelli I, Van de Velde F. Inactivesubretinal neovascularization in age-related maculardegeneration. Ophthalmology 1990; 97:1614–9.

62. Yeo JH, Marcus S, Murphy RP. Retinal pigment epithelialtears: patterns and progression. Ophthalmology 1988;95:8–13.

63. Marmor MF, McNamara JA. Pattern dystrophy of theretinal pigment epithelium and geographic atrophy. AmJ Ophthalmol 1996; 122:382–92.

64. Krill AE, Archer D. Classification of the choroidal atrophies.Am J Ophthalmol 1971; 72:562.

65. Holz FG, Bellmann C, Margaritidis M, Schutt F, Otto TP,Volcker HE. Patterns of increased in vivo autofluorescencein the junctional zone of geographic atrophy of the retinalpigment epithelium associated with age-related maculardegeneration. Graefe’s Arch Clin Exp Ophthalmol 1999;237:145–52.

66. Delori FC, Dorey CK, Staurenghi G, Arend O, Goger DG,Weiter JJ. In vivo fluorescence of the ocular fundus exhibitsretinal pigment epithelium lipofuscin characteristics. InvestOphthalmol Vis Sci 1995; 36:718–29.

67. Holz FG, Bellman C, Staudt S, Schutt F, Volcker HE. Fundusautofluorescence and development of geographic atrophyin age-related macular degeneration. Invest OphthalmolVis Sci 2001; 42:1051–6.

68. Deckert A, Deckert A, Schmitz-Valckenberg S, et al. Auto-mated analysis of digital fundus autofluorescence imagesof geographic atrophy in advanced age-related maculardegeneration using confocal scanning laser ophthal-moscopy (cSLO). BMC Ophthalmol 2005; 5(1):8.

69. Sunness JS, Ziegler MD, Applegate CA. Issues in quantify-ing atrophic macular disease using retinal autofluorescence.Retina 2006 Jul–Aug; 26(6):666–72.

70. Age-Related Eye Disease Study Research Group. Arandomized, placebo-controlled, clinical trial of high-dosesupplementation with vitamins C and E, beta carotene,and zinc for age-related macular degeneration and visionloss: AREDS report no. 8. Arch Ophthalmol 2001;119:1417–36.

71. Weisz JM, deJuan E, Humayun MS, et al. Allogenic fetalretinal pigment epithelial cell transplant in a patient withgeographic atrophy. Retina 1999; 19:540–5.


8

Exudative (Neovascular) Age-RelatedMacular DegenerationJennifer I. LimUniversity of Illinois School of Medicine, Department of Ophthalmology, Eye and Ear Infirmary,


Jerry W. TsongDoheny Eye Institute and Department of Ophthalmology, Keck School of Medicine,


INTRODUCTION

Exudative age-related macular degeneration (AMD)was first described and illustrated in the literature in1875 by Pagenstecher (1). Pagenstecher termed thecondition chorioidioretinitis in regione maculaeluteae. Then in 1905, Oeller first used the name disci-form degeneration (degeneratio maculae luteaedisciformis) (2). Later, Julius and Kuhnt in 1926further elaborated on this condition and establishedit as a disease (3). Further study by clinicians andpathologists over the next several decades resulted inthe understanding that choroidal neovascularization(CNV) was responsible for the manifestations ofexudative AMD. The fact that disciform scars hadassociated CNV was revealed in 1928 by Hollowayand Verhoeff who described eight eyes with disclikedegeneration of the retina (4); histopathology showedCNV. In 1937, Verhoeff and Grossman similarlydemonstrated CNV in their cases of macular degener-ation and emphasized that blood vessels eruptedthrough Bruch’s membrane (5). It was not until 1951that clinicopathologic correlations by Ashton andSorsby demonstrated that CNV with breaks inBruchs membrane results in subretinal fluid (6).Finally, in 1967 Gass implicated CNV as having aprimary role in what was then called “senile disciformmacular degeneration” (7,8). In 1971, Blair and Aabergshowed the clinical and fluorescein angiographiccharacteristics of CNV in these eyes with “senilemacular degeneration” (9). In 1976, Small publisheda clinicopathologic correlation of the evolution of alesion, comprised of CNV with a serous pigmentepithelial detachment (PED), to a disciform scar (10).In 1977, Green and Key (11) studied the histopatho-logic features of 176 eyes from 115 patients withAMD. Their results supported the view that drusenpredispose to development of CNV. Since then,numerous studies have given us ample histopatho-logic data on CNV (see Grossniklaus chapter).

Since the earliest description of AMD, there havebeen numerous refinements in the categorization ofthe types of AMD. In fact, even the term, AMD, is arelatively recent development. Prior to 1990, the term“senile macular degeneration” was used to refer towhat we now know as AMD. More recently, the twomain types of AMD, non-exudative or non-neovas-cular AMD and exudative or neovascular AMD, havebeen referred to colloquially as dry AMD and wetAMD respectively. Although non-exudative AMD istypically associated with less severe degrees of visualdisturbances than exudative AMD and may even haveno associated visual disturbance, eyes with exudativeAMD typically have some visual disturbance.

Exudative or neovascular AMD has the mostserious visual prognosis in terms of visual acuityoutcomes. An overview of the epidemiology, riskfactors associated with development of exudativeAMD, fundus findings found in exudative AMD,diagnostic tools used in evaluating exudative AMDand treatment options for the various types of exuda-tive AMD is presented.

EPIDEMIOLOGY

Exudative (neovascular) AMD, although the lesscommon form of AMD, is the leading cause of newblindness in the older age population in the UnitedStates, accounting for 16% of all new cases of blindnessover the age of 65 years. Indeed, the majority ofpatients with severe visual loss have CNV (12).In fact, 79% of eyes legally blind in the FraminghamStudy and 90% of legally blind eyes in a large casecontrol study had neovascular AMD (13,14). With theaging of the U.S. population, AMD is reachingepidemic proportions. In the United States alone,there are 50,000 new cases of CNV due to AMD eachyear. The number of persons aged 55 years or older is38 million in the United States (U.S. census 2000 data)and is projected to increase to 88 million by 2030.

The prevalence of AMD, in general, and that of the lateforms of AMD increase with advancing age. In a studyby Friedman and colleagues, the prevalence ofadvanced AMD (defined as presence of CNV orfoveal geographic atrophy) is estimated to be 1.75million in the United States. By 2020, the prevalenceof advanced AMD is expected to reach 2.95 million.Currently, 1.22 million have neovascular AMD in atleast one eye; 973,00 have geographic atrophy in atleast one eye; 7.3 million individuals have large drusenin one or both eyes and are at high risk for progressionto advanced AMD (15). There is indeed a strong needfor the identification of risk factors for exudative AMDand for preventive therapies.

Risk FactorsThere are numerous risk factors associated with thedevelopment of CNV in AMD. These risk factorsinclude ocular and non-ocular factors; they arediscussed in great depth in the chapter by Au-Eongand Haller. Of the non-ocular risk factors, it appearsthat the strongest epidemiological associations are age,race, smoking and genetics. Of the ocular risk factors,soft large drusen, retinal pigment epithelium (RPE)pigmentary changes and presence of CNV in thefellow eye are strong associations for developmentof CNV. It is important to remember that these areassociations and do not imply causation (causeand effect).

Non-ocular Risk Factors Associated withExudative (Neovascular) AMDIncreased age is associated with increasing risk ofneovascular AMD. Patients with exudative AMDhave a mean age of 70.5 years versus 56.8 years fornon-exudative AMD (16). Racial differences in theprevalence of exudative AMD (and also early AMD)exist. Gregor and Joffe (17) found that the prevalenceof disciformAMDwas 3.5% of the white South Africanpatients compared to 0.1% of the black South Africanpatients (p!0.001). The Baltimore Eye Survey revealeda prevalence ratio of 8.8 for white to black neovascularAMD (18). Over a nine-year period, the incidenceof AMD in the Barbados Eye Study was 12.6% forearly AMD but only 0.7% for late AMD (19).In National Health and Nutritional ExaminationSurvey-III (NHANES-III), the odds ratio for lateAMD was 0.34 for non-Hispanic blacks compared tonon-Hispanic whites (20). The prevalence of neovas-cular AMD is higher in Caucasians than AfricanAmericans.

The prevalence rates of neovascular AMD inother racial groups have recently been investigated.In the Latino Eye Study (6,357 Latino patients aged 40years of age and older), the prevalence of early AMDincreased from 6.2% in the 40 to 49 years old group to

29.7% in the 80 years or older group but that ofadvanced AMD increased from 0% in the 40 to 49years old group to only 8.5% in the 80 years of age orolder group (21). Similarly, in the Proyecto group, theprevalence of late AMD increased from 0.1% in 50 to59 years old to 4.3% of those 80 years or older (22).These rates of advanced AMD are lower than thosein Caucasians.

In the Multi-ethnic Study of Atherosclerosis(MESA), the prevalence of AMD in four ethniccohorts (whites, blacks, Hispanics, and Chinese) wasdetermined. The prevalence of any AMD was 5.4% inwhites, 2.4% in blacks, 4.2% in Hispanics, and 4.6%in Chinese aged 75 to 84 years of age. The prevalenceof exudative AMD was highest for Chinese in whichthe odds ratio (OR) was 4.3 compared with Caucasians(23). Most of the Chinese in MESA were born outsideof the United States. Further work on Asian AMD isneeded to draw definitive conclusions about thisgroup. Overall, several studies corroborate the racialdifferences in the prevalence of AMD in general andthe neovascular or exudative form.

Family history is a risk factor for the develop-ment of AMD, including neovascular AMD. The BlueMountains Eye Study showed an odds ratio of 4.30 forneovascular AMD in patients with a family history(24). Klaver and colleagues found the lifetime riskestimate of late AMD to be 50% for relatives of patientsversus 12% for relatives of controls (25).

Cigarette smoking has been associated withexudative AMD in most studies, although it was notlinked to AMD in the Framingham Study (13) and theNHANES-III Study (26). The Beaver Dam Eye Study(27) linked smoking to exudative AMD with a relativerisk of 3.29 for current smokers and a relative risk of2.50 for former smokers compared to those who hadnever smoked. In the Blue Mountain Eye Study (28),the odds ratio for exudative AMD was 4.46 for currentsmokers compared with those who never smokedand 1.83 for former smokers compared with thosewho never smoked. In the Pathologies OculairesLiees a l’Age study, the odds ratio for exudativeAMD increased with the number of pack-yearssmoked (29). This higher risk of exudative AMDpersisted until 20 years after cessation of cigarettesmoking. In a case control study by Khan et al. (30)smoking more than 40 pack years of cigarettes wasassociated with an odds ratio of 2.49 (95% CI 1.06 to5.82) for CNV. Stopping smoking was associated withreduced odds and the risk in those who had notsmoked for over 20 years was comparable to non-smokers. Even passive smoking exposure was associ-ated with an increased risk of AMD (OR 1.87; 95% CI1.03–3.40) in the non-smokers.

In an animal model of rats fed a high-fat diet,exposure to cigarette smoke or the smoke-related

126 LIM AND TSONG

redox molecule, hydroquinone, resulted in the forma-tion of sub RPE deposits, thickening of Bruch’smembrane, and accumulation of deposits withinBruch’s membrane. This animal model shows thatcigarette smoking results in molecules that can causeoxidative injury to the choriocapillaris and RPE, andmay explain the association between cigarettesmoking and AMD (31).

The association between sunlight exposure andlate AMD is not clear. The Chesapeake Bay WatermenStudy found an association between late AMD andsunlight (32) as did the Beaver Dam Eye Study (33).Yet, the Eye Disease Case-Control Study (34) and theAustralian case-control study on sun exposure andAMD (35) did not show this same association. Sincethe use of sunglasses (ultraviolet blocking) is relativelyinexpensive and also protective against cataractformation, it is reasonable to recommend sunglassprotection for older patients.

There have been reports of progression of earlyto late AMD following cataract surgery. The BeaverDam Eye Study showed an odds ratio of 2.80 forprogression of AMD to late AMD after cataractsurgery (and after controlling for age) (36). Pollackand colleagues also noted progression to exudativeAMD occurred in 19.1% of eyes operated on forcataracts versus 4.3% of the fellow eye (37,38).Further investigation is needed in this area.

Ocular Risk Factors Associated withExudative (Neovascular) AMDThe risk of CNV developing in a patient’s eye has beenlinked to the presence of soft drusen, pigmentarychanges, status of the fellow eye and hypertension.Lanchoney and associates calculated the risk of CNVin patients with bilateral soft drusen to range from8.6% to 15.9%within 10 years, depending upon the ageand sex of the patient (39). These projections werebased upon natural history studies of Smiddy and Fine(16) and Holz (40).

The Macular Photocoagulation Study (MPS)group has determined the ocular risk factors fordevelopment of CNV in the fellow eye (when theopposite eye already has CNV) to include the presenceof five or more drusen, focal hyperpigmentation,one or more large drusen (O63 mm) and systemichypertension (41). The five year incidence rate fordevelopment of CNV ranged from 7% if none ofthese risk factors was present to 87% if all four riskfactors were present. (This was based upon follow-upof patients with juxtafoveal CNV).

The Age-Related Eye Disease Study (AREDS) hasalso yielded predicative rates of CNV for a patientbased upon the ocular findings of both eyes. TheAREDS group created a simplified scale for deter-mining the risk of development of neovascular

AMD. One risk factor is assigned for the presence oflarge soft drusen and one risk factor for any pigmentabnormality in each eye. Intermediate drusen alone inboth eyes is counted as one risk factor. AdvancedAMD counts as two points for an affected eye. Thevalues are then summed for both eyes. The five yearrisk of developing advanced AMD was 0.5% for zerorisk factors, 3% for one risk factor, 12% for two riskfactors, 25% for three risk factors and 50% for fourrisk factors (42). The AREDS group has recentlydetermined the 10 years risk of developing advancedAMD is less than 1% for no risk factor, 10% for one riskfactor, 30% for two risk factors, 50% for three riskfactors and 70% for four risk factors. (oral communi-cation, Retina Subspecialty Day presentation byDr. Frederick Ferris, AAO Annual Meeting, LasVegas, Nevada U.S.A., November 10, 2006).

Prevention of CNV and Genetics in AMDThe protective role of antioxidants and vitamins in theprevention of AMD was shown in the AREDS Study(43). The AREDS study enrolled 3640 participants andrandomized them to antioxidants alone, antioxidantsplus zinc, zinc alone and placebo. The results showedthat overall betacarotene (15 mg), vitamin C (500 mg),vitamin E (400 IU), zinc oxide (80 mg) and cupricoxide (2.2 mg) decreased the relative risk of pro-gression of AMD by 28% (OR 0.72, 43% to 32%absolute risk) and the risk of moderate visual loss by27% (OR 0.73). Patients with at least one large drusenin either eye, intermediate drusen in both eyes,noncentral geographic atrophy in one or both eyesand those with visual loss in one eye were consideredat high risk for advanced AMD. Patients with thesefeatures are recommended to take the supplements.However, those who smoke cigarettes are at increasedrisk of lung cancer from the betacarotene componentand it is therefore not recommended that they use theAREDS supplements. AREDS2 is a multicenteredinterventional trial that will investigate the protectiveeffects of supplements on CNV development when nobetacarotene is used in AMD patients who arecigarette smokers. AREDS2 will investigate the effectof oral supplementation with lutein, zeaxanthin andomega three fatty acids (in addition to the AREDSsupplements) on the development of advanced AMD(CNV and geographic atrophy).

The Physicians’ Health Study II is also evaluatingthe role of vitamin E, vitamin C, beta-carotene and adaily multivitamin. The Vitamin E, Cataract and Age-related macular degeneration Trial (VECAT) and theWomen’s Health Study (WHS) are two other random-ized trials assessing the risk and benefits ofantioxidant vitamins for AMD (44,45). The Caroten-oids in Age-Related Eye Disease Study (CAREDS) waspart of the WHS. In the CAREDS, the prevalence of

8: EXUDATIVE (NEOVASCULAR) AGE-RELATED MACULAR DEGENERATION 127

intermediate AMD in 1787 participants was foundto not statistically differ with respect to lutein andzeaxanthin intake (45).

Recently, genetic markers for both the risks ofAMD and CNV are being determined. (The geneticsof AMD are discussed in detail in Chao et al.’s chapter.)One highly associated genetic finding is that of comp-lement factor H (CFH). The inflammatory cascade hasbeen found to play a role in the pathogenesis of AMD.(The chapter byCsaky andCousinsdiscusses the role ofinflammation in AMD). Examination of the Subma-cular Surgery Trial (SST) specimens has revealed thatinflammatory cells are found throughout the CNVspecimens (46,47). The finding of CFH’s associationwith AMD and advanced AMD supports the role ofthe inflammatory cascade (48–51). (CFH is normally aregulator of the complement cascade and limits theimmune reaction to spare host cells. When there is aCFHmutation, the CFH can no longer protect host cellsand the host cell undergoes lysis through activation ofcomplement.) In addition to CFH mutations, there areother complement factors being evaluated in the searchfor the inflammatory component of exudative AMD.

It is also known that an imbalance in stimulatorsand inhibitors of AMD are involved in development ofCNV. For example, pigment epithelial derived factor(PEDF) is a naturally found inhibitor of angiogenesis(52). It is manufactured by RPE cells in the eye. InAMDCNVMs, levels of PEDF are markedly lower andvascular endothelial growth factor (VEGF) levelsmarkedly higher compared to normal controls.Further details on angiogenesis and rationale of anti-angiogenesis treatments are found in the chapter byCampochiaro and Kane.

Until all risk factors (and subsequent proof ofcausation) and the genetics of AMD are fullydetermined, prevention of exudative AMD remainsan enigma. Further clinical trials proving benefit ofintervention are necessary before recommendationscan be made. However, at this time, modifiable riskfactors (such as smoking, hypertension) should beaddressed as they are also linked to systemic diseases.Future application of genetic therapy and targetedantiangiogenesis treatments are now beginningtoplay a role in the prevention of exudative AMD andattendant visual acuity loss. It is possible that futuretreatments of AMD will include targeted genetictherapy to replace defective genes.

SYMPTOMS AND MONITORING FOR EXUDATIVE(NEOVASCULAR) AMD

Symptomatic patients with exudative AMD typicallypresent complaining of sudden onset decreased visualacuity, metamorphopsia, and central or paracentralscotomas (53,54). Not infrequently, patients are

unaware that one of their eyes has already hasalready lost vision until he or she covers the unaffectedeye. Other times, patients present with loss of vision inthe previously “good eye” and may have beenunaware of the visual symptoms in the fellow eyecontaining a macular scar (55). Yet other patients maybe asymptomatic and routine ophthalmoscopy mayreveal the CNV in the second eye (prior CNV in thefellow eye) (56). Thus, patients who are at risk for CNVshould be periodically screened for development ofCNV and should be encouraged to self monitor theirvision daily. Monitoring options include using anAmsler grid or the preferential hyperacuity perimeter(PHP) (57,58).

Patients with known cataracts may attributeblurred vision to their cataract and not suspect AMDas the cause. In some patients with dense cataracts,cataract extraction without pre-operative detection ofthe CNV may occur. After the cataract extraction, theophthalmologist then detects the CNV. A careful pre-operative examination for exudativeAMDor advancednon-exudative AMD is therefore of utmost importancein patients with knownAMD. Preoperative fluoresceinangiography (FA) and optical coherence tomo-graphy (OCT) may help detect CNV. Alternatively, ifthe cataract density precludes ophthalmology, FA orOCT imaging, an ultrasound examination may beuseful in screening for macular fluid or subretinalscar formation to rule out advanced AMD (59).

The Amsler grid is a useful test for detecting theearly visual symptoms of exudative AMD in patientswith high risk AMD (60). Each box on the gridrepresents one degree of visual field. Thus, theAmsler grid tests the central 10 degrees of visualfield beyond fixation. The patient is asked to fixateon the central black dot and to note whethersurrounding lines are wavy, missing or obscured byscotomas (dark areas). If these findings are present, thepatient should be instructed to seek attention urgentlywith his or her ophthalmologist as it is likely that thecause is neovascular AMD. There are limits to Amslergrid testing which includes the cortical completionphenomenon, crowding phenomenon and lack offorced fixation.

A newly developed computer-automated, three-dimensional, threshold, Amsler grid visual field testhas been shown to be useful in earlier detection ofAMD (61). The 3-D Amsler grid utilizes thresholdtesting. There appear to be different signatures basedupon the type of AMD present. The PHP (PreViewPHP, Carl Zeiss Meditec, Dublin, California) machine(Fig. 1) has shown promise in the early detection ofexudative AMD (57,58). The PHP is based upon theconcept of vernier (hyperacuity) acuity, the abilityto detect a subtle misalignment of an object. Thethreshold of vernier acuity is three to six seconds of

128 LIM AND TSONG

arc in the fovea—10 fold smaller than to resolve anobject clearly on the fovea. When photoreceptors aremisaligned because of edema, CNV and or RPEelevation, the brain is able to detect the misalignment.The PHP is useful even in patientswithmedia opacitiesdue to its resistance to retinal image degradation. Thecentral 14 degrees are tested in about five minutes.Patients are shown a series of linear dotswith an area ofartificial distortion (Fig. 2). The artificial distortion isprogressively made smaller. If a patient has CNV, theCNV results in a true area of distortion of the dots.When their distortion is larger than the artificial dis-tortion, the patient preferentially chooses that area.A computerized map of these areas is created.

A study comparing retinal specialists’ gradingsof stereoscopic color fundus photos to the PHP for thedetection of CNV has been performed. The goldstandard for the determination of CNV was FA. In 64patients with recent onset CNV and 56 patients withintermediate AMD, retina specialists had a sensitivityof 70%, specificity of 95%, and an overall accuracy of

0.82 in detecting recent onset CNV. In comparison,PHP had a sensitivity of 83%, specificity of 87%, andoverall accuracy of 0.85 in detecting the same lesions.The PHP may indeed be a useful diagnostic device forpatient monitoring.

CLINICAL FEATURES OF EXUDATIVE(NEOVASCULAR) AMD

The major clinical features of active exudative AMDinclude subretinal fluid, subretinal hemorrhage,sub-RPE fluid, sub-RPE hemorrhage, RPE pigmentalterations and hard exudates. Chronic exudativeAMD is characterized mainly by the presence ofsubretinal fibrosis with or without the other featuresof active exudation. These features may appear clini-cally as any one or any combination of the following:a serous or a hemorrhagic PED (Figs. 3 and 4), grayishsubretinal membrane (Fig. 5), area of subretinal fluid,area of RPE alteration (Fig. 6), subretinal hemorrhage(Fig. 7), or hard exudates (Fig. 8). The latemanifestationof exudative AMD is a disciform scar (Fig. 9) orgeographic atrophy (Fig. 10), with or without sub-retinal fluid or subretinal blood. Spontaneousinvolution of CNV may manifest as any of the abovefindings with RPE alterations and or scar formation.

Stereoscopic fundus examination is the bestmethod for examining a patient with suspected CNV.A fundus contact or non-contact lens in conjunctionwith slit lamp biomicroscopy should be utilized for theexam. For those less comfortable with the non-contactfundus macular lenses, a fundus contact lens is easiestto use. The fundus contact lens or the 78 diopter lens

Figure 3 A patient with age-related macular degeneration and aserous pigment epithelial detachment (PED). Note the sharply

demarcated borders of the PED.

Figure 1 Preferential hyperacuity perimeter (PHP) machine:

PreView PHP, Carl Zeiss Meditec. Abbreviation: PHP, prefer-ential hyperacuity perimeter.

Figure 2 The preferential hyperacuity perimeter machine

displays a series of dots with an area of artificial distortion onthe screen. The patient is asked to select the area of distortion on

the touch screen. The artificial distortion height is progressivelydecreased. The patient will preferentially select the area of

distortion that is more severe.


(A) (B)

Figure 4 A patient with polypoidal choroidal vasculopathy with serous pigment epithelial

detachments (PEDs). (A) Color fundus photo shows sharply outlines of the PEDs. (B) Thecorresponding fluorescein angiogram shows uniform filling of the PED. Note the sharp borders

and absence of leakage.

(A) (B)

(C)

Figure 5 Grayish subretinal membrane fluorescein angiographychoroidal neovascularization (CNV). (A) Color fundus photo showssubretinal blood and a subretinal pigmented lesion. (B) Fluoresceinangiogram reveals the subfoveal classic CNV. Note the hyperfluor-escent border and the central vessels within the membrane. (C) Laterfluorescein angiogram phase showing leakage of the CNV.

130 LIM AND TSONG

offers more magnification than the 90 diopter lens.During the exam, it is helpful to have the patient lookdirectly at the thin slit lamp beam and to ask thepatient whether the beam appears distorted. Elevationof the RPE or retina (due to underlying CNV) causesthe patient to perceive distortion of the slit beam.

Using biomicroscopy with a macular lens, theseparation of the retina from the underlying RPE, dueto underlying subretinal fluid, can be seen. The over-lying retina may have cystic changes and may showcystoid macular edema. Sub RPE fluid appears as a

PED and typically has more sharply demarcatedborders as compared to subretinal (subneurosensory)fluid (Fig. 14). Often, there is a combination of sub-RPEand subretinal fluid associated with the CNV. TheCNV itself may be visible as an area of discoloration(Fig. 5). Other times, overlying subretinal blood orlipid may be the only clinical clue to the presenceof an acute CNV. The definitive test for the presence ofCNV has been FA. This is further discussed below.

Figure 6 Retinal pigment epithelium mottling. Note the area of

increased pigmentation in this eye with subfoveal choroidalneovascularization. There are numerous large drusen also

present.

Figure 7 Subretinal hemorrhage. Extensive subretinal hemor-

rhage is present. Note that this myopic patient has large softdrusen as well as an adjacent area of retinal pigment epithelial

detachment temporal to the hemorrhage.

(A)

(B)

(C)

Figure 8 Hard exudates in an eye with choroidal neovascular-ization. (A) Fundus photograph shows subretinal hemorrhage

and the extensive hard exudates deposits. On the (B) vertical and(C) horizontal optical coherence tomography scans, the hard

exudates appear as focal intraretinal areas of high backscattering(orange color). Beneath the hard exudates, there is a markedly

decreased signal (black). The subretinal fluid appears as anelevation of the neurosensory retina above an optically clear

space (black). Intraretinal cysts are also seen in the inner retinallayers.


OCT has been an extremely useful tool in thedetection and management of CNV in AMD patients.The OCT resolution may be 3 to 5 mm for high res-olution OCT and 10 mm for the third generation OCTmachine (62). Microscopic areas of subretinal fluid andareas of elevation can be detected on OCT imaging ofthe macular area of AMD patients. Areas of CNVappear as RPE thickening with or without intraretinalcysts and subretinal fluid (Figs. 11–13). PEDs areclearly seen on the OCT (Fig. 12). OCT has been usedin the recent anti-angiogenesis clinical trials as anothermeasurement of treatment outcome. Successful treat-ment of PEDs and CNVMs has been shown to result innormalization of the OCT appearance (Fig. 13). Recur-rence of the CNV can appear as slight areas of elevationof the RPE, neurosensory retina or presence of cysticretinal change. Details of the usefulness of OCTimaging in management of AMD patients is found inthe chapter by Reichel and coworkers.

Further refinements in OCTcontinue; volumetricevaluation of the CNV lesions is becoming a reality.Other instruments can image the lesions and provide

Figure 9 Disciform scar. There is extensive subretinal fibrosis

(white areas) on the color fundus photograph of this end-stageage-related macular degeneration eye. Note the sharp margins of

the fibrotic lesion. There are areas of pigmentary abnormality withsome retinal pigment epithelial atrophy in addition to the whitish

fibrotic scar tissue. Note the residual central hemorrhage.

(A) (B)

(C)

Figure 10 (A) Color fundus photo shows a well demarcated area ofretinal pigment epithelial and choriocapillaris atrophy. Note the orangecolor of the atrophic lesion and the visibility of the deep choroidalblood vessels within the area. There are no drusen in the atrophic areabut soft drusen in the area adjacent to the lesion. (B) Thecorresponding early fluorescein angiogram frame shows hyperfluor-escence corresponding to the atrophic area. (C) The correspondinglate fluorescein angiogram frame shows well-demarcated borders thatmatch the area of atrophy seen clinically (staining). No fluorescein dyeleakage is seen (no blurring of image). The borders remain sharplydemarcated. Note that the area stained on the angiogram corre-sponds with the clinically visible lesion borders.

132 LIM AND TSONG

quantitative information. The retinal thicknessanalyzer can determine lesion dimensions of CNV(63). Quantitative measurements of the retina willundoubtedly prove useful in the management ofpatients with CNV. Further information on quan-titative imaging can be found in the chapter byEsmaili and colleagues.

Pigment Epithelial DetachmentThe borders of a PED are usually sharply demarcated(Fig. 3). Clinically, hemorrhage or hard exudates mayor may not be present depending upon the presence orabsence of associated CNV.A fluorescein angiogram orindocyanine green (ICG) angiogram is clinically usefulto detect the presence of associated CNV. A serous PEDshows early hyperfluorescence and uniform fluor-escence on the late frames of the angiogram (Fig. 4B).The dye pools in the PED on the late phase. Theborders remain sharp and the area does not increasein size. On ICG angiography, the PED is hypofluor-escent (see ICG chapter by Oliver and colleagues).Whereas a serous PED will show uniform filling of

the PED, a vascularized PED shows irregular filling,notching of the PED (Fig. 14) or irregular margins onthe FA. On the OCT, the RPE elevation is readily seen.

If there is CNV present with the PED, occult CNVwill frequently show associated subretinal fluid, hardexudate, or subretinal blood. The fluorescein angio-gram typically demonstrates irregular filling of thePED and the PED borders may be blurred in the areaof the CNV. Leakage on the late frames of the FA iscommonly noted (Fig. 14). ICG angiography has beenshown to be helpful in this regard (64). As shown inthe chapter by Oliver and colleagues, ICG can identifyareas of CNV associated with the PED. Laser photo-coagulation of the hot spots may result in resolution ofthe PED, subretinal fluid, blood and lipid (64–66).CNV has been associated with 28% to 58% of PEDs(65). A study by Elman and colleagues showed that32% of serous PEDs develop CNV at a mean of 19.6months (67). Risk factors associated with CNV in theseeyes included patient age greater than 65 years,associated sensory retinal detachments and fluor-escein findings of hot spots, notches, late or irregularfilling. The association of CNV with PED increases thechance for visual acuity loss (67–69).

In a natural history study by Poliner and associ-ates, the risk of developing CNV was 26% at one year,

Figure 11 Optical coherence tomography from an age-related

macular degeneration patient with subfoveal choroidal neovas-cularization (CNV). The overlying retina shows multiple cysts.

The CNV is seen as an area of thickening in the subfoveal zone at

the retinal pigment epithelial level.

Figure 12 Optical coherence tomography from an age-relatedmacular degeneration patient with a subfoveal pigment epithelial

detachment (PED) and adjacent choroidal neovascularization.There is a dome-shaped elevation of the retinal pigment epi-

thelium (RPE) reflective border (orange). This represents thePED. The adjacent dark areas under the neurosensory retina and

above the RPE represent subretinal fluid.

(A)

(B)

Figure 13 Vertical optical coherence tomography scans before

and after ranibizumab (Lucentis, Genentech, South San Fran-cisco, California, U.S.A.) treatment of a subfoveal choroidal

neovascularization. (A) Prior to ranibizumab injection, there isextensive intraretinal cystic edema and underlying subretinal

fluid. The visual acuity is 20/400. (B) One month after the firstranibizumab intravitreal injection, the subretinal fluid is resolved.

Visual acuity is 20/200. The intraretinal cystic edema is markedly

decreased in severity. The patient continues with the intravitrealranibizumab injection therapy.


42% at two years and 49% at three years in eyes withPEDs followed for 12 or more months (68). The risk of20/200 or worse visual acuity increased from 17% atone year to 33% at two years and 39% at three years.The majority of eyes (78%) that developed CNV were20/200 or worse, while only 3% of eyes that did notdevelop CNV lost vision to that level.

Even with spontaneous flattening of PEDs, thevisual acuity outcome is poor (67–70). Unfortunately,most clinical trials have excluded PEDs with CNVandthere remains no good treatment for this group of eyes.Currently, off label treatments are being applied to

these CNV lesions with PEDs. Anti-VEGF therapiessuch as ranibizumab (Lucentis) and off-label bevaci-zumab (Avastin) have been used to treat CNV withPED with some success. The risk of an RPE tear/ripoccurring in this setting is a real concern in theseeyes (Fig. 15). An RPE tear is readily identifiable as asharply-demarcated area of bare choroid with astraight, linear edge. This straight, linear edge corre-sponds to the location of the associated retracted,scrolled RPE. The fluorescein angiogram showsblocked fluorescence in the area of scrolled RPEand hyperfluorescence in the area without RPE.

(A)

(B) (C)

Figure 14 Pigment epithelial detachment (PED) with associated choroidal neovascularization(CNV). (A) (Stereo photo pair.) The PED is elevated on the stereo images. There is subretinal fluid

overlying the entire lesion in addition to the sub retinal pigment epithelium PED fluid. Note thegrayish area on the superonasal edge of the PED. This grayish area corresponds to the CNV. (B)

The corresponding fluorescein angiogram from the late transit phase shows a notch of the PEDsuperonasally. (C) The corresponding fluorescein angiogram from the late phase shows

fluorescein dye leakage in the area corresponding to the CNV. The adjacent PED shows sharpedges in the areas not involving the CNV.

134 LIM AND TSONG

The natural history of PEDs includes RPE tears, buttreatment of CNV with PEDs has also been tempor-arily associated with RPE tears (71).

Choroidal NeovascularizationThe MPS group has defined the various forms andcomponents of CNV (72). The entire complex ofcomponents termed a “CNV lesion” includes theCNV itself, blood, elevated blocked fluorescence (dueto a pigment or scar that obscures the neovascularborders), and any serous detachment of the RPE.

The classic clinical description of a choroidalneovascular membrane is that of a dirty gray-coloredmembrane (Fig. 5). There is associated subretinal fluidand there may or may not be subretinal blood andlipid. Sometimes the outlines of the CNV are clearlyvisible with the subretinal vessels readily seen. Othertimes, the CNV is manifest only by a neurosensorydetachment or even subretinal blood.

The fluorescein angiogram is a key test in theevaluation of patients with CNV. On FA, a well-

demarcated area of choroidal hyperfluorescence isseen early (Figs. 5 and 16). The MPS group charac-terized classic CNVas only occasionally showing a lacypattern of hyperfluorescence in the early fluoresceinphases. In the later frames of the angiogram, theboundaries of the CNV are obscured by progressivepooling of dye in the subneurosensory space. Withthe advent of photodynamic therapy (PDT), the term“predominantly classic” was coined. A predominantlyclassic lesion is one inwhich the lesion ismore than 50%classic CNV in composition (Fig. 16).

Occult CNV has been classified as either fibro-vascular PED (FVPED) (Fig. 17) or late leakage ofundetermined source (LLUS) (Fig. 18). These types ofoccult CNVs are differentiated on the basis of thefluorescein angiogram. A stereoscopic FA is veryhelpful in recognizing occult CNV. FVPEDs showearly hyperfluorescence with irregular elevation ofthe RPE. These areas are not as bright or as discreteas the classic CNV seen on the transit phases.Within one to two minutes, an area of stippled

(A) (B)

(C)

Figure 15 Retinal pigment epithelium (RPE) tear occurring afterseveral anti-vascular endothelial growth factor (VEGF) treatments.(A) Color fundus photo shows the area of exposed choroid where theRPE tear is located. Note that this area has one linear edge. Thislinear edge is formed by the scrolled RPE. There are also choroidalfolds in the macular area. (B) Early fluorescein angiography (FA)phase frame shows hyperfluorescence in the area of denuded RPEand blocked fluorescence in the area of the scrolled RPE tear. (C)Late FA phase frame shows hyperfluorescent staining in the area ofthe RPE rip. There is blocked fluorescence corresponding to thescrolled RPE. The adjacent choroidal neovascularization showssome mild leakage.


hyperfluorescence is present. By ten minutes there ispersistent fluorescein staining or leakage within thesubneurosensory detachment. The borders of theoccult CNV may be either well-demarcated or poorly

demarcated (72). Late leakage is present, although it isnot as intense as that seen in classic CNV (72).

LLUS in contrast does not show early hyperfluor-escence. LLUS appears as speckled hyperfluorescence

(A) (B)

(C)

Figure 16 Predominantly classic choroidal neovascularization(CNV). (A) Color fundus photo of a patient with subfoveal CNV.There is subretinal fluid overlying the subfoveal lesion. Areas ofhemorrhage and hard exudates are seen in the macular lesion. (B)The corresponding early fluorescein angiographic frame shows anarea with bright, well demarcated fluorescence (classic CNV).There is blocked fluorescence corresponding to the areas ofhemorrhage. There are also areas of speckled fluorescence(occult CNV) beyond the area of classic CNV. Since the area ofclassic CNV occupies more than 50% of the entire lesion, the lesionis predominantly classic CNV in composition. (C) The corre-sponding late fluorescein angiographic frame shows intenseleakage from the entire CNV component. The areas of blockedfluorescence corresponding to the hemorrhage remain unchanged.

(A) (B)

Figure 17 Fibrovascular pigment epithelial detachment. (A) The subfoveal area shows hyper-

fluorescence. The early fluorescein angiographic frame shows hyperfluorescence with somestippled fluorescence. On stereo viewing (not shown) irregular elevation of the retinal pigment

epithelium is seen within the area of leakage. The leakage is not as bright as that seen by the samephase of classic choroidal neovascularization. (B) On the late fluorescein angiographic frame, at

five minutes, more intense fluorescein leakage is seen in this area.

136 LIM AND TSONG

with pooling of dye in the overlying neurosensoryspace; choroidal leakage is apparent betweentwo and five minutes after fluorescein injection.The boundaries of this type of occult CNV are neverwell-demarcated. In fact, the later frames showhyperfluorescent leakage in an area that showed nohyperfluorescence on the early frames (Fig. 18) (72).

Lastly, there is a slow-filling form of classic CNVin which hyperfluorescence is not seen until twominutes. However, in this form of CNV, the lateframes of leakage and pooling of the dye in the sub-neurosensory space correspond with the area seen attwo minutes.

Using ICG angiography, occult CNVs can befurther classified into those with hot spots, plaques,

combination of these two types, retinal-choroidalanastomosis and polypoidal-type CNV. Using ICGangiography, about one third of eyes with occultCNV become eligible for treatment (73). ICG angio-graphy is also useful for evaluating eyes withsubretinal hemorrhage for the presence of CNV.Further details of the usefulness of ICG angiographyand ICG-guided laser photocoagulation of CNV inAMD can be found in the chapter by Oliver andcolleagues. With the recent finding that pegaptanibsodium (Macugen) and ranibizumab (Lucentis) workequally well for different types of CNV, the classi-fication of the CNV may become less important inthe future. The prognostic implication of CNV type onresponse to treatment remains to be determined.

(A)

(C)

(B)

(D)

Figure 18 Late leakage of undetermined source. (A, B) The earlier angiographic frame show no

evidence of leakage. (C, D) On the late angiographic frames, areas of fluorescein dye leakageappear. This area shows no corresponding area of leakage on the early angiographic frames. The

leakage is extrafoveal.


CNV lesions are further characterized by theirlocation in relation to the foveal center. The location ofthe CNV is divided into extrafoveal, juxtafoveal andsubfoveal. These definitions were created by theMacular Photocoagulation Study Group and are asfollows:

Location of CNVDistance from fovealavascular zone center

Extrafoveal 200–2500 mm

Juxtafoveal 1–199 mm

Subfoveal 0 mm

A lesion is juxtafoveal if the CNV border closestto the foveal center is within (but not involving thefoveal center itself) the foveal avascular zone (FAZ)(Fig. 19). A lesion whose closest border to the fovealedge is beyond the FAZ is considered extrafoveal.

Disciform ScarA disciform scar shows an area of subretinal fibrosisor subRPE fibrosis. Dull, white fibrous tissue is seenand may accompany the CNV lesion or replace it overtime (Fig. 9). Areas of retinal pigment epithelialatrophy may or may not be present. FA may showleakage associated with the scar if active CNV ispresent. The fibrotic scar may otherwise show onlystaining of the fibrotic tissue.

Patients with CNV typically present with symp-toms of metamorphopsia, decreased vision, unioculardiplopia, Amsler grid distortion, scotoma ormacropsia. The severity of the symptoms varies

depending upon the location of the CNV. Obviouslylesions closer to fixation will cause more noticeablesymptoms in the patient’s visual field. Patients com-plaining of such symptoms require prompt clinicalevaluation and FA to detect any CNV or PED and tocharacterize the CNV by type, location and size.Treatable lesions should undergo laser photocoagula-tion within 72 hours of the fluorescein angiogramfor extrafoveal and juxtafoveal CNV. Subfoveallesions can now be treated by PDT within one weekif the lesion is eligible for PDT or by a variety ofantiangiogenesis therapies. Anti-angiogenesis treat-ments should be preceded by a fluoresceinangiogram for diagnosis and determination of theextent of the CNV. OCT prior to treatment andduring follow-up is useful to gauge the clinicalresponse. Central retinal thickness, presence of subret-inal fluid and retinal cysts are all parameters that canbe monitored using the OCT.

Recently, the anti-angiogenesis therapies haveshown visual improvement is possible even afterweeks of untreated disease (74). Thus, because of thispotential visual recovery, as long as subfoveal scarringis absent, it seems reasonable to treat active CNVevenif it is not of recent onset. In the phase I/II ranibizumabtrials, nine of the eleven eyes in the untreated groupwere switched to treatment at day 98. Even at thatdelayed time interval between onset of CNV andtherapy, these eyes experienced a mean visual acuityimprovement at six months (7.3G13.1 letters for thoseswitched to 0.3 mg ranibizumab and 3.2G9 letters inthe 0.5 mg group).

Feeder VesselsA feeder vessel is a choroidal vessel that connects theCNV to the underlying choroidal vasculature thussupplying blood to the CNV membrane. Green hassuggested that there are two to three feeder vesselscrossing Bruch’s membrane per CNV (75). Feedervessels are sometimes ophthalmoscopically visiblewithin the CNV lesion (see chapter by Flower).Recent work has focused on applying laser photocoa-gulation to feeder vessels in an attempt to close theCNV. Feeder vessels have been reported to be presentin 15% of cases of CNV. Shiraga and coworkers firstreported identification and photocoagulation of feedervessels using ICG videoangiography via a scanninglaser ophthalmoscope (76). In 70% of the patients, theexudative findings resolved; visual acuity improvedor stabilized in 68% of patients. Later, Staurenghi andcoworkers verified the superiority of dynamic ICGangiography with an scanning laser opthalmoscopesystem for identifying feeder vessels in subfovealCNV (77). Dynamic ICG can detect smaller feedervessels and enables more targeted treatment of thesevessels with a 75% success rate (78).

Figure 19 Juxtafoveal choroidal neovascularization (CNV).The fluorescein angiography (FA) shows that the closest edge

to the foveal center is within the foveal avascular zone. Notehowever that the edge of the CNV does not involve the foveal

center itself.

138 LIM AND TSONG

Most recently, using high speed high resolutiondigital angiography, it is possible to detect more feedervessels in eyes with CNV. The combined ICG angio-graphy/dye-enhanced photocoagulation systemallows one to synchronize photocoagulation with thearrival of the dye bolus at a targeted vessel site.

PATHOGENESIS OF CNV

The pathogenesis of CNV is not fully understood atthis time. However it is well accepted that angiogenicfactors, such as VEGF, has a primary role in theinitiation and maintenance of CNV. The primarystimulus resulting in the increase in angiogenicfactors remains unknown. The angiogenesis factorsinvolved in the neovascular response are discussedin detail in the chapter by Campochiaro and Kane. Ofthe known isoforms of VEGF, the most important inthe eye is VEGF A. VEGF A has five known isoforms.VEGF 165 and VEGF 121 are important in ocularneovascularization as well as the cleaved VEGF break-down product, VEGF 110.

Clues to the pathogenesis of CNV are availablefrom surgically excised membranes (46,47). Themost consistent pathological finding is accumulationof abnormal extracellular matrix (ECM) resulting indiffuse thickening of Bruch’s membrane (49). Focalareas of thickening form drusen and this diffusethickening suggests an altered metabolism of theECM. There is data to suggest that altered ECM ofRPE cells causes increased secretion of angiogenicgrowth factors that could contribute to the growthof CNV (46,47). Iatrogenic breaks of Bruch’smembrane in animals have led to animal modelsof CNV (79). Laser-induced CNV in primates hasbeen used to investigate the mechanisms of CNVproduction and the role of the RPE (80). The chapterby Kang and Grossniklaus discusses these changesin detail.

It is known that RPE cells produce VEGF andfibroblast growth factor 2. Both are present in fibro-blastic cells and in transdifferentiated RPE cells ofsurgically excised CNV (81–83). Healthy photo-receptors are needed to prevent the choriocapillarisfrom responding to excess VEGF (84). In addition,inflammatory cells are now felt to be key ingredientsto CNV development. In a murine laser-inducedmodel of CNV, CNV volume was significantlysuppressed when inflammatory mechanisms wereinhibited (85). Angiotensin II type I receptor (AT1-R)signaling blockade with telmisartan inhibited macro-phage infiltration and upregulation of VEGF,intercellular adhesion molecule-1 (ICAM-1), monocytechemoattractant protein-1, and Interleukin-6 in theretinal pigment epithelium-choroid complex. The

research showed that AT1-R-mediated inflammationplays a pivotal role in the development of CNV.Perhaps, AT1-R blockade may serve as another thera-peutic strategy to inhibit CNV.

The fact that VEGF is present in CNV has ledto the development of drugs that bind VEGF(antibodies or aptamer) or its receptor or drugs thatblock VEGF signaling. Blockage of VEGF signalinghas been shown to inhibit the development of CNVin the laser-induced CNV mouse model (86,87).Initially it was thought that VEGF 165 was the mostsignificant VEGF isoform for CNV. This led to thedevelopment of an aptamer to bind VEGF 165 andthus inhibit VEGF 165 as a treatment for CNV inAMD patients (88). Anti-VEGF antibodies were alsodeveloped as a treatment for CNV (89,90). The resultsof the clinical trials evaluating these treatmentswill be discussed later in this chapter. Furtherdetails are found in the chapter by Klesert andcolleagues.

IDIOPATHIC POLYPOIDAL CHOROIDAL VASCULO-PATHY OR POLYPOIDAL CHOROIDALVASCULOPATHY

Idiopathic polypoidal choroidal vasculopathy (PCV)has recently been classified as a form of CNV that mayoccur in elderly patients. A recent study by Yannuzziand colleagues determined the frequency and natureof PCV in patients suspected of harboring exudativeAMD (91). In their prospective study of 167 newlydiagnosed patients with exudative AMD, CNV wasdiagnosed in 154 (92.2%) and PCV in 13 (7.8%). Non-white race (23.1%), absence of drusen (16.7% haddrusen) and peripapillary location were felt to dis-tinguish between PCV and AMD. Since then, it is nowrecognized that PCV occurs in all races (92). PEDs arecommonly seen in PCV (Fig. 4). ICG angiography isuseful in the diagnosis of this entity. Further infor-mation is found in the ICG chapter by Oliverand colleagues.

RETINAL ANGIOMATOUS PROLIFERATION

One other distinct type of neovascular AMD isretinal angiomatous proliferation (RAP). This entityis characterized by an anomalous retinal vascularcomplex which is most commonly associated withretinal and subretinal neovascularization (93,94). Ithas been described predominantly in elderly Cauca-sians and is often seen bilaterally. While its naturalhistory is not fully understood, it is thought toprogress ultimately to a disciform scar. Prior to the


recognition of this entity, it was often misdiagnosed asoccult CNV.

A three-stage classification system of RAP hasbeen proposed by Yannuzzi and colleagues to describethe various clinical presentations and to theorize onthe disease’s natural history (94). In stage I, a nodularmass of intraretinal neovascularization is seen andoriginates from the deep capillary plexus in theparamacular area. There is usually one or moreassociated retinal vessels which either perfuse ordrain the vascular complex. Intraretinal hemorrhagesand intraretinal edema are often present. FA typicallyshows a focal area of staining corresponding to theintraretinal neovascularization. Surrounding leakageis present and often misinterpreted as occult CNV.ICG angiography can aid in the diagnosis by identi-fying the neovascularization as a focal “hot spot”and intraretinal cystic spaces as focal hyperfluorescentareas.

Stage II, subretinal neovascularization, involvesboth retinal and subretinal vascular proliferation. Theneovascularization occurs in a tangential directionwith minimal horizontal extension. Other commonsigns include increased intraretinal edema, neurosen-sory retinal detachment, serous PED, and preretinaland subretinal hemorrhages. In many cases, a clearretinal–retinal anastomosis can be seen. FA oftenshows a diffuse area of leakage which is, again, oftenmisinterpreted as occult CNV.

Stage III of RAP is defined by the Stage IIfindings plus the clear presence of CNV (Fig. 20).This is most often documented by the presence of a

FVPED or a predisciform scar. Occasionally, thepresence of a retinal-choroidal anastomosis helpsconfirm the staging. OCT images representative ofRAP are found in the chapter by Oliver and colleagues(Figs. 3 and 4 in the chapter 9).

PROGNOSTIC IMPLICATIONS OF EXUDATIVE AMD:NATURAL HISTORY OF UNTREATED CNV

The natural history of untreated CNV in the setting ofAMD is well established in both retrospective reviewsand prospective randomized controlled clinical trials.Untreated, eyes with CNV often loose visual acuity.The location (extrafoveal, juxtafoveal, subfoveal) ofthe CNV is linked with the visual acuity prognosis.Obviously, subfoveal CNV causes more immediatevisual symptoms than lesions further from thefoveal center.

Natural History of Extrafoveal CNVThe Macular Photocoagulation Study on extrafovealCNV provides us with robust natural history outcomedata for eyes with similar baseline characteristics andextrafoveal CNV. In the MPS untreated group, initialvisual acuity was 20/100 or better in thesesymptomatic eyes.

When untreated, 50% of patients with extra-foveal CNV had, by the time of the first follow-upvisit (three months after enrollment for 98 eyes),already lost two or more lines of visual acuity; 10%had suffered a loss of six or more lines of visual acuity(95). Thus, eyes with classic extrafoveal CNV are at

(A) (B)

Figure 20 Retinal angiomatous proliferation (RAP). (A) Early and late fluorescein angiography

images of a patient with RAP. The early frame shows a focal area of staining around a retinalvessel off of the infero-temporal arcade. There are associated pinpoint areas of hypofluorescence

from blockage from blood. (B) The late frame shows increasing fluorescence and leakageassociated with the choroidal neovascularization in this RAP lesion. Patient’s visual acuity was

20/100.

140 LIM AND TSONG

high risk for visual acuity loss without prompt treat-ment. Patients remain in this non-subfoveal phase foronly a short time after the onset of symptoms (96).

At the conclusion of the extrafoveal study, inthe untreated (natural history) group, 80% of womenand 67% of men lost two or more lines of visual acuityfrom baseline; 43% of women and 47% of men lostsix or more lines of visual acuity from baseline (95).Thus, the natural course of extrafoveal CNV may bevisually devastating. Although photocoagulation isnot a cure for the majority of eyes, the MPS resultsas summarized below show that there is a statisticallysignificant benefit to photocoagulation versusobservation.

Natural History of Juxtafoveal CNVThirteen percent of patients with juxtafoveal CNV inthe natural history arm (249 eyes) of theMPS lost six ormore lines of visual acuity by threemonths after enroll-ment and 58% lost six or more lines by 36 months. Thejuxtafoveal study included eyes with visual acuity20/400 or better at entry (97). By five years 61% sufferedsix or more lines of visual acuity loss (98). Only 9.6%of eyes remained unchanged and 5.9% of eyes gainedtwo or more lines of visual acuity by five years.

Natural History of Subfoveal CNVThe MPS Subfoveal Study is the largest study of thenatural history of eyes with subfoveal CNVand initialvisual acuity of 20/100 or better. This study found thata majority of eyes will loose significant amounts ofvisual acuity over time if untreated. In fact, 77% ofpatients lost four or more lines of visual acuity at 24months and 64% lost six or more lines. The smaller thelesion at baseline, the better the initial visual acuity (99).

In the MPS subfoveal trials, eyes with subfovealCNVwere enrolled if initial visual acuity was between20/40 and 20/320. The visual acuity outcomes weredependent upon the baseline visual acuity and thelesion size. For all of the groups (A–D), visual acuity inthe natural history group continued to drop duringfollow-up.

For lesions one disc area or smaller in size withvisual acuity 20/125 or worse and for lesions greaterthan one and up to two disc areas with visual acuity20/200 or worse (Group A), 14% of untreated eyes lostsix or more lines of visual acuity at three months afterenrollment. By one year, 25% lost six or more lines ofvisual acuity. By four years, 35% lost six or more linesof visual acuity. These were eyes with small lesionsand poor visual acuities.

For lesions one disc area or smaller in size withvisual acuity 20/100 or better and for lesions greaterthan one and up to two disc areas with visual acuity20/160 or better (Group B), 11% of the natural historygroup lost six or more lines of visual acuity at three

months, 19% at six months, 38% at one year, 52% attwo years and three years and 55% at four years afterenrollment. Thus these eyes with better initial acuityand smaller lesion size had more visual acuity to looseover time.

For lesions two or more disc areas in size andinitial visual acuity 20/200 or worse (Group C), 8% lostsix or more lines of visual acuity at six months, 15% atone year, 13% at two years, 16% at three years and 25%at four years. Thus eyes with larger lesions and poorerinitial acuity had less visual acuity to loose over time.

For lesions more than two disc areas is size andinitial visual acuity 20/160 or better (Group D), 13%lost six or more lines of visual acuity by three months,26% at six months, 31% at one year, 54% at two years,45% at three years and 55% at four years. Eyes withlarger lesions and better visual acuity had more visualacuity to loose over time.

MACULAR PHOTOCOAGULATION STUDY

The MPS studies represent the first randomizedclinical trials for evaluation of treatments of neovas-cular AMD. The MPS studies gave us robust naturalhistory data for eyes with classic CNV in variouslocations. Although laser is now rarely used fortreatment of CNV, the results will be summarizedhere for historic purposes. (Further details are foundin the chapter by Cukras and colleagues.) The MPSenrolled patients with classic CNV. However, analysisof the data revealed that eyes were enrolled that hadclassic CNV associated with occult CNV. The MPStreatment recommendations, however, should applyto eyes with classic CNV. Only about 13% of AMDpatients with CNV are eligible for treatment by theMPS criteria (100). Patients should be informed thatlaser treatment results in a permanent scotoma(location, size, effect on central vision function suchas reading) and that there is a high risk of persistentCNV (CNV seen within six weeks of treatment afterclosure) or recurrent CNV (CNV developing after sixweeks of treatment and initial closure). In the currentera, laser photocoagulation should be considered onlyfor non-subfoveal lesions.

Extrafoveal CNVThe original MPS report on the efficacy of laserphotocoagulation for extrafoveal CNV in the settingof AMD was published in 1982 (95). That studyshowed an overwhelmingly positive effect of lasertreatment for extrafoveal CNV (200–2500 mm fromthe foveal center). Eligibility criteria included patientage of 50 years or more, best corrected visual acuity atleast 20/100, presence of drusen, symptoms due to theCNV, no prior laser treatment, and no other eyediseases that could affect visual acuity. Treatment


was applied to the entire CNV and all surroundingblocked fluorescence (based on FA) or subretinalblood. The treatment, performed with 200 mm spotsof 0.5 second duration argon blue-green laser,extended 100 to 125 mm on all sides of the CNVbeyond blood, pigment or blocked fluorescence. Theintention was to treat any occult CNV in those regions.

After 18 months follow-up, 60% of untreatedeyes versus 25% of treated eyes suffered severevisual loss (p!0.001) (severe visual loss was definedas a loss of six or more lines of visual acuity). Thestudy recruitment was halted at 18 months because ofthis overwhelming treatment benefit and controlgroup patients were offered laser treatment if therewere eligible lesions. This report which was the firstrandomized controlled multicenter clinical trial fortreatment of AMD lead to a firm treatment recommen-dation of laser photocoagulation for extrafoveal CNVdue to AMD (95). Three years and five years data laterconfirmed the long term efficacy of laser photocoagu-lation in treated versus control, despite developmentof recurrent CNV in the treated eyes (101). RecurrentCNVoccurred in 54% of the eyes; these eyes had worsevisual acuity outcomes than eyes without recurrences.At the time of the study, no treatment was possible forthe eyes that developed subfoveal recurrent CNV.Now, with our current armamentarium of subfovealtreatment, better visual results would be expectedeven in eyes with the subfoveal CNV (88–90,102).

Juxtafoveal CNVThe MPS juxtafoveal AMD studies showed thatkrypton laser photocoagulation for juxtafoveal CNVwas effective for prevention of moderate and severevisual acuity loss. This study incorporated krypton redlaser because the red wavelength would not beabsorbed by the xanthophyll as much as blue laserlight and was thus felt to be safer. CNV lesionsbetween one to 199 mm of the foveal center or CNVbetween 200 and 2500 mm of the center with associatedblood or blocked fluorescence within 200 mm of theFAZ center were enrolled. Peripapillary CNV waseligible if the required laser photocoagulation wouldspare at least 1.5 clock hours along the temporal half ofthe disc. Treatment of the entire CNV with a 100 mmborder was required on the non-foveal border and inareas of blood or blocked fluorescence.

Eighty-six of 174 (49%) treated eyes versus 98 of169 (58%) observed eyes lost six or more lines of visualacuity at three years (97). At the 36 month visit, 62% ofuntreated and 49% of treated eyes had visual acuityworse than 20/200 (pZ0.02). The treatment effectdepended strongly on the presence or absence ofhypertension. Untreated eyes without hypertensionwere twice as likely to lose six or more lines of visualacuity compared to treated eyes (64% vs. 31%). This

effect was only 1.5 times for the eyes with hyperten-sion (70% vs. 46%). At five years, 71 (52%) of treatedeyes versus 83 (61%) of untreated eyes lost six ormore lines of visual acuity (95). The effect wasgreater for normotensive (RRZ1.82) than hyperten-sive (RRZ0.93) patients.

The MPS study found 32% of treated eyesshowed persistence and an additional 47% of treatedeyes developed recurrent CNV within five years afterkrypton laser to juxtafoveal CNV (103). Eyes withoutpersistent or recurrent CNV maintained 20/80 to20/100 visual acuity.

Persistent CNV was twice as high when therewas 10% or more of the foveal side of the CNV nottreated. Central leakage in the MPS studies was notlinked to a worse outcome. Forty-one percent of theeyes in the juxtafoveal group did not have adequatecoverage of the CNV on the foveal edge in contrast to14% for the extrafoveal group. TheMPS recommendedthat the visual loss may be reduced by covering theentire CNV lesion with laser treatment. More than 90%of recurrences are on the foveal side following lasertreatment of extrafoveal and juxtafoveal CNV.

Subfoveal CNVSubfoveal CNV was investigated by the MPS groupbeginning in 1986. Investigators felt that the poornatural history of eyes with subfoveal CNV (99,104)and scattered reports of outcomes of subfoveal lasernot resulting in uniformly poor visual acuitywarranted trial of photocoagulation of subfovealCNV lesions (105). In Jalkh et al.’s study of 94 eyesfollowed up for an average of 15 months, CNV wasclosed in 88 eyes and visual acuity was stabilized orimproved in those eyes (105).

Patients were eligible for inclusion into the MPSsubfoveal study if there was some classic CNV, thelesion borders were well-defined, the lesion was 3.5disc areas or less in size, or less than six disc areas (newarea of treatment plus old scar) if recurrent CNV waspresent. Visual acuity had to be at least 20/320 but nobetter than 20/40. A total of 373 eyes (371 patients)with new onset CNV and a total of 206 eyes (206patients) from 13 clinical centers were randomized tolaser treatment (argon green or krypton red asassigned during randomization) versus observation.Treatment was performed to all areas of classic andoccult CNV within the lesion. Treatment included aborder 100 mm beyond the margins for initial treat-ments or 300 mm into the old treatment scar forrecurrent CNV; treatment was based on a fluoresceinangiogram not more than 96 hours old. Posttreatmentphotographs were taken to check adherence to theMPS standards (106).

For treated eyes, visual acuity usually decreasedthree lines from baseline within three months after

142 LIM AND TSONG

treatment and then was stable for 42 months aftertreatment. In contrast, untreated eyes had lessdecreased visual acuity initially but continued todecrease throughout the follow-up period. Treatedeyes lost 3.3 lines at 12 months versus 3.7 lines foruntreated eyes. At 24months, treated eyes lost 3.0 linesversus 4.4 lines for untreated eyes. At three months,20% of treated eyes lost six or more lines of visualacuity loss; this remained stable at 42 months. Foruntreated eyes, 11% at three months had lost six ormore lines of vision, but increased to 48% at the 42months follow-up with pZ0.006. At the 42 monthsfollow-up, reading speed and contrast sensitivity werebetter for the treated than untreated eyes (106).

The persistence ratewas 24%; recurrence ratewas32% at three years. There was no difference betweenthe argon and the krypton groups. However, persis-tence and recurrence did not affect visual acuityoutcomes, unlike in the extrafoveal and juxtafovealgroups. The three year rate for subfoveal persistent orrecurrent CNV was 56%.

Subgroup analysis showed that treated eyes withsmaller CNV lesions (one disc diameter or less) experi-enced an earlier treatment benefit. Eyes with 20/40 to20/100 visual acuity lost on average more than fourlines of vision posttreatment and did not experienceany treatment benefit until 18 months later.

For the recurrent CNV subfoveal group, the MPSfound a similar treatment benefit. Ninety-seven eyeswere treated (49 argon, 48 krypton) and 109 wereobserved. Treated eyes lost approximately 2.5 lines ofvisual acuity three months after treatment followed bystable vision for 30 months. Untreated eyes continuedto lose visual acuity throughout follow-up such thatthe average loss was 1.1 lines more than the treatedeyes at 24 months. Six or more lines of visual acuitywere lost in 14% of treated versus 9% of untreated eyesat three months. This remained about 10% for thetreated group but increased to 32% for the untreatedgroup at 18 months of follow-up. Treated eyes retainedcontrast sensitivity whereas untreated eyes lostcontrast sensitivity.

After thermal laser photocoagulation, patientsrequire close monitoring consisting of visual acuity,Amsler grid, biomicroscopy, and FA to help detectpersistent or recurrent CNV. Usually patients arechecked three weeks after laser for extrafoveal/subfo-veal CNV and two to three weeks after laser forjuxtafoveal CNV. The second visit is typically four tosix weeks after laser, the third visit is six to twelveweeks after the laser and the fourth visit three to sixmonths after laser. Any symptomatic patient should beexamined immediately.

Overall, we no longer recommend thermal laserphotocoagulation to patients with subfoveal CNV.PDT and newer anti-angiogenesis treatments are

more efficacious and safe. Instead of an immediatevisual loss with laser photocoagulation, these newertreatments slow down the progression of visual loss inmost patients and sometimes even result in improvedvisual acuity.

OCCULT CHOROIDAL NEOVASCULARIZATIONNatural HistoryThe MPS group reviewed the results of the juxtafovealstudy with respect to the presence or absence ofoccult CNV. In the subgroup, they noted that therewere eyes with only occult CNV, occult and classicCNV and only classic CNV. For eyes with occult CNVthat were untreated, 41% within 12 months lost signi-ficant visual acuity. Of the 26 symptomatic eyes withoccult only lesions, classic CNV developed in 23%within three months and an additional 23% developedclassic CNV by 12 months. For the eyes whichdeveloped classic CNV, 58% developed severe visualloss. In the group that did not develop classic CNV,only 18% developed severe visual loss. Overall, 23% ofeyes which initially had occult-only CNV lesionsremained stable or improved at the 36 monthsfollow-up. Of these, 5% of the occult-only group hada two lines or more increase in visual acuity at 36months follow-up (107).

Bressler et al. performed a natural history studyof 84 eyes in 74 patients with poorly-defined fluor-escein angiographic CNV (108). The lesions weresubfoveal in 89% of the eyes. Initial visual acuityaveraged 20/80 and 93% had no classic CNV com-ponent. Over a follow-up ranging from six months to53 months (mean 28 months), 14% remained stable orimproved, 21% lost three to six lines of visual acuityand 42% lost more than six lines of visual acuity.Additional analysis, which included only those 46eyes with two or more years of follow-up, similarresults were found: 17% remained stable or improved,22% lost three to six lines of visual acuity and 48% lostmore than six lines of visual acuity. Eyes whichdeveloped disciform scars had worse visual acuitiescompared with eyes which had poorly-defined CNVand leakage.

Soubrane et al. (109) analyzed visual and angio-graphic outcomes of 156 patients (82 untreated) withsymptomatic occult CNV and initial visual acuity of20/100 or better. This series excluded eyes with turbidfluid, subretinal blood, PED or visible CNV. Follow-upranged from one to eight years. There was nodifference between treated and untreated eyes withCNV. Sixty-five percent of eyes with presenting subfo-veal CNV had initial visual acuity of 20/50 or better.In this natural history group, visual acuity fell from20/40 to 20/70. Similar to Bressler et al.’s results (108),when visible new vessels developed, the visual acuity


decreased. Treatment, when compared to the naturalhistory group, did not result in better visual acuityoutcomes over time (109).

Bressler et al. also evaluated macular scatter(grid) laser treatment of symptomatic eyes withpoorly demarcated subfoveal CNV. Visual acuityranged form 20/25 to 20/320 visual acuity in the 51treated eyes and the 52 observed eyes (110). Forobserved eyes, median visual acuity was 20/80initially and decreased to 20/320 at 24 months. Thedifference in visual acuity loss was significant betweenthe treated and observed groups only at six months(1.8 lines lost in the observed vs. 3.8 lines lost in thetreated group). At six months follow-up, treated eyeslost two more lines of visual acuity compared withobserved eyes. However, by 24months, approximately40% had severe visual acuity loss in both groups; meanchange was a loss of 4.3 lines for observed eyes and 4.6lines for treated eyes. At 12 months, 35% observed and29% treated had improved or remained stable, with37% observed and 30% treated losing two to five linesand 28% observed and 41% treated losing six or morelines of visual acuity. At 24 months, 31% observed and31% treated had improved or remained stable, 31%observed and 28% treated lost two to five lines and38% observed and 42% treated lost six or more lines ofvisual acuity.

Thus, conventional laser photocoagulation(either confluent or scatter) is not beneficial for AMDpatients with subfoveal occult CNV. However, we arenow in an era with several effective alternatives tolaser photocoagulation. These treatments are notunique to occult CNV and apply to other CNVlesion subtypes.

CURRENT THERAPEUTIC OPTIONS FOR CNV

The last few years have witnessed an explosion in theavailable non-ablative therapies for subfoveal CNV.The first proven and effective alternative treatment tolaser photocoagulation of subfoveal CNV was PDT.Large studies on the efficacy of antiangiogenesis agentshave not been performed on non-subfoveal lesions.More recently, the anti-angiogenic agents have beenshown to be effective treatments. These agents arecapable not only of stabilizing visual acuity but alsoimproving visual acuity. Gene therapy and com-bination treatments are investigational treatments.

Phototodynamic TherapyPDT utilizes a photosensitizer drug that is givenintravenously. The drug preferentially localizes to theCNV. A diode laser is used to deliver the diodewavelength of light (689 nm) to the lesion. Details ofthe application of PDT is found in the chapter byBlumenkranz and colleagues. PDT was initially

demonstrated to be effective for treatment of subfovealpredominantly classic (classic CNV comprises greaterthan 50% of the CNV lesion) CNV. PDT treatment ofsubfoveal CNVoffers an obvious advantage over laserphotocoagulation, which causes an irreversiblescotoma and hence visual acuity loss. The Treatmentof AMD with Photodynamic Therapy (TAP) studyshowed PDT with verteporfin dye (Visudyne,Novartis, East Hanover, New Jersey, U.S.A.) for subfo-veal CNV in AMD patients resulted in 61% of treatedeyes versus 46% of placebo eyes losing less than 15letters (approximately three lines of visual acuity) atone year. Subgroup analysis showed that eyes withpredominantly classic CNV (defined as 50% ormore ofthe lesion as classic CNV) had the best treatmentbenefit (67% treated vs. 39% placebo lost less thanthree lines of visual acuity) (111). Two-year resultsdemonstrated continued efficacy with PDT treatmentversus placebo (112). For eyes with 50% or less classicCNV, 66% of treated versus 32.5% placebo lost less thanthree lines of visual acuity at 24 months. Only a smallpercentage of patients (16% vs. 7%) gained one ormorelines of visual acuity. Thus patients should be told that,although PDT can help prevent severe visual loss,improvement of visual acuity is indeed rare. DespitePDT treatment a significant proportion of patients willstill lose visual acuity. PDT is also useful in situationswhere the treating ophthalmologist feels that conven-tional laser (113) could lead to visual loss (e.g.,juxtafoveal lesions close to the foveal center).

Eyes with occult lesions were studied as part ofthe Verteporfin in Photodynamic Therapy Study (VIP).The VIP study showed PDT for occult CNV lesionswasbeneficial at year 2; the one-year data showed nostatistically significant difference between Visudyneand placebo, but did show a trend in favor of Visudynetherapy (114). At year 2, 45% of Visudyne eyes versus31% of placebo eyes (pZ0.03) lost less than three linesof visual acuity and 71% of Visudyne versus 53% ofplacebo eyes (pZ0.004) lost less than six lines of visualacuity. The treatment effect was best for eyes with20/50 or less visual acuity and lesion size smallerthan four MPS disc areas. Further evaluation ofsmaller lesion sizes in the Visudyne in Occult CNV(VIO) showed no benefit at year 1 or year 2 (115). PDTtreatment is not very beneficial for occult CNV lesions.

The visudyne in minimally classic (VIM) studyevaluated the use of Visudyne PDT using reducedfluence (RF) and standard fluence (SF). RF was inves-tigated as a way to increase selectivity while limitingpotential adverse effects to normal tissue.At 12months,the VIM study showed 14% (5 of 36) of RF eyes and 28%(10 of 36 eyes) of SF eyes, comparedwith 47% (18 of 38)of placebo eyes (RF, pZ0.002; SF, pZ0.08; RFCSF,pZ0.004) lost three or more lines of visual acuity.At month 24, this loss occurred in 26% (nine of 34) of

144 LIM AND TSONG

RF eyes and 53% (17 of 32) of SF eyes, compared with62% (23 of 37) of placebo eyes (RF, pZ0.003; SF, pZ0.45;RFCSF, pZ0.03) (116). There were more eyes thatprogressed from minimally to predominantly classicCNV by 24 months in the placebo group than eithertreatment groups [11 (28%) of 39 patients comparedwith 2 (5%) of 38 in the RF group (pZ0.007) and 1 (3%)of 37 in the SF group (pZ0.002)]. No unexpected ocularor systemic adverse events were identified. It wasconcluded that PDT with Visudyne safely reducedthe risks of moderate visual loss and progression topredominantly classic CNV for at least two years inindividuals with subfoveal minimally classic lesionsdue to AMD measuring six MPS disc areas or less.

More recently, combination therapy using PDTwith triamcinolone or with anti-angiogenic agentshas been tried. Limited case reports show a decreasednumber of required PDT treatments as well asimproved visual acuity results from combined intra-vitreal steroid injections with PDT (117–119). Onegroup however found no visual benefit but adecreased number of required PDT treatments toclose the CNV (119). There are currently several multi-center studies investigating the benefit of intravitrealtriamcinolone combinations with PDT. Further detailsof PDTare found in the chapter by Jain and colleaguesin this book. Others are now combining PDTwith anti-angiogenesis treatments.

Anti-angiogenesis TreatmentsAs mentioned previously, VEGF plays a major rolein angiogenesis. Inhibition of VEGF is therefore arational treatment approach for CNV therapy. Thefirst anti-VEGF treatment study, the VEGF InhibitionStudy in Ocular Neovascularization (V.I.S.I.O.N.) trialused a VEGF aptamer (pegaptanib sodium, Macugen)to treat subfoveal CNV. All lesion subtypes andlesions up to 12 disc areas were included. Patientswere randomized to receive an intravitreal injection ofMacugen (three does) or to a usual care group. Theusual care group allowed the use of PDT for predo-minantly classic lesions.

In this study of 1186 patients, the 0.3 mg pegap-tanib dose was effective in the prevention of visualloss: 70% versus 55% (p!0.001) lost less than threelines of visual acuity. Overall, 6% of treated eyesgained three or more lines of visual acuity in the0.3 mg dose versus 2% in the usual care group.Twenty-two percent of eyes gained one or more linesof vision in the 0.3 mg dose group versus 12% in thesham group. Angiographically, there was a slowing inthe rate of the CNV lesion growth, CNV size andleakage by 30 and 54 weeks. No antibodies weredetected against pegaptanib. Significant ocularadverse events in the Macugen treated eyes includedendophthalmitis in 1.3% patients, vitreous floaters in

33% and anterior chamber inflammation. The 0.3 mgdose of Macugen was approved for use for subfovealCNV due to AMD in 2005.

Exploratory subgroup analysis of the V.I.S.I.O.N.trial results showed Macugen was effective for allsubgroups and that no single subgroup drove theefficacy results. The use of PDT could have enhancedthe usual care group results and lessened thedifferences between treatment and usual care. Thiswas the first time that a CNV treatment was indepen-dent of lesion subtype. The benefit was maintained atyear 2 (120). Macugen has been shown to be quite safe,with low rates of endophthalmitis. The continuoustreatment group of eyes did better than the group ofeyes in which treatment was halted after one year(and then allowed to resume treatment if losing tenor more letters).

A subgroup analysis of early lesions was subse-quently performed (121). Two groups of early lesionswere identified: group 1 included small lesions lessthan two disc areas in size, with O54 letters, withoutprior therapy and without scarring or atrophy. Group2 included occult only lesions without lipid and withbetter visual acuity than the fellow eye. For these earlylesions, there was a better response to Macugentherapy. For group 1, 76% of treated eyes versus 50%of usual care eyes lost !15 letters (pZ0.03). Twelvepercent of treated and 4% of usual care eyes gained 15or more letters. For group 2 eyes, 80% of treated versus57% of usual care eyes lost !15 letters (pZ0.05).Twenty percent of treated eyes and 0% of usual careeyes gained 15 or more letters.

Another anti-VEGF treatment approach is thatof an antibody to VEGF. The VEGF antibody, rhuFAb(ranibizumab, Lucentis) is a humanized monoclonalantibody antigen-binding fragment (Fab) that binds toand neutralizes the biological activities of all knownhuman VEGF-A isoforms including its proteolyticcleavage products. Lucentis blocks all VEGF isoforms,unlike Macugen which is an aptamer that specificallybinds VEGF 165. Ranibizumab was given as an intra-vitreal injection monthly in the initial clinical trials.In these clinical trials, visual acuity was maintainedwithin three lines of baseline in over 90% of eyes.Visual acuity improved three or more lines in 33% to40% of patients (89,90). These one and two year studiesshowed the best visual acuity results to date for anyrandomized multicenter clinical trial on exudativeAMD (89,90).

The MARINA (Minimally Classic/Occult Trial ofAnti-VEGF Ranibizumab In Antibody the Treatmentof Neovascular AMD) study was a phase III, multi-centered, randomized trial comparing the efficacyof monthly intravitreal injections of ranibizumabcompared to sham injections in 716 patients withminimally classic or occult CNV. Eyes were


randomized equally to one of three groups: the 0.3 mgranibizumab, 0.5 mg of ranibizumab, or sham.Approximately 95% (0.3, 0.5 mg) of ranibizumab eyesversus 62% of control eyes maintained or lost less than15 letters from baseline (p!0.0001) at one year.Twenty-five percent of 0.3 mg ranibizumab dose eyesand 34% of 0.5 mg dose ranibizumab eyes versus 5% ofsham eyes gained three or more letters from baseline atmonth 12. The results were maintained at two years.Ninety percent of ranibizumab eyes compared with53%of control eyes lost less than 15 letters (p!0.0001) atyear 2 for the 0.5 mg dose. Thirty-three percent ofranibizumab eyes versus 3.8% of sham eyes gainedthree of more lines of visual acuity (p!0.0001) at year 2for the 0.5 mg dose. Ranibizumab eyes gained anaverage of 7.2 letters at year 1 and 6.6 letters at year 2versus control eyes which lost 10.4 letters at year 1 and14.9 letters at year 2 (89).Anatomic findings (lesion size,area of leakage, OCT thickness) were in favor ofranibizumab over sham.

The ANCHOR (ANti-VEGF Antibody for theTreatment of Predominantly Classic CHORoidalNeovascularization in AMD) study was a phase III,multicenter, randomized clinical trial comparingefficacy of monthly intravitreal injections of two dosesof ranibizumab (0.3 mg, 0.5 mg) with PDT in 423patients with predominantly classic CNV. Ninety-fourpercent (0.3 mg) and 96% (0.5 mg) of ranibizumab eyesversus 64% of PDT eyes lost less than 15 letters (p!0.0001). Forty percent of ranibizumab eyes versus 5.6%of PDT eyes improved at least 15 letters (p!0.0001)from baseline at one year. The average change was again of 11 letters for ranibizumab eyes versus a loss of9.5 letters for sham eyes at one year (90).

The PIER (A Phase IIIb, Multicenter, Random-ized, Double-Masked, Sham Injection-ControlledStudy of the Efficacy and Safety of Ranibizumab inSubjects with Subfoveal Choroidal Neovascularizationwith or without classic CNV Secondary to Age-Related Macular Degeneration) study was a phase IIIstudy that evaluated the safety and efficacy of intra-vitreal injections of two different doses (0.3 mg,0.5 mg) of ranibizumab administered monthly forthree doses and then every three months comparedwith sham for eyes with subfoveal CNV in 184patients. The one year results showed no gain invisual acuity gain from baseline in the treated groupsversus a loss of 16.3 letters in the sham group (p!0.0001). Ninety percent of the ranibizumab 0.5 mgdose lost less than 15 letters versus 49% of the shameyes (p!0.0001). There was no difference in thepercentage of eyes with improvement of 15 or moreletters: 15% of ranibizumab eyes versus 10% of shameyes (pZ0.71) (122). Based on these positive results,the 0.5 mg dose of Lucentis was approved for treat-ment of subfoveal CNV by the FDA in 2006.

The FOCUS study was a phase I/II random-ized, multicentered, single-masked, controlled studycomparing the safety and efficacy of monthly intra-vitreal injections of ranibizumab (0.5 mg dose) incombination with verteporfin PDT to PDT alone inpatients with predominantly classic CNV. PDT wasgiven at baseline and then on an as needed basisevery three months. Eyes receiving ranibizumab andPDT did better than the eyes receiving PDT alone.90.5% of combination eyes lost less than 15 lettersfrom baseline as compared with 68% of PDT eyes(pZ0.0003) (123). Approximately 24% of ranibi-zumab and PDT eyes gained 15 letters frombaselines as compared with 5% of PDT eyes (pZ0.0033).

In all of the ranibizumab studies, the rates ofserious adverse events were low. The incidenceof hypertension and thromboembolic events did notdiffer significantly between the ranibizumab treatedpatients and the sham or PDT groups. The incidenceof endophthalmitis was low between 1% and 2%.Ocular serious adverse events occurred in !0.1% ofintravitreal injections.

Bevacizumab (Avastinw) has recently been usedoff-label for treatment of CNV in AMD patients.Avastin is a full length anti-VEGF antibody, whichcontrasts with ranibizumab, which is a VEGF antibodyfragment specifically developed for intraocular use.Bevacizumab is FDA approved for use in metastaticcolorectal cancer and has significant systemic sideeffects, including hypertension and increased throm-boembolic events, when given intravenously in cancerpatients (124,125). In the open label Systemic Avastinfor Neovascular AMD (SANA) trial, patients withprogressive visual loss who were ineligible for PDTwere given intravenous bevacizumab (126,127). Basedon the risks of bevacizumab therapy, patients wereexcluded if they had uncontrolled hypertension,a history of thromboembolic events, current anti-coagulation therapy, or proteinuria, or if electivesurgery was planned within three months. Patientswere followed weekly initially and then monthly.Significant improvements in visual acuity anddecreased retinal thickness on OCT were seen. Ten ofthe 18 patients required medication or adjustmentof existing medication for systemic hypertension.These systemic side effects led to the exploratoryuse of intravitreal injection of bevacizumab to treatCNV in AMD patients.

Intravitreal use of bevacizumab involves both anoff-label application of the drug and an alternativeroute of drug delivery. Rapid visual acuity improve-ment and decreased retinal thickness (126–129) haveled to widespread use in the retinal community.The main force driving intravitreal bevacizumabusage is the high percentage of patients that

146 LIM AND TSONG

experience symptomatic relief from active subfovealCNV. Currently, the National Eye Institute (NEI) issponsoring the Comparison of Treatment Trial (CATT)study, which will directly compare the efficacy andsafety of bevacizumab to ranibizumab in monthly andalternate dosing regimens in patients with subfovealCNV in AMD eyes.

The current treatment paradigm using thesemedical treatments is illustrated in Figure 21. Treat-ments with proven efficacy are shown in bold.

ThermotherapyOther treatments for subfoveal exudative AMDinclude thermotherapy, radiation therapy, otherforms of antiangiogenesis therapy, submacularsurgery, submacular surgery with RPE transplantationsurgery and translocation surgery. Some of thesealternative therapies have shown no benefit whereasothers, such as antiangiogenesis treatments, haveshown great visual acuity improvement in eyeswith CNV.

Subfoveal

Extrafoveal:Laser PhotocoagulationPhotodynamic Therapy

Anti-VEGF Drug ?

PredominantlyClassic:

Photodynamic TherapyLucentis, Macugen

Avastin

Occult withNo Classic or

Minimally Classic

No:Photodynamic Therapy

Lucentis, MacugenAvastin

Yes, < 4 MPS Disc Areas

Yes:Visual Acuity < 20/50

Photodynamic TherapyLucentis, Macugen

Avastin

No:Lucentis, Macugen

Avastin

Recent Disease Progression?

Lesion Size < 4 MPS Disc Areas?

Visual Acuity < 20/50?

Lesion Location?

Type of Leakage?

ChoroidalNeovascularizationSecondary to AMD

Juxtafoveal:Laser PhotocoagulationPhotodynamic Therapy

Anti-VEGF Drug ?

Yes, RecentDisease Progression

No:Observation

Figure 21 Flowchart illustrating treatment options for choroidal neovascular lesions in age-related macular degeneration (AMD). BoldZefficacy proven in phase 3 clinical trials.


Thermotherapy (TTT) is the treatment by whicha modified diode laser is used to deliver heat to thechoroid and RPE through the pupil. Prior success hasbeen demonstrated for treatment of small choroidalmelanoma and retinoblastoma with this method(130,131). Reichel and colleagues (132) performeda pilot study in which 16 eyes of 15 patients withoccult CNV were treated with TTT. Over a meanfollow-up time of 12 months, 19% improved by twoor more lines of visual acuity, 56% remained the same,and 25% lost two ormore lines of visual acuity. Ninety-four percent showed decreased exudation despite thevisual acuity results. Reichel and coworkers have sinceperformed a multicenter, double-masked, placebo-controlled trial, the TTT4CNV trial. The TTT4CNVtrial enrolled 303 patients with small (%3 mm) occultsubfoveal CNV with visual acuities ranging from20/50 to 20/400. Patients randomized to treatmentreceived 800 mW TTT over 60 seconds using a 3 mmspot size from the Iris OcuLight SLx 810 nm laser andthe large spot size slit lamp adapter. Compared toplacebo eyes, treated eyes did not show a beneficialeffect on prevention of moderate visual loss at oneyear. However, for the subgroup of eyes (41% atbaseline) with 20/100 or worse visual acuity at base-line, 23% of treated eyes gained one or more lines ofvisual acuity and 14% gained three of more lines ofvisual acuity at one year. No further studies areplanned using TTT (133).

RadiationRadiation therapy has been advocated as a therapyfor exudative AMD. Low dose radiation inhibitsneovascularization (134–137). The key factor in theuse of radiation therapy is achieving a balancebetween destruction of abnormal CNV tissue andpreservation of normal retinal and choroidal bloodvessels (138). Since proliferating tissues are moreradiosensitive, this balance is theoretically achievable.Conflicting data about the efficacy and morbidity ofradiation therapy has led to the Age-related MacularDegeneration Radiation Trial (AMDRT) an NEI-spon-sored pilot study comparing observation to radiation.The AMDRT enrolled patients with lesions not amen-able to laser treatment (classic, occult or mixed). TheAMDRT found no clinically significant differencebetween eyes assigned to radiation and those toobservation (138). At 6 months, 9 radiated eyes (26%)and 17 observed eyes (49%) lost R3 lines of visualacuity [pZ0.04; stratified chi-square test]. At 12months, 13 radiated eyes (42%) and 9 observed eyes(49%) lost R3 visual acuity lines (pZ0.60). Theradiated group demonstrated smaller lesions andless fibrosis than the nonradiated group (pZ0.05 and0.004, respectively) at 12 months. Experimental work

continues in this area. More information on the use ofradiation for treatment of CNV in AMD can be foundin the chapter by Flaxel and Finger and in that ofChong and Scartozzi.

Surgical TreatmentsPilot studies evaluating submacular surgery for CNVin AMD patients showed some promise and led to thedefinitive Submacular Surgery Trails (SSTs). The SSTwas comprised of four studies, which included a pilottrial. One of the arms of the SST study, the recurrentCNV arm, showed no difference between surgeryor laser treatment and no further study was rec-ommended (139). The other three arms included newsubfoveal CNVassociated with AMD (Group N), largesubfoveal hemorrhage associated with AMD (GroupB) and subfoveal histoplasmosis CNV and idiopathicCNV (Group H). Long-term visual outcomes andrecurrence rates after submacular surgery comparedsimilarly with the natural history of untreated CNV.Thus, the SSTs did not show a benefit for removal ofCNV and a modest benefit for subretinal hemorrhageevacuation.

Some groups are combining submacular surgerywith RPE transplantation. Since the RPE is oftenremoved during submacular surgery or since RPEatrophy often follows submacular surgical procedures,researchers are evaluating the efficacy of transplantingRPE cells to repopulate the RPE layer. Loss of the RPEleads to choriocapillaris loss. The details of this tech-nique and the rationale for RPE transplantation isgiven in detail in Del Priore and colleagues’ chapterin this book.

Retinal translocation and limited macular trans-location surgery (140–146) have been described fortreatment of subfoveal CNV. The rationale behindthese surgical techniques is to move the macular areafrom the underlying CNV to a healthier RPE environ-ment. The underlying CNV is thus moved relative tothe foveal center and can be treated with conventionallaser or surgically removed. Limited macular translo-cation surgery (141–142) and 360 degree translocationsurgery (143–146) is discussed in detail in the chapterby Au Eong and colleagues. In the current era of anti-VEGF therapy, translocation surgery is used less oftenthan in the past.

Emerging TreatmentsNewer mechanisms of antiangiogenesis treatmentscontinue to be developed. Small interfering (shortinhibitory) RNA technology (147–150) targetedagainst VEGF is being evaluated. VEGF trap hasbeen evaluated in phase I and II studies (151). Non-RNA inhibitors of VEGF receptor tyrosine kinaseactivity are in development (152,153). Tubulin-binding

148 LIM AND TSONG

agents, such as combretastatin A-4 phosphate, are inearly clinical trials (154). Some emerging drugs indevelopment have recently been pulled from thedevelopmental pipeline because the efficacy is muchlower than that of anti-VEGF treatments. This hasoccurred with squalamine lactate (155–158). Excitingwork with gene therapy using adenoviral vectors(159–164) is in progress.

Small Interfering RNAIn 1998, Fire and Mello (147) discovered that injectionof gene-specific double stranded RNA into cellsresults in potent silencing of that gene’s expression.This RNA interference is one of the fundamentalmechanisms by which a cell regulates gene expressionand protects itself against viral infection. (Fire andMello were awarded the Nobel Prize in Physiologyand Medicine for 2006.)

Double-stranded RNA binds to a proteincomplex called a Dicer. The Dicer cleaves the double-stranded RNA into multiple smaller fragments.A second protein complex called RNA-inducedsilencing complex (RISC) then binds these RNA frag-ments and eliminates one of the strands. Theremaining strand stays bound to RISC, and serves asa probe that recognizes the corresponding messengerRNA transcript in the cell. When the RISC complexfinds a complementary messenger RNA transcript, thetranscript is cleaved and degraded, thus silencing thatgene’s expression.

Reich and Tolentino (148,149) used a small inter-fering RNA (siRNA) inhibitor of VEGF designed forintravitreal injection for treatment of CNV. A phase I,open-label, dose escalation study of 15 patientsrevealed no serious ocular or systemic adverse effectsat a dose up to 3.0 mg. The drug was later namedbevasiranib/Cand5 (Acuity Pharmaceuticals, Phila-delphia, Pennsylvania, U.S.A.) in the phase II trial,known as the CARE study (Cand5 Anti-VEGF RNAiEvaluation). In this multicenter, randomized, double-masked, trial of bevasiranib/Cand5 in patients withCNV secondary to AMD (Brucker et al., Retina SocietyMeeting, Cape Town, October 2006. Thompson et al.,AAO Meeting Las Vegas, November 2006), 127patients with predominantly classic, minimallyclassic, or RAP lesions (occult no classic lesionsexcluded) were randomized to receive one of threedoses of the drug (0.2, 1.5, and 3.0 mg) at baseline andat six weeks. The primary endpoint was the meanchange in Early Treatment Diabetic Retinopathy Study(ETDRS) visual acuity from baseline at 12 weeks. Thedrug was found to be safe. At 12 weeks, mean changein ETDRS visual acuity was K4 letters (0.2 mg), K7letters (1.5 mg), and K6 letters (3.0 mg) for the drugdoses respectively. The authors theorized that thedisappointing visual results resulted from the fact

that bevasiranib/Cand5 only blocks the productionof new VEGF and not existing VEGF. They postulatedthat a baseline combination treatment with a VEGFblocker may be required to “mop up” the preexistingVEGF load. However, the half-life of VEGF is shortand it does not explain why the results were notseen by the 12 weeks time point. Further workremains to be done to determine efficacy for theproposed treatment combination.

Other therapeutic targets for siRNA are beinginvestigated. siRNA directed against the vascularendothelial growth factor receptor-1 (VEGFR-1) haveshownpromise in amousemodel of CNV (150), and arecurrently in clinical development (Sirna-027—SirnaTherapeutics, Boulder).

VEGF TrapVEGF Trap (Regeneron, Tarrytown, New York, U.S.A.)is an experimental new drug designed to inhibit allmembers of the VEGF family: VEGF-A, -B, -C, -D, andplacental growth factor-1 and -2. VEGF Trap is arecombinant, chimeric, VEGF receptor fusion protein.The binding domains of VEGFR-1 and -2 are combinedwith the Fc portion of immunoglobulin G to create astable, soluble, high-affinity inhibitor. VEGF trap bindsVEGF-A with a higher affinity (Kd !1 pmol/L) thancurrently available anti-VEGF drugs (151). Whetherthe broader spectrum and higher affinity of VEGF Trapequates to improved efficacy in the treatment of CNVsecondary to AMD is yet to be determined.

The CLEAR-AMD 1 Study is a randomized,multicenter, placebo-controlled, dose-escalationstudy designed to assess the safety, tolerability andbioactivity of VEGF Trap (151). In this study, 25 AMDpatients with subfoveal CNV with lesions less than12 disc areas in size and more than 50% active leakagewere enrolled. Patients were randomized to receiveeither placebo or one of three doses of VEGF Trap(0.3-, 1.0-, or 3.0-mg/kg) as a single IV dose, followedby a four week observation period. Three additionaldoses were given two weeks apart. Dose-limitingtoxicity was observed in two of five patients treatedwith the 3.0 mg/kg dose: one patient developedgrade 4 hypertension and the other developed grade2 proteinuria. The maximum tolerated IV dose ofVEGF Trap was 1.0 mg/kg. Reduced leakage on FAand reduced retinal thickening on OCTwere observedin the treated patients. No corresponding reductionin CNV lesion size or improvement in visual acuitywas observed in these patients over the short 71 daystudy period.

The CLEAR-IT 1 Study is now underway andwill assess the safety, tolerability and bioactivity ofVEGF Trap through the intravitreal route of adminis-tration (Nguyen et al., abstract, Retina Society Meeting,Cape Town, October 2006). The study has enrolled


21 patients using the same inclusion criteria asCLEAR-AMD 1, and randomized them to receiveone of six doses of VEGF Trap as single intravitrealinjection: 0.05, 0.15, 0.5, 1.0, 2.0, or 4.0 mg. After43 days of follow-up, no adverse ocular or systemicevents were observed. Mean decrease in excess fovealthickness for all patients was 72%. The mean increasein ETDRS visual acuity was 4.75 letters and visualacuity remained stable or improved in 95% of patients.Notably, three out of six patients treated withthe higher doses (2.0 or 4.0 mg) gained R3 lines ofvisual acuity by day 43. VEGF Trap shows promise as anovel treatment for CNV in AMD patients.

Receptor Tyrosine Kinase InhibitorsNon-RNA inhibitors of VEGF receptor tyrosine kinaseactivity havebeen identified. The anti-angiogenic prop-erties are being investigated for use in the treatmentof systemicmalignancy, aswell asCNV.One advantageof this class of drugs is the possibility of an oral route ofadministration, thereby avoiding the ocular com-plications associated with intravitreal injections.

One promising compound is PTK787, which is anon-selective inhibitor of all known VEGF receptors.PTK787 has been shown to inhibit retinal neovascular-ization in a hypoxic mouse model (152,153). Phase I/IIclinical trials of PTK787 (Vatalanib, Novartis, EastHanover), have been done in patients with both solidand hematologic malignancies. A multicenter phase Itrial of PTK787/Vatalanib in patients with maculardegeneration—the ADVANCE study—is currentlyenrolling patients. Patients with all CNV lesion typeswill receive PDT with Visudyne at baseline, andrandomized to receive concurrent treatment witheither 500 or 1000 mg of oral PTK787/Vatalanib, orplacebo, once daily for three months (Joondeph,abstract, Retina Society Meeting, Cape Town, October2006). ADVANCE is designed to assess the safety andefficacy of the drug.

AG-013958 (Pfizer, San Diego, California, U.S.A.)is a selective VEGFR and platelet-derived growthfactor receptor inhibitor that is currently in Phase Iand II clinical testing. The drug is administered as asubtenons injection. Preliminary results of 21 patientswith subfoveal CNV indicated that adverse eventswere mild (165).

Squalamine LactateSqualamine lactate (Evizon, Genaera, PlymouthMeeting, Pennsylvania, U.S.A.) is another antiangio-genic compound which was investigated as a potentialtreatment of CNV. It is a chemically-synthesizedaminosterol which was originally isolated from theliver of the dogfish shark Squalus acanthias (155).It is thought to inhibit angiogenesis by acting onthe sodium-hydrogen antiporter sodium-proton

exchangers (specifically the NHE3 isoform) to inhibitendothelial cell proliferation. It works intracellularlyand binds calmodulin. It inhibits VEGF signaling andintegrin expression, and reverses cytoskeletal forma-tion. These effects result in endothelial inactivationand apoptosis. It has the greatest effect on new vesselformation with no appreciable effect on unstimulatedendothelial cells (156). In addition to its antiangiogenicproperties, it also has direct antitumor effects and iscurrently being studied in human clinical trials as atreatment for advanced cancers.

In rats with laser-induced trauma, systemicsqualaminewas shown to reduce histological evidenceof CNV development (157). In a small, four-monthstudy of forty humans with AMD-associated CNV,squalamine was found to improve mean visualacuity in ten (26%) of the subjects by R3 lines(approximately R15 letters). Vision was stable in74% of subjects (%G2 lines) (158). A subsequentstudy is investigating the use of monthly infusions ofsqualamine lactate in combination with PDT. Threedifferent doses of squalamine are being investigated.Preliminary safety data on the 46 subjects showedno drug-related serious adverse events during thefirst 29 weeks. However, of the ten “probablyrelated” adverse events, there was one retinal detach-ment and one incidence of prolonged prothrombintime, and eight infusion site reactions. The interimresults of this study have led to a decision to haltfurther investigation as the efficacy does not near thatseen with the anti-VEGF drug Lucentis treatment forAMD-associated CNV.

Anti-VEGF treatment has enabled a sizeableproportion of treated patients to attain significantvisual improvement or to maintain vision. Futureresearch will hopefully continue to build on theseadvances and make restoration of vision a reality forthe majority of these patients.

Tubulin Binding AgentsVascular targeting agents are also being investigatedas treatments for CNV. These agents have shownefficacy in causing tumor regression. CombretastatinA-4 is a naturally occurring agent that binds tubulinand causes necrosis and shrinkage of tumors bydamaging their blood vessels. A CA-4 prodrug,combretastatin A-4-phosphate (CA-4-P), has beentested in two models of ocular neovascularization.CA-4-P is a novel agent that bind tubulin and causesendothelial cells which are normally flat to becomeround, resulting in narrowing of the lumen andcessation of blood flow in the vessels (160). It isonly effective on newly formed blood vessels whichhave no actin and are therefore susceptible to tubulinstructure disruption. CA-4-P increases endothelial cellpermeability, while inhibiting endothelial cell

150 LIM AND TSONG

migration and capillary tube formation predomi-nantly through disruption of vascular endothelial-cadherin/beta-catenin/Akt signaling pathway,thereby leading to rapid vascular collapse andtumor necrosis (161).

Nambu and coworkers quantitatively assessedthe effect of CA-4-P to suppress CNVin transgenicmicewith overexpression of VEGF in the retina (rho/VEGFmice) andmicewith CNVdue to laser-induced ruptureof Bruch’smembrane. CA-4-P suppressed the develop-ment of VEGF-induced neovascularization in theretina. CA-4-P blocked development and promotedregression of CNV. Therefore, CA-4-P shows potentialfor both prevention and treatment of ocular neovascu-larization (162).

Gene TherapyGene therapy approaches include delivery of adeno-viral vectors contained antiangiogenic proteinssuch as PEDF or RNA that can attenuate VEGF. Useof adenoviral vectors to continuously deliver theprotein would eliminate the need for multiple injec-tions. One approach is to use short hairpin RNA(shRNA) that could attenuate VEGF as a potentialtherapy for AMD (163). Cashman and colleaguesdeveloped several shRNAs from recombinant adeno-virus. The investigators found potent shRNAsequences that were able to silence VEGF in humanRPE cells by 94% at a 1:5 molar ratio (VEGF to shRNA)and 64% at a 1:0.05 molar ratio. Co-injection of VEGF-expressing viruses into mice with shRNA targetingVEGF led to a substantial (84%) reduction in CNV.shRNA may hold promise as a therapy for AMD.

Campochiaro and colleagues have studied theuse of human PEDF to treat CNV. PEDF is one of themost potent known antiangiogenic proteins found inhumans. Campochiaro and colleagues use adenovirusadsorbed PEDF, Ad(GV)PEDF.11D. cDNA for PEDF isthe transgene on E1-, partial E3-, E4- deleted replica-tion-deficient, adenovirus serotype 5, the gene transfervector. The natural blood retinal barrier limits theability of Ad(GV)PEDF.11D to affect tissues otherthan in the eye. Intravitreal administration ofAd(GV)PEDF.11D is a convenient means of deliveringPEDF “factories” to the relevant cells within the eye,thus resulting in local PEDF production. In threemurine disease models (the laser-induced CNVmodel, the VEGF transgenic model, and the retino-pathy of prematurity model) significant inhibition ofneovascularization (up to 85%) was shown with dosesof Ad(GV)PEDF vectors ranging from 1!108 to1!109 particle units (PUs). Toxicology studies inCynomolgus monkeys showed a dose-related inflam-matory response to Ad(GV)PEDF. A dose of 1!108 PUcaused no adverse effects, while the inflammatory

response observed at 1!109 PU was minimal andfully reversible. Higher doses produced increasinglysevere inflammatory responses (164).

An open-label, dose-escalation, phase I studyinvestigated the safety, tolerability and potentialactivity of intravitreal injection of Ad(GV)PEDF.11Din patients with advanced AMD and CNV with visualacuity 20/200 or worse (165). Twenty-eight patientsreceived a single intravitreal injection of Adenovectorpigment epithelium-derived factor 11(AdPEDF.11), atdoses ranging from 106 to 109.5 PU. No serious adverseevents related to AdPEDF.11 and no dose-limitingtoxicities were found. Signs of mild, transient intra-ocular inflammation occurred in 25% of patients, butthere was no severe inflammation. All adenoviralcultures were negative. Six patients had increasedintraocular pressures, which were controlled withtopical medications. At three and six months afterinjection, 55% and 50%, respectively, of patientstreated with 106–107.5 PU and 94% and 71% of patientstreated with 108–109.5 PU had no change or improve-ment in lesion size from baseline. The median increasein lesion size at 6 and 12 months was 0.5 and 1.0 diskareas in the low-dose group compared with 0 and 0disk areas in the high-dose group. These data suggestthe possibility of antiangiogenic activity that may lastfor several months after a single intravitreal injectionof doses greater than 108 PU of AdPEDF.11. This studyprovided evidence that adenoviral vector-mediatedocular gene transfer is a viable approach for thetreatment of ocular disorders. Further studies investi-gating the efficacy of AdPEDF.11 in AMDpatients withCNV are planned.

Low VisionFor patients in whom visual acuity is impaired and notreatment is possible or for whom no treatmentpossibilities remain, visual rehabilitation is of utmostimportance. Low vision rehabilitation may help thesepatients best utilize their remaining visual acuity andteach them to utilize ancillary tools such as closedcircuit television and magnifiers. These patientsshould also be reminded that AMD affects centraland not peripheral vision. Expectations of the magni-tude benefit of low vision rehabilitation should berealistically explained to the patient. Further detailsof the devices and the services available for the lowvision patient are detailed in the chapters by Primo.

SUMMARY

The era of improvement of visual acuity as a goal inthe treatment of AMD patients with CNV has arrivedwith the advent of the anti-angiogenesis agents.Ultimately, prevention of CNV must be our goal in


order to prevent visual loss in patients with AMD.Some experimental approaches have included laserto drusen, which did not prove beneficial. At present,the anecortave acetate risk reduction trial (AART) isin progress. AART is a prospective study in whicheyes at high risk for CNV are randomized either toanecortave acetate (delivered via a juxtascleral injec-tion) or observation. The Age-Related Eye DiseaseStudy 2 (AREDS 2) will explore nutritional supple-mentation for prevention of CNV in high risk AMDeyes. (Further details of these studies can be found inthe non-exudative AMD chapter by Bhagat andFlaxel). Perhaps gene therapy will play an importantrole in the future as discussed further in the chapter byChao and colleagues.

The next decade will undoubtedly usher in evenmore exciting developments in our battle againstexudative (neovascular) AMD. Basic science researchinto the pathophysiology of AMD and CNV hasresulted in, and hopefully will continue to result in,translational discoveries for preventive and innovativetargeted treatments against CNV in patients withAMD. Clinical trials investigating the efficacy andsafety of these new treatments will continue to deci-pher useful from non-useful therapy. (The role ofclinical studies is further elaborated in the chapterby Walonker and Diddie.) Perhaps, a combinationapproach using two or more of the following: anti-angiogenesis agents, anti-inflammatory therapy, genetherapy and basement membrane stabilizers may oneday be preferred treatments for CNV. It is evenpossible that such agents may some day be used asprophylactic treatment in high risk eyes. Althoughgreat progress has been achieved in the last decade,much remains to be done in the battle against visualacuity loss from exudative AMD.

SUMMARY OF MAIN POINTS

& Exudative form of AMD is the major cause ofvisual blindness in patients with AMD.

& Systemic risk factors associated with CNV includeincreased age, Caucasian race, smoking.

& Ocular risk factors associated with increased risk ofCNV include large drusen (O5), confluent drusen,hyperpigmentation and hypertension.

& The simplified AREDS scale predicts the risk ofCNV over the next 5 years and 10 based upon thepresence of drusen and pigment abnormalities ineach eye.

& Symptoms may be absent in the presence of CNV.Amsler grid testing and PHP may help to detectproblems earlier.

& Prompt evaluation of symptomatic patients is essen-tial for preventing visual loss due to CNV.

& FA is used to characterize the location (extrafoveal,juxtafoveal, subfoveal), type of CNV (classic, occult-FVPED, LLUS) and to ascertain treatment effects.

& SignsofCNVinclude subretinalfluid,hardexudates,subretinal hemorrhage or intraretinal hemorrhage,pigmented subretinal lesions, and subRPE fluid.

& The MPS showed a beneficial effect of laser photo-coagulation for extrafoveal or juxtafoveal CNV(classic or well-demarcated forms). Persistent andrecurrent CNV risk is high for extrafoveal andjuxtafoveal CNV treated with thermal laser. Sub-foveal laser should not be used in the current era ofeffective non-ablative therapy.

& ICG is useful for evaluating eyes with occult CNVorPEDs, or subretinal hemorrhage.

& Occult CNV has a better natural history than that ofclassic CNV.

& OCTimaging isuseful for thedetectionof intraretinalcystic changes, subretinal fluid, sub-RPE fluid andfor monitoring CNV treatment responses.

& PDT is useful for: (i) eyes with subfoveal CNV thatare at least 50% or more classic in composition and(ii) for eyes with subfoveal minimally classic CNV(less than six disc areas) and eyes with occult CNVwith visual acuity less than 20/50 and lesion size lessthan four MPS disc areas.

& Antiangiogenesis treatments cannowresult invisualacuity improvements in eyes with subfoveal CNV.Ranibizumab (Lucentis) can results in visual acuityimprovement in about one-third of patients withnew onset subfoveal CNV. Avastin is being usedoff-label. The CATT will further investigate theefficacy and safety of Avastin and compare Avastinto Lucentis.

& Thermotherapydidnot preventmoderate visual lossfor treated eyes versus placebo in the TTT4CNVstudy. A subgroup of eyes with 20/100 or worsevisual acuity, however, showed some visual benefitcompared to placebo.

& Radiation therapy was not useful in the AMDRT.Other methods radiation therapy (plaque, betaparticle) are still being investigated.

& Clinical trials investigating agents to prevent CNVinhigh risk eyes are in progress (AREDS 2, AART).

& Several new avenues of treatment for CNV includeVEGF receptor inhibitors, tubulin binding agents,RNA interference and other gene therapy.

REFERENCES

1. Pagenstecher H, Genth CP. Atlas Der PatholischenAntomie Des Augapfels. Wiesbaden: CW Kreiden, 1875.

2. Oeller J. Atlas Seltener Ophthalmoscopischer Bufunde.Wiesbaden: Bergmann JF, 1905.

3. Junius P, Kuhnt H. Die Scheibenformige Entartung DerNetzhautmitte (Degeneratio Maculae Luteae Disciformis).Berlin: Karger, 1926.

152 LIM AND TSONG

4. Holloway TB, Verhoeff FH. Disk-like degeneration of themacula. Trans Am Ophthalmol Soc 1928; 26:206.

5. Verhoeff FH, Grossman HP. Pathogenesis of disciformdegeneration of the macula. Arch Ophthalmol 1937;35:262–94.

6. Ashton N, Sorsby A. Fundus dystrophy with unusualfeatures: a histological study. Br J Ophthalmol 1951; 35:751.

7. Gass JDM. Pathogenesis of disciform detachment of theneuroepithelium. III. Senile disciform macular degener-ation. Am J Ophthalmol 1967; 63:617.

8. Gass JDM. Pathogenesis of disciform detachment of theneuroepithelium. IV. Fluorescein angiographic study ofsenile disciform macular degeneration. Am J Ophthalmol1967; 63:645.

9. Blair CJ, Aaberg TM. Massive subretinal exudation associ-ated with senile macular degeneration. Am J Ophthalmol1971; 71:639–48.

10. Small ML, GreenWR, Alpar JJ, Drewry RE. Senile maculardegeneration: a clinicopathologic correlation of two caseswith neovascularization beneath the retinal pigment epi-thelium. Arch Ophthalmol 1976; 94:601–7.


12. Ferris FL, Fine SL, Hyman L. Age-related maculardegeneration and blindness due to neovascular maculo-pathy. Arch Ophthalmol 1984; 102:1640–2.

13. Leibowitz HM, Krueger DE, Maunder LR, et al. TheFramingham Eye Study monograph: an ophthalmologicaland epidemiological study of cataract, glaucoma, diabeticretinopathy, macular degeneration, and visual acuity in ageneral population of 2631 adults, 1973–1975. SurvOphthalmol 1980; 24:335–610.

14. Hyman LG, Lilienfeld AM, Ferris FL, III, Fine SL. Senilemacular degeneration: a case-control study. AmJ Epidemiol 1983; 118:213–27.

15. Friedman DS, O’Colmain BJ, Munoz B, et al. Prevelance ofage-related macular degeneration in the United States.Arch Ophthalmol 2004; 122:564–72.


17. Gregor Z, Joffe L. Senile macular changes in the blackAfrican. Br J Ophthalmol 1978; 62:547–50.

18. Friedman DS, Katz J, Bressler NM, Rahmani B,Tielsch JM. Racial differences in the prevalence of age-related macular degeneration. Ophthalmology 1999;106:1049–55.

19. Leske MC, Wu SY, Hennis A, et al. Nine-year incidence ofage-related macular degeneration in the Barbados EyeStudies. Ophthalmology 2006; 113:29–35.

20. Klein R, Rowland ML, Harris MI. Racial/ethnicdifferences in age-related maculopathy: third NationalHealth and Nutrition Examination Survey. Ophthal-mology 1995; 102:371–81.

21. Varma R, Fraser-Bell S, Tan S, Klein R, Azen SP, LosAngeles Latino Eye Study Group. Prevalence of age-related macular degeneration in Latinos; the Los AngelesLatino Eye Study. Ophthalmology 2004; 111:1288–97.

22. Munoz B, Klein R, Rodriguez J, Snyder R, West SK.Prevalence of age-related macular degeneration in apopulation-based sample of Hispanic people in Arizona:Proyecto VER. Arch Ophthalmol 2005; 123:1575–80.

23. Klein R, Klein BEK, Knudson MD, et al. Prevalence of age-related macular degeneration in 4 racial/ethnic groups inthe multi-ethnic study of atherosclerosis. Ophthalmology2006; 113:373–80.

24. Smith W, Mitchell P. Family history and age-relatedmaculopathy: the Blue Mountain Eye Study. Aust NZJ Ophthalmol 1998; 26:203–6.

25. Klaver CCW, Wolfs RCW, Assink JJM, Duijn CM,Hofman A, deJong PT. Genetic risk of age-related maculo-pathy: population-based familial aggregation study. ArchOphthalmol 1998; 116:1646–51.

26. Klein R, Klein BEK, Jensen SC, Mares-Perlman JA,Cruickshanks KJ, Palta M. Age-related maculopathy in amultiracial United States population: the National Healthand Nutrition Examination Survey III. Ophthalmology1999; 106:1056–65.

27. Klein R, Klein BEK, Linton KLP, Demets DL. The BeaverDam Eye Study: the relation of age-relatedmaculopathy tosmoking. Am J Epidemiol 1993; 137:190–200.

28. Smith W, Mitchell P, Leeder SR. Smoking and age-relatedmaculopathy: the Blue Mountain Eye Study. ArchOphthalmol 1996; 114:1518–23.

29. Delacourt C, Diaz JL, Ponton-Sanchez A, Papoz L.Smoking and age-related macular degeneration. ThePola Study. Pathologies oculaires liees a l’age. ArchOphthalmol 1998; 116:1031–5.

30. Khan JC, Thurlby DA, Shahid H, et al. Smoking and agerelated macular degeneration: the number of pack yearsof cigarette smoking is a major determinant of risk forboth geographic atrophy and choroidal neovascularisa-tion. Br J Ophthalmol 2006; 90:75–80.

31. Espinosa-Heidmann DG, Suner IJ, Catanuto P,Hernandez EP, Marin-Castano ME, Cousins SW. Cigarettesmoke-related oxidants and the development of sub-RPEdeposits in an experimental animal model of dry AMD.Invest Ophthalmol Vis Sci 2006; 47:729–37.

32. Taylor HR, West SK, Munoz B, Rosenthal FS, Bressler SB,Bressler NM. The long-term effects of visible light on theeye. Arch Ophthalmol 1992; 110:99–104.

33. Cruickshanks KJ, Klein R, Klein BEK. Sunlight and age-related macular degeneration: the Beaver Dam Eye Study.Arch Ophthalmol 1993; 111:514–8.

34. The Eye Disease Case-Control Study Group (EDCCS). Riskfactors for neovascular age-related macular degeneration.Arch Ophthalmol 1992; 110:1701–8.

35. Darzins P, Mitchell P, Heller RF. Sun exposure and age-related macular degeneration: an Australian case-controlstudy. Ophthalmology 1997; 104:770–6.

36. Klein R, Klein BEK, Jensen SC, Cruickshanks KJ. Therelationship of ocular factors to the incidence and pro-gression of age-related maculopathy. Arch Ophthalmol1998; 116(4):506–13.

37. Pollack A, Marcovich A, Bukelman A, Oliver M. Age-related macular degeneration after extracapsular cataractextraction with intraocular lens implantation. Ophthal-mology 1996; 103:1546–54.

38. Pollack A, Marcovich A, Bukelman A, Zalish M, Oliver M.Development of exudative age-related macular degener-ation after cataract surgery. Eye 1997; 11:523–30.

39. Lanchoney DM, Maguire MG, Fine SL. A model of theincidence and consequences of choroidal neovasculariza-tion secondary to age-related macular degeneration:comparative effects of current treatment and potentialprophylaxis on visual outcomes in high-risk patients.Arch Ophthalmol 1998; 116:1045–52.

40. Holz FG, Wolfensberger TJ, Piguet B, et al. Bilateralmacular drusen in age-related macular degeneration:prognosis and risk factors. Ophthalmology 1994;101:1522–8.

41. Macular Photocoagulation Study Group. Risk factors forchoroidal neovascularization in the second eye of patients


with juxtafoveal or subfoveal choroidal neovasculariza-tion secondary to age-related macular degeneration. ArchOphthalmol 1997; 115:741–7.

42. Ferris FL, Davis MD, Clemons TE, et al. A simplifiedseverity scale for AMD: AREDS Report No. 18. ArchOphthalmol 2005; 123:1570–4.

43. Age-Related Eye Disease Study Research Group. Arandomized, placebo-controlled, clinical trial of high-dose supplementation with vitamins C and E, betacar-otene, and zinc for age-related macular degeneration andvision loss. AREDS Report No. 8. Arch Ophthalmol 2001;119:1417–36.

44. Christen WG, Gaziano JM, Hennekens CH. Design ofPhysicians’ Health Study II—a randomized trial of beta-carotene, vitamins E and C, and multivitamins, in preven-tion of cancer, cardiovascular disease, and eye disease, andreview of results of completed trials. Ann Epidemiol 2000;10:125–34.

45. Moeller SM,ParekhN,TinkerL, et al.Associations betweenintermediate age-related macular degeneration and luteinand zeaxanthin in the carotenoids in age-related eyedisease study (CAREDS). Ancillary study of the women’shealth initiative. Arch Ophthalmol 2006; 124:1151–62.

46. Grossniklaus HE, Miskala PH, Green WR, et al. Histo-pathologic and ultrastructural features of surgicallyexcised subfoveal choroidal neovascular lesions: subma-cular surgery trials Report No. 7. Arch Ophthalmol 2005;123:914–21.

47. Grossniklaus HE, Wilson DJ, Bressler SB, et al. Clinico-pathologic studies of eyes that were obtained postmortemfrom four patients who were enrolled in the submacularsurgery trials: Report No. 16. Am J Ophthalmol 2006;141:93–104.

48. Klein RJ, Zeiss C, Chew EY, Tsai JY, Sackler RS, Haynes C,Henning AK, SanGiovanni JP, Mane SM, Mayne ST, et al.Complement factor H polymorphism in age-relatedmacular degeneration. Science 2005; 308:385–9.

49. Edwards AO, Ritter R, III, Abel KJ, Manning A,Panhuysen C, Farrer LA. Complement factor H poly-morphism and age-related macular degeneration.Science 2005; 308:421–4.

50. Haines JL, Hauser MA, Schmidt S, et al. Complementfactor H variant increases the risk of age-related maculardegeneration. Science 2005; 308:419–21.

51. Hageman GS, Anderson DH, Johnson LV, Hancox LS,Taiber AJ, Hardisty LI, Hageman JL, Stockman HA,Borchardt JD, Gehrs M, et al. A common haplotype inthe complement regulatory gene factor H (HF1/CFH)predisposes individuals to age-related macular degener-ation. Proc Natl Acad Sci 2005; 102:7227–32.

52. Mousa SA, Lorelli W, Campochiaro PA. Extracellularmatrix-integrin binding modulates secretion of angiogenicgrowth factors by retinal pigmented epithelial cells. J CellBiochem 1999; 74:135–43.

53. Bressler NM, Bressler SB, Gragoudas ES. Clinical charac-teristics of choroidal neovascular membranes. ArchOphthalmol 1987; 105:209–13.

54. Bressler NM, Bressler SB, Fine SL. Age-related maculardegeneration. Surv Ophthalmol 1988; 32:375–413.

55. Moisseiev J, Bressler NM. Asymptomatic neovascularmembranes in the second eye of patients with visual lossfrom age-related macular degeneration. Invest Ophthal-mol Vis Sci 1990; 31:462.

56. Alster Y, Bressler NM, Bressler SB, et al. Preferentialhyperacuity perimeter (PreView PHP) for detecting chor-oidal neovascularization study. Ophthalmology 2005;112:1758–65.

57. Goldstein M, Loewenstein A, Barak A, et al. Results of amulticenter clinical trial to evaluate the preferentialhyperacuity perimeter for detection of age-relatedmacular degeneration. Retina 2005; 25:296–303.

58. Loewenstein A, Malach R, Goldstein M, et al. Replacingthe Amsler grid: a new method for monitoring patientswith age-related macular degeneration. Ophthalmology2003; 110:966–70.

59. Valencia M, Green RL, Lopez PF. Echographic findings inhemorrhagic disciform lesions. Ophthalmology 1994;101:1379–83.

60. Fine AM, Elman MJ, Ebert JE, Prestia PA, Starr JS, Fine SL.Earliest symptoms caused by neovascular membranes inthe macula. Arch Ophthalmol 1986; 104:513–4.

61. Nazemi PP, Fink W, Lim JI, Sadun AA. Electronic Amslergrid scotomas of age-related macular degenerationdetected and characterized by means of a novel three-dimensional computer-automated visual field test. Retina2005; 25:446–53.

62. Srinivasan VJ, Wojtkowski M, Witkin AJ, et al. High-definition and 3-dimensional imaging of macularpathologies with high-speed ultrahigh-resolution opticalcoherence tomography. Ophthalmology 2006; 113:2054–65.

63. Shakoor A, Shahidi M, Blair NP, Gieser JL, Zelkha R.Macular thickness mapping in AMD. Retina 2006;26:44–8.

64. Lim JI, Aaberg TM, Sr., Capone A, Jr., Sternberg P, Jr.Indocyanine green angiography-guided photocoagulationof choroidal neovascularization associated with retinalpigment epithelial detachment. Am J Ophthalmol 1997;123:524–32.

65. Guyer DR, Yannuzzi LA, Slakter JS, Sorenson JA, Hope-Ross M, Orlock DR. Digital indocyanine green videoan-giography of occult choroidal neovascularization.Ophthalmology 1994; 101:1727–37.

66. Slakter JS, Yannuzzi LA, Sorenson JA, Guyer DR, Ho AC,Orlock D. A pilot study of indocyanine green videoangio-graphy-guided laser photocoagulation of occult choroidalneovascularization in age-related macular degeneration.Arch Ophthalmol 1994; 112:465–72.

67. Elman MJ, Fine SL, Murphy RP, Patz A, Auer C. Thenatural history of serous retinal pigment epitheliumdetachment in patients with age-related macular degener-ation. Ophthalmology 1986; 93:224–30.

68. Poliner LA, Olk RJ, Burgess D, Gordon ME. Naturalhistory of retinal pigment epithelial detachments in age-related macular degeneration. Ophthalmology 1986;93:543–51.

69. Bird AC, Marshall J. Retinal pigment epithelial detach-ments in the elderly. Trans Ophthalmol Soc UK 1986;105:674–82.

70. Casswell AG, Kohen D, Bird AC. Retinal pigment epi-thelial detachments in the elderly: classification andoutcome. Br J Ophthalmol 1985; 69:397–403.

71. Dhalla MS, Blinder KJ, Tewari A, Hariprasad SM, Apte RS.Retinal pigment epithelial tear following intra-vitreal pegaptanib sodium. Am J Ophthalmol 2006;141(4):752–4.

72. Macular Photocoagulation Study Group. Subfovealneovascular lesions in age-related macular degeneration:guidelines for evaluation and treatment in the macularphotocoagulation study. Arch Ophthalmol 1991; 109:1242–57.

73. Lim JI, Sternberg P, Jr., Capone A, Jr., Aaberg TM, Sr.,Gilman JP. Selective use of indocyanine green angio-graphy for occult choroidal neovascularization. AmJ Ophthalmol 1995; 120:75–82.

154 LIM AND TSONG

74. Heier JS, Antoszyk AN, Pavan PR, et al. Ranibizumabfor treatment of neovascular age-related macular degener-ation: a phase I/II multicenter, controlled, multidosedegeneration: a phase I/II multicenter, controlled, multi-dose study. Ophthalmology 2006; 113:633–42.

75. Green WR, Enger C. Age-related macular degeneration:histopathologic studies. The 1992 Lorenz E Zimmermanlecture. Ophthalmology 1993; 100:1519–35.

76. Shiraga F, Ojima Y, Matsuo T, Takasu I, Matsui N. Feedervessel photocoagulation of subfoveal choroidal neovascu-larization secondary to age-related macular degeneration.Ophthalmology 1998; 105:662–9.

77. Staurenghi G, Orzalesi N, La Capria A, Aschero M.Laser treatment of feeder vessels in subfoveal choroidalneovascular membranes. A revisitation using dynamicindocyanine green angiography. Ophthalmology 1998;2297–305.

78. Flower RW. Optimizing treatment of choroidal neovascu-larization feeder vessels associated with age-relatedmacular degeneration. Am J Ophthalmol 2002; 134:228–39.

79. Ryan SJ. Subretinal neovascularization: natural history ofan experimentalmodel. ArchOphthalmol 1982; 100:1804–9.

80. Miller H, Miller B, Ryan SJ. The role of the retinalpigmented epithelium in the involution of subretinalneovascularization. Invest Ophthalmol Vis Sci 1986;27:1644–52.

81. Amin R, Pulkin JE, Frank RN. Growth factor localizationin choroidal neovascular membranes of age-relatedmacular degeneration. Invest Ophthalmol Vis Sci 1994;35:3178–88.

82. Frank RN, Amin RH, Eliott D, Puklin JE, Abrams GW.Basic fibroblast growth factor and vascular endothelialgrowth factor are present in epiretinal and choroidalneovascularization membranes. Am J Ophthalmol 1996;122:393–403.

83. Lopez PF, Sippy BD, Lambert HM, Thach AB, Hinton DR.Transdifferentiated retinal pigment epithelial cells areimmunoreactive for vascular endothelial growth factorin surgically excised age-related macular degenerationchoroidal neovascular membranes. Invest OphthalmolVis Sci 1996; 37:855–68.

84. Yamada H, Yamada E, Kwak N, et al. Cell injury unmasksa latent proangiogenic phenotype in mice with increasedexpression of FGF2 in the retina. J Cell Physiol 2000;185:135–42.

85. Nagai N, Oike Y, Izumi K, et al. Angiotensin II type 1receptor-mediated inflammation is required for choroidalneovascularization. Arterioscler Thromb Vasc Biol 2006;26:2252–9.

86. Seo M-S, Kwak N, Ozaki H, et al. Dramatic inhibition ofretinal and choroidal neovascularization by oral adminis-tration of a kinase inhibitor. Am J Pathol 1999; 154:1743–53.

87. Kwak N, Okamoto N, Wood JM, Campochiaro PA. VEGFis an important stimulator in a model of choroidalneovascularization. Invest Ophthalmol Vis Sci 2000;41:3158–64.

88. Gragoudas ES, Adamis AP, Cunningham ET, Feinsod M,Guyer DR, VEGF Inhibition Study in Ocular Neovascular-ization Clinical Trial Group. Pegaptanib for neovascularage-related macular degeneration. N Engl J Med 2004;351(27):2805–16.

89. Rosenfeld P, Brown DM, Heier J, et al. Ranibizumab forneovascular age-related macular degeneration. N EnglJ Med 2006; 355:1419–31.

90. Brown DM, Kaiser PK, Michels M, et al. Ranibizumabversus verteporfin for neovascular age-related maculardegeneration. N Engl J Med 2006; 355:1432–44.

91. Yannuzzi LA, Wong DWK, Sforzolini BS, et al. Polypoidalchoroidal vasculopathy and neovascularized age-relatedmacular degeneration. Arch Ophthalmol 1999; 117:1503–10.

92. Lafaut BA, Leys AM, Snyders B, Rasquin F, DeLaey JJ.Polypoidal choroidal vasculopathy in Caucasians. GraefesArch Clin Exp Ophthalmol 2000; 238:752–9.

93. Latfaut BA, Aisenbrey S, Broeck CV, Bartz-Schmidt KU.Clinicopathological correlation of deep retinal vascularanomalous complex in age-related macular degeneration.Br J Ophthalmol 2000; 84:1269–74.

94. Yannuzzi LA, Negrao S, Iida T, et al. Retinal angiomatousproliferation in age-related macular degeneration. Retina2001; 21:416–34.

95. Macular Photocoagulation Study Group. Argon laserphotocoagulation for senile macular degeneration.Results of a randomized clinical trial. Arch Ophthalmol1982; 100:912–8.

96. Grey RHB, Bird AC, Chisholm IH. Senile disciformmacular degeneration: features indicating suitability forphotocoagulation. Br J Ophthalmol 1979; 63:85–9.

97. Macular Photocoagulation Study Group. Krypton laserphotocoagulation for neovascular lesion of age-relatedmacular degeneration. Results of a randomized clinicaltrial. Arch Ophthalmol 1990; 108:816–24.

98. Macular Photocoagulation Study Group. Laser photocoa-gulation for juxtafoveal choroidal neovascularization.Five-year results from randomized clinical trials. ArchOphthalmol 1994; 112:500–9.

99. Guyer DR, Fine SL, Maguire MG, Hawkins BS, Owens SL,Murphy RP. Subfoveal choroidal neovascular membranesin age-related macular degeneration. Visual prognosis ineyes with relatively good initial visual acuity. ArchOphthalmol 1986; 104:702–5.

100. Freund KB, Yannuzzi LA, Sorenson JA. Age-relatedmacular degeneration and choroidal neovascularization.Am J Ophthalmol 1993; 115:786–91.

101. Macular Photocoagulation Study Group. Argon laserphotocoagulation for neovascular maculopathy: five-yearresults from randomized clinical trials. Arch Ophthalmol1991; 109:1109–14.

102. Macular Photocoagulation Study Group. Recurrentchoroidal neovascularization after argon laser photocoa-gulation for neovascular maculopathy. Arch Ophthalmol1986; 104:503–12.

103. Macular Photocoagulation Study Group. Persistent andrecurrent neovascularization after krypton laser photo-coagulation for neovascular lesions of age-relatedmacular degeneration. Arch Ophthalmol 1990; 108:825–31.

104. Bressler SB, Bressler NM, Fine SL, et al. Natural course ofchoroidal neovascular membranes within the foveal avas-cular zone in senile macular degeneration. AmJ Ophthalmol 1982; 93:157–63.

105. Jalkh AE, Avila MP, Trempe CL, Schepens CL. Manage-ment of choroidal neovascularization within the fovealavascular zone in senile macular degeneration. AmJ Ophthalmol 1983; 95:818–25.

106. Macular Photocoagulation Study Group. Laser photocoa-gulation of subfoveal neovascular lesions in age-relatedmacular degeneration. Results of a randomized clinicaltrial. Arch Ophthalmol 1991; 109:1220–31.

107. Macular Photocoagulation Study Group. Occult choroidalneovascularization. Influence on visual outcome inpatients with age-related macular degeneration. ArchOphthalmol 1996; 114:400–12.


108. Bressler NM, Frost LA, Bressler SB, Murphy RP, Fine SL.Natural course of poorly defined choroidal neovascular-ization associated with macular degeneration. ArchOphthalmol 1988; 106:1537–42.

109. Soubrane G, Coscas G, Francais C, Koenig F. Occult subret-inal new vessels in age-related macular degeneration.Natural history and early laser treatment. Ophthalmology1990; 97:649–57.

110. Bressler NM, Maguire MG, Murphy PL, et al. Macularscatter (“grid”) laser treatment of poorly demarcatedsubfoveal choroidal neovascularization in age-relatedmacular degeneration. Arch Ophthalmol 1996; 114:1456–64.

111. Treatment of Age-related Macular Degeneration withPhotodynamic Therapy (TAP) Study Group Principal Inves-tigators. Photodynamic therapy of subfoveal choroidalneovascularization in age-related macular degenerationwith verteporfin. One-year results of 2 randomized clinicaltrials—TAP Report 1. Arch Ophthalmol 1999; 117:1329–45.

112. Treatment of Age-Related Macular Degeneration withPhotodynamic Therapy (TAP) Study Group. Photody-namic therapy of subfoveal choroidal neovascularizationin age-related macular degeneration with verteporfin.Two-year results of 2 randomized clinical trials—TAPReport 2. Arch Ophthalmol 2001; 119:198–207.

113. Verteporfin Roundtable 2000 Participants, Treatmentof Age-related Macular Degeneration with Photo-dynamic Therapy (TAP) Study Group PrincipalInvestigators, and Verteporfin in Photodynamic therapy(VIP) Study Group Principal Investigators. Guidelines forusing verteporfin (Visudynee) in photodynamic therapyto treat choroidal neovascularization due to age-relatedmacular degeneration and other causes. Retina 2002;22:6–18.

114. Verteporfin In Photodynamic Therapy (VIP) Study Group.Verteporfin therapy of subfoveal choroidal neovascular-ization in age-related macular degeneration: two-yearresults of a randomized clinical trial including lesionswith occult with no classic choroidal neovascularization–verteporfin in photodynamic therapy report 2. AmJ Ophthalmol 2001; 131:541–60.

115. VIO Study Group (in press).116. Azab M, Boyer DS, Bressler NM, et al. Visudyne in

Minimally Classic Choroidal Neovascularization StudyGroup. Verteporfin therapy of subfoveal minimallyclassic choroidal neovascularization in age-relatedmacular degeneration: 2-year results of a randomizedclinical trial. Arch Ophthalmol 2005; 123:448–57.

117. Spaide RF, Sorenson J, Maranan L. Combined photody-namic therapy with verteporfin and intravitrealtriamcinolone acetonide for choroidal neovascularization.Ophthalmology 2003; 110:1517–25.

118. Rechtman E, Danis RP, Pratt LM, Harris A. Intravitrealtriamcinolone with photodynamic therapy for subfovealchoroidal neovascularization in age related maculardegeneration. Br J Ophthalmol 2004; 88:344–7.

119. Ergun E, Maar N, Ansari-Shahrezaei S, et al. Photo-dynamic therapy with verteporfin and intravitrealtriamcinolone acetonide in the treatment of neovascularage-related macular degeneration. Am J Ophthalmol 2006;142:10–6.

120. VEGF Inhibition Study in Ocular Neovascularization(V.I.S.I.O.N.) Clinical Trial Group. Year 2 efficacy resultsof 2 randomized controlled clinical trials of pegaptanib forneovascular age-related macular degeneration. Ophthal-mology 2006; 113:1508–21.

121. The VEGF Inhibition Study in Ocular Neovascularization(V.I.S.I.O.N) Clinical Trial Group. Enhanced efficacy

associated with early treatment of neovascular age-related macular degeneration with pegaptanib sodium:an exploratory analysis. Retina 2005; 25:815–27.

122. Regillo CD, Brown DM, Abraham H, Kaiser PK,MielerWF. Randomized, double-masked, sham-controlledtrial of ranibizumab for neovascular age-related maculardegeneration: PIER study year 1. Am J Ophthalmol 2007(in press).

123. Heier JS, BoyerDS,Ciulla TA, et al. Ranibizumab combinedwith verteporfin photodynamic therapy in neovascularage-related macular degeneration: year 1 results of theFOCUS Study. Arch Ophthalmol 2006; 124:1532–42.

124. Ferrara N, Hillan KJ, Novotny W. Bevacizumab(Avastin), a humanized anti-VEGF monoclonal antibodyfor cancer therapy. Biochem Biophys Res Commun 2005;333:328–35.

125. Salesi N, Bossone G, Veltri E, et al. Clinical experience withbevacizumab in colorectal cancer. Anticancer Res 2005;25:3619–23.

126. Michels S, Rosenfeld PJ, Puliafito CA, Marcus EN,Venkatraman AS. Systemic bevacizumab (Avastin)therapy for neovascular age-related macular degenerationtwelve-week results of an uncontrolled open-label clinicalstudy. Ophthalmology 2005; 112:1035–47.

127. Moshfeghi AA, Rosenfeld P J, Puliafito CA, et al. Systemicbevacizumab (Avastin) therapy for neovascular age-related macular degeneration: twenty-four-week resultsof an uncontrolled open-label clinical study. Ophthal-mology 2006; 113:2002–11.

128. Rosenfeld PJ, Moshfeghi AA, Puliafito CA. Optical coher-ence tomography findings after an intravitreal injection ofbevacizumab (Avastin) for neovascular age-relatedmacular degeneration. Ophthalmic Surg Lasers Imaging2005; 36:331–5.

129. Rich RM, Rosenfeld PJ, Puliafito CA, et al. Short-termsafety and efficacy of intravitreal bevacizumab (Avastin)for neovascular age-related macular degeneration. Retina2006; 26:495–511.

130. Godfrey DG, Waldron RG, Capone A, Jr. Transpupillarythermotherapy for small choroidal melanoma. AmJ Ophthalmol 1999; 128:88–93.

131. Shields CL, Santos MC, Diniz W, et al. Thermotherapy forretinoblastoma. Arch of Ophthalmol 1999; 117:885–93.

132. Reichel E, Berrocal AM, Ip M, et al. Transpupillarythermotherapy of occult subfoveal choroidal neovascular-ization in patients with age-related macular degeneration.Ophthalmology 1999; 106:1908–14.

133. Reichel E, Musch DC, Blodi BA, Mainster MA,TTT4CNV Study Group. Results from the TTT4CNVClinical Trial. Invest Ophthalmol Vis Sci 2005; 46(E-abstract 2311).

134. Archambeau JO, Mao XW, Yonemoto LT, et al. What is therole of radiation in the treatment of subfoveal membranes:review of radiobiologic, pathologic, and other consider-ations to initiate a multimodality discussion. Int J RadiatOncol Biol Phys 1998; 40:1125–36.

135. Del Gowin RL, Lewis JW, Hoak JC, Mueller AL,Gibson DP. Radiosensitivity of human endothelial cellsin culture. J Lab Clin Med 1974; 84:42–8.

136. Hosoi Y, Yamamoto M, Ono T, Sakamoto K. Prostacyclinproduction in cultured endothelial cells is highly sensitiveto low doses of ionizing radiation. Int J Radiat Biol 1993;63:631–8.

137. Sagerman RH, Chung CT, Alterti WE. Radiosensitivityof ocular and orbital structures. In: Alberti WE,Sagerman RH, eds. Radiotherapy of Intraocular andOrbital Tumors. Berlin: Springer, 1993:375–85.

156 LIM AND TSONG

138. Marcus DM, Peskin E, Maguire M, et al. The age-relatedmacular degeneration radiotherapy trial (AMDRT): oneyear results from a pilot study. Am J Ophthalmol 2004;138:818–28.

139. Submacular Surgery Trials Pilot Study Investigators.Submacular surgery trials randomized pilot trial of laserphotocoagulation versus surgery for recurrent choroidalneovascularization secondary to age-related maculardegeneration: I. ophthalmic outcomes. Am J Ophthalmol2000; 130(4):387–407.

140. de Juan E, Jr., Loewenstein A, Bressler NM, Alexander J.Translocation of the retina for management of subfoveal-choroidal neovascularization II: a preliminary report inhumans. Am J Ophthalmol 1998; 125:635–46.

141. Fujii GY, Pieramici D, Humayun MS, et al. Complicationsassociated with limited macular translocation. AmJ Ophthalmol 2000; 130:751–62.

142. Lewis H, Kaiser PK, Lewis S, Estafanous M. Maculartranslocation for subfoveal choroidal neovascularizationin age-related macular degeneration: a prospective study.Am J Ophthalmol 1999; 128:135–46.

143. Pertile G, Claes C. Macular translocation with 360 degreeretinotomy for management of age-related maculardegeneration with subfoveal choroidal neovasculariza-tion. Am J Ophthalmol 2002; 134(4):560–5.

144. Mruthyunjaya P, Stinnett SS, Toth CA. Change in visualfunction after macular translocation with 360 retinectomyfor neovascular age-related macular degeneration.Ophthalmology 2004; 111(9):1715–24.

145. Fujikado T, Asonuma S, Ohji M, et al. Reading abilityafter macular translocation surgery with 360-degree reti-notomy. Am J Ophthalmol 2002; 134(6):849–56.

146. Toth CA, Lapolice DJ, Banks AD, Stinnett SS. Improve-ment in near visual function after macular translocationsurgery with 360-degree peripheral retinectomy. GraefesArch Clin Exp Ophthalmol 2004; 242(2):541–8.

147. Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE,Mello CC. Potent and specific genetic interference bydouble-stranded RNA in Caenorhabditis elegans. Nature1998; 391:806–11.

148. Reich SJ, Fosnot J, Kuroki A, et al. Small interferingRNA (siRNA) targeting VEGF effectively inhibits ocularneovascularization in a mouse model. Mol Vis 2003;9:210–6.

149. Tolentino MJ, Brucker AJ, Fosnot J, et al. Intravitrealinjection of vascular endothelial growth factor smallinterfering RNA inhibits growth and leakage in a non-human primate, laser-induced model of choroidalneovascularization. Retina 2004; 24:132–8.

150. Shen J, Samul R, Silva RL, et al. Suppression of ocularneovascularization with siRNA targeting VEGF receptor1. Gene Ther 2006; 13:225–34.

151. Nguyen QD, Shah SM, Hafiz G, et al. A phase I trial ofan IV-administered vascular endothelial growth factortrap for treatment in patients with choroidal neovas-cularization due to age-related macular degeneration.Ophthalmology 2006; 113:1522–32, 33.

152. Maier P, Unsoeld AS, Junker B, et al. Intravitreal injectionof specific receptor tyrosine kinase inhibitor PTK787/ZK222 584 improves ischemia-induced retinopathy inmice. Graefes Arch Clin Exp Ophthalmol 2005; 243:593–6.

153. Ozaki H, Seo MS, Ozaki K, et al. Blockade of vascularendothelial cell growth factor receptor signaling is suf-ficient to completely prevent retinal neovascularization.Am J Pathol 2000; 156:697–707.

154. Nambu H, Nambu R, Melia M, Campochiaro PA. Com-bretastatin A-4 phosphate suppresses development andinduces regression of choroidal neovascularization. InvestOphthalmol Vis Sci 2003; 44:3650–5.

155. Moore KS, Wehrli S, Roder H, et al. Squalamine: anaminosterol antibiotic from the shark. Proc Natl Acad SciUSA 1993; 90:1354–8.

156. Sills AK, Jr., Williams JI, Tyler BM, et al. Squalamineinhibits angiogenesis and solid tumor growth in vivoand perturbs embryonic vasculature. Cancer Res 1998;58:2784–92.

157. Ciulla TA, CriswellMH,Danis RP,Williams JI,McLaneMP,Holrovd KJ. Squalamine lactate reduces choroidal neovas-cularization in a laser-injury model in the rat. Retina 2003;23:808–14.

158. Garcia CA, Connolly B, Thomas E, et al. A phase 2multi-dose pharmacokinetic study of MSI-1256F (squa-lamine lactate) for the treatment of subfoveal choroidalneovascularization associated with age-related maculardegeneration (AMD). Invest Ophthalmol Vis Sci 2005;46:E-Abstract 206.

159. Cashman SM, Bowman L, Christofferson J, Kumar-Singh R. Inhibition of choroidal neovascularization byadenovirus-mediated delivery of short hairpin RNAstargeting VEGF as a potential therapy for AMD. InvestOphthalmol Vis Sci 2006; 47:3496–504.

160. West CM, Price P. Combretastatin A4 phosphate. Antic-ancer Drugs 2004; 15:179–87.

161. Vincent L, Kermani P, Young LM, et al. CombretastatinA4 phosphate induces rapid regression of tumor neo-vessels and growth through interference with vascularendothelial-cadherin signaling. J Clin Invest 2005; 115:2992–3006.

162. Mori K, Gelbach P, Ando A, McVey D, Wei L,Campochiaro PA. Regression of ocular neovascularizationin response to increased expression of pigment epithelium-derived factor. Invest Ophthalmol Vis Sci 2002; 43:2428–34.

163. Rasmussen H, Chu KW, Campochiaro P, et al. Clinicalprotocol. An open-label, phase I, single administration,dose-escalation study of ADGVPEDF.11D (ADPEDF) inneovascular age-related macular degeneration (AMD).Hum Gene Ther 2001; 2:2029–32.

164. Campochiaro PA, Nguyen QD, Shah SM, et al. Adenoviralvector-delivered pigment epithelium-derived factor forneovascular age-related macular degeneration: results of aphase I clinical trial. Hum Gene Ther 2006; 17:177–9.

165. Kaiser PK. Antivascular endothelial growth factor agentsand their development: therapeutic implications in oculardiseases. Am J Ophthalmol 2006; 142:660–8.


Part III: Imaging Techniques for the Clinical Evaluationof Age-Related Macular Degeneration

9

Indocyanine Green Angiography in Age-RelatedMacular DegenerationScott C. N. OliverDepartment of Ophthalmology, Rocky Mountain Lions Eye Institute, University of Colorado School

of Medicine, Aurora, Colorado, U.S.A.

Antonio P. CiardellaDepartment of Ophthalmology, Denver Health Hospital Authority, Denver, Colorado, U.S.A.

Daniela C. A. C. Ferrara, Jason S. Slakter, and Lawrence A. YannuzziThe LuEsther T. Mertz Retinal Research Department, Manhattan Eye, Ear, and Throat Hospital,

New York, New York, U.S.A.

INTRODUCTION

Fluorescein angiography (FA) revolutionized the diag-nosis of retinal disorders (1,2). However, there arecertain limitations to this technique. Overlying hemor-rhage, pigment, or serosanguineous fluid can blockunderlying pathologic changes and prevent adequatevisualization by FA. Indocyanine green (ICG) is a Foodand Drug Administration-approved tricarbocyaninedye that has several advantageous properties oversodium fluorescein as a dye for ophthalmic angio-graphy. The clinical usefulness of indocyanine greenangiography (ICGA) in the past has been limited byour inability to produce high-resolution images.However, enhanced high-resolution ICG angiogramscan now be obtained owing to the technologicaladvance of coupling digital imaging systems to ICGcameras (3,4). Thus, digital ICGA finally allows thetheoretical advantages of ICG as an ophthalmic dye tobe realized.

SPECIAL PROPERTIES OF ICG

The ICG absorbs and fluoresces in the near-infraredrange. Owing to the special characteristics of the dye,there is less blockage by the normal eye pigments,which allows enhanced imaging of the choroid andchoroidal abnormalities. For example, Geeraets andBerry (5) have reported that the retinal pigmentepithelium (RPE) and choroid absorbs 59% to 75%of blue–green (500 nm) light, but only 21% to 38%of near-infrared (800 nm) light. The activity of ICG inthe near-infrared range also allows visualization of

pathologic conditions through overlying hemorrhage,serous fluid, lipid, and pigment that may blockstructures by FA. This property allows enhancedimaging of occult choroidal neovascularization(CNV) and pigment epithelial detachment (PED) inage-related macular degeneration (AMD) (4,6).

A second special property of ICG is that it ishighly protein-bound (98%). Therefore, less dyeescapes from the choroidal vasculature, which allowsenhanced imaging of choroidal abnormalities.

HISTORICAL PERSPECTIVES

ICG dye was first used in medicine in the mid-1950s attheMayoClinic to obtain bloodflowmeasurements (7).In 1956, ICG was used for determining cardiac outputand characterizing cardiac valvular and septal defects.In 1964, studies of systemic arteriovenous fistulasand renal blood flow were reported. The findingthat exclusively the liver excreted the dye soon led todevelopment of its application for measuring hepaticfunction. Recently, the use of real-time intraoperativeICGA provided information about vessel patencyduring neurosurgical aneurysm repair (8,9).

ICG first became attractive to ophthalmologistsinterested in better ways to image the choroidalcirculation because of its safety and its particularoptical and biophysical properties. Kogure and cowor-kers (10) in 1970 first performed choroidal absorptionangiography in monkeys, using intraarterial ICGinjection. The first ICG angiogram in a human wasperformed by David (11) during carotid angiography.

In 1971, Hochheimer (12) described choroidal absorp-tion angiography in cats using intravenous ICGinjections and black-and-white infrared film insteadof color film. One year later, Flower and Hochheimerperformed intravenous absorption ICGA for the firsttime in a human (13,14). These same investigators thendescribed the superior technique of ICG fluorescenceangiography (15,16). Further technological improve-ments followed (17), and, in 1985, Bischoff and Flower(18) reported on their 10-year experience with ICGA,which included 500 angiograms of various disorders.

However, the sensitivity of infrared film wastoo low to adequately capture the low-intensity ICGfluorescence, as the fluorescence efficacy of ICG isonly 4% of that of sodium fluorescein. The resolutionof ICGA was improved in the mid-1980s by Hayashiand coworkers, who developed improved filtercombinations and described ICG videoangiography(19–21). However, their video system lacked freeze-frame image recording and possessed cumulative lighttoxicity potential due to its 300-W continuous halogenlamp illumination. In 1985, Destro and Puliafito (22)described a similar video system. In 1989, Scheiderand Schroedel (23) reported the use of the scanninglaser ophthalmoscope for ICG videoangiography;refinements of their technique allowed for improvedimaging of choroidal neovascular membranes (24,25).

In 1992, Guyer and coworkers (3) and Yannuzziand associates (4) introduced the use of a 1024-linedigital imaging system to produce high-resolutionenhanced ICG images. These systems have improvedthe resolution of ICGA such that this technique is nowof practical clinical value.

PHARMACOLOGY

ICG is a sterile, water-soluble tricarbocyaninedye, which is anhdyro-3,3,3 0,3 0-tetramethyl-1-1 0-di-(4-sulfobutyl)-4,5,4 0,5-dibenzoindotricarbocyaninehydroxide sodium salt. Its empirical formula isC43H47N2NaO6S2 and its molecular weight is775 Daltons (26). It is highly protein-bound (98%).Although it has been thought that ICG is primarilybound to albumin in the serum (27), 80% of ICG inthe blood is actually bound to globulins, such as A1-lipoproteins (28).

ICGs spectral absorption is between 790 and805 nm (28–30). The dye is excreted by the liver viabile. ICG is not reabsorbed from the liver, is notdetected in cerebrospinal fluid (31,32) and does notcross the placenta (33).

TOXICITY

ICG is a relatively safe dye, with only a few side effectsreported in clinical use (7,27,34–36). In our experience,

it is safer than sodium fluorescein. In contrast to FA,nausea and vomiting are extremely uncommon duringICG angiography. We have observed two seriousvasovagal-type reactions during ICGA.

No complications were reported in one studyusing intravenous ICG doses of 150 to 200 mg. Noside effects were noted in another series of 700procedures (18). In a study 1226 consecutive patientsundergoing ICGA, there were three (0.15%) mildadverse reactions, four (0.2%) moderate reactions,one (0.05%) severe reaction, and no deaths (36).

ICGA should not be performed on patientsallergic to iodide, since it contains approximately5% iodide by weight. In addition, it should not beperformed on patients who are uremic (18) or whohave impaired hepatic clearance. Appropriate emer-gency equipment should be readily available, aswith FA.

TECHNIQUE OF INJECTION

ICGA can be performed immediately before orafter FA. We inject intravenously 25 to 50 mg ofICG (Cardio-Green: Hynson, Westcott & DunningProducts, Cockeysville, Maryland, U.S.A.) which hasbeen diluted in the aqueous solvent supplied by themanufacturer. Rapid injection is essential and shouldbe followed by a 5-mL normal saline flush. For wideangle angiography, the dosage is increased to 75 mg.

Bindewald and associates (37) recently testedthe lower limits fluorescein and ICG dye doses forangiography. Using a confocal scanning laser ophthal-moscope (cSLO) (Heidelberg retina angiograph 2,Heidelberg Engineering, Dossenheim, Germany),they found that a fluorescein dose as low as 166 mg,and an ICG dose as low as 5 mg, allowed adequateresolution for diagnosis and management of neovas-cular AMD. Resolution was impaired, however, in latephase images, compared to standard doses.

DIGITAL IMAGING SYSTEMS

The coupling of a digital imaging system with anICG camera allows production of enhanced high-res-olution (1024-line) images, which are necessary forICGA. The instantaneous images from these systemsproduce images which decrease patient waiting timeand expedite treatment. Digital imaging systems alsoallow image archiving, hard-copy generation, anddirect qualitative comparison between fluoresceinand ICGA findings. These systems are useful forplanning preoperative treatment strategies and formonitoring the adequacy of treatment postoperatively.

Imaging systems contain film, video, or digitalcameras with special antireflective coatings andappropriate excitatory and barrier filters. Flash

160 OLIVER ET AL.

synchronization allows high-resolution image capture.The digitally charged coupling device camera capturesthe digitized images and transmits them to a digitalimaging workstation. These images are captured atone frame per second, stored in buffer memory, anddisplayed on a high-contrast, high-resolution videomonitor. The images can be printed to photographsor slides, transferred via a variety of storage media, ornetworked to other stations in treatment areas and inother offices.

INTERPRETATION OF ICGA FINDINGS IN AMD

DefinitionsThe terminology used to describe the angiographicmanifestations of AMD corresponds, with certainexceptions described below, to definitions previouslyreported by the Macular Photocoagulation StudyGroup (38). Most relevant to the interpretation ofICGA in AMD are the definitions of serous pigmentepithelium detachment (SPED), vascularized pigmentepithelial detachment (VPED), classic CNV, and occultCNV (4,19,22,39).

Serous Pigment Epithelial DetachmentThe SPED is an ovoid or circular detachment of theRPE. On FA study there is rapid filling with dye of thefluid in the subRPE space. This corresponds to earlyhyperfluorescence beneath the PED, which increasesin intensity in the late phase of the study resulting in abright and homogeneous well-demarcated pattern.ICGA reveals a variable, minimal blockage of normalchoroidal vessels, more evident in the mid-phase ofthe angiogram. Thus, a SPED is bright (hyperfluor-escent) on FA and dark (hypofluorescent) on ICG. Thisdifference is caused by the fact that ICG molecules arelarger and almost completely bound to plasmaproteins, which prevents free passage of ICG dyethroughout the fenestrated choriocapillaris in thesubRPE space. Also, it is important to remember thedifference of appearance on ICGA between a SPED inAMD and a SPED in central serous chorioretinopathy(CSC). In fact, in CSC there is increased permeability ofthe choriocapillaris that causes leakage of ICGmolecules under the PED. As a result, a SPED inCSC appears bright (hyperfluorescent) with ICGA.Approximately 1.5% of newly diagnosed patientswith exudative AMD present with a pure SPED.

Choroidal NeovascularizationCNV is defined as a choroidal capillary proliferationthrough a break in the outer aspect of Bruch’smembrane under the RPE and/or the neurosensoryretina. CNV is divided into classic and occult based onthe FA angiography appearance.

Classic CNVClassic CNV is an area of bright, fairly uniformhyperfluorescence identified in the early phase of theFA. The fluorescence increases through the transitphase with leakage of dye obscuring the boundariesof this area by the late phase of the angiogram. WithICGA, a classic CNV has a similar appearance to thatseen with FA angiography, but is usually less welldelineated (Fig. 1) and exhibits little or no leakage inthe late phases of the ICG study. Only 12% of newlydiagnosed patients with exudative AMD present withclassic CNV.

Occult CNVOccult CNV is characterized as either fibrovascularpigment epithelial detachment (FVPED) or lateleakage of undetermined source (LLUS). FVPEDconsists of irregular elevation of the RPE consistingof stippled hyperfluorescence not as bright or discreteas classic CNV within one to two minutes afterfluorescein injection, with persistence of fluorescence10 minutes after injection. LLUS consists of areas ofleakage at the level of the RPE in the late phase of theangiogram not corresponding to an area of classicCNV or FVPED discernible in the early or middlephase of the angiogram to account for the leakage.Also any area of blocked fluorescence contiguous tothe CNV is considered occult CNV. More than 85%of newly diagnosed patients with exudative AMDpresent with occult CNV (Fig. 2). Two main types ofoccult CNV are recognized on ICGA.

Figure 1 Classic choroidal neovascularization. Early phase

indocyanine green angiogram shows a well-defined hyperfluor-escent vascular network consistent with a classic choroidal

neovascular membrane.

9: INDOCYANINE GREEN ANGIOGRAPHY IN AGE-RELATED MACULAR DEGENERATION 161

Without SPED. The first type of occult CNV iscaused by subRPE CNV that is not associated with aPED. The early stages of FA study reveal minimalsubretinal hyperfluorescence of undetermined sourcethat slowly increases over a period of several minutesto produce an irregular staining of the subRPE tissue.The ICG angiogram reveals early vascular hyperfluor-escence and late staining of the abnormal vessels. Ifthe ICG angiographic image has distinct margins, it isconsidered to be a well-defined CNV on ICGA. Two-third of newly diagnosed patients with an occult CNVpresent without an associated SPED.

With SPED. The second type of occult CNV isassociated with a SPED of at least 1-disc diameter insize. Combined CNV and SPED are called a VPED.This lesion is the result of subRPE neovascularizationassociated with a serous detachment of the RPE. One-third of newly diagnosed patients with AMD havean associated SPED. The determination of whether aSPED is present is best made on the basis of the FAstudy. FA may also demonstrate occult vessels as late,indistinct, subretinal hyperfluorescence beneath, or atthe margin of the SPED. ICGA reveals early vascularhyperfluorescence and late staining of the CNV. TheSPED, as noted previously, is comparatively hypo-fluorescent, because only minimal ICG leakage occursbeneath the serous detachment. ICG is more helpfulthan FA in differentiating between a SPED and aVPED. It also permits better identification of thevascularized and serous component of VPEDs. Thesedifferentiations between the vascularized and serouscomponents are often not possible with FA alone

because the serous and vascularized portions of aPED demonstrate late hyperfluorescence and leakagerespectively. Although fluorescein staining is moreintense in the serous portion of the detachment thanin the vascularized component, differences in intensityare often too minimal for accurate interpretation.However, the ICG angiographic findings are infinitelymore reliable for this differentiation; the serouscomponent of a PED is hypo-32#fluorescent and thevascularized component is hyperfluorescent.

Occult CNV is also sub grouped in two types,one with a solitary area of well-defined focal neovas-cularization (hot spot) and the other with a larger anddelineated area of neovascularization (plaque).

Hot Spot (Focal CNV)Focal CNV or a “hot spot” is an area of occult CNVthat is both well-delineated and no more than 1-discdiameter in size on ICGA. Also a hot spot representsan area of actively proliferating and more highlypermeable areas of neovascularization (active occultCNV). Chorioretnal anastomosis and polypoidal-type CNV may represent two subgroups of hot spots(see below).

PlaqueA plaque is an area of occult CNV larger than 1-discdiameter in size. A plaque often is formed by late-staining vessels, which are more likely to be quiescentareas of neovascularization that are not associatedwith appreciable leakage (inactive occult CNV).

(A) (B)

Figure 2 Occult choroidal neovascularization. Midphase fluorescein angiogram (A) demon-

strates hyperfluorescent drusen, while the late phase indocyanine green angiogram (B) reveals ahyperfluorescent occult choroidal neovascular membrane.

162 OLIVER ET AL.

Plaques of occult CNV seems to slowly grow indimension with time. Well-defined and ill-definedplaques are recognized on ICG study. A well-definedplaque has distinct borders throughout the study andthe full extent of the lesion can be assessed. An ill-defined plaque has indistinct margins or may be one inwhich any part of the neovascularization is blockedby blood.

In a review of our first 1000 patients with occultCNV by FA, which were imaged by ICGA, we categor-ized occult CNV into three morphologic categories:focal CNV or hot spots, plaques (well-defined andill-defined), and combination lesions in which bothhot spots and plaques were noted (39). The results ofthat study are discussed later in this chapter underclinical applications.

Two other forms of occult CNV are identified byICGA: polypoidal choroidal vasculopathy (PCV) andretinal angiomatous proliferation (RAP).

POLYPOIDAL CHOROIDAL VASCULOPATHY

PCV is a primary abnormality of the choroidal circula-tion characterized by an inner choroidal vascularnetwork of vessels ending in an aneurysmal bulgeor outward projection, visible clinically as a reddishorange, spheroid, polyp-like structure. The disorder isassociated with multiple, recurrent, serosanguineousdetachments of the RPE and neurosensory retina,secondary to leakage and bleeding from the peculiarchoroidal vascular abnormality (40,41).

ICGA has been used to detect and characterizethe PCV abnormality with enhanced sensitivity andspecificity (Fig. 3) (42–55). In the initial phases ofthe ICG study, a distinct network of vessels withinthe choroid becomes visible. Optical coherencetomography (OCT) (Fig. 3C) delineates the polypoidalextensions of the choroidal vasculature. In patientswith juxtapapillary involvement, the vascularchannels extend in a radial, arching pattern and areinterconnected with smaller spanning branches thatbecome more evident and numerous at the edge of thePCV lesion (Fig. 4).

Early in the course of the ICG study, the largervessels of PCV network start to fill before the retinalvessels, but the area within and surrounding thenetwork is relatively hypofluorescent compared withthe uninvolved choroid. The vessels of the networkappear to fill more slowly than the retinal vessels.Shortly after the network can be identified on theICG angiogram, small hyperfluorescent “polyps”become visible within the choroid.

These polypoidal structures correspond to thereddish, orange choroidal excrescence seen on clinicalexamination. They appear to leak slowly as the sur-rounding hypofluorescent area becomes increasingly

hyperfluorescent. In the later phase of the angiogramthere is a uniform disappearance of the dye(“washout”) from the bulging polypoidal lesions.The late ICG staining characteristic of occult CNV isnot seen in the PCV vascular abnormality.

While the first reports of PCV were in middle-aged black females, it is now recognized that PCVmaybe a variant of CNV seen in white patients with AMD,it may be localized in the macular area without any

(A)

(B)

(C)

Figure 3 Polypoidal choroidal vasculopathy. Color photograph(A) demonstrates hemorrhagic detachment of the macula. Late-

phase indocyanine green study (B) reveals a peripapillary polyp-like vascular network. Note central hypofluorescence indicative

of a pigment epithelial detachment. Optical coherencetomography (C) delineates the polypoidal extensions of the

choroidal vasculature.


peripapillary component (Figs. 5 and 6), and it may beformed by a network of small branching vesselsending in polypoidal dilation difficult to imagewithout ICGA (Fig. 6).

Ahuja and colleagues sought to determine theprevalence of PCV among British patients in their

practice (50). Of 40 consecutive patients with hemor-rhagic or exudative PEDs, 34 (85%) were attributed toPCV. Of those with PCV, 65% were female, the meanage was 65 years (range 44–88), 74% were white, 20%black, and 6% Asian. Eight had a history of hyper-tension. Sixty-eight percent of lesions were located inthe macula.

RETINAL ANGIOMATOUS PROLIFERATION

RAP is a distinct subgroup of neovascular AMD,manifested by intraretinal neovascularization (IRN)that extends into the deep retinal, subretinal, and sub-RPE spaces.

Clinical evidence of pre-, intra-, or subretinalhemorrhage, sometimes with associated exudatesor cystoid macular edema, in the setting of a PEDsuggests a RAP lesion. Often dilated compensatoryvessels perfuse and drain the neovascularization,forming a retinal–retinal anastomosis. Extension ofthe neovascular complex to the subretinal space mayresult in a retino-choroidal anastomosis (RCA).

On FA, indistinct RPE staining, often withassociated PED, resembles occult CNV. Presence ofactive IRN extending into a PED is difficult todistinguish from a standard VPED. ICG allowsbetter characterization of a VPED, revealing theneovascular hotspot contained within the hypofluor-escent PED (Fig. 7A,B). The OCT (Fig. 7C) shows

(A) (B)

Figure 4 Polypoidal choroidal vasculopathy. Peripapillary hyperfluorescent lesions are apparent

in the midphase fluorescein angiogram (A); however, the indocyanine green (B) delineates a moreextensive vascular network.

Figure 5 Macular polypoidal choroidal vasculopathy. Midphaseindocyanine green angiogram demonstrates a prominent lesion

of polypoidal channels in the macula.

164 OLIVER ET AL.

intraretinal cystic changes overlying a PED alongwith a hyperreflective area suggestive of a RCA.Late intraretinal leakage may arise from the IRN.ICG permits visualization of the direct communi-cation between the retinal and the choroidalcomponents of the neovascularization as they forman RCA (Fig. 8) Lafaut and coworkers (56) docu-mented the histopathology of an RCA, in whichneovascularization grows out from the neuroretinainto the subretinal space.

Kuhn et al. (57), in 1995, first identified RCA as apotential manifestation of this form of neovascularAMD in the setting of a VPED. With ICGA forenhanced choroidal imaging, this group found RCAin 50 of 186 (28%) patients with AMD and an associ-ated VPED. Slakter et al. (58) detected RCA in 34 of150 eyes (21%) with occult AMD and a focal hot spotin ICG. Fernandez and coworkers (59) reported aseries of 190 patients with neovascular AMD inwhich ICGA revealed 34 eyes (16%) with RAP lesions.

(A) (B)

(C) (D)

Figure 6 Macular polypoidal choroidal vasculopathy. Despite a fundus appearance (A) only ofmild pigment epithelial change, the indocyanine green reveals progressive macular hyperfluor-

escence of polypoidal lesions in the early (B), mid (C) and late (D) phases of the angiogram.


Yannuzzi and colleagues (60) classified RAP intothree stages: stage I involves IRN, stage II results fromextension to subretinal neovascularization, and stageIII occurs once CNV is documented.

Clinical knowledge and recognition of RAP isimportant because this form of neovascular AMDmayhave a natural course, visual prognosis, and responseto treatment distinct from other forms of neovascularAMD. Different forms of treatment may be preferable

for each stage of the disorder. For example, we havefound that an uncomplicated focal area of IRN may beamenable to conventional thermal laser treatment;whereas, a more advanced stage involving a VPEDand an RCA is less likely to respond to any form ofcurrently available treatment.

Bottoni and colleagues (61) retrospectivelyreported results of 99 eyes of 81 patients with RAPtreated with direct laser photocoagulation of the

(A) (B)

(C)

Figure 7 Retinal angiomatous proliferation. Fluorescein angio-

gram (A) indicates a pigment epithelial detachment, while theindocyanine green (B) reveals a focal area of hyperfluorescence

adjacent to two retinal arterioles. The optical coherencetomography (C) shows intraretinal cystic changes overlying a

pigment epithelial detachment along with a hyperreflective areasuggestive of a retino-choroidal anastomosis.

(A)

(B)

Figure 8 Retinal angiomatous proliferation (RAP) vasculature.Retino-choroidal anastomosis stands out in this indocyaninegreen angiogram (A) that reveals a larger underlying choroidalneovascularization. Optical coherence tomography (B) of thisstage III RAP lesion demonstrates vessels from a low neuro-sensory detachment diving towards a subretinal choroidalneovascular membrane.

166 OLIVER ET AL.

vascular lesion, laser photocoagulation of the feederretinal arteriole, scatter gridlike laser photocoagula-tion, photodynamic therapy (PDT), or transpupillarythermotherapy. Complete obliteration was achievedin 24 (57%) cases of stage I lesions (73% closure fromdirect laser and 45% closure from PDT, 11 (26%) ofstage II lesions (38% closure from scatter gridlikephotocoagulation and 17% closure from direct photo-coagulation of the vascular lesion), and only 3 (15%)of stage III lesions. The uncontrolled use of thera-peutic interventions in this study makes it difficult todraw definitive conclusions about a superior treat-ment modality, but the study makes clear thedifficulty in effectively treating more advancedstages of RAP.

More recent series using PDT alone or PDTwith triamcinolone confirm the challenge in effec-tively treating RAP lesions. Boscia and colleagues(62) treated 21 eyes with stage II or III RAP usingPDT alone and reported an overall decline in visionfrom 20 out of 80 to 20 out of 174, stabilization ofvision in six eyes (29%), occlusion of RAP and PEDflattening in three eyes (14%), and an RPE tear infour eyes (19%). Nicolo and colleagues (63) reported10 eyes with stage II RAP treated with 20 mg ofintravitreal triamcinolone acetonide (IVTA) followedone month later by PDT. All patients experiencedflattening of the PED prior to PDT, six patients(60%) showed improved vision of at least threeearly treatment of diabetic retinopathy study(ETDRS) lines at three, six, and nine months, andfour patients (40%) maintained visual improvementat 12 months.

PDT with a photosensitizing dye such asverteporphin may have a different effect on RAPthan on classic-CNV or occult-CNV lesions (64,65).Given the tendency for ICG dye to stain the retinain eyes with RAP, there is a possibility that similarstaining may occur with the verteporphin molecule,theoretically predisposing the retina to photoche-mical damage when exposed to the excitatory lightused in PDT. This possibility is speculative, sinceverteporphin has not yet been imaged successfullywith good spatial and temporal definition inthe human.

Because eyes with RAP are generally classified aspure occult CNV based on FA, it is possible thatpatients with RAP were actually treated in the treat-ment of age-related macular degeneration withphotodynamic therapy (TAP) trial (65). ICGA wasnot used in the TAP trial, so the frequency of RAP inthe subset of patients classified as occult-CNV isunknown. Future studies of AMD that use ICG willbe able to delineate between these two distinct formsof macular degeneration.

CLINICAL APPLICATION OF ICGATO THE STUDY OF AMD

Patz and associates (26) were the first to study CNV byICG videoangiography. They could resolve only 2 of 25CNVs with their early model. Bischoff and Flower (18)studied 100 ICG angiograms of patients with AMD.They found “delayed and/or irregular choroidalfilling” in some patients. The significance of thisfinding is unclear, however, because these authorsdid not include an age-matched control group.Tortuous vessels and marked dilation of macularchoroidal arteries, often with loop formation, werealso observed.

Hayashi and associates (19,21) found that ICGvideoangiography was useful in the detection of CNV.ICG videoangiography was able to confirm the fluor-escein angiographic appearance of CNV in patientswith well-defined CNV. It revealed a more well-defined neovascularization in 27 eyes with occultCNV by FA. In a subgroup of patients with poorlydefined occult CNV, the ICG angiogram, but not theFA, imaged a well-defined CNV in 9 of 12 (75%) cases.ICG videoangiography of the other three eyes revealedsuspicious areas of neovascularization. Hayashi andcoworkers (19,21) were also the first to show thatleakage from CNV with ICG was slow compared tothe rapid leakage of sodium fluorescein. While theresults of these investigators concerning ICG videoan-giographic imaging of occult CNV were promising,the spatial resolution that they could obtain waslimited by the 512-line video monitor and analogtape of their ICG system.

Destro and Puliafito (22) reported that ICGvideoangiography was particularly useful in studyingoccult CNV with overlying hemorrhage and recurrentCNV. Guyer and coworkers (3) used a 1024-line digitalimaging system to study patients with occult CNV.These authors reported that ICG videoangiographywas useful in imaging occult CNV and that thistechnique could allow photocoagulation of otherwiseuntreatable lesions. Scheider and coinvestigators (25)have reported enhanced imaging of CNV in a study of80 patients using the scanning laser ophthalmoscopewith ICG videoangiography.

Yannuzzi and associates (4) have shown thatICGA is extremely useful in reclassifying occult CNVinto “well-defined CNV.” In their study, 39% of 129patients with occult CNV were reclassified as well-defined CNV based on information added by ICGA.Five of seven (72%) cases of occult CNV with SPEDwere reclassified as “well demarcated” CNV by ICG.In 17 of 38 (45%) VPED cases and in 11 of 19 (58%)combined VPED and SPED cases, ICGA allowedoccult CNV to be reclassified as well defined CNV.These authors concluded that ICGA was especially


useful in identifying occult CNV in patients with SPEDor with recurrent CNV.

Lim et al. found that ICG demonstrated well-demarcated hyperfluorescence in 50% of eyes thoughtto have occult CNV by FA and in 82% of eyes with PED(66). Baumal et al. found that ICG demonstratedunderlying CNV in 19 of 23 eyes (83%) with an isolatedPED and in 21 of 21 eyes (100%) with PED and occultCNV (67).

Yannuzzi and coworkers (68) studied with ICGA235 consecutive AMD patients with occult CNV andassociated VPED. These eyes were divided into twogroups, depending on the size and delineation of theCNV. Of the 235 eyes 89 (38%) had a solitary area ofneovascularization that was well delineated, no morethan 1-disc diameter in size, and defined as focal CNV.The other 146 (62%) eyes had a larger area of neovas-cularization, with variable delineation defined as aplaque CNV.

In a further report, 657 consecutive eyes withoccult CNV determined by FA were studied withICGA. Of 413 eyes with occult CNV without pigmentepithelium detachments, focal areas of neovascular-ization were noted in 89 (22%). Overall, 142 (34%) eyeshad lesions that were potentially treatable by thermallaser photocoagulation based on additional infor-mation provided by ICGA. Of the 235 eyes withoccult CNV and VPEDs, 98 (42%) were eligible forphotocoagulation therapy based on ICGA findings.The authors calculated that ICGA enhances the treat-ment eligibility by approximately one-third (69).

In a expanded series (39) the same authorsreported their results on ICGA study of 1000 consecu-tive eyes with occult CNV by FA. They recognizedthree morphologic types of CNV, which included focalspots, plaques (well defined and poorly defined), andcombination lesions (in which both focal spots andplaques are noted). Combination lesions were furthersubdivided into marginal spots (focal spots at the edgeof plaque of neovascularization), overlying spots (hotspots overlying plaques of neovascularization), orremote spots (a focal spot remote from a plaque ofneovascularization).

The relative frequency of these lesions wasas follows: focal spots 29%, plaques 61% consisting27% of well-defined plaques and 34% of poorlydefined plaques, and combination lesions 8%,consisting of 3% of marginal spots, 4% of overlyingspots and 1% of remote spots (39). A follow-up studyfrom the same authors of patients with newly diag-nosed unilateral occult CNV secondary to AMDshowed that the patients tended to develop the samemorphologic type of CNV in the fellow eye (70).

Chang et al. (71) reported on the clinicopatho-logic correlation of AMDwith CNV detected by ICGA.Histopathologic examination of the lesion revealed a

thick subRPE CNV corresponding to the plaque-likelesion seen with ICGA.

Watzke and colleagues analyzed 104 consecutiveAMD patients to determine the sensitivity of ICGin detecting lesions originally identified by FA (72).ICG hyperfluorescence was present in 87% of eyeswith classic CNVand in 93% of eyes with fibrovascularpigment epithelium detachments (FVPEDs). Of eyesdiagnosed with LLUS by FA, 50% were hyperfluor-escent and 50% were isofluorescent by ICG.Additionally, three fellow eyes with dry AMD hadhyperfluorescent lesions by ICG, but it is unknownwhether these eyes progressed to neovascular AMD.

Finally, Lee et al. (73) reported on 15 eyes withsurgically excised subfoveal CNV that underwentpreoperative and postoperative ICGA. All excisedmembranes were examined by light microscopy, andall surgically excised ICG-imaged membranes corre-sponded to subRPE and subneurosensory CNV.

The above studies demonstrate that ICGA is animportant adjunctive study to FA in the detection ofCNV. FA is more sensitive than ICGA in imaging finecapillaries that connect larger vessels and capillaries atthe proliferating edge of well-defined CNV. While FAimages well-defined CNV better than ICGA in somecases, ICGA allows reclassification of FA-definedoccult CNV into well-demarcated CNV eligible forICG-guided thermal laser treatment in about 30% ofcases (74).

The best imaging strategy to thoroughly classifyCNV is the combination of FA and ICGA. Helbig et al.studied 502 patients using simultaneous FA and ICGto characterize AMD, and found that 3% of eyes had ahot spot within an occult lesion, 4% had plaqueswithin an occult lesion, 9% had RAP, and 6% hadPCV (75). Yanagi and colleagues (76) compared simul-taneous fluorescein and ICG injection with FA-guidedICGA, in which FA was used to detect an area ofleakage, allowing a lower dose ICG injection andfocusing the ICG detector only on the lesion in ques-tion. Overall detection of feeder vessels (FVs) wassimilar between the simultaneous and FA-guidedICG groups, but the latter group required lowerquantities of ICG and had shorter examination times.The benefits of simultaneous procedures, such asconvenience and accurate diagnosis of treatablecases, must be weighed against the disadvantages ofincreased cost and adverse effects.

RECURRENT CNV IN AMD

Recurrent CNV following photocoagulation treat-ment is a major cause of treatment failure. Althoughmost recurrences can be detected and imaged withclinical biomicroscopic examination and FA, a signi-ficant number of patients demonstrate new exudative

168 OLIVER ET AL.

manifestations and visual symptoms without aclearly defined area of recurrent neovascularizationidentified by FA. These patients may exhibit diffusestaining and leakage at the site of previous treatmentor may demonstrate no FA evidence of recurrencedespite the new exudative manifestations identifiedclinically. ICGA has proven to be often useful indetecting the recurrence.

Sorenson et al. (77) reported the use of ICG-guided laser treatment in 66 cases of recurrent occultCNV secondary to AMD. Only 29 (44%) were eligiblefor laser retreatment, and of these 29 eyes 18 (62%) hadanatomic success with an average follow-up of sixmonths (54). Similar results were reported by Reichelet al., who reported 58 eyes with recurrent CNV fromAMD (78). In 14 eyes (24%), a well-defined recurrentCNV could be identified by evaluating the fluoresceinangiogram. In 6 (14%) of the remaining 44 eyes, a well-defined recurrent CNV was identifiable by ICGA.

However, clinical evidence of recurrence mustaccompany a hot spot detected by ICG. Chen andcolleagues (79) performed ICGA two weeks afterkrypton laser treatment on 230 consecutive eyes withexudative AMD. Forty patients (18%) developed ICGhot spots after treatment, and these hot spot spon-taneously resolved without development of CNV in31 patients. Recurrent CNVwas present at the hot spotin four patients and away from the hot spot infive patients.

ICG-ASSOCIATED TREATMENT STRATEGIESFOR CNV IN AMD

In the past, patients were considered potentiallyeligible for laser photocoagulation therapy by ICGguidance if they had clinical and FA evidence ofoccult CNV. Of the two types of occult CNV identi-fiable by ICG study, hot spots and plaques, direct laserphotocoagulation was recommended only to hotspots. In fact as mentioned above, hot spots representareas of actively leaking neovascularization that canbe obliterated by laser photocoagulation in attemptto eliminate the associated serosanguineous compli-cations, and stabilize or improve the vision. On thecontrary, plaques seem to represent a thin layer ofneovascularization, which is not actively leaking, andwhich may benefit from PDT (80) or intravitrealantiangiogenic agents (81–86).

In the case of a lesion comprised of a hot spot anda plaque, and in which the hot spot is at the marginof the plaque (that may extend under the fovea),laser photocoagulation to the extrafoveal hot spotspares the fovea. This treatment approach wassuccessful in obliterating the CNV and stabilizing thevision in 56% of a consecutive series of AMD patients(74). On the contrary we have had poor success with

direct laser treatment of hot spots overlying plaques,or confluent treatment of the entire plaque.

Slakter and associates (87) performed ICG-guided laser photocoagulation in 79 eyes with occultCNV.The occult CNVwas successfully eliminatedwithstabilized or improved visual acuity in 29 (66%) of 44eyes with occult CNV associated with neurosensoryretinal elevations, and in 15 (43%) of 35 eyeswith occultCNV associated with PED. This study demonstratedthat in some cases ICGA imaging can successfullyguide laser photocoagulation of occult CNV.

Another pilot study of ICG-guided laser treat-ment of occult CNV had similar results (88).

Guyer and coworkers (74) reported a pilot studywith ICG-guided laser photocoagulation of 23 eyeswith occult CNV secondary to AMDwith focal spots atthe edge of a neovascular plaque of the ICG study.ICG-guided laser photocoagulation was applied solelyto the focal spot at the edge of the plaque. At 24months of follow-up anatomic success with resolutionof the exudative findings was obtained in 6 (37.5%) of16 eyes. Importantly, these studies set the foundationfor future prospective studies of ICG-guided lasertreatment. In addition, they proved that the presenceof a PED is a poor prognostic factor in the treatment ofexudative AMD.

Lim et al. reported the visual acuity outcomeafter ICGA-guided laser photocoagulation of CNVassociated with PED in 20 eyes with AMD (89). Atthree months after laser photocoagulation, visualacuity had improved two or more Snellen lines intwo eyes (10%), worsened by two or more lines in 10(50%), and remained unchanged in eight of 20 (40%).At nine months after laser photocoagulation, visualacuity had improved by two or more lines in one eye(9%), worsened by two or more lines in nine (82%),and remained unchanged in one of 11 (9%). Theyconcluded that ICG-guided laser photocoagulationmay temporarily stabilize visual acuity in some eyeswith CNVassociated with PED, but final visual acuitydecreases with time.

More recently, Da Pozzo and associates evalu-ated the efficacy of ICG-guided photocoagulation in86 eyes with occult CNV and a hot spot on ICG (90).Of the 53 eyes without PED, 32 (60%) had stable orimproved vision at one year, but 27 (51%) had recur-rence of the CNV. Of the 33 eyes with PED, only five(15%) had stable or improved vision at one year, and23 (70%) had CNV recurrence.

Another potential therapeutic applicationusing ICG is ICG dye-enhanced diode laser photo-coagulation. The peak absorption of ICG (795 to810 nm) is at a similar wavelength as the peak emissionof the diode laser (805 nm). Thus, dye-enhanced laserphotocoagulation may allow selective ablation of theICG-containing CNV with relative sparing of the


normal neighboring retina. However, leakage of ICGinto the intraretinal space,which occurred in 11%of 149eyes in a series reported byHo and colleagues,may be acontraindication to ICG dye-enhanced diode photo-coagulation (91).

A pilot study by Reichel and associates of 10patients with poorly defined CNV resulted in closureof the CNV in all cases, but a severe immediate visionloss occurred in one patient (92). A larger series byObana et al. studied 38 eyes with classic or occult CNV,and found that CNVocclusion was achieved in 92% ofeyes, with and 18% recurrence rate over an averagefollow-up of 26 months (93). Ten eyes (26%) showedimproved visual acuity, 16 (42%) showed no change,and 12 eyes (32%) worsened.

A pilot study by Arevalo et al. compared ICGdye-enhanced diode laser photocoagulation alonewith dye-enhanced laser combined with IVTA (94).In their initial series of 19 eyes selected irrespectiveof lesion subtype, none of the 9 eyes receiving com-bination therapy required retreatment at sevenmonths, while 4 of 10 eyes receiving ICG dye-enhanced laser alone needed retreatment. A followup paper by the same group reported 31 eyes treatedwith dye-enhanced laser and 4 mg of IVTA followedfor a mean of nine months (95). Nineteen eyes (61%)showed stable vision, seven eyes (23%) improved, andfive eyes (16%) worsened. In the occult subgroup,however, the proportion of patients who worsenedwas greater (33%). No severe acute vision loss wasreported. Topically treatable glaucoma occurred infive eyes.

NEW TECHNIQUES IN ICG ANGIOGRAPHY

Recent advances in ICGA are real-time angiography,contrast enhancement ICGA, wide-angle angiography,digital subtraction-indocyanine green (DS-ICG) angio-graphy, dynamic ICG-guided FV laser treatment ofCNV, ICGA for dry AMD, and use of cSLO-ICG.

Contrast Enhanced ICG AngiographyContrast enhancement of ICG angiographic imagesusing digital imaging software may enhance the diag-nostic sensitivity and specificity of the study. Maberleyand Cruess compared nonenhanced and contrast-enhanced ICG angiographic images of 50 consecutivepatients with occult CNV from AMD (96). Only 36% ofthe nonenhanced images demonstrated well-definedmembranes, whereas 58% were well defined with thecontrast-enhanced images.

Real-Time ICGAReal-time ICGA (97) uses a modified Topcon 50IAcamera with a diode laser illumination system that

has an output at 805 nm (Topcon 50IAL camera), canproduce images at 30 frames per second, and allowscontinuous recording. The images can be acquiredeither as a videotape, or as single image at afrequency of 30 images per second. To makeprinted copies of these images single frames aredigitized, but the resolution is limited to 640 by480 pixels.

Wide-Angle ICGAWide-angle images of the fundus can be obtainedby performing ICG videoangiography with the aid ofwide angle contact lenses. The contact lenses used arethe Volk SuperQuad 160, the Volk Quadraspheric, orthe Volk Transequator (Volk, Mentor OH). Because theimage formed by these lenses lies about 1 cm in frontof the lens, the fundus camera is set on A orC so thatthe camera is focused on the image plane of thecontact lens.

This technique allows instantaneous imagingof a large area of the fundus. The combined use ofthe contact lens and of the laser illumination systemin a high-speed digital fundus camera allows real-time imaging of a 1608 of field of view. Staurenghiand colleagues recently developed a combined contactand noncontact system to achieve wide-field imagesup to 1508 with a cSLO ICG (98).

Digital Subtraction-Indocyanine Green AngiographyDS-ICGA uses DS of sequentially acquired ICG angio-graphic frames to image the progression of the dyefront in the choroidal circulation (99,100). A methodof pseudocolor imaging of the choroid allows differ-entiation and identification of choroidal arteries andveins. DS-ICGA allows imaging of occult CNV withgreater detail and in a shorter period of time than withconventional ICGA.

Matsumoto et al. performed DS-ICGA on 20patients with CNV accompanied by subretinalhemorrhage (101). In six of the 20 eyes, DS-ICGAdistinguished hyperfluorescence due to a slowlyexpanding, poorly defined, large lesion from simpleleakage with a well-defined lesion. The DS-ICGAtechnique made clear the expanding wave ofhyperfluorescence from a more slowly filling, ill-defined lesion.

FV TherapyStaurenghi et al. (102) considered a series of 15 patientswith subfoveal CNVM in whom FVs could be clearlydetected by means of dynamic ICGA but not necess-arily with FA. Based on the pilot study, the authorssimultaneously reported a second series of 16 patientswith FVs smaller than 85 mm. FV were treatedwith argon green laser. The ICGA was repeated

170 OLIVER ET AL.

immediately after treatment, and at two, seven, andthirty, and then every three months, to assess FVclosure. A FV that remained patent was immediatelyretreated, and ICGA follow-up started again. In thepilot study, 40% of FVs were successfully occluded;this result was affected by the width and number ofFVs. The occlusion success rate in the second series,with FVs under 85 mm, was 75%. The authorsconcluded that dynamic ICGA may detect small FVsthat are more successfully occluded by argonphotocoagulation.

ICG FOR DRY AMD

Hanutsaha et al. studied 432 patients by ICGwith exudative AMD in one eye and drusen withoutexudation in the fellow eye (103). Eighty-nine percentof eyes with drusen had normal fluorescence on ICG,while 11% of eyes with drusen had focal hot spots orhyperintense plaques. Over an average follow-up of22 months, 27% of eyes with drusen and an abnormalICG developed CNV, while only 10% of druseneyes and a normal ICG developed exudative AMD.The authors suggested that ICG may be a predictiveindicator of future exudative changes in eyes withdrusen.

Patchy and slow choroidal filling on FA, inassociation with reduced choroidal fluorescence onICGA, was associated by Pauleikoff and associateswith early changes in AMD (104). One hundred eyeswith early AMD were studied for the above charac-teristics, termed a prolonged choroidal filling phase(PCFP), which was associated with confluent drusenin the study eye, focal RPE atrophy in the study eye,and geographic atrophy in the fellow eye. The grouppostulated that PCFP was a clinical indicator of Bruchmembrane deposits and a predictor for geographicatrophy from AMD.

Ultra-late phase ICGA, performed 24 hoursafter dye injection, demonstrates hypofluorescentgeographic lesions in patients with both exudativeand dry AMD, as shown by Mori and associates.They demonstrated that 95% of AMD eyes withCNV had geographic hypofluorescent lesions, andthat all CNV detectable by FA or ICG was containedwithin these lesions. In 73% of eyes without CNV, thesame geographic areas were present, while age-matched normal subjects did not have the lesions.Mean fluorescence intensity was higher in a normalgroup older than 62 years, compared tonormal subjects less than 36 years. The authorspostulated that these geographic hypofluorescentareas may represent areas predisposed to CNVdevelopment.

CONFOCAL SCANNING LASEROPHTHALMOSCOPE ICG

With the relatively recent availability of cSLO toperform ICG, many retinal physicians are choosingthis modality over high-speed digital angiographybecause of the ability to use the cSLO for otherfunctions, such as autofluorescence and FA. Geliskenand colleagues simultaneously compared cSLO ICGwith high-resolution fundus camera ICG in 100 eyeswith occult CNV (105). Confocal SLO was superior indelineating vessel architecture of the neovascularlesion; however fundus photography was muchmore sensitive than cSLO in detecting focal lesions(52% vs. 37%, respectively) and plaques (35% vs. 13%,respectively).

CONCLUSION

The role ICGA in the treatment of AMD is inevolution. As photocoagulation of extrafoveal CNVgave way to treatment of all types of CNV withPDT, ICG angiography has proven very useful inadding information to FA about lesion subtype. Theability of ICG to identify subtypes of occult CNV,such as VPED, hot spots, plaques, and RCA, allowstargeted and sometimes effective therapy for theserefractory types of CNV. Given that approximately87% of new CNV from AMD is minimally classic oroccult (106), many patients have derived somebenefit from the additional information obtainableby ICGA.

The approval of pegaptanib sodium (Macugen)heralded a new era in AMD treatment (81). Shown tobe equally efficacious for all lesion subtypes, pegap-tanib was a departure from traditional laser-based,destructive procedures. Ranibizumab (Lucentis) hasbeen shown to be even more efficacious and beavci-zumab (Avastin) appears to show similar results toranibizumab (82–85). Further research is necessary todetermine whether lesion subtype remains an import-ant predictor of treatment response with thesenew modalities.

A systematic evidence based review of thePubMed indexed literature in English or with anEnglish abstract yielded a strong recommendationfor the use of ICGA for the following conditions:PCV, occult CNV, neovascularization associated withPED, and recurrent choroidal neovascular membranes(107). The same review reported only modest evidencesupporting the use of ICGA for routine choroidalneovascular membranes and for identifying FVs inAMD. Future advances in ICGA, such as wide angle,real-time, and DS techniques may improve our diag-nostic ability in AMD.


SUMMARY POINTS

& ICGA is a useful adjunctive technique to FA for thediagnosis of AMD. This is especially true in thepresence of occult CNV. ICG allows better recog-nition of subtypes of occult CNV such as VPED,hot spots, plaques and retinal-choroidal anasto-mosis.

& ICGA is useful in the diagnosis of PCV, RAP, andrecurrent choroidal neovascular membranes.

& Preliminary studies suggested that ICG-guidedlaser photocoagulation was beneficial in the treat-ment of CNV prior to the era of antivascularendothelial growth factor therapy.

& Further research is necessary to improve ourunderstanding of all the information obtainedby ICGA and its potential role in new therapeuticregimens.

& Real-time ICGA, wide-angle ICGA, and DS-ICGAmay improve our diagnostic ability in AMD.

REFERENCES

1. Schatz H, Burton TC, Tannuzzi LA, Rabb MF. Interpre-tation of Fundus Fluorescein Angiograph. St. Louis:Mosby-Year Book, 1978.

2. Yannuzzi LA. Laser Photocoagulation of the Macula.Philadelphia, PA: JB Lippincott, 1989.

3. Guyer DR, Puliafito CA, Mones JM, Friedman E, ChangW,Verdooner SR. Digital indocyanine-green angiography inchorioretinal disorders. Ophthalmology 1992; 99(2):287–91.

4. Yannuzzi LA, Slakter JS, Sorenson JA, Guyer DR,Orlock DA. Digital indocyanine green videoangio-graphy and choroidal neovascularization. Retina 1992;12(3):191–223.

5. Geeraets WJ, Berry ER. Ocular spectral characteristicsas related to hazards from lasers and other light sources.Am J Ophthalmol 1968; 66(1):15–20.

6. Fox IJ, Wood EH. Applications of dilution curves recordedfrom the right side of the heart or venous circulation withthe aid of a new indicator dye. Mayo Clin Proc 1957;32(19):541–50.

7. Fox IJ, Wood EH. Indocyanine green: physical and physio-logic properties. Mayo Clin Proc 1960; 35:732–44.

8. Kuroiwa T, Kajimoto Y, Ohta T. Development andclinical application of near-infrared surgical microscope:preliminary report. Minim Invasive Neurosurg 2001;44(4):240–2.

9. Raabe A, Nakaji P, Beck J, et al. Prospective evaluationof surgical microscope-integrated intraoperative near-infrared indocyanine green videoangiography duringaneurysm surgery. J Neurosurg 2005; 103(6):982–9.

10. Kogure K, David NJ, Yamanouchi U, Choromokos E.Infrared absorption angiography of the fundus circula-tion. Arch Ophthalmol 1970; 83(2):209–14.

11. David NJ. Infrared absorption fundus angiography. In:proceedings of the International Symposium on Fluor-escein Angiography. Albi, France, 1969.

12. Hochheimer BF. Angiography of the retina with indocya-nine green. Arch Ophthalmol 1971; 86(5):564–5.

13. Flower RW, Hochheimer BF. Clinical infrared absorptionangiography of the choroid. Am J Ophthalmol 1972;73(3):458–9.

14. Flower RW. Infrared absorption angiography of thechoroid and some observations on the effects of highintraocular pressures. Am J Ophthalmol 1972; 74(4):600–14.

15. Flower RW, Hochheimer BF. A clinical technique andapparatus for simultaneous angiography of the separateretinal and choroidal circulations. Invest Ophthalmol VisSci 1973; 12(4):248–61.

16. Flower RW, Hochheimer BF. Indocyanine green dye fluor-escence and infrared absorption choroidal angiographyperformed simultaneously with fluorescein angiography.Johns Hopkins Med J 1976; 138(2):33–42.

17. Hyvarinen L, Flower RW. Indocyanine green fluore-scence angiography. Acta Ophthalmol (Copenh) 1980;58(4):528–38.

18. Bischoff PM, Flower RW. Ten years experience withchoroidal angiography using indocyanine green dye: anew routine examination or an epilogue? Doc Ophthalmol1985; 60(3):235–91.

19. Hayashi K, Hasegawa Y, Tazawa Y, de Laey JJ. Clinicalapplication of indocyanine green angiography to chor-oidal neovascularization. Jpn J Ophthalmol 1989;33(1):57–65.

20. Hayashi K, Hasegawa Y, Tokoro T. Indocyanine greenangiography of central serous chorioretinopathy. IntOphthalmol 1986; 9(1):37–41.

21. Hayashi K, de Laey JJ. Indocyanine green angiography ofsubmacular choroidal vessels in the human eye. Ophthal-mologica 1985; 190(1):20–9.

22. Destro M, Puliafito CA. Indocyanine green videoangio-graphy of choroidal neovascularization. Ophthalmology1989; 96(6):846–53.

23. Scheider A, Schroedel C. High resolution indocyaninegreen angiography with a scanning laser ophthalmoscope.Am J Ophthalmol 1989; 108(4):458–9.

24. Scheider A. Indocyanine green angiography with aninfrared scanning laser ophthalmoscope. Initial clinicalexperiences. Ophthalmologe 1992; 89(1):27–33.

25. Scheider A, Kaboth A, Neuhauser L. Detection of sub-retinal neovascular membranes with indocyanine greenand an infrared scanning laser ophthalmoscope. AmJ Ophthalmol 1992; 113(1):45–51.

26. Patz A, Flower RW, Klein ML, Orth DH, Fleishman JA,MacLeod D. Clinical applications of indocyanine greenangiography. Doc Ophthalmol Proc Ser 1976; 9:245–51.

27. Cherrick GR, Stein SW, Leevy CM, Davidson CS. Indo-cyanine green: observations on its physical properties,plasma decay, and hepatic extraction. J Clin Invest 1960;39:592–600.

28. Baker KJ. Binding of sulfobromophthalein (BSP) sodiumand indocyanine green (ICG) by plasma alpha-1 lipopro-teins. Proc Soc Exp Biol Med 1966; 122(4):957–63.

29. Leevy CM, Bender J. Physiology of dye extraction by theliver: comparative studies of sulfobromophthalein andindocyanine green. Ann N YAcad Sci 1963; 111:161–76.

30. Goresky CA. Initial distribution and rate of uptake ofsulfobromophthalein in the liver. Am J Physiol 1964;207:13–26.

31. Ketterer SG, Weigand BD. The excretion of indocyaninegreen and its use in the estimation of hepatic blood flow.Clin Res 1959; 7:71.

32. Ketterer SG, Weigand BD. Hepatic clearance of indocya-nine green. Clin Res 1959; 7:289.

33. Probst P, Paumgartner G, Caucig H, Frauolich H,Grabner G. Studies on clearance and placental transfer ofindocyanine green during labor. Clin Chim Acta 1970;29:157.

172 OLIVER ET AL.

34. Leevy CM, Smith F, Kiernan T. Liver function test. In:Bockus HL, ed. Gastroenterology. 3rd ed., Vol. 3.Philadelphia, PA: Saunders, 1976:68.

35. Shabetai R, Adolph RJ. Principles of cardiac catheteriza-tion. In: Fowler NO, ed. Cardiac Diagnosis and Treatment.3rd ed. Hagerstown, MD: Harper & Row, 1980:117.

36. Hope-Ross M, Yannuzzi LA, Gragoudas ES, et al. Adversereactions due to indocyanine green. Ophthalmology 1994;101(3):529–33.

37. Bindewald A, Stuhrmann O, Roth F, et al. Lower limits offluorescein and indocyanine green dye for digital cSLOfluorescence angiography. Br J Ophthalmol 2005;89(12):1609–15.

38. Macular Photocoagulation Study Group. Occult choroidalneovascularization. Influence on visual outcome inpatients with age-related macular degeneration. ArchOphthalmol 1996; 114(4):400–12.

39. Guyer DR, Yannuzzi LA, Slakter JS, et al. Classification ofchoroidal neovascularization by digital indocyanine greenvideoangiography. Ophthalmology 1996; 103(12):2054–60.

40. Kleiner RC, Brucker AJ, Johnston RL. The posterior uvealbleeding syndrome. Retina 1990; 10(1):9–17.

41. Yannuzzi LA, Sorenson J, Spaide RF, Lipson B. Idiopathicpolypoidal choroidal vasculopathy (IPCV). Retina 1990;10(1):1–8.

42. Spaide RF, Yannuzzi LA, Slakter JS, Sorenson J, Orlach DA.Indocyanine green videoangiography of idiopathicpolypoidal choroidal vasculopathy. Retina 1995;15(2):100–10.

43. Phillips WB, II, Regillo CD, Maguire JI. Indocyanine greenangiography of idiopathic polypoidal choroidal vasculo-pathy. Ophthalmic Surg Lasers 1996; 27(6):467–70.

44. Yannuzzi LA, Ciardella A, Spaide RF, Rabb M, Freund KB,Orlock DA. The expanding clinical spectrum of idiopathicpolypoidal choroidal vasculopathy. Arch Ophthalmol1997; 115(4):478–85.

45. Yannuzzi LA, Nogueira FB, Spaide RF, et al. Idiopathicpolypoidal choroidal vasculopathy: a peripheral lesion.Arch Ophthalmol 1998; 116(3):382–3.

46. Moorthy RS, Lyon AT, Rabb MF, Spaide RF, Yannuzzi LA,Jampol LM. Idiopathic polypoidal choroidal vasculopathyof the macula. Ophthalmology 1998; 105(8):1380–5.

47. Schneider U, Gelisken F, Kreissig I. Indocyanine greenangiography and idiopathic polypoidal choroidal vasculo-pathy. Br J Ophthalmol 1998; 82(1):98–9.

48. Yannuzzi LA, Wong DW, Sforzolini BS, et al. Polypoidalchoroidal vasculopathy and neovascularized age-related macular degeneration. Arch Ophthalmol 1999;117(11):1503–10.

49. Yannuzzi LA, Freund KB, Goldbaum M, et al. Polypoidalchoroidal vasculopathy masquerading as central serouschorioretinopathy. Ophthalmology 2000; 107(4):767–77.

50. Ahuja RM, Stanga PE, Vingerling JR, Reck AC, Bird AC.Polypoidal choroidal vasculopathy in exudative andhaemorrhagic pigment epithelial detachments. Br JOphthalmol 2000; 84(5):479–84.

51. Escano MF, Fujii S, Ishibashi K, Matsuo H, Yamamoto M.Indocyanine green videoangiography in macular variantof idiopathic polypoidal choroidal vasculopathy. Jpn JOphthalmol 2000; 44(3):313–6.

52. Uyama M, Wada M, Nagai Y, et al. Polypoidal choroidalvasculopathy: natural history. Am J Ophthalmol 2002;133(5):639–48.

53. Sho K, Takahashi K, Yamada H, et al. Polypoidal choroidalvasculopathy: incidence, demographic features, andclinical characteristics. Arch Ophthalmol 2003; 121(10):1392–6.

54. Nishijima K, Takahashi M, Akita J, et al. Laser photo-coagulation of indocyanine green angiographicallyidentified feeder vessels to idiopathic polypoidal chor-oidal vasculopathy. Am J Ophthalmol 2004; 137(4):770–3.

55. Nakajima M, Yuzawa M, Shimada H, Mori R. Correlationbetween indocyanine green angiographic findings andhistopathology of polypoidal choroidal vasculopathy.Jpn J Ophthalmol 2004; 48(3):249–55.

56. Lafaut BA, Aisenbrey S, Vanden Broecke C, Bartz-Schmidt KU. Clinicopathological correlation of deep ret-inal vascular anomalous complex in age related maculardegeneration. Br J Ophthalmol 2000; 84(11):1269–74.

57. Kuhn D, Meunier I, Soubrane G, Coscas G. Imaging ofchorioretinal anastomoses in vascularized retinalpigment epithelium detachments. Arch Ophthalmol1995; 113(11):1392–8.

58. Slakter JS, Yannuzzi LA, Schneider U, et al. Retinalchoroidal anastomoses and occult choroidal neovas-cularization in age-related macular degeneration.Ophthalmology 2000; 107(4):742–53 (discussion 753-44).

59. Fernandes LH, Freund KB, Yannuzzi LA, et al. The natureof focal areas of hyperfluorescence or hot spots imagedwith indocyanine green angiography. Retina 2002;22(5):557–68.

60. Yannuzzi LA, Negrao S, Iida T, et al. Retinal angiomatousproliferation in age-related macular degeneration. Retina2001; 21(5):416–34.

61. Bottoni F, Massacesi A, Cigada M, Viola F, Musicco I,Staurenghi G. Treatment of retinal angiomatous prolifer-ation in age-related macular degeneration: a series of 104cases of retinal angiomatous proliferation. ArchOphthalmol 2005; 123(12):1644–50.

62. Boscia F, Parodi MB, Furino C, Reibaldi M, Sborgia C.Photodynamic therapy with verteporfin for retinal angio-matous proliferation. Graefes Arch Clin Exp Ophthalmol2006; 244(10):1224–32.

63. Nicolo M, Ghiglione D, Lai S, Calabria G. Retinal angio-matous proliferation treated by intravitreal triamcinoloneand photodynamic therapywith verteporfin. Graefes ArchClin Exp Ophthalmol 2006; 244(10):1336–8.

64. Schmidt-Erfurth U, Hasan T. Mechanisms of action ofphotodynamic therapy with verteporfin for the treatmentof age-related macular degeneration. Surv Ophthalmol2000; 45(3):195–214.

65. Treatment of age-related macular degeneration withphotodynamic therapy (TAP) Study Group. Photody-namic therapy of subfoveal choroidal neovascularizationin age-related macular degeneration with verteporfin:one-year results of 2 randomized clinical trials–TAPreport. Arch Ophthalmol 1999; 117(10):1329–45.

66. Lim JI, Sternberg P, Jr., Capone A, Jr., Aaberg TM, Sr.,Gilman JP. Selective use of indocyanine green angio-graphy for occult choroidal neovascularization. AmJ Ophthalmol 1995; 120(1):75–82.

67. Baumal CR, Reichel E, Duker JS, Wong J, Puliafito CA.Indocyanine green hyperfluorescence associated withserous retinal pigment epithelial detachment in age-related macular degeneration. Ophthalmology 1997;104(5):761–9.

68. Yannuzzi LA, Hope-Ross M, Slakter JS, et al. Analysis ofvascularized pigment epithelial detachments using indo-cyanine green videoangiography. Retina 1994; 14(2):99–113.

69. Guyer DR, Yannuzzi LA, Slakter JS, Sorenson JA, Hope-Ross M, Orlock DR. Digital indocyanine-green videoan-giography of occult choroidal neovascularization.Ophthalmology 1994; 101(10):1727–35 (discussion 1735-7).


70. Chang B, Yannuzzi LA, Ladas ID, Guyer DR, Slakter JS,Sorenson JA. Choroidal neovascularization in second eyesof patients with unilateral exudative age-related maculardegeneration. Ophthalmology 1995; 102(9):1380–6.

71. Chang TS, Freund KB, de la Cruz Z, Yannuzzi LA,Green WR. Clinicopathologic correlation of choroidalneovascularization demonstrated by indocyanine greenangiography in a patient with retention of good visionfor almost four years. Retina 1994; 14(2):114–24.

72. Watzke RC, Klein ML, Hiner CJ, Chan BK, Kraemer DF.A comparison of stereoscopic fluorescein angiographywith indocyanine green videoangiography in age-relatedmacular degeneration. Ophthalmology 2000; 107(8):1601–6.

73. Lee BL, Lim JI, Grossniklaus HE. Clinicopathologicfeatures of indocyanine green angiography-imaged, surgi-cally excised choroidal neovascular membranes. Retina1996; 16(1):64–9.

74. Guyer DR, Yannuzzi LA, Ladas I, Slakter JS, Sorenson JA,Orlock D. Indocyanine green-guided laser photocoagu-lation of focal spots at the edge of plaques of choroidalneovascularization. Arch Ophthalmol 1996; 114(6):693–7.

75. Helbig H, Niederberger H, Valmaggia C, Bischoff P. Simul-taneous fluorescein and indocyanine green angiographyfor exudative macular degeneration. Klin MonatsblAugenheilkd 2005; 222(3):202–5.

76. Yanagi Y, Tamaki Y, Sekine H. Fluorescein angiography-guided indocyanine green angiography for the detectionof feeder vessels in subfoveal choroidal neovasculariza-tion. Eye 2004; 18(5):474–7.

77. Sorenson JA, Yannuzzi LA, Slakter JS, Guyer DR, Ho AC,Orlock DA. A pilot study of digital indocyanine greenvideoangiography for recurrent occult choroidal neovas-cularization in age-related macular degeneration. ArchOphthalmol 1994; 112(4):473–9.

78. Reichel E, Pollock DA, Duker JS, Puliafito CA. Indo-cyanine green angiography for recurrent choroidalneovascularization in age-related macular degeneration.Ophthalmic Surg Lasers 1995; 26(6):513–8.

79. Chen CJ, Chen LJ, Miller KR. Clinical significance ofpostlaser indocyanine green angiographic hot spots inage-related macular degeneration. Ophthalmology 1999;106(5):925–9 (discussion 929–31).

80. Photodynamic Therapy Study Group. Verteporfin therapyof subfoveal choroidal neovascularization in age-relatedmacular degeneration: two-year results of a randomizedclinical trial including lesions with occult with no classicchoroidal neovascularization-verteporfin in photody-namic therapy report 2. Am J Ophthalmol 2001; 131(5):541–60.

81. Gragoudas ES, Adamis AP, Cunningham ET, Jr.,Feinsod M, Guyer DR. Pegaptanib for neovascular age-related macular degeneration. N Engl J Med 2004;351(27):2805–16.

82. Avery RL, Pieramici DJ, Rabena MD, Castellarin AA,Nasir MA, Giust MJ. Intravitreal bevacizumab (Avastin)for neovascular age-related macular degeneration.Ophthalmology 2006; 113(3):363–72. e5.

83. Michels S, Rosenfeld PJ, Puliafito CA, Marcus EN,Venkatraman AS. Systemic bevacizumab (Avastin)therapy for neovascular age-related macular degenerationtwelve-week results of an uncontrolled open-label clinicalstudy. Ophthalmology 2005; 112(6):1035–47.

84. Spaide RF, Laud K, Fine HF, et al. Intravitreal bevaci-zumab treatment of choroidal neovascularizationsecondary to age-related macular degeneration. Retina2006; 26(4):383–90.

85. Heier JS, Antoszyk AN, Pavan PR, et al. Ranibizumab fortreatment of neovascular age-related macular degener-ation: a phase I/II multicenter, controlled, multidosestudy. Ophthalmology 2006; 113(4):642. e1–4.

86. Rosenfeld PJ, Heier JS, Hantsbarger G, Shams N. Toler-ability and efficacy of multiple escalating doses ofranibizumab (Lucentis) for neovascular age-relatedmacular degeneration. Ophthalmology 2006; 113(4):632.e1.

87. Slakter JS, Yannuzzi LA, Sorenson JA, Guyer DR, Ho AC,Orlock DA. A pilot study of indocyanine green videoan-giography-guided laser photocoagulation of occultchoroidal neovascularization in age-related maculardegeneration. Arch Ophthalmol 1994; 112(4):465–72.

88. Regillo CD, Benson WE, Maguire JI, Annesley WH, Jr.Indocyanine green angiography and occult choroidalneovascularization. Ophthalmology 1994; 101(2):280–8.

89. Lim JI, Aaberg TM, Capone A, Jr., Sternberg P, Jr. Indo-cyanine green angiography-guided photocoagulation ofchoroidal neovascularization associated with retinalpigment epithelial detachment. Am J Ophthalmol 1997;123(4):524–32.

90. Da Pozzo S, Parodi MB, Ravalico G. A pilot study ofICG-guided laser photocoagulation for occult choroidalneovascularization presenting as a focal spot in age-related macular degeneration. Int Ophthalmol 2001;24(4):187–94.

91. Ho AC, Yannuzzi LA, Guyer DR, Slakter JS, Sorenson JA,Orlock DA. Intraretinal leakage of indocyanine green dye.Ophthalmology 1994; 101(3):534–41.

92. Reichel E, Puliafito CA, Duker JS, Guyer DR. Indocyaninegreen dye-enhanced diode laser photocoagulation ofpoorly defined subfoveal choroidal neovascularization.Ophthalmic Surg 1994; 25(3):195–201.

93. Obana A, Gohto Y, Nishiguchi K, Miki T, Nishi S, Asada A.A retrospective pilot study of indocyanine green enhanceddiode laser photocoagulation for subfoveal choroidal neo-vascularization associated with age-related maculardegeneration. Jpn J Ophthalmol 2000; 44(6):668–76.

94. Arevalo JF, Mendoza AJ, Fernandez CF. Indocyaninegreen-mediated photothrombosis with and without intra-vitreal triamcinolone acetonide for subfoveal choroidalneovascularization in age-related macular degeneration:a pilot study. Retina 2005; 25(6):719–26.

95. Arevalo JF, Garcia RA, Mendoza AJ. Indocyanine green-mediated photothrombosis with intravitreal triamcinoloneacetonide for subfoveal choroidal neovascularization inage-related macular degeneration. Graefes Arch Clin ExpOphthalmol 2005; 243(11):1180–5.

96. Maberley DA, Cruess AF. Indocyanine green angio-graphy: an evaluation of image enhancement for theidentification of occult choroidal neovascular membranes.Retina 1999; 19(1):37–44.

97. Spaide RF, Orlock DA, Herrmann-Delemazure B, et al.Wide-angle indocyanine green angiography. Retina 1998;18(1):44–9.

98. Staurenghi G, Viola F, Mainster MA, Graham RD,Harrington PG. Scanning laser ophthalmoscopy andangiography with a wide-field contact lens system. ArchOphthalmol 2005; 123(2):244–52.

99. Miki T, Shiraki K, Kohno T, Moriwaki M, Obana A.Computer assisted image analysis using the subtractionmethod in indocyanine green angiography. EurJ Ophthalmol 1996; 6(1):30–8.

100. Spaide RF, Orlock D, Yannuzzi L, et al. Digital subtractionindocyanine green angiography of occult choroidalneovascularization. Ophthalmology 1998; 105(4):680–8.

174 OLIVER ET AL.

101. Matsumoto M, Shiraki K, Obana A. Detection of choroidalneovascularization by subtraction indocyanine greenangiography. Osaka City Med J 2003; 49(2):85–91.

102. Staurenghi G, Orzalesi N, La Capria A, Aschero M.Laser treatment of feeder vessels in subfoveal choroidalneovascular membranes: a revisitation using dynamicindocyanine green angiography. Ophthalmology 1998;105(12):2297–305.

103. Hanutsaha P, Guyer DR, Yannuzzi LA, et al. Indocyanine-green videoangiography of drusen as a possible predictiveindicator of exudative maculopathy. Ophthalmology 1998;105(9):1632–6.

104. Pauleikhoff D, Spital G, Radermacher M, Brumm GA,Lommatzsch A, Bird AC. A fluorescein and indocyanine

green angiographic study of choriocapillaris in age-relatedmacular disease. Arch Ophthalmol 1999; 117(10):1353–8.

105. Gelisken F, Inhoffen W, Schneider U, Stroman GA,Kreissig I. Indocyanine green videoangiography of occultchoroidal neovascularization: a comparison of scanninglaser ophthalmoscope with high-resolution digital funduscamera. Retina 1998; 18(1):37–43.

106. Freund KB, Yannuzzi LA, Sorenson JA. Age-relatedmacular degeneration and choroidal neovascularization.Am J Ophthalmol 1993; 115(6):786–91.

107. Stanga PE, Lim JI, Hamilton P. Indocyanine green angio-graphy in chorioretinal diseases: indications andinterpretation: an evidence-based update. Ophthalmology2003; 110(1):15–21 (quiz 22–3).


10

Optical Coherence Tomography in the Evaluation andManagement of Age-Related Macular DegenerationDavid Eichenbaum and Elias ReichelNew England Eye Center, Tufts University School of Medicine, Boston, Massachusetts, U.S.A.

The widespread use of optical coherence tomography(OCT) has changed the way ophthalmologists evalu-ate and treat age-related macular degeneration(AMD). OCT has been added to the armamentariumof macular imaging that now includes color fundusphotography, fluorescein angiography (FA), and indo-cyanine green angiography. Although each of thesemodalities is important in the management of maculardegeneration, OCT provides useful informationregarding retinal structure. Cross-sectional imagingwith commercially available units gives an axial res-olution of 10 to 15 mm, and ultrahigh resolution OCT(UHR-OCT) provides 2- to 3-mm resolution. This datacan be used in the diagnosis and management ofAMD; reliance on OCT as part of the decisionmaking process in treatment with different anti-vascular endothelial growth factor (VEGF) agents israpidly becoming standard of care.

OCT IMAGING PRINCIPLES ANDTHE NORMAL OCT IMAGE

The different layers of the retina have a characteristicappearance on the OCT scan. The principle used increating OCT images is Michelson interferometry,which uses the property of light passing through theeye and producing different reflections from differentcell layers. A split beam of infrared light in commer-cially available OCTunits reflects light from the layersof the retina and interacts with light reflected from areference mirror (1). The interference patternproduced is digitally processed and the false-colormap shown on the scan is a representation of thereflection characteristics of the cells in each layer,with hyperreflective areas being bright and hypore-flective areas being dark. When compared to in vivospecimens, layers of relative high reflectivity corre-spond to horizontally aligned retinal components (2).

The innermost layers of the retina, the nerve fiberlayer, and ganglion cells, are bright on the false-colormap. As the scan progresses to the outer retina, the

more densely packed nuclear layers are hyporeflectiveand dark, and the horizontally oriented plexiformlayers are hyperreflective and bright. In the outermostretina, the photoreceptors are hyporeflective, althoughthe junction between the inner segment (IS) and outersegment (OS), as best visualized with UHR-OCT, isbright. The outermost layers, those further distal thanthe OS of the photoreceptors, are hyperreflective andbright, and pathologic studies have shown the tissuesresponsible for this signal are the basement membraneof the retinal pigment epithelium (RPE) and the innerchoroid (3).

The OCT image also reflects the topographicalshape of the normal retina, and the OCT software cancombine the topographical radial scans to create avolumetric measurement of retinal tissue. The retinalthickness is displayed on a separate false color map,with cool colors (greens and blues) representing areasof less retinal volume, and warm colors (yellows andreds) representing areas of greater retinal volume. Thenormal foveal depression of the umbo is visible ontopographic and volumetric OCT, and the quantitativethickness of a normal fovea can be compared withknown approximate normal central field (central1-mmdiameter) thickness, which is about 175 mm (4,5).

OCT in Assessment of ChoroidalNeovascularization in AMDOCT can help characterize retinal pathology, evenwhen this information is difficult to discern on clinicalexamination or angiography. It is possible to define thelocation of choroidal neovascular membranes above orbelow the RPE. Solid fibrous tissue can be differen-tiated from subretinal fluid when these findings maybe angiographically equivocal. Other features ofAMD, including cystoid macular edema (CME),drusenoid RPE detachments and RPE tears can beimaged by OCT (6).

Optical coherence is useful for quantitativeassessment of retinal thickness and subretinal fluidwhen associated with choroidal neovascularization(CNV). A characteristic appearance of CNV has also

been described, consisting of thickening and fragmen-tation of the reflective layer corresponding to the RPEand choriocapillaris (Figs. 1–3) (7). The extent andlocation of subretinal fluid associated with CNV canbe used to assess whether the pathology is subfoveal,as long as there is preservation of some fovealarchitecture (8).

As noted on clinical examination and FA, CME isfrequently associated with CNV in wet AMD (9).However, the presence of CME may be difficult todefinitively diagnose through those modalities alone.In addition to imaging subretinal fluid, OCT is effec-tive in identifying intraretinal edema, compared to

both clinical stereoscopic images (10) and FA (3,11).The appearance of CME on OCT images is seen ashyporeflective, dark spaces within retinal tissue. Itspresence is important clinically, since CME as seen onOCT scan in wet AMD correlates with decreasedvisual acuity (Figs. 4–7) (9).

RPE detachments and sub-RPE neovasculariza-tion has been associated with occult CNV in AMDas defined histopathologically. This finding has beencorroborated in studies using OCT imaging. In oneseries, new neovascular AMD lesions were charac-terized as occult or classic according to theirangiographic findings (12). OCT scans of those samelesions revealed subretinal opacities separate fromthe RPE present in over 87% of lesions characterizedas classic and only 13% of lesions characterized asoccult. OCT findings consistent with RPE detachmentwere present in none of the lesions characterized asclassic and in one-third of those characterized as

Figure 1 Color picture of choroidal neovascularization in apatient’s left eye.

Figure 2 Fluorescein angiography of the eye in Figure 1exhibiting classic choroidal neovascularization.

Figure 3 Cross-sectional optical coherence tomographyappearance of eye in Figures 1 and 2 revealing hyporeflective

subretinal fluid and thickening of adjacent layer just anterior to theintensely hyperreflective retinal pigment epithelium. There is

essentially preservation of the normal foveal anatomy and onlya minimal increase in retinal thickness.

Figure 4 Transit phase fluorescein angiography of a right eyeexhibiting a large neovascular membrane.

178 EICHENBAUM AND REICHEL

occult (13). The relationship of CNV and RPE detach-ments has been further elucidated in another OCTseries, and a pattern of double RPE detachmentsseparated by a notch, as well as highly reflectivetissue beneath the dome of the RPE detachmenthave been correlated with CNV that was observedon angiography (Figs. 8–11) (14).

OCT in Assessment of Non-exudativeMacular DegenerationAlthough the utility of commercially available OCTscanning in the clinical diagnostic setting has beenexamined mostly for neovascular AMD, there arecharacteristic OCT findings in nonneovascularforms of AMD. “Drusenoid” RPE detachments,which appear angiographically as staining of drusen,

are documented by OCT as elevations of the pigmentepithelium itself (Figs. 12–14) (15).

The UHR-OCT is an experimental imagingmodality which builds upon the interferometryprinciples of commercially available OCT imagingdevices. The standard OCT axial resolution of 10 to15 mm is improved to 3 mm using a femtosecondtitanium-sapphire laser with a broader bandwidth.This improvement in resolution allows for betterdelineation and characterization of changes withinthe intraretinal layers, especially the IS and OSphotoreceptors and the IS–OS junction (also knownas the external limiting membrane) (16,17), which arethe site of many early changes in AMD. In UHR-OCT,the outermost retina is seen as the hyperreflective RPEunderlying hyporeflective OS photoreceptors, which

Figure 5 Recirculation phase fluorescein angiography of thesame eye as Figure 4 revealing late leakage and cystoid macular

edema from the membrane.

Figure 6 Cross-sectional optical coherence tomography (OCT)

of the lesion in Figures 4 and 5, revealing large hyporeflectivespaces typical of cystoid edema. It is present with thickening and

fragmentation of the hyperreflective layer corresponding to theretinal pigment epithelium, which is the characteristic OCT

appearance of choroidal neovascularization.

258

360

274382407295210

316

225

Microns

Figure 7 Topographical optical coherence tomography map of

the lesion shown in Figures 4–6, showing marked thickening ofretinal tissue.

Figure 8 Color picture of a left eye showing a complexneovascular membrane. Lesion components include a large

retinal pigment epithelium detachment, subretinal blood, andsubretinal exudate.

10: OCT IN THE EVALUATION AND MANAGEMENT OF AMD 179

are in turn distinguished from hyporeflective IS photo-receptors by the intensely hyperreflective IS–OSjunction.

UHR-OCT has been used to evaluate eyes withdry AMD and subtle patterns associated with drusenhave been observed. One pattern shows changessimilar to those seen with drusen in commerciallyavailable OCT, with the RPE excrescences overlyingreflective material consistent with drusen. A secondpatter seen on UHR-OCT in dry AMD is a saw-toothed

configuration or bunching of the RPE. This pattern isseen associated with atrophy of the inner and outerphotoreceptors and the outer nuclear layer, althoughthere is no change in retinal tissue further inward(Fig. 15A–C). The third UHR-OCT pattern in dryAMD shows discrete nodular drusen which actuallydisrupt as opposed to distort the RPE and are associ-ated with collections of reflective material. This thirdpattern corresponds to large hard drusen that areobserved on clinical examination. Patients with dryAMD can have all three patterns present, or a varietyof combinations. Avery small percentage of patients inthe series of dry AMD eyes studied with UHR-OCTwere noted to have findings consistent with earlyCNV, despite no clinical or angiographic evidence

Figure 9 Transit phase of the eye shown in Figure 12. There issome early filling of a superotemporal retinal pigment epithelium

detachment with relative hypofluorescence of the central,neovascular component of the lesion.

Figure 10 Late recirculation phase of the eye shown in Figures

12 and 13. There has been filling of the retinal pigment epitheliumdetachment, as well as filling of cystoid spaces in the center of the

macula. There is leakage from the central and inferior occultchoroidal neovascularization and blockage from the subretinal

blood.

Figure 11 Cross-sectional optical coherence tomography of

the lesion in Figures 12–14. There is diffuse retinal thickeningand numerous hyporeflective spaces consistent with severe

cystoid macular edema. There is a small amount of subretinalfluid. The retinal pigment epithelium (RPE) detachment is very

apparent, underlying the retina and the hyperreflective bandcorresponding to the RPE.

Figure 12 Color picture of a right eye showing moderatepigment atrophy centrally and confluent soft drusen.


of CNV. This finding is particularly interesting, as veryearly diagnosis of neovascular AMDmay be facilitatedby UHR imaging (18).

Geographic atrophy has also been examined,both with conventional OCT and UHR-OCT.A typical pattern of geographic atrophy consisting ofenhanced reflectivity of the choroid and significantthinning of overlying retinal tissue has been describedusing conventional OCT. The bright, well demarcatedsignal of the choroid is felt to be due to loss of the RPE,a common feature of geographic atrophy that has beenwell described in histopathologic studies (19). Thesefindings are similar to those seen on UHR-OCT, withassociated generalized retinal thinning and increasedreflectivity from tissue below the absent RPE, which islikely choriocapillaris (19).

OCT in the Treatment of AMDThe ability of OCT to accurately map and quantitateretinal findings has added an entirely new dimensionfor monitoring the response of neovascular AMDto treatment. Prior to OCT, qualitative clinical exami-nation and FA were the only means by which onecould assess disease progression, stability, orregression. In addition to the quantitative accuracy ofOCT, its noninvasive nature, and absent risk of allergicresponse, OCT appears to be surpassing the use ofangiographic techniques as the test of choice for visit-to-visit monitoring of neovascular AMD. (See alsoChapter 11 on quantitative imaging).

Visudyne photodynamic therapy (PDT), whichhas been a Food and Drug Administration (FDA)approved treatment for neovascular AMD since 2000,was the first new AMD treatment introduced in the eraof OCT, and the first pharmacologic treatment for CNVsecondary to AMD. OCT findings 6- and 12-monthfollowing the initiation of CNV treatment with PDTshow macular thickness declining after treatment, butdid not show a decrease in the thickness of the CNVfollowing treatment. However, that same study alsofound that OCT has an excellent sensitivity but only afair specificity for monitoring CNV activity, though itserved as a useful adjunct for verifying intraretinal orsubretinal fluid, especially when angiography wasinconclusive (20). Many of PDT effects have beenshown by OCT, including a transient increase inintraretinal and subretinal fluid in the first weekafter treatment (21), which has also been describedas an early response stage in an OCT gradingsystem developed to monitor treatment by PDTand help clarify angiographic changes seen aftertreatment (22).

Pegaptanib (Macugenw) is a new class of drugfor neovascular AMD, a molecular agent known asan aptamer, which targets VEGF isoform 165. OCThas been used to evaluate patients’ response to thistherapy. Since the introduction of pegaptanib treat-ment, OCT has been used to document complicationsof therapy, and RPE rips have been documented withOCT after a single pegaptanib treatment of CNVwith turbid pigment epithelial detachments (23,24).(It is important to note that RPE rips have also beendocumentedwith thermal laser treatment (25) andPDTtreatment (26) of CNV, as well as other anti-VEGFagents including bevacizumab and ranibizumab.)Evaluation of eyes treated with a single intravitrealinjection of pegaptanib showed that there was nodifference in OCTanatomy compared to baseline (27).

OCT has played a pivotal role in the introductionof VEGF inhibitors in the treatment of neovascularAMD. Following the initial treatment of neovascularAMD with systemic bevacizumab (28), a full-lengthantibody approved by the FDA for the treatment of

Figure 13 Recirculation phase fluorescein angiography of theeye shown in Figure 12. There is staining of the soft drusen and

transmission of dye fluorescence through the retinal pigmentepithelium atrophy.

Figure 14 Cross-sectional optical coherence tomography ofthe eye shown in Figures 13 and 14. There are sub-RPE

hyporeflective spaces typical of “drusenoid” RPE detachments.There is preservation of the retinal architecture. Of note, there is

a mild epiretinal membrane.


colon cancer, studies of intravitreal bevacizumabemerged in the ophthalmic literature supporting thisoff-label treatment. The results of intravitreal bevaci-zumab were also striking, with initially a single casereport (29) followed by small prospective studiesshowing OCT evidence of improvement or resolutionof intraretinal, subretinal, and sub-RPE fluid in alarge percentage of patients as early as four weeks

after treatment that was also associated with improve-ment in visual acuity. Treatment effects persisted up to12 weeks after treatment (30–32). In all of thesestudies, OCT results supported anatomical efficacyof treatment.

The PrONTO study, a small, single center,prospective, nonrandomized efficacy study, showedthat patients given a “loading dose” of ranibizumab

(C)

(A) (B)

200μm

200μm

Figure 15 (A) Color photo of a left eyewith retinal pigment epithelium (RPE)

changes and pigment clumping. (B) Ultra-high resolution optical coherence

tomography (UHR-OCT) showing thicken-ing and bunching of the RPE. There is a

loss of both the outer segment (OS) andinner segment (IS) of the photoreceptors,

with a fragmenting of the IS–OS junction.Intraretinal pigment migration is noted by

the arrow. (C) Stratus OCT showing lessdistinction of intraretinal pigment migration.

The photoreceptors cannot be distin-guished, although the thickening of the

RPE remains apparent. Abbreviations: IS,inner segment; OS, outer segment Source:

From Ref. 18.

Figure 16 Optical coherence tomography of the baseline lesion,showing hyperreflective retinal pigment epithelium (RPE) detach-

ments, thickening of the tissue adjacent to the RPE, andintraretinal hyporeflective fluid with thickening of retinal tissue.

The first treatment with intravitreal ranibizumab is administered.

Figure 17 Optical coherence tomography one month after thefourth ranibizumab treatment. There is complete resolution of the

intraretinal and sub-retinal pigment epithelium (RPE) fluid, nosubretinal scarring, and a normal contour to the retinal tissue.

Note that the RPE band has a normal thickness to it.


monthly for three months and then given a custo-mized dosing scheme based upon visual acuity andOCT findings had equivalent efficacy compared tomonthly dosing with ranibizumab after one year offollow-up (33). This study is remarkable because itallowed a variable dosing regimen for all types oflesions at the discretion of the treating physicianthat was primarily guided by OCT. The followingsequence (Figs. 16 and 17) is an example of aranibizumab treatment of CNV with OCT used asguidance to retreat.

CONCLUSIONS

OCT has the ability to verify the presence ofneovascular disease and quantitatively evaluatetreatment response. The utility of OCT as a guidefor treatment is suggested by early clinical experi-ence from many centers and results from thePrONTO study. Further studies using UHR-OCTmay determine the role of screening high risk eyesfor the early detection of neovascular AMD. OCThas been utilized as supportive data in some trialsfor the treatment of macular degeneration, though itsincorporation as a key endpoint in the evaluation oftreatments for neovascular AMD has yet to beconclusively established. OCT will continue to be akey component in the evaluation, treatment anddevelopment of new therapeutic modalities inpatients with AMD.

SUMMARY POINTS

& Cross-sectional imaging with commerciallyavailable units give an axial resolution of 10to 15 mm, and UHR-OCT provides 2- to 3-mmresolution.

& OCT is useful to verify the presence of neovasculardisease and quantitatively evaluate treatmentresponse as suggested by clinical trials experience.

& Further studies using UHR-OCT may determinethe role of screening high risk eyes for the earlydetection of neovascular AMD.

& OCT will continue to be a key component in theevaluation, treatment and development of newtherapeutic modalities in patients with AMD.

REFERENCES

1. Huang D, Swanson EA, Lin CP, et al. Optical coherencetomography. Science 1991; 254:1178–81.

2. Toth CA, Narayan DG, Boppart SA, et al. A comparison ofretinal morphology viewed by optical coherencetomography and by light microscopy. Arch Ophthalmol1997; 115:1425–8.

3. Ghazi N, Dibernardo C, Ying HS, Mori K, Gehlbach PL.Optical coherence tomography of enucleated human eyespecimens with histological correlation: origin of the outer“red line”. Am J Ophthalmol 2006; 141:719–26.

4. Hee MR, Puliafito CA, Duker JS, et al. Topography ofdiabetic macular edema with optical coherence tomo-graphy. Ophthalmology 1998; 105:360–70.

5. Massin P, Erginay A, Haouchine B, Mehidi AB, Paques M,Gaudric A. Retinal thickness in healthy and diabeticsubjects measured using optical coherence tomographymapping software. Eur J Ophthalmol 2002; 12:102–8.

6. Voo I, Mavrofrides EC, Puliafito CA. Clinical applications ofoptical coherence tomography for the diagnosis andmanagement of macular diseases. Ophthalmol Clin NorthAm 2004; 17:21–31.

7. Hee MR, Baumal CR, Puliafito CA, et al. Optical coherencetomography of age-related macular degeneration and chor-oidal neovascularization. Ophthalmology 1996; 103:1260–70.

8. Jaffe GJ, Caprioli J. Optical coherence tomography to detectand manage retinal disease and glaucoma. Am J Opthalmol2004; 137:156–69.

9. Ting TD, Oh M, Cox TA, Meyer CH, Toth CA. Decreasedvisual acuity associated with cystoid macular edema inneovascular age-related macular degeneration. ArchOphthalmol 2002; 120:731–7.

10. Strom C, Sander B, Larsen N, Larsen M, Lund-Andersen H.Diabetic macular edema assessed with optical coherencetomography and stereo fundus photography. InvestOphthalmol Vis Sci 2002; 43:241–5.

11. Antcliff RJ, Stanford MR, Chauhan DS, et al. Comparisonbetween optical coherence tomography and fundus fluores-cein angiography for thedetection of cystoidmacular edemain patients with uveitis. Ophthalmology 2000; 107:593–9.

12. Treatment of Age-Related Macular Degeneration withPhotodynamic Therapy (TAP) Study Group. Photodynamictherapy of subfoveal choroidal neovascularization in age-related macular degeneration with verteporfinone-yearresults of 2 randomized clinical trials-TAP report. ArchOphthalmol 1999; 117:1329–45.

13. Hughes EH, Khan J, Patel N, Kashani S, Chong NV. In vivodemonstration of the differences between classic and occultneovascularization using optical coherence tomography.Am J Ophtalmol 2005; 139:344–6.

14. Sato T, Iida T, Hagimura N, Kishi S. Correlation of opticalcoherence tomographywith angiography in retinal pigmentepithelial detachments associated with age-related maculardegeneration. Retina 2004; 24:910–4.

15. Emfietzoglou I, Grigoropoulos V, Kipioti A, Alimisi S,Theodossiadis PG, Theodossiadis GP. Optical coherencetomography appearance of “drusenoid” pigment epithelialdetachments. Ophthalmic Surg Lasers Imaging 2005;36:147–50.

16. DrexlerW,Morgner U, Ghanta RK, Kartner FX, Schuman JS,Fujimoto JG. Ultrahigh-resolution optical coherencetomography. Nat Med 2001; 7:502–7.

17. Drexler W, Sattmann H, Hermann B, et al. Enhancedvisualization of macular pathology with the use of ultra-high-resolution optical coherence tomography. ArchOphthalmol 2003; 121:695–706.

18. Pieroni CG, Witkin AJ, Ko TH, et al. Ultrahigh resolutionoptical coherence tomography in non-exudative age relatedmacular degeneration. Br J Ophthalmol 2006; 90:191–7.

19. Hassenstein A, Ruhl R, Richard G. Optical coherencetomography in geographic atrophy—a clinicopathologicalstudy. Klin Monatsbl Augenheilkd 2001; 218:503–9.


20. Salinas-Alaman A, Garcıa-Layana A,MaldonadoMJ, Sainz-Gomez C, Alvarez-Vidal A. Using optical coherence tomo-graphy to monitor photodynamic therapy in age relatedmacular degeneration. Am J Ophthalmol 2005; 140:23–8.

21. Ozdemir H, Karacorlu SA, Karacorlu M. Early opticalcoherence tomography changes after photodynamictherapy in patients with age related macular degeneration.Am J Ophthalmol 2006; 141:574–6.

22. Rogers AH,Martidis A, Greenberg PB, Puliafito CA. Opticalcoherence tomography findings following photodynamictherapy of choroidal neovascularization. Am J Ophthalmol2002; 134:566–76.

23. Dhalla MS, Blinder KJ, Tewari A, Hariprasad SM, Apte RS.Retinal pigment epithelial tears following intravitrealpegaptanib sodium. Am J Ophthalmol 2006; 141:752–4.

24. Singh RP, Sears JE. Retinal pigment epithelial tears afterpegaptanib injection for exudative age-related maculardegeneration. Am J Ophthalmol 2006; 142:160–2.

25. Gass JDM. Retinal pigment epithelial rip during krypton redlaser photocoagulation. Am J Ophthalmol 1984; 98:700–6.

26. Gelisken F, Inhoffen W, Partsch M, Schneider U, Kreissig I.Retinal pigment epithelial tear after photodynamic therapyfor choroidal neovascularization. Am J Ophthalmol 2001;131:518–20.

27. Schuman S, Rogers AH, Duker JS, Reichel E, Baumal CR.Six-week outcomes after pegaptanib. Ophthalmology 2006;113:501.

28. Michels S, Rosenfeld PJ, Puliafito CA, Marcus EN,Venkatraman AS. Systemic bevacizumab (Avastin) therapyfor neovascular age-related macular degeneration: twelve-week results of an uncontrolled open-label clinical study.Ophthalmology 2005; 112:1035–47.

29. Rosenfeld PJ, Moshfeghi AA, Puliafito CA. Optical coher-ence tomography findings after an intravitreal injection ofbevacizumab (Avastin) for neovascular age-relatedmaculardegeneration. Ophthalmic Surg Lasers Imaging 2005;36:331–5.

30. Avery RL, Pieramici DJ, Rabena MD, Castellarin AA,Nasir MA, Giust MJ. Intravitreal bevacizumab (Avastin)for neovascular age-related macular degeneration.Ophthalmology 2006; 113:363–72.

31. Spaide RF, Laud K, Fine HF, et al. Intravitreal bevacizumabtreatment of choroidal neovascularization secondary toage-related macular degeneration. Retina 2006; 26:383–90.

32. Rich RM, Rosenfeld PJ, Puliafito CA, et al. Short-termsafety and efficacy of intravitreal bevacizumab (Avastin)for neovascular age-related macular degeneration. Retina2006; 26:495–511.

33. Rosenfeld PJ. An OCT-guided variable-dosing regimenwith Lucentis(TM) (ranibizumab) in neovascular AMD:one year results from the PrONTO study. In: Program andabstracts of the 24th Annual American Society of RetinaSpecialists and 6th Annual European Vitreoretinal SocietyMeeting. Cannes, France, September 9–13, 2006.


11

Quantitative Retinal ImagingDaniel D. EsmailiDoheny Eye Institute and Department of Ophthalmology, Keck School of Medicine,


Roya H. GhafouriDepartment of Ophthalmology, Boston University Medical Center, Boston University School of Medicine,

Boston, Massachusetts, U.S.A.

Usha ChakravarthyThe Queen’s University of Belfast and Royal Hospitals, Belfast, Northern Ireland



INTRODUCTION

Historically, clinical ophthalmology developedrapidly as the transparency of the ocular mediapermitted direct examination of the intraocularstructures. The development of methods to examinethe fundus of the eye considerably improved theunderstanding of retinal anatomy, structure, andcirculation. Consequently, great advantages weregained in the diagnosis of retinal disease and compre-hension of systemic disorders.

The development of innovative imaging tech-niques such as fluorescein and indocyanine greenangiography followed by tomographic image capturehas further improved the examination of the retinaand its vasculature. These advances have allowedophthalmologists to better describe retinal diseasesand to develop classification systems in areas such asdiabetic retinal disease and age-related maculardegeneration (AMD). Investigators, for example,have developed systematic grading systems based onstereoscopic examination of color fundus and angio-graphic images to characterize disease progressionand response to therapy (1,2). However, these classi-fication systems are descriptive and categorical. Bycontrast, other medical specialties have made enor-mous strides in the development of quantitativemethods to both describe and categorize diseasesand to assess the efficacy of intervention. Forexample, an internist can use quantifiable markerssuch as blood pressure, cholesterol, and cardiac ejec-tion fraction to predict patients at highest risk for acardiovascular event and then offer preventive

treatments. Likewise, neurologists can image thebrain with precision and accuracy to identify volu-metric changes in tumor size and contour.

Similar advances in retinal imaging are clearlypossible and a number of quantitative methods arecurrently being developed. To understand thecomplexities of the tissue layers and vasculature thatcan be imaged, it is necessary to review the variousimaging modalities and techniques for interpretationthat are currently available.

FUNDUS PHOTOGRAPHY

Color fundus photography has been a significant toolin the documentation of macular disease for manyyears (1,2). Clinical trials have used photographicdocumentation to ensure quality assurance and adher-ence to standards. Advances in image capture,allowing the transition from film-based photographyto high-resolution digital acquisition, have resulted inmany advantages. First, retrieval of information iseasier. Second, computer-assisted image analysis canbe easily applied. Third, parametric descriptors can bedeveloped. All these contribute toward providinggreater accuracy, objectivity, and reproducibility.

Application of quantitative techniques to evaluateeyes with AMD is becoming a reality. Quantitativetechniques are useful for evaluating the type andnumber of drusen and retinal pigment epitheliumpigmentary changes, both of which are key features ofearly AMD (3,4). The careful analysis of drusen number,size, area, and morphology allows clinicians to assess

disease severity and may aid in predicting conversionfrom non-neovascular to neovascular forms (5).

The goals of quantifying macular pathology suchas drusen are multifactorial. Quantified data provideobjective information that can be compared over time,giving an index of progression. Such data can be usedto evaluate the response to treatment as well as tograde disease progression. Quantification of drusenwould also allow for better designed epidemiologicaland clinical studies of the natural history of AMD.

Manual grading is the oldest form of photo-graphic quantitation and is accurate, although laborand resource intensive (1,2). Currently, several semi-automated and automated modalities have beendeveloped to address the shortcomings of manualgrading (6–9). Shin and coworkers describe a super-vised method of automated drusen grading withgood correlation to manual grading, with interclasscorrelation coefficients of 0.92 and 0.93 (10). Semi-quantitative methods have also been shown to havegood interobserver reproducibility by graders atdifferent institutions (11).

Several limitations have prevented the wide-spread adoption of such technology. To date, noautomated system free of observer supervision existsthat can accurately and reproducibly assess drusenburden. Obstacles that have limited progress in thisfield include difficulties in distinguishing featuressuch as soft drusen with indistinct borders fromother pale retinal lesions. The presence of branchingand crossing points in retinal blood vessels also detractfrom detection of drusen as the former produce darkhaloes around the lighter retinal background. Thisphenomenon causes lesion thresholds to be calculatedbelow retinal backgrounds, thus creating a propensityfor the false detection of drusen (8).

A major technical obstacle has been accountingfor differences in macular reflectivity (6). As onemoves toward the central macula, normal backgroundreflectivity decreases in intensity. Thus, if there existedidentical soft drusen, with one located in the centralmacula and the other parafoveally, the difference innormal background reflectance would affect thethreshold at which the drusen could be detected.Semi-automated methods based on the geometry offundus reflectance as well as newer, more automatedtechnology utilizing mathematical modeling to recon-struct and then level the background reflectance arebeing explored to overcome such limitations (6,12).

The lack of true automation may limit the utilityof drusen quantification in clinical practice. However,an automated method that is precise and reproduciblewould allow large-scale population-based studies tobe performed with fewer resource implications.Furthermore, the technology could also be applied intrials assessing therapeutic interventions.

OPTICAL COHERENCE TOMOGRAPHY

Optical coherence tomography (OCT) was firstdescribed by Huang and colleagues as a method ofutilizing near infrared light to provide a noninvasivemeans of evaluating ocular structures including theretina (13). In many ways, OCT is an ideal modality forobtaining quantifiable data for the evaluation of retinalpathology (14). The current generation of OCT scan-ners is able to provide topographic numericaldeterminations of retinal thickness, which in recentyears has become an increasingly popular means ofevaluating disease and, particularly, the localization offluid within the different tissue layers of the fundus(15,16).

The role of OCT in quantifying changes due toAMD has yet to be established. This may partly be dueto the fact that OCT is a relatively new modality, andthat until recent years, the drive to create a numericalindex of disease was perceived as irrelevant sincetreatment options were less than satisfactory.Currently, new therapeutic modalities leading toimpressive improvements in visual acuity have beenintroduced in the management of neovascular AMD.OCT is now recognized as having an important role inthe measurement of retinal thickness and subretinalpathology in monitoring choroidal neovascularization(CNV) activity after appropriate therapy (17). With anincreasing number of treatment options availableincluding combination therapies, a quantifiable under-standing of the therapeutic response will be needed toallow meaningful comparisons of the morphologicaland functional outcomes.

The potential role of quantitative OCT in neo-vascular AMD is substantial. Automated imageanalysis algorithms are being developed to betterdefine retinal layers (18). Software like the OCTORsystem developed at the Doheny Eye Institute allowsfor delineation of retinal layers as well as pathologicentities such as subretinal fluid, area occupied bychoroidal neovascular membranes, and the size ofserous retinal pigment epithelial detachments. Thissoftware is currently employed as a research tool toevaluate the efficacy of various treatment modalities inneovascular AMD.

FLUORESCEIN ANGIOGRAPHY

Fluorescein angiography (FA) is a well-establishedmodality for assessment of neovascular AMD. Quan-titative FA can also provide similar benefits as OCT inthe management and treatment of AMD. Currentsoftware allows for the measurement of neovascularmembrane size, which is a parameter that can influ-ence the therapeutic outcome. One such example is theuse of photodynamic therapy (PDT) in the treatment

186 ESMAILI ET AL.

of occult CNV. Optimal outcomes were detected ineyes with lesions smaller than four disc diameters,while larger lesions were associated with a worseoutcome than natural history (19).

Semi-automated detection and quantificationof hyperfluorescent leakage has been described byPhillips et al., who manipulated digitized fluoresceinangiograms via gradient threshold and region-growing techniques to detect leakage (20). Thismethod superimposes paired images (early and lateangiograms) to produce a composite with the area ofleakage mapped onto a frame. A numerical value canthen be determined to describe a total edema value.

Chakravarthy et al. (21) have described quan-titative FA to assess the multiple pathologicalcomponents of CNV using a unique algorithm thatsubtracts the background and adds a term for thepositive change in fluorescence corrected to back-ground [positive fluorescence quotient (PFQ)]. Thiswork has shown that the PFQ for CNV and leakageare important determinants of visual function, andare better correlated with functional measures suchas visual acuity and reading ability.

Shah and coworkers used quantitative methodsto demonstrate the utility of quantitating the amountof hyperfluorescence intensity and area seen on FAimages of AMD patients before and after treatment forCNV (22). The investigators used image processing tomeasure the area of hyperfluorescence and fluor-escence intensity above background fluorescence.Values for each image were plotted against time afterdye injection to generate curves. Each area under thecurve (AUC) was calculated. The investigators foundan 11% decrease in AUC for fluorescence area and a32% decrease in AUC for fluorescence intensity in thepatients who clinically improved with treatment, butincreases of 131% and 292% in the patients whoworsened after PDT. Similarly, a 38% decrease inAUC for fluorescence intensity and a 19% decreasein AUC for fluorescence area were observed inpatients who received vascular endothelial growthfactor trap compared with increases of 66% (pZ0.004, Mann–Whitney U-test) and 21% (pZ0.07) forpatients who received placebo.

FA quantification is still being developed andwill undoubtedly serve as a means to furnish import-ant clinical information regarding the progression ofAMD. Like quantified fundus photography and OCT,developments in computer technology hold thepromise of creating truly automated methods tobetter understand retinal pathology.

MICROPERIMETRY

Microperimetry, also known as fundus perimetry, is anoninvasive method by which focal areas of retinal

sensitivity loss can be measured in those with maculardisease. Several systems are currently in practice,including the scanning laser ophthalmoscope (Roden-stock, Germany) and the Micro Perimeter 1 (NidekTechnologies, Italy) (23). By integrating real-timefundus imaging and computerized threshold peri-metry, these systems can provide fixation controlby accounting for eye movement disturbances thatcan be common in patients with central visual loss.As a result, this technology can provide point-to-pointcorrelation of the area and magnitude of retinalsensitivity loss at a precise location in the macula.In other words, this technology serves to delineateabsolute and relative scotomas while allowing forelucidation of preferential fixation location andfixation stability.

Traditionally, distance visual acuity has been thegold standard for assessment of macular function inthose with AMD. Although visual acuity is a usefuland easily assessed parameter, it does not provide acomplete description of visual function, and corre-lations with self-reported visual functioning aregenerally poor. Performance in daily activities suchas reading are better correlated to the integrity of thecentral visual field (24). Microperimetry has shownitself to be a useful tool in assessing the functionaldeficits due to AMD beyond that of visual acuity asit generates information on the location and depthof relative and absolute scotomas. For example, thistechnology has revealed changes in retinal sensitivityover drusen (25). Sunness and coworkers have alsodescribed changes in fixation patterns and quantifiedthe area of scotomas in those with geographic atrophy(26,27).

Microperimetry has proved useful in the evalu-ation of neovascular AMD. Absolute scotomas havebeen measured over CNV, subretinal hemorrhage,and chorioretinal scars (28). Fujii and colleaguesevaluated the characteristics of visual loss in subfo-veal CNV and suggest that functional deteriorationmay begin with a mild decrease in retinal sensitivitythat later evolves into fixation instability and eventualabsolute central scotoma with subsequent eccentricfixation (29). This technology has also been used toassess macular function following treatment.Schmidt-Erfurth et al. quantified the size of absoluteand relative scotomas in patients with CNV atbaseline and following treatment with PDT (30).Microperimetry has also been applied to assessfunctional changes with other treatment modalitiesfor CNV including laser photocoagulation, subma-cular surgery, and macular translocation (31–34).

In contrast to the other imaging modalities thatassess structural integrity, microperimetry providesa functional measure of macular function. Much likethe Humphrey visual field is used to assess visual loss

11: QUANTITATIVE RETINAL IMAGING 187

in glaucoma, microperimetry can functionally assessmacular deterioration in AMD. As newer treatmentsemerge for AMD, this modality may supplement theanatomic evaluation of drug efficacy by providing afunctional measure of the area and degree of retinalrecovery or deterioration. Furthermore, microperi-metric analysis to evaluate progression of maculardeterioration may aid in clinical decision making.

CONCLUSION

The value of quantitative retinal imaging willundoubtedly increase as the image processing soft-ware and the quantification methods improve overtime. These advances are already assisting ophthal-mologist through improved education of the patienton the nature and progression of AMD disease withimportant implications for better physician–patientrelationships and compliance with treatment. Itnow seems likely that quantitative retinal imagingwill also provide robust endpoints for assessing theeffectiveness of interventions in clinical trials andepidemiological studies.

SUMMARY POINTS

& Quantitative techniques are useful for objectivelyevaluating the type and number of drusen, retinalpigment epithelium pigmentary changes, andcomponents of CNV lesions.

& OCT based software, such as the OCTOR systemdeveloped at the Doheny Eye Institute, allows fordelineation of retinal layers as well as pathologicentities such as subretinal fluid, area occupied bychoroidal neovascular membranes, and the size ofserous retinal pigment epithelial detachments.

& Quantification of fluorescein angiographic infor-mation is being developed and will provideimportant clinical information regarding the pro-gression of AMD and response to therapy.

& Microperimetry is a useful tool in assessingfunctional deficits due to AMD beyond that ofvisual acuity since it provides information on thelocation and depth of relative and absolutescotomas.

REFERENCES

1. Klein R, Davis MD, Magli YL, et al. The Wisconsin age-related maculopathy grading system. Ophthalmology 1991;98:1128–34.

2. Bird AC, Bressler NM, Bressler SB, et al. An internationalclassification and grading system for age-related maculo-pathy and age-related macular degeneration. SurvOphthalmol 1995; 39:367–74.

3. Bressler NM, Bressler SB, Seddon LM, et al. Drusen charac-teristics in patients with exudative versus non-exudativeage-related macular degeneration. Retina 1988; 8:109–14.

4. Bressler NM, Maguire MG, Bressler SB, et al. Relationshipof drusen and abnormalities of the retinal pigmentepithelium to the prognosis of neovascular macular degen-eration. The Macular Photocoagulation Study Group. ArchOphthalmol 1990; 108:1442–7.

5. Ferris FL, Davis MD, Clemons TE, et al. A simplifiedseverity scale for AMD: AREDS Report No. 18. ArchOphthalmol 2005; 123:1570–4.

6. Smith RT, Nagasaki T, Sparrow JR, et al. A method ofdrusen measurement based on the geometry of fundusreflectance. Biomed Eng Online 2003; 2:10.

7. Peli E, Lahav M. Drusen measurement from fundus photo-graphs using computer image analysis. Ophthalmology1986; 93:1575–80.

8. Morgan WH, Cooper RL, Constable IJ, et al. Automatedextraction and quantification of macular drusen fromfundal photographs. Aust NZ J Ophthalmol 1994; 22:7–12.

9. Kirkpatrick JN, Spencer T, Manivannan A, et al. Quan-titative image analysis of macular drusen from fundusphotographs and scanning laser ophthalmoscope images.Eye 1995; 9:48–55.

10. Shin DS, Javornik NB, Berger JW. Computer-assisted, inter-active fundus image processing for macular drusenquantification. Ophthalmology 1999; 106:1119–25.

11. Sivagnanavel V, Smith RT, Lau GB, et al. An interinstitu-tional comparative study and validation of computer aideddrusen quantification. Br J Ophthalmol 2005; 89:554–7.

12. Smith RT, Chan JK, Nagasaki T, et al. A method of drusenmeasurement based on reconstruction of fundus back-ground reflectance. Br J Ophthalmol 2005; 89:87–91.

13. Huang D, Swanson EA, Lin CP, et al. Optical coherencetomography. Science 1991; 254:1178–81.

14. Puliafito CA, Hee MR, Lin CP, et al. Imaging of maculardiseases with optical coherence tomography. Ophthal-mology 1995; 102:217–29.

15. Hee MR, Puliafito CA, Wong C, et al. Quantitative assess-ment ofmacular edemawith optical coherence tomography.Arch Ophthalmol 1995; 113:1019–29.

16. Hee MR, Puliafito CA, Duker JS, et al. Topography of dia-betic macular edema with optical coherence tomography.Ophthalmology 1998; 105:360–70.

17. Salinas-Alamon A, Garcia-Layana A, Maldonado MJ, et al.Using optical coherence tomography to monitor photo-dynamic therapy in age related macular degeneration.Am J Ophthalmol 2005; 140:23.e1–23.7.

18. Shahidi M, Wang Z, Zelkha R. Quantitative thicknessmeasurement of retinal layers imaged by optical coherencetomography. Am J Ophthalmol 2005; 139:1056–61.

19. Verteporfin Photodynamic Therapy (VIP) Study Group.Verteporfin therapy of subfoveal choroidal neovasculariza-tion in age related macular degeneration: two-year resultsof a randomized clinical trial including lesions with occultwith no classic choroidal neovascularization—Verteporfinin Photodynamic Therapy Report 2. Am J Ophthalmol 2001;131:541–60.

20. Phillips RP, Ross PG, Tyska M, Sharp PF, Forrester JV.Detection and quantification of hyperfluorescent leakageby computer analysis of fundus fluorescein angiograms.Graefes Arch Clin Exp Ophthalmol 1991; 229:329–35.

21. Chakravarthy U, Walsh AC, Muldrew A, Updike PG,Barbour T, Sadda SR. Quantitative fluorescein angiographicanalysis of choroidal neovascular membranes: validationand correlation with visual function. Invest Ophthalmol VisSci 2007; 48:349–54.

188 ESMAILI ET AL.

22. Shah SM, Tatlipinar S, Quinlan E, et al. Dynamic andquantitative analysis of choroidal neovascularization byfluorescein angiography. Invest Ophthalmol Vis Sci 2006;47:5460–8.

23. Rohrschneider K, Springer C, Bultmann S, et al. Micro-perimetry—comparison between the micro perimeter 1 andscanning laser ophthalmoscope—fundus perimetry. AmJ Ophthalmol 2005; 139:125–34.

24. Sunness JS, Applegate CA, Haselwood D, et al. Fixationpatterns and reading rates in eyes with central scotomasfrom advanced atrophic age-related macular degenerationand Stargardt disease. Ophthalmology 1996; 103:1458–66.

25. Takamine Y, Shiraki K, Moriwaki M, et al. Retinal sensi-tivity measurement over drusen using scanning laserophthalmoscope microperimetry. Graefes Arch Clin ExpOphthalmol 1998; 236:285–90.

26. Sunness JS, Bressler NM, Maguire MG. Scanning laserophthalmoscopic analysis of the pattern of visual lossin age-related geographic atrophy of the macula. AmJ Ophthalmol 1995; 119:143–51.

27. Sunness JS, Bressler NM, Tian Y, et al. Measuringgeographic atrophy in advanced age-related maculardegeneration. Invest Ophthalmol Vis Sci 1999; 40:1761–9.

28. Tezel TH, Del Priore LV, Flowers BE, et al. Correlationbetween scanning laser ophthalmoscope microperimetry

and anatomic abnormalities in patients with subfovealneovascularization. Ophthalmology 1996; 103:1829–36.

29. Fujii GY, De Juan E, Jr., Humayun MS, et al. Characteristicsof visual loss by scanning laser ophthalmoscope microperi-metry in eyes with subfoveal choroidal neovascularizationsecondary to age-related macular degeneration. AmJ Ophthalmol 2004; 136:1067–78.

30. Schmidt-Erfurth UM, Elsner H, Terai N, et al. Effects ofverteporfin therapy on central visual field function.Ophthalmology 2004; 111:931–9.

31. Rohrschneider K, Gluck R, Becker M, et al. Scanning laserfundus perimetry before laser photocoagulation of welldefined choroidal neovascularization. Br J Ophthalmol1997; 81:568–73.

32. Fujii GY, de Juan E, Jr., Sunness J, et al. Patient selection formacular translocation surgery using the scanning laserophthalmoscope. Ophthalmology 2002; 109:1737–44.

33. Hudson HL, Frambach DA, Lopez PF. Relation of thefunctional and structural fundus changes after submacularsurgery for neovascular age-related macular degeneration.Br J Ophthalmol 1995; 79:417–23.

34. Loewenstein A, Sunness JS, Bressler NM, et al. Scanninglaser ophthalmoscope fundus perimetry after surgery forchoroidal neovascularization. Am J Ophthalmol 1998;125:657–65.

11: QUANTITATIVE RETINAL IMAGING 189

12

Fundus Autofluorescence in Age-RelatedMacular DegenerationRishi P. Singh and Jeffrey Y. ChungCole Eye Institute, Cleveland Clinic Foundation, Cleveland, Ohio, U.S.A.

Peter K. KaiserDigital Optical Coherence Tomography Reading Center, Cleveland, Ohio, U.S.A.

INTRODUCTION

Autofluorescence (AF) is the intrinsic fluorescenceemitted by a substance after being stimulated byexcitation energy. Ocular structures that autofluoresceinclude the corneal epithelium and endothelium,lens, macular and retinal pigment epithelium (RPE)pigments, optic nerve drusen and RPE deposits inBest’s disease. The AF emitted by macular pigmentsis in the 520 to 800 nm range with peak emission at 590to 630 nm. Clinically, AF of macular pigments can beproduced in vivo by an exciting light source with awavelength between 400 and 590 nm with peakexcitation occurring between 490 and 510 nm (1).This can be achieved with a modified fundus cameraor scanning laser ophthalmoscope (SLO).

The SLO uses blue laser light at 488 nm forillumination and a 500 nm barrier filter to isolate lightfrom other ocular autofluorescent structures (2,3). Theuse of confocal scanning laser ophthalmoscope (cSLO)is considered superior to modified fundus cameraimages because cSLO helps eliminate the competingAF of the lens. As the plane of the cSLO detectionsystem is conjugate to the plane of fundus, competinglight signals from other planes are reduced (4). Inaddition, cSLO images have a higher reliability in com-paring images in one patient from one visit to the nextthan modified fundus camera imaging (5). However, acSLO requires the purchase of a new imaging devicewhereas most offices already have a fundus camera.Spaide has described an easy and inexpensive modifi-cation to preexisting fundus camera by adding 580 nmexcitation and 695 nm barrier filters. With the properbarrier filter signals, wavelengths shorter than 695 nmincluding fluorescence emitted by the lens from 510 to670 nm, would be blocked (4).

The predominant source of AF in the macula islipofuscin, a complex mixture of fluorophores. Whenthe RPE phagocytize photoreceptor outer segments,consisting of retinoids, fatty acids, and proteins,

lipofuscin accumulates as an oxidative byproductwithin the RPE cells (1,6,7). Lipofuscin has an affinityfor acidic organelles and thus accumulates in RPElysosomes and it can account for as much as 20% ofthe free cytoplasmic space of a RPE cell (8,9). A loss ofRPE cells has been shown to be accompanied bysubstantial loss of AF content (10).

The pigment within lipofuscin that causes thisfluorescence was isolated and characterized by Eldredto be A2E, named for its derivation from twomolecules of vitamin A aldehyde and one moleculeof ethanolamine (Fig. 1) (12,13). A2E has been shownto inhibit human RPE cell growth and induce apop-tosis in vitro. It exhibits detergent-like activity,disrupting membrane bound ATPase that maintainlysosomal pH (13–15). In mitochondria, A2E inhibitsoxygen consumption synergistically with light byinhibiting cytochrome c oxidase (16). By mobilizingcytochrome c and apoptosis-inducing factor frommitochondria into the cytoplasm and nucleus,apoptosis is induced in RPE cells (17). A2E has alsobeen shown to confer a dose related sensitivity toblue light damage in RPE cells via oxidativemechanisms (18).

AF is a rapid noncontact, noninvasive way toevaluate RPE function. AF can evaluate the amount oflipofuscin that is accumulated in RPE. By evaluatingfundus autofluorescence images and thus lipofuscinaccumulation, disturbances within the RPE can bereadily detected. In a normal retina, lipofuscin ismost concentrated in the macula with the exceptionof the fovea and decreases towards the periphery (1).The highest lipofuscin AF level in the eye is found 78 to138 away from the fovea, correlating with the areawith the highest distribution of rod photoreceptors(19). Even though high inter-subject variability isseen in the distribution of macular AF, Delori foundsimilarity of AF level within the same retina, as well asthat of the fellow eye (20,21). Lipofuscin fluorescence

levels increases linearly with age. In humans, intra-cellular lipofuscin levels occupy 1% of cell volumeduring the first decade, increasing to 12% to 13% in the50- to 80-year-olds and reaching 19% of the cell volumein the 81- to 90-year-olds (9).

AF AND AMD

Several studies have found that the accumulation oflipofuscin over time may promote the developmentof AMD (Figs. 2 and 3). The age, spatial, and racialdistribution of lipofuscin correlates well with the AFpatterns seen with AMD. Dorey and colleagues foundsignificant correlation between photoreceptor loss and

elevated lipofuscin levels in RPE within donor eyesof Caucasians over 50 years old. They hypothesizedthat lipofuscin accumulation may be indicative ofincreased phagocytic and metabolic stress on theRPE cells leading to photoreceptor death (22,23).Excessive lipofuscin accumulation may precede thedevelopment of GA and the enlargement of preex-isting GA (24). Thus, AF imaging may be useful inevaluating the risk of AMD progression by mappingretinal AF and lipofuscin accumulation over extendedperiods. Previous work has shown a significant corre-lation in the amount of large, foveal, soft drusen andpatterns of increased AF (5). Spaide reported greaterlevels of AF in fellow eyes of patients with exudative

1'

2'3'

4'

7'8'

9'10'

11'12'

14'

15'

13 11 9 7 1

2

34

8101214N

CH2CH2OH CH2CH2OH

A2E

1'

2'3'

4'

7'8'

9'10'

11'12'

14'

15' N

14

12

13

11

10

8

9

7

4

32

1

iso-A2E

iso-A2EA2E

Figure 1 Structure of A2E and isomer iso-A2E from human retinal pigment epithelium. Source: From Ref. 11.

192 SINGH ET AL.

AMD than in patients without a history of AMD (4).Delori and colleagues identified that RPE overlyingdrusen have a central area of decreased AF withsurrounding ring of increased AF, suggestingdamage to RPE health (7).

An International Fundus AutofluorescenceClassification Group (IFAG) was organized by theFundus autofluorescence in Age-Related MacularDegeneration (FAM)group to establish an internationalclassification system to describe the abnormal fundus

Figure 2 Lipofuscin accumulation in a 68-year-old patient over three years. Source: Photos courtesy ofL. Yanuzzi, R. Spaide, and P. Bhatnagar.

Abnormal AF GA formation

Figure 3 Progression to geographic atrophy over three years with abnormal autofluorescence. Abbreviations:

AF, autofluorescence; GA, geographic atrophy. Source: Photos courtesy of L. Yanuzzi, R. Spaide, andP. Bhatnagar.

12: FUNDUS AUTOFLUORESCENCE IN AGE-RELATED MACULAR DEGENERATION 193

AF patterns seen in AMD (25). The multinationalgroup of clinicians established eight distinct patternsof AF in AMD: normal AF, minimal change patternAF, focal increased pattern, patchy pattern AF, linearpattern AF, lacelike pattern AF, reticular pattern AF,and speckled pattern AF. Standardized photos of theseAF patterns were established (Fig. 4).

1. Normal pattern. The normal AF pattern is charac-terized by a homogeneous background AF with agradual decrease from the innermacula toward the

fovea due to the masking effect of yellow lutealmacular pigment.

2. Minimal change pattern. Theminimal changepatternis characterized by very small irregular increases ordecreases of background AF without an obvioustopographic pattern.

3. Focal increased pattern. Focal increased AF isdescribed as the presence of at least one spot (lessthan 200 mm diameter) of markedly increased AFbrighter than the surrounding fluorescence. Theborders are well defined and some areas of focal

(B)

(A)

Figure 4 (Continued ) Patterns of fundus autofluorescence (AF) as established by IFAG. (A–H) (A) Normal pattern;(B) minimal change pattern—very limited irregular increases or decreases of AF due to multiple small hard drusen.

194 SINGH ET AL.

increased AF may be surrounded by a darker-appearing halo. Visible alterations (focal hyper-pigmentation or drusen) seen on color fundusphotos may or may not correspond to areas of AF.

4. Patchy pattern. Patchy AF is defined as at least onelarger area (greater than 200 mm diameter) of mark-edly increasedAFwhere the borders of the areas aretypically lesswell-defined than thepreviouspattern.There is a gradual increase in AF from the back-

ground to the patchy area. This pattern may alsocorrespond to large drusen, soft drusen and areasof hyperpigmentation seen on color photographs.

5. Linear pattern. The linear pattern describes thepresence of at least one linear area of markedlyincreased AF with well-demarcated borders andno gradual decrease in AF. These AF areas usuallycorrespond to hyperpigmented lines on the colorfundus photograph.

(C)

(D)

Figure 4 (Continued ) (C) focal increased pattern—several well-defined spots with markedly increased AF;

(D) patchy pattern—multiple large areas of increased AF corresponding to multiple large soft drusen and/orhyperpigmentation in the fundus photograph.


6. Lacelike pattern. The lacelike pattern typically exhi-bits numerous branching linear structures ofincreased AF that form a lace pattern. The bordersare poorly defined and a decline in AF is observedfrom the center of the AF areas to the surroundingareas. These lacelike areas can correspond tohyperpigmentation on the color image, but it maycorrespond to normal fundus areas as well.

7. Reticular pattern. The reticular pattern is exemplifiedby the presence of multiple small areas (less than

200 mm diameter) of decreased AF with poorlydefined borders. Fundoscopically, there are usuallyvisible small soft drusen, hard drusen, or areas withpigmentary changes overlying these areas, but thefundus can be normal as well.

8. Speckled pattern. The speckled AF pattern has thesimultaneous presence of a variety of AF abnorm-alities that extend beyond the macular area. Therecan be multiple, small areas of irregularly increasedand decreased AF which appear punctuate or

(F)

(E)

Figure 4 (Continued) (E) linear pattern—at least one linear area with marked increased AF;

(F) lacelike pattern—multiple branching linear structures of increased AF.

196 SINGH ET AL.

resemble linear structures. Color fundus photo-graphs may include corresponding hyper- andhypopigmentation and multiple subconfluent andconfluent drusen.

Of 149 eyes studied within the IFAG group’sinitial studyof 107patients (44male, 63 femalepatients)with unilateral or bilateral GA, the diffuse patternwas the most common (57%), followed by the banded,focal and normal patterns at about 12% per group.The other patterns were less commonly seen (25).

AF with DrusenThere have been numerous reports examining AF indrusen associated with dry AMD. Delori and associ-ates identified a specific pattern of AF spatiallyassociated with hard and soft drusen rangingbetween 60 and 175 mm in size. The pattern is charac-terized by a central area of decreased AF surrounded,in most cases, by an annulus of increased AF aroundthe drusen (7). It was hypothesized by Delori that thisAF pattern is due to RPE impairment with secondaryaccumulation of lipofuscin around drusen, with RPE

(G)

(H)

Figure 4 (Continued ) (G) reticular pattern—multiple small areas of decreased AF with bright lines in between; (H)

speckled pattern—presence of a variety of AF abnormalities, which extend beyond the macular area to the posterior pole.


atrophy overlying drusen. Spade interprets theappearance of this pattern as secondary to thinnerRPE cells on top of the drusen, and thicker RPE cellsaround the base (26). These areas of increased AFaround drusen showed normal or near-normalphotopic sensitivity, but moderately reduced scotopicsensitivity (27). Soft drusen larger than 175 mm andconfluent soft drusen show either a heterogenousdistribution of AF or multifocal areas of decreasedAF (7). But, both normal, hyper- and hypofluorescenceof the RPE cells overlying soft drusen has beenreported with no proven biochemical explanationand thus future larger studies are needed to refinethis classification (3,5,28,29).

AF in Geographic AtrophyWith loss of RPE containing lipofuscin, areas of GAappear dark under AF. Holz and colleagues found 83%of the geographic atrophic from AMD has increasedAF pattern at the border (30). This ring of elevatedAF from lipofuscin bordering the GA supports theconcept that excessive lipofuscin may be associatedwith RPE damage (21). In fact, the increased AF in thejunctional zone around GA is thought to be charac-teristic for AMD since only 9% of geography atrophyfrom other causes exhibit similar findings (3,30). Whentested with fundus perimetry, a significant degree ofretinal sensitivity loss is found in the junctional areabetween the inner dark zone and ring of increased AF(31). Photopic and scotopic fine matrix mapping ofthese areas has shown a scotopic sensitivity lossdemonstrating a preferential loss of rods (27). Thiscorrelation of AF abnormality to a loss of function mayfurther suggest a relationship between GA and itsincreased AF border. Holz found that 90% of theAMD patients with bilateral GA exhibited the sameAF pattern in both eyes (30).

Longitudinal studies have demonstrated that AFis useful for the precise mapping andmeasuring of GAareas (Fig. 5) (24). Moreover, progression of increasedAF from GA has been described in several studies(3,30). It has been noted that the rate of GA spreadaccelerates with expansion of GA area, then levels offat five disc areas (32,33). However, the natural pro-gression of GA is poorly understood. Manualmeasurement of GA is time consuming and resultsin significant inter-observer variability (31). Previousstudies have found that automated quantification anddelineation of AF images is superior to fundus pho-tography or fluorescein angiography in the delineationof GA (31). When combined with cSLO, the measure-ment of GA area improves significantly (34). Thequantification of these lesions adds to the under-standing of the natural history of GA formation andallows for the monitoring of future therapeutics toslowdown its progression.

In a three year prospective study of threepatients, Holz and colleagues found the developmentand enlargement of GA within areas of increased AF(24). In another study by Holz, eyes with only diffusepatterns of AF were examined and found to havehigher levels of AF on the transitional area betweenGA and healthy RPE. The study determined that thegreater the area of increased AF adjacent to the GA, thefaster the expansion of GA. They concluded a positivecorrelation between increased AF and GA expansionin this population (35).

With evidence linking lipofuscin to cell deathin a dose-dependent manner, it is plausible topropose the zone of increased AF around GA maybe the advancing border of GA. As GA advances,RPE cells absorb lipofuscin materials from adjacentdead RPE cells, thus increasing its susceptibility fordestruction. This effect spread like dominos, andaccelerates as the area of GA containing lipofuscingrows exponentially. Expansion of GA slows onlywhen the spread reaches less vulnerable and heal-thier RPE cells.

However, in a retrospective study of AF photo-graphs with differing baseline patterns of AF, Hwangand colleagues found that only 34% to 50% of the newareas of GA fell into an area of increased AF. Thepositive predictive value for increased AF to form newGA was no better than chance (36). Given these twoconflicting reports, it difficult to form a consensus onwhether increased AF at the border of GA trulypredicts future progression of GA to this area.

AF in Choroidal NeovascularizationWhile the majority of studies with AF in AMD haveconcentrated on the dry form, there has been limitedwork on determining a correlation between AF andchoroidal neovascularization (CNV). Eyes with earlyCNV lesions show various patterns of increased AF(Fig. 6). Einbock and colleagues studied eyes withexudative changes and found that 17% of eyes exhib-ited “patchy” AF pattern, as well as some with “focalincreased” plaque and “reticular” patterns. Other lessintense patterns were not associated with progressionto late AMD during 18-month follow-up (37).Dandekar and colleagues studied AF in 65 consecutiveeyes with CNV secondary to AMD (38). Patients werestratified by age of lesion. Eyes with recent onsetlesions (one to six months) showed no AF abnormal-ities indicating relatively healthy RPE initially. OlderCNV lesions (greater than six months) exhibiteddecreased AF levels indicating RPE damage andphotoreceptor loss. Similar to pattern seen in GA,increased AF in the junctional zone is also seen insome eyes with CNV. Eyes with better visual acuity arethose with intact AF in early onset lesions and others

198 SINGH ET AL.

with intact AF in the foveal area. This study high-lighted the possible use of AF to determine visualprognosis for AMD lesions.

In an observational case series examiningpigment epithelial detachments (PEDs) associatedwith AMD, increased AF was seen in all serousPEDs regardless of whether there was an underlyingCNV in the area of detachment (Fig. 7). The authorsconcluded that the increased AF seen with serousPEDs may be due to AF of sub-retinal pigmentepithelial fluid. In the case of a drusenoid PED, AFlevels were dependent on pigment clumping withincreased pigment correlating with lower AF levels.Larger numbers of patients are needed to verify thismorphological features (39).

A few studies have attempted to classify CNVlesion type based on the AF pattern seen.

In a study examining AF of 68 eyes undergoingphotodynamic therapy (PDT) treatment, Frammefound 79% of the untreated classic lesions wereassociated with decreased AF and a junctional zoneof increased or normal AF. In untreated occultmembranes, a normal or mottled AF pattern withfoci of hyper- and hypofluorescence were seen. AfterPDT treatment, 90% of the classic CNV lesions showeddecreased AF signal. There appeared to be no AFchange in occult parts of CNV lesions after PDT (40).These baseline AF patterns were also described byMcBain and collegues in patients with exudativeAMD. Low AF signal at the site of classic CNV wasdetected in 90% of exudative AMD lesions. Whilemultiple foci of low AF was seen in half of occultCNV lesions, focally increased AF was rarely seenwith CNV lesions (41).

Figure 5 Precise mapping of geographic atrophy over time with autofluoresence. Source: Photos courtesyof L. Yanuzzi, R. Spaide, and P. Bhatnagar.


Figure 6 Fundus autofluorescence corresponding to subretinal fluid with choroidal neovascularization.

Abbreviation: AF, autoflourescence. Source: Photos courtesy of L. Yanuzzi, R. Spaide, and P. Bhatnagar.

Red Free FA ICG Autoflurescence

Figure 7 Serous pigment epithelial detachment imaged with various modalities. Abbreviations: FA, fluorescein

angiography; ICG, indocyanine green. Source: Photos courtesy of L. Yanuzzi, R. Spaide, and P. Bhatnagar.

200 SINGH ET AL.

AF after Laser TreatmentAF has shown to be useful tools in localizing laserlesions after treatment. In a pilot study, Frammeattempted Nd:YAG RPE selective laser treatment fordiabetic maculopathy and central serous chorioretino-pathy. Changes to AF levels around laser lesions10 minutes after treatment were seen in 22 out of 26patients. These laser lesions later exhibit a lower AFlevel in comparison to surrounding tissues at three tosix months (5,42). In a study of subthreshold infrareddiode laser for the reduction of drusen in dry AMD,visualization of laser lesions using AFwas shown to bemore sensitive than fluorescein angiogram in 75% ofthe eyes immediately post laser treatment, and 55% ofthe eyes three months after treatment (43).

LIMITATIONS

There are several limitations in the use of AF in theevaluation of patients for AMD. AF detection islimited in patients with significant media opacitiessuch as cataract and vitreous hemorrhage. Further-more, the comparative quantification of AF imagescannot occur amongst patients. Rather, only imagesfrom the same patient can be compared to determinechanges in intensities of AF seen over time. Thus, theuse of AF is still in its infancy and further studies needto be performed to evaluate its role in the diagnosisand management of AMD.

Reports on the use of AF in the diagnosis andmanagement of AMD are preliminary at best andsometimes conflicting. There remains a need forfurther data to clarify the relationship of AF patternsin formation and expansion of GA. More informationis needed to substantiate the relationship between AFpatterns and risk of CNV. There is a lack of studiesidentifying AF criteria useful in the classification ofCNV subtype. The current literature thus far consistsmostly of non-randomized small case series limitingits applicability to the general population. To addressthese limitations, the FAM study group, a multi-centerstudy is currently under way, with some resultsalready in press. The goal of the group is to investigatethe correlation between fundus AF and the naturalhistory of AMD. The group also intends to identifyhigh-risk AF characteristics that can predict patientswho will progress to late AMD (37).

SUMMARY POINTS

& The predominant source of AF in the macula islipofuscin, a complex mixture of fluorophores.

& The pigment within lipofuscin that causes thisfluorescence is A2E (named for its derivationfrom two molecules of vitamin A aldehyde andone molecule of ethanolamine).

& Fundus autofluorescence is a useful modality toimage lipofuscin in RPE cells and is a unique wayto assess RPE function in AMD.

& In GA, automated imaging analysis by AF has beenshown superior to fundus photo or FA in assessingthe extent of the atrophy. Increased AF, especiallyat the edge of GA area, may predict GA formationand expansion.

& AF has been helpful in assessing RPE health inexudative AMD, and can consistently visualizeserous PED.

& AF can also be helpful in the early localizationof previous retinal laser treatments where itmay be up to three times more sensitive thanfluorescein angiogram.

& AF is shown to be helpful in indicating graftvisualization in RPE-choroidal grafts for AMDpatients.

& Larger randomized controlled studies using AF areneeded to further assess its potential in the detec-tion and management of AMD.

REFERENCES

1. Delori FC, Dorey CK, Staurenghi G, Arend O, Goger DG,Weiter JJ. In vivo fluorescence of the ocular fundus exhibitsretinal pigment epithelium lipofuscin characteristics. InvestOphthalmol Vis Sci 1995; 36:718–29.

2. von Ruckmann A, Fitzke FW, Bird AC. Distribution offundus autofluorescence with a scanning laser ophthal-moscope. Br J Ophthalmol 1995; 79:407–12.

3. von Ruckmann A, Fitzke FW, Bird AC. Fundus autofluor-escence in age-related macular disease imaged with a laserscanning ophthalmoscope. Invest Ophthalmol Vis Sci 1997;38:478–86.

4. Spaide RF. Fundus autofluorescence and age-relatedmacular degeneration. Ophthalmology 2003; 110:392–9.

5. Lois N, Owens SL, Coco R, Hopkins J, Fitzke FW, Bird AC.Fundus autofluorescence in patients with age-relatedmacular degeneration and high risk of visual loss. AmJ Ophthalmol 2002; 133:341–9.

6. Gaillard ER, Atherton SJ, Eldred G, Dillon J. Photophysicalstudies on human retinal lipofuscin. Photochem Photobiol1995; 61:448–53.

7. Delori FC, Fleckner MR, Goger DG, Weiter JJ, Dorey CK.Autofluorescence distribution associated with drusen inage-related macular degeneration. Invest Ophthalmol VisSci 2000; 41:496–504.

8. Kennedy CJ, Rakoczy PE, Constable IJ. Lipofuscin of theretinal pigment epithelium: a review. 1995; 9(Pt 6): 763–71.

9. Feeney-Burns L, Hilderbrand ES, Eldridge S. Aging humanRPE: morphometric analysis of macular, equatorial, andperipheral cells. Invest Ophthalmol Vis Sci 1984; 25:195–200.

10. Katz ML, Eldred GE. Retinal light damage reduces auto-fluorescent pigment deposition in the retinal pigmentepithelium. Invest Ophthalmol Vis Sci 1989; 30:37–43.

11. Parish CA, Hashimoto M, Nakanishi K, Dillon J, Sparrow J.Isolation and one-step preparation of A2E and iso-A2E,fluorophores from human retinal pigment epithelium. ProcNatl Acad Sci USA 1998; 95:14609–13.


12. Sakai N, Decatur J, Nakanishi K, Eldred GE. Ocular agepigment “A2E”: an unprecedented pyridinium bisretinoid.J Am Chem Soc 1996; 118:1559–60.

13. Eldred GE. Age pigment structure. Nature 1993; 364:396.14. Mellman I, Fuchs R, Helenius A. Acidification of the

endocytic and exocytic pathways. Annu Rev Biochem1986; 55:663–700.

15. Sparrow JR, Parish CA, Hashimoto M, Nakanishi K. A2E, alipofuscin fluorophore, in human retinal pigmented epi-thelial cells in culture. Invest Ophthalmol Vis Sci 1999;40:2988–95.

16. Shaban H, Gazzotti P, Richter C. Cytochrome c oxidaseinhibition by N-retinyl-N-retinylidene ethanolamine, acompound suspected to cause age-related macula degener-ation. Arch Biochem Biophys 2001; 394:111–6.

17. Suter M, Reme C, Grimm C, et al. Age-related maculardegeneration. The lipofusion component N-retinyl-N-retinylidene ethanolamine detaches proapoptotic proteinsfrom mitochondria and induces apoptosis in mammalianretinal pigment epithelial cells. J Biol Chem 2000;275:39625–30.

18. Nilsson SE, Sundelin SP, Wihlmark U, Brunk UT. Aging ofcultured retinal pigment epithelial cells: oxidativereactions, lipofuscin formation and blue light damage.Doct Ophthalmol 2003; 106:13–6.

19. Delori FC, Goger DG, Dorey CK. Age-related accumulationand spatial distribution of lipofuscin in RPE of normalsubjects. Invest Ophthalmol Vis Sci 2001; 42:1855–66.

20. Bellmann C, Jorzik J, Spital G, Unnebrink K, Pauleikhoff D,Holz FG. Symmetry of bilateral lesions in geographicatrophy in patients with age-related macular degeneration.Arch Ophthalmol 2002; 120:579–84.

21. Robson AG, Moreland JD, Pauleikhoff D, et al. Macularpigment density and distribution: comparison of fundusautofluorescence with minimum motion photometry.Vision Res 2003; 43:1765–75.

22. Dorey CK, Wu G, Ebenstein D, Garsd A, Weiter JJ. Cell lossin the aging retina relationship to lipofuscin accumulationand macular degeneration. Invest Ophthalmol Vis Sci 1989;30:1691–9.

23. Dorey K, Staurenghi G, Delori FC. Lipofuscin in age andARMD eyes. In: Hollyfield JG, ed. Retinal Degeneration.New York: Plenum Pub Corp., 1993:3–14.

24. Holz FG, Bellman C, Staudt S, Schutt F, Volcker HE. Fundusautofluorescence and development of geographic atrophyin age-related macular degeneration. Invest OphthalmolVis Sci 2001; 42:1051–6.

25. Bindewald A, Schmitz-Valckenberg S, Jorzik JJ, et al. Classi-fication of abnormal fundus autofluorescence patterns inthe junctional zone of geographic atrophy in patients withage related macular degeneration. Br J Ophthalmol 2005;89:874–8.

26. Spaide RF. Macular autofluorescence. AAO Retina Subspe-cialty Day 2005; 1:192–7.

27. Scholl HP, Bellmann C, Dandekar SS, Bird AC, Fitzke FW.Photopic and scotopic fine matrix mapping of retinal areasof increased fundus autofluorescence in patients with age-related maculopathy. Invest Ophthalmol Vis Sci 2004;45:574–83.

28. Solbach U, Keilhauer C, Knabben H, Wolf S. Imaging ofretinal autofluorescence in patients with age-relatedmacular degeneration. Retina 1997; 17:385–9.

29. Sunness JS, Ziegler MD, Applegate CA. Issues in quanti-fying atrophic macular disease using retinalautofluorescence. Retina 2006; 26:666–72.

30. Holz FG, Bellmann C, Margaritidis M, Schutt F, Otto TP,Volcker HE. Patterns of increased in vivo fundus autofluor-escence in the junctional zone of geographic atrophy of theretinal pigment epithelium associated with age-relatedmacular degeneration. Graefes Arch Clin Exp Ophthalmol1999; 237:145–52.

31. Schmitz-Valckenberg S, Jorzik J, Unnebrink K, Holz FG,FAM Study Group. Analysis of digital scanning laserophthalmoscopy fundus autofluorescence images ofgeographic atrophy in advanced age-related maculardegene 132#ration. Graefes Arch Clin Exp Ophthalmol2002; 240:73–8.

32. Schatz H, McDonald HR. Atrophic macular degenerationrate of spread of geographic atrophy and visual loss.Ophthalmology 1989; 96:1541–51.

33. Sarks SH. Drusen patterns predisposing to geographic atro-phy of the retinal pigment epithelium. Aust J Ophthalmol1982; 10:91–7.

34. Deckert A, Schmitz-Valckenberg S, Jorzik J, Bindewald A,Holz FG, Mansmann U. Automated analysis of digitalfundus autofluorescence images of geographic atrophy inadvanced age-related macular degeneration using confocalscanning laser ophthalmoscopy (cSLO). BMC Ophthalmol2005; 5:8.

35. Schmitz-Valckenberg S, Bindewald-Wittich A, Dolar-Szczasny J, et al. Correlation between the area of increasedautofluorescence surrounding geographic atrophy anddisease progression in patients with AMD. InvestOphthalmol Vis Sci 2006; 47:2648–54.

36. Hwang JC, Chan JW, Chang S, Smith RT. Predictive value offundus autofluorescence for development of geographicatrophy in age-related macular degeneration. InvestOphthalmol Vis Sci 2006; 47:2655–61.

37. EinbockW,Moessner A, Schnurrbusch UE, Holz FG,Wolf S,FAM Study Group. Changes in fundus autofluorescence inpatientswith age-relatedmaculopathy. Correlation to visualfunction: a prospective study. Graefe’s Arch Clin ExpOphthalmol 2005; 243:300–5.

38. Dandekar SS, Jenkins SA, Peto T, et al. Autofluorescenceimaging of choroidal neovascularization due to age-relatedmacular degeneration. Arch Ophthalmol 2005; 123:1507–13.

39. Karadimas P, Bouzas EA. Fundus autofluorescence imagingin serous and drusenoid pigment epithelial detachmentsassociated with age-related macular degeneration. AmJ Ophthalmol 2005; 140:1163–5.

40. Framme C, Bunse A, Sofroni R, et al. Fundus autofluores-cence before and after photodynamic therapy for choroidalneovascularization secondary to age-related macular dege-neration. Ophthalmic Surg Lasers Imaging 2006; 37:406–14.

41. McBain VA, Townend J, Lois N. Fundus autofluorescencein exudative age-related macular degeneration. Br JOphthalmol 2006; 91(4):491–6.

42. Framme C, Brinkmann R, Birngruber R, Roider J. Auto-fluorescence imaging after selective RPE laser treatment inmacular diseases and clinical outcome: a pilot study. BrJ Ophthalmol 2002; 86:1099–106.

43. Bessho K, Rodanant N, Bartsch DU, Cheng L, Koh HJ,Freeman WR. Effect of subthreshold infrared laser treat-ment for drusen regression on macular autofluorescence inpatients with age-related macular degeneration. Retina2005; 25:981–8.

202 SINGH ET AL.

Part IV: Medical Treatment for Age-RelatedMacular Degeneration

13

Laser Photocoagulation for ChoroidalNeovascularizationCatherine Cukras and Stuart L. FineDepartment of Ophthalmology, Scheie Eye Institute, University of Pennsylvania, Philadelphia,

Pennsylvania, U.S.A.

INTRODUCTION

Until the initial Macular Photocoagulation Study (MPS)outcome data were published in June 1982, there wereno reported treatments of proven benefit for patientswith choroidal neovascularization (CNV) secondary toage-related macular degeneration (AMD). The MPStrials conducted from 1979 to 1994 showed that laserphotocoagulation was preferable to observation forseveral categories of well-defined CNV based on thefluorescein angiographic location of the CNV withrespect to the geometric center of the fovea, i.e., extra-foveal, juxtafoveal, and subfoveal (1–5). The MPSpublications also described the factors which limitedthe utility of laser photocoagulation treatment.

1. Only a small proportion of symptomatic AMD eyesmet MPS eligibility criteria as being appropriate forlaser treatment (6,7);

2. There was a high rate of persistent and recurrentleakage even after initially successful closure of theCNV (8,9);

3. Laser photocoagulation caused immediate andpermanent damage to the retina in the area treatedand this damage typically resulted in an immediatedecrease in visual acuity (VA) (2);

4. Treated as well as untreated eyes continued to losecentral vision over time, despite initial closure of theCNV in laser-treated eyes.

In addition, because laser photocoagulation is afocal treatment, there is no expected beneficial effectbeyond the area of laser application. Specifically, laserphotocoagulation does not inhibit the development ofnew areas of CNV. With the advent of safe and effectiveantiangiogenic therapies which treat not only theexisting neovascularization but also reduce the risk ofdeveloping CNV, it would appear that laser photocoa-gulation will have an extremely limited role in the

management of CNV secondary to AMD. Thus, thefollowing narrative is presented primarily for anhistorical perspective on how treatment for CNVsecondary to AMD developed over the last quartercentury.

EPIDEMIOLOGY AND NATURAL HISTORY

AMD is a leading cause of severe and irreversiblecentral vision loss in the developed world amongpeople over the age of 55 (10–13). Up to 90% of thesevere vision loss in AMD is caused by CNV (14–16).

Before the MPS was initiated in 1979, there wereseveral natural history studies which documented theunfavorable visual prognosis of eyes with untreatedCNV secondary to both AMD and ocular histoplas-mosis (17,18). These natural history data weresubstantiated by the visual outcomes of untreatedeyes among participants in the MPS. In the MPS trialof juxtafoveal CNV, 65% of untreated eyes lost six ormore lines of acuity after five years follow up, and 93%progressed from juxtafoveal to subfoveal CNV (4,19).

The initial component of theMPS evaluated argonlaser photocoagulation in patients with extrafovealCNV secondary to AMD. At the time, this trial wasknown as the Senile Macular Degeneration Study(SMDS). Eyes with extrafoveal CNV were assignedrandomly to immediate argon laser treatment or toobservation. By 18 months after enrollment, 60% ofuntreated eyes had lost six or more lines of VA. Byone year after enrollment, fluorescein angiographyshowed that 73% of untreated eyes had progressedfrom extrafoveal to subfoveal CNV (20).

In 1985, Guyer et al. reported that among 92 AMDpatients with subfoveal neovascular lesions, 64% lostsix or more lines of vision within two years (18). In theMPS trial of subfoveal lesions, 30% of untreated eyeslost six or more lines of VA at 12 months follow up, and39% lost six or more lines of vision by two years (2).

Table1

SummaryoftheMajorResultsoftheMacularPhotocoagulationStudy

MPSstudylesiontype

CNVdescription

Location

Size

Age

VA

Exclusion

Extrafoveal(5)

Angiographicevidenceof

leakingCNVwith“well-

demarcatedborders”

200–2500mmfromcenterof

FAZ

R50yr

R20/100

VA!20/400,prior

laser,otherocular

disease,systemic

steroids

Juxtafoveal—AMDS-K

(4)

Angiographicevidenceof

leakingCNVwith“well-

demarcatedborders”

1–199mmfromcenterof

FAZorO200mmfrom

FAZifadjacentbloodor

pigmentextendedto

within200mm

R50yr

R20/100

VA!20/400,prior

laser,otherocular

disease

NewsubfovealCNV(1,2)

FAwithin96hrof

randomization;leaking

CNVwith“well-

demarcatedborders”;

mostoflesioneither

classicoroccult

NewvesselsunderFAZ

center

!3.5MPSstandarddisc

area(1MPSstandard

areaZ1.77mm2);some

areawithin2discdiameters

ofretinamustbeabletobe

leftuntreated

R50yr

20/40–20/320

inclusive

Priorlaser,other

oculardisease,

systemicsteroids

RecurrentsubfovealCNV(1,3)

FAwithin96hrof

randomization;leaking

CNVwith“well-demarcated

borders”;contiguoustothe

scarfromearliertreatment

NewvesselsunderFAZ

centerorCNVwithin

150mmofFAZscarunder

FAZcenter

Areaoftreatmentplusscar

%6MPSdiscareas

(10.6mm2)andsome

portionofretinawithin

1-discdiameter(1.5mm)of

FAZmustremainuntreated

Previoustreatment

directlytothecenter

oftheFAZ,other

oculardisease,

systemicsteroids

Abbreviations:AMDS-K,age-relatedmaculardegenerationstudy-kryptonlaser;CNV,choroidalneovascularization;FA,fluoresceinangiography;FAZ,fovealavascularzone;MPS,MacularPhotocoagulationStudy;

VA,visualacuity.

Source:FromRef.26.Copyright2007fromThermallasertreatmentinAMD:therapeuticandprophylactic.InternationalOphthalmologyClinics.ReproducedbypermissionofLippincottWilliamsandWilkins.

204 CUKRAS AND FINE

The MPS TrialsThe MPS documented that the visual outcome of lasertreatment for eyes with extrafoveal CNV was betterthan the natural history (21–23). In fact, recruitment intothe argon laser trial of extrafoveal CNV (SMDS) washalted early because 18 months after enrollment, only25% of laser treated eyes compared to 60% of observedeyes had lost six or more lines of VA (21). Althoughlaser treatment did not reverse or stop progression ofvision loss, laser treated eyes continued to have bettervision than untreated eyes even after five years offollow up (5). Trials of similar design conducted atMoorfields Eye Hospital in London, England and byCoscas and Soubrane in Creteil, France also demon-strated a benefit of laser treatment versus observationin AMD patients with selected CNV lesions (24,25).Several MPS trials reported that the difference in visionloss between laser-treated and untreated eyes wasmaintained over a four to five years course of followup. The patient eligibility criteria defining the studypopulation as well as the results from the key trials aresummarized in Tables 1 and 2 (1–5,27).

Decreased Vision after Laser TreatmentThe studies which showed a benefit of laser treatmentcompared toobservation foreyeswith studyeligibleCNVlesions also documented that laser treatment did notprevent the progressive vision loss associated with CNV.Significant vision loss occurred over time in most treatedeyes. Followup also showed that persistent and recurrentCNV were responsible for the progressive loss of vision.For example, 24 months after laser treatment of extra-foveal CNV lesions, 52% of eyes showed evidence ofrecurrence (28). Even for subfoveal lesions, after threeyears of follow up, nearly half the treated eyes hadpersistent or recurrent CNV (9). One MPS trial reportedthat eyes with recurrent CNV had less vision loss withlaser treatment than with observation (Table 2) (3).

It must be noted that in eyes with subfoveallesions and relatively good VA, there is greater loss of

vision in laser-treated versus untreated eyes within thefirst three months after laser treatment (3). This obser-vation documents the immediate harmful effects oflaser treatment to the fovea. However, when patientswith subfoveal CNV were followed for longer periods,it became evident that laser treated eyes had less visionloss than observed eyes, indicating some long-termbenefit of laser treatment even when applied to subfo-veal CNV (3). This benefit was maintained over thethree years course of follow up.

As indicated in the opening paragraphs, thisreview and the accompanying tables are provided forhistorical perspective. At present, anti vascular endo-thelial growth factor (AntiVEGF) therapy appears to bethe preferredmanagement strategy for all forms ofCNVsecondary to AMD irrespective of the geographiclocation of the CNV with respect to the foveal center.The reasons are listed below (20,29–33).

1. AntiVEGF therapy is not associatedwith immediatelossofvisiondue todestructionofvisual elements inthe retina.

2. AntiVEGF therapy is more effective than laserphotocoagulation or photodynamic therapy.

3. AntiVEGF therapy discourages the formation ofnew vessels as well as treating the new vessels.

SUMMARY POINTS

& TheMPS trialswere conducted from1979 to1994andshowed that laser photocoagulation was preferableto observation for several categories of well-definedCNV based on the fluorescein angiographic locationof theCNVwith respect to thegeometric centerof thefovea, i.e., extrafoveal, juxtafoveal, and subfoveal.

& The MPS studies also documented that laser treat-ment did not prevent the progressive vision lossassociated with CNV.

& Significant vision loss occurred over time in mosttreated eyes.

Table 2 Percentage Progressing to Severe Vision Loss Defined as Loss of More than Six Lines of Visual Acuity

One year Two yearsThree years for all (exceptfour years “subfoveal new”) Five years

MPS AMD study Treated (%) Control (%) Treated (%) Control (%) Treated (%) Control (%) Treated (%) Control (%)

Extrafoveal

CNV (5)

24 41 33# 51# 45 63 46 64

Juxtafoveal

CNV (4)

31 45 45 54 51 61 55 65

Subfoveal CNV

(new) (1)

24 (20)C 30 (11)C 23 39 23 45

Subfoveal CNV

(recurrent) (1)

11 29 9 28 17 39

C, 3 months; #, 18 months.Abbreviations: AMD, age-related macular degeneration; CNV, choroidal neovascularization; MPS, Macular Photocoagulation Study.Source: From Ref. 27. Copyright 2002 from Age-Related Macular Degeneration by J Lim editor. Reproduced by permission of Routledge/Taylor & FrancisGroup, LLC.

13: LASER PHOTOCOAGULATION FOR CHOROIDAL NEOVASCULARIZATION 205

REFERENCES

1. Macular Photocoagulation Study Group. Laser photocoagu-lation of subfoveal neovascular lesions of age-relatedmacular degeneration. Updated findings from two clinicaltrials. Arch Ophthalmol 1993; 111(9):1200–9.

2. Macular Photocoagulation Study Group. Laser photocoagu-lation of subfoveal neovascular lesions in age-relatedmacular degeneration. Results of a randomized clinicaltrial. Arch Ophthalmol 1991; 109(9):1220–31.

3. Macular Photocoagulation Study Group. Laser photocoagu-lation of subfoveal recurrent neovascular lesions in age-related macular degeneration. Results of a randomizedclinical trial. Arch Ophthalmol 1991; 109(9):1232–41.

4. Macular Photocoagulation Study Group. Laser photo-coagulation for juxtafoveal choroidal neovascularization.Five-year results from randomized clinical trials. ArchOphthalmol 1994; 112(4):500–9.

5. Macular Photocoagulation Study Group. Argon laser photo-coagulation for neovascular maculopathy. Five-year resultsfrom randomized clinical trials. Arch Ophthalmol 1991;109(8):1109–14.

6. Ciulla TA, Danis RP, Harris A. Age-related macular degener-ation: a review of experimental treatments. Surv Ophthalmol1998; 43(2):134–46.

7. Freund KB, Yannuzzi LA, Sorenson JA. Age-related maculardegeneration and choroidal neovascularization. AmJ Ophthalmol 1993; 115(6):786–91.

8. Macular Photocoagulation Study Group. Persistent andrecurrent neovascularization after krypton laser photocoa-gulation for neovascular lesions of age-related maculardegeneration. Arch Ophthalmol 1990; 108(6):825–31.

9. Macular Photocoagulation Study Group. Persistent andrecurrent neovascularization after laser photocoagula-tion for subfoveal choroidal neovascularization of age-related macular degeneration. Arch Ophthalmol 1994;112(4): 489–99.

10. Fine SL, Berger JW, Maguire MG, Ho AC. Age-relatedmacular degeneration. N Engl J Med 2000; 342(7):483–92.

11. Evans J, Wormald R. Is the incidence of registrable age-related macular degeneration increasing? Br J Ophthalmol1996; 80(1):9–14.

12. Vingerling JR, Dielemans I, Hofman A, et al. The prevalenceof age-related maculopathy in the Rotterdam Study.Ophthalmology 1995; 102(2):205–10.

13. Klein R, Klein BE, Jensen SC, Meuer SM. The five-yearincidence and progression of age-related maculopathy: theBeaver Dam Eye Study. Ophthalmology 1997; 104(1):7–21.

14. Leibowitz HM, Krueger DE, Maunder LR, et al. TheFramingham Eye Study monograph: an ophthalmologicaland epidemiological study of cataract, glaucoma, diabeticretinopathy, macular degeneration, and visual acuity in ageneral population of 2631 adults, 1973–1975. SurvOphthalmol 1980; 24:335–610.

15. Ferris FL, III, Fine SL, Hyman L. Age-related maculardegeneration and blindness due to neovascular maculo-pathy. Arch Ophthalmol 1984; 102(11):1640–2.

16. Hyman LG, Lilienfeld AM, Ferris FL, III, Fine SL. Senilemacular degeneration: a case-control study. Am J Epidemiol1983; 118(2):213–27.

17. Bressler SB, Bressler NM, Fine SL, et al. Natural courseof choroidal neovascular membranes within the foveal

avascular zone in senile macular degeneration. AmJ Ophthalmol 1982; 93(2):157–63.

18. Guyer DR, Fine SL, Maguire MG, et al. Subfoveal choroidalneovascular membranes in age-related macular degener-ation. Visual prognosis in eyes with relatively good initialvisual acuity. Arch Ophthalmol 1986; 104(5):702–5.

19. Macular Photocoagulation Study Group. Krypton laserphotocoagulation for idiopathic neovascular lesions.Results of a randomized clinical trial. Arch Ophthalmol1990; 108(6):832–7.

20. Barbazetto I, Burdan A, Bressler NM, et al. Photodynamictherapy of subfoveal choroidal neovascularization withverteporfin: fluorescein angiographic guidelines for evalu-ation and treatment—TAP and VIP Report No. 2. ArchOphthalmol 2003; 121(9):1253–68.

21. Macular Photocoagulation Study Group. Argon laser photo-coagulation for senile macular degeneration. Results of arandomized clinical trial. Arch Ophthalmol 1982;100(6):912–8.

22. Macular Photocoagulation Study Group. Argon laserphotocoagulation for ocular histoplasmosis. Results of arandomized clinical trial. Arch Ophthalmol 1983; 101(9):1347–57.

23. Macular Photocoagulation Study Group. Argon laser photo-coagulation for idiopathic neovascularization. Results of arandomized clinical trial. Arch Ophthalmol 1983; 101(9):1358–61.

24. The Moorfields Macular Study Group. Treatment of seniledisciform macular degeneration: a single-blind randomisedtrial by argon laser photocoagulation. Br J Ophthalmol 1982;66(12):745–53.

25. Coscas G, Soubrane G. Argon laser photocoagulation ofsubretinal neovascularization in senile macular degener-ation. Results of a randomized study of 60 cases. Bull MemSoc Fr Ophtalmol 1982; 94:149–54.

26. Cukras C, Fine SL. Thermal laser treatment in AMD:therapeutic and prophylactic. Int Ophthalmol Clin 2007Winter; 47(1):75–93. Review.

27. Yoken J, Duncan JL, Berger JW, et al. Laser photocoagulationfor choroidal neovascularization in age-related maculardegeneration. In: Lim JI, ed. Age-Related Macular Degener-ation. New York: Marcel Dekker, 2002:181–201.

28. Macular Photocoagulation Study Group. Recurrent chor-oidal neovascularization after argon laser photocoagulationfor neovascular maculopathy. Arch Ophthalmol 1986;104(4):503–12.

29. BrownDM, Kaiser PK,Michels M, et al. Ranibizumab versusverteporfin for neovascular age-related macular degener-ation. N Engl J Med 2006; 355(14):1432–44.

30. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab forneovascular age-relatedmacular degeneration. N Engl JMed2006; 355(14):1419–31.

31. Rosenfeld PJ, Rich RM, Lalwani GA. Ranibizumab: phase IIIclinical trial results. Ophthalmol Clin North Am 2006;19(3):361–72.

32. Ferrara N, Damico L, Shams N, et al. Development ofranibizumab, an anti-vascular endothelial growth factorantigen binding fragment, as therapy for neovascular age-related macular degeneration. Retina 2006; 26(8): 859–70.

33. Ferrara N. Role of vascular endothelial growth factor inphysiologic and pathologic angiogenesis: therapeutic impli-cations. Semin Oncol 2002; 29(6 Suppl. 16):10–4.

206 CUKRAS AND FINE

14

Photocoagulation of AMD-Associated CNV FeederVessels: An Optimized ApproachRobert W. FlowerDepartment of Ophthalmology, University of Maryland School of Medicine, Baltimore, Maryland and

Department of Ophthalmology, New York University School of Medicine and the Macula Foundation,

Manhattan Eye, Ear, and Throat Hospital, New York, New York, U.S.A.

INTRODUCTION

Feeder vessel treatment (FVT) of age-related maculardegeneration (AMD)-related choroidal neovasculari-zation (CNV); that is, occlusion of just the vesselsdelivering blood to CNV membranes has long beenviewed as an attractive clinical approach, particularlywhen the neovascular membrane is very near orunderlies the fovea. Available clinical evidence clearlyindicates that this treatment approach is a successfulone that, beyond lesion stabilization, often results invisual improvement. While in recent years the focusof an aggressive quest for an efficacious treatment forAMD-related CNV has been dominated by drug-basedapproaches, refinement of FVT has continued andreached the point that results from its applicationappear to rival and in some cases exceed those ofother currently available methodologies. Moreover,as reaction to the unmet expectations of single-drug-based approaches has led to investigation of combiningextant therapies, FVT—its attributes as a stand-alonetreatment notwithstanding—is an attractive com-bination candidate, because it is not drug based, andit acts directly on the source of blood flow that must bepresent in every viable CNV membrane.

Although elegantly simple as a concept, success-fully implementing a routine FVT method has been aprotracted process. The history of its developmentspans a period of nearly 30 years, and the case canbe made that its development has been coupled tothe evolution of fundus angiography technology,especially choroidal angiography. Today FVT hasbeen refined to take advantage of improvements notonly in the devices used for angiogram acquisitionand application of laser photocoagulation energy, butalso in diagnostic angiogram analysis. In one methodof FVT described here, even the method of applyinglaser energy to feeder vessels (FVs) has been optimizedby introduction of dye-enhanced photocoagulation(DEP),wherein indocyanine green (ICG) dye transitingtargeted FVs at the instant of photocoagulation acts

to selectively enhance absorption of the laser energy,thereby focusing the thermal tissue damage onto thetargeted FV and sparing the surrounding tissues.

ORIGINS OF THE CONCEPT

Perhaps the earliest description of FVT in ophthal-mology was in 1972 by Behrendt, who discussed argonlaser photocoagulation of intraretinal and vitreous FVsof neovascular membranes associated with diabeticretinopathy (1). The then recent availability of visiblelight wavelength lasers led to numerous such novelapproaches aimed at controlling ocular neovascular-ization. Understandably, all of those were related toretinal and anterior segment neovascularization,since they could be directly visualized by means ofreadily available optical devices. The choroidal vascu-lature, on the other hand, was not a popular target ofinterest, since direct visualization of it was obscuredby retinal and choroidal pigments, and in sodiumfluorescein angiography images it appeared mostlyonly as a diffuse “choriocapillaris (CC) flush.” Thedeeper-lying vascular layers remained obscured so faras routine clinical observations were concerned.

At about that same time, in the early 1970s, theconcept of routine clinical angiography of the chor-oidal circulation using ICG dye was being developed.ICG fluorescence angiography initially had beenexplored as an investigative tool for studying chor-oidal blood flow in animal experiments. However,since ICG dye already had a long documentedhistory of biocompatibility, exploring its use inhuman subjects as well was compelling. Since up tothat time, relatively little attention had been paid tothe choroidal circulation compared to the retina,there was no well-defined clinical goal at first invisualizing human choroidal blood flow beyondacademic curiosity, so a rudimentary survey ofboth normal and diseased eyes was undertaken (2).One of the first groups of patients considered in thesurvey was those with macular degeneration.

Figure 1 shows simultaneously acquired fluor-escein and ICG angiogram images of the first patientsuccessfully studied by that methodology. The greatlyimproved ability to visualize the angioarchitectureof AMD-associated CNV lesions afforded by ICGangiography, coupled with the concept of FV photo-coagulation, led to the first attempts at ICG-guidedphotocoagulation of CNV-FVs. Unfortunately, theresults of those first attempts were not encouraging:clear differentiation between CNVafferent and efferentvessels was not easy—or in most cases not possible—since both spatial and temporal resolution of the earlyICG fluorescence angiogram images was limited, thespot size and aiming precision of the first visible lightlaser photocoagulation delivery systems also waslimited, and perhaps most importantly, the laser lightwavelengths available were not ideally suited tothe task.

For some time thereafter, the concept of FVphotocoagulation was not seriously pursued as aclinical tool. Instead, the dominant treatmentapproach for AMD-associated CNV came to bebased on Macular Photocoagulation Study (MPS)recommendations (3). These included destruction ofthe entire CNV membrane—as delineated by fluor-escein angiography—along with an additional marginaround the CNV, even when the procedure resulted inan immediate, non-recoverable additional loss ofvisual acuity (VA). The results of the MPS suggested

that despite an immediate vision loss, three years latera patient so treated statistically would have better VAthan if untreated. Those results notwithstanding,few ophthalmologists remained comfortable with thenotion of having to destroy the retina in order to saveit, preferring for the most part to avoid photocoagula-tion near the fovea.

REVISITING THE CONCEPT

The first notable clinical application of ICG fluor-escence angiography was its use in guiding laserphotocoagulation of CNV. This method was appliedto patients whose clinical and fluorescein angio-graphic features did not meet the eligibility criteriafor laser therapy defined by the MPS recommen-dations; generally it was applied to cases of poorlydefined, or occult, CNV. In this application, use of ICGangiography resulted in improved localization ofabnormal choroidal vessels, thereby making treatmentby photocoagulation possible (4,5). Whereas thisclinical use of ICG undoubtedly contributed tosustaining interest in ICG angiography, arguably itwas the commercial availability of the scanning laserophthalmoscope (SLO) that contributed to increasinginterest in ICG angiography. Compared to the predom-inantly available commercial ICG angiographysystems based on fundus camera optics capable ofacquiring images at a rate of about one per second,

Figure 1 Simultaneously-acquired fluorescein (left frame) and indocyanine green (1CG) (right frame)

angiogram images of the first patient with choroidal neovascularization studied by use of ICG fluorescenceangiography.

208 FLOWER

the SLO afforded the ability to perform high-speedimaging. Ready access to high-speed ICG imageacquisition systems was an important component ofrenewed interest in FV photocoagulation treatment.

The concept of FVphotocoagulationwas revisitedas a treatment for AMD-associated CNV in February,1998 by Shiraga and coworkers; they reported theresults of a pilot trial to assess the feasibility of extra-foveal photocoagulation of subfoveal CNV secondarytoAMD(6). Their use of SLO ICGangiography resultedin the identification of FVs in 37 of 170 consecutivepatients (22%). In 70% of those 37 cases (26 cases)extrafoveal photocoagulation of the FVs, using 575- to630-nm wavelength light, resulted in resolution of theexudative manifestations and improved or stabilizedVA. The following December, Straurenghi and cowor-kers (7), also using SLO ICG angiography, reportedfinding treatable FVs in 15 of 22 patients havingsubfoveal CNV not amenable to the treatmentsuggested by theMPS (3). They successfully obliteratedthe FVs in 40% of the cases, resulting in improved orstabilized VA after more than two years. In a secondgroup of 16 patients they reported a much highersuccess rate of 75%, attributed to the smaller FVdiameters (less than 85 mm) found in this group. InDecember 1997, yet another series of FV treatmentswasbegun using a high-speed, pulsed-laser (HSPL) funduscamera system for FV identification. (Flower RW,Glaser BM, Murphy RP, Macula Soc. Presentation,1999.) The HSPL used in this study consisted of aZeiss fundus camera modified to include a pulsed805-nm-wavelength diode laser for excitation of ICGdye fluorescence in the choroidal circulation; imageswere acquired at a rate of 30 per second (8). In this latterstudy, a higher incidence (about 66%) of FV identifi-cationwas achieved, apparently due to use of theHSPLsystem and different angiogram analysis techniques.Nevertheless, treatment success of the latter studyappears to be equivalent to that of the other groups,even though the follow-up period was shorter and itfocused on occult CNV, whereas the other studiesappear to have focused on classic CNV.

The common experience of all these studieswas that FV photocoagulation appeared to be aviable treatment approach and worthy of continuedpursuit, even though the exact nature of the vesselsbeing treated and the most efficacious application oflaser energy remain to be determined. Clearly, there isa catch-22 associated with this methodology. There areno histological data on treated CNV-FVs, per se, andthe only proof currently available of the accuracy ofangiographic CNV-FV identification is improvementor stabilization of the patient’s VA following treat-ment. But this standard of proof is biased towardfailure, since conventional laser photocoagulation ofCNV-FVs already has proven to be difficult or

incomplete in some cases. Therefore, if the fullpotential of FV treatment is to be accurately assessedand eventually realized, a more consistently successfulapproach to laser photocoagulation must be devised.And at the same time, a much better understandingof the hemodynamic consequences of FV photocoagu-lation must be developed in order to facilitate rationalanalysis of treatment successes and failures.

WHAT IS A FV?

Properly characterizing CNV-FVs in terms of theirlocations within the choroid, their vessel wall structureand the blood flow, is a necessary step in developingthe most efficacious photocoagulation method. Inthat regard, however, histological data about CNVangioarchitecture appear to be at odds with the angio-graphic appearance of the so-called FVs being treated.

Histological Appearance of CNV-FVsThe vessels passing through breaks in Bruch’smembrane and connecting a CNV to the choroidalblood supply can be capillaries, arteries, or veins, asdetermined by the vessel wall structure. In general,CNV complexes up to 300 mm diameter have onlyone break containing a capillary-like vessel (9,10).Complexes on the order of 500 mm have two to fourbreaks, and at least one or two contain a capillary-likevessel; the others transmit only cells. CNV complexesof these dimensions consist of a single layer of capillaryvessels on the inner surface of Bruch’s membrane,and they arise from a layer of vessels which lies justbeneath, instead of between, the intercapillary tissuepillars; so it is assumed these are new vessels replacingthe choroidal capillaries. Because many tissue sectionsmust be cut to find and track these vessels, there areonly a few examples in which the vessels can actuallybe tracked in the choroid, and even then it is not alwaysclear whether they lead to an artery or a vein. (Sarks JPand Sarks SH, written communication, March 14,1999). Complexes on the order of 2000 mm have morethan four breaks, and the vessels passing through areof medium size. These complexes usually are twolayers thick, but still lie beneath the retinal pigmentepithelium (RPE), and they can be served by well-formed arterial and venous vessels. Complexes frompatients with disciform scars have breaks containinglarger arteries and veins; these disrupt the RPE andinvaded the retina. (Sarks SP and Sarks SH, writtencommunication, March 14, 1999).

It has been suggested that on average, there are2.3 vessels passing through Bruch’s membrane andconnecting each CNV to the underlying choroidalvasculature (11). The frequency with which thesevessels are capillaries, arteries, or veins has not yet

14: PHOTOCOAGULATION OF CNV FEEDER VESSELS 209

been reported, but it is clear that the majority ofpenetrating vessels encountered are relatively shortcapillary-like vessels. It is clear also that such smallvessels are not likely to be recognized in ICGangiogram images.

Angiographic Appearance of CNV-FVsThe most frequently identified and treated FVsreported in studies to date appear to be on the orderof one to several millimeters long, a dimension quitelarge with respect to the penetrating vessels mostfrequently found in histological preparations.Figure 2 shows examples of FVs, identified using theHSPL fundus camera system, which have beensuccessfully photocoagulated, resulting in improvedor stabilized vision. In using that system, identificationof FVs is made by first carefully examining the areasurrounding the location of a known or suspectedCNV complex in high-speed ICG angiogram images,

since the most obvious characteristic of a FV isproximity to CNV. Some FVs are easily identified, asin Figure 2 (top left and right), when they are promi-nent and easily distinguishable from adjacentchoroidal vessels. Often, however, FVs are less promi-nent, as in Figure 2 (bottom left and right), andidentification requires use of an analytical techniquesuch as phi-motiona angiography, which helps

Figure 2 Examples of choroidal neovascularization feeder vessels, identified using the high-speed,

pulsed-laser fundus camera system, that were successfully photocoagulated, resulting in improved orstabilized vision. In each case, arrows indicate the course of the feeder vessel. Source: From Ref. 12.

a Phi-motion is a phenomenon first identified byWertheimer in 1912(13); it refers to visual perception of motion where none exists. In asituation where there is a gap in visual information, the brain fillsin what is missing. An example of the case in point is theappearance of two spatially separated points of light whereinfirst one is illuminated and, a finite time later, the second one isilluminated. The perception is that of a single point moving fromthe location of the first point to that of the second. By repeatedlyviewing an appropriate segment of a high-speed angiogram imagesequence in continuous loop fashion and at an appropriate speed,the phi-motion phenomenon accentuates perception of the move-ment of dye through vessels.

210 FLOWER

differentiate a FV from its surroundings by enhancingvisualization of blood flow through it, toward theCNV. Determining direction of flow is essential tocorrectly identifying CNV-FVs—as opposed to theirdraining vessels—even when their angioarchitectureseems obvious.

Reconciling Histological and Angiographic DataClearly, the vessels identified in histological specimensas the conduits of blood from the CC to CNVs appearto be different from the so-called FVs identified in

angiograms. Typically “FV” refers to an afferent vesselsupplying blood to a particular vascular complex, onedirectly connected to the complex. To be precise, in thecase of CNV that definition should apply to the shortcapillary-like vessels that penetrate Bruch’s membraneand form a CNV/CC connection. The vessels in ICGangiograms dubbed FVs in the recently reportedstudies of CNV-FV photocoagulation—especially inthe case of occult CNV—meet the criterion of beingafferent, but they appear to be much larger than thecapillary-like vessels seen in the histological speci-mens. Strictly speaking, therefore, the term “CNV-FV,” as applied in angiographic descriptions, appearsto be amisnomer for some other choroidal vessel; mostlikely Sattler’s layer arterioles.

The so-called FVs seen in angiograms verymuch resemble vessels of the choroidal middle layer,or Sattler’s layer, which lies just beneath the CC.Comparison of the ICG angiogram images of the FVsin Figure 2 to the scanning electron micrographs ofcorrosion casts of the anterior aspect of the CC inFigure 3 demonstrates this similarity. Therefore, itseems a reasonable assumption that the FVs identifiedin ICG angiograms and reported to have been success-fully treated by photocoagulation are Sattler’s layerarteriolar vessels.

There is additional evidence to support thenotion that the angiographically-defined CNV-FVsare Sattler ’s layer vessels: a commonly observedcharacteristic of successfully treated FVs is their“beaded” appearance in ICG angiograms (RPMurphy, symposium presentation, Chicago, June

Figure 3 A scanning electron micrograph of a corrosion cast of

the posterior (Sattler’s) layer of small diameter choroidal arteriesand veins that feed and drain the choriocapillaris, which can be

seen in the background. For the most part, the veins are orientedfrom the upper left-hand corner of the image toward the lower

right-hand corner; they overlie the arteries. Source: From Ref. 12.Courtesy of Dr. Andrzej W. Fryczkowski.

(A) (B)

Figure 4 (A) Indocyanine green (ICG) angiogram demonstrating the commonly observed “beaded”appearance of choroidal neovascularization feeder vessels. (B) The same beaded appearance seen

more prominently in the high-speed ICG angiograms of rhesus monkey eye following carotid arterial dyeinjection. When crossed by small non-dye-filled vessels, the intersections appear as dark segments along

the feeder vessel (vessel indicated by the lower arrow); when crossed by small dye-filled vessels, theintersections appear hyperfluorescence due to additivity of fluorescence from the overlapping vessels

(vessel indicated by the upper arrow). Source: From Ref. 14.


3, 2000); an example of that appearance is shown inFigure 4A. The most likely explanation for the beadedappearance is that the dye-filled FV is crossedthroughout its length by smaller non-dye-filledchoroidal vessels. This same phenomenon is morepronounced in high-speed ICG angiograms of rhesusmonkey eyes following carotid arterial dye injection, asdemonstrated in Figure 4B, wherein carotid dye injec-tion improves dye wave front definition, enhancingobservation of the temporal filling differencesbetween various layers of choroidal vessels. Whencrossed by small non-dye-filled vessels, the crossingsresult in dark segments along the FV; when crossed bysmall dye-filled vessels, the crossings result in hyper-fluorescence, due to additivity offluorescence from theoverlapping vessels. The presence of small vesselsbetween the FV and the CC fixes the FV location wellbelow the CC, consistent with the notion that CNV-FVsare Sattler’s layer vessels.

Additionally, Arnold and coworkers (15) haveshown the choroids of AMD eyes to be as little ashalf the thickness of those in age-matched normaleyes (e.g., 90 mm compared to 180 g), primarily dueto a significant decrease in the number of vessels thatnormally occupy the middle choroidal layers (Sattler’slayer). So it is possible that the relatively high contrastof some FVs (Fig. 2) is a result of there being fewer thannormal adjacent vessels in the same layer, and in theabsence of the normal number of adjacent vessels, theFVs may have become preferential channels for bloodflow through a diminished Sattler’s layer. Therefore,the assumption that many of the FVs investigatorshave identified and photocoagulated are Sattler’slayer arteriolar vessels is at least consistent with theevidence at hand.

THE RELATIONSHIP BETWEEN SATTLER’S LAYERVESSELS (FVs) and CNVs

The explanation for apparently successful photocoa-gulation treatment of so-called CNV-FVs (i.e., Sattler’slayer vessels) lies in the hemodynamic relationshipbetween the Sattler’s layer vessels and the capillary-like vessels that form the CC/CNV communication.

An Anthropomorphic Model ofthe CC/CNV ConnectionThe relationship proposed to exist between these twotypes of vessels is modeled in Figure 5, wherein thereis no anatomical continuity between them, althoughfunctionally they behave as if there were. The figurealso demonstrates how blood could move in a func-tionally contiguous manner from a Sattler’s layer FV,into the CC, and then through a nearby capillary vessel

leading to the CNV during the systolic phase of theintraocular pressure pulse.

By comparison to the sinusoid-like structure ofthe CC vascular plexus, it is likely that resistance toblood flow would be higher through a parallel CNVcomplex, connected to the CC by the capillary-likevessels that penetrate Bruch’s membrane. In thismodel, blood flow through the CNV would occur,but it would not be as great as through the underlyingCC. In keeping with the pulsatile nature of CC bloodflow shown to exist as the result of the perpendicularinterface of arterioles and the wide, flat choriocapil-laries (8,16), a high hydrostatic pressure head mustexist at the interface early during systole, relative tothe surrounding CC (as indicated by the collapsedstate of the choriocapillaries and the CNV vessels inFig. 5A,B). In addition to pushing blood into thechoriocapillaries, the pressure head would be partiallydissipated in forcing some blood into the adjacentpenetrating vessel. Thus, a small, pulsatile pressuregradient would be established through the CNV, eventhough the majority of flow would be through the CC.In this model, closure of the FV or even significantpartial closure would have the effect of reducing thepressure head available at the penetrating vessel to alevel so low that resistance to flow through the CNVcould not be overcome, thereby effectively closing theCNV as well.

Thus, there is considerable evidence to supportthe hypothesis that ultimately the source of bloodsupplying a CNV is a Sattler’s layer arteriole whoseentry into the CC is situated near one of the capillary-like vessels that penetrate Bruch’s membrane, forminga CC/CNV communication. That is, the FVs identifiedfor focal photocoagulation treatment of CNV appearto be Sattler’s layer arterioles that are functionally—but not directly physically—connected to the CNV.Throughout the rest of this discussion, the term CNV-FV is intended to imply a Sattler’s layer vessel that isfunctionally contiguous with a CNV. This leads to thepossibility that in some case a direct, anatomicallycontiguous connection between a Sattler ’s layervessel and a CNV eventually could evolve, obviatingany CC involvement at all; indeed, such an evolutionmight be the path leading from occult to classic CNV.

A Model of the FV/CC/CNV HemodynamicRelationshipThe simple anthropomorphic model of FV/CNVblood flow described above was conceived toaccount for the clinically observed resolution ofretinal edema following FV photocoagulation, evenwhen only partial FV vessel closure is achieved (12).However, since the submacular CC is a true vascularplexus—fed and drained by multiple interspersedarteries and veins—a much more sophisticated

212 FLOWER

(A)

(B)

(C)

(D)

Figure 5 A schematic representation of the presumed relationship between a vessel penetrating Bruch’s

membrane (penetrating vessel) and connecting a choroidal neovascularization (CNV) membrane to thechoriocapillaris (CC). The posterior margin of Bruch’s membrane is represented by the dark horizontal line.

A Sattler’s layer choroidal arteriole (presumably a feeder vessel) is shown entering the CC from beneath.The four frames of the figure indicate how blood would move in a functionally contiguous manner from a

Sattler’s layer feeder vessel, into the CC, and then though a nearby penetrating vessel during the systolicphase of the intraocular pressure pulse even though the penetrating and feeder vessels are not anatomically

contiguous: (A) At the onset of the blood pressure pulse, a high hydrostatic pressure head of blood(represented by the black dots) would develop at the perpendicular interface of arteriole and the wide, flat

CC (as indicated by the collapsed state of the choriocapillaries and the CNV membrane). (B) Slightly laterduring the pulse, In addition to pushing blood into the choriocapillaries, part of the pressure head would be

dissipated in forcing some blood into the adjacent penetrating vessel. Thus, a small pressure gradient wouldbe established through the CNV. (C) Still later, blood flow through the CNV would occur, but it would not be

as great as through the underlying CC, because by comparison to the sinusoid-like structure of the CCvascular plexus, it is likely that resistance to blood flow through a parallel CNV complex, connected to the

CC by capillary-like penetrating vessels, would be higher. (D) Eventually, flow through the CNV would becomplete. Source: From Ref. 12.


model is needed to describe the changes in CC bloodflow beneath the CNV following FV photocoagulation.Therefore, a theoretical model for the human CC,based on available histologic and hemodynamicdata, was developed to simulate the CC blood flowfield before and after FV photocoagulation.b

Known angioarchitectural and hemodynamicparameters for the CC and CNV from the literaturewere used to construct the theoretical model of asection of submacular CC and a small overlyingCNV membrane shown in Figure 6. The CC plexusconsists of two parallel sheets separated by 7.5 mm,between which 10 in. diameter columns are placed atregular intervals, leaving 15 mm wide channels inbetween to simulate the CC plexus. Isolated, but well

separated, precapillary arterioles and venules commu-nicate with the CC plexus and perfuse it with blood.The cross-sectional dimensions of the arterioles andvenules are of the same order as the CC thickness, h.The center-to-center spacing between adjacent arter-ioles and venules is much larger than h. Therefore, theCC was modeled as a planar porous mediumcontaining a widely dispersed set of fluid inflowsand outflows, simulating the feeding and drainingvessels of Sattler’s layer. Feeding arteriolar anddraining venous vessels consist, respectively, of 7.5and 15 mm diameter tubes entering the CC frombeneath.

An overlying CNV membrane was modeled asa parallel miniature version of the CC, but withsmaller dimensions that will result in a significantlyhigher resistance to fluid flow. The communicationbetween the CNV and the CC is by way of twocapillary-dimensioned vessels that penetrate Bruch’smembrane. In the model, the position of the CNV

Figure 6 Schematic representation of the computer simulated model of the choriocapillaris (CC) and anoverlying choroidal neovascular (CNV) membrane. The CC segment is represented by the thin green

rectangular box; the red disks within the volume of the box represent the interstitial spaces surrounded bythe network of choriocapillaries. One Sattler’s layer arteriolar (red cylinder) and one venous (blue cylinder)

vessel are shown connected to the posterior CC. A CNV membrane is represented by the very thin purple

rectangular box. Two capillary-like vessels (green cylinders) penetrate Bruch’s membrane (not depicted)and form the CC/CNV connection (penetrating vessels) is shown; in the text, these are referred to as

penetrating vessels. In the simulation, the position of the penetrating vessels with respect the Sattler’s layervessels was varied. Source: From Ref. 14.

b This model was developed in collaboration with C. von Kerczek.L. Zhu, A. Ernest, C. Eggleton, and L.D.T. Topoleski from theDepartment of Mechanical Engineering University of Maryland,Baltimore, Maryland, U.S.A.

214 FLOWER

could be changed in order to achieve various spatialrelationships between the penetrating vessels and theSattler’s layer vessels that feed and drain the CC.

This theoretical model became the basis forcomputer simulation of blood flow distribution in asegment of human sub-foveal CC approximately1300!1000 mm in area. The actual placement of themultiple Sattler’s layer vessels to feed and drain bloodfrom the simulated CC plexus segment was madeaccording to the histologically determined locationsof those vessels in one normal human eye (17). Figure 7shows the anterior aspect of the computer simulatedsegment of that human submacular CC, marked withthe actual locations of arteriolar and venous vessels,Sattler’s layer vessels connected to its posterior aspect;the figure also shows the simulated CNV in twodifferent locations. Blood flow rates in the feedingarterioles and venules were then estimated bymatching the predicted precapillary arteriole andvenule pressure difference to experimentallymeasured data; the experimentally measuredmaximum pressure difference between a feeding arter-iole and venule was found to be 4.5 mmHg (18).

CNV1

CNV2

500 μm

V2

A 4

A 3

A2

V1

A1

Figure 7 The anterior aspect of the computer simulatedsegment of a human submacular choriocapillaris, marked with

the actual locations of arteriolar and venous vessels Sattler’s

layer vessels connected to its posterior aspect; the figure alsoshows the simulated choroidal neovascularization in two different

locations. Abbreviation: CNV, choroidal neovascularization.Source: From Ref. 14.

Normal Condition

V1V1

3.45

3.45

4.0353.84.645

452.28

2.282.4752.572.863

3.2553.45

2.0851.89

1.6951.499

1.304

1.81.61

2553.06

3.06

3.2553.255 3.45

3.84

0.9140.329

2.0852.281.89

3.2553.06

2.865

2.67

2.4752.8652.65

2.475

3.645

0.914

W

A2

V1

A2

V1

0.3250.643

0.643

0.6431.28

1.281.599

2.8732.554

2.2363.511.917

1.280.643

0.962 0.325

0.643 0.9622.236

1.9171.599

1.280.643

2.5545.42

0.962

0.6430.3252.2361.917

1.599

1.9171.5990.962

0.962

0.962

1.280.962 0.325

Artery #2 Closed 100%

A2V1

G2

A2A2V1

3.0282.887

2.6062.7472.887

3.168

2.7472.466

2.3252.0441.094

1.0942.044

2.7473.028

2.8872.606

2.887

3.028

3.309 3.4493.028

3.168 2.8872.7472.6062.4662.3252.1852.044

1.9041.7631.6231.432

1.2011.342

1.365

2.4662.325

1.7631.623

2.185

2.6062.466

2.325

2.1852.185

2.044

0.217

–0.064

A2

V1

W2

A2

V1

0.322

0.6390.9561.273

1.2732.8593.812.5420.639

0.956

0.322

2.2253.4930.9561.591.59

1.907

1.590.6391.273

1.907

0.322

0.3220.6390.956 0.6390.956

1.2731.591.9073.816.982.542

0.9560.639

0.322

Vein #1 Closed 100%

G4

A2V1

A2V1

3.245

3.479

3.479

3.245

3.011

3.011 3.245 2.777 2.544 2.31 2.076 1.842 1.6081.374

1.140.966

–0.263

2.0762.31

2.5442.777

3.0113.245

3.479

3.4793.2453.011

3.7133.94

W4

V1

A2

V1

A2

0.267

1.8611.5951.0641.33

0.799

0.799

1.0641.338.51.861

1.595

0.7991.0641.064

0.799

0.799

0.799

0.533

0.533

0.533

0.533

0.533

1.331.5951.861.3722.3923.845

2.1261.064

3.845

0.5330.7990.267

0.267

0.267

0.267

Figure 8 Isogramic maps of the blood pressure and blood speed fields of the choriocapillaris (CC)

segment shown in Figure 7 under normal and simulated vascular photocoagulation conditions. Theisogramic lines in the left-hand two frames identify locations of constant pressure (upper frame) and

flow (lower frame) throughout the CC segment under normal conditions. The pattern of these lines change,as shown in the other pairs of frames, when either the underlying Sattler’s layer arteries (middle frames)

or veins (right-hand frames) are occluded. The particular vessels occluded in these examples are aretrioleA1 and venule V1 identified in Figure 7. Source: From Ref. 14.


Experimentally measured pressures and pressuredifferences were applied across the feeding anddraining vessels in order to generate maps of bloodflow through the computer-simulated model CCsegment. Figure 8 shows the normal isobar and iso-blood-speed distributions in the computer simulatedsegment of CC from Figure 7; it also shows how thosedistributions are altered when one of the Sattler’s layerfeeding arterioles is completely occluded.

A significant reduction in the local CC pressureprobably results in significant changes in the bloodflow through an overlying CNV network, since thedriving force for CNV blood flow is the pressuredifference between the capillary-like vessels that pene-trate Bruch’s membrane, forming the CC/CNVcommunication. Clinical observations indicate thatpartial—as well as complete—photocoagulation ofthe (presumed Sattler’s layer) FV adjacent to aCNV’s penetrating vessel(s) is an effective means ofdecreasing the blood flow in the CNV (BM Glaser, RPMurphy, G Staurenghi, personal communications,1999). Therefore, the model also was used to simulateblood flow through a CNV before and after FV laserphotocoagulation; the simulation was performed forthe CNVmembrane situated in two different locations,as indicated in Figure 7. The first location, CNV #1,was between arteriole #2 and venule #1, while thesecond, CNV #2, was between arteriole #3 and a pointequidistant from venules #1 and #2. Photocoagulationof arteriole #2 and of venule #1 resulted in significantreduction of CNV #1 blood flow (71% and 79%,respectively), with similar results in CNV #2 whenarteriole #3 was photocoagulated (84% reduction). Onthe other hand, even the complete closure of venules#1 or #2 produced less than 30% decrease in bloodvelocity through CNV #2.

Implications of the FV/CC/CNVHemodynamic RelationshipThis model predicts that even 50% closure of a bloodvessel entering the posterior aspect of the CC in thevicinity of a capillary-like vessel leading to a CNV canbe effective in reducing or possibly stopping CNVblood flow, regardless of whether that vessel is afeeding arteriole or a draining venule. In otherwords, the important hemodynamic event withrespect to reducing or stopping CNV blood flow issignificant reduction of the blood pressure—hence,blood flow as well—in the local underlying CC.Thus, the predictions of the present computersimulated model support the novel approach toCNV management made previously, namely that(i) rather than total obliteration of a CNV (whichfrequently results in recurrence), the end point oflaser photocoagulation treatment can be reduction ofCNV blood flow to the extent that undesirable

manifestations of the CNV—most notably retinaledema—are halted or reversed and (ii) that CNVblood flow reduction can be mediated by reductionof blood flow through the underlying CC (12).

There are two important implications to thatnovel approach, one related to FV treatment and theother related to the mechanics of successful CNVtreatments in general. Regarding FV photocoagulationtreatment of CNV, the selection criterion for targetedFVs might be extended to include venous as well asarteriolar vessels entering the posterior CC in thevicinity of a CNV membrane. If, indeed, reductionof the underlying CC blood flow is the importanttreatment goal, then depending upon the orientationof the CNV’s penetrating vessels with respect to thefield of vessels feeding and draining the CC, targetingveins or veins in conjunction with arteries may yieldthe best results. After all, the ramifications ofoccluding a venous drainage channel to a truevascular plexus, like the posterior pole CC, is notthe same as occlusion of the drainage vein of a trueend-arteriolar vascular complex. In the former case,blood is diverted to adjacent venous channels,without excessive increase in capillary transmuralpressure; whereas in the latter case, venous occlusionlikely results in blood flow stasis and elevation ofcapillary transmural pressure to a level near thatacross the feeding arterial vessel wall.

Since the predicted relationship between CCand CNV blood flows actually is independent of thespecific means by which CC blood flow is reduced,the second implication of the results is that reductionof CC blood flow underlying a CNV membrane maybe a component mechanism common to successfulCNV photocoagulation treatments, including photo-dynamic therapy (PDT), transpupillaty thermaltherapy (TTT), and drusen photocoagulation. It iswell established that post-PDT angiograms routinelyevidence reduced CC fluorescence (19), and thatappears also to be the case following TTT (20).In the case of TTT, reduced CC blood flow may bedue to increased resistance to plexus blood flowresulting from heat-induced interstitial tissue swel-ling and concomitant reduction of CC lulninal space.Angiographic data specifically related to submacularblood flow following photocoagulation destruction ofmacular drusen have not been presented anywhere;however, it has been demonstrated that CC oblitera-tion occurs with application of moderate to heavylaser burns and that loss of choriocapillaries can addsignificant resistance to blood flow through the CCplexus (8).

If reduced CC blood flow is a componentmechanism of successful CNV treatment, regardlessof the photocoagulation modality used, then FVphotocoagulation arguably might be viewed as the

216 FLOWER

most effective method. The difference between FVphotocoagulation and the other methods is analogousto removing a weed from a lawn by pulling out itsroots (FV) versus just cutting off the weed’s leaves.It can be argued that FV photocoagulation is the mostprecise of the various methods in terms of manipu-lating CC blood flow, and it minimizes the area oftissue–laser interaction. Moreover, since blood flowthrough a particular CC area apparently can bemanipulated by modulation of adjacent venous or asarteriolar vessels connected to the plexus’ anteriorside, it may be that the most precise manipulation ofCC blood flow—and hence, treatment of CNV—will beby controlled, partial photocoagulation of carefullyselected combinations of arterioles and venules inSattler’s layer vessels.

DEVELOPMENT OF A MORE EFFICACIOUSMETHOD OF FV TREATMENT

Themodels of CNV-FVs are consistent with the clinicalobservation that often, even incomplete closure of aFV produces reduction of CNV dye filling, resolutionof associated edema, and improved VA. Of course,partial closure of targeted FVs at present is an un-intended end-point of Argon and Krypton laserphotocoagulation application. In such cases, failureto completely close the relatively deep-lying targetedvessels may be attributable to generation of an insuffi-ciently high temperature gradient, emanating fromthe RPE where laser light-to-heat transductionoccurs. The temperature gradient that is produceddoes extend into the sensory retina and can producesignificant damage there, so the location for FV photo-coagulation must be chosen so as not to involve thefovea. It would be desirable, therefore, to avoid theconcomitant retinal damage and to make FV photo-coagulation more efficient and predictable. This wouldhave the additional potential benefit of allowing suchtreatment to be applied much closer to the fovea thanis presently possible, thereby increasing the numberof patients whomight benefit from CNV-FV treatment.

The Concept of ICG-DEPAn example of a successfully treated FV is shown inFigure 9, and it also shows an undesirable side effect aswell: damage to the nerve fiber layer overlying the siteof FV photocoagulation. Since CNV-FVs apparentlylie below the plane of the CC, a method of photo-coagulation that moves the epicenter of the laser-generated heat closer to those vessels and away fromthe sensory retina would be an improvement over thepresently available method. The concept of ICG-DEPhas that potential and, therefore, should be revisitedfor this application, bearing in mind that its use mustbe optimized to accommodate characteristics of the

targeted choroidal vasculature. The main premise ofDEP is that application of laser light energy witha wavelength matched to the primary wavelengthabsorbed by a bolus of dye passing through thetarget blood vessel produces the most efficient

∗

∗

(A)

(B)

(C)

∗

Figure 9 Post-treatment indocyanine green angiogram images

of a successfully treated feeder vessel. (A) Pre-treatment: theFV is indicated by asterisk. (B) Post-treatment: note lack of CNV

filling. (C) Image shows an undesirable side effect as well:damage to the nerve fiber layer overlying the site of FV photo-

coagulation. Source: From Ref. 12.


photocoagulation burn in terms of vessel closure withminimum damage to surrounding tissue. Figure 10demonstrates the main aspects of ICG-DEP andcompares it to FV photocoagulation by conventionallaser light photocoagulation. The concept ofimproving the efficiency of the photocoagulationprocess by ICG-dye enhancement is not new to

treatment of AMD-related CNV, as Reichel and cowor-kers utilized it for treating poorly defined subfovealCNV. Eventually they reported their initial clinicalinvestigation in 10 patients (21), but in terms ofvisual outcome, their results were equivocal, and thetechnique did not achieve widespread use. Theparticular dye-enhancement technique they usedrelied on absorption of infrared laser light energy bydye-stained choroidal blood vessel walls minutesfollowing dye injection. That apparently is a veryinefficient process, compared to the one in whichthe same laser energy is absorbed by dye moleculeswithin the target vessels during transit of a high-concentration dye bolus (12).

A Combined ICG Angiography/DEP SystemPerformance of ICG-DEP requires use of a laserdelivery system that permits visualization of inter-venously-injected ICG dye as it traverses thevasculature. Such a system was constructed from aZeiss fundus camera (Carl Zeiss, Oberkochen,Germany) modified to include a pulsed diode laserlight source and a synchronized, gated CCD camerafor performing high-speed ICG angiography, as pre-viously described (8,22). The fundus camera wasfurther modified so that the output tip of the fiberoptic of an 810 nm diode laser photocoagulator(Oculight SLx, Iris Medical Instruments, MountainView, California, U.S.A.) can be positioned in theplane of the fundus illumination optics pathwaynormally occupied by the internal fixation pointer;that plane is conjugate to the fundus of the subject’seye. The He–Ne aiming beam emitted by the photo-coagulator appears as a sharply focused spot whenviewed through the fundus camera’s video system,and the position of the fiber optic with respect tothe subject’s fundus can be controlled by the micro-manipulator’s X- and Y-adjustments. With thisconfiguration, it is possible to deliver 810 nm photo-coagulation light pulses to precisely located areas ofthe fundus while observing the fundus with visiblelight through the fundus camera eyepiece, making itpossible to synchronize photocoagulation laser pulsedelivery with arrival of a dye bolus at a targetedvessel site. The fundus camera/laser photocoagula-tion system is shown in Figure 11.

Clinical Application of ICG-DEPUse of the ICG dye-enhanced camera system isdemonstrated in the three frames of Figure 12, whichshow ICG angiogram images from a patienttreated with ICG-DEP. Incarceration of ICG dyeimmediately following laser photocoagulation(center frame) not only provides immediate feed-back as to the success of the procedure, but the

Conventional Visible Light Photocoagulation

Retina

RPE

(A)

ICG Dye Enhanced Photocoagulation

Retina

RPE

(B)

(C)

Figure 10 Schematic comparison of choroidal vessel photo-

coagulation by (A) conventional laser and (B) ICG dye-enhancedlaser and (C) ICG angiogram image made immediately post-

treatment with ICG-DEP demonstrating incarceration of ICG dyein the treated feeder vessel (arrow) and choroidal neovascular-

ization membrane (circle). Abbreviations: ICG, indocyaninegreen; RPE, retinal pigment epithelium. Source: From Ref. 12.

Courtesy of Dr. B. Eric Jones, Baltimore, Maryland, U.S.A..

218 FLOWER

Figure 12 Demonstration of use of the indocyanine green (ICG) dye-enhanced camera system. Left : The

site of application of laser energy during subsequent transit of a high concentration dye bolus (arrow).Middle: Incarceration of ICG dye in the choroidal neovascularization (CNV) (circle) distal to the burn site.

Right: Validation of vessel closure by follow-up ICG angiography a week later (the circle indicates thelocation of the now non-perfused CNV).

(A)

(B)

Figure 11 The fundus camera/photocoagulation system. (A) On the left side of the fundus camera body isa joystick control for positioning the 810 nm wavelength photocoagulation laser beam on the patient’s

fundus. (B) The photocoagulation laser aiming beam (red spot) is visualized on the patient’s live indocyaninegreen (ICG) angiogram, which is seen in the left pane of the monitor located above the patient’s head.

Reference ICG angiogram image from a previously made diagnostic study to determine the location ofa treatable feeder vessel (FV); the targeted FV is indicated on the reference image by a white cross.


incarcerated dye constitutes as a strongly absorbingtarget for further laser application without the need toinject additional dye boluses. The reduction in retinaltissue damage concomitant to FV laser photocoagula-tion using ICG dye-enhancement is demonstrated inFigure 13, which compares the extent of RPE damageresulting from application of identical laser burns toidentical choroidal arteries of a rabbit eye, one withand one without presence of a transiting high-concen-tration dye bolus.

Recently, a single center, prospective, random-ized study of FVT using ICG-DEP was conducted byDr. G. Staurenghi (University of Brescia, Italy) underthe auspices of Novadaq Technologies, Inc. (Toronto,Canada). The objective of the study was to evaluate thesafety and effectiveness of choroidal FV closure in thepresence of ICG using the above described funduscamera/laser photocoagulation system. In the study,forty patients were evaluated for presence of visibleFVs associated with CNV. Upon identification of theFVs, the patients were randomized into one of twotreatment arms: one group of 20 patients was treatedby choroidal FV photocoagulation during ICG dyetransit (ICG-DEP arm), the other group of 20 patients(Control arm) was identically using the same devicesystem, but FV photocoagulation was done withoutICG-DEP, using the laser energy alone. All patientswere followed and/or treated at 2, 4, 8, 12 weeks, and6 months; with 1 additional follow-up at 12 monthspost-first treatment.

The study demonstrated that the funduscamera/laser photocoagulation system was easy touse, and that treatment session times decreased withexperience with the system. The entire diagnostic,treatment and post-treatment confirmation ICG angio-graphy took 21 to 23 minutes; this was similar for bothtreatment arms. On average, four to five treatmentsessions were required for complete treatment in botharms over the course of the study. And on average, theICG-DEP arm used approximately seven times lessenergy/treatment session than the Control arm (5.7 Jper treatment session versus 38.9 J per treatmentsession) to close targeted choroidal FVs.

Importantly, treatment was more effective andmore durable in the ICG-DEP arm, as 90% of thepatients were able to have their choroidal FVs closedor partially closed, with 70% of those vessels remainingclosed at the last treatment assessment, compared to77% and 44%, respectively, in the Control arm. Duringthe course of the study, 45% fewer patients in theICG-DEP armwent on to require alternative treatmentsfor their wet AMD than patients in the Control arm.VA at the end of the treatment phase of this trial, asmeasured by the Early Treatment Diabetic RetinopathyScale, showed that, for the whole treated population,on average the VA was stable, and 29% of all patientsseen at this study milestone had an improvement inVA. Of those patients who completed the study as perthe study prescribed treatment regimen, at the lastscheduled treatment visit, 67% had stable or improved

(A) (B)

Figure 13 Demonstration of the reduction in retinal tissue damage concomitant to feeder vessel laserphotocoagulation using indocyanine green (ICG) dye-enhancement, using identical choroidal arteries

arising from a common origin in a pigmented rabbit eye as a model. (A) Arrows indicate locations oflaser burns of identical energy on the two identical choroidal arterioles. The left-hand burn was applied

without use of ICG dye-enhancement, and the right-hand burn was placed during transit of a high-concentration bolus of ICG dye. (B) Comparison of the extent of retinal pigment epithelium damage

resulting from application of the identical laser burns inferior to the medullary rays.

220 FLOWER

VA, with 42% having one to four line improvement inVA, while 33% had a decrease of more than three linesof VA: none experienced severe vision loss (morethan six lines of VA). Of the nine patients whofollowed the study prescribed treatment regimen andhad a VA equal or better than 20/100 at entry, seven(78%) had stable or improved VA at the last treatmentvisit, with four (44%) having a one to four line improve-ment in VA and two (22%) had more than three linedecrease in VA.

Overall, the study added to the body of datademonstrating the efficacy of the concept of FVTof wetAMD. In addition, it demonstrated that the funduscamera/laser photocoagulation system simplifies FVtreatment by allowing for real-time visualization ofchoroidal FVs during treatment. Moreover, FV photo-coagulation with ICG-DEP produced a more effectiveand more durable treatment outcome than FV photo-coagulation using laser only.

THE FUTURE OF CNV-FV TREATMENT

The current anti-vascular endothelial growth factor(anti-VEGF) drugs, Avastin or Lucentis, have experi-enced significant clinical success to date. It appears thatthese anti-VEGF drugs have such strong anti-per-meability effects on CNV membrane vessels that fluidoutflow into surrounding tissues is reducedor stopped,resulting in stabilized or even improved VA. This canoccur early, before the CNV angioarchitecture issubstantially changed in the process, leaving someCNV blood flow intact; but repeated injections areneeded. Interestingly, this is analogous to whathappens in FVT, where only partial FV closure occursor where reperfusion recurs following completeclosure. Apparently, even partial FV closure results inreduced transmural pressure across CNV membranevessels, which in turn reduces fluid outflow. In thesecases, CNV angioarchitecture also appears substan-tially unchanged, and there is no concomitant recur-rent edema, resulting in stabilized or improved VA. Ithas been postulated that during the period of reducedtransmural pressure, neovascular membrane matura-tion progresses to a level of vessel structural integritysuch that fluid outflowno longer occurs once the higherpre-FVT transmural pressures are reestablished.

If the foregoing understanding of the methods ofaction of the anti-VEGF and FVTs continues to holdtrue, then their use in combination might prove to besymbiotic in a way that leads to a very effectivetreatment approach, since both act to reduce edemaresulting from CNV membrane fluid outflow, but bydifferent pathways. However, as a stand-alone treat-ment, FVT ultimately still may prove to be the mostdesirable approach, since even when repeated treat-ments are applied, those treatments are totally non-

invasive with respect to the peripheral retina and wallof the eye itself, and they are inexpensive. Moreover,consideration should be given to the long-term ramifi-cations of successfully achieving the currently soughtclinical treatment endpoint, namely CNVobliteration.

To the extent that CNV (especially “occult” CNV)serves to augment or replace functionally compro-mised choriocapillaries, successful destruction ofthe CNV ultimately would leave the adjacent sensoryretina and RPE without adequate metabolic supportfrom the choroidal circulation. In that situation,the RPE and retina likely would atrophy. It is for thisreason that CNV blood flow obliteration as the treat-ment endpoint should be reconsidered in favor ofmodulating CNV blood flow just to the point thatretinal edema is ameliorated, since that leaves a levelof choroidal metabolic support for the RPE and retinain place. Owing to FVT’s highly localized applicationand the ability it affords for immediate CNV bloodflow assessment, DEP-FVT allows for a level of indi-vidual patient treatment titration that drug-basedtreatment cannot provide.

Aggressive CNV behavior—rapid membranegrowth, edema formation, etc.—has been viewed asa destructive event, and conventional treatment aimsto remedy such behavior by complete CNV oblitera-tion. But the frequent recurrence of CNV followingsuch treatment could be nature’s continuing effortto compensate for the original—and perhaps nowexacerbated—defect. Instead, such aggressive CNVbehavior could be viewed as an over compensationfor some metabolic or other blood flow related defect.And if laser treatment were to be applied in such away as to just reduce the blood flow to aggressiveCNV by an appropriate amount—perhaps until theCNV vasculature matures—then further aggressivebehavior might be avoided; those cases of inadvertentincomplete FV closure resulting in improved visionwould be examples.

Photocoagulating the FVs supplying CNVassoci-ated with AMD not only can be a successful treatmentmethod (6,7) especially for occult CNV. Indeed, theremay be an important difference between the responseof CNV evoked by direct application of laser energy,as in conventional treatment, and that evoked byreducing blood flow through the otherwise undis-turbed membrane. If ultimately FV photocoagulationtreatment were to be refined along these lines, thelaser would become more a precision instrument tomodulate blood flow than a weapon for destructionof the very retinal tissue whose function we are tryingto conserve. Additionally, because of the pre- and post-treatment high-speed ICG angiograms the methodrequires, information about choroidal hemodynamicsis being accrued that otherwise probably would neverbe available.


SUMMARY POINTS& Identification of FVs is made by first carefully

examining the area surrounding the location of aknown or suspected CNV complex in high-speedICG angiogram images.

& It is a reasonable assumption that the FVs ident-ified in ICG angiograms and reported to have beensuccessfully treated by photocoagulation areSattler’s layer arteriolar vessels.

& Clinical observations indicate that partial—as wellas complete—photocoagulation of the (presumedSattler’s layer) FV adjacent to a CNV’s penetratingvessel(s) is an effective means of decreasing theblood flow in the CNV.

& FV photocoagulation may be the most precisemethod of manipulating CC blood flow and mini-mizes the area of tissue/laser interaction.

& CNV blood flow eradication as the treatmentendpoint should be reconsidered and replacedwith modulation of the CNV blood flow just tothe point that retinal edema is ameliorated, sincethat leaves a level of choroidal metabolic supportfor the RPE and retina in place.

REFERENCES

1. Behrendt T. Therapeutic vascular occlusions in diabeticretinopathy. Arch Ophthalmol 1972; 87:629–33.

2. Patz A, Flower RW, Klein ML, et al. Clinical application ofindocyanine green angiography. In: DeLaey JJ, ed. Inter-national Symposium on Fluorescein Angiography.Documenta Ophthalmologica Proceedings Series. Vol. 9.The Hague: Dr. W. Junk b.v., 1976:245.

3. Macular Photocoagulation Study Group. Subfoveal neo-vascular lesions in age-related macular degeneration:guidelines for evaluation and treatment in the macularphotocoagulation study. Arch Ophthalmol 1991;109:1242–57.

4. Slakter JS, Yannuzzi LA, Sorensen JS, et al. A pilot study ofindocyanine green videoangiography-guided laser treat-ment of primary occult choroidal neovascularizaton. ArchOpthtalmol 1994; 112:465–72.

5. Schwartz S, Guyer DR, Yannuzzi LA, et al. Indocyaninegreen videoangiography guided laser photocoagulation ofprimary occult choroidal neovascularizaton in age-relatedmacular degeneration. Invest Ophthalmol Vis Sci 1995;36:S244.

6. Shiraga F, Ojima Y, Matsuo T, et al. Feeder vessel photo-coagulation of subfoveal choroidal neovascularizationsecondary to age-related macular degeneration. Ophthal-mology 1998; 105:662–9.

7. Staurenghi G, Orzalesi N, La Capria A, et al. Laser treat-ment of feeder vessels in subfoveal choroidal neovascularmembranes: a revisitation using dynamic indocyaninegreen angiography. Ophthalmology 1998; 105:2297–305.

8. Flower RW. Extraction of choriocapillaris hemodynamicdata from ICG fluorescence angiograms. Invest Ophthal-mol Vis Sci 1993; 34:2720–9.

9. Sarks SH. Aging and degeneration in the macular region: aclinicopathological study. Br J Ophthalmol 1976; 60:324–41.

10. Schneider S, Greven CM, Green WR. Photocoagulation ofwell-defined choroidal neovascularization in age-relatedmacular degeneration. Retina 1998; 18:242–50.

11. Green WR, Enger C. Age-related macular degeneration:histopathologic studies. The 1992 Lorenz E Zimmermanlecture. Ophthalmology 1993; 100:1519–35.

12. Flower RW. Experimental studies of indocyanine greendye-enhanced photocoagulation of choroidal neovascular-ization feeder vessels. Am J Ophthalmol 2000; 129:501–12.

13. Wertheimer M. Experimentelle Studien ueber das Sehenvon Bewegung. Z Psychol 1912; 61:161–265.

14. Flower RW, von Kerczek C, Zhu L, Ernest A, Eggleton C,Topoleski LDT. A theoretical investigation of the role ofchoriocapillaris blood flow in treatment of sub-fovealchoroidal neovascularization associated with age-relatedmacular degeneration, copyright 2001 (in press).

15. Arnold JJ, Sarks SH, Killingsworth MC, et al. Reticularpseudodrusen: a risk factor in age-related maculopathy.Retina 1995; 15:183–91.

16. Flower RW. High-speed ICG angiography. In: Yannuzzi LA,Flower RW, Slakter JS, eds. Indocyanine Green Angio-graphy. Mosby: St. Louis, 1997:86–94.

17. Fryczkowski AW, Sherman MD. Scanning electronmicroscopy of human ocular vascular casts: the submacularchoriocapillaris. Acta Anat 1988; 132:265–9.

18. Maepea O. Pressures in the anterior ciliary arteries, chor-oidal veins and choriocapillaris. Exp Eye Res 1992; 54:731–6.

19. Flower RW, Snyder WA. Expanded hypothesis on themechanism of photodynamic therapy action on choroidalneovascularization. Retina 1999; 19:365–9.

20. Reichel E, Berrocal AM, Ip M, et al. Transpupillarythermotherapy (TTT) of occult subfoveal choroidal neo-vascularization in patients with age-related maculardegeneration. Ophthalmology 1999; 106:1908–14.

21. Reichel E, Puliafito CA, Duker JS, et al. Indocyaninegreen dye-enhanced diode laser photocoagulation ofpoorly defined subfoveal choroidal neovascularization.Ophthalmic Surg 1994; 25:195–201.

22. Flower RW. Variability in choriocapillaris blood flow distri-bution. Invest Ophthalmol Vis Sci 1995; 36:1247–58.

222 FLOWER

15

Photodynamic TherapyATul JainDepartment of Ophthalmology, Stanford University Medical Center, Stanford, California, U.S.A.

Darius M. MoshfeghiAdult and Pediatric Vitreoretinal Surgery, Stanford University Medical Center,

Stanford, California, U.S.A.

Mark S. BlumenkranzVitreoretinal Surgery, Department of Ophthalmology, Stanford University Medical Center,

Stanford, California, U.S.A.

INTRODUCTION

Photodynamic therapy (PDT) is a therapeuticmodality that entails the administration of a photo-sensitizer with its subsequent accumulation in thetarget tissue and then its activation by non-thermalmonochromatic light corresponding to the sensitizer’sabsorption profile (1). Powerful oxidizing agents suchas cytotoxic singlet oxygen and free radicals areproduced causing irreversible cellular damage. PDThas traditionally focused on the treatment of cancer(2), but the potential for selective destruction ofdiseased vessels, while sparing normal overlyingtissues, coupled with promising clinical efficacy,resulted in its use for the treatment of age-relatedmacular degeneration (AMD), particularly subfovealchoroidal neovascularization (CNV). PDT selectivityfor the CNV is achieved both through photosensitizerretention in CNV new vessels and through targetedlight application. Illumination is restricted to thediseased area and the limited depth of lightpenetration restricts damage to underlying tissues.

VASCULAR TARGETING

PDT has been used successfully in the treatment ofcertain cancers due to the remarkable selectivity ofmany photosensitizers for tumor tissue. PDT causesdirect cellular injury in addition to microvasculardamage or “shutdown” within the illuminatedtumor. Uptake is considered to be due to the increasedexpression of low-density lipoprotein receptors ontumor cells and neovascular endothelial cells.Porphyrin photosensitization in mammals wasstudied as early as 1910 when Hausmann investigatedthe effects of hematoporphyrin and light on mice (3).The results established the phototoxic propensity of

porphyrins, and Hausmann concluded that theperipheral vasculature was one of the primary PDTtargets. In 1963, Castellani and coworkers demon-strated the microvasculature to be a crucial target (4).PDT-mediated neovascular damage became a main-stay in the treatment of wet AMD and has onlyrecently began to be replaced by newer anti-vascularendothelial growth factor (VEGF) therapies.

Endothelial cells accumulate certain photosensi-tizers and are susceptible to PDT-induced destruction.The subcellular localization of motexafin lutetium(Lu-Tex) was determined in human umbilical veinendothelial cells using fluorescence microscopy.Lu-Tex exhibits a fluorescence emission profile at750 nm and this signature fluorescence marker isused to characterize and quantify sensitizer concentra-tions within the tissues. Lu-Tex was found to localizewithin the lysosomes and endoplasmic reticulum asevidenced by co-staining with organelle-specificfluoroprobes. Following illumination, some relocaliza-tion of the sensitizer occurred with partitioning beingobserved in the mitochondria, suggesting that theprimary subcellular localization site could notpossibly fully account for all of the PDT-induceddamages. Sensitizer-alone and light administration-alone treatment groups did not induce any changesin the cell viability. Significant cell death due toLu-Tex-mediated PDT was observed in endothelialcells producing a steep dose response.

Vascular occlusion following PDT is marked bythe release of vasoactive molecules, vasoconstriction,blood cell aggregation, endothelial cell damage, bloodflow stasis, and hemorrhage. The response is dependenton sensitizer type, concentration, and the time intervalbetweenadministrationand treatment.Benzoporphyrinderivative monoacid ring A (BPD-MA)-induced

PDT resulted in selective destruction of tumormicrovasculature in a chrondosarcoma rodent modelwhen compared with the surrounding normal micro-vasculature; illuminationwasappliedwithin 30minutesfollowing sensitizer administration (5). However, noacute change in vascular status was observed whenillumination occurred at three hours. The vascular shut-down results correlated with the anti-tumor effect sincetumor-bearing animals treated at five minutesresponded more positively than those treated atthree hours.

LIGHT APPLICATION

The light used for ophthalmic applications is non-thermal monochromatic laser light matched to thesensitizer’s far-red (infrared) absorbance profile. Infra-red light possesses greater transmission through bothblood and tissue than light at lower wavelengthsthereby enabling the treatment of pigmented orhemorrhagic lesions. The energy at which light isdelivered is a product of the radiant power (expressedin milliwatts per square centimeter, mW/cm2) and thetime of illumination. The radiant energy, often termedfluence, is expressed as joules per square centimeter(J/cm2). Therefore, to deliver a fluence of 50 J/cm2

light at a power density of 600 mW/cm2, an illumina-tion time of 83 seconds is required.

Upon illumination, photons (hy) interact with theground singlet state sensitizer (1Sensitizer) causing itto undergo an electronic transition to an activatedshort-lived excited singlet state (1Sensitizer*). Thesinglet state can then either convert back to theground state causing fluorescence or undergo inter-system crossing to generate the longer-lived excitedtriplet state sensitizer (3Sensitizer*). From the tripletstate, a photon can be emitted causing phosphor-escence with conversion to the ground state or thetriplet state can interact with oxygen or biologicalsubstrates leading to microvascular damage (6,7).Two photooxidation processes can occur between thetriplet state and molecular oxygen (3O2) causing irre-versible damage to vascular components. The directinteraction of the excited triplet state with biomole-cular substrates is termed the type-I mode and isfavored in areas with low oxygen concentrations.Biomolecular radicals are generated and react withoxygen-forming cytotoxic oxidizing products. Thetype-II mechanism entails interaction from theexcited triplet state sensitizer to ground stateoxygen-producing singlet oxygen ð1OÞwith theoreticalregeneration of the ground state sensitizer. However,photobleaching and photoproduct formation candeplete the ground state sensitizer concentration.

Singlet oxygen is highly electrophilic, oxidizingbiological substrates and initiating a cascade of radical

chain reactions that damage cellular components.Singlet oxygen production is thought to be responsiblefor most of the damage induced by PDT. Singletoxygen possesses a reactive path length of less than0.02 mm so that any effect has a limited potency (2). Thephotochemical processes involved are complex andare different for each sensitizer and are also subject tothe microenvironment. Intersystem crossing is kineti-cally important for the formation of the excited tripletstate and for PDT potency. Molecules with highfluorescence quantum yields will generate lowertriplet quantum yields and are more likely to beused as diagnostic agents. Conversely, moleculeswith low fluorescence quantum yields will generatehigh triplet quantum yields and therefore shouldproduce a high yield of cytotoxic species.

PDT AGENTS

The ideal photosensitizer should be chemically pureand possess the appropriate physical and biologicproperties that make it inherently non-toxic untilactivated by light. The agent should possess strongabsorption properties in the far-red spectral region(660–780 nm) where light has greatest penetrationinto blood and tissue and possess efficient photophy-sical properties for destroying neovascular endothelialcells. The sensitizer should also localize selectively inthe neovasculature while being rapidly cleared fromthe blood and overlying photoreceptors. In addition,rapid cutaneous clearance would limit cutaneousphotosensitivity. Several photosensitizers wereexplored and underwent different stages of preclinicaland clinical development.

Photosensitizing candidate molecules aregenerally related to porphyrins. Porphyrins are fusedtetrapyrrolic macrocycles that are omnipresent innature as major biological pigments. ProtoporphyrinIX, a typical porphyrin molecule, forms the non-protein portion of hemoglobin. Reduction, oxidation,or expansion of the macrocyclic ring leads to differentmolecular subclasses. A reduction at one of the fourpyrrole rings in the porphyrin macrocycle yields achlorin molecule. The electronic conjugation systemis altered causing further absorption into the far-redwavelength region, from 630 to approximately 660 to690 nm. Increasing the macrocycle conjugation systemfurther, by the formation of a pentadentate, metallo-photosensitizer yields a texaphyrin molecule andresults in even further absorption in the far-redspectral region (700–760 nm). Phthalocyanines aretetrapyrrolic structures fused together by nitrogenatoms instead of carbon bridges; absorption is exhib-ited in the 650- to 700-nm wavelength region.Purpurins possess a reduced pyrrole ring and also

224 JAIN ET AL.

an extended ring conjugation system; the absorptionmaxima is between 650 and 690 nm.

BENZOPORPHYRIN DERIVATIVE MONOACID(VERTEPORFIN, VISUDYNEe, BPD-MA)

BPD-MA consists of equal amounts of two regio-isomers that differ in the location of the carboxylicacid and methyl ester on the lower pyrrole rings of thechlorin macrocycle. BPD-MA, due to its hydrophobi-city, is formulated with liposomes. The monoacidanalogues were developed because they producedgreater PDT responses compared with the diacids (8).The monoacid regioisomers are converted, in the liver,to the diacids. The regioisomers responded similarlyin experimental efficacy settings; however, the phar-macokinetic properties were different in the rat, dog,and monkey but not in humans, where the plasmahalf-life was five to six hours (9,10). It is thought thelatter may be due to differences in plasma esterases orlipoprotein profiles.

PDT studies undertaken using experimentallyinduced CNV in primates resulted in closure of theneovasculature and choriocapillaris, but not theretinal vasculature.

Liposomal BPD-MA was infused at a dose of0.375 mg/kg for 10 to 32 minutes. Illumination withinfrared light at a fluence of 150 J/cm2 (689–692 nmlaser light at 600 mW/cm2) occurred 30 to 55 minutesfollowing the start of the infusion (7). When the sametreatment parameters were performed on normalprimate eyes, some retinal pigment epithelium (RPE)damage and choriocapillaris closure occurred locallywith little damage in contiguous tissues. When lightwas delivered within 30 to 45 minutes followingsensitizer delivery, sensitizer administration rateshad little effect on vascular occlusion rates. BPD-MAlocalization in the choroid and RPE was confirmedusing fluorescence microscopy in rabbits. Retentionoccurred within five minutes with progression to theouter segments within 20 minutes. No BPD-MA wasdetected within the choroid or photoreceptors at twohours; however, a small trace was detected in the RPEat 24 hours (11). A similar pharmacokinetic patternwas observed in monkeys using in vivo fluorescenceimaging (12).

The long-term effects on the retina and choroidwere evaluated in cynomolgus monkeys with experi-mental CNV (13). Fundus photography andangiography analyses were performed at 24 hoursand then weekly for four to seven weeks following atreatment with 0.375 mg BPD-MA/kg and a fluence of150 J/cm2. Eyes were examined histologically at theend of the follow-up period. CNV closure also resultedin the closure of the choriocapillaris with damageoccurring to RPE cells. However, these areas appeared

to regenerate somewhat in the four to seven weeksstudy period. Of 28 CNV lesions followed for fourweeks, 72% remained closed.

However, lesion retreatment was necessary tosustain vascular closure. The effect of three differentdosing treatments was evaluated in disease-freeprimate eyes (14). Treatments, using sensitizer dosesof 6, 12, or 18 mg/m2, 20 minutes after drug infusionand a fluence of 100 J/cm2 were performed every twoweeks. A cumulative dose response was observed.Damage to the retina, choroid, and optic nerve waslimited in the 6 mg/m2 sensitizer subgroup. The higherdose groups exhibited severe choriocapillaris andphotoreceptor damage at six weeks.

Many other photosensitizer agents [i.e., tin ethyletiopurpurin (Purlytine, SnET2), (Optrine, lutetiumtexaphyrin, Lu-Tex), mono-L-aspartyl chlorin e6 (NPe6or MACE), chloroaluminum sulfonated phthalocya-nine (AlPcS4), and ATX-S10] have been explored forthe treatment of exudative AMD and other retinalconditions; however, they have not been used inclinical practice (15–27). Verteporfin PDT hasemerged as the dominant therapeutic option forexudative AMD since the publication of the previousedition of this book, and we will focus most ofour discussion to the clinical results with thisphotosensitizer.

LIGHT CONSIDERATIONS

Generally any light source that is matched to thephotosensitizer’s absorption profile can be used forPDT. For ophthalmology, fiberoptic delivery of a lasersource is required to permit focusing on the retina witha slit lamp system. Lasers are needed because high-energymonochromatic collimated light can be coupledefficiently to fiber optics allowing delivery within anacceptable time frame. Diode lasers that are stable,compact, and relatively inexpensive in the 630- to 730-nm wavelength range are readily available.

CLINICAL OUTCOMES

PDT is a superior alternative to laser photocoagula-tion for subfoveal CNV. Using preclinical CNVmodels, the neovascularization and normal chorioca-pillaris can be closed while preserving the outer andinner retina. In contrast, during the process ofdestroying neovascularization lying beneath the RPEand sensory retina with laser photocoagulation,thermal conductance to the retina results in acutenecrosis of all layers of the retina. This later resultsin atrophy leading to loss of vision. However, withPDT treatment, visual acuity generally remains stableimmediately after treatment and has been shown,in a minority of patients, to improve immediately.

15: PHOTODYNAMIC THERAPY 225

This suggests that the photoreceptors and innerretinal elements are generally preserved (28).

Verteporfin Human TrialsThe safety and efficacy of verteporfin (BPD-MA,Visudynee) have been confirmed (Table 1) in phaseI, II, and III clinical trials (28,34,35). The phase I andphase II studies proved that a single treatment ofverteporfin PDT could occlude CNV vessels for oneto four weeks following administration, as measuredby fluorescein angiography (34). The maximal toler-ated light dose, defined by retinal closure, was150 J/cm2. The minimal light dose required toachieve closure of the vessels was 25 J/cm2.

Treatment of Age-Related Macular Degenerationwith Photodynamic Therapy TrialThe one-year results of the Treatment of Age-RelatedMacular Degeneration with Photodynamic Therapy

(TAP) Study were published in 1999 (28). The studyconsisted of two multicenter, double-masked, placebo-controlled randomized trials with identical protocols.Eligible AMD patients had subfoveal CNV whosegreatest linear dimension was up to 5400 mm andbest-corrected visual acuity ranged from 20/40 to20/200. Verteporfin at 6 mg/m2 was infused intrave-nously for 10 minutes. Then, a diode laser was used toactivate the dye (689-nm diode laser, 50 J/cm2,600 mW/cm2, 83-second duration, spot size 1000 mmlarger than greatest linear diameter of the CNV lesion)15 minutes after the start of infusion. Patients wereevaluated by clinical examination and fluoresceinangiography approximately every three months, andretreated at the discretion of the treating ophthalmol-ogist. Of the 609 eyes enrolled in the study (402treatment and 207 placebo), 94% completed the 12months follow-up. In the treatment group, 246 (61%)of 402 eyes lost fewer than 15 letters of visual acuity

Table 1 Summary of Treatment of Age-Related Macular Degeneration with Photodynamic Therapy Reports 1–6

TAP Report # Study design Follow-up (months) Main outcomes

1 (28) Two multicenter, double-masked, placebo-

controlled randomized clinical trials

12 61% PDT versus 47% placebo had less than 15

ETDRS letters loss (p!0.001)Predominantly classic SFCNV subgroup: 67%

PDT versus 39% placebo had less than 15

ETDRS letters loss (p!0.001)2 (29) Two multicenter, double-masked, placebo-

controlled randomized clinical trials

24 53% PDT versus 38% placebo had less than 15

ETDRS letters loss (p!0.001)Predominantly classic SFCNV subgroup: 59%

PDT versus 31% placebo had less than 15

ETDRS letters loss (p!0.001)3 (30) Subgroup analysis of TAP 24 Predominantly classic SFCNV subgroup at

12 mo: 33% PDT versus 61% placebo had at

least 15 ETDRS letters loss (p!0.001)Predominantly classic SFCNV subgroup at

24 mo: 41% PDT versus 69% placebo had at

least 15 ETDRS letters loss (p!0.001)Predominantly classic SFCNV subgroup at

24 mo: 55% PDT versus 32% placebo greater

than 20/200 visual acuity (p!0.001)4 (31) Subgroup analysis of TAP 24 Predominantly classic SFCNV subgroup at

24 mo: loss of six or more letters of contrast

sensitivity was 21% PDT versus 45% placebo

(p!0.05)Predominantly classic SFCNV subgroup at

24 mo: loss of 15 or more letters of contrast

sensitivity was 7% PDT versus 12% placebo

(p!0.05)5 (32) Open-label extension of TAP 36 Predominantly classic SFCNV subgroup treated

with PDT: 37.5% at 24 mo versus 41.9% at

36 mo lost at least 15 ETDRS letters

Predominantly classic SFCNV subgroup treated

with PDT: visual acuity change ofK1.9 lines

at 24 mo versusK2.0 lines at 36 mo

6 (33) Natural history data from TAP 40% of patients in the placebo arm with

minimally classic disease converted to

predominantly classic SFCNV

Abbreviations: ETDRS, Early Treatment Diabetic Retinopathy Study; PDT, photodynamic therapy; SFCNV, subfoveal choroidal neovascularization; TAP,treatment of age-related macular degeneration with photodynamic therapy.

226 JAIN ET AL.

from baseline, compared with 96 (47%) of 207 placeboeyes, a difference that was statistically significant(p!0.001) (28). Subgroup analysis demonstrated thegreatest benefit (67% vs. 39% losing less than 15 lettersof visual acuity, p!0.001) for those eyes with predo-minantly classic CNV (greater than 50% of the entirelesion being classic CNVat baseline before treatment).No significant lasting adverse effects were reported(28). The results at various time points followingenrollment are summarized in Table 1 (28–33). Theaverage number of verteporfin PDT treatments was 3.4by 12 months and 5.6 by 24 months (28,29). Thetreatment effect for verteporfin PDT of predominantlyclassic subfoveal CNV persisted at 24 months (29).Additionally, for the subgroup of predominantlyclassic CNV patients, those treated with PDT weremore likely to have visual acuity greater than 20/200at the 24-month follow-up (30).

Verteporfin in Photodynamic Therapy TrialIn the Verteporfin in Photodynamic Therapy (VIP)Study, patients with pathologic myopia, occult CNV,and classic CNV (with visual acuity better than 20/40)

were evaluated (Table 2) (36–39). For pathologicmyopia with subfoveal CNV, patients treated withverteporfin PDT were less likely than placebo to lose8 and 12 Early Treatment Diabetic Retinopathy Study(ETDRS) letters at the 12-month follow-up (36).Additionally, at the 24-month follow-up, the distri-bution of change in visual acuity favored the PDTgroup over placebo (38). One arm of the VIP trialevaluated verteporfin for treatment of occult-onlyCNV with at least 50 letters on the ETDRS scale orsome classic component with at least 70 letters (betterthan 20/40) on the ETDRS scale (37).

For the subgroup of occult-only CNV, the PDTgroup was less likely than placebo to lose 15 and 30ETDRS letters at the 24-month follow-up. For thesubgroup of patients with a visual acuity score ofless than 65 ETDRS letters or lesion size less than orequal to four disc areas, verteporfin PDT-treatedpatients were less likely than placebo to lose 15and 30 ETDRS letters at the 24-month follow-up.While the overall safety profile was favorable, 4.4%of PDT-treated patients lost at least 20 ETDRS letterswithin seven days of treatment (37). This loss of at

Table 2 Summary of Verteporfin in Photodynamic Therapy Reports 1–4

VIP Report # Study designFollow-up(months) Main outcomes

1 (36) Multicenter, double-masked, placebo-

controlled randomized clinical trial for

treatment of patients with SFCNV due

to pathologic myopia

12 72% PDT versus 44% placebo lost fewer than eight

ETDRS letters (p!0.01)86% PDT versus 67% placebo lost fewer than 15

ETDRS letters (pZ0.01)2 (37) Multicenter, double-masked, placebo-


treatment of patients with occult

SFCNV (at least 50 ETDRS letters) or

some classic CNV (at least 70 EDTRS

letters)

24 54% PDT versus 67% placebo lost at least 15 ETDRS

letters (pZ0.023)30% PDT versus 47% placebo lost at least 30 ETDRS

letters (pZ0.001)Occult-only subgroup: 55% PDT versus 68% placebo

lost at least 15 ETDRS letters (pZ0.032)Occult-only subgroup: 29% PDT versus 47% placebo

lost at least 30 ETDRS letters (pZ0.004)Subgroup visual acuity score less than 65 ETDRS letters

or lesion size less than or equal to four disc areas at

baseline: 49% PDT versus 75% placebo lost at least

15 ETDRS letters (p!0.001)Subgroup visual acuity score less than 65 ETDRS letters

or lesion size less than or equal to four disc areas at

baseline: 21% PDT versus 48% placebo lost at least

30 ETDRS letters (p!0.001)4.4% of PDT versus 0% of placebo lost at least 20

ETDRS letters within 7 days of treatment

3 (38) Multicenter, double-masked, placebo-


treatment of patients with SFCNV due

to pathologic myopia

24 36% PDT versus 51% placebo lost at least eight ETDRS

letters (pZ0.11)Distribution of change in vision favored PDT (pZ0.05)

4 (39) Prospective non-comparative case

series looking at patients from control

group in VIP trial

24 Continued monitoring for patients with occult with no

classic lesions

If acuity decreases or predominantly classic features

develop, PDT should be considered

Abbreviations: CNV, choroidal neovascularization; ETDRS, Early Treatment Diabetic Retinopathy Study; PDT, photodynamic therapy; SFCNV, subfovealchoroidal neovascularization; VIP, verteporfin in photodynamic therapy.


least 20 ETDRS letters from baseline visual acuitywithin seven days was termed acute severe visualdecrease (40).

TAP and VIP TrialsTheTAPandVIP trial datawere combinedandanalyzedin a series of reports (Table 3) (40–43). The most signi-ficant data to be gleaned from these reports was thatbaseline lesion size was the most important predictor ofvisual acuity following verteporfin PDT, regardless oflesion composition (41). Size was a significant factor forpatients with predominantly classic lesions greater thanone disc area, minimally classic lesions less than fourdisc areas, and occult-only lesions less than five discareas (42).

Verteporfin PDT was also evaluated for thetreatment of subfoveal CNV secondary to pathologicmyopia, ocular histoplasmosis, angioid streaks, andidiopathic causes (44,45). The main findings of these

papers were that verteporfin was well tolerated,effective in decreasing fluorescein leakage, andbroadly applicable to subfoveal CNV, regardless ofetiology. Unfortunately, both studies are limited bytheir small numbers and lack of controls.

The Photodynamic Therapy of Ocular Histoplas-mosis Study trial has yielded two reports to date(46,47). In a non-comparative, prospective study, 56%and 45% of patients gained at least seven ETDRSletters following verteporfin for subfoveal CNVsecondary to ocular histoplasmosis at the 12- and24-month follow-up periods, respectively.

Evolution of PDT TreatmentSpaide and colleagues popularized the use of conco-mitant intravitreal triamcinolone acetonide and PDT(48). They demonstrated that combination therapyresulted in improved visual acuity and lack of fluor-escein leakage following therapy, with the greatest

Table 3 Summary of Treatment of Age-Related Macular Degeneration with Photodynamic Therapy and Verteporfin in PhotodynamicTherapy Reports 1–4

TAP and VIPReport # Purpose Follow-up (months) Outcomes

1 (41) To determine the effect of lesion size and

visual acuity in patients with SFCNV

treated with verteporfin PDT

24 Baseline: mean predominantly classic lesion

(3.4 DA) smaller than occult-only (4.3 DA) and

minimally classic (4.7 DA)

Visual acuity change from baseline to 24 mo:

significant treatment effect for lesion size effect

(smallerOlarger), but not for composition orbaseline visual acuity (pZ0.01)

For the entire TAP and VIP population, only

lesion size was a significant predictor

following treatment (pZ0.032 and 0.043 withand without last observation carried forward

respectively)

Lesion size was significant at p!0.05 for thefollowing lesion compositions: predominantly

classic greater than 1 DA, minimally classic

less than 4 DA, and occult-only less than 5 DA

Lesion size was a significant predictor of at least

15 ETDRS letters loss at 24 mo (pZ0.009)2 (42) To describe angiographic guidelines for PDT 24 Guidelines presented and examples given for

interpretation of angiograms following PDT

3 (40) To describe acute severe visual acuity loss

(20 or more ETDRS letters) within 2–

4 days of PDT

24 15 occurrences in 14 eyes of 14 patients (0.7%

in TAP and 4.4% in VIP trial)

11 events occurred following first treatment

4 (43) To determine safety data in the TAP and VIP

trials

24 Ocular and non-ocular adverse events: 92.3%

PDT and 89.1% placebo (pZ0.114)Higher “visual disturbances” following

verteporfin: 22.1%PDT versus 15.5% placebo

(pZ0.054) in TAP trial and 41.7% PDT versus

22.8% placebo (p!0.001) in VIP trial

Injection site reactions: 13.1% PDT versus 5.6%

placebo (p!0.001)Photosensitivity reactions: 2.4% PDT versus

0.3% placebo (pZ0.016)Infusion-related back pain: 2.4% PDT versus

0.0% placebo (pZ0.004)

Abbreviations: DA, disk areas; PDT, photodynamic therapy; SFCNV, subfoveal choroidal neovascularization; TAP, treatment of age-related macular degenerationwith photodynamic therapy; VIP, verteporfin in photodynamic therapy.

228 JAIN ET AL.

effect seen in treatment-naıve patients (48). Theseresults were durable, lasting out to 12 months, andthe most frequent side effect was increased intraocularpressure in 38.5%. These results have been supportedby similar work, which suggests that the results arebroadly applicable to all sub-types of AMD (49–55).

Other combination therapies include othersteroids and anti-VEGFdrugs. Recently, retinal special-ists have been utilizing dexamethasone in combinationwith PDT. The data from this and other steroid com-bination studies should be published in the near future.The FOCUS Study studied the combination of rani-bizumab (Lucentis, Genentech, South San Francisco,California, U.S.A.) and PDT versus PDT alone as atreatment for AMD patients with predominantlyclassic CNV. The FOCUS Study showed that thecombined use of PDT with ranibizumab (Lucentis)was better than PDT alone (56). Further details on theFOCUS Study are found in Chapter 8 of this book.

SUMMARY POINTS

& Verteporfin PDT has proven itself useful in thetreatment of subfoveal CNV of several etiologies.

& There are many exciting reports indicating that theefficacy of verteporfin PDT can be enhanced withthe concomitant intravitreal triamcinolone aceto-nide for all types of subfoveal and non-subfovealCNV (46,48,50,53,57–60).

& While recent anti-VEGF therapies hold muchpromise for the treatment of CNV, there areinitial reports that verteporfin PDT might play asignificant role as an adjunctive treatment used inconjunction with these new therapies (56,61–65).

& There is a need for long-term prospective studies toquantify and validate these initial reports.

& Verteporfin PDT has been shown to be useful in thetreatment of subfoveal CNV for AMD as well asother etiologies.

& Use of verteporfin PDT can be enhanced with theconcomitant use of intravitreal triamcinoloneacetonide.

& Many studies are currently underway evaluatingthe use of verteporfin PDTwith other steroids andanti-VEGF agents with initially promising initialresults.

& It may be that, in the future, the role of verteporfinPDTwill be as an adjunct in combination with anti-VEGF or other intravitreal therapies.

REFERENCES

1. Henderson BW, Dougherty TJ. How does photodynamictherapy work? Photochem Photobiol 1992; 55(1):145–57.

2. Dougherty TJ, Gomer CJ, Hender BW, et al. Photodynamictherapy. J Natl Cancer Inst 1998; 90(12):889–905.

3. Hausmann W. Die sensibilisierende wirkung des hamato-porphyrins. Biochem Z 1911; 30:276–316.

4. Castellani A, Page G, Concioli M. Photodynamic effect ofhaematoporphyrin on blood microcirculation. J PatholBacteriol 1963; 86:99–102.

5. Fingar VH, Kik PK, Haydon PS, et al. Analysis of acutevascular damage after photodynamic therapy using benzo-porphyrin derivative (BPD). Br J Cancer 1999; 79(11–12):1702–8.

6. Fingar VH. Vascular effects of photodynamic therapy. J ClinLaser Med Surg 1996; 14(5):323–8.

7. Husain D, Miller JW, Kenney AG, et al. Photodynamictherapy and digital angiography of experimental irisneovascularization using liposomal benzoporphyrinderivative. Ophthalmology 1997; 104(8):1242–50.

8. Richter AM, Waterfield E, Jain AK, et al. Photosensitisingpotency of structural analogues of benzoporphyrin deriva-tive (BPD) in a mouse tumour model. Br J Cancer 1991;63(1):87–93.

9. Levy J, Chan A, Strong A. The clinical status of benzopor-phyrin derivative. Proc SPIE 1995; 2625:86–95.

10. Richter AM, Jain AK, Canaan AJ, et al. Photosensitizingefficiency of two regioisomers of the benzoporphyrinderivative monoacid ring A (BPD-MA). Biochem Phar-macol 1992; 43(11):2349–58.

11. Haimovici R, Kramer M, Miller JW, et al. Localization oflipoprotein-delivered benzoporphyrin derivative in therabbit eye. Curr Eye Res 1997; 16(2):83–90.

12. Husain D, Miller J. Photodynamic therapy of exudativeage-related macular degeneration. Semin Ophthalmol 1997;12:14–25.

13. Husain D, Kramer M, Kenny AG, et al. Effects of photo-dynamic therapy using verteporfin on experimentalchoroidal neovascularization and normal retina andchoroid up to 7 weeks after treatment. Invest OphthalmolVis Sci 1999; 40(10):2322–31.

14. Reinke MH, Canakis C, Husain D, et al. Verteporfinphotodynamic therapy retreatment of normal retina andchoroid in the cynomolgus monkey. Ophthalmology 1999;106(10):1915–23.

15. Mori K, Yoneya S, Ohta M, et al. Angiographic andhistologic effects of fundus photodynamic therapy with ahydrophilic sensitizer (mono-L-aspartyl chlorin e6).Ophthalmology 1999; 106(7):1384–91.

16. Kilman GH, Puliafito CA, Grossman GA, et al. Retinal andchoroidal vessel closure using phthalocyanine photody-namic therapy. Lasers Surg Med 1994; 15(1):11–8.

17. Kilman GH, Puliafito CA, Stern D, et al. Phthalocyaninephotodynamic therapy: new strategy for closure of chor-oidal neovascularization. Lasers Surg Med 1994; 15(1):2–10.

18. Asrani S, Zeimer R. Feasibility of laser targeted photo-occlu-sion of ocular vessels. Br J Ophthalmol 1995; 79(8):766–70.

19. Obana A, Gohto Y, Kanai M, et al. Selective photodynamiceffects of the new photosensitizer ATX-S10(Na) on chor-oidal neovascularization in monkeys. Arch Ophthalmol2000; 118(5):650–8.

20. Mang TS, Allison R, Hewson G, et al. A phase II/III clinicalstudy of tin ethyl etiopurpurin (Purlytin)-induced photo-dynamic therapy for the treatment of recurrent cutaneousmetastatic breast cancer. Cancer J Sci Am 1998; 4(6):378–84.

21. Primbs GB, Casey R, Wamser K, et al. Photodynamictherapy for corneal neovascularization. Ophthalmic SurgLasers 1998; 29(10):832–8.

22. Blumenkranz MS, Woodburn KW, Quing F, et al. Lutetiumtexaphyrin (Lu-Tex): a potential new agent for ocularfundus angiography and photodynamic therapy. AmJ Ophthalmol 2000; 129(3):353–62.


23. Rockson SG, Lorenz DP, Cheong WF, et al. Photoangio-plasty: an emerging clinical cardiovascular role forphotodynamic therapy. Circulation 2000; 102(5):591–6.

24. Kereiakes DJ, Szyniszewski AM,Whar D, et al. Phase I drugand light dose-escalation trial of motexafin lutetium and farred light activation (phototherapy) in subjects with coron-ary artery disease undergoing percutaneous coronaryintervention and stent deployment: procedural and long-term results. Circulation 2003; 108(11):1310–5.

25. Verigos K, Stripp H, Mick R, et al. Updated results of aphase I trial of motexafin lutetium-mediated interstitialphotodynamic therapy in patients with locally recurrentprostate cancer. J Environ Pathol Toxicol Oncol 2006;25(1–2):373–87.

26. Graham KB, Arbor JD, Coonnolly EJ, et al. Digital angio-graphy using lutetium texaphyrin in a monkey model ofchoroidal neovascularization. Invest Ophthalmol Vis Sci1999; 40:402 (Abstract).

27. Arbor JD, Connolly EJ, Graham K, et al. Photodynamictherapy of experimental choroidal neovascularizationusing intravenous infusion of lutetium texaphyrin. InvestOphthalmol Vis Sci 1999; 40:401 (Abstract).

28. Treatment of Age-Related Macular Degeneration withPhotodynamic Therapy (TAP) Study Group. Photodynamictherapy of subfoveal choroidal neovascularization in age-related macular degeneration with verteporfin: one-yearresults of 2 randomized clinical trials-TAP Report. ArchOphthalmol 1999; 117(10):1329–45.

29. Bressler NM. Photodynamic therapy of subfoveal choroidalneovascularization in age-related macular degenerationwith verteporfin: two-year results of 2 randomized clini-cal trials-TAP Report 2. Arch Ophthalmol 2001; 119(2):198–207.

30. Bressler NM, Arnold J, Benchaboune M, et al. Verteporfintherapy of subfoveal choroidal neovascularization inpatients with age-related macular degeneration: additionalinformation regarding baseline lesion composition’s impacton vision outcomes-TAP Report No. 3. Arch Ophthalmol2002; 120(11):1443–54.

31. Rubin GS, Bressler NM. Effects of verteporfin therapy oncontrast on sensitivity: results from the treatment of age-related macular degeneration with photodynamic therapy(TAP) investigation-TAP Report No. 4. Retina 2002; 22(5):536–44.

32. Blumenkranz MS, Bressler NM, Bressler SB, et al. Verte-porfin therapy for subfoveal choroidal neovascularizationin age-related macular degeneration: three-year results ofan open-label extension of 2 randomized clinical trials-TAPReport No. 5. Arch Ophthalmol 2002; 120(10):1307–14.

33. Bressler SB, Pieramici DJ, Koester JM, et al. Natural historyof minimally classic subfoveal choroidal neovascular lesionsin the treatment of age-related macular degenerationwith photodynamic therapy (TAP) investigation: outcomespotentially relevant to management-TAP Report No. 6. ArchOphthalmol 2004; 122(3):325–9.

34. Miller JW, Schmidt-Erfurth U, Sickenberg M, et al. Photo-dynamic therapy with verteporfin for choroidalneovascularization caused by age-related macular degener-ation: results of a single treatment in a phase 1 and 2 study.Arch Ophthalmol 1999; 117(9):1161–73.

35. Schmidt-Erfurth U, Miller JW, Sickenberg M, et al.Photodynamic therapy with verteporfin for choroidalneovascularization caused by age-related macular degener-ation: results of retreatments in a phase 1 and 2 study. ArchOphthalmol 1999; 117(9):1177–87.

36. Verteporfin in Photodynamic Therapy (VIP) StudyGroup. Photodynamic therapy of subfoveal choroidal

neovascularization in pathologic myopia with verte-porfin. 1-year results of a randomized clinical trial-VIPReport No. 1. Ophthalmology 2001; 108(5):841–52.

37. Verteporfin in Photodynamic Therapy (VIP) Study Group.Verteporfin therapy of subfoveal choroidal neovasculariza-tion in age-related macular degeneration: two-year resultsof a randomized clinical trial including lesions with occultwith no classic choroidal neovascularization—Verteporfinin Photodynamic Therapy Report 2. Am J Ophthalmol 2001;131(5):541–60.

38. Blinder KJ, Blumenkranz MS, Bressler NM, et al. Verte-porfin therapy of subfoveal choroidal neovascularizationin pathologic myopia: 2-year results of a randomizedclinical trial-VIP Report No. 3. Ophthalmology 2003;110(4):667–73.

39. Pieramici DJ, Bressler SB, Koester JM, et al. Occult withno classic subfoveal choroidal neovascular lesions in age-related macular degeneration: clinically relevant naturalhistory information in larger lesions with good visionfrom the Verteporfin in Photodynamic Therapy (VIP)Trial: VIP Report No. 4. Arch Ophthalmol 2006;124(5):660–4.

40. Arnold JJ, Blinder KJ, Bressler NM, et al. Acute severevisual acuity decrease after photodynamic therapy withverteporfin: case reports from randomized clinical trials-TAP and VIP Report No. 3. Am J Ophthalmol 2004;137(4):683–96.

41. Blinder KJ, Bradley S, Bressler NM, et al. Effect of lesionsize, visual acuity, and lesion composition on visual acuitychange with and without verteporfin therapy for choroidalneovascularization secondary to age-related maculardegeneration: TAP and VIP Report No. 1. AmJ Ophthalmol 2003; 136(3):407–18.

42. Barbazetto I, Burdan A, Bressler NM, et al. Photodynamictherapy of subfoveal choroidal neovascularization withverteporfin: fluorescein angiographic guidelines for evalu-ation and treatment-TAP and VIP Report No. 2. ArchOphthalmol 2003; 121(9):1253–68.

43. Azab M, Benchabourne M, Blinder KJ, et al. Verteporfintherapy of subfoveal choroidal neovascularization in age-related macular degeneration: meta-analysis of 2-yearsafety results in three randomized clinical trials: treatmentof age-related macular degeneration with photodynamictherapy and verteporfin in photodynamic therapy studyReport No. 4. Retina 2004; 24(1):1–12.

44. Sickenberg M, Schmidt-Erfurth U, Miller JW, et al. Apreliminary study of photodynamic therapy using verte-porfin for choroidal neovascularization in pathologicmyopia, ocular histoplasmosis syndrome, angioid streaks,and idiopathic causes. Arch Ophthalmol 2000; 118:327–36.

45. Lim JI, Flaxel CJ, Labree L. Photodynamic therapy forchoroidal neovascularisation secondary to inflammatorychorioretinal disease. Ann Acad Med Singapore 2006;35(3):198–202.

46. Rosenfeld PJ, Saperstein DA, Bressler NM, et al. Photo-dynamic therapy with verteporfin in ocular histoplasmosis:uncontrolled, open-label 2-year study. Ophthalmology2004; 111(9):1725–33.

47. Saperstein DA, Rosenfeld PJ, Bressler NM, et al. Photo-dynamic therapy of subfoveal choroidal neovascularizationwith verteporfin in the ocular histoplasmosis syndrome:one-year results of an uncontrolled, prospective case series.Ophthalmology 2002; 109(8):1499–505.

48. Spaide RF, Sorenson J, Maranan L. Combined photody-namic therapy with verteporfin and intravitrealtriamcinolone acetonide for choroidal neovascularization.Ophthalmology 2003; 110(8):1517–25.

230 JAIN ET AL.

49. Augustin AJ, Schmidt-Erfurth U. Verteporfin and intra-vitreal triamcinolone acetonide combination therapy foroccult choroidal neovascularization in age-related maculardegeneration. Am J Ophthalmol 2006; 141(4):638–45.

50. Augustin AJ, Schmidt-Erfurth U. Verteporfin therapycombined with intravitreal triamcinolone in all types ofchoroidal neovascularization due to age-related maculardegeneration. Ophthalmology 2006; 113(1):14–22.

51. Chan WM, Lai TY, Wong AL, et al. Combined photody-namic therapy and intravitreal triamcinolone injection forthe treatment of subfoveal choroidal neovascularisation inage related macular degeneration: a comparative study. BrJ Ophthalmol 2006; 90(3):337–41.

52. Ergun E, Maar N, Ansari-Shahrezaei S, et al. Photody-namic therapy with verteporfin and intravitrealtriamcinolone acetonide in the treatment of neovascularage-related macular degeneration. Am J Ophthalmol 2006;142(1):10–6.

53. Nicolo M, Ghiglione D, Lai S, et al. Occult with no classicchoroidal neovascularization secondary to age-relatedmacular degeneration treated by intravitreal triamcinoloneand photodynamic therapy with verteporfin. Retina 2006;26(1):58–64.

54. Ruiz-Moreno JM, Montero JA, Barile S. Triamcinolone andPDT to treat exudative age-related macular degenerationand submacular hemorrhage. Eur J Ophthalmol 2006;16(3):426–34.

55. Schmidt-Erfurth U, Michels S, Augustin A. Perspectives onverteporfin therapy combined with intravitreal corticoster-oids. Arch Ophthalmol 2006; 124(4):561–3.

56. Heier JS, Boyer DS, Ciulla TA, et al. Ranibizumab combinedwith verteporfin photodynamic therapy in neovascularage-related macular degeneration: year 1 results of theFOCUS Study. Arch Ophthalmol 2006; 124(11):1532–42.

57. Marticorena J, Gomez-Ulla F, FernandezM, et al. Combinedphotodynamic therapy and intravitreal triamcinolone acet-onide for the treatment of myopic subfoveal choroidalneovascularization. Am J Ophthalmol 2006; 142(2):335–7.

58. Marticorena J, Gomez-Ulla F, Fernandez M, et al. Photo-dynamic therapy and high-dose intravitreal triamcinoloneto treat exudative age-related macular degeneration: 1-yearoutcome. Retina 2006; 26(6):602–12.

59. SpaideRF, Sorenson J,MarananL.Combinedphotodynamictherapy and intravitreal triamcinolone for nonsubfovealchoroidal neovascularization. Retina 2005; 25(6):685–90.

60. Spaide RF, Sorenson J, Maranan L. Photodynamic therapywith verteporfin combined with intravitreal injection oftriamcinolone acetonide for choroidal neovascularization.Ophthalmology 2005; 112(2):301–4.

61. Aggio FB, Melo GB, Hofling-Lima AL, et al. Photodynamictherapy with verteporfin combined with intravitreal injec-tion of bevacizumab for exudative age-related maculardegeneration. Acta Ophthalmol Scand 2006; 84(6):831–3.

62. Brown DM, Kaiser PK, Michaels M, et al. Ranibizumabversus verteporfin for neovascular age-related maculardegeneration. N Engl J Med 2006; 355(14):1432–44.

63. Kim IK, Husain D, Michaud N, et al. Effect of intravitrealinjection of ranibizumab in combination with verteporfinPDT on normal primate retina and choroid. InvestOphthalmol Vis Sci 2006; 47(1):357–63.

64. Moshfeghi AA, Rosenfeld PJ, Pulifito CA, et al. Systemicbevacizumab (Avastin) therapy for neovascular age-relatedmacular degeneration: twenty-four-week results of anuncontrolled open-label clinical study. Ophthalmology2006; 113(11):2002–11.

65. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab forneovascular age-related macular degeneration. N EnglJ Med 2006; 355(14):1419–31.


16

Radiation Treatment in Age-RelatedMacular DegenerationChristina J. FlaxelCasey Eye Institute, Oregon Health & Science University, Portland, Oregon, U.S.A.

Paul T. FingerNew York University School of Medicine, The New York Eye Cancer Center, New York,

New York, U.S.A.

INTRODUCTION

Age-related macular degeneration (AMD) is aleading cause of rapid and severe visual loss andlegal blindness in developed countries (1,2). Tenmillion Americans are visually disabled due toAMD and 10% of patients aged 66 to 74 showsigns of AMD (3,4). Estimated prevalence is from7% to 30% in persons aged 75 to 85 years (4–6). The“wet” form of AMD is responsible for the mostsevere and rapid vision loss. In North America,200,000 to 400,000 people will develop this form ofAMD each year. Wet AMD accounts for 12% of casesoverall but 90% of cases of legal blindness (see chap8 on Wet AMD) (4).

Vision loss due to neovascular (wet) AMDinvolves the growth of abnormal “new” vesselsthrough breaks in Bruch’s membrane from thechoroid and under the retinal pigment epithelium(RPE). Unfortunately these new vessels, called chor-oidal neovascular membranes (CNV), leak serum,blood, and other exudates, resulting in many of theproblems related to the wet form of the disease (refer tothe chap 1onpathologyofAMD). It is these newvesselsthat have been the primary target of most currenttherapies (see Chapters 13–18). Of these therapies,laser treatment, photodynamic therapy (PDT), intra-vitreal injections with pegaptanib sodium andranibizumab have been proven by prospectiverandomized clinical trials to be effective in treatingCNV (7–35). Other anti-angiogenic agents that aretargeting these new vessels are undergoing clinicaltrials (36–40) (Chapter 17).

RATIONALE FOR RADIATION THERAPY FOR AMD

When compared with proven and experimental treat-ment methods, theoretical advantages of radiationtherapy include absence of iatrogenic mechanical orthermal laser damage and systemic side effects (41).

An additional advantage is that eyes with primarilyoccult CNV are eligible for treatment (41) as are eyeswith extensive subretinal hemorrhage. In addition,unlike PDT, pegaptanib or ranibizumab sodium treat-ment, radiation is known to exert a longer-term effecton tissues. Thus repeated treatments may not benecessary. Radiation therapy for AMD has beenstudied for the past 10 years. Eight randomized,controlled trials have evaluated the use of variousradiation types and methods of delivery for treatingAMD. Thus far, studies have shown varying degreesof benefit from radiation therapy, with a trend towardbetter results with higher radiation doses and fewer(larger) treatment fractions.

The scientific rationale for using radiationtherapy for a benign disease characterized by neovas-cular growth is based on experimental and clinicalevidence. Radiation is known to potentially destroyvascular tissue (42–45). Specifically, low-dose radi-ation has been shown to inhibit neovascularization(46–49). For example, in plaque-irradiated choroidalmelanomas, a ring of chorioretinal atrophy iscommonly found around the tumor’s base anddecreased or absent blood flow are demonstrated byfluorescein angiography (Fig. 1). These findingsdemonstrate the ability of radiation to destroynormal and neovascular blood vessels, but the resul-tant chorioretinal atrophy is an unacceptable endpointwhen treating macular degeneration (49,50).

If relatively low-dose radiotherapy could inhibitCNVand secondary disciform scars, this would lead tobetter visual outcomes (Fig. 2A,B). The main questionpersists: is there a therapeutic window in which thedose of radiation used is high enough to induceregression of CNV but low enough to spare thenormal retina and choroid? Radiation specialistsbelieve this is possible. Proliferating endotheliumis more susceptible to radiation damage than non-proliferating capillary endothelial cells and largervessels, and thus neovascular endothelial cells and

inflammatory cells are particularly radiosensitive (51).There is also the potential for radiotherapy to inhibitfurther neovascular growth and induce neovascularregression by inducing programmed cell death andmodifying the growth factor profiles of the neovascularcomplexes (44,52). This has been shown by theregression of both benign intracerebral arteriovenousmalformations and choroidal hemangiomas afterradiation therapy (53–55). Finally, it is thought thatinflammation may play a role in neovascularizationand, as noted, radiation inhibits the inflammatoryresponse (44,56).

Radiation ToxicityDoses of ionizing radiation absorbed by the body arereported in conventional units called grays (Gy) orSysteme Internationale (SI) units called rads, repre-senting a given quantity of energy delivered per gramof tissue. A rad is 100 ergs of energy per gram of tissue,while a gray (Gy) equals 100 rads [as does a GrayEquivalent, used when the type of radiation is notstandard (i.e., the charged particles of proton beamirradiation)]. The potential toxicity of radiation is wellknown (45,47,49,57–61). However, studies have shownthat the normal neural retina and choroid are relativelyradiation resistant (60,47). It is also known that factorsinfluencing the development of radiation retinopathyinclude total dose delivered, daily fraction size, pre-existing microangiopathy, and diabetes or priorchemotherapy (50,51).

Fractionation of the radiation involves dividingthe total amount of radiation into smaller doses anddelivering these doses over an extended period oftime. These small, frequent doses allow healthy cellstime to grow back, repairing damage inflicted by theradiation. However, fraction size affects dose; forexample, 400 cGy delivered in one over a fiveminutes period does not equal to 200 cGy per day fortwo days. The 400 cGy delivered over a short period oftime will deliver a higher overall amount of radiationthan the second dosage method.

While radiation-induced retinopathy has beenreported at doses of 30 to 35 Gy, it is more commonlyassociated with doses of 45 to 60 Gy (Fig. 3). Radi-ation optic neuropathy is rare at doses below 50 Gy(Fig. 4A–C) (50,51,57–59). Fraction sizes greater than2.5 Gy may predispose to toxicity, especially withtotal doses greater than 45 Gy (58,59). However,there is increasing evidence that fractionated doses

Figure 1 Example of chorioretinal atrophy at the edge of amelanoma treated with radioactive plaque.

(A) (B)

Figure 2 (A) Late phase fluorescein angiogram showing subfoveal choroidal neovascular membrane prior

to treatment. (B) Late phase fluorescein angiogram of same lesion six months post-14 gray equivalent (GE) protonbeam irradiation without leakage.

234 FLAXEL AND FINGER

with larger daily fraction sizes are lower than stan-dard overall doses can be delivered safely andeffectively to small regions. Lens doses of 15 Gy ormore will induce cataract and transient dry eye;keratitis and epiphora are expected complications(57–59,62). Other concerns with external beamtherapy are radiation exposure of the brain andcontralateral eye (Fig. 5) (56–59).

PRIOR STUDIES AND ALTERNATE DELIVERYMETHODS FOR RADIATION TREATMENT

Initial reports regarding radiation for AMD began toappear in the literature in 1993 (63). Chakravarthy’spreliminary results described 19 patients who weretreated with radiation therapy for subfoveal CNVdue to AMD. The study also included sevenmatched control subjects. At one year, 63% oftreated patients showed stabilization of vision,while there was deterioration of acuity in all controleyes over the same time period. By image analysis,this study also showed significant neovascularmembrane regression in 77% of treated patients atone year, with concurrent progressive enlargement ofthe neovascular membranes in all control subjects(63). These results and those from other centers ledto a prospective, randomized British trial of radiationtherapy which reported results in 2002 and is dis-cussed below in more detail (64). Multiple additionalreports on external beam radiotherapy (EBRT) andplaque radiotherapy have showed promising butvariable results.

In 1996, Finger and colleagues reported theresults of low-dose EBRT of 12 to 15 Gy and plaque

Figure 3 Radiation retinopathy one year following proton beamradiation using 14 gray equivalent (GE).

(A)

(B)

(C)

Figure 4 (A) Post-radiation optic neuropathy after 14 grayequivalent (GE) proton beam treatment, one year after treat-

ment. (B,C) Note in these early and late fluoresceinangiography’s that the choroidal neovascular membrane

(CNVM) appears dry.

16: RADIATION TREATMENT IN AMD 235

radiotherapy with equivalent dosage in 137 patients.They found decreased subretinal hemorrhages,exudates, and leakage of neovascular membraneswith maintenance of visual acuity (65). Subsequently,Stalmans et al. reported failure to control CNV withradiation dosage of 20 Gy in 2 Gy fractions in 111patients (66). Spaide and colleagues reported similarfindings in 1997, when 10 Gy delivered in 5 Gy frac-tions that failed to control neovascular growth inAMD. This study never disclosed what percentageof treated patients had recurrent CNV (previouslytreated by laser photocoagulation) (67). Severalfurther reports in 1998 and 1999 reported possiblebeneficial effects of radiation. Conducted in France,the Radiotherapy Study, conducted in France,reported potential benefit to 16 Gy delivered in four-sessions of 4 Gy each with mean follow-up of 6.4months (68). In 1999, a second French group fromFrance also reported stabilization of visual acuity andanatomical outcome in eyes with AMD (69). However,this group also reported a significant rate of compli-cations, including radiation retinopathy, opticneuropathy, choroidal vasculopathy, and branchretinal vein occlusion when patients received dosesof either 20 Gy in five fractions via lateral beam(effectively, a 30 Gy dose) or 16 to 20 Gy in four tofive fractions delivered via lateral arc (69). This studydid not include a control group. Follow-up timeranged from 12 to 24 months (69).

Chakravarthy undertook a meta-analysis ofPhase I clinical trials utilizing low-dose externalbeam radiotherapy. Results were published in 2000(70). This report suggested that low-dose EBRTinhibited exudative AMD, but that higher doseswere more effective in preventing severe vision loss

(O6 lines on the Snellen visual acuity chart) (70). Insupport of this conclusion, Berginks and colleaguesreported good results with relatively high-radiationdoses of 24 Gy and concluded that there was a dose–response effect, with more favorable effects at higherdosages (71,72).

The only published study to evaluate treatmentof recurrent CNV with radiation was by Marcus andcolleagues who reported the safety and visualoutcome of radiation treatment (73). They treated 18eyes consecutively with seven fractions of 2 Gy for atotal dose of 14 Gy, then treated the next 16 eyes withfive fractions of 3 Gy for a total dose of 15 Gy. Theyfound no radiation toxicity, but also no significantdifferences in contrast sensitivity or fluorescein angio-graphy stabilization rates, though they noted a trendfor palliative benefit with higher fraction sizes of 4 Gyor higher (73).

Implant Radiation Therapy (Brachytherapy)Several groups have used brachytherapy to deliver arelatively high dose of radiation to the involvedmaculawith less irradiation of surrounding structures, usingmethods developed for localized treatment of oculartumors (Fig. 6A,B) (74–76). Finger and his groupemployed plaque radiotherapy in eyes with neovas-cular AMDwith no adverse effects (65,74). They foundno sight-limiting complications in this Phase I clinicaltrial, in which they treated 23 eyes with palladium-103plaques (Figs. 7 and 9) (74). Encouraged by theseearly results, they enrolled an additional eight eyesand treated all eyes with a mean dose of 17.62 Gyby palladium-103 plaque (76). Their seven yearresults were reported in 2003, with the conclusionthat most patients experienced decreased exudationor stabilization with the dosage employed. They rec-ommended a randomized clinical trial to evaluatebrachytherapy for AMD treatment (Fig. 8A,B) (76).Since this report, the group has increased their dose(35 Gy–2 mm from the inner sclera) utilizing 10 mmeye plaques and pallidium-103 seeds. They have notedno adverse effects and everyone had promising results(personal communication).

Charged Particle Radiation TherapyIn June 2000, Friedrichsen and Flaxel published onthe use of proton beam irradiation for subfovealCNV in AMD along with their data from the PhaseI/II planned dose-escalation clinical trial (77). Thismethod of irradiation allows a higher dose (and doserate) to be delivered to a specific volume of tissue.Like most forms of external beam radiation therapy,proton beam therapy requires an entry site andirradiates all the tissue in its path. However, protonH dose volumes are limited to a section of the eye,decreasing irradiation of normal tissues outside the

Figure 5 External beam radiotherapy (EBRT) dose overlay

showing radiation delivered to other structures during EBRT.Source: From Ref. 65.


beam, and in the contralateral eye. Proton beamirradiation was delivered as a single dose, utilizinglight field patient orientation with temporal beamentry, initially with 8 GE (Grey equivalent) beginningin March 1994 and increasing to 14 GE in March1995. No acute radiation-related adverse effectswere noted. Twenty-one eyes were treated with8 GE followed by an initial stabilization of subretinalleakage on fluorescein angiography (FA) in 50% ofeyes at 12-month follow-up but with regrowth in allbut three eyes at 15-month follow-up. However, inthe 14 GE-treated eyes, 83% showed no leakage after12 months of follow-up and 78% of eyes hadunchanged or improved vision. For those eyesfollowed for longer than nine months (in the 14GE-treated group), 83% with 20 out of 100 or bettervision prior to proton beam treatment showedimprovement in vision. Also, severe visual loss

increased up to 37% at two years with 8 GE-treatedeyes, while with 14 GE, the incidence of severe visualloss was 3.7% throughout the follow-up period.There were no cases of cataract, dry eye, lash loss,or optic neuropathy in any of the study eyes and noradiation retinopathy in the 8-GE group; however,radiation retinopathy was found in 48% of eyestreated with 14 GE at a mean of 14 months. Therewas one case outside the study of severe proliferativeradiation retinopathy and optic neuropathy withinone year of treatment, with severe visual loss. Theauthors concluded that their preliminary datasuggest that proton beam irradiation correlates withCNV regression, maintains visual function, is moreeffective at 14 GE, is less beneficial in larger lesionsand that radiation complications are more commonwith longer follow-up but only in the 14-GE group(77). Because of the significant risk of complications,proton beam treatment is not recommended untilfurther studies can be done regarding dose deliveryand with consideration of fractionization ofthe dosage.

REVIEW OF CONTROLLED RADIATIONSTUDIES FOR AMD

There are now several completed studies in the use ofradiation in AMD (Table 1). Reports on randomizedtrials of radiation treatment include those from Holz’sRAD Study Group in Germany (78), Kobayashi andcolleagues in Japan (79), the United Kingdom group

(A)

(B)

Figure 6 (A) Dose overlay of plaque radiation delivery. (B)

Typical plaque used for brachytherapy treatment for AMD(10 mm). Source: From Ref. 65.

Figure 7 A palladium-103 plaque assembly with seeds prior toimplantation.


(64), Valmaggia and colleagues from Switzerland (80),Marcus et al. from the Medical College of Georgia(81,82), and the Age-related Macular DegenerationRadiation Trial (AMDRT) study group report fromthe United States (84,85).

The RAD study is a randomized, prospective,double-blind, placebo-controlled trial performed atnine centers throughout Germany (78). This studyenrolled 205 patients who were treated with eithereight fractions of 2 Gy (101 eyes), or eight fractions of0 Gy (104 eyes). At one-year follow-up, no benefit wasseen in either classic or occult subfoveal CNV due toAMD (approximately 50% of treated eyes had only

occult CNV, while the other half had a combination ofclassic and occult disease). There have been no seriouscomplications relating to the radiation treatment todate (78).

A randomized, prospective, placebo-controlledtrial was also carried out at a center in Japan (79).This study enrolled 101 patients and followed themfor two years. They also reported no significant treat-ment-related side effects from a total dose of 20 Gydelivered in 10 divided doses over a period of 14 days,with irradiation through a single lateral port.The investigators concluded that radiotherapyshowed a beneficial effect comparedwith no treatment,with favorable factors being smaller area of CNV,higher degree of occult CNV, and better initial visualacuity (79). Both groups are continuing follow-up onall patients.

Hart et al. reported the results of a large, multi-center randomized trial in theUnitedKingdom in 2002.This trial included 203 patients randomly assignedeither to radiotherapy using 12 Gy of 6 mV photons(delivered in six fractions) or observation (64). Theydid not find a statistically significant benefit to radi-ation treatment and felt their results did not supportthe routine use of radiation treatment for AMD (64).

A Swiss group reported the results of 18-monthfollow-up in 161 patients with subfoveal CNV whowere enrolled in a prospective study (80). The exam-iners treated the posterior pole of the affected eyewith 1 Gy (4!0.25 Gy) in the control group and 8 Gy(4!2 Gy) or 16 Gy (4!4 Gy) in the treatment groups.They found that patients with classic CNV, or withinitial distance visual acuityR20/100, benefited morefrom treatment. However, no significant differencewas found between control and treatment groups inreading ability and size of CNV (80). They alsoreported no radiation treatment side effects in anygroup (80).

At Association for Research in Vision andOphthalmology (ARVO) in 2003, Marcus andcolleagues from the Medical College of Georgiareported the four-year results of a small, double-masked clinical trial that included 42 observed and 41treated eyes (81,82). They used low-dose external beamirradiation at 14 Gy in seven fractions of 2 Gy, andreported no benefit and possible detriment to visionin long-term (2–4 years) follow-up (81,82).

Another Japanese study published in 2004 withtwo-year follow-up utilized external-beam radiationtherapy in 21 eyes of 18 patients, with a group of 15non-treated controls (83). This group reportedimproved or maintained visual acuity rates of 81% inthe treated group versus 40% in the control group (83).This study, however, was non-randomized. Anothernon-randomized trial from Japan also reported short-term benefit to low-dose radiation in 68 eyes (86).

(A)

(B)

Figure 8 (A) Fluorescein angiography of an eye with CNV

before implantation of palladium-103 plaque. (B) Fluoresceinangiography of the same eye following treatment with

palladium-103 plaque. Abbreviation: CNV, choroidal neovascularmembranes.


The problem of conflicting data from multiplestudies led the National Eye Institute to sponsor aprospective randomized pilot study in the UnitedStates (84,85). This non-funded, multi-center pilotstudy included two groups of patients randomized

to either treatment or observation, and was called theAMDRT (84,85). Eligibility criteria for the new subfo-veal CNV study included lesions not amenable toMacular Photocoagulation Study (MPS) laser treat-ment, classic, mixed or occult CNV by FA, blood

(B)(A)

(C)

Figure 9 TheraSightw Ocular Brachy-therapy System. (A) Assembled

TheraSight System. (B) Representationof device behind the macula. (C) Close-

up of applicator with lever engagedretract the shield.

Table 1 Comparison of Published Clinical Trial Results Utilizing Radiation in AMD

Author No. of patients Radiation dosage (total) Conclusion (follow-up)

Holz (78) 205 8 fractions 2 Gy (16 Gy) No benefit (1 yr)

Kobayashi (79) 101 10 fractions 2 Gy (20 Gy) C benefit (2 yr) stable vision and stable lesion size

Hart (64) 203 6 fractions 6 mV photons (12 Gy) No benefit (2 yr)

Valmaggia (80) 161 4 fractions of 2 Gy (8 Gy) or 4

fractions of 4 Gy (16 Gy)

C benefit (18 mo) less lines of vision lost in both treated

groups

Marcus (81,82) 83 7 fractions of 2 Gy (14 Gy) No benefit (2 yr), possible detriment to vision in long term

(4 yr)

Churei (83) 36 10 fractions of 6 mV X rays (20 Gy) C benefit (2 yr) improved or maintained vision

Marcus (73) 34 7 fractions of 2 Gy (14 Gy) or 5

fractions of 3 Gy (15 Gy)

No benefit (1 yr)

AMDRT

(84,85)

88 5 fractions of 4 Gy (20 Gy) No benefit (1 yr) (modest short-lived benefit at 6 mo)

Abbreviations: AMD, age-related macular degeneration; AMDRT, age-related macular degeneration radiotherapy trial.


obscuring!50% of the lesion, visual acuity (VA)O20/320, and no contraindication to EBRT (i.e., priorchemotherapy, diabetes, or history of periorbital orocular radiation). Randomization was to either EBRT(five daily sessions of 4 Gy for a total dose of 20 Gy) orobservation. The primary outcome measure was athree-line or greater loss of visual acuity over thefive-year follow-up period. There was also a recurrentCNV study arm with similar criteria (84,85). Eighty-

eight patients were enrolled through 10 sites and wererandomized to either radiotherapy [20 Gy delivered infive daily fractions of 4 Gy each; 6 mV (NZ41)] or noradiotherapy (sham NZ22 or observation NZ25). Theresults were reported in 2004 and concluded thatexternal beam radiation at 5!4 Gy may have amodest and short-lived (six-month) benefit in preser-ving visual acuity. There were no safety concerns(84,85).

(A)

(B)

Figure 10 (A) Immonen’s round strontium-90 plaque applicator. (B) Freire’s strontium-90 plaqueapplicator. Source: From Ref. 92.


Multimodality Treatment and Novel Methodsfor Radiation DeliveryMarcus and colleagues from the Medical College ofGeorgia submitted an ARVO abstract in 2002, updatedin 2004, on the use of transpupillary thermotherapy(TTT) and radiotherapy of CNV in AMD (87,88). Theinitial report was a safety evaluation in which foureyes of four patients were treated with TTT at 810 nmfor 60 seconds at a power of 360 to 1000 mW, followedwithin eight hours by administration of 6 mV photonbeam to deliver 20 Gy in five fractions at 4 Gy perfraction over five days (87). They found no safety riskand proceeded with a prospective non-randomizedcase series including eight patients following the sameprotocol (88). They found mixed results with thetreatment of occult subfoveal CNV, but again, therewere no safety concerns (88). This study has, however,been halted due to little subject interest in undergoinga combination of two experimental therapies (personalcommunication).

In 2002, Tong and colleagues from the Universityof California at Davis reported on the use of stereo-tactic external beam radiation to treat eyes with AMD(89,90). This method allows radiation to be deliveredto a smaller, better-defined area than standard EBRT.Patients treated with varying doses of radiation werefollowed for 24 months (89). They concluded that themethod was safe at all studied dosages with stablevisual acuity until 12 to 18 months posttreatment, atwhich time the effect of the radiation appeared to cease(89). At doses of 28–32 Gy, vision tended to stabilizefor a longer period (at least 24 months) (89). The one-year data from this pilot study werepublished in 2005(90). The investigators found no significant acute sideeffects and no benefit in either VA or membrane sizefrom increasing the radiation dosage. They concludedthat their results were consistent with trends inpalliative benefit and that there was no evidence thattherapeutic effectiveness is dose-dependent. There-fore, they found no justification for potentiallydangerous escalations in radiation dosage for treat-ment of neovascular AMD (90).

Other novel methods of delivering radiation arebeing studied, including a spoon-shaped device madeby Theragenics. This is inserted through a conjunctivalincision and traverses to an episcleral submacularposition, where the radiation source is uncovered fora short (minutes) period of time (75,91). Hubbardreported results of preliminary work with the Thera-genics device (Therasight Ocular BrachytherapySystem) at ARVO 2005 and found the device to bewell tolerated by patients and readily positioned andinserted by clinicians. No adverse events werereported after short post-operative follow-up (75). Ina personal communication, the company has suppliedthe following device description and Figure 9: “The

TheraSightw Ocular Brachytherapy System (Thera-Sight System) is a radiation device that primarilyconsists of a sealed palladium–103 source on thedistal end of an insertion applicator. The devicedelivers high dose rate, low energy X rays of21–23 keV in a minimally invasive procedure wherethe source is inserted in the retrobulbar space behindthe eye. The energy is deposited locally to the targettissue, consisting of new choroidal blood vesselsintruding into the subretinal space. The radiationtherapy is intended to reduce neovascularization.”This device delivers about 14 Gy to the inner retina.

Other applicators include Immonen’s applicatorand Freire’s applicator, both for strontium-90 (92).Immonen’s applicator was calculated to deliver 15 Gyto the inner retina (Fig. 10A) and Freire’s applicatorallows a dose at 1.5 mm depth of 6 cGy/second, thusallowing the total dose delivered to be altered based onexposure time (Fig. 10B). Finger has pointed out thatlow-energy photo-emitting palladium-103 will depositless radiation to the subjacent sclera, choroids, andretina than the beta-particle emitting 90Sr, possiblyexplaining why Immonen et al. noted increased andearlier chorioretinal atrophy within the targeted zone(Fig. 11A,B) (92).

(A)

(B)

Figure 11 (A) Immonen pre-treatment fluorescein angiography

(FA) utilizing strontium-90. (B) Immonen posttreatment FA.


Baseline

20/250

28 let.

3 Months

20/200

36 let.

6 Months

20/200

35 let.

15 Gy Dose 7-002(A)

Baseline

20/250

3 Months

20/200

6 Months

20/200

15 Gy Dose 7-002(B) (C)

Figure 12 (A) Subretinal radiation dose of 15 Gy, FA at baseline, three and six months demonstrating

development of inactivity and subretinal fibrosis without signs of radiation toxicity, respectively. (B) Sameeye as (A) OCT at same time periods showing subretinal fibrosis and inactivity of lesion. Abbreviations:

OCT, optical coherence tomography; FA, fluorescein angiography.


Fujii et al. evaluated the feasibility and initialsafety of retinal-sparing subretinal delivery of stron-tium beta-radiation using a novel-selective subretinalbrachytherapy system (Neovista, Atlanta, Georgia) on90 rabbit and 4 dog eyes (92). The surgery involvedvitrectomy, creation of a subretinal bleb, and introduc-tion of the probe, which was calculated to deliver aradiation dose of 0–246 Gy into the subretinal space(92). Lim and co-workers from Los Angeles presentedfurther work with the Neovista device in 10 patients atthe 2005 ARVO meeting (91). This was a tolerabilityand safety study that compared two probe designsdelivering 26 Gy to the CNV over a period of two tothree minutes, reportedly sparing the overlying retina.They found no retinal detachments or endophthal-mitis complications; however, there were threeadverse events that led to further device modifications(91). Figure 12 was supplied by the Neovista companydemonstrating the results in one of their initial trialeyes. The group has since switched to an epiretinalradiation delivery device.

CONCLUSIONS

Several well-organized, multi-center clinical trialsconducted in the United States and Europe haveshown no benefit to EBRT (64,78–84). Most of thesestudied doses or dose rates less than those used inbrachytherapy or proton irradiation studies. Three ofthese studies did show evidence of some benefit inlimiting lesion size and vision loss, mainly withradiation dosages of 20 Gy and higher (Table 1)(79,80,83).

In addition, it is possible that radiation treatmentmight be of benefit when combined with PDT or anti-angiogenic drugs, or with TTTas described by Marcusand colleagues (87,88). Combined treatment wouldpotentially allow complete closure of the neovascularcomplex, with PDT or injection of an anti-angiogenicagent followed by radiation to extend the effects oftreatment. Similarly, groups studying low-dose protonbeam radiation combined with PDT hope that thiswill limit CNV recurrence. This approach mightavoid the complications seen with higher doses ofradiation using the proton beam, or of multiple laseror pharmacologic treatments. Finally, other groups areevaluating different ways to deliver radiation in orderto limit toxicity and allow higher radiation doses(75,91).

This review has found significant evidencethat radiation can halt the growth of choroidal neovas-cularization. However, the prospective randomizedevidence-based studies reported to date do notsupport the widespread treatment of patients.Further prospective randomized studies are needed

to actually determine whether a different method ofdelivering the radiation will offer longer-term benefitwith less chance of toxicity, or whether a moreefficacious method might involve combining radiationwith another treatment modality.

SUMMARY POINTS

& Proliferating endothelium is more susceptible toradiation damage than are non-proliferating capil-lary endothelial cells and larger vessels, and thusneovascular endothelial cells and inflammatorycells are particularly radiosensitive.

& Radiation may induce programmed cell deathand modify the growth factor profiles ofthe neovascular complexes as well as limit theinflammatory response.

& Factors influencing the development of radiationretinopathy include total dose delivered, dailyfraction size, preexisting microangiopathy, anddiabetes or prior chemotherapy.

& While radiation-induced retinopathy has beenreported at doses of 30–35 Gy, it is more commonlyassociated with doses of 45–60 Gy (Fig. 5). Radi-ation optic neuropathy is rare at doses below50 Gy.

REFERENCES

1. Ferris FL, Fine SL, Hyman L. Age-related macular degener-ation and blindness due to neovascular maculopathy. ArchOphthalmol 1984; 102:1640–2.

2. American Academy of Ophthalmology. Age-relatedmacular degeneration, Preferred Practice Pattern. San Fran-cisco: American Academy of Ophthalmology, 1998.

3. Leibowitz HM, Krueger DE, Maunder LR, et al. TheFramingham Eye Study monograph: an ophthalmologicaland epidemiological study of cataract, glaucoma, diabeticretinopathy, macular degeneration, and visual acuity in ageneral population of 2631 adults, 1973–1975. SurvOphthalmol 1980; 24(Suppl.):335–610.

4. Kahn HA, Leibowitz HM, Ganley JP, et al. The Framinghameye study II. Association of ophthalmic pathology withsingle variables previously measured in the Framinghamheart study. Am J Epidemiol 1977; 106:33–41.

5. Klein BE, Klein R. Cataracts and macular degeneration inolder Americans. Arch Ophthalmol 1982; 100:571–3.

6. Vinding T. Age related macular degeneration, macularchanges, prevalence and sex ratio. An epidemiologicalstudy of 1000 aged individuals. Acta Ophthalmol 1989;67:609–16.

7. Macular Photocoagulation Study Group. Argon laserphotocoagulation for AMD: results of a randomized clinicaltrial. Arch Ophthalmol 1982; 100:912–8.

8. Macular Photocoagulation Study Group. Krypton laserphotocoagulation for neovascular lesions of AMD: resultsof a randomized clinical trial. Arch Ophthalmol 1990;108:816–24.


9. Macular Photocoagulation Study Group. Subfoveal neovas-cular lesions in AMD: guidelines for evaluation andtreatment in the Macular Photocoagulation Study. ArchOphthalmol 1991; 109:1242–57.

10. Macular Photocoagulation Study Group. Recurrent CNVafter argon laser treatment for neovascular maculopathy.Arch Ophthalmol 1986; 104:503–12.

11. Macular Photocoagulation Study Group. Occult choroidalneovascularization. Influence on visual outcome in patientswith age-related macular degeneration. Arch Ophthalmol1996; 114:400–12.

12. Macular Photocoagulation Study Group. Laser photocoa-gulation of subfoveal neovascular lesions in AMD: resultsof a randomized clinical trial. Arch Ophthalmol 1991;109:1220–31.

13. Macular Photocoagulation Study Group. Visual outcomeafter laser photocoagulation for subfoveal CNV secondaryto AMD: the influence of initial lesion size and initial visualacuity. Arch Ophthalmol 1994; 112:480–8.

14. Freund KB, Yanuzzi LA, Sorenson JA. Age-related maculardegeneration and choroidal neovascularization. AmJ Ophthalmol 1993; 115:786–91.

15. Algvere PV, Berglin L, Gouras P, Sheng Y. Transplantation offetal retinal pigment epithelium in age-related maculardegeneration with subfoveal neovascularization. GraefesArch Clin Exp Ophthalmol 1994; 232:707–16.

16. Rezai KA, Kohen L, Wiedermann P, Heimann K. Irispigment epithelium transplantation. Graefes Arch ClinExp Ophthalmol 1997; 235:558–62.

17. Thomas MA, Grand MG, Williams DF, Lowe MA. Surgicalmanagement of subfoveal choroidal neovascularization.Ophthalmology 1992; 99:952–68.

18. Scheider A, Gundisch O, Kampik A. Surgical extraction ofsubfoveal choroidal new vessels and submacular haemor-rhage in age-related macular degeneration; results of aprospective study. Graefes Arch Clin Exp Ophthalmol1999; 237:10–5.

19. Machemer R, Steinhorst UH. Retinal separation, reti-notomy, and macular relocation: II. A surgical approachfor age-related macular degeneration? Graefes Arch ClinExp Ophthalmol 1993; 231:635–41.

20. Ninomiya Y, Lewis JM, Hasegawa T, Tana Y. Retinotomyand foveal translocation for surgical management of subfo-veal choroidal neovascular membranes. Am J Ophthalmol1996; 122:613–21.

21. Eckardt C, Eckardt U, Conrad HG. Macular rotation withand without counter-rotation of the globe in patients withage-related macular degeneration. Graefes Arch Clin ExpOphthalmol 1999; 237:313–25.

22. Lewis H, Kaiser PK, Lewis S, Estafanous M. Maculartranslocation for subfoveal choroidal neovascularizationin age-related macular degeneration: a prospective study.Am J Ophthalmol 1999; 128:135–46.

23. Husain D, Miller JW, Michaud N, Connolly E, Flotte TJ,Gragoudas ES. Intravenous infusion of liposomalbenzoporphyrin derivative for photodynamic therapyof experimental choroidal neovascularization. ArchOphthalmol 1996; 114:978–85.

24. Kliman GH, Puliafito CA, Stern D, Borirakchanyavat S,Gregory WA. Phthalocyanine photodynamic therapy: newstrategy for closure of choroidal neovascularization. LasersSurg Med 1994; 15:2–10.

25. Lin SC, Lin CP, Feld JR, Duker JS, Puliafito CA. Thephotodynamic occlusion of choroidal vessels using benzo-porphyrin derivative. Curr Eye Res 1994; 13:513–22.

26. Miller JW, Walsh AW, Kramer M, et al. GragoudasESPhotodynamic therapy of experimental choroidalneovascularization using lipoprotein-delivered benzopor-phyrin. Arch Ophthalmol 1995; 113:810–8.

27. Peyman GA, Moshfeghi DM, Moshfeghi A, et al. Photo-dynamic therapy for choriocapillaris using tin ethyletiopurpurin (SnET2). Ophthalmic Surg Lasers 1997;28:409–17.

28. Schmidt-Erfurth U, Miller J, Sickenberg M, et al. Photody-namic therapy of subfoveal choroidal neovascularization:clinical and angiographic examples. Graefes Arch Clin ExpOphthalmol 1998; 236:365–74.

29. Treatment of Age-Related Macular Degeneration withPhotodynamic Therapy (TAP) Study Group. Photodynamictherapy of subfoveal choroidal neovascularization in age-related macular degeneration with verteporfin: one-yearresults of 2 randomized clinical trials—TAP report 1. ArchOphthalmol 1999; 117:1329–45.

30. Verteporfin in Photodynamic Therapy (VIP) Study Group.Verteporfin therapy of subfoveal choroidal neovasculariza-tion in AMD 2 year results of a randomized clinical trialincluding lesions with occult with no classic choroidalneovascularization-VIP Report No. 2. Am J Ophthalmol2001; 131:541–60.

31. Verteporfin in Photodynamic Therapy (VIP) Study Group.Verteporfin therapy of subfoveal choroidal neovasculariza-tion in pathologic myopia, 2 year results of a randomizedclinical trial—VIP Report No. 3. Ophthalmology 2003;110:667–73.

32. Gragoudas ES, Adamis AP, Cunningham ET, et al. Pegap-tanib for neovascularAMD.NEngl JMed 2004; 351:2805–16.

33. Mainster MA, Reichel E. Transpupillary thermotherapy forage-related macular degeneration: long-pulse photocoagu-lation, apoptosis and heat-shock proteins. Ophthalmic SurgLasers 2000; 31:359–73.

34. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab forneovascular age-related macular degeneration. N EnglJ Med 2006; 355:1419–31.


36. Gillies MC, Simpson JM, Luo W, et al. A randomizedclinical trial of a single dose of intravitreal triamcinoloneacetonide for neovascular age-related macular degener-ation (one-year results). Arch Ophthalmol 2003; 121:667–73.

37. Amato RJ, LoughanMS, Flynn E, Folkman J. Thalidomide isan inhibitor of angiogenesis. Proc Natl Acad Sci USA 1994;91:4082–5.

38. Pharmacological Therapy for Macular Degeneration StudyGroup. Interferon-2a is ineffective for patients withchoroidal neovascularization secondary to age-relatedmacular degeneration. Results of a prospective random-ized placebo-controlled clinical trial. Arch Ophthalmol1997; 115:865–72.

39. Thomas MA, Ibanez HE. Interferon alfa-2a in the treat-ment of subfoveal choroidal neovascularization. AmJ Ophthalmol 1993; 115:563–8.

40. Ciulla TA, Danis RP, Harris A. Age-related maculardegeneration: review of experimental treatments. SurvOphthalmol 1998; 32:134–46.

41. Finger PT, Augsberger JJ. Controversies, radiotherapy andthe treatment of age-related macular degeneration. ArchOphthalmol 1998; 116:1507–11.

42. Baker DG, Krochak RJ. The response of the microvascularsystem to radiation: a review. Cancer Invest 1989; 7:287–94.


43. Krishnan L, Krishnan EC, Jewell WR. Immediate effect ofirradiation on microvasculature. Int J Radiat Oncol BiolPhys 1988; 15:147–50.

44. Finger PT, Immonen I, Freire J, Brown G. Brachytherapy formacular degeneration associated with subretinal neovascu-larization. In: Alberti WE, Richard G, Sagerman RH, eds.Age-Related Macular Degeneration. Germany: Springer,2001:167–72.

45. Chakravarthy U, Gardiner TA, Archer DB, Maguire CJF. Alight microscopic and autoradiographic study of non-irradiated and irradiated ocular wounds. Curr Eye Res1989; 8:337–48.

46. Archambeau JO, Mao XW, Yonemoto LT, et al. What is therole of radiation in the treatment of subfoveal membranes:review of radiobiologic, pathologic, and other consider-ations to initiate a multimodality discussion. Int J RadiatOncol Biol Phys 1998; 40:1125–36.

47. De Gowin RL, Lewis JL, Hoak JC, Mueller AL, Gibson DP.Radiosensitivity of human endothelial cells in culture.J Lab Clin Med 1974; 84:42–8.

48. Hosoi Y, Yamamoto M, Ono T, Sakamoto K. Prostacyclinproduction in cultured endothelial cells is highly sensitiveto low doses of ionizing radiation. Int J Radiat Biol 1993;63:631–8.

49. Johnson LK, Longenecker JP, Fajardo LF. Differentialradia-tion response of cultured endothelial cells and smoothmyocytes. Anal Quant Cytol 1982; 4:188–98.

50. Finger PT. Radiation therapy for choroidal melanoma. SurvOphthalmol 1997; 42:215–32.

51. Sagerman RH, Chung CT, Alberti WE. Radiosensitivity ofocular and orbital structures. In: Alberti WE, Sagerman RH,eds. Radiotherapy of Intraocular and Orbital Tumors.Berlin: Springer, 1993:375–85.

52. Langley RE, Bune EA, Quartuccio SG, Medeiras D,Braunhut SJ. Radiation induced apoptosis in microvascularendothelial cells. Br J Cancer 1997; 75:666–72.

53. Perez CA, Brady LW. Principles and Practice of RadiationOncology. 2nd ed. Philadelphia, PA: JB Lippincott Co., 1992.

54. Schilling H, Sauerwein W, Lommatzsch A, et al. Long-termresults after low dose ocular irradiation for choroidalhaemamgiomas. Br J Ophthalmol 1997; 81:267–73.

55. Engenhart R, Wowra B, Debus J, Kimmig BN, et al. The roleof high-dose, single-fraction irradiation in small and largeintracranial arteriovenous malformations. Int J RadiatOncol Biol Phys 1994; 30:521–9.

56. Finger PT, Chakravarthy Y. External beam radiation therapyis effective in the treatment of age-related macular degener-ation. Arch Ophthalmol 1998; 116:1507–9.

57. Scott TA, Augsburger JJ, Brady LW, Hernandez C,Woodleigh R. Low dose ocular irradiation for diffusechoroidal hemangiomas associated with bullous nonrheg-matogenous retinal detachment. Retina 1991; 11:389–93.

58. BrownGC, Shields JA, Sanborn G, Augsburger JJ, Savino PJ,Schatz NJ. Radiation retinopathy. Ophthalmology 1982;89(12):1494–501.

59. Chan RC, Shukovsky LJ. Effects of irradiation on the eye.Radiology 1976; 120:673–5.

60. Parsons JT, Fitzgerald CR, Hood CI, Ellingwood KE,Bova FJ, Million RR. The effects of irradiation on theeye and optic nerve. Int J Radiat Oncol Biol Phys 1983;9:609–22.

61. Archer DB, Amoaku SMK, Gardinier TA. Radiation retino-pathy, clinical, histological and ultrastructural correlations.Eye 1991; 5:239–51.

62. Plowman PN, Harnett AN. Radiotherapy in benign orbitaldisease. I: complicated ocular angiomas. Br J Ophthalmol1988; 72:286–8.

63. Chakravarthy U, Houston RF, Acher DB. Treatment of age-related subfoveal neovascular membranes by teletherapy: apilot study. Br J Ophthalmol 1993; 77:265–73.

64. Hart P, Chakravarthy U, Mackenzie G, et al. Visualoutcomes in the subfoveal radiotherapy study a random-ized controlled trial of teletherapy for age-related maculardegeneration. Arch Ophthalmol 2002; 120:1029–38.

65. Finger PT, Berson A, Sherr D, Riley R, Balkin RA,Bosworth JL. Radiation therapy for subretinal neovascular-ization. Ophthalmology 1996; 103:878–89.

66. Stalmans P, Leys A, Van Limbergen E. External beamradiation therapy (20 Gy, 2 Gy fractions) fails to controlthe growth of choroidal neovascularization in age-relatedmacular degeneration: a review of 111 cases. Retina 1997;17:481–92.

67. Spaide RF, Guyer DR, McCormick B, et al. External beamradiation therapy for choroidal neovascularization.Ophthalmology 1998; 105:24–30.

68. Donati G, Soubrane D, Quaranta M, Coscas G, Soubrane G.Radiotherapy for isolated occult subfoveal neovascularisa-tion in age-related macular degeneration: a pilot study. BrJ Ophthalmol 1999; 83:646–51.

69. Mauget-Faysse M, Chiquet C, Milea D, et al. Long termresults of radiotherapy for subfoveal choroidal neovascu-larisation in age related macular degeneration. BrJ Ophthalmol 1999; 83:923–8.

70. Chakravarthy U. External beam radiotherapy in exudativeage-related macular degeneration: a pooled analysis ofphase-1 data. Br J Radiol 2000; 73:305–13.

71. Bergink GJ, Deutman AF, van den Broek JFCM, vanDaal WA, van der Maazen RW. Radiation therapy for age-related subfoveal choroidal neovascular membranes, a pilotstudy. Doc Ophthalmol 1995; 90:67–74.

72. Bergink GJ, Hoyng CB, van der Maazen RWM. Arandomized controlled trial on efficacy of radiationtherapy in the control of subfoveal choroidal neovascu-larization in age-related macular degeneration: radiationversus observation. Graefes Arch Clin Exp Ophthalmol1998; 236:321–5.

73. Marcus DM, Sheils WC, Young JO, et al. Radiotherapy forrecurrent choroidal neovascularization complicating age-related macular degeneration. Br J Ophthalmol 2004;88:114–9.

74. Finger PT, Berson A, Ng T, Szechter A. Ophthalmic plaqueradiotherapy for age-related macular degeneration associ-ated with subretinal neovascularization. Am J Ophthalmol1999; 127:170–7.

75. Hubbard G, Ciulla T, Marcus D, et al. A new ocularbrachytherapy system for the treatment of exudativeAMD. Invest Ophthalmol Vis Sci 2005; 46 (E-abstract 2425).

76. Finger PT, Gelman YP, Berson AM, Szechter A. Pallidium-103 plaque radiation therapy for macular degeneration:results of a 7 year study. Br J Ophthalmol 2003; 87:1497–503.

77. Flaxel CJ, Friedrichsen EJ, Smith JO, et al. Proton beamirradiation of subfoveal choroidal neovascularization inage-related macular degeneration. Eye 2000; 14:155–64.

78. The Radiation Therapy for Age-related Macular Degener-ation (RAD) Study Group. A prospective, randomized,double-masked trial on radiation therapy for neovascularage-related macular degeneration (RAD Study). Ophthal-mology 1999; 106:2239–47.

79. Kobayashi H, Kobayashi K. Age-related macular degener-ation: long-term results of radiotherapy for subfovealneovascular membranes. Am J Ophthalmol 2000;130:617–35.


80. Valmaggia C, Reis G, Ballinari P. Radiotherapy for subfo-veal choroidal neovascularization in age-related maculardegeneration: a randomized clinical trial. Am J Ophthalmol2002; 133:521–9.

81. Marcus DM, Sheils C, Johnson MH, et al. External beamirradiation of subfoveal choroidal neovascularizationcomplicating age-related macular degeneration. One-yearresults of a prospective, double masked randomizedclinical trial. Arch Ophthalmol 2001; 119:171–80.

82. Lott M, Marcus D, Sheils W, Johnson M, Samy C. Externalbeam irradiation of subfoveal CNV complicating AMD: 4year results of a prospective, double masked, randomizedclinical trial. Invest Ophthalmol Vis Sci 2003; 44 (E-abstract5007).

83. Churei H, Ohkubo K, Nakajo M, et al. External-beamradiation therapy for age-related macular degeneration:two years’ follow-up results at a total dose of 20 Gy in 10fractions. Radiat Med 2004; 22:398–404.

84. Marcus D, Peskin E, Alexander J, et al. The age-relatedmacular degeneration radiotherapy trial (AMDRT): 1—yearresults. Invest Ophthalmol Vis Sci 2003; 44 (E-abstract 3158).

85. Marcus D, Peskin E, AMDRT Study Group. The age-relatedmacular degeneration radiotherapy trial (AMDRT): one yearresults froma pilot study. Am JOphthalmol 2004; 138:818–28.

86. Tamai M. The results of randomized controlled trial (RCT)of low-dose radiation for wet-AMD on 1 year term basis.Invest Ophthalmol Vis Sci 2003; 44 (E-abstract 3157).

87. Lee S, Sheils W, Redd J, Samy C, Marcus D. Multimodalitytranspupillary thermotherapy and radiotherapy of chor-oidal neovascular membranes in age-related maculardegeneration: a phase I safety study. Invest OphthalmolVis Sci 2002; 43 (E-abstract 4411).

88. Ying M, Fuller J, Alexander J, Sheils W, Lee Y, Marcus D.Multimodality transpupillary thermotherapy and radio-therapy of occult subfoveal choroidal neovascularmembranes in AMD. Invest Ophthalmol Vis Sci 2004; 45(E-abstract 5136).

89. Tong A, Hauser D, Barak A, et al. Stereotactic external beamradiation treatment in eyes with AMD CNV—a 2-yearfollow-up. Invest Ophthalmol Vis Sci 2002; 43 (E-abstract1221).

90. Barak A, Hauser D, Yipp P, et al. A phase I trial ofstereotactic external beam radiation for subfoveal choroidalneovascular membranes in age-related macular degener-ation. Br J Radiol 2005; 78:827–31.

91. Lim JI, DeJuan E, Sadda V, et al. Subretinal radiationtreatment of occult CNV due to AMD. Invest OphthalmolVis Sci 2005; 46 (E-abstract 1384).

92. Finger PT, Immonen I, Freire J, Brown G. Implant radio-therapy for exudative macular degeneration associatedwith subretinal neovascularization. In: Peyman G,Meffert S, Conway MD, Chou F, eds. Vitreoretinal SurgicalTechniques. London: Martin Dunitz, 2001:555–60.


17

Anti-VEGF Drugs and Clinical TrialsTodd R. KlesertDoheny Eye Institute, University of Southern California, Los Angeles, California, U.S.A.



Phillip J. RosenfeldBascom Palmer Eye Institute, Miami, Florida, U.S.A.

INTRODUCTION

In 1989, Ferrara and Henzel (1) isolated a diffusibleprotein from bovine pituitary follicular cells thatshowed cell-specific mitogenic activity for vascularendothelium. They named this protein vascular endo-thelial growth factor (VEGF). Further research showedthat VEGF was in fact Michelson’s factor X, whichwas the postulated diffusible angiogenesis factor (2).As discussed in Chapter 5, VEGF was then shown tohave a major role in choroidal neovascularization(CNV) (3,4).

The human VEGF-A gene, located on chromo-some 6p21.3, consists of eight exons and seven introns.Alternative splicing produces mRNA transcripts thatcode for at least six different protein isoforms: 121, 145,165, 183, 189, and 206 amino acids in length (5). Thesedifferent isoforms vary in their affinity for heparinbinding, and as such, in their affinity for the extra-cellular matrix. The larger isoforms, such as VEGF189and VEGF206, bind heparin with high affinity, and aretherefore almost completely sequestered in the extra-cellular matrix. The smaller isoform, VEGF121, doesnot bind heparin and is freely diffusible. All VEGFisoforms contain a plasmin cleavage site. Cleavage atthis site creates a freely diffusible, 110 kD, bioactiveform of VEGF (VEGF110). Plasmin-mediated extra-cellular proteolysis may therefore be an importantregulator of VEGF bioavailablility (6).

CURRENT ANTI-VEGF THERAPIES

Aptamers: Pegaptanib Sodium(Macugen, New York)The first anti-VEGF therapy to undergo clinical testingwas a VEGF aptamer. Approved by the Food andDrug Administration (FDA) in 2004, Pegaptanib(Macugen—OSI/Eyetech Pharmaceuticals, NewYork) was the first anti-VEGF agent with provenefficacy for the treatment of CNV secondary to

age-related macular degeneration (AMD). Pegaptanibis an aptamer—a short single-stranded oligonucleo-tide sequence that functions as a high affinity inhibitorof a specific protein target. Aptamers are created by aform of in vitro evolution called systematic evolutionof ligands by exponential enrichment (SELEX) (7).

Pegaptanib is a 28-base RNA oligonucleotide thatis covalently linked to two 20 kD polyethylene glycolmoieties to extend the half-life. Pegaptanib selectivelybinds to the heparin-binding domain of VEGF165 andlarger isoforms, preventing ligand-receptor binding.The smaller VEGF isoforms and proteolytic fragmentsare therefore not inhibited by pegaptanib (7).

Safety and efficacy of pegaptanib for the treat-ment of neovascular AMDwas established through theVEGF Inhibition Study in Ocular Neovascularization(VISION) study (8). VISION consisted of two phase IIIprospective, multicenter, randomized, controlled,double-masked trials comparing intravitreal injectionsof pegaptanib with sham injections. Patients (1186total) were randomized to receive pegaptanib (at adose of 0.3, 1.0, or 3.0 mg) or sham injection (usualcare), every six weeks for a total of 54 weeks. Theprimary end point of the study was the number ofpatients losing less than 15 letters of Early TreatmentDiabetic Retinopathy Study (ETDRS) visual acuity at54 weeks. Patients with all CNV lesion subtypes withsizes up to and including 12 disc areas in size wereincluded. Concomitant photodynamic therapy (PDT)with verteporfin (Visudynew, Novartis, East Hanover,New Jersey, U.S.A.) was allowed at the physician’sdiscretion. Twenty-five percent of the VISION patientsreceived PDT during the study period.

In the pooled analysis, efficacywas demonstratedfor all three doses, without a dose–response relation-ship. Seventy percent of pegaptanib-treated patientslost less than 15 letters, compared with 55% of usualcare patients. More pegaptanib-treated patientsmaintained or gained visual acuity (33%) at 54 weeksthan usual care patients (23%). In addition, the usual

care group was twice as likely to experience severevision loss (R30 letters) during the study period thanpegaptanib-treated patients. However, only 6% ofpegaptanib-treated patients in the study gained R15letters at 54 weeks (compared with 2% of usual carecontrols), and as a group, the pegaptanib-treatedpatients lost an average of eight letters over the studyperiod (compared with 15 letters in the usual caregroup). Adverse ocular events in the VISION trialresulted in severe vision loss in 0.1% of patients.These adverse events included endophthalmitis(1.3%), traumatic lens injury (0.6%), and retinal detach-ment (0.6%).

For year 2 of the VISION study, patients werere-randomized to the treatment and usual care arms(9). The results indicated that those patients continuingwith pegaptanib treatment for a second year did betterthan those reassigned to the usual care control arm at54 weeks, and better than those assigned to the usualcare arm for the entire two years. The percentage ofpegaptanib-treated patients who progressed tomoderate visual loss (from baseline) during thesecond year of treatment was half (7%) that of thosereassigned to the control group at 54 weeks (14%), andthose who continued in the control group for thesecond year (14%). Of note, however, patients whohad benefited from their year 1 treatment assignment(defined as %0 letters of vision loss from baseline),and who subsequently lost R10 letters of vision afterre-randomization at 54 weeks, were allowed to receive“salvage therapy” (a reassignment back to theiroriginal year 1 treatment arm). Year 2 safety datacontinue to show that pegaptanib is a relatively safedrug. Non-ocular hemorrhagic events were not signi-ficantly different from the usual care group (10).

Studies with pegaptanib continue. The Verte-porfin Intravitreal Triamcinolone Acetonide Study(VERITAS) is a phase III prospective, multicenter,randomized, double-masked trial comparing PDTcombined with one of two doses of intravitrealtriamcinolone (1 mg, 4 mg) versus PDT combinedwith 0.3 mg of intravitreal pegaptanib. Approxi-mately 100 patients have been enrolled, includingall CNV lesion subtypes (11).

Studies are also ongoing at OSI/Eyetech to createa sustained-release form of pegaptanib, with the goalof reducing the frequency of intravitreal injectionsrequired for treatment, and thereby reducing therisk of serious adverse events associated withintravitreal injections, such as endophthalmitis andretinal detachment. Preliminary animal work withpoly(lactic-co-glycolic) acid (PLGA)-based micro-sphere encapsulation suggests that sustained-releaseof pegaptanib for greater than six months is possiblewith a single intravitreal injection (12).

Monoclonal Antibodies: Ranibizumab (Lucentis)In June 2006, ranibizumab (Lucentis—Genentech,South San Francisco, California, U.S.A.) became thesecond VEGF inhibitor approved by the FDA foruse in the treatment of CNV secondary to AMD.Ranibizumab is a humanized, affinity-maturated Fabfragment of a murine monoclonal antibody directedagainst human VEGF-A. Ranibizumab is a potent,non-selective inhibitor of all VEGF-A isoforms andbioactive proteolytic products. Ranibizumab wasspecifically designed as a molecule smaller than itsparent full-size precursor anti-VEGF antibody,because it was felt that the full-sized antibody wasunable to cross the inner retina and choroid, assuggested by a histologic study of the Herceptinantibody (13). More recent histologic analysis of beva-cizumab in rabbits by Sharar et al. (14) however,suggests that the full-length antibody actually doespenetrate all layers of the retina quite effectively.Because ranibizumab is missing the Fc region, it isalso felt that the molecule will be less likely to incitean immune response, as it can no longer bind tocomplement C1q or Fc gamma receptors (15).

Efficacy and safety of ranibizumab has thusfar been established through two large prospective,multicenter, randomized, double-masked, controlledclinical trials: Minimally Classic/Occult Trial of Anti-VEGF Antibody Ranibizumab in the Treatment ofNeovascular Age-Related Macular Degeneration(MARINA) (16) and Anti-VEGF Antibody for theTreatment of Predominantly Classic CNV in AMD(ANCHOR) (17). The MARINA trial was limited topatients with subfoveal occult or minimally classicCNV, either primary or recurrent, with evidence ofrecent disease progression. In MARINA, 716 patientswere randomized 1:1:1 to receive monthly intravitrealinjections of ranibizumab (either 0.3 or 0.5 mg) orsham injections. The primary outcome measure wasthe proportion of patients losing less than 15 ETDRSletters at 12 months. 94.5% of patients assigned to the0.3 mg group and 94.6% of patients assigned to the0.5 mg ranibizumab treatment arms, compared with62.2% in the sham-treatment arm, met this endpoint.More eyes gained 15 or more letters of visual acuityby month 12 in the ranibizumab treatment arms thanthe control arms: 24.8% in the 0.3 mg group, 33.8%in the 0.5 mg group, 5.0% in the sham-treated group.Mean visual acuity increased by 6.5 letters in the0.3 mg group and 7.2 letters in the 0.5 mg group at12 months. In contrast, mean visual acuity dropped by10.4 letters in the sham-treated group. In general,vision gains were maintained throughout year two ofthe MARINA trial in ranibizumab-treated patients,whereas vision continued to decline in the sham-treated patients; mean loss was 14.9 letters in thesham group.

248 KLESERT ET AL.

There was also a difference in the lesion sizeoutcomes between the ranibizumab and controlgroups. While lesion size on average remainedstable in the ranibizumab-treated patients, lesionsize increased by about 50% in the sham-treatedpatients at 12 months. The area of leakage in theranibizumab-treated lesions decreased on average byapproximately 50%.

Adverse ocular events in ranibizumab-treatedpatients in the MARINA trial over 24 months inclu-ded presumed endophthalmitis in 1.0% of patientsand serious uveitis in 1.3% of patients. No retinaldetachments were observed in the ranibizumab-treated patients, although retinal tears were identifiedin two patients (0.4%). Lens damage as a result ofintravitreal injection was seen in one patient (0.2%).No statistically significant difference in serioussystemic adverse events was observed between thetreatment and control arms of the study, althoughthere was a trend toward the increased rate ofserious (1.3% in 0.3 mg group; 2.1% in 0.5 mg group;0.8% in sham group) and non-serious (9.2% in 0.3 mggroup; 8.8% in 0.5 mg group; 5.5% in sham group)non-ocular hemorrhages.

The ANCHOR trial—has likewise demonstratedefficacy of ranibizumab for the treatment of predomi-nantly classic CNV lesions secondary to AMD.ANCHOR was designed as a head-to-head compari-son between ranibizumab and PDT with verteporfin(Visudyne), which was then the standard of care forsubfoveal CNV. 423 patients were randomized 1:1:1 toreceive monthly intravitreal injections with ranibi-zumab 0.3 mg and sham PDT, ranibizumab 0.5 mgwith sham PDT or monthly sham injections plusactive verteporfin PDT. The primary end point wasthe number of patients losing fewer than 15 letters ofbaseline visual acuity at 12 months. This end point wasachieved in 94.3% of the patients receiving 0.3 mgranibizumab and 96.4% of patients receiving 0.5 mgranibizumab versus 64.3% of the verteporfin group.The percentage of patients experiencing an improve-ment over baseline visual acuity of at least 15 letterswas 35.7% and 40.3% respectively, in the ranibizumab-treated patients, versus only 5.6% in the verteporfin-treated patients. Mean visual acuity increased by8.5 letters in the 0.3 mg ranibizumab group and 11.3letters in the 0.5 mg ranibizumab group at 12 months.In contrast, mean visual acuity dropped by 9.5 lettersin the verteporfin PDT group at 12 months.

Measurement of CNV lesion size throughout theANCHOR study revealed positive morphologic effectssimilar to those observed in the MARINA study. Ingeneral, average total lesion size remained relativelystable in the ranibizumab-treated patients over oneyear, while increasing significantly in the verteporfin-treated patients. Moreover, average total area of

leakage and average total area of classic CNVleakage both decreased significantly at one year inthe ranibizumab-treated patients, while they increasedin the verteporfin-treated group.

No statistically significant difference in serioussystemic adverse events was observed between theranibizumab and verteporfin arms of the study, but asin the MARINA trial, there was a trend toward anincreased rate of serious (1.5% in 0.3 mg group; 2.1% in0.5 mg group; 0% in PDT group) and non-serious(5.1% in 0.3 mg group; 6.4% in 0.5 mg group; 2.1% inPDT group) non-ocular hemorrhages. Serious adverseocular events in the ranibizumab-treated ANCHORtrial patients over 12 months included presumedendophthalmitis in 0.7% of patients and significantuveitis in 0.4% of patients. One patient each developeda retinal detachment (0.4%) or vitreous hemorrhage(0.4%). There were no cases of lens damage as aresult of the intravitreal injection. The most commonadverse event (12% patients) was mild post-injectioninflammation.

The PIER study is a phase IIIb, prospective,multicenter, randomized, double-masked, controlledstudy of 184 patients with predominantly classic oroccult CNV randomized to receive ranibizumab orsham injections monthly for the first three months,followed by once every three months for a total of24 months. The purpose of PIER is to help determinethe optimal dosing schedule for ranibizumab. The one-year results of the PIER study showed that 83%(0.3 mg) and 90% (0.5 mg) of ranibizumab-treatedeyes lost less than 15 letters of visual acuity, comparedto 49% of sham eyes. However, the percentage of eyesimproving 15 or more letters was only 12% (0.3 mg)and 13% (0.5 mg) in ranibizumab-treated eyes,compared with 10% of sham eyes (18).

Prospective optical coherence tomography(PRONTO) imaging of patients with neovascularAMD treated with intraocular ranibizumab is atwo-year, single site, open-label, uncontrolled studyof 40 patients designed to evaluate the durability ofresponse of ranibizumab and whether optical coher-ence tomography (OCT) can be used to guidetreatment of neovascular AMD (19). As in the PIERstudy, patients receive monthly injections of ranibi-zumab for the first three months. Thereafter,re-treatment with ranibizumab is performed if one ofthe following changes were observed between visits:a loss of 5 letters in vision in conjunction with fluidon OCT, increase in OCT central retinal thickness ofat least 100 mm, new onset classic CNV, new macularhemorrhage, or persistent macular fluid detectedby OCT at least 1 month after the previous injectionof ranibizumab. At 12 months, mean visual acuityimproved by 9.3 letters (p!0.001) and the meanOCT central retinal thickness decreased by 178 mm

17: ANTI-VEGF DRUGS AND CLINICAL TRIALS 249

(p!0.001). Visual acuity improved 15 or more letters in35% of patients. These visual acuity andOCToutcomeswere achieved with an average of 5.6 injections over 12months. Once a fluid-free macula was achieved, themean injection-free interval was 4.5 months beforeanother reinjection was necessary. Unlike the PIERstudy, visual acuity gains did occur despite the lessfrequent dosing scheme. PRONTO outcomes suggestthat OCT can be useful for guiding re-treatment withintravitreal ranibizumab in neovascular AMD, andthat use of an OCT-guided variable-dosing regimencould decrease the injection burdenwithout sacrificingimprovements in visual acuity.

Monoclonal Antibodies: Bevacizumab (Avastin)Bevacizumab (Avastinw, Genentech, South San Fran-cisco, California, U.S.A.) is a full-length humanizedmurine monoclonal antibody directed against humanVEGF-A. It was FDA approved in 2004 for the intra-venous treatment of metastatic colorectal cancer.Its potential for use in the treatment of CNV wasfirst tested by Michels et al. (20) via intravenousinfusion in a 12-week open-label uncontrolled study.Striking effects were observed on both visual acuity,and the OCT and angiographic characteristics of theneovascular lesions. However, patients experienced amean increase of 12 mmHg in systolic blood pressure,which was felt to be a deterrent to its common use.

This systemic side effect, combined with thepromising visual and anatomic results from the intra-venous infusion of bevacizumab, led investigators toconsider intravitreal injection of bevacizumab (21).Since then, several retrospective, uncontrolled, open-label case series have been published regarding theuse of intravitreal bevacizumab for the treatment ofCNV secondary to AMD (22–25). As with ranibi-zumab, the effect of bevacizumab has been impressive.

Avery and colleagues (22) treated 79 patientswith 1.25 mg of intravitreal bevacizumab monthlyand reported the early results at three monthsfollow-up. Many of these patients had prior failedtreatment with verteporfin or pegaptanib. At threemonths, median Snellen visual acuity improved from20/200 at baseline to 20/80. Mean central retinalthickness by OCT decreased by 67 mm at 3 months.No ocular or systemic adverse events were observed.

Spaide and colleagues treated (23) 266 patientswith monthly 1.25 mg of intravitreal bevacizumab.By three months, Snellen visual acuity improvedfrom a mean of 20/184 at baseline to 20/109, with38.3% of patients experiencing some improvementin visual acuity. Mean central retinal thickness byOCT improved from 340 mm at baseline to 213 mm at3 months. Again, no adverse ocular or systemicadverse events were observed.

In contrast to the intravenous administrationof bevacizumab, intravitreal injection of bevaci-zumab did not result in the systemic side effect ofhypertension in any of these studies. The systemicconcentration of bevacizumab when given intrave-nously is obviously several times larger thanthe systemic concentrations seen after intravitrealinjections, and no elevation in blood pressure hasyet been reported in patients treated with intra-vitreal bevacizumab.

Animal and in vitro studies published thus farhave failed to identify any specific toxicity associatedwith bevacizumab use. Luthra et al. (26) demonstratedthat viability of human RPE cells, rat neurosensorycells and human microvascular endothelial cells inculture was normal after exposure to bevacizumab atconcentrations of up to 1 mg/mL. Rabbit studies byManzano et al. (27) found no changes in the electro-retinogram (ERG) patterns of eyes injected withintravitreal bevacizumab at doses up to 5.0 mg. Mildvitreous inflammation was seen at 5.0-mg dose, butnot at lower doses. Bakri et al. (28) looked at retinalhistology of rabbit eyes injected with bevacizumab,and again found no histologic changes compared withcontrol eyes.

One important aspect in which ranibizumab andbevacizumab may differ is their pharmacokinetics.Because of its larger molecular weight, it is assumedthat bevacizumab has a significantly longer half-lifein the vitreous, and possibly systemically as well.A longer half-life may allow for less frequent injectionsto achieve the same biologic effect. Recent unpub-lished data, however, indicate that the half-lives ofthe two drugs may actually be quite similar. Per thepackage insert for Lucentis, the half-life of ranibi-zumab in the vitreous is approximately 3 days basedon animal studies. Pharmacokinetic studies in rabbitsreveal that the half-life of bevacizumab in the vitreousis only marginally longer at 4.3 days (29).

Although the limited data thus far suggest thatbevacizumab is highly effective and safe for thetreatment of CNV secondary to AMD, without ahead-to-head prospective clinical trial, the relativeefficacy and safety of bevacizumab compared withranibizumab will remain unknown. Fortunately,The National Eye Institute has agreed to sponsor atrial comparing bevacizumab with ranibizumab inAMD patients with subfoveal CNV. This study,the Comparison of Treatment Trial (CATT) study,will randomize patients into one of four treatmentarms: monthly intravitreal injection of ranibizumab,monthly injection of bevacizumab, monthly injectionof ranibizumab followed by as-needed treatment,and monthly injection of bevacizumab followed byas-needed treatment. Until the results of the CATTstudy are available, bevacizumab is nonetheless

250 KLESERT ET AL.

an attractive treatment option due to its cost advantageover ranibizumab, especially for those patientswithout drug insurance coverage or with large drugco-payment requirements.

Because bevacizumab has been FDA approvedonly for the intravenous treatment of metastatic coloncancer, intravitreal injection of bevacizumab is anoff-label use of drug by an altered route of adminis-tration. This makes documentation of the informedconsent process especially important when using beva-cizumab. During informed consent, the physicianshould explain to patients that the safety and efficacyofbevacizumabhavenotbeenestablishedwith certainty,and that there may be unknown risks with its use. Abevacizumab-specific consent form is recommended,and can be found on the website of the OphthalmicMutual Insurance Company (OMIC) (30,31).

Bevacizumab comes in preservative-free 100 mgvials, containing 4 cc of a 25 mg/cc solution,intended for one-time use only for treatment of asingle cancer patient. A single vial can theoreticallybe aliquoted out to provide up to eighty individual0.05 cc intravitreal doses in 1 cc tuberculin syringes.The pharmacy should confirm the dose and sterility,provide proper storage instructions, and mark allaliquots with an expiration date. Although bevaci-zumab is a very stable drug with a shelf-life of manymonths, compounded aliquots will usually have anexpiration date due to sterility concerns.

Combination Therapy with PDTThe RhuFab V2 Ocular Treatment Combining the Useof VISUDYNEw to Evaluate Safety (FOCUS) study (32)is a two-year, phase I/II, multicenter, randomized,single-masked, controlled study of 162 patients withpredominantly classic CNV. FOCUS compared thesafety and efficacy of intravitreal ranibizumab(0.5 mg) combined with verteporfin PDT versus verte-porfin PDT alone (combined with sham injection).Patients received monthly ranibizumab (0.5 mg)(nZ106) or sham (nZ56) injections. The PDT wasperformed seven days before initial ranibizumab orsham treatment and then quarterly as needed. Theprimary outcome measure was the proportion ofpatients who lost fewer than 15 letters from baselineat 12 months. At 12 months, 90.5% of the ranibizumab-treated patients and 67.9% of the control patients lostfewer than 15 letters (p!0.001).

The most frequent ranibizumab-associatedserious ocular adverse events were intraocular inflam-mation (11.4%) and endophthalmitis (1.9%; 4.8%if including presumed cases). On average, patientswith serious inflammation had better visual acuityoutcomes at 12 months than did controls. Keyserious non-ocular adverse events included myo-cardial infarctions in the PDT-alone group (3.6%) and

cerebrovascular accidents in the ranibizumab-treatedgroup (3.8%). Notably, ranibizumab-treated patientsexperiencing intraocular inflammation still had bettervisual acuity outcomes at 12 months than the controlpatients. Thus, ranibizumab combined with PDTwas more efficacious than PDT alone for treatingneovascular AMD.

In addition, the FOCUS study showed thatdespite a history of prior PDT therapy, a significantproportion of these patients were able to gain visualacuity when treated with ranibizumab and PDT. Theneed for additional PDT was 27.6% for the combinedgroup but 91.1% for the PDT group. A difference in therate of PDT re-treatment was seen by the 3-monthfollow-up period and maintained for the study.

The FOCUS study however did not compare theranibizumab plus PDT combination to ranibizumabalone. The DENALI study is a randomized, controlled,multicenter clinical trial that will perform thecomparison study. The study will gauge the safety,efficacy and impact on re-treatment rates of Visudyne(verteporfin, Novartis) and Lucentis (ranibizumab,Genentech, South San Francisco, California, U.S.A.)as a combination therapy against wet AMD.DENALI is expected to enroll 300 wet AMD patientsat 45 centers in the United States and five centers inCanada. The two-year study will investigate whetherpatients receiving the combination therapy requirefewer re-treatments than control patients treatedwith Lucentis monotherapy.

Intravitreal Injection TechniqueIt appears that the greatest risks associated withthe use of current anti-VEGF therapies for the treat-ment of AMD (endophthalmitis, retinal detachment,lens trauma) come from the intravitreal injection itself.Therefore, proper injection technique and careful anti-septic practices are important.

Supplies that are recommended for prepping theeye include 5% povidine-iodine solution, povidine-iodine sticks, and a sterile lid speculum. At ourcenter, we use sterile gloves, a sterile drape, andan empty sterile 1 cc tuberculin syringe to mark thesclera. Alternatively, one can use a caliper to markthe location for the injection procedure. The drug isdrawn from the drug vial using a filtered needleattached to a tuberculin syringe. The needle is thenchanged to a sterile 30-gauge needle prior to theinjection. Preinjection prophylactic antibiotic dropsmay also be used, although no benefit of antibioticprophylaxis has been established.

The eye should first be anesthetized. In ourhands, topical anesthesia appears to work just aswell as subconjunctival injection of lidocaine, buteither method can be used. For topical anesthesia, acotton tip applicator is soaked with tetracaine and


placed under the upper or lower eyelid in theconjuntival fornix, so that it rests against the super-otemporal or inferotemporal bulbar conjuctiva atthe site where the injection is planned. The patientshould be instructed to look in the opposite directionand remain that way, so as not to scratch the corneaon the cotton tip applicator. After three to fiveminutes, the applicator can be removed and the eyeprepped with 5% povidine-iodine solution placeddirectly on the eye, and povidine sticks used toclean the eyelids, lashes and periocular skin. Glovesare worn and a sterile lid speculum is insertedbetween the eyelids. A sterile drape may be usedover the eye if desired, but is not necessary. Thepatient is then asked to fix his or her gaze in thedirection opposite to where the injection is planned,so as to provide the best possible exposure. Providingthe patients with an object to fixate upon, such astheir own raised thumb, can improve stability of theeye during the injection.

A sterile 1 cc syringe hub or a sterile caliper canbe used to mark the site of injection. The safest pointof injection in phakic patients is 4 mm posterior tothe limbus, and the round tip of the tuberculin 1 ccsyringe happens to be 4 mm in diameter. The drug isthen injected into the vitreous cavity through thepars plana using a 30-gauge needle (0.05 cc totalvolume in the case of ranibizumab or bevacizumab,0.1 cc total volume in the case of pegaptanib). Theneedle is withdrawn and a dry cotton tip applicatoris immediately applied over the injection site for afew seconds to help prevent prolapse and incarcera-tion of vitreous in the wound, which can serve as apossible wick for the introduction of bacteria into theeye. Antibiotic drops are then placed in the eye andthe lid speculum is removed. The eye pressure ismonitored following the injection to confirm that itreturns to normal. Finally, the patient is sent homewith prophylactic antibiotic drops to be used forthree days.

Most compliant patients do not need to berechecked in the clinic until they are due for theirnext injection four to six weeks later, presuming yougive them clear instructions on the signs and symp-toms of infection or retinal detachment and areconfident that they will call you immediately if theywere to develop these symptoms. Povidine-iodine canbe quite irritating to the corneal epithelium. It istherefore normal for patients to have some degree ofirritation, burning and tearing following their injec-tion, in addition to varying amounts of subconjuctivalhemorrhage. The wise physician will warn theirpatients of these possibilities at the time of injectionin order to prevent the inevitable after-hours tele-phone call. However, any antiseptic-associateddiscomfort should resolve by the following day.

Therefore, any pain or decreased vision reported bythe patient on post-injection day one or later should betaken very seriously.

Safety ConsiderationsThe observation that injection of intravitreal bevaci-zumab (33) or pegaptanib (34) for the treatment ofproliferative diabetic retinopathy results in regressionof neovascularization in the fellow eye providescompelling evidence that these molecules are indeedabsorbed systemically to levels that are clinicallyrelevant. Although no serious systemic concernswere raised by the MARINA, ANCHOR or VISIONstudies, it should be remembered that studies of thissize are powered to detect only relatively largedifferences in rare events between the study groups.A modest increase in the risk of heart attack or stroke,for example, might not be detected by these studies. Inthis regard, both the MARINA (16) and ANCHOR (17)trials revealed a non-statistically significant trendtoward an increased risk of serious systemic hemor-rhage. In MARINA, the incidence of such events was1.3% in 0.3 mg group, 2.1% in 0.5 mg group, versus0.8% in sham group at 24 months. In ANCHOR, theincidence of such events was 1.5% in 0.3 mg group,2.1% in 0.5 mg group, versus 0% in sham group at 12months. A similar trend was observed for non-serioussystemic hemorrhages. No such trend was observed inthe VISION trial (10) of pegaptanib, in which theincidence of serious systemic hemorrhage was 0.5%in the treatment arm, versus 1.9% in the sham arm.

These data simply underscore the fact that anti-VEGF agents are potent drugs, and they shouldalways be used with due caution and consideration.

FUTURE ANTI-VEGF THERAPIES

VEGF TrapPegaptanib, ranibizumab and bevacizumab all actthrough inhibition of VEGF-A; they do not bindother members of the VEGF family. VEGF Trap(Regeneron, Tarrytown, New York, U.S.A.) is anexperimental new drug designed to inhibit allmembers of the VEGF family: VEGF-A, -B, -C, -D,and Placental growth factors (PlGF-1 and PlGF -2).VEGF Trap is a recombinant chimeric VEGF receptorfusion protein in which the binding domains of VEGFreceptors 1 and 2 are combined with the Fc portion ofimmunoglobulin G to create a stable, soluble, high-affinity inhibitor. VEGF Trap also binds VEGF-A withhigher affinity (kD!1 pmol/L) than any of thecurrently available anti-VEGF drugs (35). Whetherthe broader spectrum and higher affinity of VEGFTrap equates to improved efficacy in the treatment ofCNV secondary to AMD remains to be determined.

252 KLESERT ET AL.

The CLEAR-AMD 1 study is a randomized,multicenter, placebo-controlled, dose-escalation studydesigned to assess the safety, tolerability and bioac-tivity of VEGF Trap (35). The study enrolled 25 patientswith CNV secondary to AMD with lesions %12 discareas is size and with R50% active leakage, and withETDRS visual acuity of 20/40 or worse. Patients wererandomized to receive either placebo or one of threedoses of VEGF Trap (0.3, 1.0, or 3.0 mg/kg). The VEGFtrap was given as a single intravenous dose, followedby a four-week observation period, followed by threeadditional doses two weeks apart. Dose-limitingtoxicity was observed for two of the five patientstreatedwith the 3.0 mg/kgdose: onepatient developedgrade 4 hypertension and the other developed grade 2proteinuria. Although reduced leakage on fluoresceinangiography and reduced retinal thickening on OCTwas observed in the treated patients, there was nocorresponding reduction in CNV lesion size orimprovement in visual acuity observed in thesepatients over the short 71-day study period. It wasconcluded that the maximum tolerated IV dose ofVEGF Trap was 1.0 mg/kg.

The CLEAR-IT 1 study is similarly designed toassess the safety, tolerability and bioactivity of VEGFTrap through the intravitreal route of administration(36). The study enrolled 21 patients using the sameinclusion criteria as CLEAR-AMD 1, and randomizedthem to receive one of six doses of VEGF Trap as singleintravitreal injection: 0.05, 0.15, 0.5, 1.0, 2.0 or 4.0 mg.After 43 days of follow-up, no adverse ocular orsystemic events were observed. Mean decrease inexcess foveal thickness for all patients was 72%. Themean increase in ETDRS visual acuity was 4.75 lettersand visual acuity remained stable or improved in 95%of patients. Notably, 3 out of 6 patients treated with thehigher doses (2.0 or 4.0 mg) gainedR3 lines of visualacuity by day 43. Clearly, VEGF Trap given intravi-treally shows promise as a novel treatment for CNV inAMD patients.

Small Interfering RNAs (siRNAs)The therapeutic potential of RNA interference wasborn in 1998, when Fire and Mello (37) discoveredthat injection of gene-specific double stranded RNAinto cells resulted in potent silencing of that gene’sexpression. They had discovered one of fundamentalmechanisms by which the cell regulates geneexpression and protects itself against viral infection:RNA interference. Fire and Mello were awarded theNobel Prize in Physiology and Medicine for 2006.

The components of the RNA interferencemachinery have since been identified. Double-stranded RNA binds to a protein complex calledDicer, which cleaves it intomultiple smaller fragments.A second protein complex called RNA induced

silencing complex (RISC) then binds these RNAfragments and eliminates one of the strands. Theremaining strand stays bound to RISC, and serves asa probe that recognizes the corresponding messengerRNA transcript in the cell. When the RISC complexfinds a complementary messenger RNA transcript, thetranscript is cleaved and degraded, thus silencing thatgene’s expression (38).

Small interfering RNAs (siRNAs) have quicklybecome important tools in genetic research, and theirpotential as therapeutic agents is being explored inmany areas of medicine. Reich and Tolentino (38,39)were the first to apply siRNA technology toward thetreatment of CNV. Bevasiranib/Cand5 (AcuityPharmaceuticals, Philadelphia, Pennsylvania, U.S.A.)is a siRNA inhibitor of VEGF, which is given as anintravitreal injection. A phase I, open-label, dose esca-lation study of 15 patients revealed no serious ocularor systemic adverse effects at a dose up to 3.0 mg.

The CARE study (Cand5 Anti-VEGF RNA:Evaluation) is a phase II multicenter, randomized,double-masked, trial of bevasiranib/Cand5 in patientswith CNV secondary to AMD (40). 127 Patients withpredominantly classic, minimally classic, or retinalangiomatous proliferation lesions (occult no classiclesions excluded) were randomized to receive one ofthree doses of the drug (0.2, 1.5, and 3.0 mg) at baselineand at 6 weeks. The primary endpoint was themean change in ETDRS visual acuity from baselineat 12 weeks, which was 4 letters (0.2 mg), 7 letters(1.5 mg), and 6 letters (3.0 mg). The authors havetheorized that these disappointing results stem fromthe fact that bevaciranib/Cand5 only blocks theproduction of new VEGF–VEGF already present atthe time of injection was not inhibited. The investi-gators postulated that a baseline combinationtreatment with a VEGF protein blocker may berequired to “mop up” the preexisting VEGF load.However, the half-life of VEGF is short and it doesnot explain why the results were not seen by 12 weekstime point with siRNA treatment. Efficacy for theproposed treatment combination remains to beshown. The investigators envision a role of bevasir-anib/Cand5 as a long-term “maintenance” drug. TheCARE trial raised no safety concerns, with only onepatient developing uveitis.

Other therapeutic targets for siRNAs are beinginvestigated. siRNAs directed against the VEGFR-1receptor have shown promise in a mouse model ofCNV (41), and are currently in clinical development(Sirna-027, Sirna Therapeutics, Boulder, Colorado,U.S.A.).

Receptor Tyrosine Kinase InhibitorsNon-RNA inhibitors of VEGF receptor tyrosine kinaseactivity have been identified, and their anti-angiogenic


properties are being investigated for use in the treat-ment of systemic malignancy, as well as CNV. Oneadvantage of this class of drugs over those discussedthus far in this chapter is the possibility of an oralroute of administration, thereby avoiding the ocularcomplications associated with frequent intravitrealinjections.

One promising compound is PTK787, which isa non-selective inhibitor of all known VEGF receptors(42). PTK787 has been shown to inhibit retinalneovascularization in a hypoxic mouse model(43,44). Phase I/II clinical trials of PTK787 (Vatalanib,Novartis, East Hanover, New Jersey, U.S.A.), havebeen done in patients with both solid and hemato-logic malignancies, such as the randomized, double-masked, multicenter, phase I/II study of the safety ofvatalanib administered in conjunction with photody-namic therapy with verteporfin to patients withpredominantly classic, minimally classic or occultwith no classic subfoveal CNV secondary to AMD.A multicenter phase I trial of PTK787/Vatalanib inpatients with AMD is the ADVANCE study. Patientswith all CNV lesion types will receive PDT withVisudyne at baseline, and will be randomized toreceive concurrent treatment with either 500 or1000 mg of oral PTK787/Vatalanib or placebo, oncedaily for three months (45). ADVANCE is designed toassess the safety and efficacy of the drug.

AG-013958 (Pfizer, San Diego, California, U.S.A.)is a selective VEGFR and PDGFR inhibitor that iscurrently in phase I/II testing. The route of adminis-tration being examined is subtenon injection.Preliminary results of 21 patients with subfovealCNV indicated that adverse events were mild (15).

Anti-VEGF treatment has enabled a sizeableproportion of treated patients to attain significantvisual improvement or to maintain vision. Futureresearch will hopefully continue to build on theseadvances and make restoration of vision a reality forthe majority of these patients.

SUMMARY POINTS

& The only two anti-VEGF agents currently approvedby the FDA for treatment of CNV are pegaptanib(Macugen, New York, U.S.A.) and ranibizumab(Lucentis).

& Pegaptanib, an aptamer (short oligonucleotide)that specifically binds and inhibits VEGF isoformscontaining at least 165 amino acids, was shown toslow the rate of vision loss in a large, prospective,randomized clinical trial.

& Ranibizumab, an antigen binding fragment of ahumanized monoclonal antibody directed againstall the biologically active forms of VEGF, includingthe known active proteolytic breakdown products,

was shown to slow the rate of vision loss in twolarge, prospective, randomized clinical trials.

& Bevacizumab is a full-sized humanizedmonoclonalantibody with VEGF binding characteristics similarto ranibizumab, is approved by the FDA for sys-temic treatment of metststic colorectal cancer andlung cancer, but is used off-label for the treatment ofneovascular AMD.

& Efficacy and safety of bevacizumab for the treat-ment of neovascular AMD have been reported inseveral retrospective case series and two smallprospective studies, but a large, prospective,randomized, controlled clinical trial has not yetbeen performed.

& Additional anti-VEGF drugs are in different stagesof development but none have yet entered phase IIItrials as of January 2007.

REFERENCES

1. Ferrara N, Henzel WJ. Pituitary follicular cells secrete anovel heparin-binding growth factor specific for vascularendothelial cells. Biochem Biophys Res Commun 1989;161:851–8.

2. Michaelson IC. The mode of development of the vascularsystem of the retina with some observations on its signi-ficance for certain retinal disorders. Trans Ophthalmol SocUK 1948; 68:137–80.

3. Kvanta A, Algvere PV, Berglin L, et al. Subfoveal fibrovas-cular membranes in age-related macular degenerationexpress vascular endothelial growth factor. InvestOphthalmol Vis Sci 1996; 37:1929–34.

4. Wells JA, Murthy R, Chibber R, et al. Levels of vascularendothelial growth factor are elevated in the vitreous ofpatients with subretinal neovascularisation. Br J Ophthal-mol 1996; 80:363–6.

5. Robinson C, Stinger S. The splice variants of vascularendothelial growth factor (VEGF) and their receptors.J Cell Sci 1991; 114:853–65.

6. Keyt BA, Berleau LT, Nguyen HV, et al. The carboxyl-terminal domain (111–165) of vascular endothelial growthfactor is critical for its mitogenic potency. J Biol Chem 1996;271:7788–95.

7. Ng EW, Shima DT, Calias P, et al. Pegaptanib, a targetedanti-VEGF aptamer for ocular vascular disease. Nat RevDrug Discov 2006; 5:123–32.

8. Gragoudas ES, Adamis AP, Cunningham ET, Jr., et al.Pegaptanib for neovascular age-related macular degener-ation. N Engl J Med 2004; 351:2805–16.

9. Chakravarthy U, Adamis AP, Cunningham ET, Jr., et al.Year 2 efficacy results of 2 randomized controlled clinicaltrials of pegaptanib for neovascular age-related maculardegeneration. Ophthalmology 2006; 113:1508–21.

10. D’Amico DJ, Masonson HN, Patel M, et al. Pegaptanibsodium for neovascular age related macular degeneration:two-year safety results of the two prospective, multi-center, controlled clinical trials. Ophthalmology 2006; 113:992–1001.

11. Kaiser PK, VERITAS Study Group. Abstract of Papers,Combined Meeting of Club Jules Gonin and The RetinaSociety, Cape Town, South Africa, October 15–20, 2006.

254 KLESERT ET AL.

12. Adamis AP, Cook G, Shima D, et al. Abstract of Papers,Combined Meeting of Club Jules Gonin and The RetinaSociety, Cape Town, South Africa, October 15–20, 2006.

13. Mordenti J, Cuthbertson RA, Ferrara N, et al. Comparisonsof the intraocular tissue distribution, pharmacokinetics,and safety of 125I-labeled full-length and fab antibodiesin rhesus monkeys following intravitreal administration.Toxicol Pathol 1999; 27:536–44.

14. Shahar J, Avery RL, Heilweil G, et al. Electrophysiologicand retinal penetration studies following intravitreal injec-tion of bevacizumab (Avastin). Retina 2006; 26:262–9.

15. Kaiser PK. Antivascular endothelial growth factor agentsand their development: therapeutic implications in oculardiseases. Am J Ophthalmol 2006; 142:660e1–10.

16. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab forneovascular age-related macular degeneration. N EnglJ Med 2006; 355:1419–31.


18. Regillo CD, Brown DM, AbrahamH, Kaiser PK, Mieler WF.Randomized, double-masked, sham-controlled trial of rani-bizumab for neovascular age-relatedmacular degeneration:PIER study year 1. Am J Ophthalmol 2007 (in press).

19. Fung AE, Lalwani GA, Rosenfeld PJ, et al. An OCT guided,variable dosing regimen with intravitreal ranibizumab(Lucentis) for Neovascular age-related macular degener-ation. Am J Ophthalmol (in press).

20. Michels S, Rosenfeld JR, Puliafito CA, et al. Systemicbevacizumab (Avastin) therapy for neovascular age-related macular degeneration: twelve-week results of anuncontrolled open-label clinical study. Ophthalmology2005; 112:1035–47.

21. Rosenfeld PJ, Moshfeghi AA, Puliafito CA. Optical coher-ence tomography findings after an intravitreal injection ofbevacizumab (Avastin) for neovascular age-relatedmaculardegeneration. Ophthalmic Surg Lasers Imaging 2005;36:331–5.

22. Avery RL, Pieramici DJ, Rabena MD, et al. Intravitrealbevacizumab (Avastin) for neovascular age-relatedmacular degeneration. Ophthalmology 2006; 113:363–72.

23. Spaide RF, Laud K, Fine HF, et al. Intravitreal bevacizumabtreatment of choroidal neovascularization secondary toage-related macular degeneration. Retina 2006; 26:383–90.

24. Rich RM, Rosenfeld PJ, Puliafito CA, et al. Short-term safetyand efficacy of intravitreal bevacizumab (Avastin) forneovascular age-related macular degeneration. Retina2006; 26:495–511.

25. Bashshur ZF, Bazarbachi A, Schakal A, et al. Intravitrealbevacizumab for the management of choroidal neovascu-larization in age-related macular degeneration. AmJ Ophthalmol 2006; 142:1–9.

26. Luthra S, Narayanan R, Marques LE, et al. Evaluation ofin vitro effects of bevacizumab (Avastin) on retinal pigmentepithelial, neurosensory retinal, and microvascular endo-thelial cells. Retina 2006; 26:512–8.

27. Manzano RP, Peyman GA, Khan P, et al. Testing intravitrealtoxicity of bevacizumab (Avastin). Retina 2006; 26:257–61.

28. Bakri SJ, Cameron JD, McCannel CA, et al. Absence ofhistologic retinal toxicity of intravitreal bevacizumab in arabbit model. Am J Ophthalmol 2006; 142:162–4.

29. Bakri SJ, Snyder MR, Pulido JS, et al. Abstract of Papers,Combined Meeting of Club Jules Gonin and The RetinaSociety, Cape Town, South Africa, October 15–20, 2006.

30. www.omic.com (Accessed on July 15, 2007).31. Klesert TR. So you want to try intravitreal Avastin. Retina

Times 2006; 14:18–21.32. Heier JS, Boyer DS, Ciulla TA, et al. Ranibizumab combined

with verteporfin photodynamic therapy in neovascularage-related macular degeneration: year 1 results of theFOCUS study. Arch Ophthalmol 2006; 124:1532–42.

33. Avery RL, Pearlman J, Pieramici DJ, et al. Intravitreal bevaci-zumab (Avastin) in the treatment of proliferative diabeticretinopathy. Ophthalmology 2006; 113:1695e1–15.

34. AdamisAP,AltaweelM,BresslerNM,etal.Changes inretinalneovascularization after pegaptanib (Macugen) therapyin diabetic individuals. Ophthalmology 2006; 113:23–8.

35. Nguyen QD, Shah SM, Hafiz G, et al. A phase I trial of anIV-administered vascular endothelial growth factor trap fortreatment in patients with choroidal neovascularizationdue to age-related macular degeneration. Ophthalmology2006; 113:1522–38.

36. Nguyen QD, Hariprasad S, Shar SM, et al. Abstract ofPapers, Combined Meeting of Club Jules Gonin and TheRetina Society, Cape Town, South Africa, October 15–20,2006.

37. Fire A, Xu S, Montgomery MK, et al. Potent and specificgenetic interference by double-stranded RNA in Caenor-habditis elegans. Nature 1998; 391:806–11.

38. Reich SJ, Fosnot J, Kuroki A, et al. Small interfering RNA(siRNA) targeting VEGF effectively inhibits ocular neovas-cularization in a mouse model. Mol Vis 2003; 9:210–6.

39. Tolentino MJ, Brucker AJ, Fosnot J, et al. Intravitrealinjection of vascular endothelial growth factor small inter-fering RNA inhibits growth and leakage in a nonhumanprimate, laser-induced model of choroidal neovasculariza-tion. Retina 2004; 24:132–8.

40. Brucker AJ, The Cand5 Study Group. Abstract of Papers,Combined Meeting of Club Jules Gonin and The RetinaSociety, Cape Town, South Africa, October 15–20, 2006.

41. Shen J, Samul R, Silva RL, et al. Suppression of ocularneovascularization with siRNA targeting VEGF receptor 1.Gene Ther 2006; 13:225–34.

42. Wood JM, Bold G, Buchdunger E, et al. PTK787/ZK 222584,a novel and potent inhibitor of vascular endothelial growthfactor receptor tyrosine kinases, impairs vascular endo-thelial growth factor-induced responses and tumorgrowth after oral administration. Cancer Res 2000; 60:2178–89.

43. Maier P, Unsoeld AS, Junker B, et al. Intravitreal injec-tion of specific receptor tyrosine kinase inhibitor PTK787/ZK222 584 improves ischemia-induced retinopathy inmice. Graefes Arch Clin Exp Ophthalmol 2005; 243:593–6.

44. Ozaki H, Seo MS, Ozaki K, et al. Blockade of vascularendothelial cell growth factor receptor signaling is suf-ficient to completely prevent retinal neovascularization.Am J Pathol 2000; 156:697–707.

45. Joondeph BC, Szczesny P, Sforzolini B. Abstract of Papers,Combined Meeting of Club Jules Gonin and The RetinaSociety, Cape Town, South Africa, October 15–20, 2006.


18

Laser Prophylaxis for Age-RelatedMacular DegenerationJason Hsu and Allen C. HoRetina Service, Wills Eye Hospital, Philadelphia, Pennsylvania, U.S.A.

INTRODUCTION

Age-related macular degeneration (AMD) is theleading cause of visual loss in people older than 65years in the United States (1–6). Approximately200,000 Americans per year lose central vision due toAMD and 50,000 will lose vision in both eyes.Currently, there are an estimated 38 million Americanseniors with a projected 88 million by 2030, which willlead to a proportional increase in the population at riskfrom vision loss due to AMD. Ninety percent of thesevere visual loss from AMD results from choroidalneovascularization (CNV) (2,7,8). Although thermallaser photocoagulation, photodynamic therapy, andvarious drug therapies to treat neovascular AMD areavailable or on the horizon, they have only proven tobe moderately effective and applicable to a subset ofpatients (9–18). As a result, the development ofpreventive strategies for patients at high risk of devel-oping CNV is very desirable. Even a modestly effectivebilateral preventive treatment can have a substantialimpact on the development of late AMD (geographicatrophy and/or CNV) and the rate of legal blindnesscaused by CNV. According to one estimate, an inter-vention that reduced the risk of developing CNV byjust 30% in eyes of patients with bilateral large drusencould eventually halve the rate of bilateral blindnessfrom AMD (19).

Several natural history studies have identifiedthe presence of large, soft drusen as a significant riskfactor for the development of late complications ofAMD (20–22). In 1973, Gass first described the disap-pearance of drusen after laser photocoagulation (23).Subsequently, laser photocoagulation to promotedrusen resorption has been examined in numerousstudies as a potential prophylaxis against late compli-cations of AMD.

ANATOMY AND PATHOPHYSIOLOGY

In order to rationalize the potential therapeutic role ofprophylactic laser photocoagulation for drusenresorption, it is necessary to define drusen and

understand the anatomy and pathophysiology of theouter retina, retinal pigment epithelium (RPE), Bruch’smembrane and choriocapillaris. The RPE, a monolayerof hexagonal-shaped cells external to the neurosensoryretina and internal to Bruch’s membrane, is intrinsi-cally involved in the outer retina’s metabolism. Itsfunctions include phagocytosis of photoreceptorouter segments, maintenance of the blood–retinalbarrier and the transportation of nutrients and wasteproducts (24–26). Bruch’s membrane is not a truemembrane but a five-layered connective tissue sheet(27). The basal lamina of the RPE is the most internallayer. The inner collagenous layer, elastic lamina andouter collagenous layer comprise the middle elements.The basal lamina of the choriocapillaris is the finalstructure. The choriocapillaris is the innermost layer ofthe choroid and is comprised of an anastomosing sheetof large, fenestrated capillaries. The blood flow in thechoroid is one of the highest in the body, largely tomeet the high metabolic needs of the outer retina andRPE. Nutrients and waste products pass through thefenestrations of the choriocapillaris. Typically, Bruch’smembrane is not a barrier to these molecules and theRPE transports them to and from the outer retina viaactive and passive mechanisms (28).

Druse (plural drusen) is a German-derived wordthat means “nodule.” Literally, drusen are crystallinenodules found in stones. In the ophthalmic literature,there have been numerous clinical and histopathologicdefinitions of drusen (27). The lack of standard termi-nology for drusen makes interpretation of theliterature difficult. Recently, a clinical classificationand grading for AMD was developed. In this system,drusen are whitish-yellow spots external to the retinaor RPE (29). Hard drusen are less than 63 mm, welldefined and yellow–white. Soft drusen are greater than63 mm and are often also referred to as large drusen.They can have indistinct and distinct borders, maycoalesce to form larger, confluent drusen and typicallyarewhite–yellow in color. Pathologically, three types ofsoft drusen have been described: (i) localized detach-ments of RPE and basal linear deposit in eyes with

diffuse basal linear deposits; (ii) localized detachmentsof the RPE and basal laminar deposit in eyes withdiffuse basal laminar deposits; and (iii) localized RPEdetachments due to focal accumulation of basal lineardeposit in eyes without diffuse basal linear deposits(30,31). Ultrastructurally, basal laminar depositsconsist of membrane-bound vesicles, wide-spacedcollagen and amorphous, granular material locatedbetween the plasma membrane and basal lamina ofthe RPE. Basal linear deposits are located external tothe RPE’s basal lamina in the inner collagenous zone.They consist of vesicular and granular electron-densematerial and small foci of wide-spaced collagen(30–35). Histochemically, drusen have been shown toconsist of lipids, mucopolysaccharides, and glycocon-jugates (36–38).

The RPE is a metabolically active tissue layerand, most likely, drusen are derived from RPE (39–41).Studies have demonstrated that RPE cells over timeaccumulate intracellular lipofuscin and other by-products of the catabolism of photoreceptor outersegments (42). It has been shown that the RPE depositscellular material into the sub-RPE space via evagina-tion of its plasma membrane. This probably representsthe deposition of the intracellular accumulation of itsphagocytic by-products. These plasma membrane-bound vesicles break down into drusenoid material(41). With normal aging, Bruch’s membrane alsoundergoes ultrastructural and histochemical changes(43–46). Bruch’s membrane increases in thickness,accumulates lipids and develops protein cross-linking. The hydraulic conductivity (flow per unitpressure) of Bruch’s membrane in normal eyesdecreases with age (45). Similar to drusen, thesealterations in Bruch’s membrane may also representthe accumulation of waste products from the RPE. Thebasal linear and laminar deposits and the alterations inBruch’s membrane may impair the flow of fluid to andfrom the choriocapillaris. The reduced flow of nutri-ents and oxygen and the impaired removal of wasteproducts may impose a metabolic strain on the outerretina and RPE. The relative hypoxia of the RPE andouter retina from an enlarged, hydrophobic (lipid-laden) Bruch’s membrane and drusen may inducethe formation of angiogenic factors and may promotethe formation of CNV (47).

DRUSEN AS A RISK FACTOR FOR CNV

Laser-induced drusen regression has generated inves-tigation because soft drusen have been identifiedas risk factors for CNV and subsequent visual loss.In 1973, Gass noted that 9 of 49 patients (18%) withbilateral macular drusen developed visual loss in oneeye due to “disciform detachment or degeneration”over an average of 4.5 years (23). Smiddy followed 71

patients with bilateral macular drusen for an averageof 4.3 years (20). Eight eyes of seven patients (9.9%)developed exudative maculopathy. Severe visual loss(more than six lines) occurred in seven eyes and thefive-year cumulative risk of developing severe visualloss was 12.7%. Holz prospectively followed 126patients with bilateral drusen and “good visualacuity” (48). The three-year cumulative incidence ofdeveloping CNV or pigment epithelial detachmentwas 13.3%.

The risk for CNV is higher in patients withdrusen in one eye and CNV in the other eye. In Gass’study, 31 of 91 patients lost central vision from CNV intheir fellow eye over an average of four years (23). TheMacular Photocoagulation Study Group followed 127patients who had extrafoveal CNV in one eye (21). Inthe fellow eye, the risk of developing CNV was 58%over five years if large drusen and RPE hyperpigmen-tation were present. The risk dropped to 10% if nodrusen or hyperpigmentation was present. In anotherstudy, the Macular Photocoagulation Study Groupverified that large drusen are a significant indepen-dent risk factor for CNV (49). In this same study, therisk for CNV jumped to 87% in eyes with five or moredrusen, focal hyperpigmentation, one or more largedrusen and systemic hypertension. In Sandberg’sstudy, 127 patients with unilateral CNV were followedfor an average of 4.5 years (50). CNV developed in thefellow eyes at a rate of 8.8% per year. Macularappearance, which included large drusen, was signi-ficantly associated with CNV. One prospective studyfollowed 101 patients with unilateral CNVand drusenin the fellow eye for up to nine years (51). The yearlyincidence of CNV varied between 5% and 11%. Signi-ficant risk factors were the number, size andconfluence of drusen.

Numerous pathologic studies have shown acorrelation between drusen and AMD. Spraul andGrossniklaus examined 51 eyes with AMD and 40age-matched control eyes (34). Soft, confluent andlarge drusen as well as basal (linear) deposits corre-lated with AMD. Curcio demonstrated that basallinear deposits and large drusen are 24 times morelikely to be found in eyes with AMD than age-matchedcontrol eyes (32).

IMPACT OF LASER PHOTOCOAGULATIONON PRESENCE OF DRUSEN

In order to understand how laser results in drusenresorption, it is necessary to examine the cellulareffects of laser on the outer retina, RPE, Bruch’smembrane and choriocapillaris. Laser energy islargely absorbed by the melanin of the RPE andchoroid with shorter wavelengths (e.g., 514 nm argongreen laser) having better absorption compared to

258 HSU AND HO

longer wavelengths (e.g., 810 nm diode laser).Absorption of the laser light elevates the tissuetemperature and causes denaturation of proteins.This thermal effect is called photocoagulation (52,53).The histopathologic characteristics of a laser burndepend on the power, spot size and duration.Smiddy examined the light microscopic changes to ahuman retina 24 hours after argon laser application(54). The juxtafoveal region was treated with laserspots 200 mm in size and 0.5 seconds in duration. Thepower ranged between 200 and 400 milliwatts (mW).Histopathologically, there was a choroidal infiltrate ofmononuclear and polymorphonuclear cells. The chor-iocapillaris was acellular at the center of the burn. TheRPE was disrupted and the outer and inner retinalnuclear layers were pyknotic. The ganglion and nervefiber layers were also affected. Thomas conducted asimilar study examining a human eye 24 hours afterargon laser (55). One laser spot with a power of310 mW, spot size of 100 mm and duration of 0.5seconds was applied in the superonasal quadrant.Variable RPE necrosis and advanced choriocapillarisnecrosis was seen. A second argon laser burn with apower of 210 mW, spot size of 500 mm and duration of0.5 seconds in the peripapillary region demonstratedsignificant RPE disruption, choriocapillaris necrosisand Bruch’s membrane disruption.

A number of studies have been performed withlaser on cynomologus monkeys whose fovea is similarto the human fovea. Smiddy placed a 13-spot burn inthe juxtafoveal region of a cynomologus monkey withargon green laser and examined the histopathologiceffects at one and seven days (56). He used a 200 mmspot size, 0.2 second duration and power between 100and 200 mW. The desired reaction was a laser burnthat turned the retina light gray. At day one, theganglion cell layer was partially preserved but alldeeper layers were necrotic with RPE hyperplasia.At day seven, there was disruption of the retina tothe level of the ganglion cell layer. In a second study,Smiddy demonstrated that the RPE undergoes cellularproliferation after argon laser (57). Peyman examinedthe histopathologic effects of argon blue–green laser tothe parafoveal area of cynomologus monkeys (58). Heused a 100 mm spot size, 0.1 second duration andpower of 100 mW. At day one, there was coagulativenecrosis of the RPE, outer nuclear layer and outerplexiform layer. The choroid was minimally affected.At days 12 and 21, glial tissue had replaced the outerretina. There was an inflammatory infiltrate and theRPE was hyperplastic. If the power was increased to320 mW, the basement membrane was ruptured andchoroidal hemorrhages developed. Coscas treated theparafoveal region of adult baboons with argon greenlaser and examined the light and electron microscopicchanges at one hour, three weeks and six weeks (59).

As in the above studies, they showed disruption of theouter retina, necrosis of the RPE and a macrophageresponse. Depending on the laser settings, there wasvariable involvement of the choriocapillaris. In areview of macular photocoagulation, Swartz foundthat the histologic characteristics of a moderateargon-green burn showed a typical cone-shapedlesion sparing the inner retina (60). The laser intensi-ties of these studies exceed those in most human laserto drusen trials.

No histopathologic studies have been performedon human eyes examining the effects of laser ondrusen. However, a limited number of experimentalanimal studies have been reported. Duvall and Tsoapplied argon green laser directly to drusen in twoeyes of a rhesus monkey and noted the light micro-scopic and ultrastructural characteristics of drusenresorption (61). At zero to two days, outer segmentretinal disruption, RPE necrosis and fibrin depositionwere noted. The drusen were still present. At three toeight days, two types of macrophages were present.One type was in the outer retina and subretinal spaceand had an appearance that was consistent withblood-borne monocytes. The second type of macro-phage contained cell processes that surrounded thedrusen material. These cell processes were traced byserial sectioning to the pericytes of the choriocapillaris.At nine days and beyond, there was resorption of thedrusen. Blood-borne monocytes were densely packedin the subretinal space. The cell processes of thechoroidal pericytes contained drusenoid material.The authors postulated that the fibrin depositionfrom the laser photocoagulation initiated a phagocyticresponse, which resulted in clearance of the drusen bychoroidal pericytes. Perry examined the choroidalmicrovascular response to argon laser in cats (62).He demonstrated activation of the endothelial cellsin the choriocapillaris after laser photocoagulation.Della treated a rhesus monkey with soft large drusen(63). He used an argon laser to apply a grid pattern inthe macula. Six weeks after laser, the directly treateddrusen had disappeared.

THEORIES ON DRUSEN REDUCTIONAND CNV PREVENTION

Drusen disappearance after laser photocoagulationis clearly documented in the literature (64–76).However, the mechanism of drusen disappearanceis not well understood. Several theories have beenproposed: (i) phagocytosis of drusen; (ii) decreaseddeposits by removal of RPE; (iii) release of solublemediators; (iv) thinning of Bruch’s membrane; and(v) mechanical alteration of the structure of Bruch’smembrane. It is clear from the above studies that laserinduces an inflammatory response and the intensity

18: LASER PROPHYLAXIS FOR AGE-RELATED MACULAR DEGENERATION 259

of the reaction depends on the laser settings as wellas the laser subjects. The differences in these factorsbetween various studies make interpretation difficult(25,54–57,59–62). Furthermore, in most of the clinicalstudies of laser to drusen, the calibrated intensity isminimal whitening. This is different from the abovestudies where stronger intensities were evaluated.However, despite these limitations, we can postulatethat laser-induced phagocytosis of drusen occurs.Blood-borne inflammatory cells may ingest thedrusen material. Studies certainly indicate theirpresence after laser. Duvall and Tso noted drusenoidmaterial in cell processes after laser photocoagulationand attributed the origin of these cell processesto choroidal pericytes (61). Dysfunctional RPE,destroyed by laser, is replaced by proliferating RPE(57). The RPE has phagocytic ability and the prolifer-ating RPE may be involved in drusen resorption (69).Also, the removal of dysfunctional RPE cells may haltfurther drusen development and allow removal ofaccumulated material. After laser-induced tissuedamage, the RPE and other cells may producesoluble mediators. For instance, Glaser showed thatRPE cells release an inhibitor of neovascularization(77). These soluble mediators may enhance thenatural processes that result in spontaneous drusenresorption (23,70,78). They might also account for theobservation that drusen distant from laser burnsdisappear after photocoagulation.

Bruch’s membrane in AMD eyes is diffuselythickened and hydrophobic. The structural effect onBruch’s membrane by laser is variable. Thomasshowed the integrity of Bruch’s membrane dependedon the energy density of the laser (55). Photocoagula-tion may thin the abnormally thick Bruch’s membraneand, in theory, improve its hydraulic conductivity. Theincreased metabolic transport could improve drusenclearance and decrease drusen formation. The lasercould also exert a mechanical effect on Bruch’smembrane, causing contraction of collagen andelastin (similar to laser trabeculoplasty) andimproving egress of material through a more per-meable Bruch’s membrane. Peyman showed thatphotocoagulation may improve perioxidase diffusionfrom the vitreous to the choroid (79).

However, it is important to note that drusenreduction seems to occur during the natural courseof AMD. Soft drusen often progress to confluence,drusenoid PED, and fading which leads to RPEdisturbances or atrophy in some cases. Over thecourse of five years, large drusen have been seen todisappear in 34% of eyes with very early changesconsisting of one or a few large drusen (78). Also,among fellow eyes of patients enrolled in the MacularPhotocoagulation Study with CNV in one eye, areas oflarge drusen disappeared from one or more areas and

new large drusen appeared in an additional 13% ofeyes (80). Nevertheless, large spontaneous reductionsof greater than 50% in drusen area are uncommon inpatients with 10 or more large drusen (81).

Similar to drusen reduction, it is unclear howlaser to drusen might prevent CNV. Some of the sametheories on the mechanism of drusen reduction applyto CNV prevention. Improved transport of nutrientsacross Bruch’s membrane might reduce the metabolicstrain on the RPE/outer retina and stop the productionof angiogenic factors from the RPE. Indeed, lasermight even induce the production of vasoinhibitorygrowth factors from the RPE. Gass postulated thatlaser “tacks” down the RPE to Bruch’s membrane,eliminating a potential cleavage plane for CNV (23).Proliferating RPE, induced by the laser, may envelopearly CNV and prevent further growth.

UNCONTROLLED STUDIES AND CASE REPORTS

Since Gass first described the disappearance of drusenafter laser photocoagulation, there have been anumber of case reports and uncontrolled clinicalstudies that have examined the prophylactic treatmentof drusen. Sigelman published a case report of a58-year old woman with a disciform scar secondaryto AMD in the right eye and confluent soft drusen inthe left eye (82). The patient’s vision dropped to 20/40with metamorphopsia in the left eye. There was noCNV but an increased density and size of fovealdrusen. Using a wavelength of 576 nm (yellow),power of 180 mW, duration of 0.3 seconds, and spotsize of 200 mm, he directly treated drusen and alsoapplied a parafoveal grid for a total of 56 spots.Treated and untreated drusen disappeared and thevision returned to 20/20 one year after treatment.

Hyver reported laser photocoagulation in apatient with CNV in one eye and large, confluentsoft drusen in the fellow eye (83). Using a wavelengthof 630 nm, spot size of 200 mm, duration of 0.05seconds and power of 200 mW, 24 burns were placedin the temporal macula with no direct drusen treat-ment. Burn intensity was calibrated to create barelyvisible whitening. Ten months after treatment thevisual acuity had dropped from 20/25 to 20/60,which was felt to be due to the development of agranular subfoveal material. No CNV was noted onfluorescein angiography.

Cleasby treated 29 eyes in patients with “exuda-tive senile maculopathy (ESM)” in the fellow eye (65).In addition, one eye of 25 patients with “nonexuda-tive senile maculopathy (NSM)” in both eyes wastreated. The criteria for NSM included the presenceof drusen, retinal pigment epithelial atrophy andclumping and/or cholesterol deposits in the maculain individuals older than 50. Argon laser was used to

260 HSU AND HO

directly treat drusen “within a broad ring around thefovea.” The desired intensity was a minimally visiblereaction in the retina. The laser parameters were aspot size of 50 to 100 mm, power between 100 and150 mW and duration of 0.05 to 0.1 seconds. Thenumber of applications was approximately 200 to300 shots. In the group of 29 patients with ESM inone eye, three developed ESM in the treated eye overan average follow-up of 28.4 months. This rep-resented a 4.4% yearly rate of ESM formation,which is less than the natural history of AMD. Inthe NSM group, neither the control eyes nor thetreated eyes developed ESM over an average follow-up of 27.3 months. All 25 treated eyes and five controleyes showed a reduction in drusen. There were noreported complications from the laser. Despite a smallnumber of patients, no control group for the ESMeyes and no randomization for NSM eyes, this studysuggested prophylactic laser to drusen mightbe beneficial.

Wetzig treated 42 eyes of 27 patients withprophylactic laser in a retrospective, nonrandomizedstudy (75). All patients had macular soft drusen andrecent visual changes (visual loss or metamorphopsia).The vision ranged from 20/20 to 20/400. Only 25%of eyes had a best-corrected pre-laser visual acuity of20/40 or better. The mean age at treatment was 69years. Eyes with CNV or hemorrhagic exudativechanges were excluded. Thirty-one eyes were treatedwith krypton red laser, one eye with a combination ofxenon and krypton, eight eyes with argon laser andtwo eyes with a combination of argon and kryptonlaser. Both eyes were treated in some patients andseveral eyes were retreated. The desired intensity ofthe laser reaction was a faint, white gray spot.Approximately 50 to 75 spots were applied in ascatter pattern around the fovea. The vision improved,remained stable or worsened by only one line in 22eyes (52%) over an average follow-up of 3.7 years.CNV developed in 12%. Drusen disappeared in thesetreated eyes, usually beginning at three months.Wetzig published a follow-up of these patients sixyears after the original publication (76). The averagefollow-up time was 120 months. Of the treated eyes,33% remained stable or lost one line of visual acuity,21% lost two to three lines and 46% lost three or morelines. CNV developed in 21% of treated eyes duringthe follow-up and several patients had progressiveenlargement of the treatment scars. While no controlgroup was designated, seven eyes with drusen hadgone untreated. In this untreated group, three eyesretained 20/40 or better visual acuity, two eyes losttwo or more lines and two eyes worsened to 20/400 orless. Overall, no clear beneficial effect of prophylacticlaser was demonstrated. However, the retrospective,nonrandomized design with a small number of eyes

limits the conclusions that can be drawn from thisstudy. Also, it included many patients withpoor vision and selected patients with visual symp-toms. These patients may have harbored subtleoccult CNV.

Figueroa treated 20 patients with argon laser (66).Group 1 consisted of 14 patients with bilateral drusenwith one eye randomly assigned to receive lasertreatment. Group 2 consisted of six patients withCNV in one eye and drusen in the fellow eye. Thepatients ranged in age from 55 to 80 years and theaverage follow-up was 18 months. Drusen temporal tothe fovea were directly treated with the argon greenlaser with a spot size of 100 mm, duration of 0.1seconds, power of 100 mW, and mean number oflaser spots of 30. The desired laser intensity wascalibrated to achieve a light gray–white lesion.Treated drusen disappeared by approximately twomonths while surrounding, untreated drusen disap-peared at a mean of 10 months. Visual acuityimproved in 30% of eyes by one line or more. Thiswas secondary to the resorption of untreated subfo-veal drusen. The visual acuity remained unchanged in65% of eyes and decreased in 5% (one eye). The oneeye that worsened developed a choroidal neovascularmembrane away from the laser scars. Figueroaupdated these results and presented new data in asecond publication with 30 patients in Group 1 and 16patients in Group 2 (67). The laser settings were thesame as described above. All treated drusen disap-peared at an average of 3.5 months. In all but threepatients, the untreated drusen resolved by an averageof 8.6 months. The drusen disappearance progressedin a temporal to nasal direction. Superonasal drusenpersisted for the longest amount of time. Two of the 30control eyes in Group 1 (bilateral drusen) demon-strated spontaneous drusen resolution. After anaverage of three years, one control eye but no treatedeyes developed CNV. Three fellow eyes (18%) inGroup 2 developed CNV. In one eye, the CNVdeveloped adjacent to the laser scars. Again, due tothe small number of patients, interpretation of theseresults should be approached with caution.

Sarks treated 18 eyes of 16 patients with bilateraldrusen and one eye of 10 patients with exudativechanges in the other eye (74). Patients were 55 yearsor older and followed for a mean of 16.8 months.Inclusion criteria included visual acuity 20/40 orbetter and no evidence of atrophy or CNV. A ring of40 to 50non-confluent laser burnswas applied approxi-mately 1500 mmfromthe foveal center.Drusenwerenotdirectly targeted. Argon green laser was used with aspot size of 100 mm, duration of 0.05 to 0.1 seconds andpower calibrated to produce “a barely discernablewhitening of the RPE.” In 14 of the 16 patients withbilateral drusen, only one eye was treated. In these


treated eyes, the vision remained stable in 10 eyes andimproved in four eyes. Vision decreased in four eyesand remained stable in 10 eyes in the untreated group.Overall, in the two treated groups, visual acuityimproved in 12 eyes (40%), remained unchanged in16 eyes (53%) and worsened in 2 eyes (7%). Visualimprovement was related to foveal drusen resorption,which occurred in all treated eyes but none of theuntreated eyes. Two treated patients developed CNVat seven and eight months post-treatment in retinaadjacent to laser burns. Expansion of laser-inducedatrophy was minimal in this study.

Guymer treated one eye of 12 patients at highrisk for visual loss secondary to AMD (71). All 12treated eyes demonstrated macular drusen and visualacuity of 20/40 or better. Ten patients had end-stagelesions in one eye and two patients had bilateral softconfluent drusen. Twelve laser spots were placed in aring 750 to 1000 mm from the fovea. Argon green laserwas used with a spot size of 200 mm, duration of 0.2seconds and power calibrated to achieve faintblanching of the RPE (80–300 mW). The averagefollow-up was 16 months. Visual acuity remained thesame or improved in 11 patients. Nine of the 11patients had a reduction in drusen size, number andconfluence. One patient lost four lines due to develop-ment of CNV that did not originate from a laser site.Two patients developed profound atrophy at the lasersite and four others developed RPE pigmentarychanges at the laser sites. This study showed that asmall number of laser applications could promotedrusen disappearance. It also showed no correlationbetween drusen resolution and improvement ordeterioration of dark-adapted retinal thresholds.

Ruiz Moreno performed a prospective, nonran-domized clinical study of laser photocoagulation andlooked at the development of macular atrophy in aconsecutive series of patients with soft drusen whounderwent argon green laser photocoagulation (84).Eyes had to have documented recent loss of visualacuity preceding treatment in order to be included in

the study. Fifty-two consecutive eyes of 52 patientsreceived direct photocoagulation to drusen. Laserparameters included a spot size of 200 mm, durationof 0.2 seconds and power titrated to a light gray–whiteretinal reaction. Treatment was performed greater than500 mm from the foveal center with a mean of 117 spotsand was completed over two sessions. Macularatrophy occurred in nine eyes (17.7%) about 37.2months after photocoagulation (range 7–75 months)and was associated with a significant decrease invisual acuity. There was no significant correlationbetween the areas of atrophy and the number oftreatment spots (pZ0.97) or the intensity of treatmentspots (pZ0.09). Due to the uncontrolled nature of thisstudy, it is unclear if the macular atrophy is attribu-table to the laser treatment or related to the naturalcourse of AMD.

CONTROLLED STUDIES

Information from the above studies confirmed thatlaser promoted drusen reduction. However, thevisual benefit of this prophylactic laser was stillunclear. These studies provided the impetus for morecontrolled studies and larger clinical trials. Tables 1and 2 summarize the findings from these studies.

Frennesson conducted a randomized, prospec-tive study of prophylactic laser treatment (68). One eyeof 13 patients with bilateral soft drusen was treated. Ina second group, the fellow eye of six patients with adisciform lesion in the other eye was treated. Thecontrol group consisted of 19 patients who had beenrandomized to observation. The groups were matchedfor age and visual acuity but there were more men inthe treatment group. The visual acuity in all treatedeyes was 20/25 or better. Patients with macularpigment clumping, atrophy, pigment epithelialdetachments or exudative AMD were excluded. Ahorseshoe-shaped grid pattern with direct drusentreatment as well as scatter treatment was appliedwith argon green laser. Laser parameters included

Table 1 Controlled Bilateral Drusen Studies

Mean follow-up(mos)

Drusen regression CNV development

Study Laser type N (pts) Treated Control Treated Control

Cleasby (65) Argon—threshold 25 27 25 5 0 0

Figueroa (66) Argon—threshold 30 36 30 2 0 1

Frennesson (68) Argon—threshold 13 96 2 4

Little (73) Dye—threshold 27 38 2 4

Olk (85) Diode—threshold and

subthreshold

77 24 71 7 3 3

Scorolli (86) Argon—threshold/

Diode—subthreshold

78/66 18 3/1 4

Choroidal neovascularization

prevention trial (64)

Argon—threshold 156 30 77 12 4 2

Drusen laser study (87) Argon/dye—threshold 105 36 12 7

262 HSU AND HO

a spot size of 200 mm, duration of 0.05 seconds andpower of 100 to 200 mW. The number of laser spotsvaried from 51 to 154 spots with intensity calibrated toachieve a “grayish reaction.” Drusen area on colorfundus photographs and fluorescein angiogramswere calculated at baseline and follow-up for bothgroups. Follow-up results were published at 6months, 12 months, 3 years, and 8 years (68–70,88).The mean drusen area significantly decreased in thetreated eyes and significantly increased in the controleyes. Over three years, five eyes (33%) in the controlgroup but none in the treatment group developedCNV. By eight years of follow-up, 29 patientsincluding 13 treated and 16 controls remained in thestudy. Nine of the 16 controls (56%) developed CNV(five fellow eyes and four bilateral drusen eyes) andonly 2 of 13 treated eyes (15%) developed CNV (twobilateral drusen eyes). While vision decreased signi-ficantly in both groups, the magnitude was greater inthe control groupwith quadrupling of minimum angleof resolution in controls versus doubling of minimumangle of resolution in treated patients. This studydemonstrated that laser treatment promotes drusenresorption, which had also been shown in the abovestudies. Importantly, it suggested that laser prophy-laxis might prevent the exudative complications ofAMD. However, as with the above studies, thesample size was small and the confidence intervallarge, making it difficult to draw valid conclusions.

Little randomized one eye of 27 patients withbilateral confluent soft drusen to prophylactic treat-ment (73). Mean age of patients was 69.7 years. Theminimum visual acuity was 20/60with amean follow-up of 3.2 years. Foveal atrophy, pigment epithelialdetachments and exudative changes were exclu-sionary criteria. Drusen were directly treated. Lasersettings for the dye laser (577–620 nm) were a spot sizeof 100–200 mm, duration of 0.05 to 0.1 seconds andpower of 100–200 mW calibrated to induce a slightlightening of the RPE/outer retina. No laser spotswere applied within 300 mm of the foveal center andrarely within 500 mm. A total of 23 to 526 laser spotswere applied, and 37% of eyes were treated over morethan one session. Six treated eyes and no control eyesimproved two or more lines, 16 treated and 17 control

eyes remained stable, and five treated and 10 controleyes lost two or more lines. Drusen resorption within1500 mm of the fovea occurred more completely in thetreated eyes than control eyes. In five eyes of bothgroups, there was equal drusen disappearance. Fourcontrol patients and two treated patients developedCNV. Laser scar enlargement occurred in three eyes. Itis again difficult to draw conclusions from this studydue to the small sample size but visual acuity anddrusen resorption were significantly better in thetreated eyes.

Olk studied the use of diode laser photocoagula-tion for 152 patients with macular drusen (bilateraldrusen, 77 patients; fellow eye, 75 patients) (85). Theseinvestigators also compared the ability of subthreshold(invisible) diode laser photocoagulationwith threshold(visible) laser photocoagulation to reduce the numberof large drusen. Visual acuity was 20/63 or better atbaseline. During the first 12 months of follow-up,threshold laser photocoagulation appeared to inducea more rapid disappearance of drusen compared withsubthreshold laser. By 18 months, no difference wasnoted between the two groups. During the 24 monthsof follow-up, laser treated eyes had significant drusenreduction and improvement in visual acuity comparedwith observed eyes. CNV occurred at similar rates inboth treated and observed eyes.

Scorolli compared using argon laser withsubthreshold 810 nm diode-laser in 144 patients withbilateral macular drusen (78 eyes received argon, 66eyes received diode laser) (86). One eye of each patientwas treated with the second eye serving as control.During a mean of 18 months follow-up, best-correctedvisual acuity was statistically significantly improvedin both treatment groups compared with controls,with no significant difference between the argon anddiode groups. Drusen reduction occurred in bothtreated groups as well compared with controls. CNVdeveloped in three eyes receiving argon laser, one eyereceiving diode laser, and four eyes in the untreatedgroup. Visual field testing revealed minor but statisti-cally significant reductions in the argon group but notin the diode group. A slight reduction in contrastsensitivity was also noted in the argon group but notseen in the diode group. However, it should be noted

Table 2 Controlled Fellow Eye Studies

CNV development

Study Laser type N (pts) Mean follow-up (mos) Treated Control

Frennesson (68) Argon—threshold 6 96 0 5

Olk (85) Diode—threshold and

subthreshold

75 24 8 7

Choroidal neovascularization

prevention trial (64)

Argon—threshold 120 30 10 2

Drusen laser study (87) Argon/dye—threshold 177 36 27 15


that the treatment protocol in the argon group (greenlaser with power titrated to graying effect, 0.2 secondduration, 100 mm spot size, and 200 spots placed500 mm outside the foveal avascular zone) differedfrom the diode group (150 mW power, 0.1 secondduration, 125 mm spot size, 48 spots placed 750 to2250 mm outside the foveal avascular zone).

In 1994, the largest randomized pilot trial to date,the Choroidal Neovascularization Prevention Trial(CNVPT) was begun (64,72,81,89,90). A total of 312eyes of 156 patients without exudative AMD andmorethan 10 large (more than 63 mm) drusen in each eyewere enrolled in the Bilateral Drusen Study and 120eyes of 120 patients with exudative AMD in one eyeand more than 10 large drusen in the other eye wereenrolled in the Fellow Eye Study. Study eyes wererequired to have visual acuity of 20/40 or better withno evidence of current or past CNV and progressiveocular disease. Fluorescein angiography was used toexclude CNV in the study eye at baseline. Patients inthe bilateral drusen arm had one eye randomized tolaser treatment with the second eye serving as thecontrol. Patients in the fellow eye arm had the non-exudative AMD eye randomized to laser treatment orcontrol. Laser parameters included a spot size of100 mm, duration of 0.1 seconds, and power titratedto a light gray–white lesion. Figure 1 shows thetreatment protocol for 85% of the patients, whichconsisted of 20 laser spots placed in three rows tempo-ral to the fovea and greater than 750 mm from thecenter. Figure 2 shows the treatment protocol for theremaining patients, who received 24 laser spots ofthe same intensity placed in two rows temporal tothe fovea and greater than 750 mm from the center.

The CNVPT protocol specified that eyes assignedto treatment be retreated at 6 months nasal to the fovea

in a mirror image of the first treatment if the area ofdrusen had not decreased by 50% from baseline. At sixmonths, 28% of 78 eyes in the Bilateral Drusen Studyand 41% of 37 eyes in the Fellow Eye Study had a 50%reduction in drusen and were exempt from retreat-ment. By 12 months, 54% of 35 eyes in the BilateralDrusen Study and 27% of 11 eyes in the Fellow EyeStudy had a 50% reduction. One eye in the observedgroup had a 50% reduction in drusen area. Less than10% of treated eyes and more than 90% of observedeyes showed no reduction in the area of drusen at 12months (64). Laser-treated eyes with a 50% or morereduction in drusen at this follow-up were more likelyto have improved contrast sensitivity as well as one-and two-line increases in visual acuity compared withlaser-treated eyes with less drusen reduction orobserved eyes (pZ0.001) (72).

Enrollment was suspended early due to theapparent increase in CNV development within thefirst 12 months of follow-up in the Fellow Eye Study.CNV was seen in 10 of 59 treated eyes versus only 2 of61 control eyes (pZ0.02). In the Bilateral Drusen Study,CNV developed in 4 of 156 treated eyes and 2 of 156control eyes (pZ0.62). With additional follow-up, thesignificant increase in CNV incidence in treated felloweyes compared with control eyes was maintainedthrough 18 months but by 30 months the incidenceof CNV was the same in both groups (91). Moreover,there were no statistically significant differences inthese fellow eyes compared with controls in terms ofchange in visual acuity, contrast threshold, criticalprint size, or incidence of geographic atrophy.

Owens who reported the results of a randomized,controlled clinical trial, the Drusen Laser Study, sawsimilar findings (87,92,93). A total of 177 eyes of 177patients with exudative AMD in one eye and drusenwith or without pigment clumping were enrolled in thefellow eye group and 210 eyes of 105 patients with soft

100

750–1000Microns

Fovea

300

300300

Figure 1 Configuration of burns in the Laser 20 treatment

protocol of the choroidal neovascularization prevention trial.Source: From Ref. 64.

Figure 2 Configuration of burns in the Laser 24 treatmentprotocol of the choroidal neovascularization prevention trial.

Source: From Ref. 64.

264 HSU AND HO

drusen with or without pigment clumping wereenrolled in the bilateral group. Baseline visual acuitiesof the study eyes were 20/40 or better, and fluoresceinangiographywasperformedatbaseline to excludeCNV.Argongreen or yellowdye laserwasusedwith a 200 mmspot size, 0.2 second duration, and 65 to 120 mWpower. Twelve spots at a distance of 1000 mm from thecenter of the foveola were applied in a circular protocolpattern. Over three years of follow-up in the bilateraldrusen group, CNV developed in 12 of 103 treated eyes(11.6%) and 7 of 103 observed eyes (6.8%, pZ0.23).Visual acuity decreased by 15 or more letters in 6 of 72treated eyes (8.3%) and 10 of 72 observed eyes (13.9%,pZ0.39). During three years of follow-up in thefellow eye group, CNV developed in 27 of 91 treatedeyes (29.7%) and only 15 of 85 observed eyes (17.7%,pZ0.06). Visual acuity decreased in 21 of 73 treated eyes(28.8%) and 13 of 66 observed eyes (19.7%, pZ0.21).Neither of these results was statistically significant, butthe investigators felt compelled to halt recruitment intothe trial due to concern for laser-induced CNV in thefellow eye group. In the final analysis, one of the mostsignificant findings from the fellow eye group wasthat the incidence of CNV occurred six months earlierin the laser treated eyes comparedwith the no laser eyes(pZ0.05). This finding was maintained throughout thethree years of follow-up.

The increased incidence of CNV in laser-treatedfellow eyes has been somewhat unexpected. It isknown that fellow eyes are at higher risk for CNVdevelopment than bilateral drusen eyes. One possi-bility is that some of the fellow eyes had undetectedCNV at baseline that was stimulated by the lasertreatment. These eyes may simply harbor moreadvanced AMD that is less amenable to prophylaxis.Differences in laser treatment strategy may also play arole. While some groups have specifically targetedmacular drusen, the CNVPT and Drusen Laser StudyGroup followed a pattern that resulted in laser beingdirected either between or directly on drusen. Theintensity of laser photocoagulation may also play arole. Using a computerized method of laser burnquantitation, Kaiser demonstrated that patients inthe CNVPT who received more intense burns weremore likely to have greater drusen resolution (90).However, a higher laser burn intensity seemed tocorrelate with increased risk of CNV development.Ultimately, one major challenge may be to deliver asufficient amount of energy to promote a protectiveeffect while limiting the risk of CNV stimulation.

FUTURE DIRECTIONS: MULTICENTEREDCLINICAL TRIALS

Based on the favorable data from Olk using diodelaser, a larger multicenter, randomized, prospective

clinical trial known as the Prophylactic Treatment ofAMD (PTAMD) Trial is currently in progress tocompare subthreshold infrared (810 nm) diode laserphotocoagulation with observation. Enrolled patientshad visual acuities of 20/63 or better. Figure 3 demon-strates the laser protocol, which consisted of a gridof 48 sub-threshold 810-nm diode laser spots with aspot size of 125 mm applied in four concentric circlesoutside the foveal avascular zone. Only one lasertreatment was applied throughout the study duration.Patients with at least five large drusen (more than63 mm) within 2250 mm of the foveal avascular zone inboth eyes were placed in the bilateral arm of the studywith one eye being randomized to treatment and theother serving as control. Approximately 600 patientswere enrolled into this arm by November 2001. A sub-study of 100 eyes (50 patients) enrolled in this bilateralarm of the PTAMD revealed that the number oflaser-induced lesions and the surface area of thelaser-induced RPE changes on fluorescein angio-graphy at three months post-treatment were strongpredictors of major drusen reduction by 18 monthspost-treatment (94). This finding may explain thehigher rate of drusen reduction in patients who weretreated with threshold diode laser in the pilot studyand echoes the findings of Kaiser from the CNVPT.

The PTAMD also enrolled patients with neovas-cular AMD in one eye and at least five large drusenin the fellow eye. These patients were placed in theunilateral arm with the eligible fellow eye beingrandomized to treatment or observation. Enrollmentin the unilateral arm was suspended in April 2000after 242 patients were enrolled due to an increasedincidence of CNVand higher rates of worsening visualacuity in treated eyes (95). Follow-up of these patientsis on-going.

Figure 3 Artist’s illustration of 48 diode-laser lesions in a grid

pattern of four concentric circles 750 to 2250 mm from the centerof the foveal avascular zone. Source: From Ref. 86.


Based on the findings of the CNVPT, themulti-center randomized clinical trial known as theComplications of Age-Related Macular DegenerationPrevention Trial (CAPT) was proposed and is beingconducted with support from the National Eye Insti-tute. The goal of CAPT is to determine whetherprophylactic low-intensity laser treatment to theretina can prevent vision loss associated with thecomplications of advanced AMD (96,97). Whilefellow eyes in the CNVPT that were treated showeda higher rate of CNV development, in patients withbilateral drusen, the risk was found to be relativelylow and similar between treated eyes and control eyes.As a result, only patients with bilateral high-riskdrusen (10 or more drusen larger than 125 mm within3000 mm of fovea) were incorporated into the CAPTdesign. Baseline visual acuity was 20/40 or better. The

laser treatment protocol also was modified based onthe CNVPT findings. Given an apparent increasedincidence of CNV in eyes that received more intenselaser burns, the burn intensity was reduced to a barelyvisible lesion (90). The initial treatment consisted of 60burns (100 mm spot size, 0.1 seconds duration) in a gridpattern within an annulus between 1500 and 2500 mmfrom the fovea. Retreatment could be performed at 12months if 10 or more drusen 125 mm or largerremained within 1500 mm of the fovea. Figure 4shows the follow-up treatment protocol, whichconsisted of 30 burns in the 1000 to 2000 mm annuluscentered on the fovea with drusen being treateddirectly.

A total of 1052 patients were recruited by March2001 with one eye being randomized to receive lasertreatment and the other eye to observation. Patients

(A) (B)

2500 μm

1500 μm

2000 μm

Fovea1000 μm

= Drusen

Fovea

Figure 4 (A) Initial laser treatmentprotocol in the complications ofage-related macular degenera-tion prevention trial. (B) Repeat(12-month) protocol. Source: FromRef. 96.

(A) (B)

Figure 5 (A) Extensive, confluent drusen in a 51-year-old woman at the time of enrollment incomplications of age-related macular degeneration prevention trial; visual acuity was 20/40. (B) Marked

regression of drusen in the same eye one year after laser treatment according to trial protocol; visual acuityhad improved to 20/25. Note that the reddish discoloration in the central macula is not representative of

hemorrhage.

266 HSU AND HO

will be followed for a minimum of five years withthe primary outcome measure being change in visualacuity. Secondary outcome measures include theincidence of advanced stage AMD changes (CNV,pigment epithelial detachments, geographicatrophy), contrast threshold, and critical print size.Sample fundus photographs at baseline and 12

months follow-up from two patients in the laser-treatment group are depicted in Figures 5 and 6.Drusen regression is demonstrated in Figure 5,and CNV development with subsequent disciformscarring is shown in Figure 6. While studies havedemonstrated regression of drusen in laser-treatedeyes, it is important to remember that drusen

(A) (B)

Figure 6 (A) Extensive drusen in the macula in a 64-year-old man at the time of enrollment incomplications of age-related macular degeneration prevention trial; visual acuity was 20/20. (B) Large,

fibrous, disciform scar in the macula of the same eye one year after laser treatment according to trialprotocol; visual acuity had decreased to 20/400.

(A) (B)

Figure 7 (A) Extensive macular drusen in a 75-year-old woman at the time of enrollment in complications ofage-related macular degeneration prevention trial; visual acuity was 20/25. (B) Substantial regression of

drusen in the same eye one year after enrollment; visual acuity was 20/25C. The eye had been assigned tothe control group.


regression can also occur spontaneously though typi-cally at a lower rate compared to laser-treated eyes.Fundus photographs from one patient in the obser-vation group demonstrating spontaneous drusenregression are shown in Figure 7.

Currently, many questions remain unansweredwith regard to the use of laser photocoagulation inpatients with high-risk drusen. Most of the studiesreviewed support the fact that drusen number isreduced in patients who receive laser treatments.Furthermore, it seems that the greater the intensity oftreatment, the faster the resolution. However, thisincreased intensity may also correlate with a higherrisk of CNV (90). The clinical significance of drusenreduction is also unclear at this time. While severalsmaller studies have demonstrated a correlationbetween drusen reduction and improvement invisual acuity, this finding has yet to be confirmed bya large, randomized, controlled study.

Another important finding from the CNVPT,PTAMD, and Drusen Laser Study has been theincreased risk of CNV in patients with neovascularAMD in one eye who underwent laser treatment inthe fellow eye with high-risk drusen. Moreover, theDrusen Laser Study demonstrated a six-month earlieronset of CNV in the laser-treated eyes compared withcontrols. As a result, these fellow eyes should not beconsidered for laser prophylaxis using these proto-cols. At this point, clinical studies on laserprophylaxis seem best reserved for patients withbilateral high-risk drusen in the absence of neovas-cular complications.

Based on the multitude of laser treatment regi-mens, no single method has proven superior. Drusenreduction has been found with varying wavelengths,burn intensities, and treatment area of laser. Also inquestion is whether laser should be applied to drusendirectly, indirectly, or both. As results become avail-able from the CAPT and PTAMD trials as well asother ongoing studies, the effect of laser prophylaxisand drusen reduction on the natural course of AMDshould become clarified.

SUMMARY POINTS

& A prophylactic treatment for AMD is highly desir-able and would have a significant publichealth impact.

& Laser photocoagulation can induce drusenregression.

& The long-term effect of laser-induced drusenreduction on the natural history of AMD andvisual function remains unclear.

& In patients with neovascular AMD in one eye,prophylactic laser appears to increase both therisk of CNV development as well as promote

earlier CNV development when performed in thefellow eye and should be avoided.

& Results from randomized clinical trials are necess-ary before laser prophylaxis for eyes with drusenshould be recommended outside this context.

REFERENCES

1. Kini MM, Leibowitz HM, Colton T, et al. Prevalance ofsenile cataract, diabetic retinopathy, senile maculardegeneration and open-angle glaucoma in the FraminghamEye Study. Am J Ophthalmol 1978; 85:28–34.

2. Leibowitz HM, Krueger DE, Maunder LR, et al. TheFramingham Eye Study monograph. Surv Ophthalmol1980; 24:335–610.

3. The Eye Diseases Prevalence Research Group. Causes andprevalence of visual impairment among adults in theUnited States. Arch Ophthalmol 2004; 122:477–85.

4. Sommer A, Tielsch JM, Katz J, et al. Racial differences inthe cause-specific prevalence of blindness in East Baltimore.N Engl J Med 1991; 14:1412–7.

5. Klein R, Klein BE, Linton KLP. Prevalence of age-relatedmaculopathy: the Beaver Dam Eye Study. Ophthalmology1992; 99:933–45.

6. National Eye Institute. Age-related Macular Degeneration:Status of Research. Bethesda, MD: National Eye Institute/National Institutes of Health/U.S. Department of Healthand Human Services, 1997.

7. Ferris FL. Senile macular degeneration: review of epide-miologic features. Am J Ophthalmol 1983; 118:132–51.

8. Ferris FL, Fine SL, Hyman L. Age-related macular degener-ation and blindness due to neovascular maculopathy. ArchOphthalmol 1984; 102:1640–2.

9. Macular Photocoagulation Study Group. Argon laserphotocoagulation for neovascular maculopathy after fiveyears. Results from randomized clinical trials. ArchOphthalmol 1991; 109:1109–14.

10. Macular Photocoagulation Study Group. Subfoveal neovas-cular lesions in age-related macular degeneration:guidelines for evaluation and treatment in the MacularPhotocoagulation Study. Arch Ophthalmol 1991; 109:1242–57.

11. Macular Photocoagulation Study Group. Laser photocoa-gulation for subfoveal neovascular lesions of age-relatedmacular degeneration: updated findings from two clinicaltrials. Arch Ophthalmol 1993; 111:1200–9.

12. Macular Photocoagulation Study Group. Laser photocoa-gulation for juxtafoveal choroidal neovascularization: fiveyear results from randomized clinical trials. ArchOphthalmol 1994; 112:500–9.

13. Treatment of Age-related Macular Degeneration withPhotodynamic Therapy (TAP) Study Group. Photodynamictherapy of subfoveal choroidal neovascularization in age-related macular degeneration with verteporfin: one-yearresults of 2 randomized clinical trials—TAP report 1. ArchOphthalmol 1999; 117:1329–45.

14. Treatment of Age-related Macular Degeneration withPhotodynamic Therapy (TAP) Study Group. Photodynamictherapy of subfoveal choroidal neovascularization in age-related macular degeneration with verteporfin: two-yearresults of 2 randomized clinical trials—TAP report 2. ArchOphthalmol 2001; 119:198–207.

15. Verteporfin in Photodynamic Therapy Study Group. Verte-porfin therapy of subfoveal choroidal neovascularization inage-related macular degeneration: two-year results of a

268 HSU AND HO

randomized clinical trial including lesions with occult withno classic choroidal neovascularization—verteporfin inphotodynamic therapy report 2. Am J Ophthalmol 2001;131:541–60.

16. Margherio RR, Margherio AR, DeSantis ME. Laser treat-ments with verteporfin therapy and its potential impact onretinal practices. Retina 2000; 20:325–30.

17. Gragoudas ES, Adamis AP, Cunningham ET, et al. Pegap-tanib for neovascular age-related macular degeneration. NEng J Med 2004; 351:2805–16.

18. Spaide RF, Sorenson J, Maranan L. Photodynamic therapywith verteporfin combined with intravitreal injection oftriamcinolone acetonide for choroidal neovascularization.Ophthalmology 2005; 112:301–4.

19. Lanchoney DM, Maguire MG, Fine SL. A model of theincidence and consequences of choroidal neovasculariza-tion secondary to age-related macular degeneration.Comparative effects of current treatment and potentialprophylaxis on visual outcomes in high-risk patients.Arch Ophthalmol 1998; 116:1045–52.


21. Bressler SB, Maguire MG, Bressler NM, et al. The MacularPhotocoagulation Study Group. Relationship of drusen andabnormalities of the retinal pigment epithelium to theprognosis of neovascular macular degeneration. ArchOphthalmol 1990; 108:1442–7.

22. Klein R, Klein BEK, Jensen SC, et al. The five-year incidenceand progression of age-related maculopathy. The BeaverDam Eye Study. Ophthalmology 1997; 104:7–21.

23. Gass JD. Drusen and disciform macular detachment anddegeneration. Arch Ophthalmol 1973; 90:206–17.

24. Bok D. Retinal photoceptor-pigment epithelium inter-actions. Invest Ophthalmol Vis Sci 1985; 26:1659–94.

25. Marshall J. Interactions between sensory cells, glial cellsand the retinal pigment epithelium and their response tophotocoagulation. Dev Ophthalmol 1981; 2:308–17.

26. Young R. The daily rhythm of shedding and degradation ofrod and cone outer segment membranes in the chick retina.Invest Ophthalmol Vis Sci 1978; 17:105–16.

27. Loeffler KU, Lee WR. Terminology of sub-RPE deposits: dowe all speak the same language? Br J Ophthalmol 1998;82:1104–5.

28. Foulds W. The choroidal circulation and retinal metab-olism: an overview. Eye 1990; 4:242–8.

29. Bird AC, Bressler NM, Bressler SB, et al. An internationalclassification and grading system for age-related maculo-pathy and age-related macular degeneration. SurvOphthalmol 1995; 39:367–74.

30. Bressler NM, Silva JC, Bressler SB, et al. Clinicopathologiccorrelation of drusen and retinal pigment epithelialabnormalities in age-related macular degeneration. Retina1994; 14:130–42.

31. Green WR, Enger C. Age-related macular degenerationhistopathologic studies: the 1992 Lorenz E. ZimmermanLecture. Ophthalmology 1993; 100:1519–35.

32. Curcio C, Millican C. Basal linear deposit and large drusenare specific for early age-related maculopathy. ArchOphthalmol 1999; 117:329–39.

33. Russell SR, Mullins RF, Schneider BL, et al. Location,substructure and composition of basal laminar drusencompared with drusen associated with aging and age-related macular degeneration. Am J Ophthalmol 2000;129:205–14.

34. Spraul C, Grossniklaus H. Characteristics of drusen andBruch’s membrane in post-mortem eyes with age-relatedmacular degeneration. Arch Ophthalmol 1997; 115:267–83.

35. van der Schaft TL, de Bruijn WC, Mooy CM, et al. Is basallaminar deposit unique for age-related macular degener-ation. Arch Ophthalmol 1991; 109:420–5.

36. Farkas TG, Syvlester V, Archer D. The ultrastructure ofdrusen. Am J Ophthalmol 1971; 71:1196–205.

37. Farkas TG, Sylvester V, Archer D, et al. The histochemistryof drusen. Am J Ophthalmol 1971; 71:1206–15.

38. Mullins R, Johnson L, Anderson D. Characterization ofdrusen-associated glycoconjugates. Ophthalmology 1997;104:288–94.

39. Sarks J, Sarks S, Killingsworth M. Evolution of soft drusenin age-related macular degeneration. Eye 1994; 8:269–83.

40. Ishibashi T, Sorgente N, Patterson R, et al. Pathogenesis ofdrusen in the primate. Invest Ophthalmol Vis Sci 1986;27:184–93.

41. Ishibashi T, Patterson R, Ohnishi Y, et al. Formation ofdrusen in the human eye. Am J Ophthalmol 1986;101:342–53.

42. Hogan MJ. Role of the retinal pigment epithelium inmacular disease. Trans Am Acad Ophthalmol Otolaryngol1972; 76:64–80.

43. Feeney-Burns L, Ellersieck M. Age-related changes in theultrastructure of Bruch’s membrane. Am J Ophthalmol1985; 100:686–97.

44. Pauleikhoff D, Zuels S, Sheraidah GS, et al. Correlationbetween biochemical composition and fluorescein bindingof deposits in Bruch’s membrane. Ophthalmology 1992;99:1548–53.

45. Moore D, Hussain A, Marshall J. Age-related variation inthe hydraulic conductivity of Bruch’s membrane. InvestOphthalmol Vis Sci 1995; 36:1290–7.

46. Pauleikhoff D, Harper CA, Marshall J, et al. Aging changesin Bruch’s membrane. A histochemical and morphologicstudy. Ophthalmology 1990; 97:171–8.

47. Zarbin M. Age-related macular degeneration: review ofpathogenesis. Eur J Ophthalmol 1998; 8:199–206.

48. Holz FG,Wolfensberger TJ, Piguet B, et al. Bilateral maculardrusen in age-related macular degeneration: prognosis andrisk factors. Ophthalmology 1994; 101:1522–8.

49. Macular Photocoagulation Study Group. Risk factors forchoroidal neovascularization in the second eye of patientswith juxtafoveal or subfoveal choroidal neovascularizationsecondary to age-related macular degeneration. ArchOphthalmol 1997; 115:741–7.

50. Sandberg MA, Weiner A, Miller S, et al. High-risk charac-teristics of fellow eye of patients with unilateralneovascular age-related macular degeneration. Ophthal-mology 1998; 105:441–7.

51. Sarraf D, Gin T, Yu F, et al. Long-term drusen study. Retina1999; 19:513–9.

52. Mainster M. Wavelength selection in macular photocoagu-lation. Ophthalmology 1986; 93:952–8.

53. Peyman G, RaichandM, Zeimer R. Ocular effects of variouslaser wavelenghts. Surv Ophthalmol 1984; 28:391–404.

54. Smiddy WE, Fine SL, Green WR, et al. Clinicopathologiccorrelation of krypton red, argon blue-green, and argongreen laser photocoagulation in the human fundus. Retina1984; 4:15–21.

55. Thomas EL, Apple DJ, Swartz M, Kavka-Van Norman D.Histopathology and ultrastructure of krypton and argonlaser lesions in a human retina-choroid. Retina 1984; 4:22–39.

56. Smiddy WE, Fine SL, Quigley HA, et al. Comparison ofkrypton and argon laser photocoagulation. ArchOphthalmol 1984; 102:1086–92.

57. Smiddy WE, Fine SL, Quigley HA, et al. Cell proliferationafter laser photocoagulation in primate retina. ArchOphthalmol 1986; 104:1065–9.


58. Peyman GA, Li M, Yoneya S, et al. Fundus photocoagula-tion with the argon and krypton lasers: a comparativestudy. Ophthalmic Surg 1981; 12:481–90.

59. Coscas G, Soubrane G. The effects of red krypton and greenargon laser on the foveal region. Ophthalmology 1983;90:1013–22.

60. Swartz M. Histology of macular photocoagulation.Ophthalmology 1986; 93:959–63.

61. Duvall J, Tso MO. Cellular mechanisms of resolution ofdrusen after laser coagulation. An experimental study. ArchOphthalmol 1985; 103:694–703.

62. Perry D, Reddick R, Risco J. Choroidal microvascular repairafter argon laser photocoagulation. Invest Ophthalmol VisSci 1984; 25:1019–26.

63. Della NG, Wilson DJ, Klein ML. Clinical and pathologiceffects of grid macular laser in aged primate eyescontaining drusen. Invest Ophthalmol Vis Sci 1997;38(Suppl.):S18.

64. Choroidal Neovascularization Prevention Trial ResearchGroup. Laser treatment in eyes with large drusen. Short-term effects seen in a pilot randomized clinical trial.Ophthalmology 1998; 105:11–23.

65. Cleasby G, Nakanishi A, Norris J. Prophylactic photocoa-gulation of the fellow eye in exudative senile maculopathy.Mod Probl Ophthalmol 1979; 20:141–7.

66. Figueroa MS, Regueras A, Bertrand J. Laser photocoagula-tion to treat macular soft drusen in age-related maculardegeneration. Retina 1994; 14:391–6.

67. Figueroa MS, Regueras A, Bertrand J. Laser photocoagula-tion for macular soft drusen. Updated results. Retina 1997;17:378–84.

68. Frennesson CI, Nilsson SE. Effects of argon (green) lasertreatment of soft drusen in early age-relatedmaculopathy: a6 month prospective study. Br J Ophthalmol 1995; 79:905–9.

69. Frennesson CI, Nilsson SE. Laser photocoagulation of softdrusen in early age-related maculopathy (ARM). The one-year results of a prospective, randomised trial. EurJ Ophthalmol 1996; 6:307–14.

70. Frennesson CI, Nilsson SE. Prophylactic laser treatment inearly age related maculopathy reduced the incidence ofexudative complications. Br J Ophthalmol 1998; 82:1169–74.

71. Guymer RH, Gross-Jendroska M, Owens SL, et al. Lasertreatment in subjects with high-risk clinical features of age-related macular degeneration. Posterior pole appearanceand retinal function. Arch Ophthalmol 1997; 115:595–603.

72. Ho AC, Maguire MG, Yoken J, et al. Laser-induced drusenreduction improves visual function at 1 year. ChoroidalNeovascularization Prevention Trial Research Group.Ophthalmology 1999; 106:1367–74.

73. Little HL, Showman JM, Brown BW. A pilot randomizedcontrolled study on the effect of laser photocoagulation ofconfluent soft macular drusen. Ophthalmology 1997;104:623–31.

74. Sarks SH, Arnold JJ, Sarks JP, Gilles MC, Walter CJ. Prophy-lactic perifoveal laser treatment of soft drusen. Aust N ZJ Ophthalmol 1996; 24:15–26.

75. Wetzig PC. Treatment of drusen-related aging maculardegeneration by photocoagulation. Trans Am OphthalmolSoc 1988; 86:276–90.

76. Wetzig PC. Photocoagulation of drusen-related maculardegeneration: a long-term outcome. Trans Am OphthalmolSoc 1994; 92:299–303.

77. Glaser BM, Campochiaro PA, Davis JL, Jr., et al. Retinalpigment epithelial cells release an inhibitor of neovascular-ization. Arch Ophthalmol 1985; 103:1870–5.

78. Bressler N, Munoz B, MaguireMG. Five-year incidence anddisappearance of drusen and retinal pigment epithelialabnormalities: Waterman study. Arch Ophthalmol 1995;113:301–8.

79. Peyman G, SpitznasM, Straatsma B. Chorioretinal diffusionof perioxidase before and after photocogulation. InvestOphthalmol Vis Sci 1971; 10:489.

80. Javornik N, Hiner CJ, Marsh MJ, et al. Changes in drusenand RPE abnormalities in age-related macular degener-ation. Invest Ophthalmol Vis Sci 1992; 33(Suppl.):1230.

81. Ho AC, Javornik N, Maguire MG, et al. The ChoroidalNeovascularization Prevention Trial (CNVPT): assessmentof macular risk factors for CNVat 1 year. Invest OphthalmolVis Sci 1997; 38(Suppl.):517.

82. Sigelman J. Foveal drusen resorption one year after perifo-veal laser photocoagulation. Ophthalmology 1991; 98:1379–83.

83. Hyver SW, Schatz H, McDonald HR, et al. A case of visualacuity loss following laser photocoagulation for maculardrusen. Arch Ophthalmol 1997; 115:554–5.

84. Ruiz-Moreno JM, De La Vega C, Zarbin MA. Macularatrophy after photocoagulation of soft drusen. Retina2003; 23:315–21.

85. Olk JR, Friberg TR, Stickney KL, et al. Therapeutic benefitsof infrared (810-nm) diode laser macular grid photocoagu-lation in prophylactic treatment of nonexudative age-related macular degeneration. Ophthalmology 1999;106:2082–90.

86. Scorolli L, Corazza D, Morara M, et al. Argon laser vs.subthreshold infrared (810-nm) diode laser macular gridphotocoagulation in nonexudative age-related maculardegeneration. Can J Ophthalmol 2003; 38:489–95.

87. Owens SL, Guymer RH, Gross-Jendroska M, et al. Fluor-escein angiographic abnormalities after prophylacticmacular photocoagulation for high-risk age-related macu-lopathy. Am J Ophthalmol 1999; 127:681–7.

88. Frennesson CI. Prophylactic laser treatment in early age-related maculopathy: an 8-year follow-up in a randomizedpilot study shows a reduced incidence of exudative compli-cations. Acta Ophthalmol Scand 2003; 81:449–54.

89. The Choroidal Neovascularization Prevention TrialResearch Group. Choroidal neovascularization in the chor-oidal neovascularization prevention trial. Ophthalmology1998; 105:1364–72.

90. Kaiser RS, Berger JW, Maguire MG, et al. Laser burnintensity and the risk for choroidal neovascularization inthe CNVPT Fellow Eye Study. Arch Ophthalmol 2001;119:826–32.

91. The Choroidal Neovascularization Prevention TrialResearch Group. Laser treatment in fellow eyes with largedrusen: updated findings from a pilot randomized clinicaltrial. Ophthalmology 2003; 110:971–8.

92. Owens SL, Bunce C, Brannon AJ, et al. Prophylactic lasertreatment appears to promote choroidal neovascularizationin high-risk ARM: results of an interim analysis. Eye 2003;17:623–7.

93. Owens SL, Bunce C, Brannon AJ, et al. Prophylactic lasertreatment hastens choroidal neovascularization in uni-lateral age-related maculopathy: final results of theDrusen Laser Study. Am J Ophthalmol 2006; 141:276–81.

94. Rodanant N, Friberg TR, Cheng L, et al. Predictors ofdrusen reduction after subthreshold infrared (810 nm)diode laser macular grid photocoagulation for nonexuda-tive age-related macular degeneration. Am J Ophthalmol2002; 134:577–85.

270 HSU AND HO

95. Friberg TR, Musch D. Prophylactic treatment of age-relatedmacular degeneration (PTAMD): update on the clinicaltrial. Invest Ophthalmol Vis Sci 2002; 43 (E-abstract 2904).

96. Complications of Age-Related Macular Degeneration Pre-vention Trial study group. The complications of age-relatedmacular degeneration prevention trial (CAPT): rationale,design and methodology. Clin Trials 2004; 1:91–107.

97. Complications of Age-Related Macular DegenerationPrevention Trial study group. Baseline characteristics, the25-item National Eye Institute visual functioning question-naire, and their associations in the Complications of Age-Related Macular Degeneration Prevention Trial (CAPT).Ophthalmology 2004; 111:1307–16.


Part V: Surgical Treatment for Age-Related Macular Degeneration

19

Macular TranslocationKah-Guan Au EongDepartment of Ophthalmology and Visual Sciences, Alexandra Hospital, Department of Ophthalmology,

Yong Loo Lin School of Medicine, National University of Singapore, The Eye Institute, National

Healthcare Group, Jurong Medical Center, Singapore Eye Research Institute, and Department


Dante J. PieramiciCalifornia Retina Research Foundation and California Retina Consultants,

Santa Barbara, California, U.S.A., and Doheny Eye Institute and Department

of Ophthalmology, Keck School of Medicine, University of Southern California, Los Angeles,

California, U.S.A.

Gildo Y. FujiiVitreous and Retina Department, State University of Londrina, Londrina, Parana, Brazil

Bakthavatsalu MaheshwarDepartment of Ophthalmology and Visual Sciences, Alexandra Hospital and Jurong Medical Center,

Singapore

Eugene de Juan, Jr.Beckman Vision Center, Department of Ophthalmology, University of California, San Francisco,

California, U.S.A.

INTRODUCTION

In recent years, new treatment modalities such asphotodynamic therapy and intravitreal anti-vascularendothelial growth factor (anti-VEGF) injections havebeen added to the armamentarium of physicianstreating age-related macular degeneration (AMD).Prior to the introduction of these therapies, the treat-ment options for AMDweremore limited. At that time,only laser photocoagulation had been shown in a largerandomized controlled trial to be effective for thetreatment of subfoveal choroidal neovascularization(CNV) secondary to AMD. This trial, the MacularPhotocoagulation Study, documented that laser photo-coagulation of subfoveal CNV confers a statisticallysignificant benefit with regard to long-term visualacuity (VA) when compared to the natural history ofthe condition (1–3).

Unfortunately, the Macular PhotocoagulationStudy also showed that the treatment of subfovealCNV with laser photocoagulation was associatedwith an immediate average reduction of three Bailey–Lovie lines and the benefits of treatment over notreatment only became apparent six months after thetreatment. In addition, retention or recovery of good

vision rarely occurred in patients treated with laserphotocoagulation. For these reasons, many physiciansworldwide did not use laser photocoagulation to treatsubfoveal CNV, even at a time when it was the onlytreatment that had been proven effective by a large,well-designed, randomized clinical trial. This is nicelyillustrated by a survey in 1999 of all consultant ophthal-mologists in the United Kingdom and the Republic ofIreland by Beatty et al., which showed that only 13.6%of 339 ophthalmologists whose practice includedlaser photocoagulation of CNV secondary to AMDstated that they ablated subfoveal CNV with laserphotocoagulation (4). The main reason (73.6%) theophthalmologists gave for withholding treatment wasthat they were not prepared to accept the likelihoodof an immediate drop in VA following laser ablation.

Investigators who pursued alternative therapysuch as interferon alpha-2a (5–8), radiation (9,10),subretinal endophotocoagulation (11), and subma-cular surgery (12–17) also had no or limited success.

As a result of the limited treatment options in thepre-photodynamic therapy era, a number of investi-gators approached the management of subfovealCNV with a totally new treatment paradigm. This

treatment is known by several names including retinalrelocation (18), retinal translocation (19,20), macularrelocation (21–23), macular translocation (24–29),macular rotation (30), and foveal translocation(31–35). The term macular translocation is currentlythe most widely used (36). The popularity of maculartranslocation was highest in the few years prior tothe introduction of photodynamic therapy in 2000 buthas waned in recent years due to the wider availabilityof photodynamic therapy and the introduction ofintravitreal anti-VEGF agents such as pegaptanibsodium, bevacizumab, and ranibizumab. However, itremains a potentially useful treatment option, and isstill in use in some countries including the UnitedStates, Japan, and Germany. This chapter reviewsthe current status of macular translocation, with anemphasis on the two more widely used techniques,limited macular translocation and macular transloca-tion with 3608 retinotomy.

CLASSIFICATION AND TERMINOLOGY

There are several different macular translocation tech-niques currently in use (36). These techniques producedifferent degrees of postoperative foveal displace-ment. The various forms of macular translocationmay be broadly classified into three categoriesdepending on the size of the retinotomy/retinotomiesused: (i) macular translocation with 3608 peripheralcircumferential retinotomy (21,22,24,37), (ii) maculartrans-location with large (but less than 3608) circum-ferential retinotomy (31–35), and (iii) maculartranslocation with either small (self-sealing) or no

retinotomy/retinotomies, with or without chorioscl-eral infolding or outfolding (Fig. 1) (19,20,23,28,38,39).Macular translocation with 3608 retinotomy is alsoknown as full macular translocation while anothername for macular translocation with either smallor no retinotomy/retinotomies is limited maculartranslocation.

RATIONALE

Although the exact pathogenesis of CNV secondary toAMD is not known, the natural history of this con-dition is progressive loss of central vision over time.The initial retinal dysfunction responsible forimpaired vision in eyes with subfoveal CNV may beattributable to factors such as subretinal fluid, subret-inal hemorrhage, and retinal edema. Accordingly,visual function may recover, at least partially,if these factors were removed. This improvement inmacular function has been substantiated by focalelectroretinography performed before and aftermacular translocation (40). When fibrous proliferationand degeneration of the overlying photoreceptorsoccur during the later stages of the disease, thevisual loss becomes irreversible.

The rationale of macular translocation is thatmoving the neurosensory retina of the fovea in aneye with recent-onset subfoveal CNV to a new locationbefore permanent retinal damage occurs may allowit to recover or maintain its visual function over ahealthier bed of retinal pigment epithelium (RPE)–Bruch’s membrane–choriocapillaris complex. Ineffect, macular translocation attempts to achieve

Macular Translocation (MT)

MT withPunctate or no

Retinotomy (Limited MT)

MT withLarge Curvilinear “Incisions”

of the Retina

MT withLarge

Retinotomy

WithChorioscleral

Shortening

MT with360-degree

Retinotomy (Full MT)

WithoutChorioscleralShortening

Chorioscleral Infolding(Imbrication or

Inpouching)

ChorioscleralOutfolding

(Outpouching)

Figure 1 Classification of macular translocation.

274 AU EONG ET AL.

a more normal subretinal space beneath the fovea. Inaddition, relocating the fovea to an area outside theborder of the CNVallows ablation of the CNV by laserphotocoagulation without destroying the fovea,thereby arresting the progression of the CNV andpreserving central vision.

Some surgeons have combined macular translo-cation with submacular surgery. Thomas et al. haveshown that removal of subfoveal CNV secondary toAMD is frequently accompanied by removal of nativeRPE, accounting for the relatively poorer visualoutcome of submacular surgery for AMD whencompared with that for other etiologies such asocular histoplasmosis syndrome (15). This is becausethe CNV in AMD typically lies in the sub-RPE spacebetween the RPE and Bruch’s membrane (type 1CNV), as opposed to that found anterior to thenative RPE in the sub-neurosensory retinal space(type 2 CNV) in eyes with ocular histoplasmosis,multifocal choroiditis, and idiopathic neovascularmembranes (41). When combined with removal ofCNV, macular translocation allows the fovea to berelocated to an area outside the RPE defect created.

HISTORICAL BACKGROUND

Lindsey et al. were the first to report their experimentwith retinal relocation in 1983, but their aim was tostudy the anatomic dependency of the foveal retina onfoveal RPE and choroid (18). Their techniquesincluded creation of a retinal detachment and relaxingretinal incisions, shifting of the neurosensory retinaand retinal reattachment. These techniques wereexpanded in 1985 by Tiedeman et al. who conceivedthe idea of rotating the macula of eyes with subfovealCNV to a new area of underlying RPE–Bruch’smembrane–choriocapillaris complex as a treatmentfor the condition (42). They showed it was feasible torotate the macula approximately 458 around the opticdisc with reattachment of the fovea in animal eyes.

After developing their surgical techniques inrabbit eyes (21), Machemer and Steinhorst in 1993became the first surgeons to demonstrate the feasi-bility of macular translocation in humans (22). Theirtechnique involves lensectomy, complete vitrectomy,planned total retinal detachment by transscleralinfusion of fluid under the retina, 3608 peripheralcircumferential retinotomy, rotation of the retinaaround the optic disc, and reattachment of theretina with silicone oil tamponade. Besides allowingretinal rotation to occur, the retinotomy also providedaccess to the subretinal space to remove blood andchoroidal neovascular membranes. A number ofinvestigators subsequently modified this technique,but many of them still require large or 3608 retinotomyto allow rotation of the retina (26,30,31,37).

The early reports of proliferative vitreoretino-pathy (PVR) complicating macular translocation withlarge retinotomy and 3608 retinotomy prompted Imaiand de Juan to develop a new technique withoutthe need for any retinotomy in 1996 (23). Theirtechnique involves transscleral subretinal hydrodis-section, anterior–posterior scleral shortening near theequator and retinal reattachment. Using this tech-nique, they were able to achieve a predictablemacular relocation of greater than 500 mm in rabbiteyes. As no retinal break was created, the likelihoodof developing PVR was thought to be lower thanthat with earlier techniques. As they gained moreexperience with the surgery, de Juan et al. madeadditional modifications to their original technique(19,20,28,38). A 41-gauge retinal hydrodissectioncannula is now used to make several tiny self-sealing retinotomies for subretinal hydrodissectionto create a controlled, reproducible subtotal retinaldetachment, and scleral resection during the scleralshortening procedure has been abandoned. Theycalled this technique limited macular translocationsince the operation achieves a smaller degree ofpostoperative foveal displacement and is less exten-sive compared with other techniques requiring largeor 3608 retinotomy (43).

To increase the redundancy of the detachedretina relative to the shortened eyewall, some inves-tigators have modified the technique of scleralshortening from chorioscleral infolding to outfolding.Kamei et al. work in an animal model (44) and aclinical trial (39,45) demonstrated that radialoutfolding with clips was a predictable and effectivemethod of limited macular translocation. Since radialoutfolding technique carries the risk of the choroidalfold affecting the macula, and because it is technicallydifficult to create a sufficiently long radial fold, thesurgeons have changed their technique from radialto diagonal outfolding. Other investigators have usednonabsorbable sutures instead of clips to effect theoutfolding (46).

INDICATIONS

Most surgeons use macular translocation to treatrecent-onset exudative macular degeneration. How-ever, some have also utilized it to treat nonexudativeAMD and subfoveal RPE loss following submacularsurgery.

Exudative Macular DegenerationThe most common application of macular transloca-tion is in the management of recent-onset subfovealCNV from a variety of etiologies. AMD is the mostcommon indication given the high prevalence of thiscondition, but subfoveal CNV due to other causes

19: MACULAR TRANSLOCATION 275

such as pathologic myopia, ocular histoplasmosissyndrome, angioid streaks, and multifocal choroiditis,as well as idiopathic neovascular membranes, havealso been treated with this procedure (20). Someauthors have reported better visual improvementafter limited macular translocation for CNV secondaryto pathologic myopia than for those due to AMD (47).It is possible to perform macular translocation forrecurrent subfoveal CNV that develops after laserphotocoagulation for initial nonsubfoveal CNV,although in such cases, the planned detachment ofthe macula is more difficult to achieve because thelaser scar causes the retina to be more adherent tothe underlying RPE (48).

Non-exudative Macular DegenerationA small number of surgeons have used maculartranslocation to treat atrophic AMD (49–51). In aseries of seven patients who had non-exudativeAMD treated with macular translocation with 3608retinotomy, five of the patients had improved distanceand near vision (49). One of these patients developed anew area of geographic atrophy in the translocatedfovea 12 months after surgery. This was similar toanother patient who had apparent continued pro-gression of geographic atrophy in the newlytranslocated macular region after effective maculartranslocation with 3608 retinotomy (51).

Subfoveal RPE DefectMacular translocation is a potentially useful remedyfor eyes with subfoveal RPE defect caused by subma-cular surgery. A case of a patient who underwentsuccessful limited macular translocation for subfovealRPE defect following submacular surgery for CNVsecondary to ocular histoplasmosis syndrome hasbeen described (52).

PREOPERATIVE CONSIDERATIONS

Proper case selection is crucial to good anatomic andfunctional outcome following macular translocation.A careful and detailed preoperative evaluation istherefore very important, and attention should bepaid to the characteristics of the lesion in the maculaas well as to concurrent pathology elsewhere in theretina. A recent good quality fluorescein angiogram,preferably obtained within one week of the surgery, isnecessary to evaluate the characteristics of the CNVand its precise relationship to the geometric center ofthe foveal avascular zone. If limited macular transloca-tion is planned, special care shouldbepaid to the retinalperiphery during indirect ophthalmoscopywith scleraldepression to look for concurrent peripheral retinalpathology that may lead to operative complications.

Several preoperative pathophysiologic andanatomic factors are important in determining thepostoperative functional and anatomic outcome ofpatients undergoing the procedure.

Pathophysiologic ConsiderationsSeveral pathophysiologic mechanisms responsible forvisual loss in eyes with subfoveal CNV may havesome bearing on the functional outcome followinglimited macular translocation. These factors may bebroadly divided into “reversible” and “irreversible”components.

"Reversible" Components of Visual Loss“Reversible” components of visual loss from sub-foveal CNV secondary to AMD include (i) impairedphotoreceptor function secondary to abnormal RPEfunction and impaired nutrient/waste exchangeacross the RPE and Bruch’s membrane, (ii) relativeretinal ischemia/ hypoxia secondary to abnormalRPE–Bruch’s membrane–choriocapillaris complex,(iii) retinal edema and subretinal fluid, and(iv) retinal and subretinal hemorrhages. Theseproblems may be evident early in the course of thedisease, resulting in metamorphopsia and centralblurring. Their effects are often not immediatelydevastating, and therefore affected eyes do notusually lose foveal fixation. Theoretically, effectivemacular translocation may, by reestablishing a rela-tively more normal subretinal space and underlyingRPE–Bruch’s membrane–choriocapillaris complex,cause one or more of these factors to be reduced orreversed, thereby allowing visual recovery. The bestcandidates for surgery are therefore those with recent-onset metamorphopsia or disturbance in central visiondue to new or recurrent CNV, before massive subret-inal fibrosis and degeneration of the photoreceptorspermanently destroy the fovea.

"Irreversible" Components of Visual LossUntreated long-standing subfoveal CNV often resultsin permanent photoreceptor cell loss, an “irreversible”mechanism responsible for visual loss. This usuallyoccurs in the late stages of the disease when there isfibrovascular scarring. Histopathologic studies havedocumented that the size and thickness of the disciformscar are directly related to the loss of photoreceptors(53). The visual loss associated with photoreceptor cellloss is often severe, but metamorphopsia becomesless prominent. Loss of foveal fixation may resultfrom the severe visual impairment. Such a severelyand irreversibly damaged foveal neurosensory retina isunlikely to achieve good functional recovery even aftersuccessful relocation to a healthier bed of RPE–Bruch’smembrane–choriocapillaris complex, and thereforeis a poor candidate for limited macular translocation.

276 AU EONG ET AL.

Proper case selection, by identifying patientswith good photoreceptor function for surgery andexcluding others with irreversible photoreceptordamage, is critically important to achieving goodvisual outcomes. The foveal function can be assessedpreoperatively by a number of means includingmeasurement of VA, scanning laser ophthalmoscope(SLO) microperimetry, and focal electroretinography(54). An analysis of a large series has shown thatpreoperative VA is a significant predictor of post-operative visual outcome, with good preoperativeVA being associated with better postoperative visualresults (55). However, eyes presenting with poorervision have a greater chance of visual improvementbut less likelihood of achieving excellent vision of20/40 or better.

SLO microperimetry appears to be a useful wayof identifying eyes that have viable foveal photo-receptors (54,56). It is particularly helpful inidentifying patients who have maintained centralfixation and may be a better indicator than VA inpredicting good visual outcome following maculartranslocation.

Anatomic ConsiderationsEffective macular translocation may be defined assuccessful postoperative relocation of the fovea to anarea outside the boundary of the lesion to be treated,i.e., when a more “normal” subfoveal space has beenestablished. In the case of CNV, effective maculartranslocation is the successful postoperative relocationof the fovea to an area outside the border of theCNV i.e., a previously subfoveal CNV becomes eitherjuxtafoveal (1 to 199 mm from the foveal center) orextrafoveal (R200 mmfrom the foveal center) followingthe surgery. If submacular surgerywere combinedwithmacular translocation, then effective macular trans-location is the successful postoperative relocation ofthe fovea to an area outside the border of the RPEdefectassociated with CNV removal during the surgery.

Barring any complication, the anatomic successof macular translocation is dependent on two majorfactors: (i) the minimum desired translocation and(ii) the postoperative foveal displacement achieved. Theminimum desired translocation can be measured priorto surgery and, when taken into consideration with themedian postoperative foveal displacement normallyachieved by the surgeon, can give some idea of thelikelihood of achieving effective macular translocationfollowing the surgery.

Minimum Desired TranslocationThe minimum amount of foveal displacementrequired to achieve effective macular translocation isthe distance between the foveal center and a pointeither on the inferior or superior border of the

subfoveal lesion depending on whether the transloca-tion is inferior or superior, all of these points beingequidistant from the temporal edge of the optic disc.This distance is the minimum desired translocation(Fig. 2). The temporal edge of the optic disc ratherthan the center of the disc is taken as the pivoting pointof the fovea because the papillomacular bundle entersthe optic disc from temporally close to this point. Thisis therefore the point in which the papillomacularbundle would pivot when the fovea is relocatedduring macular translocation.

Although the size of a subfoveal lesion is intui-tively a factor in determining the minimum desiredtranslocation, other factors such as eccentricity andshape of the lesion are important too. For example, ininferior macular translocation, a lesion that is eccen-trically centered superiorly relative to the fovea has asmaller minimum desired translocation and is morelikely to become juxtafoveal or extrafoveal followingsurgery compared with another lesion of the same sizewhich is eccentrically centered downwards relative tothe fovea, assuming that the net postoperative fovealdisplacement achieved is identical in both cases(Fig. 3). Lesions of the same size but of differentshapes may also have different minimum desiredtranslocations. On the other hand, lesions of differentsizes and eccentricities may have the same minimumdesired translocation (Fig. 4).

Median Postoperative Foveal DisplacementThe median postoperative foveal displacementnormally achieved by a surgeon can be derived by

F

I

D

Figure 2 Schematic diagram showing fundus of the left eye.F is the foveal center, D is a point on the temporal edge of the

optic disc, and I is a point on the inferior border of the subfoveallesion (circle) such that DFZDI. The distance FI is the minimumdesired translocation for an inferior translocation.


analyzing data collected either retrospectively orprospectively in a series of consecutive cases operatedby the surgeon. To estimate the amount of transloca-tion achieved, we first measure on the preoperativefluorescein angiogram the distance from a predeter-mined retinal landmark (such as a retinal vascularbifurcation) located superior to the CNV to a specificpoint along the inferior edge of the CNV. We then usethe same points to obtain a similar measurementon the postoperative angiogram. The differencebetween these two measurements estimates the post-operative foveal displacement achieved (Fig. 5). If thetime difference between the preoperative and post-operative angiograms is within two weeks, the sizeand characteristics of the CNV on the postoperativeangiogram tend not to change significantly. Althoughthis method of determining the postoperative fovealdisplacement is not very precise, especially for greateramounts of translocation, it does give useful estimateswithout the need to resort to sophisticated imagingequipment.

Ideally, a surgeon should have some idea of themedian postoperative foveal displacement he or shehas achieved in his or her previous cases whenevaluating potential patients for macular transloca-tion. This is particularly relevant for limited maculartranslocation. This information, when considered

a

b

FD

I

cF

I

D

FD

I

Figure 3 Schematic diagram showing the fundi of three eyes

with subfoveal lesions (circles a, b, and c) of equal size but differenteccentricities relative to the foveal center (F). Lesion a is centered

eccentrically upwards relative to the foveal center (F), lesion b iscentered on the foveal center (F), and lesion c is centered

eccentrically downwards relative to the foveal center (F). D is apoint on the temporal edge of the optic disc and I is a point on the

inferior border of the subfoveal lesions such that DFZDI. Theminimum desired translocation (FI) for inferior translocation is

smallest for lesion a and greatest for lesion c. Lesion a is thereforemore likely to achieve effective macular translocation compared

with lesions b and c following inferior macular translocation.

CB

A

F

I

D

Figure 4 Schematic diagram showing ocular fundus with threepossible subfoveal lesions (circles a, b, and c) of different sizes

and eccentricities. F is the foveal center, D is a point on thetemporal edge of the optic disc, and I is a point on the inferior

border of the subfoveal lesions such that DFZDI. The minimumdesired translocations (FI) for inferior translocation for lesions A,

B, and C are identical. Lesions A, B, and C therefore have thesame likelihood of achieving effective macular translocation

following inferior macular translocation. Note, however, that theminimum desired translocations for superior translocation for

lesions A, B, and C are different.

278 AU EONG ET AL.

together with the minimum desired translocation ofa particular eye, gives some useful idea of thelikelihood of achieving effective macular transloca-tion. If the minimum desired translocation in an eyeis equal to the median postoperative foveal displace-ment normally achieved by the surgeon, the eye hasan approximately 50:50 chance of achieving effectivemacular translocation after the surgery, regardless ofthe other dimensions of the subfoveal lesion. If theminimum desired translocation is less than themedian postoperative foveal displacement, the eyehas a greater than 50% chance of achieving effectivemacular translocation. The chance of effectivemacular translocation is less than 50% if theminimum desired translocation is greater thanthe median postoperative foveal displacement forthe surgeon. For example, if a surgeon has a post-operative foveal displacement greater than thepatient’s minimum desired translocation in 75% ofhis previous cases, he could then tell his patient thathe has an approximately 75% chance of effectivemacular translocation following surgery in hishands. If the macular translocation is combinedwith CNV removal, this rule may not apply if thearea of the RPE defect accompanying the CNVremoval differs greatly from the area of the original

CNV. This rule is more useful for limited maculartranslocation than for macular translocation with3608 retinotomy since large amounts of postoperativefoveal displacement are more readily achievedintraoperatively during the latter surgery. It isimportant to remember that the median postopera-tive foveal displacement for a particular surgeon isnot static and may change with modifications orrefinements in techniques.

OPERATIVE TECHNIQUE AND EARLYPOSTOPERATIVE MANAGEMENT

Limited Macular TranslocationSince the initial publications of the procedure(19,20,23), the technique has seen a number of modifi-cations to improve the amount of translocation and toreduce the incidence of complications (28,38,55).Unlike other techniques that require the creation oflarge retinotomies to allow foveal displacement(22,31), limited macular translocation relies on scleralinfolding or outfolding to shorten the outer eyewall(sclera, choroid, and RPE), creating redundancy of theneurosensory retina relative to the eyewall. Instead oflarge retinotomies, several small self-sealing posteriorretinotomies are used.

R

R'

D

F'

C'

D

(A) (B)

C

F

Figure 5 Schematic diagram showing the fundus of an eye (A) before and (B) after inferior macular translocation. R is a point on aretinal vascular bifurcation (“retinal” landmark) situated superior to the subfoveal lesion (circle). C is a point on the inferior border of the

subfoveal lesion (“choroidal” landmark) such that the line RC is close to and roughly parallel to the “path” of the foveal displacement. Fand F 0 are the foveal centers before and after macular translocation, respectively. R 0 and C 0 are the same “retinal” and “choroidal”

landmarks, respectively following macular translocation. The absolute difference between the distances RC and R 0C 0 estimates thepostoperative foveal displacement achieved.


Limited macular translocation may be eitherinferior or superior. Inferior limited macular transloca-tion causes inferior movement of the neurosensorymacula relative to the underlying tissues and viceversa. Our experience with this surgery is that inferiorlimited macular translocation achieves a greatermedian postoperative foveal displacement thansuperior translocation for the same amount of scleralimbrication used. When the patient’s head is uprightpostoperatively, the buoyancy of the intravitreal airbubble supports the superior retina while the weightof the subretinal fluid stretches the retina inferiorly.These forces probably contribute to the greater down-ward displacement of the fovea during inferiormacular translocation and reduce the upward displa-cement of the fovea during superior translocation. Forthis reason, inferior limited macular translocation ismore commonly performed than superior limitedmacular translocation, which may be done for theoccasional case in which the CNV is markedly eccen-trically centered inferiorly relative to the fovea. Thetechnique described below is for inferior limitedmacular translocation with chorioscleral infolding.

Overview/EquipmentInferior limited macular translocation is essentially afive-step procedure (Table 1). The first step is placementof scleral imbricating sutures. The second step is a 3-portpars plana vitrectomy with separation of the posteriorhyaloid face from the retina. The third step is creation ofa neurosensory retinal detachment, with or withoutsubretinal manipulation. The fourth step is tighteningof the scleral imbricating sutures. The final step in theprocedure is a subtotal fluid–air exchange.

The equipment necessary to perform thisprocedure includes a standard 3-port pars planavitrectomy equipment. Additional devices that areunique to this procedure include (i) a 41-gaugeretinal hydrodissection cannula (MADLAB retinalhydrodissection cannula, Bausch & Lomb Surgical,St. Louis, Missouri, U.S.A.) for subretinal hydrodissec-tion to create a detachment of the neurosensory retina(Fig. 6), (ii) a specially designed retinal manipulator(Bausch & Lomb Surgical) for gently grasping thedetached retina, aiding in the separation of themacular neurosensory retina from the RPE and also

permitting fluid–air exchange (Fig. 7), and (iii) a sub-retinal pick for subretinal dissection to break firmsubretinal adhesions. In addition, we use an airhumidifier (MoistAire humidifying chamber, Retina-Labs.com, Atlanta, Georgia, U.S.A.) that minimizesposterior capsular opacification in phakic patients(57) and potentially reduces excessive nerve fiberlayer dehydration during the fluid–air exchanges(Fig. 8).

Operative TechniquePlacement of Imbricating SuturesWe place three imbricating sutures in the superotem-poral quadrant between the superior and lateral recti,

Table 1 Key Surgical Steps of Limited Macular Translocation

Placement of imbricating sutures

Pars plana vitrectomy

Planned subtotal neurosensory retinal detachment

Tightening of imbricating sutures

Subtotal fluid–air exchange

Figure 6 Forty-one gauge retinal hydrodissection cannula(MADLAB retinal hydrodissection cannula, Bausch & Lomb

Surgical, St. Louis, Missouri, U.S.A.).

Figure 7 Retinal manipulator (Bausch & Lomb Surgical,

St. Louis, Missouri, U.S.A.). The tip of the instrument is enlargedto show the three small openings of the retinal manipulator.

280 AU EONG ET AL.

one suture just nasal to the superior rectus in thesuperonasal quadrant and one suture just inferior tothe lateral rectus in the inferotemporal quadrant(Fig. 9). The number and actual location of the

sutures have been selected empirically and are notbased on precise data. The purpose of the imbricatingsutures is to cause anterior–posterior shortening ofthe eyewall (sclera, choroid, and RPE) relative to theneurosensory retina. The sutures are placed in amattress fashion and we use the same nonabsorbablesutures used for scleral buckling, i.e., either 4-0 silk or5-0 dexon. The sutures are placed 6 mm apart from theanterior to posterior extent with the sutures straddlingthe equator. These sutures are not tightened until lateron in the procedure.

Pars Plana VitrectomyFollowing preplacement of the imbricating sutures,vitrectomy is initiated. We prefer to fit the scleros-tomies with metal cannulas for limited maculartranslocation because a “leaky” system is desirableduring the creation of retinal detachment whenbalanced salt solution is injected into the subretinalspace and during tightening of the imbricating sutureswhen the eye is deliberately kept soft. The metalcannula also facilitates the insertion of the delicate41-gauge retinal hydrodissection cannula. Otherwise,the delicate cannula may be easily damaged duringinsertion through a sclerostomy. A subtotal vitrectomyis then performed. It is critical in these cases to becertain that the posterior hyaloid face is separatedfrom the posterior pole, preferably up to the retinalperiphery but at least past the intended positions ofthe posterior retinotomies. It appears that when theposterior hyaloid face has not been separated from theneurosensory retina, it tethers the neurosensory retinaand reduces the amount of macular translocation. It isnot necessary to trim the vitreous gel down to thevitreous base but the vitreous cavity needs to bedebulked sufficiently to achieve a good air or gas fill.

Planned Neurosensory Retinal DetachmentTo detach the retina, three to eight retinotomies areusually necessary. The preferred locations of initialretinotomy placement, which are just superior to thesuperotemporal vascular arcade and just inferior tothe inferotemporal vascular arcade (Fig. 10). A thirdretinotomy is often necessary and is placed temporalto the macula (Fig. 11). The retinal detachments shouldbe relatively large and need to extend in the super-otemporal quadrant past the zone of intendedimbrication. The 41-gauge retinal hydrodissectioncannula is connected to an infusion pump to activelyinfuse balanced salt solution under the retina (Fig. 12).Prior to entering the vitreous cavity, the rate ofinfusion is set so that there is a steady dripof approximately two or three drops of balanced saltsolution per second from the cannula. To initiate thesubretinal blister, the 41-gauge retinal hydrodissectioncannula is placed through the retina with the infusion

Figure 8 Air humidifier (MoistAire humidifying chamber,RetinaLabs Inc, Atlanta, Georgia, U.S.A.).

LR SR

Figure 9 Nonabsorbable imbricating sutures are placed strad-

dling the equator of the globe prior to pars plana vitrectomy. Theanterior scleral bites are placed 3 mm posterior to the recti

insertion and the posterior scleral bites are placed 6 mm posteriorto the anterior bites. Three imbricating sutures are placed

between the SR and LR. The fourth imbricating suture isplaced medial to the SR and the final one is placed inferior to

the LR (not shown). Abbreviations: LR, lateral rectus; SR,superior rectus.


running. The neurosensory retinal detachment willinitially progress rapidly and tends to expandtowards the retinal periphery. As the blister becomeslarger, the expansion of the blister is slower althoughthe infusion rate remains constant. If the cannulainadvertently becomes dislodged from the retinotomyduring the procedure, one can usually reenter thesame retinotomy and continue with the detachment.If this is not possible, a new retinotomy can be made inanother site nearby. It is uncommon for the macula tobecome completely detached during this maneuversince the detachments have a tendency to progressanteriorly, presumably because the macula isrelatively more adherent to the RPE than the retinalperiphery.

The key to successful macular translocation is tocompletely detach the macula up to the temporal edgeof the optic disc. At the same time, limit the detachmentof the superonasal aspect of the retina because detach-ment of this area is associated with a higher risk ofmacular fold formation. The first step in completelydetaching themacula is to perform a complete fluid–airexchange. The subretinal fluid will gravitate poster-iorly and will usually dissect the macula off theunderlyingRPE (Fig. 13).When themacula is detached,the retinal bullae should extend to the optic nerve.However, this does not assure that all subretinal adhe-sions have been released. At this point, the air isexchanged for fluid, and inspection of the posteriorretina with the aid of the retinal manipulator willconfirm whether or not the macula is completelydetached. If adhesions are present, the retinal manipu-lator can be activatedwith low suction to grasp a part ofthe detached retina. Gentle traction is then exertedwiththe retinal manipulator to release any persistent sub-retinal adhesion (Fig. 14). Care should be taken whenusing the retinal manipulator as it may result iniatrogenic retinal breaks, hemorrhage, macular hole,

Figure 10 The first retinotomy for subretinal hydrodissection is

placed near the superotemporal vascular arcade to detach thesuperior retina.

Figure 11 The third retinotomy for subretinal hydrodissection isplaced a few disc diameters temporal to the fovea to detach the

temporal retina. The inferior retina had earlier been detached witha retinotomy placed near the inferotemporal vascular arcade.

Note that the retinal detachment from the first retinotomy extendsanteriorly beyond the zone of intended scleral imbrication.

BSS CNV

RetinaRPE

Sclera

Figure 12 The retina is detached by injecting BSS betweenthe neurosensory retina and the RPE with a 41-gauge

retinal hydrodissection cannula through a tiny retinotomy.Abbreviations: RPE, retinal pigment epithelium; CNV, choroidal

neovascularization; BSS, balanced salt solution.

282 AU EONG ET AL.

and nerve fiber layer injury (48). If despite thesemaneuvers, the retina is still not completely detached,a repeat fluid–air exchange can be performed.If repeated attempts fail to free a localized area ofsubretinal adhesion, such a laser scar, a small reti-notomy is created eccentrically in the macula throughwhich a retinal pick can be used to break the adhesions(Fig. 15).

Tightening of the Imbricating SuturesFollowing the neurosensory retinal detachment(Fig. 16), when there is still fluid in the vitreous

cavity, the imbricating sutures are tightened (Fig. 17).We tighten the sutures while the eye is filled with fluidrather than air to imbricate the eyewall under thebullous retina. Tightening the sutures while the eyeis filled with air may cause the retina lying onthe eyewall to be “caught” in the crevices of theimbrication and thus reduce the amount of retinalmovement relative to the eyewall. To achieveadequate imbrication, the globe should be softened

Figure 13 A complete fluid–air exchange allows the subretinal

fluid to gravitate posteriorly (white arrow) and dissect the maculaoff the underlying retinal pigment epithelium.

Retinal ManipulatorRetina

RPEChoroid

Sclera

Figure 14 Gentle traction on the retina (white arrows) with a

retinal manipulator helps to break abnormal chorioretinal adhe-sions and fully detach the macula from the retinal pigment

epithelium.

Retinotomy

Subretinal pick

Sclera

Figure 15 Subretinal blunt dissection (white arrows) with a pickthrough a small eccentric retinotomy may be necessary to break

abnormal chorioretinal adhesions in the macula.

Figure 16 A large retinal detachment temporal to an imaginaryvertical line bisecting the optic disc is obtained following coales-

cence of the multiple smaller localized retinal detachments. It isimportant to ensure that the macula is completely detached and

that the retinal detachment extends anteriorly beyond the zone ofintended scleral imbrication.


either by clamping the fluid infusion or leaving asclerostomy open or both. There is a theoretical riskthat this state of hypotony may increase the risk ofintraocular hemorrhage such as suprachoroidalhemorrhage.

Although we perform anterior–posterior short-ening of the eyewall with scleral imbrication in themajority of our cases of inferior limited maculartranslocation, it is interesting to note that this is notalways necessary, and effective macular translocationmay still be achieved without employing scleral imbri-cation for very small subfoveal lesions (28).

Subtotal Fluid–Air ExchangeThe sclerostomy sites and peripheral retina areinspected for inadvertent retinal breaks prior to thefinal fluid–air exchange. If present, they should betreated with laser retinopexy or cryoretinopexy and alonger-acting gas such as sulfur hexafluoride is thenused instead of air for internal tamponade.

The final fluid–air exchange is performedfollowing tightening of the imbricating sutures(Fig. 18). An estimated 75% to 90% air–fluid exchangeis carried out. The subretinal fluid is not completelydrained as this tends to result in a smaller amountof macular translocation. After the sclerostomiesand conjunctival incisions have been closed, a com-bination of corticosteroid–antibiotic subconjunctivalinjection is given. Intravenous corticosteroids may begiven during the procedure to reduce the incidenceof PVR.

Patient PositioningAfter the eye is patched, the patient is turned on theoperative side for about five minutes. This allowsthe subretinal fluid to gravitate temporally to detachthe temporal peripheral retina. From this position(without turning the patient on his or her back), thepatient sat upright and instructed to keep his or herhead upright overnight. Besides allowing the temporalperipheral retina to be completely detached, thismaneuver also causes all the subretinal fluid toaccumulate in the inferior retina, reducing the inci-dence of a postoperative macular or foveal fold(Fig. 19). If the superonasal retina has been inadver-tently detached during the surgery, sitting the patientupright from the supine position may cause somesubretinal fluid to become trapped under the super-onasal retina, causing a retinal bulla or retinal fold tooverhang fromthe superonasal retina. This bulla or foldwill often cause a retinal fold to stretch from thesuperior margin of the optic disc into the macula.When such amacular or foveal fold persists postopera-tively, undesirable visual consequences occur andremedial surgery is usually necessary to unfold themacula.

The buoyancy of the intravitreal air bubble whenthe patient’s head is upright, coupled with the weightof the subretinal fluid inferiorly, stretches the retina ina downward fashion (Fig. 20). The superior retina isthe first to become reattached, and this is quicklyfollowed by the macula and the rest of the retinaover the next several days.

LR SR

Figure 17 Tightening the imbricating sutures (white arrow)

causes the sclera to be imbricated under the detached retinaand creates redundancy of the retina relative to the eyewall

(sclera, choroid, and retinal pigment epithelium). Abbreviations:LR, lateral rectus; SR, superior rectus.

Air

Figure 18 Following scleral imbrication, a final subtotal fluid–airexchange is performed without draining the subretinal fluid.

284 AU EONG ET AL.

Combined Removal of CNV and LimitedMacular TranslocationSome surgeons have advocated surgically removingthe CNV at the time of limited macular translocation(27). We tend not to favor this approach, particularly inpatients with AMD, because of the uncertainty in thesize of the RPE defect that may occur. Thus, eventhough the preoperative CNV may be of a size andlocation that effective macular translocation wouldhave a good chance of being achieved, the RPEdefect created during submacular surgical excisionmay be significantly larger and therefore jeopardizethe chances of anatomic success. We feel that laserablation is a much more precise method of treatingthe CNV.

Postoperative Review, Fluorescein Angiography,and Laser PhotocoagulationOn the first postoperative day, the macula is typicallyattached, although there is often subretinal fluid inthe inferior retina. At this time, the presence of theintravitreal air bubble usually makes the fundus viewtoo poor for fluorescein angiography. The patientcontinues to position his head upright until theretina becomes completely attached. Completeretinal reattachment generally occurs within two tothree days. By three to seven days following theprocedure, the reduced air bubble no longer covers

Figure 19 The immediate postoperative head-positioning

maneuver (see text) causes all the subretinal fluid to accumulateunder the inferior retina. The inferior retina is detached. Note the

scleral imbrication (white arrows) and the fluid–air interface in thevitreous cavity (black arrows).

Figure 20 With the head in an upright position following thesurgery, the buoyancy of the air bubble supports the superior

retina (white arrows) while the weight of the subretinal fluidstretches the retina downwards (black arrow), causing the

fovea to be displaced downwards relative to the underlyingeyewall (sclera, choroid and retinal pigment epithelium).

Clinical ExampleA 63-year-old manwith a five-month history of

decreased vision in his left eye due to neovascularAMD presented for consideration of maculartranslocation surgery. His best-corrected VA atpresentation was 20/200K1. Clinical exam-ination and fluorescein angiography confirmeda subfoveal CNV approximately one MacularPhotocoagulation Study disc area in size(Fig. 21). After written informed consent wasobtained, inferior limited macular translocationwas performed without complication. Clinicalexamination and fluorescein angiography on thethird postoperative day disclosed effectiveinferior translocation of the fovea relative to theCNV (Fig. 22). The postoperative foveal displace-ment achieved was approximately 700 mm. Laserphotocoagulation was applied to the area of theCNV. The best-corrected VA improved to 20/60C2 and 20/40 at four and eight months, respect-ively after the surgery. He had no postoperativecomplication or recurrence of the CNV during thefollow-up period (Fig. 23).


the macula when the patient is upright. At this point, itis appropriate to consider fluorescein angiography soas to identify the postoperative location of the CNV.

Interpretation of the postoperative fluoresceinangiograms can be difficult in some cases given theadditional retinal pigment epithelial changes inducedby the surgical procedure. It is particularly importantto obtain good quality angiograms and compare themwith the preoperative angiograms to determine theactual location and extent of the CNV.

Laser photocoagulation of the entire CNV lesionis considered following effective macular translocationwhen the CNV no longer lies under the geometriccenter of the foveal avascular zone. We follow theguidelines for laser treatment outlined in theMacular Photocoagulation Study (58). Followinglaser photocoagulation, the patient will be followedup in about three to four weeks with repeat fluoresceinangiography to detect persistent or recurrent CNV.

Management of Persistent orRecurrent Subfoveal CNVWhen some parts of the CNVremains under the centerof the fovea due to insufficient macular translocationor when CNV recurs subfoveally after effectivemacular translocation and laser photocoagulation,the patient and physician must choose between anumber of options including photodynamic therapy,intravitreal anti-VEGF injections, laser ablation of thefovea, or observation. Successful treatment of CNVwith photodynamic therapy following insufficientmacular translocation has been reported (59). Theuse of intravitreal anti-VEGF injections followingmacular translocation has not been reported but itcould potentially be helpful. We do not advocatepartial laser treatment of the CNV because it hasbeen shown to be ineffective by the Macular Photo-coagulation Study Group (60). Repeated attempts ofmacular translocation are also not recommendedbecause initial efforts of this resulted in retinal detach-ment with significant PVR in some patients. One mustconsider that even when a CNV has not completelymoved out of the foveal center, the partial movementmay still benefit the patient as less of the perifovealretina will need laser ablation.

(A) (B)

Figure 21 (A) Fundus photograph and (B) fluorescein angiogram at presentation demonstrates asubfoveal choroidal neovascular membrane approximately one Macular Photocoagulation Study disc

area in size under the geometric center of the foveal avascular zone in the left eye. Visual acuity is20/200K1.

Figure 22 Three days following inferior limited macular translo-cation, fluorescein angiogram demonstrates effective macular

translocation with displacement of the geometric center of thefoveal avascular zone (arrow) to an area inferior to the choroidal

neovascular membrane. The postoperative foveal displacementis approximately 700 mm.

286 AU EONG ET AL.

Macular Translocation with 3608 RetinotomyMacular translocation with 3608 retinotomy requiresmoremanipulation than limitedmacular translocation,and is often combined with phacoemulsification orlensectomy with preservation of the anterior lenscapsule (Table 2) (43,61,62). After a near-completevitrectomy, the retina is detached totally with subret-inal infusion of balanced salt solution and a 3608peripheral circumferential retinotomy is performedwith a vitreous cutter or vertical scissors near the oraserrata. Removal of the CNVor drainage of subretinalhemorrhage, if desired, is then performed under directvisualization. Some perfluorocarbon liquid is theninjected onto the posterior retina after unfolding theretina. The retina is then rotated around its optic disk,usuallywith the fovea displaced superiorly.Additionalperfluorocarbon liquid is then injected to fill the vitr-eous cavity, followed by endolaser photocoagulation tothe retinal edges. Finally, the perfluorocarbon liquid isdirectly exchanged with silicone oil before the scler-ostomies and conjunctiva are closed. The silicone oil isremoved several months later, with or without anintraocular lens implantation. Corrective surgery forglobe counter-rotationmaybedoneduring theprimarysurgery (43) or at a later date.

MANAGEMENT OF POSTOPERATIVECYCLOVERTICAL DIPLOPIA

When the macula is moved sufficiently postopera-tively, cyclovertical diplopia or awareness of a tiltedimage may occur in some patients. This is because thedisplacement of the fovea is around the optic disk andnot directly upwards or downwards. This rotation ofthe retina can be measured using the Maddox rod testor the disk–fovea angle (63). This retinal torsion,coupled with the small range of fusional amplitudein the vertical direction, causes some patients toexperience cyclovertical diplopia after successfulmacular translocation. As the degree of foveal displa-cement is relatively smaller following limited maculartranslocation compared with macular translocationwith 3608 retinotomy, the incidence of postoperativecyclovertical diplopia is lower after limited maculartranslocation, and the symptoms may disappearspontaneously within a few months in many ofthese patients (27,43). For small degrees of cyclover-tical diplopia, correcting the vertical componentof deviation with vertical prism in glasses mayallow the cyclodeviation to be compensated by thesensory fusion ability, which is driven by the centralnervous system.

Three out of 10 patients who had limited maculartranslocation by Lewis et al. experienced either distor-tion or tilting of image postoperatively and thesesymptoms persisted for six months postoperativelyin only one patient (27). Ohtsuki et al. evaluated 20patients who underwent limited macular translocationand found seven of them (35%) experienced cyclover-tical diplopia postoperatively (64). They treated thesepatients with transposition of the anterior superioroblique insertion with or without additional vertical

Table 2 Key Surgical Steps of Macular Translocation with3608 Retinotomy

Phacoemulsification or pars plana lensectomy

Pars plana vitrectomy

Planned total neurosensory retinal detachment

3608 circumferential peripheral retinotomy

Retinal rotation and reattachment with perfluorocarbon liquid and

endolaser photocoagulation

Silicone oil exchange

(A) (B)

Figure 23 (A) Postoperative fundus photograph and (B) fluorescein angiogram shows successful laserablation of the choroidal neovascular membrane with no evidence of recurrence. The geometric center of

the foveal avascular zone (arrow) is preserved. Visual acuity is 20/40.


muscle surgery. Six of the seven patients (85.7%)became unaware of tilted image while three of them(42.9%) had successful restoration of single binocularvision at distance and near.

Unlike limited macular translocation, maculartranslocation with 3608 retinotomy creates large-angle ocular torsion (30). This large magnitude andsudden onset of torsion causes disorientation andhinders use of the eye. Extraocular muscle surgery isusually required to decrease or eliminate torsion andimproves the patient’s ability to function with thetranslocated retina. Several types of torsional musclesurgery for counter-rotation of the globe, sometimeswith additional muscle surgery on the fellow eye, havebeen developed to reduce this complication (30,65–67).Following extraocular muscle surgery in a series of 63patients who had undergone macular translocationwith 3608 retinotomy, Freedman et al. were able tomake 41% (23/63) of them free of both diplopia andtilt, while 5% (3/63) of patients had both symptomsconstantly (66).

OUTCOME

Although histopathologic analyses of a single humancase and three animal models have shown someminimal changes including a decreased photoreceptordensity in the retina after macular translocation(19,68–70). One advantage of macular translocationover many other experimental or established treat-ment is that it offers the potential for improvementin VA (20,27,38,55). While some surgeons have foundthe results of macular translocation encouraging insome cases (20,55,71), others have found the surgeryunpredictable (27,61).

The largest series of limited macular transloca-tion by Pieramici et al. analyzed the outcomes of 102consecutive eyes of 101 patients aged 41 to 89 years(median, 76 years) that underwent inferior transloca-tion by one surgeon for new or recurrent AMD-relatedsubfoveal CNV (55). The median postoperative fovealdisplacement achieved in the series was 1200 mm(range, 200–2800 mm). Seventy-five percent of thecases experienced at least 900 mm of postoperativefoveal displacement and 25% achieved 1500 mm ormore of foveal displacement. Sixty-two percent ofthe cases achieved effective macular translocation.At three and six months postoperatively, 31%and 49% of the eyes, respectively, achieved a VAbetter than 20/100 while 37% and 48% of the eyes,respectively, experienced R2 Snellen lines of visualimprovement. Sixteen percent of the eyes experiencedsix or more Snellen lines of visual improvement. In afollow-up report on the same cohort of patients, 39.5%of 86 eyes with one-year follow-up experienced R2

Snellen lines of visual improvement, 29.0% remainedunchanged, and 31.4% lost R2 lines of VA (72).

Pieramici et al. found that good preoperative VA,achieving the desired amount of postoperative fovealdisplacement, a greater amount of postoperativefoveal displacement and recurrent CNV at baselinewere associated with better VA at three and six monthspostoperatively (55). The reason patients with recur-rent CNV achieved better outcome was thought to bethat this select group of patients, having undergoneprevious laser photocoagulation for a juxtafoveal orextrafoveal lesion, were better educated about thenecessity to see their ophthalmologist for any newvisual change and were already on close follow-upby their ophthalmologist treating them. The subfovealdisease in this group of patients may therefore be of ashorter duration and less severe than those seen inpatients who never had prior laser photocoagulation.Poor preoperative VA and the development of acomplication either during or after surgery wereassociated with worse VA at three and six monthspostoperatively.

Ng et al. analyzed a consecutive series of 31 eyesof 29 patients who underwent limited macular trans-location for recurrent subfoveal CNV after laserphotocoagulation for initial nonsubfoveal CNVsecondary to AMD (77.4%) and a variety of otherpathologies (22.6%) (48). They achieved effectivemacular translocation in 77.4% of eyes. The post-operative foveal displacement ranged from 0 to2230 mm (median, 1100 mm). Preoperatively, the VAranged from 20/40 to counting fingers (median,20/160), and 19% of eyes had VA better than 20/100.At six months, 54% of eyes achieved a VA better than20/100, and 46% gained the equivalent of R2 EarlyTreatment Diabetic Retinopathy Study (ETDRS) lines.Subretinal dissection during the surgery to detach themacula was required in 25.8% of eyes and was associ-ated with a significantly higher incidence of peripheralretinal breaks. Retinal detachment occurred in 19.4%of eyes, but the retinal detachment rate observedbetween the groups with and without subretinaldissection was not statistically significant (pZ0.30).

By selecting only patients with subfoveal CNVthat did not extend more than half a disk diameterinferior to the fovea for inferior limited maculartranslocation, Morizane et al. was able to achieveeffective macular translocation in all 12 eyes thatunderwent the surgery (73). This is not unexpectedsince the small minimum desired translocation ismore likely to be associated with effective maculartranslocation (36). In this group of five patients withAMD and seven with polypoidal choroidal vasculo-pathy, the VA improved by R2 lines in 92% andremained within 1 line in 8%. In 58% of the eyes, thepostoperative VAwas 20/40 or better.

288 AU EONG ET AL.

In a small series of 10 eyes of 10 patients withsubfoveal CNV secondary to AMD treated by onesurgeon, the median postoperative foveal displace-ment achieved was 1286 mm (range, 114–1919 mm)(27). The best-corrected VA, as measured with theETDRS chart, improved in four eyes (median, 10.5letters) and decreased in six eyes (median, 14.5letters). The median change in VA for the entireseries was a decrease of five letters. The final VA atsix months postoperatively was 20/80 in two eyes,20/126 in one eye, 20/160 in four eyes, 20/200 in oneeye, 20/250 in one eye, and 20/640 in one eye.

Pawlak et al. compared the visual outcome oflimited macular translocation with photodynamictherapy for subfoveal predominantly classic CNV inAMD in a nonrandomized retrospective review of 65consecutive patients with follow-up of at least sixmonths (74). A total of 29 eyes were treated withphotodynamic therapy with verteporfin and 36 eyesunderwent limited macular translocation. Both groupswere similar for age, refraction, and lesion size, but theinitial VA was lower in the macular translocationgroup (20/200) than in the photodynamic therapygroup (20/100). At one year, both groups had thesame final VA (20/200), but the improvement wasmore favorable in the macular translocation group(gain of 0.7 line in the macular translocationgroup versus loss of 3.4 lines in the photodynamictherapy group, pZ0.007). In the photodynamictherapy group, 4.3% of eyes had a gain of 3 lines ormore versus 38% in the macular translocation group.

Using chorioscleral outfolding with titaniumclips for macular translocation, Kamei et al. achievedlarger postoperative foveal displacement with theirmodification of limited macular translocation thanhas been reported using chorioscleral infolding (39).They reported a median postoperative foveal displace-ment of 1576 mm (range, 349–3391 mm) in their series of27 eyes followed up for more than two years comparedwith the 1200 mm reported by Pieramici et al. (55). Inaddition, because their outfolding technique requiredshortening of only 2 to 2.5 mm compared with 4 to9 mm shortening for the infolding technique, there isless globe deformity and less induced corneal astig-matism (39).

It has been postulated that the scleral shorteningwith chorioscleral outfolding ought to be more than 12times larger than with infolding (75). However, clinicalstudies have found the difference in techniques toresult in a less profound difference in postoperativefoveal displacement (39,45). Histopathologic analysisof the scleral imbrication site in one patient hasrevealed pleating of the sclera rather than distinctinfolding (68). It is thought that this pleating of thesclera would reduce the scleral surface area at theimbrication site more than true infolding, thereby

explaining why the amount of postoperative fovealdisplacement is greater than expected with scleralinfolding relative to outfolding.

In a series of 50 consecutive eyes with subfovealCNV from AMD that underwent macular transloca-tion with 3608 retinotomy and followed up for amedian period of 21 months (range, 12–36 months),Pertile and Claes reported that the postoperative best-corrected VA improved by R2 Snellen lines in 66%,remained stable (G1 line) in 28%, and decreased byR2 lines in 6% of eyes (76). The final best-corrected VAwas 20/50 or better in 32% of eyes while 16% had afinal best-corrected VAworse than 20/200. In anotherlarge series by Mruthyunjaya et al. of 61 AMD patientswho underwent the same operation and followed upfor 12 months, all eyes had successful translocation,and the median VA improved from approximately20/125 before surgery to approximate 20/80 aftersurgery (77). The median reading speed also improvedfrom 71 words per minute before surgery to 105 wordsper minute at 12 months after surgery. At 12 months,the VA improved R1 line in 52% of patients.A Japanese study of 23 AMD patients also showed asignificant improvement in reading ability aftermacular translocation with 3608 retinotomy despitean absence of improvement in the distance and nearVA (78). Another report by Toth et al. also showedimprovements in distance VA, near-VA, contrastsensitivity, and reading speed in a series of 25consecutive AMD patients who underwent theprocedure (79).

Cahill et al. studied 50 patients’ quality oflife (QOL) after macular translocation with 3608retinotomy for AMD (80). They found that vision-related QOL, as measured by the 25-item NationalEye Institute Visual Function Questionnaire, improvedafter the surgery. Not surprising, the largest improve-ments in QOL were seen in patients with the greatestimprovement in visual function, and the best post-operative QOL was seen in patients with the bestpostoperative visual patients.

Park and Toth evaluated the outcome of eightpatients who underwent macular translocation with3608 retinotomy for CNV secondary to AMD followingat least one episode of photodynamic therapy withverteporfin (81). All of these patients had demon-strated continued visual loss following their mostrecent photodynamic therapy treatment. They foundthe final (mean follow-upZ10 months) mean VAchange for patients who had only one prior photo-dynamic therapy session (five eyes) was C10 ETDRSletters and those who had multiple photodynamictherapy sessions (three eyes) was K1 ETDRS letter.They concluded that macular translocation with3608 retinotomy may be a viable option to stabilize


vision for patients who continue to lose vision in theirbetter eye following photodynamic therapy.

In our experience, the most important aspects ofmacular translocation are patient selection, achievingthe desired amount of macular translocation andavoiding complications. If this procedure is performedon a patient without viable foveal photoreceptors,there is no chance for visual improvement. If theminimum desired translocation is not achieved, weare left with a persistent subfoveal CNV lesion thatwill likely result in continued photoreceptor celldamage and visual deterioration. Development of acomplication is associated with a poorer prognosis,particularly when retinal detachment occurs (55). Toimprove on the outcome of this surgery, care should betaken to select the appropriate patients and to reducethe incidence of complications.

COMPLICATIONS

The usual risks inherent to pars plana vitrectomy existfor all patients undergoing macular translocationsince posterior vitrectomy is an integral part of theprocedure. In addition, for patients who undergolimited macular translocation with chorioscleralinfolding, additional risks similar to those associatedwith scleral buckling surgery are presented (Table 3).Table 4 shows the intraoperative and postopera-tive complications documented in Pieramici et al.’series (55).

IntraoperativePlacement of sutures on the sclera for scleral imbrica-tion may cause inadvertent scleral perforation. Thismay be associated with suprachoroidal hemorrhage,

vitreous hemorrhage, and retinal break. Retinal breakcan also occur during the later stages of the operation.The retina may be inadvertently cut or traumatizedduring vitrectomy. Vitreous traction near the scleros-tomies, retinal incarceration at the sclerostomies andretinal manipulation during planned retinal detach-ment (27) may also tear the retina. Unintended retinalbreaks occurred in 10 out of 102 consecutive eyes inPieramici et al.’ series (55). Unintended nonself-sealing break(s) should receive laser retinopexy orcryoretinopexy during the surgery or in the earlypostoperative period. Longer-acting gas such assulfur hexafluoride may also be necessary for internaltamponade. Macular hole formation is another com-plication that may also require longer-term internaltamponade.

During planned detachment of the retina, sub-retinal hemorrhage may occur if the retinal hydro-dissection cannula used for subretinal hydrodissectionor the subretinal pick used for blunt dissection trauma-tizes the vascular choroid. Unplanned translocation ofthe RPE can occur when a patch of RPE adherent to theunderlying surface of the neurosensory retina detacheswith the retina (38).

While the eye is deliberately kept soft momenta-rily to allow the imbricating sutures to be tightened,the eye is at risk of retinal incarceration at the scler-ostomies and severe intraocular hemorrhage includingsuprachoroidal hemorrhage.

PostoperativeRhegmatogenous retinal detachment is the mostcommon serious complication of macular transloca-tion. Nine out of 102 eyes in Pieramici et al.’ seriesdeveloped persistent postoperative retinal detachment(55). Additional surgery is usually necessary to reat-tach the retina should this complication occur.Pneumoretinopexy may be effective in treating somecases with retinal breaks in the superior two-thirds ofthe retinal periphery. The retinal detachment may beassociated with PVR, especially if a repeat limitedmacular translocation has been performed for persist-ent subfoveal CNV. Postoperative endophthalmitis isanother potentially devastating complication.

The incidence of cataract formation appears to besimilar to that following other vitrectomy procedures.Should cataract formation occur soon after limitedmacular translocation such as following intraoperativelens touch, it can impair visualization of the funduspostoperatively and interfere with clinical exami-nation, fluorescein angiography, and laser photo-coagulation. Early cataract surgery is indicated insuch cases. Postoperative vitreous hemorrhage canalso impair visualization and close follow-up withultrasonography is warranted to look for associatedretinal detachment.

Table 3 Complications Associated with MacularTranslocation

Complications

Intraoperative Scleral perforation

Unplanned retinal break

Suprachoroidal hemorrhage

Subretinal hemorrhage

Vitreous hemorrhage

Macular hole

Unplanned translocation of retinal

pigment epithelium

Postoperative Rhegmatogenous retinal detachment

Proliferative vitreoretinopathy

Endophthalmitis

Cataract

Vitreous hemorrhage

Macular or foveal fold

New choroidal neovascularization at site

of retinotomy

Transient formed visual hallucinations

(Charles Bonnet syndrome)

290 AU EONG ET AL.

Folds running across the fovea are associatedwith poor vision, and reoperation to remove the foldmay be necessary. A foveal fold formed postopera-tively in 3 of 10 eyes reported by Lewis et al. (27).A single case of a macular fold that developed afterlimited macular translocation was successfully treatedwith release of the scleral imbrication and intravitrealgas injection four days after the initial surgery (82).Interestingly, effective macular translocation wasmaintained despite the release of the scleral short-ening. Presumably, the mechanism by which thetranslocated fovea did not return to its originalposition after scleral shortening released is the redun-dancy achieved by stretching of the neurosensoryretina (82).

Induced corneal astigmatism is another compli-cation of macular translocation. Not surprisingly,induced corneal astigmatism is more common afterlimitedmacular translocation than after macular trans-location with 3608 retinotomy because chorioscleralinfolding or outfolding deforms the globe (43).Between 1.75 and 7.37 diopters (mean, 4.6 diopters) ofcorneal astigmatismwas found in a small series of eighteyes after macular translocation with chorioscleralinfolding, with steepening along the axis of scleralshortening in the superotemporal quadrant of eacheye (83).

Rarely, new CNV can occur at the site of theretinotomy used for retinal detachment, presumablyas a result of iatrogenic focal defect in Bruch’smembrane caused by the retinal hydrodissectioncannula. A case of severe hypotony has been reportedafter macular translocation with 3608 retinotomy (84).Two cases of transient formed visual hallucinations(Charles Bonnet syndrome) developing within24 hours following limited macular translocationhave been reported (85). The visual hallucinationsceased completely three to seven days postoperativelyfollowing retinal reattachment and associated visualimprovement.

CONCLUSION

In this current era of photodynamic therapy andnewer anti-VEGF injections, it is likely that the popu-larity of macular translocation for the treatment ofCNV secondary to AMD will continue to wane.However, macular translocation does provide patientswith a realistic hope of a shorter and more definitivetreatment end-point when compared with photody-namic therapy and anti-VEGF injections wheremultiple retreatments are often necessary and theend-point of treatment sometimes uncertain. In eyesthat continue to lose vision following photodynamictherapy, macular translocation may be a viable optionto stabilize vision. In some countries where photo-dynamic therapy and anti-VEGF injections are eitherunavailable or yet to be approved, macular transloca-tion remains a useful option for the treatment ofcertain subsets of patients with CNV.

SUMMARY POINTS

& Rationale. To displace the foveal neurosensoryretina in an eye with recent-onset subfoveal CNVto a presumably healthier bed of RPE–Bruch’smembrane–choriocapillaris complex devoid ofCNV before permanent retinal damage occurs;the foveal displacement allows the destruction ofthe CNV by laser photocoagulation without dama-ging the foveal center.

& Indications. Subfoveal CNV secondary to a varietyof etiologies including exudative AMD. Somesurgeons have also performed the operation onnonexudative AMD.

& Classification of macular translocation. The variousforms of macular translocation may be broadlyclassified into three categories depending on thesize of the retinotomy/retinotomies used: (i)macular translocation with 3608 peripheral circum-ferential retinotomy, (ii) macular translocation

Table 4 Intra and Postoperative Complications Associated with Inferior Limited Macular Translocation in Pieramici and Associates’Series (NZ102)

Type of complication Intraoperative (no. of eyes) Postoperative (no. of eyes) Total (no. of eyes)

Macular hole 9 0 9

Scleral perforation 2 0 2

Choroidal hemorrhage 1 0 1

Subretinal hemorrhage 1 0 1

Unintended retinal break 6 4 10

Vitreous hemorrhage 2 2 4

Unplanned retinal detachment 0 9 9

Macular fold 0 3 3

New choroidal neovascularization

at site of retinotomy

0 2 2

Source: From Ref. 55.


with large (but less than 3608) circumferentialretinotomy, and (iii) macular translocation witheither small or no retinotomy/retinotomies.

& Complications. Intraoperative complications includescleral perforation, unplanned retinal break, intra-ocular hemorrhage, macular hole, unplannedtranslocation of RPE and postoperative includerhegmatogenous retinal detachment, PVR,endophthalmitis, cataract, intraocular hemorrhage,foveal fold, new CNV at site of retinotomy, acuteangle-closure glaucoma, and transient formedvisual hallucinations.

& Role of macular translocation in current era of photo-dynamic therapy and anti-VEGF therapy. Maculartranslocation offers a realistic hope of a shorterandmore definitive treatment end-point comparedwith photodynamic therapy and anti-VEGFtherapy where multiple retreatments are oftennecessary and the end-point of treatment some-times uncertain. However, because of its higherrisk of complications, its popularity has wanedsince the advent of photodynamic therapy. Itremains auseful optionwhen thesenewer therapiesare unavailable.

REFERENCES

1. Macular Photocoagulation Study Group. Laser photocoa-gulation of subfoveal neovascular lesions in age-relatedmacular degeneration: results of a randomized clinicaltrial. Arch Ophthalmol 1991; 109(9):1220–31.

2. Macular Photocoagulation Study Group. Laser photocoa-gulation of subfoveal recurrent neovascular lesions in age-related macular degeneration: results of a randomizedclinical trial. Arch Ophthalmol 1991; 109(9):1232–41.

3. Macular Photocoagulation Study Group. Laser photocoa-gulation of subfoveal neovascular lesions of age-relatedmacular degeneration: updated findings from two clinicaltrials. Arch Ophthalmol 1993; 111(9):1200–9.

4. Beatty S, Au Eong KG, McLeod D, Bishop PN. Photocoa-gulation of subfoveal choroidal neovascular membranes inage related macular degeneration: the impact of themacular photocoagulation study in the United Kingdomand Republic of Ireland. Br J Ophthalmol 1999;83(10):1103–4.

5. Pharmacological Therapy for Macular Degeneration StudyGroup. Interferon Alfa-2a is ineffective for patients withchoroidal neovascularization secondary to age-relatedmacular degeneration: results of a prospective randomizedplacebo-controlled clinical trial. Arch Ophthalmol 1997;115(7):865–72.

6. Thomas MA, Ibanez HE. Interferon alfa-2a in the treat-ment of subfoveal choroidal neovascularization. AmJ Ophthalmol 1993; 115(5):563–8.

7. Poliner LS, Tornambe PE, Michelson PE, Heitzmann JG.Interferon alpha-2a for subfoveal neovascularization in age-related macular degeneration. Ophthalmology 1993;100(9):1417–24.

8. Chan CK, Kempin SJ, Noble SK, Palmer GA. The treatmentof choroidal neovascular membranes by alpha interferon.Ophthalmology 1994; 101(2):289–300.

9. Spaide RF, Guyer DR, McCormick B, et al. External beamradiation therapy for choroidal neovascularization.Ophthalmology 1998; 105(1):24–30.

10. Char DH, Irvine AI, Posner MD, Quivey J, Phillips TL,Kroll S. Randomized trial of radiation for age-relatedmacular degeneration. Am J Ophthalmol 1999;127(5):574–8.

11. Thomas MA, Ibanez HE. Subretinal endophotocoagulationin the treatment of choroidal neovascularization. AmJ Ophthalmol 1993; 116(9):279–85.

12. Lambert HM, Capone AJ, Aaberg TM, Sternberg PJ,Mandell BA, Lopez PF. Surgical excision of subfovealneovascular membranes in age-related macular degener-ation. Am J Ophthalmol 1992; 113(3):257–62.

13. Thomas MA, Grand MG, Williams DF, Lee CM, Pesin SR,Lowe MA. Surgical management of subfoveal choroidalneovascularization. Ophthalmology 1992; 99(6):952–68.

14. Berger AS, Kaplan HJ. Clinical experience with the surgicalremoval of subfoveal neovascular membranes. Ophthal-mology 1992; 99(6):969–76.

15. Thomas MA, Dickinson JD, Melberg NS, Ibanez HE,Dhaliwal RS. Visual results after surgical removal ofsubfoveal choroidal neovascular membranes. Ophthal-mology 1994; 101(8):1384–96.

16. de Juan E, Machemer R. Vitreous surgery for hemorrhagicand fibrous complications of age-related macular degene-ration. Am J Ophthalmol 1988; 105(1):25–9.

17. Blinder KJ, Peyman GA, Paris CL, Gremillion CM. Subma-cular scar excision in age-related macular degeneration. IntOphthalmol 1991; 15(4):215–22.

18. Lindsey P, Finkelstein D, D’Anna S. Experimental retinalrelocation. ARVO abstracts. Invest Ophthalmol Vis Sci 1983;24(Suppl.):242.

19. Imai K, Loewenstein A, de Juan E. Translocation of theretina for management of subfoveal choroidal neovascular-ization I: experimental studies in the rabbit eye. AmJ Ophthalmol 1998; 125(5):627–34.

20. de Juan E, Loewenstein A, Bressler NM, Alexander J.Translocation of the retina for management of subfovealchoroidal neovascularization II: a preliminary report inhumans. Am J Ophthalmol 1998; 125(5):635–46.

21. Machemer R, Steinhorst UH. Retinal separation, reti-notomy, and macular relocation: I. Experimental studiesin the rabbit eye. Graefes Arch Clin Exp Ophthalmol 1993;231(11):629–34.

22. Machemer R, Steinhorst UH. Retinal separation, reti-notomy, and macular relocation: II. A surgical approachfor age-related macular degeneration? Graefe’s Arch ClinExp Ophthalmol 1993; 231(11):635–41.

23. Imai K, de Juan E. Experimental surgical macular relocationby scleral shortening. ARVO abstracts. Invest OphthalmolVis Sci 1996; 37(Suppl.):S116.

24. Seaber JH, Machemer R. Adaptation to monocular torsionafter macular translocation. Graefe’s Arch Clin ExpOphthalmol 1997; 235(2):76–81.

25. Machemer R. Macular translocation. Am J Ophthalmol1998; 125(5):698–700 (Editorial).

26. Wolf S, Lappas A, Weinberger AWA, Kirchhof B. Maculartranslocation for surgical management of subfoveal chor-oidal neovascularizations in patients with AMD: firstresults. Graefe’s Arch Clin Exp Ophthalmol 1999;237(1):51–7.

27. Lewis H, Kaiser PK, Lewis S, Estafanous M. Maculartranslocation for subfoveal choroidal neovascularizationin age-related macular degeneration: a prospective study.Am J Ophthalmol 1999; 128(2):135–46.

292 AU EONG ET AL.

28. de Juan E, Vander JF. Effective macular translocationwithout scleral imbrication. Am J Ophthalmol 1999;128(3):380–2.

29. Akduman L, Karavellas MP, MacDonald CJ, Olk RJ,Freeman WR. Macular translocation with retinotomy andretinal rotation for exudative age-related macular degener-ation. Retina 1999; 19(5):418–23.

30. Eckardt C, Eckardt U, Conrad H-G. Macular rotation withand without counter-rotation of the globe in patients withage-related macular degeneration. Graefes Arch Clin ExpOphthalmol 1999; 237(4):313–25.

31. Ninomiya Y, Lewis JM, Hasegawa T, Tano Y. Retinotomyand foveal translocation for surgical management of subfo-veal choroidal neovascular membranes. Am J Ophthalmol1996; 122(5):613–21.

32. Fujikado T, Ohji M, Saito Y, Hayashi A, Tano Y. Visualfunction after foveal translocation with scleral shortening inpatients with myopic neovascular maculopathy. AmJ Ophthalmol 1998; 125(5):647–56.

33. Fujikado T, Ohji M, Hayashi A, Kusaka S, Tano Y. Anatomicand functional recovery of the fovea after foveal transloca-tion surgery without large retinotomy and simultaneousexcision of a neovascular membrane. Am J Ophthalmol1998; 126(6):839–42.

34. Ohji M, Fujikado T, Saito Y, Hosohata J, Hayashi A, Tano Y.Foveal translocation: a comparison of two techniques.Semin Ophthalmol 1998; 13(1):52–61.

35. Cekic O, Ohji M, Hayashi A, Fujikado T, Tano Y. Fovealtranslocation surgery in age-related macular degeneration.Lancet 1999; 354(9175):340.

36. Au Eong KG, Pieramici DJ, Fujii GY, et al. Maculartranslocation: unifying concepts, terminology, and classi-fication. Am J Ophthalmol 2001; 131(2):244–53.

37. Toth CA, Machemer R. Macular translocation. In:Berger JW, Fine SL, Maguire MG, eds. Age-RelatedMacular Degeneration. Philadelphia, PA: Mosby Inc.,1999:353–62.

38. Harlan JB, de Juan E, Bressler NM. Retinal translocationwith unplanned translocation of the retinal pigment epi-thelium. The Wilmer Retina Update 1999; 5(1):3–8.

39. Kamei M, Tano Y, Yasuhara T, Ohji M, Lewis H. Maculartranslocation with chorioscleral outfolding: 2-year results.Am J Ophthalmol 2004; 138(4):574–81.

40. Potter MJ, Chang TS, Lee AS, Rai S. Improvement inmacular function after retinal translocation surgery in apatient with age-related macular degeneration. AmJ Ophthalmol 2000; 129(4):547–9.

41. Gass JDM. Biomicroscopic and histopathologic consider-ations regarding the feasibility of surgical excision ofsubfoveal neovascular membranes. Am J Ophthalmol1994; 118(3):285–98.

42. Tiedeman J, de Juan E, Machemer R, Hatchell DL,Hatchell MC. Surgical relocation of the macula. ARVOabstracts. Invest Ophthalmol Vis Sci 1985; 26(Suppl.):59.

43. Ohji M, Fujikado T, Kusaka S, et al. Comparison of threetechniques of foveal translocation in patients with subfo-veal choroidal neovascularization resulting from age-related macular degeneration. Am J Ophthalmol 2001;132(6):888–96.

44. Kamei M, Roth DB, Lewis H. Macular translocation withchorioscleral outfolding: an experimental study. AmJ Ophthalmol 2001; 132(2):149–55.

45. Lewis H. Macular translocation with chorioscleraloutfolding: a pilot clinical study. Am J Ophthalmol 2001;132(2):156–63.

46. Lin SB, Glaser BM, Gould D, Baudo TA, Lakhanpal RR,Murphy RP. Scleral outfolding for macular translocation.Am J Ophthalmol 2000; 130(1):76–81.

47. Glacet-Bernard A, Simon P, Hemelin N, Coscas G,Soubrane G. Translocation of the macula for manage-ment of subfoveal choroidal neovascularization: com-parison of results in age-related macular degenerationand degenerative myopia. Am J Ophthalmol 2001;131(1):78–89.

48. Ng EWM, Fujii GY, Au Eong KG, et al. Macular transloca-tion in patients with recurrent subfoveal choroidalneovascularisation after laser photocoagulation for nonsub-foveal choroidal neovascularization. Ophthalmology 2004;111(10):1889–93.

49. Eckardt C, Eckardt U. Macular translocation in nonexuda-tive age-related macular degeneration. Retina 2002;22(6):786–94.

50. Cahill MT, Freedman SF, Toth CA. Macular translocationwith 360 degrees peripheral retinectomy for geographicatrophy. Arch Ophthalmol 2003; 121(1):132–3.

51. Khurana RN, Fujii GY, Walsh AC, HumayunMS, de Juan E,Sadda SR. Rapid recurrence of geographic atrophy after fullmacular translocation for nonexudative age-relatedmaculardegeneration. Ophthalmology 2005; 112(9):1586–91.

52. Fujii GY, de Juan E, Thomas MA, Pieramici DJ,HumayunMS, Au Eong KG. Limited macular translocationfor the management of subfoveal retinal pigment epithelialloss after submacular surgery. Am J Ophthalmol 2001;131(2):272–5.

53. Green WR, Enger C. Age-related macular degenerationhistopathologic studies. The 1992 Lorenz E Zimmermanlecture. Ophthalmology 1993; 100(10):1519–35.

54. Fujii GY, de Juan E, Sunness J, Humayun MS, Pieramici DJ,Chang TS. Patient selection for macular translocationsurgery using the scanning laser ophthalmoscope. Ophthal-mology 2002; 109(9):1737–44.

55. Pieramici DJ, de Juan E, Fujii GY, et al. Limited inferiormacular translocation for the treatment of subfoveal chor-oidal neovascularization secondary to age-related maculardegeneration. Am J Ophthalmol 2000; 130(4):419–28.

56. Loewenstein A, Sunness JS, Bressler NM, Marsh MJ, deJuan E. Scanning laser ophthalmoscope fundus perimetryafter surgery for choroidal neovascularization. AmJ Ophthalmol 1998; 125(5):657–65.

57. Harlan JB, Lee ET, Jensen PS, de Juan E. Effect of humidityon posterior lens opacification during fluid–air exchange.Arch Ophthalmol 1999; 117(6):802–4.

58. Macular Photocoagulation Study Group. Krypton laserphotocoagulation for neovascular lesions of age-relatedmacular degeneration: results of a randomized clinicaltrial. Arch Ophthalmol 1990; 108(6):816–24.

59. Sawa M, Chan W-M, Ohji M, et al. Successful photo-dynamic therapy with verteporfin for recurrent choroidalneovascularization beneath the new fovea aftermacular translocation with 360-degree retinotomy. AmJ Ophthalmol 2003; 136(3):560–3.

60. Macular Photocoagulation Study Group. Occult choroidalneovascularization: influence on visual outcome in patientswith age-related macular degeneration. Arch Ophthalmol1996; 114(4):400–12.

61. American Academy of Ophthalmology. Macular transloca-tion. Ophthalmology 2000; 107(5):1015–8.

62. Terasaki H, Ishikawa K, Suzuki T, Nakamura M, Miyake K,Miyake Y. Morphologic and angiographic assessment ofthe macula after macular translocation with 360-degreeretinotomy. Ophthalmology 2003; 110:2403–8.


63. Freedman SF, Gearinger MD, Enyedi LB, Holgado S,Toth CA. Measurement of ocular torsion after maculartranslocation: disc fovea angle and Maddox rod. J AAPOS2003; 7(2):103–7.

64. Ohtsuki H, Shiraga F, Morizane Y, Furuse T, Takasu I,Hasebe S. Transposition of the anterior superior obliqueinsertion as a treatment for excyclotorsion induced fromlimited macular translocation. Am J Ophthalmol 2004;137(1):125–34.

65. Freedman SF, Rojas M, Toth CA. Strabismus surgery forlarge-angle cyclotorsion after macular translocationsurgery. J AAPOS 2002; 6(3):154–62.

66. Freedman SF, Holgado S, Enyedi LB, Toth CA.Managementof ocular torsion and diplopia after macular translocationfor age-related macular degeneration: prospective clinicalstudy. Am J Ophthalmol 2003; 136(4):640–8.

67. Holgado S, Enyedi LB, Toth CA, Freedman SF. Extraocularmuscle surgery for extorsion after macular translocationsurgery: new surgical technique and clinical management.Ophthalmology 2006; 113(1):63–9.

68. Albini TA, Rao NA, Li A, Craft CM, Fujii GY, de Juan E.Limited macular translocation: a clinicopathologic casereport. Ophthalmology 2004; 111(6):1209–14.

69. Roig-Melo EA, Alfaro DVI, Heredia-Elizondo ML, et al.Macular translocation: histopathologic findings in swineeyes. Eur J Ophthalmol 2000; 10(4):297–303.

70. Fang X, Hayashi A, Morimoto T, et al. Retinal changes aftermacular translocation with 360-degree retinotomy inmonkey eyes. Am J Ophthalmol 2004; 137(6):1034–41.

71. Au Eong KG. Initial experience of macular translocation inSingapore—one-year results. Ann Acad Med Singapore2004; 33(5):641–8.

72. Fujii GY, de Juan E, Pieramici DJ, et al. Inferior limitedmacular translocation for subfoveal choroidal neovascular-ization secondary to age-related macular degeneration:1-year visual outcome and recurrence report. AmJ Ophthalmol 2002; 134(1):69–74.

73. Morizane Y, Shiraga F, Takasu I, Yumiyama S, Okanouchi T,Ohtsuki H. Selection for inferior limited macular transloca-tion on the basis of distance from the fovea to the inferioredge of the subfoveal choroidal neovascularization. AmJ Ophthalmol 2002; 133(6):848–50.

74. Pawlak D, Glacet-Bernard A, Papp M, Roquet W, Coscas G,Soubrane G. Limited macular translocation compared with

photodynamic therapy in the management of subfovealchoroidal neovascularization in age-related maculardegeneration. Am J Ophthalmol 2004; 137(5): 880–7.

75. Sullivan P, Filsecker L, Sears J. Limited macular transloca-tion with scleral retraction suture. Br J Ophthalmol 2002;86(4):434–9.

76. Pertile G, Claes C. Macular translocation with 360 degreeretinotomy for management of age-related maculardegeneration with subfoveal choroidal neovascularization.Am J Ophthalmol 2002; 134(4):560–5.

77. Mruthyunjaya P, Stinnett SS, Toth CA. Change in visualfunction after macular translocation with 360 retinectomyfor neovascular age-related macular degeneration.Ophthalmology 2004; 111(9):1715–24.

78. Fujikado T, Asonuma S, Ohji M, et al. Reading ability aftermacular translocation surgery with 360-degree retinotomy.Am J Ophthalmol 2002; 134(6):849–56.

79. Toth CA, Lapolice DJ, Banks AD, Stinnett SS. Improvementin near visual function after macular translocation surgerywith 360-degree peripheral retinectomy. Graefe’s Arch ClinExp Ophthalmol 2004; 242(2):541–8.

80. Cahill MT, Stinnett SS, Banks AD, Freedman SF, Toth CA.Quality of life after macular translocation with 360 periph-eral retinectomy for age-related macular degeneration.Ophthalmology 2005; 112(1):144–51.

81. Park CH, Toth CA. Macular translocation surgery with360-degree peripheral retinectomy following ocular photo-dynamic therapy of choroidal neovascularization. AmJ Ophthalmol 2003; 136(5):830–5.

82. Kadonosono K, Takeuchi S, Iwata S, Uchio E, Itoh N,Akura J. Macular fold after limited macular translocationtreated with scleral shortening release and intravitreal gas.Am J Ophthalmol 2001; 132(5):790–2.

83. Kim T, Krishnasamy S, Meyer CH, Toth CA. Inducedcorneal astigmatism after macular translocation surgerywith scleral infolding. Ophthalmology 2001; 108(7):1203–8.

84. Ichibe M, Yoshizawa T, Funaki S, et al. Severe hypotonyafter macular translocation surgery with 360-degree reti-notomy. Am J Ophthalmol 2002; 134(1):139–41.

85. Au Eong KG, Fujii GY, Ng EWM, Humayun MS,Pieramici DJ, de Juan E. Transient formed visual hallucina-tions following macular translocation for subfovealchoroidal neovascularization secondary to age-relatedmacular degeneration. Am JOphthalmol 2001; 131(5):664–6.

294 AU EONG ET AL.

20

Age-Related Macular Degeneration: Use of Adjuncts inSurgery and Novel Surgical ApproachesRichard Scartozzi and Lawrence P. ChongDoheny Retina Institute of the Doheny Eye Institute, Keck School of Medicine,


INTRODUCTION

Adjuncts that have been used in surgery for age-relatedmacular degeneration (AMD) include tissue plasmi-nogen activator (tPA), balance salt solution (BSS),and calcium- and magnesium-free retinal detach-ment-enhancing solutions. The surgeries in whichthese solution have been used include submacularsurgery to excise choroidal neovascular membranes,large-scale macular translocation surgery, limitedmacular translocation surgery, evacuation, or displace-ment of submacular hemorrhages. In addition to theseadjuncts, triamcinolone acetonide (TA) has beeninjected into the subretinal space for the treatment ofchoroidal neovascular membranes.

Novel surgical approaches include the surgicalimplantation of sustained release drug devices, thesurgical implantation of cell-based delivery systems,and the pre-retinal or subretinal delivery of radiationtherapy through a pars plana approach.

TISSUE PLASMINOGEN ACTIVATOR

tPA is a polypeptide of 527 amino acids that cleavesthe Arg560-Val561 bond of plasminogen. Because of itshigh affinity for fibrin, its enhancement of binding ofplasminogen to fibrin clots, and potentiation of itsactivity in the presence of fibrin, fibrinolysis occursalmost exclusively in fibrin clots.

Commercial tPA (Activase, Genentech, Inc.; Acti-lyse, Boehringer Ingelheim International, GmbH) is a70,000 mW, single-chain protein produced from acloned human tPA gene using Chinese hamsterovary cells (1). Endogenous tPA is secreted in itssingle-chain form to be enzymatically converted byplasmin to its two chain form. Both forms of tPAare equally active. The vehicle consists of L-argininephosphate, phosphoric acid, and polysorbate 80. tPAhas been used both intracamerally and subretinally.The utility of intracameral tPA was demonstrated inanimal models of fibrin (2–4), hyphema (5), vitreous

hemorrhage (6–8), and subretinal hemorrhage (9,10).The utility of subretinal injection of tPA was demon-strated in animal models of subretinal hemorrhage(11–13).

In the anterior chamber, 0.05 mL containing up to200 mg and 0.10 mL containing up to 360 mg have beeninjected without unusual inflammation or toxicity tothe cornea or lens. In the vitreous cavity, 0.10 mLcontaining up to 25 mg has been injected withoutcorneal or retinal toxicity. Repetitive injections (threetimes, separated by seven-day intervals) of 3 mg tPAalso did not show retinal toxicity (8). A single reportsuggested probable retinal toxicity of 0.1 mLcontaining 25 mg (14). Dose-dependent retinal toxicitywas seen with 0.10 mL injections of 50, 75, and 100 mginto the vitreous cavity (15). Tractional retinal detach-ments were seen following 100 mg (6) and 200 mg (7)tPA injections.

In the subretinal space, no retinal toxicity wasseen after subretinal injection of 25 and 50 mg of tPA in0.1 mL of volume (11,12).

Lewis and colleagues demonstrated in rabbitsthat subretinal clots, 30-minutes old, cleared fasterafter a 0.1 mL subretinal injection of 25 mg tPA ascompared to an equivalent volume of BSS (11).However, the subretinal tPA could not completelyprevent retinal damage. Both BSS and tPA decreasedthe toxic effect of blood partly on the basis of dilutionof the subretinal blood. Johnson and colleaguesshowed a similar effect for lower doses of tPA (2.5 mgin 0.05 mL) on clots that were 24-hours old, but severeprogressive retinal degeneration was still seen (12).An ultramicrosurgical approach using a microinfusionof 0.5 to 5 mg of tPA facilitated lysis of one- and two-day-old clots and their removal through micropipettesunder stereotactic control. Good preservation of theretinal architecture was seen compared to untreatedcontrols (13).

The ability of intravitreal injections of tPA to lysesubretinal clots has been explored. Coll and colleaguesfound that 0.1 mL containing 50 mg of tPA facilitated

the lysis and absorption of one-day-old subretinal clotscompared to equivalent volume injections of saline (9).Unfortunately, retinal damage was not prevented.Boone and colleagues injected 25 mg of tPA into thevitreous space and found only partial clot lysis thatwas not enough to allow removal by aspiration alone(10). The inability of labeled tPA injected into thevitreous to penetrate the intact neural retina or asubretinal clot in rabbits was demonstrated by Kameiand colleagues (16). Some labeled tPA was able topenetrate into eyes with vitreous hemorrhage presum-ably from the microdefects through which bloodescaped from the subretinal space into the vitreous.

The previous studies spurred simultaneousinterest in the clinical use of tPA to assist in theremoval of subretinal hemorrhage. These techniquesinvolved the injection of 6.25 to 12.5 mg of tPA in avolume of 0.05 mL into the subretinal space and thenwaiting 10 to 45 minutes before aspiration of theliquefied blood. Injections into the subretinal spacewere accomplished with a glass pipette (17), 33-gaugecannula (18), or bent-tipped 30-gauge needle (19,20).Aspiration was performed with double-barrel subret-inal-injector aspirator (18), soft-tipped cannula (17,21),tapered 20-gauge Charles flute needle (20), or30-gauge subretinal cannula (22). Liquefied subretinalblood was also manipulated with a small perfluor-ocarbon liquid bubble (19,23,24).

In addition to intravitreal injection of tPA duringthe pars plana vitrectomy procedure, the injection of0.1 mL of 25 mg of tPA into the subretinal clot bypassing a 30-gauge needle through the pars planaunder indirect ophthalmoscopy the day before parsplana vitrectomy has also been described (25).

An intravitreal injection consisting of 6 mg of tPAin 0.1 mL was injected into the midvitreous cavity toliquefy subretinal clots 12 to 36 hours prior tovitrectomy and removal of blood through a retinotomyusing perfluorocarbon liquid manipulation (26). Intra-vitreal injections of 0.1 to 0.2 mL containing 25 to100 mg of tPA into the vitreous cavity have beengiven either the day before (27) or immediatelybefore (28,29) injection of intravitreal gas to displacesubmacular hemorrhage. Exudative retinal detach-ments seen after 100 mg injections were attributed totPA toxicity (28).

A number of investigators have injected 25 to50 mg tPA into the subretinal space following parsplana vitrectomy (30–32).An air fluid exchange wasperformed and the patient was kept erect to pneuma-tically displace the liquefied blood from the fovea.

Lewis injected tPA into the subretinal spacebefore excision of the choroidal neovascularmembrane but found no improvement comparedwith injection of BSS into the subretinal space in arandomized trial (33).

CALCIUM- AND MAGNESIUM-FREE RETINALDETACHMENT-ENHANCING SOLUTIONS

Marmor had discovered that removing calcium andmagnesium from a solution that bathed eye wallsections in vitro weakened retinal adhesive force(34). Wiedemann described a “detachment infusion”for macular translocation surgery that was calciumand magnesium free (35). Substituted for conventionalvitrectomy infusion fluid, this solution enabled theimmediate detachment of the retina from its periph-eral, diathermy-induced perforation site to the centerof the macula or macular area. He described its use inretinal organ culture and creation of experimentalretinal detachment in rabbits and in human surgery.

We hypothesized that BSS Part A might be anideal retinal detachment-enhancing solution andstudied its safety and efficacy in rabbits before usingit clinically in humans. BSS was developed as animprovement over normal saline, lactated Ringer’s,and Plasma-lyte 148 as a physiologically compatiblesolution to be used in the eye during surgery (36,37).To further improve the physiological compatibility ofBSS, glutathione, glucose, and bicarbonate buffersystems were added (38–40) resulting in BSS Plus.BSS Plus consists of two parts, which are reconstitutedjust prior to use in surgery. These two parts consist ofPart B, a sterile 480-mL solution in a 500-mL single-dose bottle to which Part A, a sterile concentrate ina 20-mL single-dose vial, is added. Compared to BSS,BSS Part A lacks magnesium and calcium, and thecitrate and acetate buffers of BSS have been replacedwith bicarbonate buffer. BSS Part B contains calciumand magnesium as well as the dextrose and theglutathione, which are unique to BSS Plus. Wehypothesized that BSS Part A alone could be usedsafely in the human eye since it contained almost allthe ingredients of BSS except for the calcium andmagnesium with a different buffering system and apH of 7.4. A tremendous advantage to the vitreoussurgeons is the commercial availability of BSS. We feltthat all these qualities plus the historical use of thesolution in the operating room (albeit reconstitutedwith Part B) could make it an ideal solution to enhanceretinal detachment during macular translocationsurgery. We showed the safety and efficacy of acalcium- and magnesium-free macular translocationsolution by comparing the results of injecting BSS PartA or BSS solution into the subretinal space of rabbiteyes using a 39-gauge cannula (40). No difference wasseen in fundus appearance, fluorescein angiography,electroretinography, or light or electron microscopy inrabbit retinas that had been detached using retinaldetachment solution compared to commercially avail-able solution. Using a manual infusion system, nomore than 100 mg of BSS compared to a much larger

296 SCARTOZZI AND CHONG

volume of retinal detachment solution could beinfused into the subretinal space. The diameter ofBSS retinal detachments was always less than that ofBSS Part A retinal detachments after injection of 100 mgof subretinal fluid.

Aaberg et al. have similarly shown the safety ofsubretinal BSS Part A in the sub-retinal space of therabbit using transscleral infusion (41).

We have used a 39-gauge cannula to atraumati-cally infuse BSS Part A underneath the retina inmacular translocation surgery and to displacesubmacular hemorrhage.

Clinically, we have found that macular transloca-tion surgery requires only one or two penetrationsthrough the retina with a 39-gauge cannula to detachthe posterior retina sufficiently. We have used BSS PartA to displace submacular hemorrhages by performingpars plana vitrectomy, injecting the solution to detachthe posterior pole of the retina, performing partial gas–fluid exchange, and then positioning the patient in anerect position for 24 hours to displace blood away fromthe fovea.

TRIAMCINOLONE ACETONIDE

A discussion of the pharmacology and the mechanismof action of TA is presented in Chapters 8 and 15. Theintravitreal injection of TA for the treatment of chor-oidal neovascular membranes is also discussed inthose chapters. The subretinal injection of TA will bediscussed here.

SUBRETINAL INJECTION OF TA

Some current methods for treating retinal diseasesinvolve the introduction of drugs directly into thevitreous chamber of the eye by intraocular injectionor intravitreal implant. Solutions injected directly intothe vitreous chamber, however, are often rapidlyremoved by the eye’s normal circulatory processes,requiring frequent injections or sustained release ofthe drug. These large-dose injections lead to thedistribution of the drug throughout the whole eyeand can be associated with complications, such ascataract formation and glaucoma. Additionally, thesetherapies do not address the issue of large molecularweight molecules (more than 70 kDa) that are virtuallyincapable of diffusing through retinal tissues.

The delivery of TA into the subretinal space hasbeen investigated for the treatment of subfoveal chor-oidal neovascularization (CNV) due to AMD. Thesubretinal delivery of a therapeutic agent couldallow for the local, low-dose treatment of retinalpathology with fewer complications to other intra-ocular structures such as the lens and optic nerve.These subretinal injections can be delivered through

a 41-gauge cannula (Figs. 1 and 2) and the retinalopenings that are created self-seal and do not needto receive retinopexy. In our institution, we haveperformed these injections after removal of the vitr-eous by pars plana vitrectomy and through formedvitreous without vitrectomy. These injections havebeen performed both in the operating room and inthe clinic setting.

In a pilot study, two eyes of two patients under-went pars plana vitrectomy, subretinal injection of4 mg of TA (0.1 mL), and gas–fluid exchange forsubfoveal neovascular AMD (42). The first patientsustained a limited subretinal hemorrhage intraopera-tively that cleared spontaneously over approximatelythree months, as well as a rise in intraocular pressurethat required the use of two topical medications tocontrol. The second patient demonstrated progressionof his nuclear sclerosis and posterior subcapsular lens

Figure 1 Forty-one gauge subretinal cannula in the extendedposition.

Figure 2 The body of the 41-gauge subretinal cannula has asliding button which can be used to extend and retract the

cannula itself.

20: AMD: USE OF ADJUNCTS IN SURGERY AND NOVEL SURGICAL APPROACHES 297

change over the 35 months of follow-up. Best correctedvisual acuity improved from 20/400 to 20/200 in thefirst patient, and from counting fingers to 20/320 in thesecond patient. The size of the neovascular complexesincreased modestly in both patients. The authorsconclude that their complications were not prohibi-tive, and that their results may be likened to the courseseen with PDT.

A pilot study of 14 eyes of 14 patients withsubfoveal CNV from AMD was performed where 0.5to 5 mg of TA was injected subretinally, overlying theCNV (43). Three patients developed a subretinalhemorrhage in the immediate postoperative period,where two of these three patients’ final visual acuitiesimproved (Fig. 3). Also, in the immediate postopera-tive period, one patient had an elevated intraocularpressure and one patient had a retinal detachment.There were no late postoperative complications.

Although subretinal delivery of triamcinoloneseems to be safe, maximizing durability of drug andminimizing injections is desirable. Therefore, abiocompatible, sustained-release subretinal drug-delivery platform has been developed which iscapable of delivering either TA or sirolimus (44). Theprototype implants were fabricated by coating nitinol,poly(methyl methacrylate) or chromic gut core fila-ments, with a drug-eluting polymer matrix, and testedin rabbits (Fig. 4). Initial observations of the implan-tation and elution characteristics revealed that theimplants are well tolerated by the retinal tissue andthat the implant can elute TA for a period of at leastfour weeks without eliciting an inflammatoryresponse or complications.

ENCAPSULATED CELL TECHNOLOGY

Encapsulated cell technology (ECT) employs mamma-lian cells that are genetically engineered to secrete atherapeutic factor. These engineered cells are thenencapsulated in a semi-permeable polymer membranedevice which allows for the free exchange of nutrientsand metabolites to sustain the cells, while allowingfor the exit of a therapeutic factor (Figs. 5 and 6). Atthe same time, these membranes protect the engin-eered cells from host antibodies and immune cells. Thedevices are then surgically implanted into the vitreouscavity of the eye (Fig. 7). ECTallows for the continuousand long-term site-specific administration of drugsin the eye without subjecting the host to systemicexposure. Furthermore, these implants can beretrieved, providing an added level of controland safety.

ECT-CNTF (human ciliary neurotrophic factor)devices were implanted in a dog model of retinitispigmentosa (RP) (45). One eye was implanted at sevenweeks of age, leaving the contralateral eye untreated.These devices were explanted at 7 or 14 weeks post-implantation. There was significant protection of thephotoreceptors from degeneration in a dose-depen-dent and safe manner, as revealed by examining thenumber of cells in the outer nuclear layer histo-logically. Furthermore, the data from this studyconfirmed that sustained delivery of protein thera-peutics is more effective than bolus injection, whileavoiding the additive risks of frequent intraocularinjections. The authors also concluded that this tech-nology was superior to the use of viruses in genetherapy, as gene therapy tended to be effective onlyfor a few weeks, induced an immune response, andproduced unpredictable amounts of therapeutic agent.

These data from animal studies enabled aprospective phase I clinical trial which safely deliveredCNTF to the eyes of 10 subjects suffering fromadvanced RP (46). Though this nonrandomized trial

Figure 4 The drug-releasing filament lies under the retina in thisrabbit.

Figure 3 After subretinal injection, the triamcinolone is locatedinferior and temporal to the fovea. There is hypopigmentation at

the site of the injection which is located between the white massand the fovea.


had only a small number of participants and was notplacebo-controlled, several of the implanted eyesshowed a trend of better acuity on a letter recognitiontask compared with contralateral control eyes. At theend of the six-month implantation duration, allexplanted capsules contained viable cells that secretedCNTF at expected levels that were therapeutic in thercd1 (cGMP-PDE6b mutation) dog study (45).

Smaller ECT devices that can be implantedthrough a 25-gauge opening are currently beingdeveloped (47). The application of this technology toother ocular diseases such as AMD is currently anactive area of investigation.

SELECTIVE INTRAOCULAR RADIATIONBRACHYTHERAPY

Although the results of radiation treatment for neovas-cular AMD are mixed and generally unfavorable(48,49), there is data to suggest that higher dosagesmay produce better results (50–53). Flaxel andcolleagues showed promising results with protonbeam radiation at 8 to 14 Gray (Gy), but radiation

Oxygen and Nutrients

Therapeutic Factors

Immune SystemComponents

ImmunoisolatoryMembrane

Encapsulated CellProtein Delivery

Figure 5 Cells that release therapeutic molecules are protected from the host immune system by a

semipermeable membrane. The semipermeable membrane allows these cells to receive oxygen andnutrients.

Membrane

Encapsulated Cell Technology

Suture clipCells

Scaffold

Seal

Figure 6 Living cells that release therapeutic molecules areencased in a proprietary scaffold, which serves as the semiperme-

able membrane to create the cell-based drug delivery device.Figure 7 The cell-based drug delivery device is anchored bysuturing an attached titanium ring to the pars plana.


retinopathy was a serious complication and seen in50% of the 14 Gy-treated eyes (54).

To overcome this limitation in radiation deliverydosage caused by radiation retinopathy, a method ofdelivering focal radiation to the choroidal neovascularmembrane by passing an intraocular radiationdelivery device underneath the retina was developed.Retina exposure to radiation is minimized by direc-tional shielding of the subretinal radiation source andby the focal nature of the radiation delivery. Creationof a subretinal bleb using a 41-gauge needle and aretinotomy allowed for the brachytherapy probe to bein direct contact with the retinal pigment epithelium(RPE) for a set amount of time. The brachytherapydevice contains shielding which prevent radiationexposure on the retinal side.

In a phase I clinical study, 10 eyes of 10 patientsreceived 26 Gy of subretinal radiation via an angled ornon-angled probe to active subfoveal CNV, withfollow-up ranging from two to nine months (55). Byfluorescein angiography, greatest linear dimension(GLD) leakage decreased by 46% in one month, 64%in three months, and 82% in six months. By opticalcoherence tomography, total macular volumedecreased by 13% in one month, 19% in threemonths, and 30% in six months. Visual acuity wasstable or improved in 44% by two and three months.Adverse events included RPE tears, RPE atrophy, RPEhyperpigmentation, subretinal hemorrhage, cataract,pre-retinal hemorrhage and vitreous hemorrhage.

Because of these complications, efforts are nowdirected towards pre-retinal delivery of radiation.Preclinical study showed that the minimum thresholdfor acute damage using a pre-retinal focal radiationdelivery device was above 103 Gy, nearly four timesthe dosage expected to cause beneficial effectdescribed in the literature. As a result, a phase I clinicaltrial is in the planning stage to evaluate the safety andfeasibility of focal delivery of radiation from a pre-retinal position using a sealed radiation source placedtemporarily over the fovea in the vitreous cavity bymeans of a proprietary intraocular probe. The deliverydevice is a shielded canister containing a strontium90

beta-radiation source with an angled-tip that has a1.0 mm outer diameter. In the storage (retracted)position, the radiation source is surrounded by astainless steel and lead lining that effectively protectsthe surgeon and patient during the handling andinitial positioning. This tip will allow for the direc-tional delivery of approximately 24 Gy of beta-radiation via a light touch approach on the retinalsurface for a three to five minute period. In thetreatment position, the source is located within aspecially designed stainless steel tip that providesdirectional administration of the beta radiation while

shielding and protecting surrounding non-target,unaffected (i.e., disease free) tissues (Figs. 8 and 9).

SUMMARY POINTS

& Adjuncts are used primarily in the subretinal spaceduring surgery for AMD.

& tPA can be infused into the subretinal space toliquefy subretinal blood.

& tPA may penetrate human retina after injectioninto the vitreous cavity through microperforationsto liquefy subretinal blood.

& Calcium- and magnesium-free solutions enhanceretinal detachment.

& BSS Plus Part A is a safe and readily availableretinal detachment solution.

& Calcium- and magnesium-free solutions can aidmacular translocation surgery and the displace-ment of submacular hemorrhage.

Radiation Source In Storage

Position

Radiation Source In Treatment

Position

Figure 8 In a retracted position, the radiation source lies in thebody of the delivery instrument and is shielded from the environ-

ment. The sliding button is used to extend the radiation sourceinto the curved tip for radiation of the choroidal neovascular

membrane.

Figure 9 Delivery device for intraocular radiation brachy-

therapy.


& Novel surgical approaches seek a long termenhancement of existing therapies for AMD.These approaches include the surgical implan-tation of sustained release drug devices, thesurgical implantation of cell-based deliverysystems, and the pre-retinal or subretinal deliveryof radiation therapy through a pars planaapproach.

REFERENCES

1. Pennica D, Holmes WE, Kohr WJ, et al. Cloning andexpression of human tissue type plasminogen activatorwith DNA in E. coli. Nature 1983; 301:214–21.

2. Johnson RN, Olsen K, Hernandez E. Tissue plasminogenactivator treatment of postoperative intraocular fibrin.Ophthalmology 1988; 95:592–6.

3. Lambrou FH, Snyder RW, Williams GA, Lewandowski M.Treatment of experimental intravitreal fibrin with tissueplasminogen activator. Am J Ophthalmol 1987; 104:619–23.

4. Snyder RW, Lambrou FH, Williams GA. Intraocular fibri-nolysis with recombinant human tissue plasminogenactivator. Arch Ophthalmol 1987; 105:1277–80.

5. Lambrou FH, Snyder RW, Williams GA. Use of tissueplasminogen activator in experimental hyphema. ArchOphthalmol 1987; 105:995–7.

6. Johnson RN, Olsen DR, Hernandez E. Intravitreal tissueplasminogen activator treatment of experimental vitreoushemorrhage. Arch Ophthalmol 1989; 107:891–4.

7. Min WK, Kim YB, Lee KM. Treatment of experimentalvitreous hemorrhage with tissue plasminogen activator.Korean J Ophthalmol 1990; 4:12–5.

8. Min WK, Kim YB, Ahn BH, Seong GH. Repetitive low-dosetissue plasminogen activator for the clearance of experi-mental vitreous hemorrhage. Korean J Ophthalmol 1994;8:45–8.

9. Coll GE, Sparrow JR, Marinovic A, Chang S. Effect ofintravitreal tissue plasminogen activator on experimentalsubretinal hemorrhage. Retina 1995; 15:319–26.

10. Boone DE, Boldt HC, Ross RD, Folk JC, Kimura AE. The useof intravitreal tissue plasminogen activator in the treatmentof experimental subretinal hemorrhage in the pig model.Retina 1996; 16:518–24.

11. Lewis H, Resnick SC, Flannery JG, Straatsma BR. Tissueplasminogen activator treatment of experimental subretinalhemorrhage. Am J Ophthalmol 1991; 111:197–204.

12. Johnson MW, Olsen DR, Hernandez E. Tissue plasminogenactivator treatment of experimental subretinal hemorrhage.Retina 1991; 11:250–8.

13. Toth CA, Benner JD, Hjelmeland LM, Landers MB, III,Morse LS. Ultramicrosurgical removal of subretinal hemor-rhage in cats. Am J Ophthalmol 1992; 113:175–82.

14. Min WK, Kim YB. Resolution of experimental intravitrealfibrin by tissue plasminogen activator. Korean J Ophthalmol1990; 4:58.

15. Johnson MW, Olsen KR, Hernandez E, Irvine WD,Johnson RJ. Retinal toxicity of recombinant tissue plasmi-nogen activator in the retina. Arch Ophthalmol 1990;108:259–63.

16. Kamei M, Misono K, Lewis H. Study of the ability of tissueplasminogen activator to diffuse into the subretinal spaceafter intravitreal injection in rabbits. Am J Ophthalmol1999; 128:739–46.

17. Peyman GA, Nelson NC, Alturki W, et al. Tissue plasmi-nogen activating factor assisted removal of subretinalhemorrhage. Ophthalmic Surg 1991; 22:575–82.

18. Lewis H. Intraoperative fibrinolysis of submacular hemor-rhage with tissue plasminogen activator and surgicaldrainage. Am J Ophthalmol 1994; 118:559–68.

19. Vander JF. Tissue plasminogen activator irrigation to facili-tate removal of subretinal hemorrhage during vitrectomy.Ophthalmic Surg 1992; 23:361–3.

20. Moriarty AP, McAllister IL, Constable U. Initial clinicalexperience with tissue plasminogen activator (tPA) assistedremoval of submacular haemorrhage. Eye 1995; 9:582–8.

21. Manning LM, Contrad DK. Tissue plasminogen activator inthe surgical management of subretinal haemorrhage. AustNZ J Ophthalmol 1994; 22:59–63.

22. Ibanez HE, Williams DF, Thomas MA, et al. Surgicalmanagement of submacular hemorrhage: a series of 47consecutive cases. Arch Ophthalmol 1995; 113:62–9.

23. Lim JI, Drews-Botsch C, Sternberg P, Jr., Capone A, Jr.,Aaberg TM, Sr. Submacular hemorrhage removal. Ophthal-mology 1995; 102:1393–9.

24. Kamei M, Tano Y, Maeno T, Ikuno Y, Mitsuda H, Yuasa T.Surgical removal of submacular hemorrhage using tissueplasminogen activator and perfluorocarbon liquid. AmJ Ophthalmol 1996; 121:267–75.

25. Chaudhry NA, Mieler WF, Han DP, Alfaro VD, III,Liggett PE. Preoperative use of tissue plasminogen activatorfor large submacular hemorrhage. Ophthalmic Surg Lasers1999; 30:176–80.

26. Kimura AE, Reddy CV, Folk JC, Farmer SG. Removal ofsubretinal hemorrhage facilitated by preoperative intra-vitreal tissue plasminogen activator. Retina 1994; 14:83–4.

27. Heriot W. Intravitreal gas and tPA: an outpatient procedurefor subretinal hemorrhage. In: Vail Vitrectomy Meeting,Vail, Colorado, March 10–15, 1996.

28. Hesse L, Schmidt J, Kroll P. Management of acute subma-cular hemorrhage using recombinant tissue plasminogenactivator and gas. Graefes Arch Clin Exp Ophthalmol 1999;202:273–7.

29. Hassan AS, Johnson MW, Schneiderman TE, et al. Manage-ment of submacular hemorrhage with intravitreous tissueplasminogen activator injection and pneumatic displace-ment. Ophthalmology 1999; 106:1900–7.

30. Connor TB. Surgical displacement of submacular hemor-rhage. In: Vail Vitrectomy Meeting, Vail, Colorado, March15, 2000.

31. Federman IL. Variation in surgical management of sub-macular hemorrhage. In: Vail Vitrectomy Meeting, Vail,Colorado, March 15, 2000.

32. McCuen BW. A new concept in the treatment of submacularhemorrhage in AMD. In: Vail Vitrectomy Meeting, Vail,Colorado, March 14, 2000.

33. Lewis H, VanderBrug MS. Tissue plasminogen activator-assisted surgical excision of subfoveal choroidal neovascu-larization in age-related macular degeneration: arandomized, double-masked trial. Ophthalmology 1997;104(11):1847–51 (discussion 1852).

34. Yao XY, Endo EG,MarmorMF. Reversibility of retinal adhe-sion in the rabbit. Invest Ophthalmol Vis Sci 1989; 30:220–4.

35. Faude F, Reichenbach A, Wiedemann P. A detachmentinfusion for macular translocation surgery. Retina 1999;19(2):173–4.

36. Edelhauser HF, Van Horn DL, Hyndiuk RA, Schultz RO.Intraocular irrigating solutions: their effect on the cornealendothelium. Arch Ophthalmol 1975; 93:648–57.


37. Waltman SR, Carroll D, Schinimelpfenning W, Okun E.Intraocular irrigating solutions for clinical vitrectomy.Ophthalmic Surg 1975; 6(4):90–4.

38. Benson WE, Diamond JG, Tasman W. Intraocular irri-gating solutions for pars plana vitrectomy: a prospective,randomized, double-blind study. Arch Ophthalmol 1981;99:1013–5.

39. Glasser DB, Matsuda M, Ellis JG, Edelhauser HF. Effect ofintraocular irrigating solutions on the corneal endotheliumafter in vivo anterior chamber irrigation. Am J Ophthalmol1985; 99:321–8.

40. Araie M. Barrier function of corneal endothelium and theintraocular irrigating solutions. Arch Ophthalmol 1986;104:435–8.

41. Aaberg TM, Sharara NA, Edelhauser HF, Grossniklaus HE.Hydroseparation of the neurosensory retina with calciumfree BSS Plus. In: XXIInd Meeting of the Club Jules Gonin,Taormina, Italy, September 2–6, 2000.

42. Kertes PJ, Coupland SG. The use of subretinal triamcinoloneacetonide in the management of neovascular age-relatedmacular degeneration: a pilot study. Can JOphthalmol 2005;40(5):573–84.

43. Equi RA, de Juan E, Jr., Sadda SR, Varner S, Lim JI.Subretinal triamcinolone for the treatment of neovascularage-related macular degeneration: a pilot study. In: 21stAnnual Meeting of the American Society of Retina Special-ists, New York, August 16–20, 2003.

44. Beeley NR, Stewart JM, Tano R, et al. Development,implantation, in vivo elution, and retrieval of a biocompa-tible, sustained release subretinal drug delivery system.J Biomed Mater Res A 2006; 76(4):690–8.

45. Tao W, Wen R, Goddard MB, Sherman SD, O’Rourke PJ,Stabila PF, Bell WJ, Dean BJ, Kauper KA, Budz VA, et al.Encapsulated cell-based delivery of ciliary neurotrophicfactor reduces photoreceptor degeneration in animalmodels of retinitis pigmentosa. Invest Ophthalmol Vis Sci2002; 43(10):3292–8.

46. Sieving PA, Caruso RC, Tao W, et al. Ciliary neurotrophicfactor (CNTF) for human retinal degeneration: phase I trialof CNTF delivered by encapsulated cell intraocularimplants. Proc Natl Acad Sci USA 2006; 103(10):3896–901.

47. Kauper KA, Stabila P, Mills J, et al. Delivery of encapsulatedcell technology (ECT) device implants using a small gaugeneedle. Invest Ophthalmol Vis Sci 2006; 47:E-Abstract 5118.

48. The Radiation Therapy for Age-related Macular Degener-ation (RAD) Study Group. A prospective, randomized,double-masked, trial on radiation therapy for neovascularage-related macular degeneration (RAD Study). Ophthal-mology 1999; 106:2239–47.

49. Kobayashi H, Kobayashi K. Age-related macular degener-ation: long-term results of radiotherapy for subfovealneovascularmembranes. Am JOphthalmol 2000; 130:617–35.

50. Jaakkola A, Heikkonen J, Tommila P, Laatikainen L,Immomen I. Strontium plaque brachytherapy for exudativeage-related macular degeneration: three-year results of arandomized study. Ophthalmology 2005; 112:567–73.

51. Jaakkola A, Heikkonen J, Tommila P, Laatikainen L,Immomen I. Strontium plaque irradiation of subfovealneovascular membranes in age-related macular degener-ation. Graefes Arch Clin Exp Ophthalmol 1998; 236:24–30.

52. Finger PT, Berson A, Ng T, Szechter A. Ophthalmic plaqueradiotherapy for age-related macular degeneration associ-ated with subretinal neovascularization. Am J Ophthalmol1999; 127:170–7.

53. Rossi JV, Fujii GY, Humayun MS, et al. Submacular surgeryfor selective subretinal delivery of beta-radiation. InvestOphthalmol Vis Sci 2004; 45:E-Abstract 5140.

54. Flaxel CJ, Friedrichsen EJ, Smith JO, et al. Proton beamirradiation of subfoveal choroidal neovascularisation inage-relatedmacular degeneration. Eye 2000; 14(Pt 2):155–64.

55. Lim JI, deJuan E, Jr., Sadda S, et al. Subretinal radiationtreatment of occult choroidal neovascularization due toage-related macular degeneration. Invest Ophthalmol VisSci 2005; 46:E-Abstract 1384.


Part VI: Visual Rehabilitation

21

Clinical Considerations for Visual RehabilitationSusan A. PrimoDepartment of Ophthalmology, Emory University School of Medicine, Atlanta, Georgia, U.S.A.

INTRODUCTION

While the trauma of macular degeneration is difficultenough for some patients to cope with, the visualimpairment left afterwards is even tougher. Patientsmust not only learn to accept the fate of retinal disease,but must also summon the strength to accept thefact that they will surrender a certain degree ofindependence as visual acuity declines. The visualrehabilitative process helps the visually impairedpatient to regain a satisfactory level of independenceand can be achieved by assisting the patient inlearning to cope with the psychological, emotional,and economic aspects of vision loss, as well as throughthe use of optical, nonoptical, and electronic devices,where there are some exciting new technologies avail-able. Typically, this type of integrated rehabilitativeprocess is necessary for patients with severe andprofound visual impairment, i.e., legal blindness.

THE LOW VISION EVALUATION

The term legal blindness as defined is a visual acuity of20/200 or worse in the best-corrected better eye or avisual field of 208 or less in the widest diameter ofvision. A patient cannot have poor vision in one eyeonly and be considered legally blind. This classi-fication becomes a part of the patient’s permanentrecord and has implications for eligibility forstate financial assistance, tax benefits, reduced publictransportation fares, and other circumstances. Inaddition, in many states that have “commissions” forthe blind, reporting of legal blindness may cause adriver’s license to be revoked. For many people,having a driver’s license, whether actually drivingor not, has significant meaning and serves as a formof identification. The practitioner should be aware ofthese issues when designating this classification.

The low vision examiner begins the evaluationwith a complete understanding of the patient’s ocularhistory. Detailed documentation of surgical historyand stage of pathology are important components.

Typically, a low vision evaluation should notcommence until a patient has undergone all surgicaland nonsurgical attempts at restoring visual function.The reasons for this are twofold. First, the low visionexaminer is concerned with performing an extensiveevaluation often using the state-of-the art devices,which can be quite expensive. If the patient’s finalvisual acuity is in question, these devices may not besuitable once the visual acuity has reached its finallevel and has stabilized. Secondly, the patient needsto have gone through the “mourning process” oflosing sight with the understanding that the nextstep must be taken to begin the visual rehabilitativeprocess. This is not to say that if miracle breakthroughsbecome available, then a patient should not haveaccess to any possibility of restoring sight. However,success with low vision devices is completely depen-dent upon: (i) patients’ full acceptance of theirvisual impairment and (ii) the ability and desire tomove on.

During the history, the patient is asked aboutaspects of vision loss. These aspects include duration,symmetry, fluctuations, stability, loss of ability todiscriminate color, effects of various illuminations orlighting conditions, andmobility concerns. These ques-tions assist patients in learning to talk about the effectsof the visual impairment on their lifestyle, an importantstep in beginning the rehabilitative process. Whileascertaining this information, the low vision examineralso documents any current devices including glasses,which may already be in the patient’s possession.Frequently, a well-meaning spouse or relative hasalready offered the patient a magnifier of some sort. Itis important to categorize all such devices for type,style, and power. It is also important to determine theusefulness of these devices. For example, can thepatient read large print or headlines of a newspaperwith glasses and/or a magnifier? Oftentimes, patientswill say that all devices are useless, but in reality, theymay be able to see large print and not regular print.While this may be considered useless to them, it isimportant to the examiner.

Perhaps, the last and most important part of thehistory is an expression by patients of their goals andexpectations. During this portion, the examinerdetermines whether the patient has realistic goalsand expectations or whether the desire is to “just seeagain.” A detailed list of desired activities is recordedin order of importance to the patient. Sometimes ittakes a little prodding, but virtually all patients’primary desire is to be able to read again. It isimportant to determine if patients simply want toread mail or bills in order to handle their own financesand/or patients want to continue leisurely reading ofprinted materials such as newspaper, novels, etc.Second to a desire to read is usually improvement ofdistance vision. Again, specific distant activities(watching television, bird-watching, or driving) needto be discussed. The driving issue is an extremelysensitive area and the examiner uses compassion andsensitivity in discussing this topic. A more detaileddiscussion of driving will follow later in this chapter.Using a checklist approach is a quick and easy methodfor determining the patient’s current level of visionand, subsequently, any special material the patientwishes to read or certain activities the patient wishesto engage in, i.e., playing cards, sewing, drawing orpainting, golf, etc. (Fig. 1). Finally, maximizing aneducation environment is critical for young childrenor adults, as is attention to patients’ workplace ifthey are employed or seeking gainful employment.

Although there is a standard format to the lowvision evaluation, the examiner always bears in mind

a goal-orientated approach. For example, if a patientexpresses the desire to read only, the focus will be onachieving this goal. The examiner might explainpossibilities for improvement in distance vision, butif a patient is still uninterested, the telescopic evalu-ation is probably unnecessary. Likewise, if a patient’sonly desire is to drive, a short near evaluation may beperformed to demonstrate possibilities, but clearly theemphasis in this case would be on the telescopicevaluation. As the examiner notices head and bodymovements as the patient initially walks into theexamination room, these seemingly minor obser-vations provide information not only about visualstatus, but also about a patient’s level of adaptationto the vision loss. In addition, before visual acuitytesting begins, any auxiliary testing is performed,which may include contrast sensitivity, Amsler grid,visual fields, etc. These tests can shed light on the sizeand extent of the central scotoma as well as on othersubjective aspects of the acuity loss.

As clear-cut as it may seem, visual acuity testingis an extremely important (and often long) part of therehabilitative examination. Evaluating a patient withreduced visual acuity requires that basic examinationtechniques be modified. It is generally recommendedthat vision testing be done at 10 ft with a self-illumi-nated, portable eye chart. The Early TreatmentDiabetic Retinopathy Study (ETDRS) chart is themost widely used. A projector chart is the leastfavorable means of measuring acuity in a patientwith reduced vision. Not only is contrast not constantwith a projector chart depending on the level of roomillumination, but a patient would also have to bemoved closer to the chart if vision was worse than20/400. Moving a visually impaired patient onlyreinforces awareness of the vision loss and causesstress and negative feelings during the exam.

The ETDRS chart has several advantages. It isself-illuminated with high contrast and is on wheelsin order that it could be moved closer than 10 ft ifnecessary. Also, the chart has a wide spectrum ofvisual acuity values, ranging from a “Snellen 200 ft”equivalent to a 10-ft equivalent. Since the testingdistance is always recorded as the numerator of theSnellen fraction, this chart gives an acuity range(at 10 ft) from 10/200 (20/400) to 10/10 (20/20).If the chart is moved closer, the test distance is againrecorded as the numerator. It is best not to convert theacuity to the 20-ft equivalent when recording vision sothat the examiner may always know the test distancefor subsequent evaluations. “Counting fingers” visionfor measurement is generally not used during a lowvision evaluation. The fingers subtend approximatelythe same visual angle as a 200-ft Snellen figure. There-fore, a patient should be able to read the top line ofthe ETDRS chart at a closer distance. Recording visual

Checklist for Current Level of Vision/Goals

Yes No Desires

Headlines ________________________________________________________________

____

____

________________________________________________________________

____

____

________________________________________________________________

____

MagazinesRegular NewsprintLabels, Price TagsMoneyRecognize FacesWatch TVCookingSew, Knit, etc.HousekeepingHygieneHandiworkGarden/Yard WorkSports (Golf, etc.)Play CardsDrivingOther

Glare

Figure 1 Checklist of vision/goals.

304 PRIMO

acuity as “3/200” instead of counting fingers at 3 ft ismuch more accurate, which is importantin determining which optical devices may beappropriate.

Determination of eccentric view, if suspected,is done during acuity testing. The easiest method iscalled the clock-face method. Patients are asked tokeep their head still and to face straight ahead. Theexaminer then asks the patient to imagine that the eyechart is at the center of a clock. Patients are asked tomove their eye in various positions of the clock untilthe top line becomes the clearest and most complete.Typically, but not always, a patient with acquiredmacular disease will attempt to place the image onthe temporal retina where the most room is, i.e., theright eye will eccentrically view towards the right(3 o’clock) and the left eye will view towards the left(9 o’clock). This position should be demonstrated tothe patient several times and recorded next to visualacuity. The knowledge of the exact location of theeccentric view will become useful for the remainderof the evaluation with devices.

Manifest refraction in a trial frame is generallythe rule. In this case, the examiner can observe thepatient’s eyes, particularly to reinforce the eccentricview. A phoropter does not allow a patient’s eyes to beobserved and the use of an eccentric view by thepatient becomes quite difficult. In addition, the lensincrements may be too small for a patient to determineany subjective difference, i.e., a patient with 20/200vision will not appreciate a difference ofG0.25 D. Theexaminer cannot easily make large increments ofchange in the phoropter for patients with poorervision. Generally, a patient whose vision is less than20/100 will appreciate 0.50 D lens changes. For visionbetween 20/100 and 20/200, the examiner should use0.75 to 1.50 D changes. If a patient’s vision is 20/200 to20/400, 1.50 to 2.00 D lens increments should be used.This technique is called lens bracketing and is the mosttime efficient and effective. Likewise, when measuringastigmatic corrections, a higher powered JacksonCross Cylinder (0.75–2.00 D) is employed to ensurethat the patient appreciates the lens changes for powerand axis refinements. To ensure that large refractiveerrors are not missed, keratometry, retinoscopy,and/or autorefraction offer a starting point. A scrupu-lous refraction is crucial before low vision devices aredemonstrated. Spectacles should always be prescribedusing polycarbonate lenses even if there seems to be aminimal increase in visual acuity; they serve as aimportant source of protection, particularly when thepatient is engaged in activities where there may behanging branches, flying objects, chemicals, etc., orsimply unfamiliar terrain.

Depending on the patient’s expression of initialgoals, either a brief or extensive telescopic evaluation

is performed next. Improvement of distance visionmay have not initially been an expressed goal since apatient may be mostly tuned in to reading concerns. Inany event, a brief introduction of a 3 or 4! poweredtelescope in the trial frame will demonstrate not onlythe device to the patient, but also the possibility ofenhancing distance vision. Vision should generallyimprove proportionately with the power of thescope. For example, if a patient has best-correctedvision of 20/200, vision should improve to 20/50with a 4! telescope. Exceptions to this rule may be alarge or irregular central scotoma or the co-existence ofother media opacities. Generally speaking, mostdistance activities usually require visual acuity of20/30 to 20/50. It is rare that an individual wouldneed to be corrected to 20/20 or better with a tele-scope. The aim should be to prescribe the lowest-powered telescope to achieve the required vision.The reasons for this are that as a telescope powerbecomes greater, the smaller the field of view andthe more difficult it becomes to use effectively. Ifvisual acuity is near equal between the two eyes, theexaminer may choose to prescribe a binocular systemwhich will give a much larger field of view foractivities such as watching television, going toshows, etc. Another alternative for increased visualfield may be a contact lens telescope (CLT). In this case,a contact lens (high minus power) is used as the ocularin conjunction with a high plus lens as the objective ina spectacle (1). As vision approaches 20/400 andworse, standard telescopic devices may not beuseful; a CTL as well as more advanced technologicalelectronic devices should be considered.

The driving issue is one that remains controver-sial and requires special mention. Driving is animportant component of everyday life for mostpatients. The inability to drive has psychological impli-cations in terms of limited independence. The subjectmust be treated with extreme care and sensitivity.Driving remains an instrumental activity of dailyliving and should be routinely addressed during thecase history and discussion of goals and expectations.Individuals with age-related macular degeneration dodrive although their driving exposure is lowby tendingto avoid challenging and hazardous situations such asdriving at night and in inclement weather (2).

Visually impaired people are permitted to usebioptic telescopes for driving in 37 states when visualacuity falls below the state’s legal limit. The termbioptic simply implies two (bi-) optical centers. Thisform of a telescope is mounted several millimetersabove the distance optical center. Therefore, a patientlooks through his/her natural prescription throughthe carrier lens housing the telescope. When sharperacuity is needed for viewing street signs, etc., thepatient lowers the chin and spots through the scope.

21: CLINICAL CONSIDERATIONS FOR VISUAL REHABILITATION 305

This manner of use is similar to the fashion in which adriver would use the rearview mirror; the telescope isused only approximately 10% to 15% of the time whiledriving. This point is an extremely important one andoften confused because it is thought that since there isa reduced field through the telescope, one could notpossibly drive safely. Again, the driver is primarilylooking through the carrier lens, not the telescopicdevice. Bioptic telescopes do meet the self-reporteddriving needs of the majority of visually impaireddrivers and were found to be a useful device forresolving details such as road signs, etc. (3).

Although visual acuity is a fundamental part ofsafe driving, several studies have demonstrated thatperipheral field (or vision) appears to play a morecritical role in driving than visual acuity (4–6). Allstates allowing bioptic telescopes have a minimumvisual field requirement without the telescope (usuallybetween 1208 and 1408). Other requirements includemaximum acuity without the telescope (usually20/200) andminimum visual acuity with the telescope(20/40–20/60). Certainly, there are issues that gobeyond visual acuity and peripheral field in deter-mining whether any given driver will drive safely,particularly one with a visual impairment. Factorssuch as age, experience, visual attention and proces-sing, reaction times, and cognitive deficits allinarguably affect an individual’s ability to drivesafely. Recent research has led to the development ofa software called the Useful Field of View test (VisualAwareness, Inc., Birmingham, Alabama). This testrequires higher-order processing skills, and not onlydetermines a conventional visual field, but also allowsfor the assessment of the visual field area over whichrapid stimuli are flashed, i.e., a car or other objectmoved into a cluttered background. Simulating the“real” driving experience with this test, an associationhas been shown between elderly drivers who havereductions in their useful field of view and crashinvolvement (7–9). This test is invariably more import-ant in determining driving safety than traditionalassessments of visual acuity and peripheral field.

Most driver’s with visual impairment do limitdriving exposure and tend to avoid challengingdriving situations, i.e., driving at night, on interstates,during inclement weather, etc. There has been noassociation found between drivers with maculardegeneration and increased accident rates/fatalities;however, driving exposure is taken into account (10).One study has demonstrated that, although patientswith macular degeneration performed more poorly ondriver simulator and on-the-road tests compared witha control group, this did not translate into an increasedrisk of real-world accidents (11). Hence, it still remainsunclear whether reduced exposure decreases adriver’s risk or whether any association exists

between increased injurious accidents and visualimpairment secondary to macular degeneration. Thedecision to prescribe a telescopic device for a patient tolegally maintain a driver’s license is a joint decisionbest left to doctors, patient, and family; the decisionshould be made on an individual basis.

Following refraction and distance evaluation isthe near evaluation. Near visual acuity is most appro-priately measured and evaluated with continuous-textreading cards. These cards will test a patient’sfunctional ability to read versus the ability to read aline of numbers or letters. MNReade and Sloan makecontinuous-text near cards. “M” notation is generallyused for recording near acuity. This notation uses themetric system, is standardized, and does not require afixed testing distance. To begin the near evaluation, areading lens addition should always be in place whentesting patients above 50 years of age and test distancemust be appropriate for the power of the add. Forexample, the test distance for aC2.50 D add should be40 cm or 16 in. (100/2.50Z40 cm; 40/2.50Z16 in.), andthe test distance for aC4.00 D add should be 25 cm or10 in. (100/4.00Z25; 40/4.00Z10 in). Distance andnear visual acuities (with standard C2.50 D add)should be approximately the same so that if a patient’sbest-corrected vision is 20/200 in the distance, the nearvision with standard add should also be 20/200.Pupil size, asymmetry, significant media opacities,and large central scotomas may create disparities;however, large differences between distance and nearacuities should alert the examiner to an inaccuratemanifest refraction.

Once initial near acuity has been determined, theexaminer increases the power of the add until theappropriate acuity is obtained. The approximate addit will take for any given patient to read newspapersize print (1 M or 20/50) can be predicted by calcu-lating the reciprocal of the distance or near acuity. Forexample, if a patient’s vision is 20/200, it would take atleast a C10.00 D add (200/20Z10) for the patient toread newspaper size print. This value may bemodified depending upon the patient’s initialexpression of goals for reading. If a patient wishes toread the stock pages, then more plus may be needed,and if a patient only wishes to read large-print text,then less plus is needed.

Binocular adds are typically prescribed when theacuity is equal or near equal between the two eyes.Base-in prism is always required in binocular addsgreater than C6.00 D because fusional vergence isexhausted and the eyes drift towards an exophoricposture. The amount of prescribed prism is two prismdiopters more than the amount of plus. For example, ifthe examiner wishes to prescribe a C8.00 D add botheyes (OU), the prescribed prism should be 10 prismdiopters base-in total split equally between the two

306 PRIMO

eyes. Glasses should be prescribed in a half-eye framesize due to the thickness and heaviness of the lenses.Since the nasal edge of the standard lens becomesquite thick with increased prism, adds greater thanC12.00 D should be prescribed monocularly with theeye not being used either occluded or the lens frostedto avoid diplopia. Recent advances using diffractiveoptics have greatly reduced the unsightly appearanceof these half-eyes (Fig. 18). If a patient has one eye thatis considerably better, the high add is prescribedmonocularly. However, there may still be a “ghost”image or halo around letters or words coming from thepoorer eye. In this instance, the lens of the poorer canbe frosted (or occluded).

For higher adds, most patients continually needreinforcement regarding the appropriate and closeworking distance. Most people are able to conceptu-alize inches rather than centimeters. Conversion ofreading distance into inches requires the power of theadd to be divided into 40. For example, if a C7.50 Dadd has been prescribed, the patient must hold allreading material at 51⁄3 in. (40/7.5Z5.33). The patientshould begin with larger text initially to becomeadjusted to the closer than usual reading distance andprobably increased fatigue. If a patient has not yetaccepted his/her visual loss, success with high addsand close reading distances is virtually impossible.

For those patients rejecting the close readingdistance of high adds, other alternatives exist. Tele-microscopes (surgical loupes/telephoto lens) can bemade with a specified reading distance. However, thepatient must weigh the benefit of the increaseddistance versus the reduction in the field of viewexperienced. Although every attempt is made toprescribe a spectacle-borne reading device, electronicdevices such as the closed-circuit television (CCTV)are good alternatives to spectacles. A patient can sitback at a comfortable distance and magnify the printlarge enough to read easily. A CCTV has not tradition-ally been a portable device and can cost severalthousand dollars, but clearly can have a tremendousimpact in allowing a patient to read (or work) again.There have been major recent advances here as well inportable versions of the reading machines (Figs. 7, 8,13–16) allowing the devices to be carried to and fromhome, school, or work. Handheld or stand magnifierscan also increase working distances and are typicallyprescribed in conjunction with spectacles. Thesedevices are most useful for spot reading rather thanextended reading. However, for those patientsrejecting spectacles, these devices are quite effectivealthough the reading field of view is reduced.

Contrast enhancement and glare reductionprovide the final steps to the low vision evaluation.Patients with macular degeneration often experience aloss of contrast. Contrast enhancement lenses shield

the eyes from very short wavelengths of light. Theseshorter wavelengths consist of high-energy visibleblue light and can cause loss of contrast as well asglare, which reduces the eye’s overall function. TheCorningw GlareControle family of lenses consists ofseven filters which selectively block specific wave-lengths of blue light while transmitting light at otherwavelengths. There are six graduated filter levels, eachnumbered to block below the corresponding wave-length (CPFw 450, 511, 527, 527!, 550, and 550! D).The filters range in color from yellow (450) to deep red(550). The seventh filter is called the GlareCuttere lensand is for patients with initial to moderate lightsensitivity. The lens color is more cosmeticallyappealing because it has more of a brownish huerather than orange/red. All of the filters are photo-chromatic easing the transition between different lightlevels. These lenses are incredibly helpful to patientswith macular degeneration. In addition to providingbenefits of protection from ultraviolet (UV) light, theyalso provide contrast enhancement as well as glarereduction. Although the full range of filters are suit-able for many ocular pathologies, usually the CPF 511and 527 lenses work best in patients with moderate toadvanced macular degeneration. Since the lenses areglass (and not high-impact polycarbonate plastic),they are best prescribed in the clip-on variety. Thesefilters are now available in plastic also, yet are stillphotochromatic (Chadwick Optical, White River Jct.,Vermont, U.S.A.) providing a nice alternative forprescription sunwear.

Other nonoptical devices which might beconsidered are large-print books, check register,clock, watch, playing cards, etc., to name a few.There are talking books, watches, and clocks as wellas specialized appliances for diabetics. Catalogues ofsuch devices can be given to the patient and family.Occupational rehabilitation to assist in training ofoptical devices as well as activities of daily life isoften quite useful for most patients with varyinglevels of visual impairment.

FUTURE IMPLICATIONS AND IMPROVEMENTS

New ApplicationsThe Useful Field of View test as mentioned is anextremely important tool in determining drivingsafety of older patients, particularly those with visualimpairment. Many times, the patient has expressed aninterest to continue driving, but the examiner feels thatthe patient may not be a good candidate even with abioptic telescope. This test gives objective resultsinforming both patient and family whether thepatient will be at risk for injurious accidents. Many


state department of motor vehicles are beginning touse this software.

Scanning laser ophthalmoscope (SLO) macularperimetry allows for the characterization of centralfield defects, i.e., macular scotomata. The presence orabsence of macular scotomata and their characteristicsare extremely important indicators of reading successand speed with low vision devices as well as per-formance with activities of daily living (12). Theconfocal SLO has graphic capabilities which allow aretinal map of the scotomata to be drawn by deter-mining the retinal location of visual stimuli directly onthe retina. The instrument obtains retinal imagescontinuously using near-infrared (780 nm) laser whilescanning graphics onto the retina with a modulatedvisionHeNe (633 nm) laser at the same time (13). Thus,the patient can see the stimuli and the investigator canview the stimuli on the retina. From these capabilities, apreferred retinal locus (PRL) can be identified and bothrelative and dense scotomata can be mapped.

Patients with macular scotomata do not oftenperceive black spots. Rather they say that letters orwords are missing in their central vision while readingor they simply have difficulties in functioning. Thepresence of these scotomata can decrease many areasof visual performance although the specific relation-ships between macular scotoma characteristics andvisual performance have not been identified (14).Therefore, it becomes useful to be able to map out thescotomata and to know the exact location of preferredretinal loci in order to begin the rehabilitative process.Traditional approaches attempt to determine a direc-tion of the eccentric view and then to basically repeatthis direction to the patient during training, etc. Manytimes this technique is effective, but often patients donot respond as well as predicted to the devices,training, visual performance task, and/or duringactivities of daily living. The SLO can essentiallydetermine the characteristics of the scotomata andtheir relationship to thePRL.Once thePRL is identified,the rehabilitative team can instruct and train the patienton better use of the PRL. Nilsson et al. (15) have alsoshown that using the eccentric view demonstrated inthe SLO, patients with macular degeneration and largeabsolute central scotomata can be successfully trainedand reading speeds increased by them using a new andmore favorable retinal locus.

Studies with the SLO have shown that there aredifferent shapes and patterns of scotomata from roundcentered on a nonfunctioning fovea to ring scotomasurrounding a functioning fovea to highly complexamoeboid shapes (12). While the majority of patientsdo have dense scotomata, it was found that if thescotomata are complex and surround the PRL bymore than two of its borders, these patients have themost difficulties in performing visual tasks when

compared with those with less encumbered PRL (12).This knowledge will aid in the prediction ofpatient success.

A recent technique for developing low visiondevices is called vision multiplexing (16). This tech-nique attempts to avoid or reduce some of thelimitations of traditional low vision devices such asreduced field of view from telescopes and magnifiersas well as loss of resolution from minifying devicesthat help increase field of view for mobility, etc. Visionmultiplexing combines the wide field of view andhigh-resolution capabilities in ways that allow func-tion to be both separate and useful for the user (16).This technique already lends itself to many electronicdevices which are almost always more costly. The ideahere would be to utilize vision multiplexing for use inlighter weight, less expensive, and more cosmeticallyappearing spectacles. An example would be a tele-scope built into the spectacle (not drilled in like abioptic telescope) consisting of two curved lensmirrors for the ocular and objective and two additionalcurved mirrors for field enhancement (17).

New TechnologiesEnhanced Vision, Inc. (Huntington Beach, California,U.S.A.) has perhaps made some of the greatest break-throughs for enhancing the quality of life of visuallyimpaired patients. They have developed some incred-ible devices utilizing the latest advances in opticaltechnology and continue to be at the forefront inthis arena.

The JORDY 2e is an amazing “virtual reality”system that immerses the patient in a video image. Atiny, color television camera is mounted in a head-borne device weighing less than 10 oz (Fig. 2).Designed for patients whose vision is worse than 20/200, this device ranges in power from as little as 1! toas much as 24! for distance and up to 50 times forreading. It is considered the all-in-one system becauseit enhances distance, can attach to a television orcomputer, and has CCTV capabilities for reading orwriting. For distance viewing, there is a wide (44 in.)field of view. The autofocus magnification has presetsettings as well as zoom switches which are operatedfrom an easy-to-use handheld control unit. Bothcontrast and brightness controls make the colorimage quite clear. For near viewing, the CCTVfeature allows the device to be placed in a portabledocking stand. The image is magnified depending onthe size of the television screen or computer monitor.The light requirements are low with the systemrequiring no additional light thereby resulting inminimum glare. The device is not designed to beword while walking, driving, or during any mobility.

TheMaxPorte is a device that allows the visuallyimpaired patient to readwith the ease of a CCTV, but is

308 PRIMO

portable. The system consists of two components: adigital magnifier that captures the information and apair of lightweight (4 oz) glasses that display themagnified image (Fig. 3). A patient would simplyplace the magnifier on any surface, either curved orstraight, and view the magnified image on the glasses.The image can be magnified up to 28 times and isavailable in black and white or color and is mostsuitable for patients whose vision is 20/100 or worse.It is a great solution for professional, students, andseniors. Concise brightness control makes the imageclear and crisp. There is also a special tracking guide(MaxTrake) which makes the magnifier move straightacross the page. The device is incredibly easy to use asthere are no connections or assembly required. Itoperates on a rechargeable battery and comes in asleek carrying case.

The Maxe is another innovative but lowerpriced magnifying system. It is a portable, handheldmagnifier that easily connects to any television orcomputer monitor to magnify words, pictures, andmore (Fig. 4). Using a 20 in. television, the device willmagnify in both black and white and color from 16 to28!. It is quite easy to use with either the right orleft hand and the image is virtually distortion free;

Figure 2 The JORDY 2e with docking stand (left) and head-borne (right).

Figure 3 The MaxPorte.


it can also be used on any surface, curved or straight.The three viewing options (low contrast/photo,high contrast/positive, high contrast/negative)make it suitable for most patients whose vision is20/100 or worse. The device can also now beconnected to a completely portable liquid crystaldisplay (LCD) 7 or 10 in. panel (MaxPanele).

The Flippere products are a new family ofdevices that magnify distance, intermediate and nearviewing. It is an innovative rotating camera that can beconnected to a standard television or monitor, to alightweight pair of electronic glasses (FlipperPorte),to a portable LCD panel (FlipperPanele), or placed ina desktop docking stand (Fig. 5). Magnification rangesfrom 6 to 50! and is in full-color or black/whiteoptions. The Flipper is great for students andprofessionals.

The Acrobate is the newest product fromEnhanced Vision and is a unique one of a kind oftechnology that will customize and memorize favoritesettings for each of the three viewing modes: self-viewing which is a camera that gives a true mirrorFigure 4 The Maxe.

Figure 5 FlipperPanele.

310 PRIMO

image for applying makeup, etc.; distance viewing;and near viewing. The camera is attached to a portablearm which can clamp on to any table or desk. Onceattached, it can provide up to 72! magnification. TheAcrobat is completely battery operated, making ittruly a portable device (Fig. 6).

The Amigoe is a revolutionary new portablemini CCTV. It is incredibly slim at less than 2 in. andweighs a mere 1.3 pounds (Fig. 7). The viewing screenis large at 6.5 in. with a tilting function to give the usermore flexibility to view images at a comfortable angle.Magnification ranges from 3.5 to 14! and connects toany TV for increased magnification and viewing area.The device includes a writing stand and carrying case.

Ocutech, Inc. (Chapel Hill, North Carolina,U.S.A.) continues to have innovative designs withseveral telescopic devices.

The Vision Enhancing System-AutoFocus (VESw-AF) is the first autofocus telescope available. It hasextremely high optical quality and consists of 4!telescope focusing from as close as 12 in. to opticalinfinity (Fig. 8). The autofocus component worksthrough computer-controlled infrared electroopticswhich measure the focusing distance and another

computer which moves the focusing lens to theproper position. The 4! telescope has an amazing12.58 field and is lightweight (2.5 oz). The device isworn on the top of a frame and is. The VES-AF comeswith rechargeable battery pack which can be wornaround the waist or in a pocket or purse. This device isuseful for distance, intermediate, and limited nearvision including activities such as driving, card/music playing, bird-watching, golf, etc., and worksbest for patients who have vision between 20/80 and20/200.

Figure 6 The Acrobate.

Figure 7 The Amigoe.


Currently under development is a binocularautofocus telescope. This device has the same quickautomatic focusing at distance and near, but has theadded capability of binocular viewing for an enhancedfield of view through the telescopes. While still underdevelopment, the device would be most beneficial forpatients who have equal or near equal acuity in botheyes and wish to use the device for both distanceand reading.

The VESw-Sport is a bioptic telescope that wasdesigned to address some of the major drawbacks ofconventional bioptic telescope systems, namely pooracceptance by patients and fitting problems by theprescriber (Fig. 9). It offers significant improvements inthe field of view, image brightness, and contrast whencompared with conventional bioptics of similarpowers. Probably, the nicest feature is the fully adjus-table mounting design which gives the prescriber fullcontrol over the positioning of the telescope. Availablein 4! and 6!, this innovative telescope is bestprescribed for patients with visual acuity between20/80 and 20/400. Though needs to be manuallyfocused, it offers a less expensive alternative to theautofocus telescope.

The VESw-MINI, also an innovative design, is aminiature 3! expanded field telescope (Fig. 10). It hasa wide 158 field of view combined with a very compactphysical design. The telescope is equivalent in sizeto small focusing galilean telescopes, but half the sizeof most expanded field telescopes. The optics are quitecrisp and bright and can be prescribed for monocularor binocular use. Other special features include itsunique design that minimized the ring scotomawhich can be characteristic of many telescopicsystems. Also, its field of view has been expandedhorizontally to provide extra added vision in the mostimportant lateral fields. The manual focus is quite fastwith capabilities of focusing from optical infinitydown to 12 in. covered in less than one completeturn. In addition to being extremely lightweight, ithas internal refractive corrections fromC12 toK12 D;eye piece corrections are available for other refractiveerrors. This telescope is a nice option for patientswhose vision is better than 20/200.

Optelec (Chelmsford, Massachusetts, U.S.A.), aleader in CCTVs, has developed a new line of thesemost popular electronic devices. Their new ClearViewline has an ergonomic design and is user-friendly. Thebest features are the easy-to-use fingertip controls onthe easy-glide X–Y table that deliver autofocus, dial-inzoom, normal text and reverse contrast modes, andchoice of monitor types.

The ClearView 517 series has all the bells andwhistles with exceptional brightness/contrast, vibrantcolors, and magnification from 2 to 50! (Fig. 11).The device delivers a full-color performance on anintegrated 17 in. LCD thin film transister (TFT)monitor with an adjustable arm or on a color orblack and white cathode ray tube (CRT) monitor. TheCCTV has an open platform for future productenhancements as well as the ability to add optionalfeatures such as alternate colors, position locator,Figure 9 VESw-sport telescope.

Figure 10 VESw-MINI expanded field telescope.

Figure 8 VESw-autofocus telescope.

312 PRIMO

windows/line markers, and an external personalcomputer (PC) switch.

The ClearNotee is a lightweight, flexiblesolution for those who use a laptop (Fig. 12). Thesystem offers 3 to 46! magnification and full-screenreading and writing capability. It also features fullautofocus and can quickly and easily adjust to nearand distance viewing—ideal for environments likeclassrooms or offices where taking notes and readingdistance boards is often necessary.

The TravellerCe is the newest addition to theOptelec family of devices. It is a completely portableCCTV with a bright 6.4 in. screen magnifying text andpictures up to 16 times (Fig. 13). The device also“stands up” to allow the user to write letters or takenotes. Additionally, it can be connected to a televisionfor even greater magnification.

Vision Technology, Inc. (St. Louis, Missouri,U.S.A.) has its own new line of close-circuit televisions,but its greatest mark on the electronic industry thus faris a device called the VIEWe. This collapsible CCTVfeatures a pop-up design and weighs only 15 lbs. Theautofocus camera easily moves 3608 on a horizontalplane and 2408 on a vertical plane for the most flexiblepositioning system. The monitor and controls can beFigure 11 ClearView 517 with integrated tiltable monitor.

Figure 12 The ClearNotee.


positioned directly in front of the user regardless ofheight. The VIEW is battery powered and is ideal forstudents, home, office, and travel (Fig. 14).

Pulse Data Humanware (Concord, California,U.S.A.), a leader in technologies for blind and visuallyimpaired patients, has developed two excitingnew technologies.

The Pocketviewere is a portable 7! electronichandheld magnifier that offers a full-color and highcontrast black/white display. It has a retractablewriting stand, built-in rechargeable battery in acompact 4 in.!3 in. compact viewing area (Fig. 15). Ittruly is pocket size and great for looking at details andprice labels, signing checks and credit card payments,writing up brief meeting notes, reading menus, maps,instructions on packets or bottles and viewing photo-graphs. It does all this by giving a larger field of viewthan a traditional magnifier of the same power.

Truly, one of the most innovative and uniquereading devices of late is the myReadere. It is arevolutionary reading system which is the only lowvision auto-reader. Having easy push-button controls,the camera basically takes a snapshot of the readingmaterial and displays it within three seconds andreads the text back out loud. It automates thereading process by allowing the user to set the speedand amount of the page that appears on the monitor.myReader is completely transportable and compact by

folding down, and is a great device for home, school,and office (Fig. 16). It is expensive, but well worth themoney for those patients needing this type ofadvanced technology.

Figure 13 The TravellerCe.

Figure 15 The Pocketviewere.

Figure 14 The Viewe.

314 PRIMO

Eschenbach (Ridgefield, Connecticut, U.S.A.)remains at the forefront of superior optical qualityand design of magnifiers, telescopes, reading glasses,and more. Some of the newest products include theNovese diffractive technology reading lens whichgreatly decreases the weight and optical aberrationsof high plus reading glasses also enhancing thecosmetic appearance of these spectacles. These newultrathin lenses are available in both monocular andbinocular designs. Monocular systems come in 3 to 6!; binocular range from 4 to 10 D (Fig. 17). Another greataddition to their line is an light emitting diode (LED)barlight attachable via a clip to glasses, magnifiers, etc.It is of the size of a pencil with a wireless power source

containing six white LEDs (Fig. 18). This addedillumination often reduces the amount of magni-fication needed by the user.

Designs for Vision, Inc. (Ronkonkoma, NewYork, U.S.A.) has always been at the forefront forproducing high optical quality devices for visuallyimpaired patients. In addition to their traditional lineof bioptics (Figs. 19 and 20), they have also becomequite innovative with reading devices. The Clear-Image II Telephoto Microscope and HighPowerMicroscopes (Fig. 21) are high-powered readingmicroscopes available in powers 8! (C32 D) to 20!(C80 D). These lenses allow low vision patients to read

Figure 16 myReadere in use (above) and folded (right).

Figure 17 Novese prismatic half-eyes (top) compared tostandard thickness (bottom). Figure 18 Eschenbach barlight clipped to spectacles.


at a greater distance from the eye than any othercomparable systems. The fields of view are quitelarge and lenses are virtually distortion free fromedge to edge, which is what makes them innovative.Because of the higher powers, they are most suitablefor patients whose vision is worse than 20/400.

Corning Medical Optics (Corning, New York,U.S.A.) has a full range of filters in its line of GlareCon-trol lenses (Fig. 22). The CPF 527! (extra) is slightlydarker than the standard CPF 527. This filter worksextremely well for increased contrast enhancement andadds additional glare reduction in patients withmoderate to advanced macular degeneration. Thesecond newest filter is called the CPF GlareCutterlens. This lens is fantastic for patients with earlymacular degeneration who do not need quite as muchcontrast enhancement, but who definitely need glarereduction. The lens also has less color distortion and amore attractive color for patientswho reject the cosmeticappearance of the CPF 511 and 527 series. Blocking 99%UVAand 100%UVB rays, the lens transmits 18%of lightin its lightened state and 6% in its darkened state.

Chadwick Optical now has the full range ofcontrast enhancement filters available in photo-chromatic plastic providing an excellent option foreither clip-ons or prescription sunwear.

SUMMARY POINTS

& Patient success with low vision devices is depen-dent upon a number of factors including age,physical and mental status, level and stability ofvisual acuity, patient’s dependency on others, andthe interval since visual loss.

& Resistance to low vision devices and thus limitedsuccess tend to be seen in those patients who havenot yet accepted or mourned their visual loss.Generally speaking, the more profound the visualloss, the more difficult finding means of enhancingvision becomes.

& Nonoptical devices may be the only mechanismacceptable by the patient to regain a small degreeof independence.

& The role of vocational rehabilitation and occu-pational therapy for orientation/mobilitytraining, activities of daily living, etc., shouldalways be considered for patients with advancedmacular degeneration.

& Support groups may also provide comfort andnew friendships in helping to cope with thevisual impairment.

& Sometimes it is best to wait for a low visionconsultation to when the patient seeks this carevoluntarily after it has been suggested.

Figure 19 Designs for Vision Standard 2.2! BIO II bioptictelescope.

Figure 20 Designs for Vision 3! bioptic telescope.

Figure 21 Designs for Vision ClearImage II telephotomicroscope.

Figure 22 Corning family of filters.

316 PRIMO

& Success with visual rehabilitation is always basedon identification and satisfaction of the visualrequirements and goals of the patient.

& There are exciting new applications and deviceshappening in the field of low vision/visual reha-bilitation. Much of the novelty utilizes the latesttechnology and will no doubt be of great benefit tomany visually impaired patients suffering frommacular degeneration.

WEB SITES OF COMPANIES FORFURTHER INFORMATION

& Enhanced Vision Systems—www.enhancedvision.com

& Optelec—www.optelec.com& Ocutech, Inc.—www.ocutech.com& Designs for Vision—www.designsforvision.com& Corning Medical Optics—www.corning.com& Vision Technology, Inc.—www.visiontechnology.

com& Chadwick Optical—www.chadwickoptical.com& Eschenbach—www.eschenbach.com

REFERENCES

1. Lavinsky J, Tomasetto G, Soares E. Use of a contact lenstelescopic system in low vision patients. Int J Rehabil Res2001; 24:337–40.

2. DeCarlo DK, Scilley K, Wells J, et al. Driving habits andhealth-related quality of life in patients with age-relatedmaculopathy. Optom Vis Sci 2003; 80(3):207–13.

3. Bowers AR, Apfelbaum DH, Peli E. Bioptic telescopes meetthe needs of drivers with moderate visual acuity loss. InvestOphthalmol Vis Sci 2005; 46(1):66–74.

4. Kelleher DK. Driving with low vision. J Vis Impair Blind1968; 11:345–50.

5. Lovsund P, Hedin A. Effect on driving performance ofvisual field defect. In: Gale A, Freeman MH,Haslegrave CM et al, eds. Vision in Vehicles. Amsterdam:Elsevier, 1989:323–9.

6. Wood JM, Dique T, Troutbeck R. The effect of artificialvisual impairment on functional visual fields and drivingperformance. Clin Vis Sci 1993; 8:563–75.

7. Owsley C, Ball K, Sloane ME, et al. Visual/cognitivecorrelates of vehicle accidents in older drivers. PsycholAging 1991; 6:403–15.

8. Owsley C, McGwin G, Jr., Ball K. Vision impairment, eyedisease, and injurious motor vehicle crashes in the elderly.Ophthalmic Epidemiol 1998; 5(2):101–13.

9. Owsley C, Ball K, McGwin G, Jr., et al. Visual processingimpairment and risk of motor vehicle crash among olderadults. JAMA 1998; 279(14):1083–8.

10. McCloskey LW, Koepsell TD,Wolf ME, Buchner DM.Motorvehicle collision injuries and sensory impairments of olderdrivers. Age Aging 1994; 23:267–72.

11. Szkyk JP, Pizzimenti CE, Fishman GA, et al. A comparisonof driving in older subjects with an without age-relatedmacular degeneration. Arch Ophthalmol 1995; 113:1033–40.

12. Fletcher DC, Schuchard RA, Livingston CL, et al. Scanninglaser ophthalmoscope macular perimetry and applicationsfor low vision rehabilitation clinicians. Ophthalmol ClinNorth Am 1994; 7(2):257–65.

13. Schuchard RA, Fletcher DC, Maino J. A scanning laserophthalmoscope (SLO) low-vision rehabilitation system.Clin Eye Vis Care 1994; 6(3):101–7.

14. Fletcher DC, Schuchard RA. Preferred retinal loci relation-ship to macular scotomas in a low-vision population.Ophthalmology 1997; 104:632–8.

15. Nilsson UL, Frennesson C, Nilsson SEG. Patients withAMD and a large absolute central scotoma can betrained successfully to use eccentric viewing, as demon-strated in a scanning laser opthalmoscope. Vis Res 2003;43:1777–87.

16. Peli E. Vision multiplexing: an engineering approach tovision rehabilitation device development. Optom Vis Sci2001; 78:304–15.

17. Peli E, Vargas-Martin F. In the spectacle-lens telescope forlow vision. Ophthalmic Technologies XII 4611. In: Proceed-ings of SPIE, 2002:129–35.


22

Retinal Prostheses: A Possible Treatment for End-StageAge-Related Macular DegenerationThomas M. O’Hearn and Michael JavaheriDoheny Eye Institute and Department of Ophthalmology, Keck School of Medicine,


Kah-Guan Au EongDepartment of Ophthalmology and Visual Sciences, Alexandra Hospital, Department of Ophthalmology,

Yong Loo Lin School of Medicine, National University of Singapore, The Eye Institute,

National Healthcare Group, Jurong Medical Center, Singapore Eye Research Institute, and Department


James D. Weiland and Mark S. HumayunDoheny Retina Institute, Doheny Eye Institute, Department of Ophthalmology, Keck School of Medicine,


INTRODUCTION

Age-related macular degeneration (AMD) is theleading cause of irreversible blindness amongpeople older than age 65 years in Western countries,with 700,000 patients newly diagnosed annually inthe United States, 10% of whom become legally blindeach year. AMD is estimated to affect more than 8million people in the U.S., with the prevalenceexpected to increase by the year 2020, when thepopulation over age 85 is expected to increase by107% (1,2). Once photoreceptors are severelydamaged, such as in end-stage AMD, no treatmentshave been shown to restore useful vision. An emer-ging modality of treatment has been visualrestoration through the use of implantable retinalprostheses (3–9). Devices are currently under evalu-ation for use in patients with light perception or nolight perception vision from retinitis pigmentosa (RP).The challenge for these prostheses is even greater inAMD patients where central visual acuity is oftenreduced only to the 20/400 level and peripheralvision spared, thus the prosthesis must be capableof improving this vision to justify the risk of implan-tation. A variety of modalities are currently underinvestigation, including electrical, neurotransmitter,and nanoparticle-based prostheses (10). Thesemethods can be further subdivided by location ofaction within the visual system and eye: extraocularlocations include the visual cortex, optic radiations,and optic nerve, and the exterior of the globe itself(4,6,8), while intraocular sites include the epiretinaland subretinal surfaces (3,5,7).

EXTRAOCULAR APPROACHES

Cortical ProsthesisBrindley and Dobelle were early pioneers in the fieldof artificial vision, being the first to demonstrate theability to evoke phosphenes and patterned percep-tions by electrical stimulation of the occipital cortexvia chronically implanted electrodes (4,11–13). Multi-electrode arrays were placed in the subdural spaceover the occipital cortex, in individuals blind fromdamage to visual pathways anterior to the visualcortex (11–13). Dobelle’s 64-channel platinum elec-trode surface stimulation prosthesis was shown toallow blind patients to recognize 6-in. characters at5 ft (approximately 20/1200 visual acuity) (4). Despiteachieving early successes, these experiments didpossess some significant difficulties in implantedpatients. Important shortcomings included control-ling the number of phosphenes induced by eachelectrode, interactions occurring between phos-phenes, as well as the need for very high currentsand large stimulating electrodes in order to achievephosphene perception. In some patients, this lead topain from meningeal stimulation and occasional focalepileptic activity following electrical stimulation.Importantly, subjects also reported the perception ofhalos surrounding each phosphene and not focaldistinct percepts (11–13).

Subsequent attempts at developing a corticalprosthesis used an intracortical approach in hopes ofovercoming the shortcomings of surface cortical stimu-lation via a lower current, higher fidelity system. Byemploying smaller electrodes in closer apposition to

the target neurons of stimulation, it was hoped that lesscurrent could be used and that resulting stimulationcould be more localized. Initial human trials, duringwhich the intracortical prosthesis was implanted forfour months, demonstrated the ability to producephosphenes which exhibited color (14,15).

The Illinois Intracortical Visual Prosthesis projectand the Utah Electrode Array are the two types ofintracortical prosthesis currently under development(6,14,15). The Illinois device, consisting of 152 intracor-tical microelectrodes, has been chronically implantedin an animal model. Experiments have shownthat receptive field mapping could be combined witheye-tracking to develop a reward-based trainingprocedure. Furthermore, the experimental animalwas able to be trained to use electrically-inducedpoint-flash percepts in performing memory saccadetasks (14,15). The Utah device consists of multiplesilicon spikes with a platinum electrode tip at eachend organized in a square grid measuring 4.2 mmby 4.2 mm (6). The chief advantage of a cortical visualprosthesis lies in the ability to bypass all diseasedvisual pathway neurons rostral to the primary visualcortex with the potential to restore vision to the largestnumber of patients.

Optic Nerve ProsthesisThe optic nerve is another site with the potentialfor the development of a visual prosthesis, as infor-mation from the entire visual field is represented inthis small area. Although well established surgicaltechniques allow this region to be accessed readily,there remain significant hurdles to overcomeregarding this approach. First, the optic nerve is anexceedingly dense neural structure with approxi-mately 1.2 million axons confined within a 2-mmdiameter nerve trunk. As a result of this density, aswell as the difficulty in translating the exact retinotopicdistribution of retinal ganglion cell axons within theoptic nerve, achieving focal stimulation of neuronswithin the optic nerve will be extremely difficult.Furthermore, implantation within this area requiresdissection of the dura, creating possible centralnervous system (CNS) complications including infec-tion and the possibility of compromising the vascularsupply of the optic nerve itself (16). Lastly, interven-tion at this point within the visual pathway requiresintact retinal ganglion cells and therefore is limited tothe treatment of outer retinal (photoreceptor) degener-ations (RDs).

Veraart et al. reported a study in which a volun-teer, with RP and no residual vision, was chronicallyimplanted with an optic nerve electrode connected toan implanted neural stimulator and antenna. By usingan external power source and controller, the subjectwas able to perceive phosphenes when the optic nerve

was stimulated. The volunteer used a head-wornvideo camera to explore a projection screen andunderwent performance evaluations during thecourse of a specifically designed training programwith multiple simple patterns. Throughout thecourse of the stimulation the patient was able todemonstrate pattern recognition a well as a learningeffect for processing time and orientation discrimi-nation (16). Recent work has also been directed atexploring intrapapillary placement of electrode micro-wires as an alternative means of creating an opticnerve-based prosthesis. Four platinum wire micro-electrodes were placed transclerally in the opticnerve head of rabbits. Stimulation via the electrodeswas able to produce recordable electrically evokedpotentials (EEP’), however, thresholds increasedsignificantly at one month of implantation, andhistopathology revealed encapsulation of the elec-trodes (17).

Scleral Based Extraocular StimulationA group form the University of New South Waleshas proposed and carried out early feasibility studiesin cats using both single electrodes and an electrodearray mounted external to the globe and fixed to thesclera. Thresholds were determined by production ofa cortical evoked response and at best were 100 mAwith a duration of 400 mS, yielding a charge densityof 1.27 mC/cm2, which is within safe limits. Althoughpotentially avoiding some of the issues of chronicintraocular implantation facing other approaches, theability of an implant mounted at the sclera to providefocal, high resolution stimulation remains a significantquestion (18).

A second scleral based approach for a prosthesisis being pursued by Yasuo Tano’s group at OsakaUniversity. Their prosthesis relies on suprachoroidal-transcleral stimulation (STS), and in its envisionedform the electrode array is implanted in a scleralpocket with a reference electrode in the vitreouscavity. Short-term stimulation studies using a ninechannel array in two retinitis pigmentosa patientswith light perception vision demonstrated the abilityto produce distinct phosphenes with a current of 0.3to 0.5mA. Charge densities were between 0.48 and 1.27mC/cm2 (53).

INTRAOCULAR APPROACHES

Epiretinal ProsthesisThe epiretinal approach to the retinal prosthesisinvolves the capture and digitization of images fromthe external world with a device such as a digitalcamera. In electrically based prostheses, capturedimages are transformed into patterns of focal electricalstimulation which are used to excite remaining,

320 O’HEARN ET AL.

viable inner retinal neurons. Morphometric studieshave been promising, documenting the survival ofas much as 90% of inner retinal neurons in non-neovascular and neovascular AMD, however it islikely that changes in neural architectures occur aswell which could complicate creating an effectiveprosthesis (19). As a means of sensory rehabilitationthe epiretinal prosthesis is analogous to commerciallyavailable cochlear implants, where degeneratedsensory hair cells are bypassed through the electricalstimulation of more distal neural elements.

Several groups have developed various designsof epiretinal implants that differ in terms of theintraocular and external components and theirmethod of enabling vision in patients. Each prototypeis guided by similar requirements: preserving as muchof the normal anatomy and physiology of the eye aspossible, while minimizing the amount of implantedelectronics required to power the device (20).

The research team at the Doheny Eye Institute ofthe University of Southern California, in conjunctionwith Second Sight Medical Products, Inc. (Sylmar, CA)and engineers from other universities as well as theDepartment of Energy National Laboratories, have

developed a prototype intraocular retinal prosthesis(IRP). This device, named the Second Sight Argus 1,consists of two main components: first, an extraocularunit for image capture and an implanted unit forretinal stimulation. The external unit is comprised ofa small, lightweight camera that is built into a pair ofglasses, and a wearable, battery-powered visualprocessing unit (Fig. 1). The camera captures video,which is then translated into a pixilated image bysoftware algorithms located in the visual processingunit. A transmitter coil sends image data to theimplanted unit. The electronics of the implanted unitare in a ceramic case implanted in the temporal bonesimilar to a cochlear implant. The electrical stimulationpattern is delivered, via a transscleral cable, from theceramic case to the intraocular portion of the pros-thesis (20,21).

The intraocular portion consists of a speciallydesigned array of 16 microelectrodes, ranging in sizefrom 250 to 500 mm, and made of platinum. Viableinner retinal neurons are stimulated by pulses frommicroelectrodes located on the array. The array ispositioned over the posterior pole and attached tothe inner retinal surface using a single tack, which is

Laser orRF

Implant

VideoCamera

EpiretinalImplant

Retina

Area ofPhotoreceptorsDestroyed byDisease

SubretinalImplant

PhotoreceptorsGanglion Cells

Figure 1 Illustration of functioning prostheses with

representation of epiretinal and subretinal implants.Source: From Ref. 20.

22: RETINAL PROSTHESES: A POSSIBLE TREATMENT FOR END-STAGE AGE-RELATED MACULAR DEGENERATION 321

inserted through the electrode array into the scleraduring the surgical procedure (Fig. 2) (20,21).

Preliminary tests of acute (!3 hours) epiretinalstimulation were performed on humans in the oper-ating room using hand held electrodes as well asmultielectrode arrays not affixed to the patients’retina, only after studies in several different animalmodels demonstrated that epiretinal stimulation couldsafely elicit reproducible neural responses in the retina(22). These patients perceived phosphenes in responseto the electrical stimulation to the retina and were ableto detect motion as well as identify shapes, amountingto a crude form vision (23,24). In 2002, at the DohenyRetina Institute of the University of Southern Cali-fornia, clinical trials testing chronic, long-termimplantation of the IRP began as part of a Food andDrug Administration Investigational Device Exemp-tion study. To date, six patients have safely receivedthe 16-electrode, Second Sight Argus 1 implants array(Figs. 3 and 4).

Chronically implanted patients described visualperceptions of phosphenes that were seen and shownto be retinotopically consistent when local currentwas applied to the surface of the retina with theimplanted electrodes. Patients ability to distinguishthe direction of motion of objects (24) and to discrimi-nate between percepts created by different electrodesbased on their locations, implies that retinotopicorganization is not lost when a patient loses sightdue to chronic disease. In addition, varying thestimulation level correspondingly enhanced thebrightness or dimness of percepts (25).

During the postoperative follow-up periods,electrically evoked responses (EERs) from the visualcortex and psychophysical tests eliciting visualperceptions in patients were subsequently recordedand added both quantitative and qualitative measuresof visual perception in implanted patients. These EERsand other psychophysical testing have also beenutilized preoperatively in potential patients not onlyfor screening purposes, but also for improved evalu-ation of critical parameters such as stimulationthresholds and current levels necessary for visualperception after the implant has been activated post-operatively (26).

Histological examination after chronic retinalstimulation showed no evidence of rejection,

Figure 2 Schematic representation of the Second SightTM,

Argus 1 intraocular retinal prosthesis apparatus, includingcamera, connector cable and microelectrode array.

Figure 3 Photograph of the Second Sight Argus 1 epiretinalmicroelectrode array prior to insertion into the vitreous cavity in a

patient with long-standing retinitis pigmentosa.

Figure 4 Color fundus photograph of the Second Sight Argus 1

epiretinal microelectrode array in a patient with long-standingretinitis pigmentosa. Note placement of array in the macula.


inflammatory reaction, neovascularization, or encap-sulation in both normal and retinal rod-conedegenerate dog subjects implanted with an epiretinalprosthesis. There still exists a possibility that theaffixed tack, in conjunction with the foreign materialof the array as well as electrical impulses delivered tothe implant, could result in fibrous encapsulation.However, histological analysis of the mechanicaleffects of the tack two to three months after implan-tation showed minimal effects on the retinal layerswith epiretinal implantation (22).

The second generation Second Sight implant,the Argus 2, which contains a 60 electrode epiretinalarray, has received an investigational device exemp-tion from the FDA to begin testing in patients in 2007.The Argus 2 contains four times as many electrodes asthe Argus 1, yet they are packaged into an area onefourth the size of the previous implant. Futureiterations of the implant will have even greater elec-trode numbers and densities and engineering towardsthis goal is already well underway with the goal of a1000 electrode implant. Psychophysical studies insighted volunteers simulating a 1000 electrode areain a 32x32 pattern has shown the ability for facialrecognition (54).

Since 1995, Rolf Eckmiller of Germany has ledand developed the Learning Retina Implant. Similar tothe epiretinal approaches discussed so far, theirimplant consists of both intraocular and extraocularcomponents. The retina encoder (RE) attempts tosimulate the typical receptive field properties andfiltering characteristics of retinal ganglion cells andreplaces the visual processing capabilities of the retinawith 100 to 1000 individually tunable spatiotemporalfilters. The RE output is then encoded and transmittedwirelessly to the implanted retina stimulator RS. TheRS consists of a ring-shaped, soft microcontact foilcentered at the fovea and affixed to the epiretinalsurface. The REs are then used to map visual patternsonto spike trains which are transmitted to thecontacted ganglion cells. The REs not only simulatethe complex mapping operation of parts of the neuralretina, but also provide an iterative, perception-baseddialog between the RE and human subject. Via thisdialog the implant is able to tune the various receptivefield filter properties with information “expected” bythe central visual system to generate optimal ganglioncell codes for epiretinal stimulation (29).

TheRE/stimulatorhas been successfully tested inanimal models and normally sighted subjects (30).While Eckmiller and his consortium have made signi-ficant advances in themanufacturing and testing of themicrocontact foils, as well as wireless signaling andenergy transfer mechanisms, they have followed acautious approach towards implanting the device inblindpatients (31). Theirgrouphas chosen to tacklefirst

the issues surrounding the information processingrequirements of the implant, retina and brain. In thefuture, it will be necessary to define the proper stimu-lation coding of electrically induced neural signals forthe retinal ganglion cells that are in contact with the RSif the dialog between the RS and the retina is to beoptimal (29). As the thrust of the German effort thus farhas been on the retinal encoder, clinical trials, primarilyfocusing on testing of the learning implant and dialog-based RE tuning, are just being initiated.

Dr. Hornig at the University of Hamburg-Eppen-dorf and Intelligent Medical Implants have alsopursued another version of an epiretinal prosthesis.Their prosthesis is implanted entirely within the orbitand has a 49 electrode epiretinal array affixed to theretina with a tack. Data presented to date in fourchronically implanted retinitis pigmentosa patientsindicate that the array can be well tolerated for up tonine months. Patients were able to distinguish pointsboth vertically and horizontally. The current iterationof the implant, however, does not contain a camera asa part of the implant (55).

Subretinal ProsthesisIn the subretinal approach to a prosthetic design, thearray is implanted between the bipolar cell layer andthe retinal pigment epithelium. Surgical access to thesubretinal space is through either an ab externo(scleral incision) or ab interno (through the vitreouscavity and retina) approach. Alan and Vincent Chowof Optobionics Corp., were among the first to implanta subretinal prosthetic. In their original conceptthe subretinal implant would function as a simplesolar cell without the need for an external powersource or signal processing (3,32,33). Their artificialsilicon retina (ASR) Microchip, which measures twomillimeters in diameter, and contains approximately5000 microelectrode-tipped microphotodiodes, ispowered solely by incident light. The microphoto-diodes convert incident light into electrical signalswhich are used to stimulate the remaining retinalneurons. To date the ASR Microchip has beenimplanted in six patients in a subretinal locationoutside of the macula, with a follow up of 6 to 18months. Chow et al. reported gains in visual functionin all patients as well as unexpected improvements inretinal areas distal to the implantation site. The studywas potentially biased by the lack of randomization ofwhich eye received the implant surgery, thus makingboth examiner and patient aware of which eyepossessed the implant, as well as the confounder ofthe rescue effect of vitrectomy and subretinal surgeryin RD (35). They noted that a larger clinical trialwould be necessary to further demonstrate safety ofthe ASR Microchip as well as to further validate theirresults (34).


Later experiments have shown that the conceptbehind this simple approach was actually not viablesince ambient light levels do not cause the micropho-todiodes to produce current sufficient enough todirectly stimulate retinal neurons (36). In fact, Chowet al. have abandoned the notion that the ASR Micro-chip is efficacious as a prosthetic device and currentlyhypothesize that the small amount of current deliveredfrom the implant, may be therapeutic through a neuro-protective effect to the deteriorating retinalphotoreceptors. Studies by Pardue et al. are ongoingto determine whether these effects are indeed neuro-protective and whether the effects are durable andreproducible. Preliminary work in Royal College ofSurgeons (RCS) rats showed similar amounts of photo-receptor sparing in sham operated and inactive deviceimplanted eyes as compared to eyes implantedwith theactiveASR (37). Therefore, the effects documentedwiththis type of an implant likely occur indirectly througha “growth factor” that then rescues the remainingphotoreceptors. Thus, this device is not a true retinalprosthesis but is best classified as a therapeutic device.Given these results Optiobionics is reevaluating itsefforts with the ASR.

Recently, Schuchard et al. presented data evalu-ating contrast sensitivity in 20 patients with RPimplanted with the subretinal ASR device. Luminanceand color were evaluated separately through modifiedcriteria. Although patients reported subjectiveimprovements in color vision and luminance afterimplantation, contingency analysis found that thecontrolled testing and self-reported rating responsesdid not generally agree. They concluded that self-reported improvements in color and contrast visionmay not correspond to controlled testing (38).

Eberhart Zrenner and a consortium of researchuniversities in Germany have been developing anothersubretinal implant since the mid-1990s. Their implantmeasures 3 mm across and consists of approximately7000 microphotodiodes in a checker-board patternconfiguration. Each microphotodiode has an area of400 mm2, and is made of biocompatible silicon andsilicon oxide, and designed to be both insulating andpermeable to light (36,39). Zrenner et al. have demon-strated in several RD animal models that subretinalstimulation with their implant is capable of inducingneuronal activity in retinal ganglion cells. Their efforthas been helped by first defining the parametersnecessary for successful electric stimulation and thenincorporating this data into the development of theirphotodiode arrays. Implanting their prosthesis inrabbits, cats, and pigs, they attempted to detect elec-trically stimulated activity in the visual cortex as aresult of retinal stimulation as well as investigatethe long-term biocompatibility and stability of theseimplants in the subretinal space (41–43). In chronic

implantation studies recording from epidural elec-trodes, they were able to record cortical evokedpotentials during stimulation with light flashes aswell as during electrical stimulation in the subretinalspace. However, it should be noted that in roughly onehalf of the animals tested, no cortical activation wasdetected subsequent to implantation. The presence ofsubretinal fluid around the array was observed duringexamination after implantation. This could potentiallyhave interfered with communication between theelectrodes and the neuronal architecture, as well asraised stimulus thresholds by increasing the distancebetween them. After 14 months, angiography andhistological findings of the retina adjacent to and inthe vicinity of the implant site revealed no significantforeign tissue rejection reactions or occurrence ofinflammation (44,45). More recent studies haveattempted to further refine the surgical implantationusing a combined ab interno/transcleral approach inpig eyes. By minimizing the intraocular component ofthe implantation to the creation of a small retinotomynecessary only to inject viscoelestic into the subretinalspace, trauma to the retina is potentially reduced.However, in the study perfluorocarbon liquids wereused to maintain retinal-implant apposition, thuslimiting conclusions as to the potential success of thisnew technique (46).

With evidence that the subretinal approach to aretinal prosthesis is not practical without an externalsource of energy to power the implant, the feasibilityof polyimide film electrodes in a cat model wasdemonstrated and further exploration of film-boundelectrical stimulation was planned (47). Prototypesof their subretinal device have attempted to addressthis issue using an external power source connectedto the subretinal implant by fine wires that arerun extraocularly until piercing the sclera. Futuremethods of powering the implant include transpupil-lary infrared (IR) illumination of receivers close tothe chip and electromagnetic transfer. CurrentlyZrenner and colleagues are planning to conducta clinical pilot study limited to 30 days and to 8completely blind RP patients.

Recently, Wilke et al. presented data related toperception of dots and patterns in two implantedpatients with long standing RP on behalf of theZrenner group. The influence of stimulation par-ameters on perception with respect to dot shape,size, color, duration, and pattern recognition wereinvestigated. They demonstrated for the first timethat electrical stimulation with subretinal implantedelectrodes is capable of eliciting reproduciblephosphenes of well defined shape, enabling clearpattern recognition. Changes in stimulation amplitudeor frequency led to modulation of perceived bright-ness and to lesser extent in dot size while shape or


color remained unchanged. In addition, stimulation ofadjacent electrodes elicited dots of comparable sizeshape and color (48). Zrenner’s group recentlypresented more data on the successful implantationof their 16 electrode subretinal array. The arrayswere implanted in six patients legally blind fromretinitis pigmentosa. Five of the six patients had thearrays removed at one month post-implantation asplanned due to their lack of hermetic packaging. Nocomplications related to the array implantation orexplantation were encountered. One patient refusedexplantation. Of note with the surgical technique wasthe need for silicone oil tamponade for successfulimplantation (56).

A third type of subretinal prosthesis has recentlybeen developed by Rizzo and Wyatt. Named theBoston Retinal Implant, the prosthesis is in an earlystage of development. Their studies so far havefocused on biocompatibility issues surrounding theeffects of a foreign material in the subretinal space, aswell as the development of surgical techniques forimplantation. Minimally invasive surgical techniquesutilizing a posterior, ab externo approach to implantthe prosthesis within the subretinal space have beentested. Results have been encouraging to date,however long-term biocompatibility studies inanimals as well as further refinement of the implan-tation technique will need to carried out prior toconducting a clinical trial in human patients (47).

As with other methods, the subretinal prosthesisapproach has its distinct advantages and disadvan-tages. One advantage is that the microphotodiodes of asubretinal prosthesis directly replace the functions ofthe damaged photoreceptor cells while the retina’sremaining intact neural network is still capable ofprocessing electrical signals. Placement of the subret-inal prosthesis in closer proximity to remaining viableinner retinal neurons may be advantageous in possiblydecreasing currents required for effective stimulation.Experiments conducted in RCS rats with subretinalarrays have shown significant migration of retinal cellsinto perforations within an array, as well as aroundprotruding electrodes, conceivably reducing thedistance between stimulating electrode and targetcell, and therefore stimulus thresholds. However,whether migrating cells will remain viable, and withan intact neural circuitry remains a major concern (49).Furthermore, in studies of electrical stimulation innormal and rd1 degenerate mice lower thresholdswith photoreceptor side stimulation, correspondingto a subretinally positioned implant, were onlyobserved in normal mice. Measured thresholds forepiretinal versus subretinal electrode placement werenot significantly different in rd1 retinas, questioningthe potential advantage of subretinal implants in termsof lowering thresholds (50).

Advantages of a subretinal approach include lesssurgically induced trauma upon implantation due to alack of mechanical fixation and the relative ease inpositioning and affixing of the microphotodiodes inthe subretinal space. However, potential thermalinjury is a significant factor due to the fact that thereis a limited area within the subretinal space which willcontain the microelectronics.

Additionally, the lack of an external source ofenergy for the microphotodiodes has been the greatestdeficiency of currently tested subretinal prostheses.Ambient light is not sufficient for the current gener-ated by a single microphotodiode, to stimulateadjacent retinal neurons. An additional source ofenergy is needed to allow for adequate current to beavailable to modulate the prosthesis for sufficientstimulation of neurons in a retinotopic distribution.

A group at Standford University has proposedan optoelectronic retinal prosthesis which attemptsto address the shortcomings already discussed forsubretinal microphotodiode based arrays. Theimplant consists of a subretinally placed microphoto-diode array connected to a projection system, althoughit could be compatible with an epiretinal system aswell. After the image is captured by video camera it isprocessed and then sent to a liquid crystal displaymicrodisplay emitting in the near IR. The image isthen bounced off a pair of transparent goggles worn bythe patient and scanned onto, thus stimulating, thephotodiode array. The external power source thusovercomes the fact that ambient light energy is insuffi-cient to power the photodiodes. Theoretically, thearray also has the advantage of allowing the patientto scan the visual field provided by the externalcamera with eye movements, as the projection arraycould be designed to track the implanted array,rather than by moving the externally mounted videocamera with head movements. The use of IR lightfor projection, a wavelength not detected by humanphotoreceptors, allows the remaining peripheralvision in AMD patients to be utilized to its fullestwithout interference from the implant (49).

Neurotransmitter-Based ProsthesesIn an attempt to mimic physiologic chemical signalingin the retina, the microfluidic retinal prosthesis hasbeen developed at the Kresge Eye Institute of WayneState University. This team, headed by Dr. RaymondIezzi, formulated a hypothesis that digital images canbe converted into neurochemical signals through amicrochip that provides chronic input to the CNSusing naturally occurring signaling molecules. Animportant aspect of this device lies in the fact that theneurotransmitter release can be used to stimulate theretina with or without electrical stimulation. By func-tioning with a concurrent electrical device, the


neurochemical transmission may help to decreasestimulatory thresholds, therefore decreasing theamounts of generated heat and tissue breakdown.In addition, this type of prosthesis offers the potentialto deliver therapeutic drugs. Potential therapeutics canbe stored as prodrugs, which can be photo-activatedinto their biologically active component, once a certainelectrical stimulation is received (52).

Future research by all groupswill need to addressthe long-term biocompatibility of the prostheseswithin the saline environment of the eye in terms ofhermetic packaging of the microfabricated arrays aswell as maintaining the stability of the retina-arrayinterface by avoiding any robust fibrotic or glialresponse from the retina. Issues specific to electricallybased arrays also include minimization of heat pro-duction by the arrays, Neurotransmitter arraysmeanwhile need to address issues of potential excito-toxic injury. Also included in these biocompatibilityissues is the unknown effect of chronic electricalstimulation on the retina. In addition to this, significantattention needs to be given to the manner in whichvisual images will be encoded and delivered inpatterns of electrical stimulation to the retina. Plasticityof the visual system in response to electrical stimu-lation as well as how the brain interprets a pattern ofstimulation resulting from sixteen, or in the futurethousands of electrodes is still not understood butwill be crucial in the evolution of prosthetic design.

Although significant advances have been made,the field of artificial vision is still relatively young.Withongoing advances in the technology of micro-electronics, hermetic packaging, surgical techniques,and in the understanding of the visual nervous systemsresponse to chronic stimulation, there remains hopethat advancement to restoring some vision to patientssuffering from end stage AMD will be possible in thefuture.

SUMMARY POINTS

& Once photoreceptors are damaged severely, such asin end-stage AMD, no treatments have been shownto restore useful vision to these patients.

& Despite early success, cortical prosthetics haveshown limitations in their ability to precisely stimu-late certain areas using large electrodes and highcurrents to achieve adequate perception. The Illi-nois Intracortical Visual Prosthesis project and theUtah Electrode Array are the two types of intra-cortical prosthesis currently under development.

& Optic nerve prostheses have shown limitedpromisein restoring visual perception. Some significanthurdles remain in perfecting this approach,including: translating the exact retinotopic

distribution, the most minimally invasive surgicalapproach, and vascular and dural compromise.

& The Second Sight Argus 1 epiretinal IRP (DohenyEye Institute) consists of and extraocular unit forimage capture, which converts video to a pixilatedimage which is then delivered to an intraocularportion consisting of a platinum 16 microelectrodearray. To date, six patients have safely receivedthe Model 1and have described visual perceptionsof phosphenes that were seen and shown to beretinotopically consistent when local current wasapplied to the surface of the retina with theimplanted electrodes. Other epiretinal modelsinclude the Harvard group and Germany’sLearning Retinal Implant.

& Alan and Vincent Chow of Optobionics Corp., wereamong the first to implant a subretinal prosthetic.Their ASR Microchip, which measures 2 mm indiameter, and contains approximately 5000 micro-electrode tipped microphotodiodes, is poweredsolely by incident light. To date the ASR Microchiphas been implanted in six patients in a subretinallocation outside of themacula, with a follow up of 6to 18 months. For multiple reasons, they haveabandoned the notion that the ASR Microchip isefficacious as a prosthetic device and currentlyhypothesize that the small amount of currentdelivered from the implant, may be therapeuticthrough a neuroprotective effect to the otherwisedying retinal photoreceptors.

& Zrenner and colleagues in Germany have beendeveloping another subretinal implant sincethe mid-1990s and have demonstrated in severalRD animal models that subretinal stimulation withtheir implant is capable of inducing neuronalactivity in retinal ganglion cells. Future plansinclude further refining the surgical approach tocreate a better communication between the elec-trode and retina.

REFERENCES

1. BresslerNM,Bressler SB,CongdonNG, et al. Potential publichealth impact of Age-Related Eye Disease Study results:AREDS Report No. 11. Arch Ophthalmol 2003; 121:1621–4.

2. Group AREDRS. A randomized, placebo-controlled,clinical trial of high-dose supplementation with vitaminsC and E, beta carotene, and zinc for age-related maculardegeneration and vision loss: AREDS Report No. 8. ArchOphthalmol 2001; 119:1417–36.

3. Chow AY, Pardue MT, Perlman JI, et al. Subretinalimplantation of semiconductor-based photodiodes: dura-bility of novel implant designs. J Rehabil Res Dev 2002;39:313–21.

4. Dobelle WH. Artificial vision for the blind by connecting atelevision camera to the visual cortex. Asaio J 2000; 46:3–9.

5. Humayun MS. Intraocular retinal prosthesis. Trans AmOphthalmol Soc 2001; 99:271–300.


6. Maynard EM, Nordhausen CT, Normann RA. The Utahintracortical electrode array: a recording structure forpotential brain-computer interfaces. ElectroencephalogrClin Neurophysiol 1997; 102:228–39.

7. Nadig MN. Development of a silicon retinal implant:cortical evoked potentials following focal stimulation ofthe rabbit retina with light and electricity. Clin Neurophy-siol 1999; 110:1545–53.

8. Shandurina AN, Panin AV, Sologubova EK, et al. Results ofthe use of therapeutic periorbital electrostimulation inneurological patients with partial atrophy of the opticnerves. Neurosci Behav Physiol 1996; 26:137–42.

9. Veraart C, Raftopoulos C, Mortimer JT, et al. Visual sen-sations produced by optic nerve stimulation using animplanted self-sizing spiral cuff electrode. Brain Res 1998;813:181–6.

10. Park RI. The bionic eye: retinal prostheses. Int OphthalmolClin 2004; 44:139–54.

11. Margalit E, Maia M, Weiland JD, et al. Retinal prosthesis forthe blind. Surv Ophthalmol 2002; 47:335–56.

12. Uhlig CE, Taneri S, Benner FP, Gerding H. Electricalstimulation of the visual system. From empirical approachto visual prostheses. Ophthalmologe 2001; 98:1089–96.

13. Weiland JD, Humayun MS. Past, present, and future ofartificial vision. Artif Organs 2003; 27:961–2.

14. Troyk P, BakM, Berg J, et al. A model for intracortical visualprosthesis research. Artif Organs 2003; 27:1005–15.

15. Uematsu S, Chapanis N, Gucer G, Konigsmark B,Walker AE. Electrical stimulation of the cerebral visualsystem in man. Confin Neurol 1974; 36:113–24.

16. Veraart C, Wanet-Defalque MC, Gerard B, Vanlierde A,Delbeke J. Pattern recognition with the optic nerve visualprosthesis. Artif Organs 2003; 27:996–1004.

17. Fang X, Sakaguchi H, Fujikado T, et al. Electrophysiologicaland histological studies of chronically implanted intrapa-pillary microelectrodes in rabbit eyes. Graefes Arch ClinExp Ophthalmol 2006; 244(3):364–75.

18. Chowdhury V, Morley JW, Coroneo MT. Stimulation of theretina with a multielectrode extraocular visual prosthesis.ANZ J Surg 2005; 75:697–704.

19. Medeiros NE, Curcio CA. Preservation of ganglion celllayer neurons in age-related macular degeneration. InvestOphthalmol Vis Sci 2001; 42:795–803.

20. Weiland JD, LiuW, HumayunMS. Retinal prosthesis. AnnuRev Biomed Eng 2005; 7:361–401.

21. Humayun MS, de Juan E, Jr, Dagnelie G, Greenberg RJ,Propst RH, Phillips DH. Visual perception elicited byelectrical stimulation of retina in blind humans. ArchOphthalmol 1996; 114:40–6.

22. Guven D, Weiland JD, Fujii G, et al. Long-term stimulationby active epiretinal implants in normal and RCD1 dogs.J Neural Eng 2005; 2:S65–73.

23. Humayun MS, de Juan E, Jr, Weiland JD, et al. Patternelectrical stimulation of the human retina. Vision Res 1999;39:2569–76.

24. HumayunMS, Weiland JD, Fujii GY, et al. Visual perceptionin a blind subject with a chronic microelectronic retinalprosthesis. Vision Res 2003; 43:2573–81.

25. Lakhanpal RR, Yanai D, Weiland JD, et al. Advances in thedevelopment of visual prostheses. Curr Opin Ophthalmol2003; 14:122–7.

26. Yanai D, Lakhanpal RR, Weiland JD, et al. The value ofpreoperative tests in the selection of blind patients for apermanent microelectronic implant. Trans Am OphthalmolSoc 2003; 101:223–8 (discussion 228–30).

27. Rizzo JF, III, Wyatt J, Loewenstein J, Kelly S, Shire D.Methods and perceptual thresholds for short-term electricalstimulation of human retina with microelectrode arrays.Invest Ophthalmol Vis Sci 2003; 44:5355–61.

28. Rizzo JF, III, Wyatt J, Loewenstein J, Kelly S, Shire D.Perceptual efficacy of electrical stimulation of humanretina with a microelectrode array during short-termsurgical trials. Invest Ophthalmol Vis Sci 2003; 44:5362–9.

29. Eckmiller R,NeumannD, BaruthO. Tunable retina encodersfor retina implants: why and how. J Neural Eng 2005;2:S91–104.

30. Walter P, Szurman P, Vobig M, et al. Successful long-termimplantation of electrically inactive epiretinal micro-electrode arrays in rabbits. Retina 1999; 19:546–52.

31. Gerding H, Horning R, Eckmiller R, et al. Implantation,mechanical fixation, and functional testing of epiretinalmultimicrocontact arrays (MMA) in primates [abstract].Invest Ophthalmol Vis Sci 2001; 42:S814.

32. Chow AY, Peachey NS. The subretinal microphotodiodearray retinal prosthesis. Ophthalmic Res 1998; 30:195–8.

33. Peyman G, Chow AY, Liang C, Chow VY, Perlman JI,Peachey NS. Subretinal semiconductor microphotodiodearray. Ophthalmic Surg Lasers 1998; 29:234–41.

34. Chow AY, Chow VY, Packo KH, Pollack JS, Peyman GA,Schuchard R. The artificial silicon retina microchip for thetreatment of vision loss from retinitis pigmentosa. ArchOphthalmol 2004; 122:460–9.

35. Del PrioreLV. Effect of shamsurgeryon retinal function aftersubretinal transplantation of the artificial silicone retina.Arch Ophthalmol 2005; 123:1156 (author reply 1156–7).

36. Zrenner E. Will retinal implants restore vision? Science2002; 295:1022–5.

37. Pardue MT, Phillips MJ, Yin H, et al. Possible sources ofneuroprotection following subretinal silicon chip implan-tation in RCS rats. J Neural Eng 2005; 2:S39–47.

38. SchuchardRA,ChowA,BarnesC,et al.Contrast sensitivity inpeople with the subretinal artificial silicone retina microchipdevice. Invest Ophthalmol Vis Sci 2006; 47 (E-abstract 3210).

39. Zrenner E, Stett A. Can subretinal microphotodiode arrayssuccessfully replace degenerated photoreceptors. VisionRes 1999; 39:2555–67.

40. www.eye-chip.com41. Gekeler F, Kobuch K, Schwahn HN, Stett A, Shinoda K,

Zrenner E. Subretinal electrical stimulation of the rabbitretina with acutely implanted electrode arrays. GraefesArch Clin Exp Ophthalmol 2004; 242:587–96.

42. Sachs HG, Gabel VP. Retinal replacement—the develop-ment of microelectronic retinal prostheses—experiencewith subretinal implants and new aspects. Graefes ArchClin Exp Ophthalmol 2004; 242:717–23.

43. Sachs HG, Schanze T, Wilms M, et al. Subretinal implan-tation and testing of polyimide film electrodes in cats.Graefes Arch Clin Exp Ophthalmol 2005; 243:464–8.

44. Sachs HG, Gekeler F, Schwahn H, et al. Implantation ofstimulation electrodes in the subretinal space to demon-strate cortical responses in Yucatan minipig in the course ofvisual prosthesis development. Eur J Ophthalmol 2005;15:493–9.

45. Schwahn HN, Gekeler F, Kohler K, et al. Studies on thefeasibility of a subretinal visual prosthesis: data fromYucatan micropig and rabbit. Graefes Arch Clin ExpOphthalmol 2001; 239:961–7.

46. Schanze T, Sachs HG, Wiesenack C, Brunner U, Sailer H.Implantation and testing of subretinal film electrodes indomestic pigs. Exp Eye Res 2006; 82:332–40.

47. Sachs HG, Schanze T. Transscleral implantation andneurophysiological testing of subretinal polyimide film


electrodes in the domestic pig in visual prosthesis develop-ment. J Neural Eng 2005; 2:S57–64.

48. Wilke R, Kuttenkeuler C, Wilhelm B, et al. Subretinalchronic multielectrode arrays in blind patients: perceptionof dots and patterns. Invest Ophthalmol Vis Sci 2006; 47(E-Abstract 3202).

49. Palanker D, Vankov A, Huie P, Baccus S. Design of a high-resolution optoelectronic retinal prosthesis. J Neural Eng2005; 2:S105–20.

50. O’Hearn TM, Sadda SR, Weiland JD, Maia M, Margalit E,Humayun MS. Electrical stimulation in normal and retinaldegeneration (rd1) isolated mouse retina. Vision Res 2006Oct; 46(19):3198–204.

51. Peterman MC, Mehenti NZ, Bilbao KV, et al. The artificialsynapse chip: a flexible retinal interface based on directedretinal cell growth and neurotransmitter stimulation. ArtifOrgans 2003; 27:975–85.

52. Iezzi R, Safadi M, Miller J. Feasability of retinal andcortical prosthesis based upon spatiotemporally controlled

release of L-glutamate. Invest Ophthalmol Vis Sci 2001;42:S815.

53. Fujikado T, Morimoto T, Kanda H, et al. Evaluation ofphosphenes elicited by extraocular stimulation in normalsand by suprachoroidal-transretinal stimulation in patientswith retinitis pigmentosa. Graefe’s Arch Clin ExpOphthalmol 2007 Mar 7 [Epub ahead of print].

54. Thompson RW, Barnett GD, Humayun MS, Dagnelie G.Facial recognition using simulated prosthetic pixelizedvision. Invest Ophthalmol Vis Sci 2003; 44(11):5035.

55. Richard G, Hornig R, Keser M, et al. Chronic EpiretinalChip Implant in Blind Patients With Retinitis Pigmentosa:Long-Term Clinical Results. IOVS 2007; 666:ARVOE-Abstract B290.

56. Sachs HG, Bartz-Schmidt K, Gekeler F, et al. TranschoroidalImplantation of Active Subretinal Implants in BlindPatients: Experience With the New Surgical Implantationand Explantation Procedures in the First Six Patients. IOVS2007; 4446:ARVO E-Abstract.


23

Retinal Pigment Epithelial Cell Transplantationand Macular Reconstruction for Age-RelatedMacular DegenerationLucian V. Del PrioreDepartment of Ophthalmology, Columbia University, New York, New York, U.S.A.

Henry J. Kaplan and Tongalp H. TezelDepartment of Ophthalmology and Visual Sciences, University of Louisville, Louisville, Kentucky, U.S.A.

INTRODUCTION

Age-related macular degeneration (AMD) is theleading cause of blindness in the elderly populationin the Western world (1). Ninety percent of AMDpatients who experience severe vision loss do so as aresult of choroidal neovascularization (2), which rep-resents growth of neovascular tissue from thechoriocapillaris, within Bruch’s membrane, and even-tually in the sub-retinal pigment epithelium (RPE)and/or subretinal space. Developing new treatmentsthat prevent or reverse vision loss in AMD is ofparamount importance due to the severe visual lossthat occurs with this condition and the knowledge thatdisease prevalence will increase with a shift demo-graphics of western populations to older ages.

The last decade has witnessed significantadvances in the management of exudative AMD.Several drugs have become available for treatment ofthis condition; the first approved therapy in the UnitedStates was photodynamic therapy with verteporfin.The most important recent advances in the manage-ment of exudative AMD have come from thedevelopment of anti-vascular endothelial growthfactor (VEGF) drugs, such an the anti-VEGF aptamerpegaptanib (3–16), which was the first anti-VEGFcompound approved for use for exudative AMD, theanti-VEGF antibody fragment ranibizumab (17–22),and the widespread off-label use of intravitreal beva-cizumab (23–30).

Despite these significant advances in themanagement of exudative AMD, there is a largeunmet need for many patients with this condition.More than 50% of patients do not respond to therapywith anti-VEGF drugs, and many patients withadvanced disease have loss of vision due to scarformation and altered subretinal architecture. Theselimitations have led to the investigation of alternative

treatment modalities for subfoveal exudative AMD,including subfoveal membranectomy with andwithout RPE transplantation or translocation (31–35)and macular translocation (36). Initial efforts toimprove vision with cell transplantation alone havenot met with success; reconstitution of the normalsubretinal architecture is necessary for visual improve-ment in these individuals. Ultimately this will requiremaculoplasty, which is defined as reconstruction ofmacular anatomy in patients with advanced visionloss in exudative AMD (37). In our view successfulmaculoplasty will require replacing or repairingdamaged cells (using transplantation, translocation,or stimulation of autologous cell proliferation);immune suppression (if allografts are used to replacedamaged cells); and reconstruction or replacement ofBruch’s membrane (to restore the integrity of thesubstrate for proper cell attachment). Successful macu-loplasty will build on prior development of surgicaltechniques for managing severe vision loss in AMDpatients with advanced subfoveal exudation. Thesetechniques include surgical excision of choroidalneovascularization (31–35,38–40); surgical excisioncombined with allograft transplantation of adult orfetal RPE (41–52) or iris pigment epithelium (53–64) ormacular translocation with or without choroidalmembrane excision (65–83).

Simple excision of the subfoveal neovascularmembrane in AMD leaves a large RPE defect underthe fovea due to the removal of native RPE along withthe surgically excised neovascular complex (84).Resulting persistent RPE defects lead to the develop-ment of progressive choriocapillaris and photo-receptor atrophy (85). Histopathology after subfovealmembranectomy alone shows absence of largeswatches of native RPE, combined with damage tothe outer retina, choriocapillaris atrophy and absenceor damage to the inner aspects of native Bruch’s

membrane (86,87). We have previously shown that thestatus of host Bruch’s membrane has a profound effecton the behavior of RPE transplanted after subfovealmembranectomy (37,88–95). Thus reconstruction ofBruch’s membrane is a necessary component forsuccessful maculoplasty (96). Herein, we review thecurrent status of efforts directed at macular reconstruc-tion in exudative AMD.

RATIONALE FOR RPE TRANSPLANTATION IN AMD

In 1991, Thomas and Kaplan reported two patientswho experienced significant visual improvement aftersurgical excision of subfoveal choroidal neovascular-ization from presumed ocular histoplasmosissyndrome (POHS) (32). One patient improved from20/200 preoperatively to 20/40, whereas the otherpatient improved from 20/200 to 20/25. de Juan andMachemer (38) had previously performed disciformscar excision in four patients with end-stage AMD, butthe manuscript by Thomas and Kaplan was the first todemonstrate that excellent visual acuity was possibleafter submacular surgery. Since these initial publi-cations, several authors have reported small,uncontrolled series of patients undergoing subma-cular surgery for AMD (32–34,97,98). The recentSubfoveal Surgery Trial demonstrated that subfovealmembranectomy alone was better than observation forpatients with POHS and an initial visual acuity betterthan 20/100, but was not better than observation forAMD patients with subfoveal neovascularization (39).Examination of the results of prior nonrandomizedstudies of submacular surgery suggests that signi-ficant visual improvement is limited aftersubmacular surgery for AMD, and that RPE removalmay be an important factor limiting postoperativevisual recovery. There appear to be significantdifferences in the postoperative visual recovery aftersubmacular surgery for patients with AMD versusPOHS. There could be many factors responsible forthis observed difference, including advanced patientage in AMD, disease within Bruch’s membrane inAMD, the size of the choroidal neovascular complex(larger in AMD eyes compared to POHS), the locationof the ingrowth site and the relationship of the chor-oidal neovascular membrane to the native RPE (99).Patients with POHS and other disorders may have abetter prognosis because the choroidal neovascularmembrane lies anterior to the RPE and can thus beremoved while leaving the native RPE intact (99).However in AMD eyes the choroidal neovascularcomplex is frequently deep to the native RPE, so thatsurgical membrane excision denudes Bruch’smembrane of native RPE.

The consequences of RPE removal duringsubmacular surgery are significant because removalof the native RPE leads to progressive choriocapillarisatrophy and limits visual recovery after submacularsurgery (100–103). The subfoveal choriocapillaris canbe perfused one to two weeks after submacularsurgery in AMD eyes but become non-perfusedwithout further surgery or laser photocoagulation(100). Thach et al. examined the choroidal perfusionafter surgical removal of subfoveal membranes in 12eyes of 11 AMDpatients (104). Stereoscopic fluoresceinand indocyanine green angiograms of the excision bedrevealed hypofluorescence with visible perfusion inthe underlying medium and large choroidal vessels inall eyes. On the basis of these observations the authorsconcluded that the choriocapillaris and small chor-oidal vessels were frequently abnormal or absent inthe bed of the removed neovascular membrane. Wecannot exclude the possibility that some non-per-fusion of the subfoveal choriocapillaris is present inpatients before submacular surgery. However, patientswho develop subfoveal choroidal neovascularizationexperience sudden and severe visual loss, demon-strating that the perfusion of the choriocapillaris issufficient to support good visual function, even if it isnot normal.

Experimental evidence suggests that the nativeRPE is removed with the choroidal neovascularizationin AMD. Grossniklaus et al. (105) examined specimensremoved from the subretinal space as part of theSubmacular Surgery Trial. Most of these patients (61out of 78) had AMD and the balance had POHS oridiopathic neovascularization. The specimenscontained fibrovascular tissue, fibrocellular tissue,and hemorrhage. Vascular endothelium and RPEwere the most common cellular constituents. Asexpected, the membranes from AMD patients weremore likely to be beneath the RPE and the size of theRPE defect was larger in AMD eyes. Histopathologicexamination of an eye from a patient who had under-gone surgical excision of a choroidal neovascularmembrane in AMD revealed an RPE defect in thecenter of the dissection bed with incomplete resurfa-cing of the RPE defect after surgery (87). Thus, AMDpatients are more likely to have a bare area of Bruch’smembrane after surgery and the RPE defect is morelikely to persist after surgery in these eyes.

There is extensive experimental evidencesuggesting that RPE removal at the time of submacularsurgery would lead to progressive atrophy of thesubfoveal choriocapillaris. Destruction of the RPEwith sodium iodate leads to changes in the RPE andchoriocapillaris within one week and marked chorio-capillaris atrophy within one month (106). In contrast,the choriocapillaris has a normal appearance in areaswhere the RPE still appeared healthy. Similar changes

330 DEL PRIORE ET AL.

are seen after intravitreal ornithine injection and inan experimental model of thioridazine retinopathy(107–110). Repopulation of Bruch’s membrane occursafter RPE removal in the cat with choriocapillarispreservation under the areas of healed RPE andchoriocapillaris atrophy in non-healed areas (111).RPE removal in the non-tapetal porcine eye yieldssimilar results (85,112,113). Bruch’s membranebecomes repopulated with a monolayer or multilayerof variably pigmented cells one month after surgicaldebridement of the RPE. In these regions, the outernuclear layer and outer limiting membrane remainedintact and the choriocapillaris appeared patent. Inregions of poor RPE healing the lumen of the chor-iocapillaris was collapsed and the choriocapillarisendothelium was separated from its basementmembrane (85,112,113).

Thus, a combination of experimental and clinicalstudies suggests that the following sequence of eventsoccurs after choroidal neovascular membrane excision(Fig. 1). Subfoveal surgery can be performed withoutdisturbing the native RPE in some eyes, but choroidalneovascular membrane excision results in a focal RPEdefect in AMD eyes and in some younger eyes withother diseases. If the native RPE is not disturbed,the underlying choriocapillaris will not undergosecondary atrophy. If the native RPE is removed, thedefect will heal by migration and proliferation of newRPE from the edge of the epithelial defect if the nativebasal lamina is intact. The proliferating RPE arehypopigmented, making it difficult to visualize thesecells in vivo. There are no changes in the choriocapil-laris if the area of the RPE defect is completelyand rapidly repopulated by hypopigmented RPE.

Figure 1 (Top): Schematic of sub-retinal pigment epithelium (RPE) choroidal neovascular membrane typical in

age-related macular degeneration. (Middle): Membrane removal denudes the native RPE from Bruch’smembrane and excises fragment of the inner aspects of Bruch’s membrane. Native RPE cannot heal the

epithelial defect completely in the absence of native basal lamina. (Bottom): Non-pigmented RPE may healthe defect partially in the presence of residual basal lamina, but will not heal a large epithelial defect completely in

the absence of basal lamina. A persistent RPE defect leads to atrophy of the subfoveal choriocapillaris.

23: RPE CELL TRANSPLANTATION AND MACULAR RECONSTRUCTION FOR AMD 331

Incomplete or delayed healing of the RPE defect willlead rapidly to atrophy of the choriocapillarisalthough the medium and large vessels of thechoroid can remain patent.

The functional consequences of subfoveal chor-iocapillaris atrophy are significant because atrophy ofthe subfoveal choriocapillaris is correlated with poorvisual recovery after surgery. Over 90% of AMD eyesand 37% of POHS eyes have atrophy of the subfovealchoriocapillaris after submacular surgery, and the rateof visual improvement to better than 20/50 was worsefor AMD eyes than for POHS (103,100). Within thePOHS subgroup, the subfoveal choriocapillaris wasperfused in 24 out of 38 (63%) eyes and non-perfusedin 14 out of 38 (37%) eyes. Best-corrected visual acuityimproved by at least two lines in 17 out of 24 (71%)perfused eyes and 2 out of 14 (14%) non-perfused eyes(pZ0.0089). Additionally, a best-corrected vision of20/100 or better was achieved in 18 (75%) of theperfused eyes and only 4 (29%) non-perfused eyes(p!0.05). Thus, both the final visual acuity andimprovement in visual acuity were correlated withpostoperative perfusion of the subfoveal choriocapil-laris (103,100).

RPE HARVESTING TECHNIQUE

We have previously described a method for harvestingand storage of intact adult human RPE sheets prior totransplantation (95). Briefly, human cadaver eyes arecleaned of extraocular tissue and the suprachoroidalspace is sealed with cyanoacrylate glue (114). A smallscleral incision is made 3-mm posterior to the limbusuntil the choroidal vessels are exposed. Tenotomyscissors are introduced through this incision into thesuprachoroidal space and the incision is extendedcircumferentially. Four radial relaxing incisions aremade in the sclera and the sclera is peeled awayfrom the periphery to the optic nerve with care notto tear the choroid. The eye cup is then incubated with25 U/mL of Dispase (Gibco) for 30 minutes, rinsedwith carbon dioxide free medium, and a circumfer-ential incision is made into the subretinal space alongthe ora serrata. The loosened RPE sheets are separatedfrom the rest of the ocular tissue and placed on a sliceof 50% gelatin on a 25!75!1 mm glass slide (FisherScientific, Pittsburgh, Pennsylvania) with the apicalRPE surface facing upwards. Contamination withchoroidal cells is avoided by directly visualizing theRPE sheets under a dissecting microscope while theyare being harvested. The glass slide containing thegelatin film with the RPE sheet is then placed in a100!15 mm polystyrene dish and incubated in ahumidified atmosphere of 5% CO2 and 95% air at378C for five minutes to allow the gelatin to melt andencase the RPE sheet. The specimen is kept at 48C for

five minutes to solidify the liquid gelatin and thenstored in carbon dioxide free medium (pHZ7.4) at48C. Harvested sheets are stained with cytokeratin toensure purity of the cell population.

Transmission electron microscopy shows intactRPE cells with well-developed microvilli, basal infold-ings and intercellular connections (95). The initialviability of intact RPE sheets is 86% with a progressivedecline in viability with increased storage time. Cellsharvested within 24 hours after death maintain greaterviability than those harvested after 24 hours (p!0.05)and maintain 82% viability for as long as 48 hours ifstored at 48C.

BRUCH’S MEMBRANE CHANGES IN AMD

At the light microscope level Bruch’s membraneappears to be a continuous structure that extendsfrom the peripapillary area to the peripheral oraserrata. This anatomic structure was recognized bylight microscopists in the 19th century on the basis ofthe staining pattern on light microscopy. The develop-ment of transmission and scanning electronmicroscopy revealed that human Bruch’s membraneis a pentilaminar structure composed of a centralelastin membrane, surrounded by collagen layersbordered externally by the basement membrane ofthe RPE and choriocapillaris (Fig. 2). From internal toexternal, Bruch’s membrane has five anatomic layerswith known structure and function. The innermost(i.e., closest to the RPE and furthest from the externalsclera) layer of Bruch’s membrane is the RPE basallamina, which serves as the anchoring surface for theRPE. Throughout the human body, basal lamina arethin acellular membranes, approximately 50 nm inthickness, that line one side of epithelia (115). Theinner aspect of the RPE basal lamina is bordered by

RPE-BL

ICL

OCLCC-BM

EL

Figure 2 Anatomic layers of human basement membrane.Abbreviations: CC-BM, choriocapillaris basement membrane;

EL, elastin layer; ICL, inner collagen layer; OCL, outer collagenlayer; RPE-BL, basal lamina of the retinal pigment epithelium.


the RPE plasma membrane; the outer surface bordersthe inner collagen layer and fibers from the innercollagen layer extend into the basal laminar layer ofBruch’s membrane. Next comes the inner collagenlayer, which is a dense collagen matrix that intercon-nects the basal lamina and elastin layers of Bruch’smembrane. Most of the dysfunction within AMDstarts in the inner collagen layer. For example,drusen-like material can accumulate either on theinner or outer aspect of the basal lamina layer; softdrusen, which represent accumulation of abnormalmaterial within the inner collagen layer, will split theinner collagen layer from the basal lamina layer.Choroidal neovascularization often invades thistissue plane, essentially splitting the inner collagenlayer from the basal lamina as it grows and progresses(116). Proceeding externally, the elastin layer is notcontinuous in humans from the optic nerve to the ora.In advanced AMD the elastin layer becomes frag-mented (84). The outer collagen layer is similar to theinner collagen layer at the ultrastructural level. Struc-tural changes occur within the outer collagen layer as afunction of advancing patient age, including collagencross-linking. However extracellular deposits,including drusen and lipid deposits, appear to sparethis layer and accumulate mainly within the innercollagen layer as described above. Lastly, the basallamina of the choriocapillaris separates the chorioca-pillaris from the outer collagen layer. Unlike the basallamina of the RPE, there is no evidence of depositformation on either side of the choriocapillaris basallamina as a function of advancing patient age(117,118).

Basal lamina are typically thin acellularmembranes, approximately 50 nm in thickness,composed of collagen IV and XVIII, laminin,nidogen, agrin, and perlecan (115). Collagen I, collagenIII and fibronectin are present within the inner andouter collagen layer; the RPE and choriocapillarisbasal lamina are composed largely of laminin andcollagen IV (119), but also contain collagen V andheparan sulfate proteoglycan (120). The elastin layercontains elastin and collagen VI (120). Fibronectin isassociated with basement membranes, collagen fibersand elastic fibers throughout Bruch’s membrane,although precise immunolocalization of fibronectin ishampered by the fact that this is a soluble serumprotein. There is differential distribution of differentcollagen IV a-chain isoforms within human Bruch’smembrane; a1 (IV) and a2 (IV) chains were identifiedin 55% of RPE basement membranes and 100% ofchoriocapillaris basement membranes, respectively(121). RPE basement membranes also contained a3(IV), a4 (IV), and a5 (IV) chains, but these chains arenot present within choriocapillaris basal lamina. Thea6 (IV) chain was not identified in any sections (121).

Collagen XVIII is also present within the inner aspectsof human Bruch’s membrane; interestingly, endo-statin, which is the C-terminal fragment of collagenXVIII, is typically released from collagen XVIII viaproteolysis (122). Administration of intravitrealendostatin can inhibit experimental choroidal neovas-cularization (123) and gene therapy to increase levelsof endostatin prevent the development of choroidalneovascularization in AMD (124). Mice lacking base-ment membrane collagen XVIII/endostatin havemassive accumulation of sub-RPE deposits withstriking similarities to basal laminar deposits,abnormal RPE, and age-dependent loss of vision (125).

CLINICAL RESULTS OF RPE TRANSPLANTATION,RPE, AND MACULAR TRANSLOCATION

The goal of RPE transplantation is to repopulateBruch’s membrane with donor RPE prior to thedevelopment of widespread atrophy of the chorioca-pillaris. There is some preliminary experimentalevidence suggesting that RPE transplanted into adebrided bed will support the native choriocapillarisand healthy RPE may reverse choriocapillaris atrophyafter it develops (126). To date all human studies ofRPE transplantation for exudative AMD have beenperformed at the same time as submacular surgery,rather than after subfoveal choriocapillaris atrophyhas progressed. Much can be learned from clinicaltrials of macular translocation as well, since thissurgery tests the hypothesis that rotation of the foveaover areas of healthy RPE and choriocapillaris can leadto significant visual recovery. The following studieshave been reported to date:

& Peyman et al. performed submacular scar excisionwith translocation of an autologous RPE pedicleflap or transplantation of an allogeneic RPE-Bruch’s membrane explant in two patients (51).The final visual acuity was 20/400 in the firstpatient and count fingers at 2 ft in the secondpatient. Neither of these patients was immunesuppressed.

& Algvere et al. initially reported subretinalmembrane removal with transplantation of fetalhuman RPE patches in five AMD patients andsubsequently reported on a larger series of 17 eyes(41,127,128). Cystoid macular edema developedand the grafts became encapsulated by whitefibrous tissue within several months after surgerybut none of these patients received systemicimmune suppression. Scanning laser ophthalmo-scopic microperimetry demonstrated that patientswere able to fixate over the area of the RPE graftimmediately after surgery, but an absolute scotomadeveloped in this region several months after


surgery. These results are not surprising because thepatients were not immune suppressed, and RPEtransplanted into the subretinal space will berejected (46,129–131). The authors observed betterintegrity of the graft margins in geographic atrophypatients, suggesting that rejection may be morecommon in exudative AMD. Long term, there iscontinued deterioration of function and graft integ-rity in all cases of exudative AMD and five out ofnine eyes with nonexudative AMD (41,127,128).

& Subfoveal membranectomy with transplantation ofadult human RPE sheets has been performed in 11AMD patients who were immune suppressed post-operatively with prednisone, cyclosporine andimmuran (49). Eligibility criteria included thepresence of drusen, patient ageO60, a best-correctacuity of %20 out of 63 (Bailey-Lovie chart) andsubfoveal neovascularization %9 disc areas onpreoperative fluorescein angiography. The meanvisual acuity, contrast sensitivity and readingspeed did not change significantly for six monthspostoperatively. Transplants showed no signs ofrejection in patients able to continue immunesuppression for the first six months after surgerybut patients who discontinued immune suppres-sion developed signs of graft rejection two weekslater. Histopathology is available on an 85-year-oldfemale who died four months after RPE sheettransplantation (45). A complete autopsy demon-strated the cause of death to be congestive heartfailure.Apatch of hyperpigmentationwas visible atthe transplant site under the foveola after surgery.Mound-like clusters of individual round, largedensely pigmented cells were present in the subret-inal space and outer retina in this area, and thetransplant site did not contain a uniformmonolayerin most areas. There was loss of the photoreceptorouter segments and native RPE in the center of thetransplant bed, with disruption of the outer nuclearlayer predominantly over regions of multilayeredpigmented cells. Cystic spaces were present in theinner and outer retina. A residual intra-Bruch’smembrane component of the original choroidalneovascular complex was present under the trans-plant site. The poor morphology at the transplantsite was consistent with the lack of visual improve-ment seen after surgery in this patient.

& Weisz et al. delivered a patch of fetal RPE under theretina in one patient with geographic atrophy (132).Visual acuity remained stable at 20/80 one monthsurgery but deteriorated to 20/500 by five monthspostoperatively. Mild subretinal fibrosis developedafter surgery. The patient demonstrated a systemicimmune response to phosducin and rhodopsinpostoperatively in the absence of systemicimmune suppression.

& Binder and coworkers have reported on 53 eyesundergoing subfoveal surgery for choroidalneovascularization in AMD (14 undergoing subfo-veal membranectomy alone and 39 undergoingmembranectomy with transplantation of autolo-gous RPE suspensions) (43,133). There was nodifference in visual acuity postop between thegroups, but postop reading vision was better inthe transplant eyes and the recurrence rate ofchoroidal neovascularization was low (43,133).

& Van Meurs reported short-term results with patchtransplantation techniques in which a free pediclegraft was harvested from the mid periphery andplaced under the fovea immediately after subfo-veal membranectomy. Their initial reportconcluded that surgery was technically feasiblebut was associated with a high surgical compli-cation rate, with retinal detachment due toproliferative vitreoretinopathy in three out ofeight eyes (134,135). Wolf et al. reported temporaryimprovement in vision in only one out of sevenpatients (136). In a larger series, Joussen et al.reported on autologous translocation of RPE andchoroid in 45 eyes of 43 patients with subfovealAMD. Surgical complications were significant,with half the eyes requiring addition proceduresdue to retinal detachment, proliferative vitreoreti-nopathy, macular pucker, or vitreous hemorrhage.Only four eyes achieved a 15 letters increase in bestcorrected visual acuity. The authors claimed thatthe graft was revascularized on indocyanine greenangiography in most eyes, although definitiveproof of revascularization awaits animal studieswith histopathology (137).

& Stanga et al. reported on nine eyes with exudativeAMD undergoing subfoveal surgery combinedwith patch RPE transplantation (82,83). Theirinitial paper reported transient fixation over thegraft by scanning laser ophthalmoloscopy.However, long-term follow-up of four of thesepatients demonstrated that recovery of fixation istemporary, with a decline in fixation ability long-term, despite the fact that areas of hyper pigmenta-tion, interpreted by the authors as representing ahealthy graft, could still be seen ophthalmoscopi-cally.

& Prior workers have emphasized the fact that muchcan be learned about the potential of RPE trans-plantation from studying macular translocationresults. For exudative AMD, the macular transloca-tion series suggests that approximately 20% ofpatients can achieve a final vision of 20/50 orbetter (65–68,78,136,138–155). However, severalfacts should be recalled before inferring theresults of RPE transplantation on the basis of


macular translocation studies alone. First, thesurgical complication rate of macular translocationsurgery is initially quite high, with a steep learningcurve that may be surgeon-dependent. Second, inmacular translocation surgery the fovea is shiftedto a new location over healthy RPE and chorioca-pillaris; the native RPE is already attached to hostBruch’s membrane, thus avoiding issues that arisewhen RPE is translocated or transplanted to a newlocation. Third, macular translocation surgerycauses a significant decline in the global electro-retinogram (ERG) that may reflect the significanteffects of this surgery on overall retinal function(81,156,157). The situation is further complicatedby the observation that subfoveal atrophy recurredin three of four patients with geographic atrophy inAMD, thus implying that changes in the outerretina may be responsible for the development ofgeographic atrophy (140,158). This finding hassignificant implications for RPE transplantation ingeographic atrophy, since one would expect similarrapid loss of RPE in AMDwith geographic atrophyafter transplantation. Rapid recurrence ofgeographic atrophy was also observed byKhurana et al. after macular translocation (159).

MECHANISM OF RPE ATTACHMENT TOHUMAN BRUCH’S MEMBRANE

RPE Attachment in Tissue CultureSeveral investigators have characterized the ligandsavailable for surface attachment of human RPE. Thebasal surface of RPE cells contains a b1-subunit ofintegrin (160,161) and the inner aspect of Bruch’smembrane contains laminin, fibronectin, heparansulfate and collagen (120). Attachment of RPE tocoated artificial surfaces can be mediated by aninteraction between the b1-subunit of integrin andknown extracellular matrix (ECM) molecules. Forexample, RPE cells bind to Petri dishes coated withlaminin or fibronectin but do not attach to untreated,uncoated Petri dishes (161). The synthetic tetrapeptidearginine-glycine-aspartate-serine (RGDS), which isderived from the cell binding domain of fibronectin,decreases RPE binding to laminin-coated or fibro-nectin-coated dishes (162). Thus, in vitro bindingstudies suggest that RPE can attach to laminin orfibronectin coating a plastic surface via an interactionbetween the b1-integrin subunit and lamininand fibronectin.

Molecular binding studies demonstrate a role forintegrins and ECM ligands in mediating RPE attach-ment to RPE-derived ECM and human Bruch’smembrane in a more direct fashion (Fig. 3) (163). Theattachment rates of human RPE cells to RPE-derived

ECM was 66.0G6.0%. Coating the surface withalbumin or an irrelevant anti-IgG antibody did notchange the attachment rates significantly (64.5G3.0%and 63.5G3.4%, respectively; pO0.05 for eachcompared to ECM alone). The addition of fibronectin,laminin, type IV collagen or vitronectin increased theattachment rates to 79.0G7.0%, 76.0G6.0%, 80.3G9.0%, or 81.3G6.3%, respectively (p!0.05 for eachcompared to ECM alone). The addition of anti-fibro-nectin, anti-laminin, anti-collagen IV, or anti-vitronectin (1:100 dilution) decreased the attachmentrates to 56.2G3.0%, 49.4G5.0%, 55.2G4.1%, or51.0G7.3%, respectively (p!0.05 for each comparedto ECM alone). Increasing the concentration of anti-bodies to a ratio of 1:10 dilution did not inhibit RPEreattachment further (data not shown). Simultaneousaddition of anti-fibronectin, anti-laminin, anti-collagenIV and anti-vitronectin antibodies (1:100 dilution)markedly decreased the attachment rates further to25.3G9.0% (p!0.05). Treatment with RGDS, a tetra-peptide known to block the interaction between theb1-subunit of integrin and ECM proteins, markedlydecreased the RPE reattachment rate to 21.0G6.3%(p!0.05). Treatment of RPE cells with anti-b1 integrinantibodies before plating the cells decreased theattachment rate to 15.0G7.0% (p!0.05). The reattach-ment rate of RPE to uncoated tissue culture plastic was24.6G3.2%. The mechanism of attachment of RPE tohuman Bruch’s membrane explants is similar (163).

Importance of RPE Attachment for Cell SurvivalWe have previously demonstrated that RPE harvestedfor transplantation must be allowed to reattach to asubstrate to prevent RPE apoptosis (93). Secondpassage human RPE were plated onto tissue cultureplastic precoated with ECM, fibronectin, laminin,uncoated tissue culture plastic, untreated plastic anduntreated plastic coated with 4% agarose. Reattach-ment rates were determined for each substrate24 hours after plating. The TUNEL (terminal deoxy-nucleotidyl transferase-mediated dUTP nick endlabeling) technique was used to determine apoptosisrates in attached cells, unattached cells and the entirecell population. Attachment rates were as follows:ECM-coated tissue culture plastic O fibronectin-coated tissue culture plastic O laminin-coated tissueculture plastic O uncoated tissue culture plastic Ountreated plastic O agarose-coated untreated plastic.Apoptosis rates for the entire cell population wereincreased as the RPE cell attachment rate decreased,and the proportion of apoptotic cells in the entirepopulation was inversely related to the percentattached cells (rZK0.95). These results imply thatRPE cells removed from their substrate prior to trans-plantation must reattach rapidly to a substrate toprevent apoptosis.


Effects of Prior Handling In Vitro on RPE IntegrinExpressionPrior to successful cellular transplantation it is necess-ary to harvest donor RPE and passage these cells intissue culture to obtain the number of cells required fortissue transplantation. It is known that culturingconditions including the composition of the substratecan have a significant impact on cell behavior. Sinceintegrins on the RPE surface are involved in attach-ment of RPE to native basal lamina, we havedetermined the effects of passaging in tissue cultureand exposure to different ECM ligands on the

expression of integrin receptors in human RPE (88).First passage confluent human RPE were harvestedand plated into 96-well plates coated with differentECM components, including laminin (4 mg/cm2),fibronectin (10 mg/cm2), vitronectin (4 mg/cm2),collagen type IV (10 mg/cm2), or a mixture of all fourligands after nonspecific binding was blocked byadding 1% normal goat serum. Mouse antibodiesagainst specific human integrin subunits were usedas a primary antibody and alkaline-conjugated goatanti-mouse IgG was used as a second antibody. Theabsorbance at 405 nm was used to determine the

Ant

i-FN

,LN

,CIV

,VN

%A

ttach

men

t

0

10

20

30

40

50

60

70

80

90

100

BM

Onl

y

FN LN

CIV

Ant

i-FN

Ant

i-LN

Ant

i-CIV

Ant

i-VN

RG

DS

Ant

i-1

Alb

umin

Ant

i-IgG V

N

Figure 3 Human retinal pigment epithelium (RPE) cell attachment rate to Bruch’s membrane in tissueculture was determined. Data are divided into four groups: (i) group 1, open bars: the attachment rate on

extracellular matrix alone (64.0C9.5%) was not altered by the addition of albumin (63.4C5.7%) or anirrelevant anti-IgG antibody (62.8C5.3%), (ii) group 2, shaded diagonal stripes: the addition of

extracellular matrix components increased the attachment rate over baseline. The attachment rate

after the addition of fibronectin (72.4C6.8%), laminin (68.2C4.6%), collagen type IV (74.6C5.7%), orvitronectin (75.6C4.9%) was always higher than the attachment rate to Bruch’s membrane alone

(64.0C9.5%; p!0.05 for all comparisons), (iii) group 3, solid grey: the addition of antibodies againstBruch’s membrane components decreased the attachment rate over baseline. The attachment rate after

the addition of anti-fibronectin (47.0C10.0%), anti-laminin (52.6C9.6%), anti-collagen type IV (51.5C3.0%), or anti-vitronectin (42.2C3.3%) was always lower than the attachment rate to Bruch’s membrane

alone. Simultaneous addition of anti-fibronectin, anti-laminin, anti-collagen IV, and anti-vitronectinmarkedly decreased RPE reattachment to extracellular matrix (35.6C4.1%; pZ0.05), and (iv) group

4, black: striking inhibition of cell reattachment was produced by addition of the synthetic peptide RGDSor by preincubating the cells with anti-b1-integrin (33.3C3.0% and 37.4C5.9%, respectively). In addition,

cell attachment to uncoated tissue culture plastic was 25.8C4.5%. Data presented as meanGstandarddeviation, nZ9. Abbreviations: anti-b1, antibody to b1-subunit of integrin; anti-IgG, an irrelevant IgG

antibody; C IV, type IV collagen; FN, fibronectin; LM, laminin; RGDS, arginine-glycine-aspartate-serine;VN, vitronectin. Source: From Ref. 163.


relative expression of RPE integrin receptor subunitsunder each condition compared to primary RPE.Significant amounts of b1, b3, b4, and a5b1 integrinmolecules were detected on the surface of primaryRPE. Passaging RPE in tissue culture markedlyincreased the detection of b2 integrin subunits on theRPE surface, with a smaller increase in the detection ofa2, a3, and aVb6 integrin subunits and decrease in b1and a4. Coating with amixture of all four ECM ligandsincreased the amount of b1, b2, b3, b4, a5, a2b1, anda5b1 detected on the RPE surface compared to RPEseeded onto tissue culture plastic. Coating with vitro-nectin alone increased the amount of b1, b3, and b4.Coating with collagen IV alone increased theexpression of b2. b4 was increased by exposure toeither fibronectin, collagen IV, vitronectin or to amixture of all four ligands. These results demonstratethat at least two factors can control the expression ofintegrin subunits on the surface of human RPE;namely, passaging in tissue culture and seeding ofpassaged cells onto different ECM substrates. Sincethe attachment of human RPE to Bruch’s membrane ismediated partially by an interaction between integrinsubunits and the underlying surface, surface-inducedchanges in integrin distribution may have a profoundeffect on the initial attachment and subsequentbehavior of RPE seeded onto Bruch’s membrane.Pretreatment of RPE may alter the attachment andsubsequent behavior of RPE transplanted onto humanBruch’s membrane.

EFFECTS OF AGE-RELATED CHANGES WITHINBRUCH’S MEMBRANE ON RPE ATTACHMENTAND BEHAVIOR

Importance of Bruch’s Membrane Layer and AgeIn RPE AttachmentThe inner layers of Bruch’s membrane are not intactafter submacular membranectomy in AMD eyes(164,165). Histopathologic evidence suggests that theRPE basal lamina is excised with the choroidal neovas-cular membraneO90% of the time, thus exposing theinner collagen layer of Bruch’s membrane, and thedissection plane is not uniform throughout the exci-sion bed (84,105). In addition aging of human Bruch’smembrane causes numerous changes within thisstructure such as collagen cross-linking, elastin frag-mentation, and deposition of abnormal material withBruch’s membrane as outlined above (166); survival oftransplanted cells is substrate-and the dependent, andthus the age presence of disease within Bruch’smembrane plus surgical removal of the inner aspectsof human Bruch’s membrane will have a profoundand detrimental impact on transplant survival. In view

of these considerations, we have examined the effectsof Bruch’s membrane age and the layer available forcell attachment on RPE behavior (47,91,94).

We have determined the effects of Bruch’smembrane layer on RPE attachment by isolatingindividual layers of human Bruch’s membrane in thelab as described previously (Fig. 4) (91). HumanBruch’s membrane explants were prepared from 10human cadaver eyes by removing native RPE with0.02N ammonium hydroxide. Six millimeters punch ofperipheral Bruch’s membrane were stabilized on 4%agarose and placed in 96-well plates with Bruch’smembrane facing upwards. The RPE basal lamina,inner collagen layer, elastin layer and outer collagenlayer were exposed by removing each apical layersequentially by mechanical or enzymatic means. Firstpassage human RPE harvested from a single donorwere plated onto the surface (15,000 viable cells/explant) and the RPE reattachment rate to each layerof Bruch’s membrane was determined. The RPEreattachment rate was highest to the inner aspects ofBruch’s membrane and decreased as deeper layers ofBruch’s membrane were exposed (i.e., basal laminaOinner collagen layer O elastin layer O outer collagenlayer). The reattachment rate to the inner collagenlayer, elastin layer and outer collagen layer harvestedfrom elderly donors (age O60) was less than tothe corresponding layers harvested from younger(age !50) donors (Fig. 5). These results demonstratethat the ability of harvested RPE to reattachment tohuman Bruch’s membrane depends on the anatomiclayer of Bruch’s membrane present in the host tissue.The layer of Bruch’s membrane available also affectsthe morphology of the grafted RPE (Fig. 6) and theirsubsequent behavior. The apoptosis rate of attachedcells increased as deeper layers of Bruch’s membranewere exposed (94). Both the proliferation rate andmitotic index (94) of the grafted cells were higher onbasal lamina than on deeper layers. RPE cells platedonto basal lamina repopulated the explant surfacewithin 14G3 days, whereas cells plated onto innercollagen layer and elastin layer eventually died andnever reached confluence. These findings suggest thatthe ability of transplanted RPE cells to repopulate bareBruch’s membrane will depend on the layer of Bruch’smembrane available for RPE cell reattachment (167).

ANATOMIC RECONSTRUCTION OF HUMANBRUCH’S MEMBRANE

As mentioned above there are two major factorsrelated to Bruch’s membrane status that influence theability of grafted RPE to survive after subretinaltransplantation, namely, the layer of Bruch’smembrane available after subretinal membranectomy,and the presence of age-related changes within


Bruch’s membrane. Several authors have suggestedsimply replacing Bruch’s membrane with anotherbasement membrane substrate, such as thin siliconerubber, lens capsule, amniotic membrane, or a morecomplex bioengineered artificial structure (168,169).Although there is some appeal to this notion, itshould be remembered that cells are exquisitely sensi-tive to all aspects of their substrate including chemicalcomposition as well as mechanical substrate proper-ties. Finding a substrate that mimics Bruch’smembrane chemically and mechanically presents asignificant challenge; in addition, it will be difficult

to secure the artificial substrate to Bruch’s membraneitself and to establish a stable, long-term interfacebetween biologic and synthetic tissue.

We have taken a different approach to thisproblem by reconstructing Bruch’s membrane tomake it amore hospitable substrate for RPE attachment(37). These efforts have involved deposition ofexogenous attachment ligands on the inner aspectsof Bruch’s membrane; cleaning deposits from agedBruch’s membrane by treatment with sodium citrateand detergents; and a combined approach in whichdebris is removed from Bruch’s membrane, followed

3

(A)

(D)(C)

(B)

RPE-BL

ICL

EL

EL

OCL

ICL

EL

OCL

CC-BM

OCL

OCL

Figure 4 Preparation of basement membrane explants. Native retinal pigment epithelium (RPE) areremoved by treatment with ammonium hydroxide. (A) Yielding a preparation with the RPE basal lamina on

the uppermost surface. (B) RPE basal lamina is removed mechanically, exposing the inner collagen layer.(C) Addition of collagenase exposes the elastin layer; (D) Treatment with elastase exposes the outer

collagen layer. Abbreviations: CC-BM, choriocapillaris basement membrane; EL, elastin layer; ICL, innercollagen layer; OCL, outer collagen layer; RPE-BL, basal lamina of the retinal pigment epithelium.


by resurfacing with ECM ligands (37). Specifically wedetermined the effects of cleaning and/or ECMproteincoating on the reattachment, apoptosis, proliferation,and final surface coverage of the transplanted RPE.Explants of aged Bruch’s membrane with innercollagen layer exposed were prepared from fivehuman cadaver eyes (donor ageZ69–84 years) andtreated with Triton-X and/or coated with a mixture oflaminin (330 mg/mL), fibronectin (250 mg/mL), andvitronectin (33 mg/mL). 15,000 viable human fetal and(ARPE-19) a spontaneously immortalized human RPEcell line, cells were plated onto the surface and the RPEreattachment, apoptosis and proliferation ratios

were determined on the modified surfaces. Cells werecultured up to 17 days to determine the surfacecoverage. Ultrastructure of the modified Bruch’smembrane and RPE morphology were studied withtransmission and scanning electron microscopy. Thereattachment ratios of fetal human RPE and ARPE-19cells were similar on aged inner collagen layer (41.5G1.7% and 42.9G2.7%, pO0.05). The reattachment ratioincreased with ECM-protein coating and decreasedwith detergent treatment. Combined cleaning andcoating restored the reattachment ratio of fetal RPEcells, but failed to increase the reattachment ratio ofARPE-19 cells. The highest apoptosis was observed onuntreated inner collagen layer. Cleaning and thecombined procedure of cleaning and ECM-proteincoating decreased the fetal RPE apoptosis. Only RPEcells plated on cleaned or cleaned and ECM-coatedinner collagen layer demonstrated proliferation thatled to substantial surface coverage at day 17. Thus,these results demonstrate that age-related changes thatimpair RPE repopulation of Bruch’s membrane can besignificantly reversed by combined cleaning and ECMprotein coating of the inner collagen layer. Develop-ment of biologically-tolerant techniques for modifyingthe inner collagen layer in vivomay enhance the abilityof the RPE to reattach and repopulate aged innercollagen layer. Figure 7 shows the effect of differenttreatments on the ultrastructural features of the innercollagen layer (37).

IRIS PIGMENT EPITHELIAL TRANSPLANTATIONFOR AMD

Within the last decade several investigators havepioneered the use of iris pigment epithelium as areplacement for RPE in retinal degenerations,including AMD (170). This use of iris pigment epi-thelium is based upon the common embryologicalorigin of these two cell lines, the ready availability ofautologous iris pigment epithelium via iris biopsy, andthe need to replace RPE in various disease states.Application of iris pigment epithelium transplantationfor treatment of tapetoretinal degenerations due to aknown gene defect, such as Leber’s congenitalamaurosis and RPE-dependent forms of retinitispigmentosa, are not likely to be fruitful since auto-logous iris pigment epithelium and RPE would havethe same genetic defect. The largest clinical applicationfor autologous iris pigment epithelium transplantationmay be in repair of age-related cell and tissue loss inAMD; here transplanted iris pigment epitheliumcould replace native RPE removed during sub-macular surgery for exudative AMD or lost duringthe development of geographic atrophy in non-exudative AMD.

Rea

ttach

men

tRat

e

Basal Lamina ICL Elastin

20

40

60

80

(A)

Rea

ttach

men

tRat

e

(B) Basal Lamina ICL Elastin

20

40

60

80LAMC IVFN

LAMC IV

Figure 5 Ability of human retinal pigment epithelium (RPE)cells to reattach to different layers of human Bruch’s membrane

explants (four donors ageO50, three donors age!50) 24 hoursafter plating. (A) RPE reattachment rates on younger and older

basal lamina were comparable (pZ0.13). However, the reattach-ment to older inner collagen layer (ICL) was significantly lower

than to younger ICL (pZ0.02). The RPE reattachment rate to ICLwas lower than to basal lamina in older donors (p!0.01) but wassimilar in younger donors (pO0.05). The reattachment to elastinwas significantly lower than to basal lamina and ICL in both

younger and older donors (p!0.05). (B) Addition of laminin andcollagen IV to young basal lamina has no effect, but addition to

older basal lamina increases the attachment rate on olderBruch’s membrane to same level seen on younger Bruch’s

membrane. Similarly, addition of laminin, collagen IV and fibro-nectin has no effect on young ICL, but increases attachment onto

older ICL to same level as young ICL. Addition of ligands to youngor older elastin layer has no effect. Abbreviations: C IV, collagen

IV; FN, fibronectin; ICL, inner collagen layer; LAM, laminin.Source: From Ref. 47.


To date a handful of laboratory and clinicalstudies have been performed to determine the abilityof iris pigment epithelium to survive after subretinaltransplantation and perform RPE functions, includingouter segment phagocytosis, recycling of visualpigment, and release of cytokines and other growthfactors (59). Prior authors have concluded that irispigment epithelium can survive at least six monthsafter subretinal transplantation but proper interpre-tation of these results is confounded by difficulty inidentifying transplanted cells unequivocally (53,56,57,63,171). Initial studies suggest that subretinalor choroidal iris pigment epithelium transplantsmay slow the rate of photoreceptor degenerationin the Royal College of Surgeons (RCS) rat for severalmonths compared to untreated controls (59,60).However, iris pigment epithelium is not as good atrescue as RPE and no better than sham surgery (172).

Iris pigment epithelium function in vitro and aftersubretinal transplantation in vivo has also been inves-tigated by previousworkers. Iris pigment epithelium iscapable of retinol metabolism (173) and transplanted

iris pigment epithelium can ingest outer segments(172). The ability of cultured iris pigment epitheliumto phagocytose latex beads is 76% of the activity of RPE(59). Cultured iris pigment epitheliummaintains mela-nogenesis for up to five passages in tissue culture (54).Iris pigment epithelium and RPE form monolayers onDescemet’s membrane (174,175) and exhibit similargrowth on native and micropatterned human lenscapsule (176). Iris pigment epithelium can form tightjunctions thus raising the possibility that transplantediris pigment epithelium may reestablish the blood-retinal barrier normally formed by RPE (59).

To date a handful of clinical studies have beenperformed on subretinal transplantation of irispigment epithelium to replace surgically-excised RPEin patients with exudative AMD. Autologous irispigment epithelium transplantation has been per-formed in 35 patients after removal of subfovealchoroidal neovascular membranes with no significantdifference in vision between transplanted patientsversus those who underwent choroidal neovascular-ization removal alone (170). Autologous iris

(A)

(B)

(C)

(D)

Figure 6 Morphology of human retinal pigment epithelium (RPE) cells (donor ageZ80) after seeding ontodifferent layers of human basement membrane explants (six donors). (A,B) RPE plated onto basal laminareached confluence within 14G3 days in over 90% of the wells. (C,D) Cells plated onto inner collagen layer

detached from the surface, and only a few rare cells (arrow) could be seen on the surface 21 days afterplating. Cells plated onto elastin exhibited similar behavior (not shown). Source: From Ref. 94.


pigment epithelium translocation after submacularmembranectomy can preserve foveal function at alow level but does not improve visual acuity (177).These poor functional results are consistent with thepoor attachment and survival of iris pigment epi-thelium and RPE on aged Bruch’s membrane (178).Despite the lack of visual improvement, subretinal irispigment epithelium transplants in AMD patients mayprevent recurrence of subretinal neovascularization(60,170).

We have demonstrated that there are majordifferences in the gene expression profile of primaryRPE versus iris pigment epithelium harvested fromthe same donor eye, including the lack of expression iniris pigment epithelium of genes known to be criticalfor RPE function (49). For example, iris pigmentepithelium does not express the gene for retinol

dehydrogenase, whose gene product is necessary forrecycling visual pigments. Recoverin is a visual cycleprotein expressed in abundance in the RPE but not irispigment epithelium although its role in the RPEfunction is not known. Iris pigment epithelium donot express other major functional RPE genes,including angiopoietin 1, S-antigen and a transcrip-tional regulator of the c-fos promoter. Numerous celladhesion genes and additional genes related to RPEphagocytosis, tight junction formation and Vitamin Ametabolism are missing in iris pigment epitheliumcells, including thrombospondin 1 and ras-related C3botulinum toxin substrate.

In order for iris pigment epithelium to replacesurgically-excised or dysfunctional RPE, the trans-planted iris pigment epithelium should develop anexpression profile that closely resembles native RPE.

(A) (B)

(C) (D)

Figure 7 Scanning electron microscopy of inner collagen layer modification with cleaning and resurfacing;84-year-old donor. (A) Untreated inner collagen layer revealed replacement of fine interdigitating structure of

the collagen framework by unidirectionally running cross-linked bundles of collagen (white arrows).Small globular structures on the collagen fibers probably represent aggregates of extracellular matrix

(ECM)-proteins (white arrowheads). Macro deposits of lipoprotein debris filled interfibrillar spaces(asterisk). (B) ECM-protein coating without cleaning yielded an increased amount of ECM-protein aggregates

on the collagen matrix. Cross-linking of collagen fibers was not effected by the coating (white arrows). (C)Cleaning with Triton-X and sodium citrate removed the debris and resulted in gaps between collagen fibers

(white asterisk). Along with debris most of the globular ECM-proteins disappeared. Note that cross-linksbetween collagen fibers were broken yielding individual fibers (white arrowheads) and rare incompletely

separated macrofibers (white arrows). (D) Cleaning and subsequent ECM-protein coating not only broke thecross-links between collagen fibers but also allowed ECM-proteins to diffusely attach on the regenerated

collagen framework. Note that ECM-proteins were smaller in size and did not form multimeric aggregates ason the native matrix. Removal of macroaggregates also created spaces between collagen fibers that may help

to restore the hydraulic conductivity across Bruch’s membrane. (BarsZ0.5 mm). Source: From Ref. 37.


Our results suggest that the native iris pigment epi-thelium expression profile may be a potential obstacleto successful subretinal transplantation. Since themicroenvironment of cells influences their behaviorand gene expression, we cannot exclude the possibilitythat the expression profile of iris pigment epitheliummay change after subretinal transplantation to moreclosely resemble native RPE. However, our datasuggest that the expression level of many genes mustchange for iris pigment epithelium to resemble RPE.Some authors have suggested that transplanted irispigment epithelium can serve as a potential reservoirfor a single growth factor or cytokine and therebyrescue adjacent cells from the effects of progressivetapetoretinal degeneration (179). For example, irispigment epithelium induced to transcribe the brain-derived neurotrophic factor (BDNF) gene protectagainst retinal damage due to N-methyl-D-aspartate-induced neuronal death and light toxicity (58,180). Irispigment epithelium genetically modified to expresspigment epithelial derived factor inhibit choroidalneovascularization in a rat model of laser-inducedchoroidal neovascularization, and increase the survivaland preserve rhodopsin expression of photoreceptorcells in the RCS rat (181). For such applications, thestriking difference in the gene expression profilebetween RPE and iris pigment epithelium may be lessof an obstacle to successful cell-based therapy.Additional studies, including determining the geneexpression profile of iris pigment epithelium and RPEafter subretinal transplantation, are needed todetermine if the microenvironment of the subretinalspace will have a marked effect on the iris pigmentepithelium gene profile.

FUTURE DIRECTIONS

At the current time there are many unresolved issuesthat may influence the ability of transplanted cells torepopulate Bruch’s membrane and numerous ques-tions need to be addressed before successful celltransplantation can occur. In the absence of ananimal model for AMD the approach to this problemmust rely on a mixture of in vivo studies of celltransplantation in healthy animals, in vitro studies ofcell reattachment to human Bruch’s membranediseased with AMD, and a small number of clinicaltrials on AMD patients. There are several importantvariables that need to be investigated.

Source of CellsThe ideal source of cells for human transplantationstudies is not known. Adult human RPE are readilyavailable from donor Eye Bank eyes but it is notknown if these cells are the best source or whetherthe age of the donor RPE makes any difference. Fetal

human RPE may be able to repopulate Bruch’smembrane better than adult RPE but there are ethicaland legal issues involved with the use of human fetalcells, and fetal cells cannot be autologous. Use ofimmortalized human RPE cells lines has beenproposed, but the effects of immortalization or passa-ging in tissue culture on the distribution of cell surfacereceptors necessary for cell attachment to Bruch’smembrane have not been considered in prior studies.We have shown that there are significant differences inthe gene expression profile of native and immortalizedhuman RPE, thus raising the important question ofwhether the immortalized cells can replace all aspectsof cell function (in press) There is some concern abouttumorogenic potential if immortalized cells are used.Several authors have already used iris pigment epi-thelial cells because these cells are relatedembryological to the RPE, are readily available, andwill not be rejected immunologically (63,182).However, iris pigment epithelial transplantationcombined with subfoveal surgery has not led to adramatic improvement in vision to better than 20 outof 200 (62,63,183), and there are significant differencesin the gene expression profile of these cells comparedto RPE.

Several other cell sources that may be useful forRPE transplantation have not been investigated fully.First, the recent isolation of retinal progenitor cells(stem cells) raises the interesting possibility of usingthese cells to repopulate denuded areas of Bruch’smembrane (182). This is an attractive possibilitybecause a small population of such retinal progenitorcells could yield a large population of cells for trans-plantation, and isolation of progenitors from therecipient eye could avoid problems of immune rejec-tion. Second, xenotransplantation of porcine cells hasalready been performed in the management of centralnervous system disease including stroke, Parkinson’sdisease, and Alzheimer’s disease. These cells havebeen well-tolerated after transplantation into thecentral nervous system in patients and the possibilityof using xenografts could provide an attractivealternative to the use of human tissue (184).

Immune SuppressionA second issue that needs to be resolved is related tothe immune suppression necessary to ensure graftsurvival. In the original paper by Algvere et al. (41),the fundus photograph strongly suggests thatimmune rejection developed in these non-suppressedindividuals since the grafts became encapsulated andcystoid macular edema developed within threemonths. Systemic immune suppression appears to besufficient to prevent ophthalmic signs of graft rejectionbut local suppression with slow release devices


(intravitreal cyclosporin implants, for example) maybe preferable (41).

Status of Bruch’s Membrane afterSubmacular SurgeryAs mentioned above the status of Bruch’s membraneafter submacular surgery is important for the ultimatesuccess of cell transplantation. Disease within Bruch’smembrane and iatrogenic removal of the inner layersof Bruch’s membrane during submacular surgeryaffects the ability of transplanted RPE to repopulatethis structure. There are several approaches that couldbe used to rectify this problem including cleaning ofBruch’s membrane surface deposits, deposition ofsoluble ECM ligands, or placement of an artificialsubstrate such as lens capsule, ECM or healthyBruch’s membrane, into the subretinal space.Successful application of these techniques in humansin vivo has yet to be demonstrated.

Timing of Surgery/Identification ofSurgical CandidatesThe issue regarding the timing of surgery is still to beresolved. Patients with disciform scars have evidenceof significant atrophy of the outer retina over theneovascular tissue. Thus, prompt intervention maybe necessary to improve the visual prognosis. Alsochoriocapillaris atrophy may develop under the foveain patients with chronic subfoveal neovascularization,so that prompt surgery may improve preservation ofthis vascular supply as well.

SUMMARY POINTS

& The development of techniques to surgically excisechoroidal neovascular membranes has introducedthe possibility of surgically reconstructing thesubretinal space in patients who have subfovealchoroidal neovascularization in AMD, POHS, andother disorders.

& Early attempts at reconstructing the anatomy of thesubretinal space were focused on simple surgicalexcision of choroidal neovascularization.

& Subfoveal membrane excision can lead to goodvisual results if the subfoveal RPE is not removedat the time of surgery, or if the RPE is removed andadjacent RPE then repopulates the subfoveal areaof Bruch’s membrane within one weekafter surgery.

& The presence of native or regenerated RPE isrequired to prevent postoperative atrophy of thesubfoveal choriocapillaris, because the subfovealchoriocapillaris will undergo atrophy if Bruch’smembrane remains devoid of RPE for R1 weekafter subfoveal surgery.

& Persistent bare areas of Bruch’s membrane will bepresent in patients who have large defects in theRPE monolayer, or in whom advanced patientage or disease to the inner aspects of Bruch’smembrane prevents complete RPE resurfacing bymigration and proliferation of adjacent RPE.

& Initial studies on RPE cell transplantation have notlead to dramatic visual improvements, but thepresence of disease within Bruch’s membrane,iatrogenic removal of the inner layers of Bruch’smembrane, and immune rejection of the transplanthave limited visual recovery after surgery.

& The next challenges in submacular surgery is todeliver RPE into the subretinal space as an orga-nized monolayer, ensure the rapid attachment ofthese cells to Bruch’s membrane and preventimmunologic rejection of these cells.

& Cell survival immediately after transplantation isimportant to prevent atrophy of the subfovealchoriocapillaris.

& Development of an elusive animal model wouldfacilitate progress in this field, because in theabsence of an animal model, conclusions must bedrawn from a combination of in vitro studiesstudying cell attachment to normal and diseasedBruch’s membrane, in vivo studies of cell trans-plantation in normal animals, and a limitednumber of in vivo studies of RPE transplantationin individuals with AMD.

& At the dawn of the new millennium, the challengeis great but the potential benefit of success is evengreater because of the sheer number of patientswho are affected by this devastating disease.

ACKNOWLEDGMENTS

Supported in part by an unrestricted grant fromResearch to Prevent Blindness, Inc., New York.

REFERENCES

1. Defoe DM, Ahmad A, et al. Membrane polarity of the Na(C)-KC pump in primary cultures of Xenopus retinalpigment epithelium. Exp Eye Res 1994; 59(5):587–96.

2. Smith W, Assink J, et al. Risk factors for age-relatedmacular degeneration: pooled findings from three conti-nents. Ophthalmology 2001; 108(4):697–704.

3. Anon. Pegaptanib sodium (Macugen) for maculardegeneration. Med Lett Drugs Ther 2005; 47(1212):55–6.

4. Adamis AP, Altaweel M, et al. Changes in retinal neovas-cularization after pegaptanib (Macugen) therapy indiabetic individuals. Ophthalmology 2006; 113(1):23–8.

5. Cunningham ET, Jr., Adamis AP, et al. A phase II random-ized double-masked trial of pegaptanib, an anti-vascularendothelial growth factor aptamer, for diabetic macularedema. Ophthalmology 2005; 112(10):1747–57.


6. D’Amico DJ, Patel M, et al. Pegaptanib sodium for neovas-cular age-related macular degeneration: two-year safetyresults of the two prospective, multicenter, controlledclinical trials. Ophthalmology 2006; 113(6):1001e1–6.

7. Fraunfelder FW. Pegaptanib for wet macular degener-ation. Drugs Today (Barc) 2005; 41(11):703–9.

8. Gonzales CR. Enhanced efficacy associated with earlytreatment of neovascular age-related macular degener-ation with pegaptanib sodium: an exploratory analysis.Retina 2005; 25(7):815–27.

9. Gragoudas ES, Adamis AP, et al. Pegaptanib for neovas-cular age-related macular degeneration. N Engl J Med2004; 351(27):2805–16.

10. Moshfeghi AA, Puliafito CA. Pegaptanib sodium for thetreatment of neovascular age-related macular degener-ation. Expert Opin Investig Drugs 2005; 14(5):671–82.

11. Ng EW, Shima DT, et al. Pegaptanib, a targeted anti-VEGFaptamer for ocular vascular disease. Nat Rev Drug Discov2006; 5(2):123–32.

12. Pieramici DJ, Bressler SB, et al. Occult with no classicsubfoveal choroidal neovascular lesions in age-relatedmacular degeneration: clinically relevant natural historyinformation in larger lesions with good vision from theVerteporfin in Photodynamic Therapy (VIP) Trial: VIPReport No. 4. Arch Ophthalmol 2006; 124(5):660–4.

13. Rakic JM, Blaise P, Foidart JM. Pegaptanib and age-relatedmacular degeneration. N Engl J Med 2005; 352(16):1720–1(Author reply 1720–1).

14. Rosenfeld PJ, Moshfeghi AA, Puliafito CA. Optical coher-ence tomography findings after an intravitreal injection ofbevacizumab (Avastin) for neovascular age-relatedmacular degeneration. Ophthalmic Surg Lasers Imaging2005; 36(4):331–5.

15. Sullivan F. Pegaptanib was effective and safe without adose-response relation in neovascular, age-related,macular degeneration. ACP J Club 2005; 143(1):18.

16. Tobin KA. Macugen treatment for wet age-related maculardegeneration. Insight 2006; 31(1):11–4.

17. Gaudreault J, Fei D, et al. Preclinical pharmacokinetics ofRanibizumab (rhuFabV2) after a single intravitreal admin-istration. Invest Ophthalmol Vis Sci 2005; 46(2):726–33.

18. Husain D, Kim I, et al. Safety and efficacy of intravitrealinjection of ranibizumab in combination with verteporfinPDT on experimental choroidal neovascularization in themonkey. Arch Ophthalmol 2005; 123(4):509–16.

19. Kim IK, Husain D, et al. Effect of intravitreal injection ofranibizumab in combination with verteporfin PDT onnormal primate retina and choroid. Invest OphthalmolVis Sci 2006; 47(1):357–63.

20. Michels S, Rosenfeld PJ. Treatment of neovascular age-related macular degeneration with Ranibizumab/Lucentis. Klin Monatsbl Augenheilkd 2005; 222(6):480–4.

21. Rosenfeld PJ. Intravitreal Avastin: the low cost alternativeto lucentis? Am J Ophthalmol 2006; 142(1):141–3.

22. Rosenfeld PJ, Heier JS, et al. Tolerability and efficacy ofmultiple escalating doses of ranibizumab (Lucentis) forneovascular age-related macular degeneration. Ophthal-mology 2006; 113(4):632.e1.

23. Avery RL, Pieramici DJ, et al. Intravitreal bevacizumab(Avastin) for neovascular age-related macular degener-ation. Ophthalmology 2006; 113(3):363–72.e5.

24. Bakri SJ, Cameron JD, et al. Absence of histologic retinaltoxicity of intravitreal bevacizumab in a rabbit model. AmJ Ophthalmol 2006; 142(1):162–4.

25. Luke M, Warga M, et al. Effects of bevacizumab (R) onretinal function in isolated vertebrate retina. BrJ Ophthalmol 2006; 90(9):1178–82.

26. Manzano RP, PeymanGA, et al. Testing intravitreal toxicityof bevacizumab (Avastin). Retina 2006; 26(3):257–61.

27. Maturi RK, Bleau LA, Wilson DL. Electrophysiologicfindings after intravitreal bevacizumab (Avastin) treat-ment. Retina 2006; 26(3):270–4.

28. Michels S, Rosenfeld PJ, et al. Systemic bevacizumab(Avastin) therapy for neovascular age-related maculardegeneration twelve-week results of an uncontrolled open-label clinical study. Ophthalmology 2005; 112(6):1035–47.

29. Rosenfeld PJ, Fung AE, Puliafito CA. Optical coherencetomography findings after an intravitreal injection ofbevacizumab (Avastin) for macular edema from centralretinal vein occlusion. Ophthalmic Surg Lasers Imaging2005; 36(4):336–9.

30. Shahar J, Avery RL, et al. Electrophysiologic and retinalpenetration studies following intravitreal injection ofbevacizumab (Avastin). Retina 2006; 26(3):262–9.

31. Berger AS, Kaplan HJ. Clinical experience with thesurgical removal of subfoveal neovascular membranes.Short-term postoperative results. Ophthalmology 1992;99(6):969–75 (discussion 975–6).

32. Thomas MA, Kaplan HJ. Surgical removal of subfovealneovascularization in the presumed ocular histoplasmosissyndrome. Am J Ophthalmol 1991; 111(1):1–7.

33. Thomas MA, Grand MG, et al. Surgical management ofsubfoveal choroidal neovascularization. Ophthalmology1992; 99(6):952–68 (discussion 975–6).

34. Lambert HM, Capone A, Jr., et al. Surgical excision ofsubfoveal neovascular membranes in age-related maculardegeneration. Am J Ophthalmol 1992; 113(3):257–62.

35. Coscas G, Meunier I. Surgery of macular neovascularsubretinal membranes. J Fr Ophtalmol 1993; 16(11):633–41.

36. Hsiue GH, Lai JY, Lin PK. Absorbable sandwich-likemembrane for retinal-sheet transplantation. J BiomedMater Res 2002; 61(1):19–25.

37. Tezel TH, Del Priore LV, Kaplan HJ. Reengineering ofaged Bruch’s membrane to enhance retinal pigmentepithelium repopulation. Invest Ophthalmol Vis Sci 2004;45(9):3337–48.

38. de Juan E, Jr., Machemer R. Vitreous surgery for hemor-rhagic and fibrous complications of age-related maculardegeneration. Am J Ophthalmol 1988; 105(1):25–9.

39. Submacular Surgery Trials Pilot Study Investigators.Submacular surgery trials randomized pilot trial oflaser photocoagulation versus surgery for recurrent chor-oidal neovascularization secondary to age-related maculardegeneration: II. Quality of life outcomes submacular sur-gery trials pilot study report number 2. Am J Ophthalmol2000; 130(4):408–18.

40. Bressler NM, Bressler SB, et al. Submacular surgery trialsrandomized pilot trial of laser photocoagulationversus surgery for recurrent choroidal neovascularizationsecondary to age-related macular degeneration: I.Ophthalmic outcomes submacular surgery trialspilot study report number 1. Am J Ophthalmol 2000;130(4):387–407.

41. Algvere PV, Berglin L, et al. Transplantation of fetal retinalpigment epithelium in age-related macular degenerationwith subfoveal neovascularization. Graefes Arch Clin ExpOphthalmol 1994; 232(12):707–16.

42. Berger AS, Tezel TH, et al. Photoreceptor transplantationin retinitis pigmentosa: short-term follow-up. Ophthal-mology 2003; 110(2):383–91.

43. Binder S, Stolba U, et al. Transplantation of autologousretinal pigment epithelium in eyeswith foveal neovascular-ization resulting from age-related macular degeneration:a pilot study. Am J Ophthalmol 2002; 133(2):215–25.


44. Del Priore LV. Effect of sham surgery on retinal functionafter subretinal transplantation of the artificial siliconeretina. Arch Ophthalmol 2005; 123(8):1156 (Author reply1156–7).

45. Del Priore LV, Kaplan HJ, et al. Retinal pigment epithelialcell transplantation after subfoveal membranectomy inage-relatedmacular degeneration: clinicopathologic corre-lation. Am J Ophthalmol 2001; 131(4):472–80.

46. Del Priore LV, Tezel TH, Kaplan HJ. Survival of allogeneicporcine retinal pigment epithelial sheets after subretinaltransplantation. Invest Ophthalmol Vis Sci 2004;45(3):985–92.

47. Kaplan HJ, Tezel TH, et al. Retinal transplantation. ChemImmunol 1999; 73:207–19.

48. Kaplan HJ, Tezel TH, et al. Human photoreceptor trans-plantation in retinitis pigmentosa. A safety study. ArchOphthalmol 1997; 115(9):1168–72.

49. Kaplan HJ, Del Priore LV, Berger A, Tezel TH, RPEtransplantation in age-related macular degeneration. In:Panozzo G, Zarbin MA, Capone A, Del Priore LV, ProgeiEditori, eds. First International Conference on NewDevelopments in the Treatment of Age-related MacularDegeneration, Verona, Italy 1998:161–4.

50. Lois N. Transplantation of autologous retinal pigmentepithelium in eyes with foveal neovascularization. AmJ Ophthalmol 2002; 134(3):468 (Author reply 468–9).

51. Peyman GA, Blinder KJ, et al. A technique for retinalpigment epithelium transplantation for age-relatedmacular degeneration secondary to extensive subfovealscarring. Ophthalmic Surg 1991; 22(2):102–8.

52. Stur M. Transplantation of autologous retinal pigmentepithelium in eyes with foveal neovascularization. AmJ Ophthalmol 2002; 134(3):469–70 (Author reply 470–2).

53. Abe T, Tomita H, et al. Autologous iris pigment epithelialcell transplantation in monkey subretinal region. Curr EyeRes 2000; 20(4):268–75.

54. Abe T, Takeda Y, et al. Cytokine gene expression aftersubretinal transplantation. Tohoku J Exp Med 1999;189(3):179–89.

55. Abe T, Yoshida M, et al. Functional analysis after auto irispigment epithelial cell transplantation in patients withage-related macular degeneration. Tohoku J Exp Med1999; 189(4):295–305.

56. Crafoord S, Geng L, et al. Experimental transplantation ofautologous iris pigment epithelial cells to the subretinalspace. Acta Ophthalmol Scand 2001; 79(5):509–14.

57. Crafoord S, Geng L, et al. Photoreceptor survival intransplantation of autologous iris pigment epithelial cellsto the subretinal space. Acta Ophthalmol Scand 2002;80(4):387–94.

58. Hojo M, Abe T, et al. Photoreceptor protection by irispigment epithelial transplantation transduced with AAV-mediated brain-derived neurotrophic factor gene. InvestOphthalmol Vis Sci 2004; 45(10):3721–6.

59. Rezai KA, Kohen L, et al. Iris pigment epithelium trans-plantation. Graefes Arch Clin Exp Ophthalmol 1997;235(9):558–62.

60. Schraermeyer U, Kayatz P, et al. Transplantation of irispigment epithelium into the choroid slows down thedegeneration of photoreceptors in the RCS rat. GraefesArch Clin Exp Ophthalmol 2000; 238(12):979–84.

61. Jordan JF, Semkova I, et al. Iris pigment epithelial cellstransplanted into the vitreous accumulate at the opticnerve head. Graefes Arch Clin Exp Ophthalmol 2002;240(5):403–7.

62. Thumann G, Grand MG, Aisenbrey S, et al. Trans-plantation of autologous iris pigment epithelium afterremoval of choroidal neovascular membranes. ArchOphthalmol 2000; 118(10):1350–5.

63. Thumann G, Bartz-Schmidt KU, et al. Transplantation ofautologous iris pigment epithelium to the subretinal spacein rabbits. Transplantation 1999; 68(2):195–201.

64. Williams KA. Transplantation of autologous iris pigmentepithelial cells as a treatment for age-related maculardegeneration? Transplantation 1999; 68(2):171–2.

65. Aisenbrey S, Bartz-Schmidt U. Macular translocation with360-degree retinotomy for management of age-relatedmacular degeneration with subfoveal choroidal neovascu-larization. Am J Ophthalmol 2003; 135(5):748–9 (Authorreply 749).

66. Chang AA, Tan W, et al. Limited macular translocation forsubfoveal choroidal neovascularization in age-relatedmacular degeneration. Clin Experiment Ophthalmol2003; 31(2):103–9.

67. D’Amico DJ, Friberg TR. Limited inferior macular translo-cation for the treatment of subfoveal choroidalneovascularization secondary to age-related maculardegeneration. Am J Ophthalmol 2001; 132(2):289–90.

68. Fujii GY, de Juan E, Jr., et al. Limited macular translocationfor the management of subfoveal choroidal neovascular-ization after photodynamic therapy. Am J Ophthalmol2003; 135(1):109–12.

69. Fujii GY, Au Eong KG, et al. Limited macular transloca-tion: current concepts. Ophthalmol Clin North Am 2002;15(4):425–36.

70. Fujii GY, Humayun MS, et al. Initial experience ofinferior limited macular translocation for subfovealchoroidal neovascularization resulting from causesother than age-related macular degeneration. AmJ Ophthalmol 2001; 131(1):90–100.

71. Glacet-Bernard A, Simon P, et al. Translocation of themacula for management of subfoveal choroidal neovascu-larization: comparison of results in age-related maculardegeneration and degenerative myopia. Am J Ophthalmol2001; 131(1):78–89.

72. Hamelin N, Glacet-Bernard A, et al. Surgical treatment ofsubfoveal neovascularization in myopia: macular translo-cation vs surgical removal. Am J Ophthalmol 2002;133(4):530–6.

73. Lewis H, Kaiser PK, et al. Macular translocation for subfo-veal choroidal neovascularization in age-related maculardegeneration: a prospective study. Am J Ophthalmol 1999;128(2):135–46.

74. Ng EW, Fujii GY, et al. Macular translocation in patientswith recurrent subfoveal choroidal neovascularizationafter laser photocoagulation for nonsubfoveal choroidalneovascularization. Ophthalmology 2004; 111(10):1889–93.

75. Ohji M, Fujikado T, et al. Comparison of three techniquesof foveal translocation in patients with subfoveal choroidalneovascularization resulting from age-related maculardegeneration. Am J Ophthalmol 2001; 132(6):888–96.

76. Park CH, Toth CA. Macular translocation surgery with360-degree peripheral retinectomy following ocularphotodynamic therapy of choroidal neovascularization.Am J Ophthalmol 2003; 136(5):830–5.

77. Pawlak D, Glacet-Bernard A, et al. Limited macular trans-location compared with photodynamic therapy in themanagement of subfoveal choroidal neovascularization inage-related macular degeneration. Am J Ophthalmol 2004;137(5):880–7.


78. Pertile G, Claes C. Macular translocation with 360 degreeretinotomy for management of age-related maculardegeneration with subfoveal choroidal neovasculariza-tion. Am J Ophthalmol 2002; 134(4):560–5.

79. Pieramici DJ, De Juan E, Jr., et al. Limited inferior maculartranslocation for the treatment of subfoveal choroidalneovascularization secondary to age-related maculardegeneration. Am J Ophthalmol 2000; 130(4):419–28.

80. Roth DB, Estafanous M, Lewis H. Macular translocationfor subfoveal choroidal neovascularization in angioidstreaks. Am J Ophthalmol 2001; 131(3):390–2.

81. TerasakiH.Rescue of retinal function bymacular transloca-tion surgery in age-relatedmacular degeneration and otherdiseases with subfoveal choroidal neovascularization.Nagoya J Med Sci 2001; 64(1–2):1–9.

82. Stanga PE, Kychenthal A, et al. Retinal pigment epi-thelium translocation and central visual function in agerelated macular degeneration: preliminary results. IntOphthalmol 2001; 23(4–6):297–307.

83. Stanga PE, Kychenthal A, et al. Retinal pigment epitheliumtranslocation after choroidal neovascular membraneremoval in age-related macular degeneration. Ophthal-mology 2002; 109(8):1492–8.

84. Grossniklaus HE, Hutchinson AK, et al. Clinicopathologicfeatures of surgically excised choroidal neovascularmembranes. Ophthalmology 1994; 101(6):1099–111.

85. Del Priore LV, Kaplan HJ, et al. Experimental and surgicalaspects of retinal pigment epithelial cell transplantation.Eur J Implant Ref Surg 1993; 5:128–32.

86. Rosa RH, Thomas MA, Green WR. Clinicopathologiccorrelation of submacular membranectomy with retentionof good vision in a patient with age-related maculardegeneration. Arch Ophthalmol 1996; 114(4):480–7.

87. Hsu JK, Thomas MA, et al. Clinicopathologic studies of aneye after submacular membranectomy for choroidalneovascularization. Retina 1995; 15(1):43–52.

88. Del Priore LV, Geng L, et al. Extracellular matrix ligandspromote RPE attachment to inner Bruch’s membrane.Curr Eye Res 2002; 25(2):79–89.

89. Del Priore LV, Kaplan HJ, Berger A. Retinal pigmentepithelial trnasplantation inthe magagement of subfovealchoroidal neovascularization. Semin Ophthalmol 1997;12:45–55.

90. Akduman L, Del Priore LV, Kaplan HJ. Spontaneousresolution of retinal detachment occurring after macularhole surgery. Arch Ophthalmol 1998; 116(4):465–7.

91. Del Priore LV, Tezel TH. Reattachment rate of humanretinal pigment epithelium to layers of human Bruch’smembrane. Arch Ophthalmol 1998; 116(3):335–41.

92. Del Priore LV, Tezel TH, et al. Retinal pigment epithelialtransplantation in exudative age-related macular degener-ation: what do in vivo and in vitro studies teach us?. In:Coscas G, Piccolino FC, eds. Retinal Pigment Epitheliumand Macular Diseases, Documenta OphthalmologicaProceeedings Series 62. Dordecht: Kluwer AcademicPublishers, 1999:125–34.

93. Tezel TH, Del Priore LV. Reattachment to a substrateprevents apoptosis of human retinal pigment epithelium.Graefe’s archive for clinical and experimental ophthal-mology. Albrecht von Graefes Archiv feur klinische undexperimentelle Ophthalmologie 1997; 235(1):41–7.

94. Tezel TH, Del Priore LV. Repopulation of different layersof host human Bruch’s membrane by retinal pigmentepithelial cell grafts. Invest Ophthalmol Vis Sci 1999;40(3):767–74.

95. Tezel TH, Del Priore LV, Kaplan HJ. Harvest and storage ofadult human retinal pigment epithelial sheets. Curr EyeRes 1997; 16(8):802–9.

96. Tezel TH, Bora NS, Kaplan HJ. Pathogenesis of age-related macular degeneration. Trends Mol Med 2004;10(9):417–20.

97. Berger A, Del Priore LV, Kaplan HJ. Surgery for subfovealchoroidal neovascularization. Vitreo-retinal and uveitisupdate. In: The 47th Annual Symposium of the NewOrleans Academy of Ophthalmology. The Hague, TheNetherlands: Kugler Publications, 1998.

98. Berger AS, Conway M, et al. Submacular surgery forsubfoveal choroidal neovascular membranes in patientswith presumed ocular histoplasmosis. Arch Ophthalmol1997; 115(8):991–6.

99. Gass JD. Biomicroscopic and histopathologic consider-ations regarding the feasibility of surgical excision ofsubfoveal neovascular membranes. Am J Ophthalmol1994; 118(3):285–98.

100. Akduman L, Del Priore LV, et al. Perfusion of thesubfoveal choriocapillaris affects visual recovery aftersubmacular surgery in presumed ocular histoplasmosissyndrome. Am J Ophthalmol 1997; 123(1):90–6.

101. Desai VN, Del Priore LV, Kaplan HJ. Choriocapillarisatrophy after submacular surgery in presumed ocularhistoplasmosis syndrome. Arch Ophthalmol 1995;113(4):408–9.

102. Nasir MA, Sugino I, Zarbin MA. Decreased choriocapil-laris perfusion following surgical excision of choroidalneovascular membranes in age-related macular degener-ation. Br J Ophthalmol 1997; 81(6):481–9.

103. Pollack JS, Del Priore LV, et al. Postoperative abnormalitiesof the choriocapillaris in exudative age-related maculardegeneration. Br J Ophthalmol 1996; 80(4):314–8.

104. Thach AB, Marx JL, et al. Choroidal hypoperfusion aftersurgical excision of subfoveal neovascular membranes inage-related macular degeneration. Int Ophthalmol 1996;20(4):205–13.

105. Grossniklaus HE, Green WR, Submacular Surgery TrialsResearch Group. Histopathologic and ultrastructural find-ings of surgically excised choroidal neovascularization.Arch ophthalmol 1998; 116(6):745–9.

106. Korte GE, Reppucci V, Henkind P. RPE destruction causeschoriocapillary atrophy. Invest Ophthalmol Vis Sci 1984;25(10):1135–45.

107. Henkind P, Gartner S. The relationship between retinalpigment epithelium and the choriocapillaris. TransOphthalmol Soc UK 1983; 103(Pt 4):444–7.

108. Kuwabara T, Ishikawa Y, Kaiser-Kupfer MI. Experimentalmodel of gyrate atrophy in animals. Ophthalmology 1981;88(4):331–5.

109. Miller FS, III, Bunt-Milam AH, Kalina RE. Clinical-ultrastructural study of thioridazine retinopathy. Ophthal-mology 1982; 89(12):1478–88.

110. Takeuchi M, Itagaki T, et al. Changes in the intermediatestage of retinal degeneration after intravitreal injection ofornithine. Nippon Ganka Gakkai Zasshi 1993; 97(1):17–28.

111. Leonard DS, Zhang XG, et al. Clinicopathologic correlationof localized retinal pigment epithelium debridement.Invest Ophthalmol Vis Sci 1997; 38(6):1094–109.

112. Del Priore LV, Kaplan HJ, et al. Retinal pigment epithelialdebridement as a model for the pathogenesis and treat-ment of macular degeneration. Am J Ophthalmol 1996;122(5):629–43.

113. Valentino TL, Kaplan HJ, et al. Retinal pigment epithelialrepopulation in monkeys after submacular surgery. ArchOphthalmol 1995; 113(7):932–8.


114. Pfeffer B. Improvedmethodology for cell culture of humanand monkey retinal pigment epoithelium. Prog Retina Res1991; 10:251–91.

115. Halfter W, Dong S, et al. Composition, synthesis, andassembly of the embryonic chick retinal basal lamina.Dev Biol 2000; 220(2):111–28.

116. Green WR, McDonnell PJ, Yeo JH. Pathologic featuresof senile macular degeneration. Ophthalmology 1985;92(5):615–27.

117. Green WR, Enger C. Age-related macular degenerationhistopathologic studies. The 1992 Lorenz E. Zimmermanlecture. Ophthalmology 1993; 100(10):1519–35.


119. Lin WL. Immunogold localization of extracellular matrixmolecules in Bruch’s membrane of the rat. Curr Eye Res1989; 8(11):1171–8.

120. Das A, Frank RN, et al. Ultrastructural localizationof extracellular matrix components in human retinalvessels and Bruch’s membrane. Arch Ophthalmol 1990;108(3):421–9.

121. Chen L, Miyamura N, et al. Distribution of the collagen IVisoforms in human Bruch’s membrane. Br J Ophthalmol2003; 87(2):212–5.

122. Bhutto IA, Kim SY, et al. Localization of collagen XVIIIand the endostatin portion of collagen XVIII inaged human control eyes and eyes with age-relatedmacular degeneration. Invest Ophthalmol Vis Sci 2004;45(5):1544–52.

123. Shang QL, Ma JX, et al. Experimental choroidal neovascu-larization is inhibited by subretinal administration ofEndostatin. Zhonghua Yan Ke Za Zhi 2004; 40(4):266–71.

124. Mori K, Ando A, et al. Inhibition of choroidal neovascu-larization by intravenous injection of adenoviral vectorsexpressing secretable endostatin. Am J Pathol 2001;159(1):313–20.

125. Marneros AG, Keene DR, et al. Collagen XVIII/endostatinis essential for vision and retinal pigment epithelialfunction. Embo J 2004; 23(1):89–99.

126. Korte GE, Bellhorn RW, Burns MS. Remodelling of theretinal pigment epithelium in response to intraepithelialcapillaries: evidence that capillaries influence the polarityof epithelium. Cell Tissue Res 1986; 245(1):135–42.

127. Algvere PV, Berglin L, et al. Transplantation of RPE in age-related macular degeneration: observations in disciformlesions and dry RPE atrophy. Graefes Arch Clin ExpOphthalmol 1997; 235(3):149–58.

128. Algvere PV, Gouras P, Dafgard Kopp E. Long-termoutcome of RPE allografts in non-immunosuppressedpatients with AMD. Eur J Ophthalmol 1999; 9(3):217–30.

129. Jiang LQ, Jorquera M, Streilein JW. Immunologic conse-quences of intraocular implantation of retinal pigmentepithelial allografts. Exp Eye Res 1994; 58(6):719–28.

130. Ye J, Li W, Ryan SJ. Long-term studies on allotransplanta-tion of rabbit retinal pigment epithelial cells double-labelled with 5-bromodeoxyuridine and natural pigment.Chin Med J (Engl) 1998; 111(8):736–40.

131. Ye J, Wang HM, et al. Allotransplantation of rabbit retinalpigment epithelial cells double-labelled with 5-bromo-deoxyuridine (BrdU) and natural pigment. Curr Eye Res1993; 12(7):629–39.

132. Weisz JM, Humayun MS, et al. Allogenic fetal retinalpigment epithelial cell transplant in a patient withgeographic atrophy. Retina 1999; 19(6):540–5.

133. Binder S, Krebs I, et al. Outcome of transplantation ofautologous retinal pigment epithelium in age-relatedmacular degeneration: a prospective trial. InvestOphthalmol Vis Sci 2004; 45(11):4151–60.

134. van Meurs JC, ter Averst E, et al. Autologous peripheralretinal pigment epithelium translocation in patients withsubfoveal neovascular membranes. Br J Ophthalmol 2004;88(1):110–3.

135. van Meurs JC, Van Den Biesen PR. Autologous retinalpigment epithelium and choroid translocation in patientswith exudative age-related macular degeneration: short-term follow-up. Am J Ophthalmol 2003; 136(4):688–95.

136. Wolf S, Lappas A, et al. Macular translocation for surgicalmanagement of subfoveal choroidal neovascularizationsin patients with AMD: first results. Graefes Arch Clin ExpOphthalmol 1999; 237(1):51–7.

137. Joussen AM, Heussen FM, et al. Autologous translocationof the choroid and retinal pigment epithelium in age-related macular degeneration. Am J Ophthalmol 2006;142(1):17–30.

138. Abdel-Meguid A, Lappas A, et al. One year follow up ofmacular translocation with 360 degree retinotomy inpatients with age related macular degeneration. BrJ Ophthalmol 2003; 87(5):615–21.

139. Benner JD, Meyer CH, et al. Macular translocation withradial scleral ouffolding: experimental studies and initialhuman results. Graefes Arch Clin Exp Ophthalmol 2001;239(11):815–23.

140. Cahill MT, Mruthyunjaya P, et al. Recurrence of retinalpigment epithelial changes after macular translocationwith 360 degrees peripheral retinectomy for geographicatrophy. Arch Ophthalmol 2005; 123(7):935–8.

141. de Juan E, Jr., Fujii GY. Limited macular translocation. Eye2001; 15(Pt 3):413–23.

142. Eckardt C, Eckardt U. Macular translocation in nonexu-dative age-related macular degeneration. Retina 2002;22(6):786–94.

143. Haller JA, Hartranft CD, et al. Limited macular transloca-tion for neovascular maculopathy. Semin Ophthalmol2000; 15(2):81–7.

144. Koh SS, Arroyo J. Macular translocation with 360-degreeretinotomy for treatment of exudative age-related maculardegeneration. Int Ophthalmol Clin 2004; 44(1):73–81.

145. Lai JC, Lapolice DJ, et al. Visual outcomes followingmacular translocation with 360-degree peripheral reti-nectomy. Arch Ophthalmol 2002; 120(10):1317–24.

146. Lewis H. Macular translocation with chorioscleraloutfolding: a pilot clinical study. Am J Ophthalmol 2001;132(2):156–63.

147. Luke C, Alteheld N, et al. Electro-oculographic findingsafter 360 degrees retinotomy andmacular translocation forsubfoveal choroidal neovascularisation in age-relatedmacular degeneration. Graefes Arch Clin Exp Ophthalmol2003; 241(9):710–5.

148. Machemer R. Macular translocation. Am J Ophthalmol1998; 125(5):698–700.

149. McLeod D. Foveal translocation for exudative age relatedmacular degeneration. Br J Ophthalmol 2000; 84(4):344–5.

150. Mruthyunjaya P, Stinnett SS, Toth CA. Change in visualfunction after macular translocation with 360 degreesretinectomy for neovascular age-related macular degener-ation. Ophthalmology 2004; 111(9):1715–24.

151. Ninomiya Y, Lewis JM, et al. Retinotomy and fovealtranslocation for surgical management of subfoveal chor-oidal neovascular membranes. Am J Ophthalmol 1996;122(5):613–21.


152. Oyagi T, Fujikado T, et al. Foveal sensitivity and fixationstability before and after macular translocation with 360-degree retinotomy. Retina 2004; 24(4):548–55.

153. Pieramici DJ, de Juna E, Jr. Limited inferior maculartranslocation for the treatment of subfoveal choroidalneovascularization secondary to age-related maculardegeneration. Am J Ophthalmol 2001; 132(1):139–40.

154. Toth CA, Freedman SF. Macular translocation with 360-degree peripheral retinectomy impact of technique andsurgical experience on visual outcomes. Retina 2001;21(4):293–303.

155. Toth CA, Lapolice DJ, et al. Improvement in near visualfunction after macular translocation surgery with 360-degree peripheral retinectomy. Graefes Arch Clin ExpOphthalmol 2004; 242(7):541–8.

156. Terasaki H, Ishikawa K, et al. Changes in focal macularERGs after macular translocation surgery with 360degrees retinotomy. Invest Ophthalmol Vis Sci 2004;45(2):567–73.

157. Terasaki H, Miyake Y, et al. Change in full-field ERGs aftermacular translocation surgery with 360 degrees reti-notomy. Invest Ophthalmol Vis Sci 2002; 43(2):452–7.

158. Cahill MT, Freedman SF, Toth CA. Macular translocationwith 360 degrees peripheral retinectomy for geographicatrophy. Arch Ophthalmol 2003; 121(1):132–3.

159. Khurana RN, Fujii GY, et al. Rapid recurrence ofgeographic atrophy after full macular translocation fornonexudative age-related macular degeneration. Ophthal-mology 2005; 112(9):1586–91.

160. Chu P, Grunwald GB. Identification of the 2A10 antigen ofretinal pigment epithelium as a beta 1 subunit of integrin.Invest Ophthalmol Vis Sci 1991; 32(6):1757–62.

161. Chu PG, Grunwald GB. Functional inhibition of retinalpigment epithelial cell-substrate adhesion with a mono-clonal antibody against the beta 1 subunit of integrin.Invest Ophthalmol Vis Sci 1991; 32(6):1763–9.

162. Avery RL, Glaser BM. Inhibition of retinal pigment epi-thelial cell attachment by a synthetic peptide derived fromthe cell-binding domain of fibronectin. Arch Ophthalmol1986; 104(8):1220–2.

163. Ho TC, Del Priore LV. Reattachment of cultured humanretinal pigment epithelium to extracellular matrix andhuman Bruch’s membrane. Invest ophthalmol Vis sci1997; 38(6):1110–8.

164. Deberg M, Labasse A, et al. New serum biochemicalmarkers (Coll 2-1 and Coll 2-1 NO2) for studying oxi-dative-related type II collagen network degradation inpatients with osteoarthritis and rheumatoid arthritis.Osteoarthritis Cartilage 2005; 13(3):258–65.

165. Paik DC, Dillon J, et al. The nitrite/collagen reaction: non-enzymatic nitration as a model system for age-relateddamage. Connect Tissue Res 2001; 42(2):111–22.

166. Spraul CW, Lang GE, et al. Histologic and morphometricanalysis of the choroid. Bruch’s membrane, and retinalpigment epithelium in postmortem eyes with age-relatedmacular degeneration and histologic examination of surgi-cally excised choroidal neovascular membranes. SurvOphthalmol 1999; 44(Suppl. 1):S10–32.

167. Tezel TH, Kaplan HJ, Del Priore LV. Fate of human retinalpigment epithelial cells seeded onto layers of human

Bruch’s membrane. Invest Ophthalmol Vis Sci 1999;40(2):467–76.

168. Lee CJ, Vroom JA, et al. Determination of human lenscapsule permeability and its feasibility as a replacementfor Bruch’s membrane. Biomaterials 2006; 27(8):1670–8.

169. Sheridan C, Williams R, Grierson I. Basement membranesand artificial substrates in cell transplantation. GraefesArch Clin Exp Ophthalmol 2004; 242(1):68–75.

170. Thumann G, Kirchhof B. Transplantation of iris pigmentepithelium. Ophthalmologe 2004; 101(9):882–5.

171. Steinhorst UH, Amdreae A, et al. Autologous subretinaltransplantation of cultivated porcine iris pigment epi-thelial cells (IPE). Klin Monatsbl Augenheilkd 2001;218(3):192–6.

172. Schraermeyer U, Kociok N, Heimann K. Rescue effects ofIPE transplants in RCS rats: short-term results. InvestOphthalmol Vis Sci 1999; 40(7):1545–56.

173. Schraermeyer U, Thumann G, et al. Subretinally trans-planted embryonic stem cells rescue photoreceptor cellsfrom degeneration in the RCS rats. Cell Transplant 2001;10(8):673–80.

174. Hartmann U, Sistani F, Steinhorst UH. Human andporcine anterior lens capsule as support for growing andgrafting retinal pigment epithelium and iris pigmentepithelium. Graefes Arch Clin Exp Ophthalmol 1999;237(11):940–5.

175. Thumann G, Schraermeyer U, et al. Descemet’s membraneas membranous support in RPE/IPE transplantation. CurrEye Res 1997; 16(12):1236–8.

176. Lee CJ, Huie P, et al. Microcontact printing on humantissue for retinal cell transplantation. Arch Ophthalmol2002; 120(12):1714–8.

177. Lappas A, Foerster AM, et al. Translocation of iris pigmentepithelium in patients with exudative age-related maculardegeneration: long-term results. Graefes Arch Clin ExpOphthalmol 2004; 242(8):638–47.

178. Itaya H, Gullapalli V, et al. Iris pigment epithelium attach-ment to aged submacular human Bruch’s membrane.Invest Ophthalmol Vis Sci 2004; 45(12):4520–8.

179. Zhang C, Tang S, et al. Adeno-associated virus mediatedLacZ gene transfect to cultured human iris pigmentepithelium cells. Yan Ke Xue Bao 2003; 19(1):49–53.

180. Kano T, Abe T, et al. Protective effect against ischemia andlight damage of iris pigment epithelial cells transfectedwith the BDNF gene. Invest Ophthalmol Vis Sci 2002;43(12):3744–53.

181. Semkova I, Kreppel F, et al. Autologous transplantationof genetically modified iris pigment epithelial cells:a promising concept for the treatment of age-relatedmacular degeneration and other disorders of the eye.Proc Natl Acad Sci USA 2002; 99(20):13090–5.

182. Tropepe V, Coles BL, et al. Retinal stem cells in the adultmammalian eye. Science 2000; 287(5460):2032–6.

183. Thumann G. Potential of pigment epithelium trans-plantation in the treatment of AMD. Graefes Arch ClinExp Ophthalmol 2002; 240(9):695–7.

184. Deacon T, Schumacher J, et al. Histological evidence offetal pig neural cell survival after transplantation intoa patient with Parkinson’s disease. Nat Med 1997;3(3):350–3.


Part VII: Clinical Trial Design

24

Clinical Research TrialsA. Frances WalonkerDoheny Eye Institute and Department of Ophthalmology, Keck School of Medicine,


Kenneth R. DiddieRetinal Consultants of Southern California, Westlake Village, California, U.S.A.

INTRODUCTION

Historical ReviewAge-related macular degeneration (AMD) is animportant public health problem. Of the estimated34.8 million people in the United States who were 65years of age or older in 2002, approximately 1.6 millionhad some form of visual impairment. Approximately600,000 of these will have experienced a rapid,devastating loss of vision due to choroidal neovascu-larization (CNV), “wet AMD,” whereas the remaining1.0 million may experience a slow, progressive retinalatrophy and possibly a severe visual handicap “dryAMD” (1). Most may have difficulty performingroutine visual tasks, such as driving, reading printedmaterial, or recognizing the faces of their friends.

As the U.S. population continues to age, more andmorepersonswill becomevisually impaired fromAMD;more, in fact, than from any other eye disease. In AMDwith CNV, some of the worst losses of vision can occur.

Because a large number of individuals have AMDcomplicated by CNV, effective treatment of even afraction of all cases (2) can lead to significant savingsto society and can decrease the number of peoplerequiring social security and other disability payments(not to mention the effects on patients’ dignity andindependence), with savings far outweighing the costsof clinical research, management, and treatment.

Treatments studied have included photody-namic therapy, submacular surgery, external beamradiation, medications such as interferon, thalido-mide, corticosteroids, and anti-vascular endothelialgrowth factor drugs as well as various oral supple-ments that are believed to be preventative. At thepresent time a number of randomized clinical researchtrials are looking at these various therapies formacular degeneration. Basic scientists are workinghand in hand with clinicians to find a cure for thisblinding disease.

Clinical RelevancePrior to the Macular Photocoagulation Study (MPS),there was no proven treatment for AMD with CNV.The use of low vision aids and mobility training wererecommended but little could be done other thanobserve the natural history of AMD with CNV. TheMPS, a randomized, multicenter trial, showed thatlaser photocoagulation of AMD with CNV preventedthe most severe types of vision loss, compared to notreatment. The study was also important as a naturalhistory study of macular degeneration (2). Since the1980s this randomized controlled clinical trial hasserved as a benchmark for AMD research, with othertreatments evaluated in the same way.

CLINICAL RESEARCH METHODOLOGY

The path a new idea takes from the patient’s problemto the basic research laboratory to the clinical researchcenter and ultimately back to the treatment of thepatient in the clinical setting is extensive and expens-ive. The final research question can be answered andpractice guidelines established, but the cost in time,commitment, and dollars is great.

The pathway from the patient and back again tothe patient starts when the ophthalmologist sees apatient with a disease that either has no cure orwould benefit from an improved treatment. Case-series studies, in which an investigator has notedsome interesting or intriguing observation, frequentlylead to the generation of a hypothesis that will sub-sequently be investigated. The ophthalmologist thenteams up with the basic scientist to address thehypothesis. Together, they design appropriate labora-tory experiments to address the hypothesis. Resultsfrom these basic science studies lead to preliminaryclinical investigations of a possible new diagnostictechnique, a treatment, a drug or even a drug deliverydevice. A small group of carefully selected patientsparticipate in a pilot study to study the safety of these

new treatments. In addition, for drug therapies, thedose levels that may be most effective are also studied.If successful, such a pilot study generates a singlecenter clinical trial to further evaluate the tolerability,safety, and efficacy of the treatments.

Subsequently a full-scale, multicenter, random-ized clinical research trial is initiated to recruit enoughpatients to test the safety and the efficacy of the newprocedure, operation, test or drug or device. Thesenew approaches are also tested for their effects on thequality of life of patients with the initial disease. Theorder of the research steps are outlined in Table 1.

The randomized clinical research trial is the goldstandard, or reference, in medicine, as it provides thegreatest justification for concluding causality and issubject to the least number of problems or biases.Clinical trials are the best type of study to use whenthe objective is to establish efficacy of a treatment or aprocedure. Clinical trials in which patients arerandomly assigned to different treatment arms arethe strongest design of all.

These innovative approaches to clinical practiceare then presented and taught to other ophthalmologiststhrough continuing medical education courses, publi-cations in peer review journals, and presentations atnational and international scientific meetings. Finally,the new techniques, medications, or test materials areavailable to all patients under standard practice guide-lines for diagnosis and treatment for disease.

Design of a Clinical Research TrialThe initial step in determining whether a researchproposal would fulfill all the ethical and investiga-tional guidelines necessary to protect human subjectsinvolved in a clinical research trial, is to go through aformal decision making process. After all the datafrom previous observational, basic laboratory

(in vitro and animal studies) case report studies,Phase I, and Phase II studies have been analyzed,and a protocol is established under which the trialwill be conducted. This protocol is developedoutlining every detail of the research study so thatall personnel—investigator, coordinator, photogra-pher, vision specialist—every participant in thestudy, is aware of the protocol detail and is able tofollow this protocol for the length of the trial, main-taining standardization of evaluation, testing, surgery,and all other procedures. The steps that are involved inthe development of a protocol are as follows.

The RationaleThe ophthalmologist will team up with a basic scienceresearcher or will work in his/her own laboratory todesign a series of experiments that may address aspecific disease entity for which there may be noadequate treatment. The results of these experiments,done again and again and replicated in other labora-tories, may suggest an intervention or therapy thatwould be tested on some laboratory animal under thestrict guidelines of a research laboratory. The resultsserve as the basis for a limited trial on a small group ofcarefully selected patients. If these patients react wellto the therapy or tolerated the therapy with minimalside effects, clinical research proceeds to the nextphase: a single or two to three center clinical study ofthe therapy in patients with a specific disease.

This is the initial stage of the clinical research trial.All the data from prior studies are then analyzed alongwith any new information, and the rationale forconducting this particular study is outlined. The objec-tives of the study, the safety andefficacyof the treatment,the design of the experimental plan, the number ofenrolled subject required to prove the hypothesis andmost importantly, whether the research study willbenefit the population at large.

The ProtocolThe protocol for the study will include:

Background

& The background of the disease to be studied andthe results of all previous related research, bothbasic science and clinical.

& All information to support the justification of thisresearch project and the impact it will have on thepopulation in general and the population with thisspecific disease entity.

& The expected benefits to be obtained fromthe study.

& All the information about the study product, be itdrug, device, surgery, delivery system, with all the

Table 1 Development Phases of a Clinical Trial

Phase I Actions of drugs etc. in humans: looking at side

effects of dose levels/safety issues: early evidence

of efficacy: may include normal subjects as well as

patients: all subjects get study product: 5–15

patients (pilot study): can be multi or single center

Phase II Evaluate efficacy/tolerability of drug etc. for a

particular indication in patients with the disease

under consideration looking at side effects and

short-term risks: all study subjects get study

product: can be multi or single center

Phase III Expanded trials after preliminary evidence suggests

efficacy: additional evidence of overall

risk/benefit: may be randomized against standard

of care for this disease, observation or a placebo:

results submitted for approval pre-marketing

Phase VI Post-marketing to delineate additional information

about the risks and benefits and the optimal use of

the product

350 WALONKER AND DIDDIE

risks and adverse events noted during the prioruses of the product.

Objectives

& The primary objectives of the study could be: haltthe progression of the disease.

& The secondary objectives could be: improve visualacuity by O than three lines.

Study Design

& Description of the study which will include:& Type of study: randomized/open label/multi-

dose& Rationale answers why: Why that treatment?

Why that dose? Why that duration?& Outcome measures: primary measures; safety

and tolerability, secondary measures; forexample: the change in visual acuity/leakage

& Safety plan: unmasking/laboratory values/detailed adverse event evaluations

& Compliance: good clinical practice/Food andDrug Administration (FDA) guidelines/IRB guidelines.

Material and Methods

& Subject selection criteria with the inclusion andexclusion criteria with justification for both

& Justification for or against inclusion/exclusion ofvulnerable subjects

& Treatment assignment: randomized/stratified& Study treatment details: formulation/dose/storage& Excluded therapies& Study assessments: visual acuity/photography/

quality of life instruments/early exit criteria& Discontinuation: subject/study& Statistical methods: sample size/safety

analysis/efficacy analysis& Data quality assurance: DSMC/monitoring

Safety Assessment

& Adverse event reporting. Serious adverse eventsthat require hospital/surgical intervention or resultin death are immediately reported to the IRB andthe sponsor

& All adverse events are followed until resolutionor stability

& All subjects are contacted after study completion ifadverse events have occurred

& Medical condition confounders& Laboratory assessments

The Informed ConsentBefore any research trial that includes human subjectscan be instituted, an Institutional Review Board (IRB)

must approve all the components of the trial. Theresponsibility of an IRB is to establish the requirementsand procedures for requests for the performance ofhuman research, development, demonstration, orother activities involving patients or patient products,in addition to the usual scope of established andaccepted methods. The IRB monitors approvedresearch in accordance with the requirements theOffice of Protection from Research Risks, theregulations of the FDA, National Institutes of Healthand the Department of Health and Human Services.The IRB uses a group process to review researchprotocols and related material, e.g., informed consentdocuments and investigator brochures, to ensure thefollowing:

& Risks to human subjects are minimized by usingprocedures that are consistent with sound researchdesign and that do not unnecessarily exposesubjects to risk. Whenever appropriate, suchprocedures already will have been performed onsubjects for diagnostic or therapeutic purposes.

& Risks to subjects are reasonable in relation to theanticipated benefits (if any) to the subjects and theimportance of the knowledge that may be expectedfrom the result.

& The selection of the subjects is equitable, i.e., thestudy subjects are of both genders and fromdifferent racial/ethnic groups, and no age limi-tations exist other than those associated with adisease entity. This will decrease the risk of biasin patient selection.

& Informed consent will be sought from each prospec-tive subject or the subject’s legally authorizedrepresentative and will be documented in accord-ance with and to the extent required by informedconsent regulations. Provisions to prevent thesuggestion of coercion are documented.

& Where appropriate, the research plan makesadequate provision formonitoring the data collectedto ensure the safety of subjects either by using aDataSafetyMonitoringBoard that looks at thedata tonoteany untoward adverse events or even unexpectedimprovement that may determine the studyshould end.

& Adequate provisions are in place to protect theprivacy of the subjects and to maintain confidenti-ality of the data.

& Appropriate additional safeguards have beenincluded in the study to protect the rights and thewelfare of subjects who are members of a vulnerablegroup (e.g., children etc.).

The IRB has the authority to disapprove, modify,or approve studies based on consideration of human

24: CLINICAL RESEARCH TRIALS 351

subject protection aspects. It also has the authority tosuspend or terminate a study, to place restrictions on astudy, and to require progress reports and oversee theconduct of the study and the study investigators.

The informed consent should be signed by thepatient before entering into a clinical research trial.The informed consent will include the length ofpatient’s participation, the alternatives to this treat-ment modality, the risks involved in this trial, and astatement allowing the patients to withdraw from thetrial at any time without consequence.

Data CollectionIt is imperative that the collection of the research data,based on the design of the study, is accurate andcomplete. All research trials have case report formson which data is recorded. These forms do not containany patient identifying information other than aunique identifying code number and/or a com-bination of the patient’s initials. These forms are sentto the sponsor and, therefore, can only contain thisunique identifying information. The types of datacollected include:

& Results of laboratory testing& Quality of life questionnaires& Clinical evaluations& Eligibility criteria& Medications, medical history, surgical history& Detailed ophthalmic history including prior treat-

ment details of the disease entity& Adverse events both serious and non-serious& The study product& Investigator signatures

All data have to be checked and corrected beforethey are sent to the sponsor and this is done byrepresentatives of the sponsor. These monitors’responsibilities include: to make sure the data arecorrect and legible; that the patients have met theenrollment criteria and received the correct treatmentto which they were assigned. The clinical chart wherethe investigator notes the clinical examinations isknown as the source document. The data on the casereport forms are matched to the source documents.This cross check of data verifies accuracy.

All data and investigational study products arestored in a secured area with access to the area only bythe study staff, investigator, and coordinator.

The principal investigator of the clinical researchstudy is responsible for the conduct of the study, theaccuracy of the data collection and the conduct ofthe study staff and the safety of the study subjectsat all times.

Settings for Research TrialsAdvancing medical knowledge—through screening,treatment, surgical, and pharmaceutical interven-tion—has prolonged the life of many people withdisabling chronic disease conditions and increasedthe number of survivors of traumatic injury. At thepresent time, 13%of the population is over the age of 65;by the year 2040, this number will have grown to 23%of the population (3). By 2040, 70 million people willhave some form of activity limitation, whethermental, physical, or visual, that will require interven-tion from the healthcare systems in some form.Research into the most effective care for persons withchronic disease, including eye disorders in particular,and efforts in prevention will be at the forefront offuture clinical research. The projected cost of health-care in the year 2040 is $906 billion, a huge percentageof the gross national product of the United Statesand the highest of the entire world’s developedcountries (4).

With such huge expenditures anticipated forhealth care, and in response to continued pressure bygovernment regulatory agencies to drive down costs,evaluation of cost in conducting research is suggested.Researchersmust include cost research objectives, suchas costs associated with screening programs, alterna-tive treatments and procedures, use of new technologyand implementation of new regulatory measuresassociated with programs and trials. The resultsobtained from including cost analysis in research helphealth care decision makers weigh the costs andconsequences of competing treatment alternatives.Cost information provides additional data that cansupplement clinical judgment when making thera-peutic choices. Therefore, clinicians and researchers atmajor academic institutions need to focus onadvancingthe care andpreventionof eyedisease. Efforts shouldbebased on rigorous clinical methods, i.e., randomizedcontrolled clinical trials and analysis of economic andhumanistic outcomes. With research of this nature, theresults canbe applieddirectly to thepatient,where theywill accomplish the greatest good. This is especiallytrue when these outcomes may mean the differencebetween sight and blindness, andwhen they impact onthe outcome measures of quality of life and ultimately,life expectancy.

Limitations of Randomized Clinical TrialsThe cost of developing the necessary infrastructure tosupport the scientific and clinical activities involved inconducting major national and international clinicalresearchmakes it prohibitive except for large academicophthalmology centers unless under the sponsor-ship of industry. There are obvious downsides to thistype of sponsorship that is somewhat obviated by the


inclusion in the project of an independent clinicalresearch organization.

Most major academic ophthalmology centersinvolved in clinical and basic science research arereferral centers for patients with complicated diseasewho have not responded to standard therapy orwho have a disease with no known cure. However,because of the nature of this population, i.e., thosewith severe disease as well as those with rare andcomplicated disease, the numbers of patients whowould be eligible to enter a clinical research trialwould be limited, making recruitment difficult. Thisplaces a potential for selection bias on these clinicalresearch studies, such that when the studies arecompleted, they may not translate to the populationin general.

On the other hand, a more common diseaseentity, such as macular degeneration, with its potentialfor marked vision loss if untreated, offers access tomore subjects for inclusion in a clinical trial. Thesepatients are seen routinely in the private practiceophthalmologist’s office that is now involved inclinical research. The disadvantage to academic insti-tutions that have invested in the development of aninfrastructure to rigorously support all basic andclinical research is that they no longer have access tothis large patient population. The disadvantage to thepatient may be that the strict protocol that is thehallmark of academic institutional research may notbe adhered to so rigorously in a community where thatinfrastructure is not present.

Another limitation is access to the underserved—those people who have no access to health careproviders, either because of lack of insurance ordistance from those same providers. These patientsare likely to postpone needed care until their con-ditions have escalated in severity. This group wouldhave no representation in the clinical research arena,the subsequent lack of diversity in the research popu-lation may result in possible bias.

The tremendous increases in new technologyhave not been accompanied by changes in the clinicalevaluation of new approaches. As a result, newapproaches become established that may harmmany patients, and researchers may have difficultyobtaining approval to perform properly designedclinical trials from the human subjects committeesthat oversee the ethics of research because of thepresumed standard of practice that is present inthe field.

RESEARCH STAFF AND DOCUMENTATION

The goal of all clinical research is to provide infor-mation that will help the practitioner treat his or her

patients more effectively. The clinical trial provides thebest means to objectively quantify and compare thebenefits and risks of new or alternative treatments toestablish treatments for disease, especially when thedifference between a new or old treatment is not clearor when a large number of factors may influence thecourse of the disease or the outcomes of the treatment(3). To ensure that the treatment groups are comparedobjectively, standardized methods of gathering data,training and certifying the personnel who collect thedata, and treating patients either surgically or pharma-ceutically, are imperative. Continuous monitoring ofadherence to the protocol, uniform data accumulationand routine re-certification of personnel will eliminateany concerns of bias or ambiguity when the data ispresented. All data accumulated on a case report form,the form that is submitted to a central data collectionagency, must be documented in the patient file andthese two documents must be reconciled at all times.All clinical research studies are monitored at regularintervals to ensure that all information is recordedon all the legal documents and that no data aremissing or unsubstantiated. The success of all clinicalresearch is totally dependent on this accurate andstandardized collection of data, and strict adherenceto the protocols.

SUMMARY POINTS

& MPS was the first clinical study to look atmacular degeneration

& The aging population is 34.8 million with 1.2million having some form of visual impairment.

& Clinical research is the best means to quantify andobjectively compare the benefits and risks of newor alternative treatments for disease or injuryespecially when:& the difference between a new or old treatment is

not clear;& the disease naturally follows a chronic, variable

and erratic course;& a large number of factors, known or unknown,

may influence both the course of the diseaseand the outcome of the treatment.

& A well-designed and conducted randomizedclinical trial incorporates the following:& High ethical standards—of paramount import-

ance are patient welfare, informed consent,adherence to protocol, and careful data moni-toring.

& Control groups that are matched to the treat-ment groups for the baseline characteristics.

& Random assignment of patients to both studyand control groups when comparability ofresults among groups is essential.

24: CLINICAL RESEARCH TRIALS 353

& Masking to minimize bias of both the examinerand the patient, if possible.

& Enrollment of an adequate number of patientsenrolled in the trial for the results to bestatistically significant.

& Completeness of patient follow-up.& Use of statistical methods for study design and

data analysis.& Continuous monitoring of adherence to

protocol and accumulation of data by the DataSafety Monitoring Committee (DSMC), thestudy Advisory Committee, the ExecutiveCommittee and the Steering Committee toensure the safety of the subjects involved inthe trial (5).

REFERENCES

1. Administration on Aging (AoA). A Profile of Older Amer-icans. U.S. Department of Health and Human Services, 2000.

2. Macular Photocoagulation Study Group. Argon laserphotocoagulation for senile macular degeneration. Resultsof a randomized clinical trial. Arch Ophthalmol 1982;100:912–8.

3. Walonker AF, Sturrock D. The Ryan Leopold BeckmanCenter for Clinical Research. Masters Thesis School ofPublic Health, UCLA, 1999.

4. Chronic Care in America. A 21st Century Challenge. Prin-ceton, NJ: Prepared by the Institute for Health and AgingUniversity of California, San Francisco for the Robert WoodJohnson Foundation, August 1996.

5. Clinical Trials Supported by The National Eye Institute. U.S.Department of Health and Human Services, 1987.


Index

ABCR genes, 36Aberrant antigen-specific immunity, 18–19ACE. See angiotensin-converting enzyme.Acquired immunity, 12Adaptive immunity, 12Adjunct usage, age-relatedmaculardegenerationand, 295–301AF. See autofluorescence.Age, as risk-factor in macular-degeneration, 53, 54Age-related eye disease study (AREDS), 58, 69–72, 76–77,

98–99, 106, 127Age-related macular degeneration (AMD)adjunct usage, 295–301calcium and magnesium free retinal detachmentenhancing solutions, 296–297

encapsulated cell technology (ECT), 298–299selective intraocular radiation brachytherapy, 299–300tissue plasminogen activator, 295–296triamcinolone acetonide, 297–298

associated factors, 97cardiovascular disease, 98demographic characteristics, 97–98environmental influences, 98–99genetic influence, 99hypercholesterolemia, 98hypertension, 98systemic diseases, 98

Bruch’s membrane, 1, 332–333choroidalmonoctyes, 26neovascular membranes (CNVM), 23neovascularization (CNV), 1

clinical features, 97controlled radiation studies, 237–243definitions of, 51–52, 53Drusen, 99–101, 257–258feeder vessel treatment (FVT), 207–222fundus autofluorescence, 191–205gene pathogenesis, 37–39angiotensin-converting enzyme (ACE), 38–39extracellular matrix, 37inflammation, 37–38lipid metabolism, 38not associated with, 39GPR75, 39LAMC1, 39multi-candidate gene screening, 39

oxidative damage, 39genetics of, 35–42allele association studies, 40–41familial aggregation studies, 35–36hereditary retinal dystrophies, 36–37

[Age-related macular degeneration (AMD)][genetics of]linkage mapping, 39–40twin aggregation studies, 35–36

histopathology of, 1–9exudative AMD, 5–7non-exudative, 1–5

immune mechanism, 23–27evidence, 23–25

immunology, 11–27biology of, 11–18nonocular degenerative diseases, 19–23

indocyanine green angiography (ICGA), 159–172inflammatory mechanisms, 23–25injury hypothese response, 25–26laser prophylaxis, 257–268lesions, basallaminar deposits (BlamD), 1linear deposits, (BlinD), 1

ocular immune disorder, 24–25optical coherence tomography, 177–183treatment of, 181–183

photodynamic therapy (PDT) clinical results and, 226–227progression ofantigen-specific immunity and, 26–27inflammatory amplification cascades, 27

radiation treatment, 233–243delivery methods, 235–237implant therapy, 236charged particle, 236–237other types, 241–243rationale for, 233–235toxicity issues, 234–235

retinal pigment epithelial (RPE), 1cell transplantation, 329–343

risk factors, 47–85antioxidant enzymes, 64–65case-control studies, 48choroidal neovascularization, 74–77cohort studies, 51cross-sectional studies, 49–50dermal elastotic degeneration, 64environmental factors, 65–73alcohol consumption, 73carotenoids, 72cigarette smoking, 65–66dietary fish intake, 72–73nutritional factors, 68–73sunlight exposure, 66–67

epidemiologic studies of, 47problems and limitations of, 47, 51–53

[Age-related macular degeneration (AMD)][risk factors]ocular, 57–60cataract surgery, 58–59cup/disc ratio, 60iris color, 59–60macular pigment optical density, 57–58refractive error, 60

reproductive, 64sociodemographic factors, 53–57age, 53, 54gender, 54race/ethnicity, 55–56heredity, 56–57

socioeconomic status, 57systemic,BMI ratio, 63cardiovascularbiomarkers, 63disease, 60–65

Chlamydia pneumoniae infection, 63–64cigarettes, 64diabetes, 62–63dietary fat intake, 62hematologic factors, 63hyperglycemia, 62–63hypertension and blood pressure, 61–62serum lipid levels, 62waist conference, 63waist-hip ratio, 63

treatment, anti-inflammatory therapy, 27Age-related maculopathy (ARM), 51–52, 53Alcohol consumption, as risk factor in macular

degeneration, 73Allele association studies, 40–41chromosome 10q26, 41complement component 2, 41complement factor H, 40–41factor B, 41

Amplification cascades, 21Amplification immunity mechanisms, 12–14complement cascade, 12–13cytokines, 13–14oxidants, 14

Angiographic appearance, choroidal neovascularizationfeeder vessel, 210–212

Angiotensin-converting enzyme (ACE), 38–39Anthropomorphic model, choriocapillaris/choroidal

neovascularization, 212Antiangiogenesis treatment, 145–147Antiangiogenic factors, 88Antibody, 16–17cell-surface, 17extracellular-bound antigens, 17

Antigen presenting cells (APC), 15Antigen specific immunity, 21, 22activation, 18adaptive or acquired, 12age-related macular degeneration progression

and, 26–27amplification mechanism, 12–14

[Antigen specific immunity]effectors, 23–24innate vs., 11–14

Antigensextracellular-bound, 17intracellular, 17

Anti-inflammatory therapy, 27Antioxidant enzymes, age-related macular degeneration

risk factors and, 64–65Antioxidant vitamins, randomized trials, 70–72Antioxidants, 68–70age-related eye disease study (AREDS), 69

Anti-vascular endothelial growth factorcurrent therapies, 247–252aptamer, 247–248intravitreal injections, 251–252monoclonal antibodies, 248–251photodynamic combination, 251

future therapies, 252–254receptor tyrosine kinase inhibitors, 253–254small interfering RNAs, 253VEGF trap, 252–253

APC. See antigen presenting cells.ApoE. See apolipoprotein.Apolipoprotein E (ApoE), 20Aptamer, 247–248macugen, 247–248pegaptanib sodium, 247–248

ARM. See age-related maculopathy.Atherosclerosisantigen-specific immunity, 21complement activation, 21cytokines and oxidants, 21heat shock proteins (HSPs), 20infectious etiology, 20–21innate mechanisms, 19–21injury, 19–20

macrophages, 20apolipoprotein E (ApoE), 20

nonspecific amplification cascades, 21Atrophic retinal degeneration, 24–25Autofluorescence (AF), 191–205age-related macular degeneration, 192–197choroidal neovascularization, 198–199drusen, 197–198geographic atrophy, 198imaging, 119–120

Avastinw, 250–251

B lymphocytes, 16–17Basal deposits, 3–4Basal laminar deposits (BlamD), 1Drusen, 101

Basal linear deposits (BlinD), 1Basophils, 15–16Benzoporphyrin derivative monoacid, 225BPD-MA, 225verteporfin, 225Visudyne e, 225

Bevacizumab, 250–251Avastinw, 250–251

356 INDEX

Biomarkers, as risk factor in macular-degeneration, 63BlamD. See basal laminar deposits.BlinD. See basal linear deposits.Blood pressure, as risk factor in macular-degeneration,

61–62BMI ratio, as risk factor in macular-degeneration, 63BPD-MA, 225Brachytherapy. See implant radiation therapy.Bruch’s membrane, 1, 332–333anatomic reconstruction, 337–339changes in, 1retinal pigment epithelial cell attachment and, 335–337

Calcium and magnesium free retinal detachmentenhancing solutions, 296–297

Candidate genes, 36–37ABCR, 36ELOVL4, 36–37Fibulin 3/EFEMP1, 37peripherin/RDS, 37TIMP3, 37VMD2, 37

Cardiovascular disease, 98age-related eye disease study (AREDS), 98biomarkers, 63as risk factor in macular-degeneration, 60–65

Carotenoids,lutein, 72as risk factors in macular degeneration, 72zeaxanthin, 72

Case-control studies, age-related macular degenerationrisk factors, 48

Cataract surgery, as risk factors in macular-degeneration, 58–59

CC. See choriocapillaris.Cell-surface antigens, 17Cellular injury, immunity activation trigger, 17Central scotomas, 116–117Charged particle radiation therapy, 236–237Chlamydia pneumoniae infection, as risk factor in macular-

degeneration, 63–64Choriocapillaris (CC), 207changes in, 2–3choridal neovascularizationanthropomorphic model of, 212feeder vessel model, hemodynamic relationshipof, 212–217

Choroidal monoctyes, 26Choroidal neovascular membranes, (CNVM), 23Choroidal neovascularization (CNV), 1, 2, 87–91, 198–199atrophic retinal degeneration, 24–25choriocapillaris, anthropomorphic model of, 212combined growth pattern, 6drusen as risk factor for, 258feeder vessel, 138–139angiographic appearance of, 210–212vs. histological appearance, 211–212

choriocapillaris model, hemodynamic relationshipof, 212–217

histological appearance, 209–212

Choroidal neovascularization (CNV)[feeder vessel]relationship between, 212–217treatment, 207–222future of, 221

geographic atrophy and, 115–116histopathology of, 7indocyanine green angiography (ICGA) findings and, 161laser photocoagulation, 203–205neovascularization (NV) inferences, 87occult, 143–144ocular immune disorder, 24–25optical coherence tomography and assessment

of, 177–179pathogenesis of, 89–90pharmacologic treatment of, 90–91prevention, exudative (neovascular) age-related

macular degeneration and, 127–128progression of, 77as risk factor in macular degeneration, 74–77subretinal, 6pigment epithelium, 5–6

therapeutic options, 144–151anti-angiogenesis treatments, 145–147gene therapy, 151low vision rehabilitation, 151new types, 148–149photodynamic, 144–145radiation, 148receptor tyrosine kinase inhibitors, 150small interfering RNA, 149squalamine lactate, 150surgery, 148thermotherapy, 147–148tubulin binding agents, 150–151VEGF trap, 149–150

vascular endothelial growth factor, 87–91Chromosome 10q26, 41Chronic injury, glomerular diseases and, 21–22Cigarettes, as risk factor in macular-degeneration, 64Classic CNV, 161Classic subretinal choroidal neovascularization, 6, 7Clinical research trials, 349–354design of, 350limitations of randomized, 352–353methodology, 349–353settings for, 352staff and documentation, 353

CNV. See choroidal neovascularization.CNVM. See choroidal neovascular membranes.Cohort studies, age-related macular degeneration risk

factors, cohort studies, 51Combined growth pattern, 6Combretastatin A-4 phosphate, 150Complement activationatherosclerosis and, 21ocular immune disorder and, 24–25

Complement cascade, 12–13Complement component 2, 41Factor B and, 41

Complement factor H, 40–41

INDEX 357

Confocal scanning laser opthalmoscope indocyanine greenangiography, 171

Contrast enhanced indocyanine green angiography(ICGA), 170

Controlled radiation studies, review of, 237–243Cortical prostheses, 319–320Cross-sectional studies, age-related macular degeneration

risk factors, 49–50CST3, 37Cup/disc ratio, as risk factors in macular-degeneration, 60Cytokines, 13–14, 21

Degenerative disease, immune responses to, 17–19aberrant antigen-specific activation, 18–19antigen-specific activation, 18innate immunity activation, 17–18

Demographic characteristics, age-related maculardegeneration and, 97–98

Dendritic cells, 15Dermal elastotic degeneration, age-related macular

degeneration risk factors and, 64Diabetes, as risk factor in macular-degeneration, 62–63Dietary fat intake, as risk factor in macular-degeneration, 62Dietary fish intake, as risk factors in macular

degeneration, 72–73Diffuse drusen, 5Digital imaging, indocyanine green angiography (ICGA)

and, 160–161Digital subtractionindocyanine green angiography

(ICGA), 170Dimly lit environments, geographic atrophy and, 117–118Disciform scar, 7Drusen, 4–5, 99–101, 257–258autofluorescence (AF) and, 197–198as choridal neovascularization risk factor, 258diffuse, 5disappearance of, 101–103laser prophylaxis on, 258–259studies on, 262–265

nodular, 4reduction, laser prophylaxis and, 259–262soft, 4–5types, 100–101basil laminar, 101hard, 100soft, 100–101

Dry age related macular degeneration. See non-exudativeage related macular degeneration.

Dye enhanced photocoagulation (DEP), indocyanine greenangiography (ICG), 217–221

clinical application of, 218–221Dye-enhanced photocoagulation (DEP), 207Dystrophies, retinal, 36–37

Early Treatment Diabetic Retinopathy Study (ETDRS), 117ECT. See encapsulated cell technology.ELOVL4 genes, 36–37Encapsulated cell technology (ECT), 298–299End-stage age-related macular degeneration, retinal

prostheses, 319–326

Environmental factors, age-related macular degenerationrisk factors and, 65–73

alcohol consumption, 73carotenoids, 72cigarette smoking, 65–66dietary fish intake, 72–73nutritional factors, 68–73sunlight exposure, 66–67

Environmental influences, age-related macular degenerationand, 98–99

age-related eye disease study (AREDS), 98Epidemiologic studies, age-related macular degeneration

risk factors and, 47Epidemiologyexudative (neovascular) age-related macular degeneration

and, 125–128geographic atrophy and, 112–113

Epiretinal prosthesis, 320–323Estrogen-related factors, as risk factor in macular

degeneration, 64ETDRS. See Early Treatment Diabetic Retinopathy Study.Ethnicity, as risk-factor in macular-degeneration, 55–56Extracellular-bound antigens, 17Extracellular matrix (ECM), 37, 88–89CST3, 37fibulin, 37MMP-9, 37

Extrafoveal CNV, 141–142Extraocular approaches, retinal prostheses and,

319–320Exudative (neovascular) age-related macular

degeneration, 125–152clinical features of, 129–139pigment epithelial detachment (PED), 133–135choroidal neovascularization, 135–139

epidemiology, 125–128choroidal neovascularization prevention, 127–128risk factors, 126–128non-ocular, 126–127ocular, 127genetics, 127–128

idiopathic polypoidal choridal vasculopathy, 139macular photocoagulation study, 141–143natural history of, 140–141natural history of untreated, 140–141juxtafoveal CNV, 141subfoveal CNV, 141

occult choroidal neovascularization, 143–144retinal angiomatous proliferation, 139–140symptoms of, 128–129

Exudative (wet) age related macular degeneration (AMD)choroidal neovascularization, 5–7disciform scar, 7

Exudative macular degeneration, macular translocationand, 275–276

FA. See fluorescein angiography.Factor B, 41complement component 2, 41

Familial aggregation studies, 35–36Fat intake, as risk factor in macular-degeneration, 62

358 INDEX

Feeder vessel (FV)choriocapillaris, choridal neovascularization model,

hemodynamic relationship of, 212–217choroidal neovascularization and, 138–139relationship between, 212–217

definition of, 209–212Feeder vessel therapy, 170–171Feeder vessel treatment (FVT), 207–222choriocapillaris (CC), 207choroidal neovascularization and, future of, 221concept of, 207–209development of, 217–221feeder vessels (FV), 209–212

Fibulin, 373/EFEMP1 gene, 37

Fish diet, as risk factor in macular degeneration, 72–73Fluid-air exchange, 284Fluorescein angiography (FA), 159, 186, 285Focal CNV, 162Fundus autofluorescence, 191–205Fundus photography, 185–186FVT. See feeder vessel treatment.

GA. See geographic atrophy.Gender, as risk-factor in macular-degeneration, 54Gene screening, multi-candidate, 39Gene therapy, 151Genesage-related macular degeneration and, pathogenesis

of, 37–39candidate, 36–37extracellular matrix, 37

Geneticsage-related macular degeneration and, 35–42allele association studies, 40–41familial aggregation studies, 35–36hereditary retinal dystrophies, 36–37influences, age-related macular degeneration

and, 999linkage mapping, 39–40risk factors, exudative (neovascular) age-related macular

degeneration and, 127–128twin aggregation studies, 35–36

Geographic atrophy (GA), 5, 111–121, 198autofluorescence imaging, 119–120bilateral development of, 115choroidal neovascularization and, 115–116clinical features, 111–112conditions similar to, 118–119epidemiology, 112–113heredity factors, 113histopathology and pathogenesis, 112natural history, 113–115prevalence, 112–113systemic risk factors, 112–113

treatments, 120–121visual function impairment, 116–118central scotomas, 116–117dimly lit environments, 117–118other visual abnormalities, 118paracentral scotomas, 116–117

Glomerular diseases, 21–23antigen-specific immunity, 22innate immunity, 21–22chronic injury, 21–22

macrophage-mediated injury, 22nonspecific amplification mechanism, 23

GPR75 type-gene, 39

Hard drusen, 100Heat shock proteins (HSPs), 20Hematologic factors, as risk factor in macular-

degeneration, 63Hemodynamic relationship, choriocapillaris, choridal

neovascularization, feeder vessel model, 212–217Hereditary retinal dystrophies, 36–37candidate genes, 36–37ABCR, 36ELOVL4, 36–37fibulin 3/EFEMP1, 37peripherin/RDS, 37TIMP3, 37VMD2, 37

Heredity factors, geographic atrophy and, 113Heredity, as risk-factor in macular-degeneration, 56–57Histological appearance, choroidal neovascularization

feeder vessel, 209–212Histopathologyage related macular degeneration and, 1–9exudative (wet) age related macular degeneration

(AMD), 5Histopathy, geographic atrophy, 112Hot spot occult CNV, 162HSP. See heat shock protein.Hypercholesterolemia, 98Hyperglycemia, as risk factor in macular-degeneration,

62–63Hypertension, 98age-related eye disease study (AREDS), 98as risk factor in macular-degeneration, 61–62

ICGA. See indocyanine green angiography.Idiopathic polypoidal choridal vasculopathy, 139Imaging, optical coherence tomography (OCT), 177–183Imbricating sutures, 280–281, 283–284tightening of, 283–284

Immune cellsantibody, 16–17B lymphocytes, 16–17T lymphocytes, 16

Immune mechanismsantigen-specific immune effectors, 23–24atherosclerosis, 19–21antigen-specific immunity, 21nonspecific amplification cascades, 21

direct evidence for innate immune effectors, 23–24glomerular diseases, 21–23nonocular degenerative diseases and, 19–23

Immune response, antigen-specific activation, 18Immune response cells, 14–17basophils, 15–16dendritic, 15

INDEX 359

[Immune response cells]macrophages, 14–15mast cells, 15–16monocytes, 14–15

Immune system, degenerative disease responses, 17–19Immunologyage-related macular degeneration and, 11–27biology of, 11–18degenerative disease response, 17–19immune response cells, 14–17innate, antigen-specific vs., 11–14

Implant radiation therapy (brachytherapy), 236Indocyanine greenhistory of, 159–160injection technique, 160pharmacology, 160properties of, 159toxicity of, 160

Indocyanine green angiography (ICGA), 159–172associated treatment strategies, 169–170clinical application of, 167–168confocal scanning laser opthalmoscope, 171digital imaging systems, 160–161dry age-related macular degeneration, 171dye enhanced photocoagulation (DEP), 217–221clinical application of, 218–221

findings, 161–163choroidal neovascularization, 161classic CNV, 161focal occult CNV, 162hot spot occult CNV, 162occult CNV, 161–162plaque, 162–163serous pigment epithelial detachment, 161

new techniques, 170–171contrast enhanced, 170digital subtraction, 170FV therapy, 170–171real-time, 170wide-angle, 170

polypoidal choroidal vasculopathy, 163–64recurrent CNV, 168–169retinal angiomatous proliferation, 164–167

Infection,immunity activation trigger, 18as risk factor in macular-degeneration, 63–64

Infectious etiology, atherosclerosis and, 20–21Inflammation factors, 37–38Inflammatory amplification cascades, age-related macular

degeneration progression, 27Inflammatory mechanisms, nonocular degenerative

diseases and, 19–23Injury, atherosclerosis and, 19–21low density lipoprotein (LDL), 19

Injury hypothese response, 25–26Innate immunityactivation triggercellular injury, 17infection, 18

amplification mechanisms, 12–14effectors, 23–24

[Innate immunity]glomerular diseases and, 21–22natural, 11

Innate immunology, antigen-specific vs., 11–14Innate mechanisms, atherosclerosis and, 19–21Intracellular antigens, 17Intraocular radiation brachytherapy, 299–300Intraoperative complications, macular translocation

and, 290Intravitreal injections, 251–252Iris color, as risk factors in macular-degeneration, 59–60Iris pigment epithelial transplantation, 339–342

Juxtafoveal CNV, 141, 142

LAMC-type genes, 39Laminar basal deposits, 3Laser light application, photodynamic therapy and, 224Laser photocoagulation, 203–205, 285–286Macular Photocoagulation Study (MPS), 203–205results of, 205

Laser prophylaxis, 257–268drusen and, 258–259reduction, 259–262studies on, 262–265

multicentered clinical trials, 265–268LDL. See low density lipoprotein.Limited macular translocation, 279–280equipment and overview, 280fluorescein angiography, 285key surgical steps, 280–284fluid-air exchange, 284imbricating sutures, 280–281, 283–284pars plana vitrectomy, 281planned neurosensory retinal detachment, 281–283

laser photocoagulation, 285–286operative techniques, 280–284patient positioning, 284persistent subfoveal CNV, 286postoperative review, 285–286recurrent subfoveal CNV, 286

Linear basal deposits, 3–4Linkage mapping, 39–40Lipid levels, as risk factor in macular-degeneration, 62Lipid metabolism, 38Low density lipoprotein (LDL), 19Low vision evaluation, 303–307Low vision rehabilitation, 151Lucentis, 248–250Lutein, 72

Macrophage-mediated injury, glomerular diseases and, 22Macrophages, 14–15, 20antigen presenting cells (APC), 15apolipoprotein E (ApoE), 20

Macugen, 247–248Macular degeneration, non-exudative, 103–106Macular Photocoagulation Study, 203–205decreased vision, 205extrafoveal CNV, 141–142juxtafoveal CNV, 142

360 INDEX

[Macular Photocoagulation Study]subfoveal CNV, 142–143trials, 205 (MPS)

Macular pigment optical density, as risk factorsin macular-degeneration, 57–58

age-related eye disease study (AREDS), 58Macular translocation, 273–292anatomic considerations, 277–279classification and terminology of, 274complications from, 290–291intraoperative, 290postoperative, 290–291

historical background of, 275indications, 275–276exudative macular degeneration, 275–276non-exudative macular degeneration, 276subfoveal RPE defect, 276

limited, 279–280outcome, 288–290pathophysiologic considerations re visual loss, 276–277postoperative cyclovertical diplopia, 287–288preoperative considerations, 276–279rationale, 274–2753608 retinotomy, 287

Macular transplantation, clinical results, 333–335Mast cells, 15–16Micronutrients, 68Microperimetry, 187–188MMP-9, 37Monoclonal antibodies, 248–251bevacizumab, 250–251ranibizumab, 248–250

Monocytes, 14–15MPS. See macular photocoagulation study.Multi-candidate gene screening, 39

Natural history, exudative (neovascular) age-relatedmaculardegeneration and, 140–141

Natural immunity. See innate immunity.Neovascularization (NV)inferences, 87transcription factors, 89

Neurosensory retinachanges in, 3detachment, 281–283

Neurotransmitter-based prothesis, 325–326Nodular drusen, 4Non-exudative age related macular degeneration (AMD),

1–5, 103–106, 171age-related eye disease study (AREDS), 106basal deposits, 3–4Bruch’s membrane, changes in, 1choriocapillaris, changes in, 2–3choroidal neovascularization, 1–2drusen, 4–5geographic atrophy, 5macular translocation and, 276monitoring of, 103–106neurosensory retina, changes in, 3optical coherence tomography and assessment of, 179–181retinal pigment epithelium, changes in, 1–2

Nonocular degenerative diseasesimmune mechanisms, 19–23atherosclerosis, 19–21

inflammatory mechanisms, 19–23Nonocular risk factors, exudative (neovascular) age-related

macular degeneration and, 126–127Nonspecific amplification mechanism, glomerular diseases

and, 23Nutritionantioxidants, 68–70micronutrients, 68as risk factors in macular degeneration, 68–73zinc, 70

NV. See neovascularization.

Occult choroidal neovascularization, 143–144Occult CNV, 161focal, 162hot spot, 162

Occult subretinal pigment epithelium, 6OCT. See optical coherence tomography.Ocular factors, age-related macular degeneration

risk factors and, 57–60cataract surgery, 58–59cup/disc ratio, 60iris color, 59–60macular pigment optical density, 57–58refractive error, 60

Ocular histoplasmosis syndrome, 24Ocular immune disorder, 24–25atrophic retinal degeneration, 24–25choroidal neovascularization (CNV), 24–25complement activation, 24–25histoplasmosis syndrome, 24

Ocular risk factors, exudative (neovascular) age-relatedmacular degeneration and, 127

age-related eye disease study (AREDS), 127Optic nerve prostheses, 320Optical coherence tomography (OCT), 177–183age-related macular degeneration treatment,

181–183assessment ofchoroidal neovascularization, 177–179non-exudative macular degeneration, 179–181

imaging principles, 177–183normal imaging, 177–183quantitative retinal imaging and, 186

Oxidants, 14, 21Oxidative damage, age-related macular degeneration

gene pathogenesis and, 39

Paracentral scotomas, 116–117Pars plana vitrectomy, 281Pathogenesischoroidal neovascularization and, 89–90geographic atrophy, 112

Patient positioning, limited macular translocationand, 284

PDT. See photodynamic therapy.PED. See pigment epithelial detachment.

INDEX 361

Pegaptanib sodium, 247–248Peripherin/RDS gene, 37Pharmacologic treatment, choroidal neovascularization

and, 90–91Photodynamic therapy (PDT), 144–145, 223–229agents of, 224–225benzoporphyrin derivative monoacid, 225

anti-vascular endothelial growth factor and, 251clinical results, 225–229age-related macular degeneration, 226–227verteporfin trials, 226

laser light application, 224light types, 225treatment evolution of, 228–229vascular targeting, 223–224

Pigment epithelial detachment (PED), 133–135Planned neurosensory retinal detachment, 281–283Plaque, 162–163Polypoidal choridal vasculopathy, 139, 163–164Postoperative complications, macular translocation

and, 290–291Postoperative cyclovertical diplopia, management of,

287–288Proangiogenic factors, 88Prostheses, retinal, 319–326

Quantitative retinal imaging, 185–189fluorescein angiography (FA), 186fundus photography, 185–186microperimetry, 187–188optical coherence tomography (OCT), 186

Race, as risk-factor in macular-degeneration, 55–56Radiation studies, review of, 237–243Radiation therapy, 148Radiation treatment, age-related macular degeneration

and, 233–243Radiation treatment delivery methods, 235–237charged particle, 236–237implant therapy, 236

Radiation treatment toxicity issues, 234–235Ranibizumab, 248–250Lucentis, 248–250

Real-time indocyanine green angiography(ICGA), 170

Receptor tyrosine kinase inhibitors, 150, 253–254Recurrent CNV, 168–169Refractive error, as risk factors in macular-

degeneration, 60Reproduction system, age-related macular degeneration

risk factors and, 64estrogen-related, 64

Retinal angiomatous proliferation, 139–140, 164–167Retinal dystrophies, 36–37Retinal pigment epithelial cell attachmentage-related changes and effects of, 337Bruch’s membrane, 335–337cell survival, 335prior and handling and effects of, 336–337tissue culture, 335

Retinal pigment epithelial cell transplantationclinical results, 333–335harvesting technique, 332rationale for, 330–332

Retinal pigment epithelium (RPE), 1changes in, 1–2

Retinal prostheses, 319–326extraocular approaches, 319–320cortical, 319–320optic nerve, 320scleral based, 320

intraocular approaches, 320–326epiretinal prosthesis, 320–323neurotransmitter-based, 325–326subretinal, 323–325

Risk factors, age-related macular degeneration and,epidemiologic studies, 47

RNAs, small interfering, 253RPE. See retinal pigment epithelium.

Sattler’s layer vessels, 212–217Scleral based extraocular stimulation, 320Scotomas, 116–117Serous pigment epithelial detachment, 161Serum lipid levels, as risk factors in macular-

degeneration, 62siRNA. See small interfering RNAs.Small interfering RNAs (siRNAs), 253type therapy, 149

Smoking cigarettes, as risk factor in macular-degeneration, 64–66

Sociodemographic factors, age-related maculardegeneration risk factors and, 53–57

age, 53, 54heredity, 56–57race/ethnicity, 55–56socioeconomic status, 57

Socioeconomic status, as risk-factor in macular-degeneration, 57

Soft drusen, 4–5, 100–101Soluble antiangiogenic factors, 88Soluble proangiogenic factors, 88Squalamine lactate, 150Subfoveal CNV, 141, 142–143management of, 286

Subfoveal RPE defect, macular translocationand, 276

Subretinal choroidal neovascularizationclassic type, 6, 7type 2 growth pattern, 5–6

Subretinal injection, triamcinolone acetonide (TA),297–298

Subretinal pigment epithelium choroidalneovascularization

occult type, 6type 1 growth pattern, 5–6

Subretinal prosthesis, 323–325Sunlight exposure, as risk factors in macular degener-

ation, 66–67Surgical therapy for choroidal neovascularization, 148

362 INDEX

Systemic diseases, age-related macular degeneration and, 98cardiovascular diseases, 98hypercholesterolemia, 98hypertension, 98

Systemic factors, age-related macular degeneration riskfactors and

BMI ratio, 63cardiovascularbiomarkers, 63disease, 60–65

Chlamydia pneumoniae infection, 63–64cigarettes, 64diabetes, 62–63dietary fat intake, 62hematologic factors, 63hyperglycemia, 62–63hypertension and blood pressure, 61–62serum lipid levels, 62waist conference, 63waist-hip ratio, 63

Systemic risk factors, geographic atrophy and, 112–113

T lymphocytes, 16TA. See triamcinolone acetonide.Thermotherapy, 147–1483608 retinotomy, 287TIMP3 gene, 37Tissue culture, retinal pigment epithelial cell attachment

and, 335Tissue plasminogen activator, 295–396Transcription factors, neovascularization and, 89Triamcinolone acetonide (TA), 297–298subretinal injection, 297–298

Tubulin binding agents, 150–151combretastatin A-4 phosphate, 150

Twin aggregation studies, 35–36Type 1 growth pattern, 5–6Type 2 growth pattern, 5–6

Vascular endothelial growth factor (VEGF), 87–91drugs and clinical trials, 247–254extracellular matrix (ECM), 88–89solubleantiangiogenic factors, 88proangiogenic factors, 88

trap therapy, 149–150Vascular endothelial growth trap, 252–253Vascular targeting, 223–224VEGF. See vascular endothelial growth factor.Verteporfin, 225trials, 226, 227–228

Visual function impairmentcentral scotomas, 116–117dimly lit environments, 117–118geographic atrophy and, 116–118other abnormalities, 118paracentral scotomas, 116–117

Visual loss, considerations re, 276–277Visual rehabilitationclinical considerations, 303–317low vision evaluation, 303–307new applications, 307–308new technologies, 308–316

Visudyne e, 225VMD2 gene, 37

Waist conference, as risk factor in macular-degeneration, 63

Waist-hip ratio, as risk factor in macular-degeneration, 63

Wet age related macular degeneration (AMD). See exudativeage related macular degeneration (AMD).

Wide angle indocyanine green angiography (ICGA), 170

Zeaxanthin, 72Zinc, 70

INDEX 363

age related macular-degeneration__2nd_edition

Documents