DIAGNOSIS AND RISK PREDICTION OF DENTAL CARIES, VOL 2 (2000)

Front Matter

THE AXELSSON SERIES ON PREVENTIVE DENTISTRY

The world-renowned authority on preventive and community dentistry presents his life's work in this five-volume series of clinical atlases focusing on risk prediction of dental caries and periodontal disease and on needs-related preventive and maintenance programs.

Volume 1  An Introduction to Risk Prediction and Preventive DentistryProvides a general overview of current and future trends in risk prediction, control, and nonaggressive management of caries and periodontal disease; preventive dentistry methods and programs; and quality control.

Volume 2  Diagnosis and Risk Prediction of Dental CariesIncludes a comprehensive discussion of the etiology, pathogenesis, diagnosis, risk indicators and factors, individual risk profiles, and epidemiology of caries.

Volume 3  Diagnosis and Risk Prediction of Periodontal DiseasesPresents a comprehensive discussion of the etiology, pathogenesis, diagnosis, risk indicators and factors, individual risk profiles, and epidemiology of periodontal diseases. Considers periodontal diseases as a possible risk factor for systemic diseases and presents current and future trends in the management of periodontal diseases, including nonaggressive debridement and preservation of the root cementum.

Volume 4  Preventive Materials, Methods, and ProgramsDiscusses self-care and professional methods of mechanical and chemical plaque control, use of fluorides and fissure sealants, and integrated caries prevention. Addresses needs-related preventive programs based on risk prediction and computer-aided epidemiology analysis for quality control and outcome.

Volume 5  Nonaggressive Treatment, Arrest, and Control of Periodontal Diseases and Dental CariesDetails current and future trends in nonaggressive treatment methods that seek to preserve the root cementum; surgical versus nonsurgical periodontal therapy; repair and regeneration of periodontal support; management of furcation-involved teeth; restricted use of antibiotics; arrest of noncavitated enamel, dentin, and root carious lesions; nonaggressive mini-preparations; esthetic and hygienic aspects of restorations; and management of erosions. Focuses on needs-related maintenance programs to ensure the long-term success of treatment and to prevent recurrence of periodontal disease and dental caries.

TITLE PAGE

DIAGNOSIS AND RISK PREDICTION OF DENTAL CARIES, VOL 2

Per Axelsson, DDS, Odont Dr

Professor and ChairmanDepartment of Preventive DentistryPublic Dental Health Service

Karlstad, Sweden

Quintessence Publishing Co, IncChicago, Berlin, London, Tokyo, Paris, Barcelona, Sao Paulo, Moscow, Prague, and Warsaw

DEDICATION

To my wife Ingrid, my daughter Eva, and my son Torbjorn

COPYRIGHT PAGE

Library of Congress Cataloging-in-Publication Data

Axelsson, Per, D.D.S., Odont. Dr.  Diagnosis and risk prediction of dental caries / Per Axelsson.   p. cm. (The Axelsson series on preventive dentistry; vol. 2)  Includes bibliographical references and index.  ISBN 0-86715-362-8  1. Preventive dentistry. 2. Dental cariesPrevention. 3. Periodontal diseasePrevention. I. Title. II. Title: Risk prediction and preventive dentistry. III. Series: Axelsson, Per, D.D.S. Axelsson series on preventive dentistry; vol. 2.  [DNLM: 1. Dental Cariesprevention & control. 2. Periodontal Diseasesprevention & control. 3. Preventive Dentistry. 4. Risk Factors. WU 270 A969i 2000] RK60.7.A94 2000 617.6dc21 DNLM/DLC for Library of Congress         99-16511                                 CIP

2000 Quintessence Publishing Co, Inc

Quintessence Publishing Co, Inc

4350 Chandler DriveHanover Park, IL 60133www.quintpub.com

All rights reserved. This book or any part thereof may not be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, or otherwise, without the written permission of the publisher.

Editor: Cheryl Anderson-WiedenbeckProduction: Gerda Steinmeyer

Printed in Germany

CONTENTS

   Preface vii

Chapter 1 Etiologic Factors Involved in Dental Caries 1

   Role of Plaque 2   Role of the Oral Environment 14   Role of Specific Cariogenic Microflora 18   Prediction of Caries Risk 29   Conclusions 40

Chapter 2 External Modifying Factors Involved in Dental Caries 43

   Role of Dietary Factors 43   Role of Socioeconomic and Behavioral Factors 77   Conclusions 86

Chapter 3 Internal Modifying Factors Involved in Dental Caries 91

   Role of Saliva 91   Role of Chronic Systemic Diseases and Impaired Host Factors 134   Role of Tooth Size, Morphology, and Composition 139   Conclusions 146

Chapter 4 Prediction of Caries Risk and Risk Profiles 151

   Risk Groups 152   Individual Risk 155   Key-Risk Teeth and Surfaces 161   Risk Profiles 165   Cariogram Model 172   Conclusions 174

Chapter 5 Development and Diagnosis of Carious Lesions 179

   Development of Carious Lesions 181

   Diagnosis and Registration of Carious Lesions 208   Conclusions 245

Chapter 6 Epidemiology of Dental Caries 249

   Limitations of Epidemiologic Surveys 249   Prevalence of Caries 253   Incidence of Caries 269   Caries Treatment Needs 272   Reasons for Changes in Caries Prevalence 275   Conclusions 278

   References 281   List of Abbreviations 297   Index 299

PREFACE

The etiology of dental caries and periodontal diseases is well understood, and we have now developed efficient methods of preventing and controlling these diseases. Over the last 25 years in County of Varmland, Sweden, large-scale implementation of our prevention programs has led to an increase in the percentage of caries-free 3 year olds, from 30% to 97%, while reducing caries in 12 year olds from an average of 25 DFS to less than 1 (0.6). In the last 10 years, we have increased the number of remaining teeth in 65 year olds by more than 15% and reduced their loss of periodontal support by more than 20% at the same time reducing the percentage who are edentulous from 17% to 7%.

According to the principles of lege artis, all members of our profession are obliged to offer treatment based on the most current scientific and clinical knowledge available. As we enter the new millennium, we must therefore continue to concentrate our efforts on preventing, controlling, and arresting dental caries and periodontal diseases. However, needs-related preventive and maintenance programs must be cost effective and should be based on information derived from comprehensive diagnoses, histories, and risk predictions at group, individual, and tooth surface levels. For quality control and evaluation of such programs, computer-aided analytical epidemiology, using relevant variables, should be introduced.

The aim of this clinical textbookthe second volume in a series of fiveis to provide up-to-date knowledge about the etiology, modifying factors, risk evaluation, development, diagnosis, and epidemiology of dental caries. A detailed scientific background is presented for each topic discussed, as well as an illustrated guide to implementing the state-of-the-art and conclusions and future recommendations. Thus this book will be useful not only for dentists and dental hygienists, but also for undergraduate and postgraduate students and teachers.

This project could not have been completed without the assistance and support of my family, friends, and colleagues. I offer my deepest thanks to my wife Ingrid and my daughter Eva and my son Torbjorn and their families, as well as to all my other relatives and friends, for their patience and understanding throughout the last 5 years

in which I spent almost every night, weekend, and vacation preparing the material for these five volumes. I also wish to thank my wonderful staff at the Department of Preventive Dentistry, Public Dental Health Service, County of Varmland, for all their service, and particularly my assistant, Pia Hird, who typed most of my manuscript. I owe special thanks to Art Director Fredrik Persson, Dr Jorgen Paulander, and the Dumex Company for their excellent support with computer-aided illustrations, and to Associate Professor Joan Bevenius for her work in checking my English manuscript.

I am very grateful to all my colleagues and friends around the world and to several publishers (Munksgaard International, the American Academy of Periodontology, S. Karger Medical and Scientific Publishers, FDI World Dental Press, WHO Oral Health Unit), who have generously permitted me to use their illustrations (about 20% of the total). Last but not least, the excellent cooperation of the publisher is gratefully acknowledged.

Chapter 1. Etiologic Factors Involved in Dental Caries

Introduction

Dental caries is an infectious, transmissible disease. As early as 1954, Orland et al demonstrated that, although germ-free animals do not develop caries, even with frequent sugar intake, all animals in the group rapidly develop carious lesions when human cariogenic bacteria (mutans streptococci) are introduced in the mouth of one animal. Specific bacteria (acidogenic and aciduric) that colonize the tooth surfaces are recognized as etiologic factors in dental caries. Frequent intake of fermentable carbohydrates, such as sugar, is regarded only as an external (environmental) modifying risk factor or prognostic risk factor. In the presence of these and other external risk factors, the outcome may be modified by internal host factors, such as the quality of the teeth and the amount and quality of saliva (Fig 1):

1. Microflora: acidogenic bacteria that colonize the tooth surface.

2. Host: quantity and quality of saliva, the quality of the tooth, etc.

3. Diet: intake of fermentable carbohydrates, especially sucrose, but also starch.

4. Time: total exposure time to inorganic acids produced by the bacteria of the dental plaque.

The development of a clinical carious lesion involves a complicated interplay among a number of factors in the oral environment and the dental hard tissues. A simplified explanatory model of the major events is illustrated in Fig 2. The carious process is initiated by bacterial fermentation of carbohydrates, leading to the formation of a variety of organic acids and a fall in pH. Initially, H+ will be taken up by buffers in plaque and saliva; when the pH continues to fall (H+ increases), however, the fluid medium will be depleted of OH- and PO3

4-, which react with H+ to form H2O and HPO2

4- On total depletion of these compounds, the pH can fall below the critical value of 5.5, at which point the aqueous phase becomes undersaturated with respect to

hydroxyapatite. Therefore, whenever surface enamel is covered by a microbial deposit, the ongoing metabolic processes within this biomass cause fluctuations in pH, and occasional steep falls in pH, which may result in dissolution of the mineralized surface.

In their classic study of experimental caries in humans, von der Fehr et al (1970) showed that, in the absence of oral hygiene (ie, with free accumulation of plaque and rinsing nine times a day with a sucrose solution), clinical signs of enamel caries develop within 3 weeks. When the same research team repeated the study, but introduced chemical plaque control (rinsing twice a day with 0.2% chlorhexidine solution), the subjects did not develop caries, even though they rinsed with sucrose solution nine times a day for 3 weeks (Loe et al, 1972). In other words, when the etiologic factor was suppressed or eliminated, the precondition for caries did not exist, and no lesions developed, despite the subjects' very frequent exposure to sucrose.

Like the inflammation induced in the gingival soft tissues adjacent to the gingival plaque, carious lesions of enamel, which develop on individual tooth surfaces beneath the undisturbed bacterial plaque, represent the net result of an extraordinarily complex interplay among "harmless" and "harmful" bacteria, antagonistic and synergistic bacterial species, their metabolic products, and their interaction with the many salivary and other host factors. This explains why combinations of different nonspecific plaque control programs have been so effective against caries, gingivitis, and periodontitis (for review, see Axelsson, 1994, 1998). However, more recently, there has been intense interest in the role plaque (amount, formation rate, and ecology) and specific cariogenic microflora play in the etiology of dental caries.

Fig 1 Diagram of the development of dental caries. Interaction among etiologic risk factors (microflora), external modifying risk factors (diet), internal modifying risk factors (host), and time exposed. (Modified from Keyes, 1960.)

Fig 2 Development of noncavitated enamel caries. (Modified from Fejerskov and Clarkson, 1996.)

Role of Plaque

Development of plaque

According to Dawes et al (1963), dental plaque is "the soft tenacious material found on tooth surfaces which is not readily removed by rinsing with water." It is estimated that 1 mm3 of dental plaque, weighing about 1 mg, will contain more than 200 million bacteria. Other microorganisms, such as mycoplasma, "yeasts," and protozoa, also occur in mature plaque; sticky polysaccharides and other products form the so-called plaque matrix, which constitutes 10% to 40% by volume of the supragingival plaque.

The most readily discernible plaque on the smooth surfaces of the teeth, along the gingival margin, is termed dentogingival plaque. Dentogingival plaque on the approximal surfaces, apical to the contact points, is called approximal dental plaque.

Plaque occurring below the gingival margin, in the gingival sulcus or in the periodontal pocket, is known as subgingival plaque (Theilade and Theilade, 1976). Occlusal or fissure plaque may also be formed, particularly in erupting molars.

Although there are more than 350 species of bacteria in the oral cavity, only a few have the ability to colonize a newly cleaned tooth surface. This initial association depends on the presentation and interaction of surface molecules on the bacteria and the pellicle-coated tooth surface. These molecules are vulnerable to alteration by chemical agents. Plaque adhesion is especially favored by high free energy of the tooth surfaces and the microorganisms.

The initial bacteria are called pioneer colonizers, because they are hardy and successfully compete with the other members of the oral flora for a place on the tooth surface (Gibbons and Van Houte, 1980). These pioneer colonizers are mainly the streptococcal strains S oralis, S mitior, and S sanguis. The deposition of these pioneer species is not a chance occurrence, but the outcome of an exquisitely sensitive interaction between protein adhesions on the surface of the colonizing bacteria and carbohydrate receptors on the salivary components adsorbed to the tooth surface.

After initial deposition, clones of pioneer colonizing bacteria, in particular Streptococcus sanguis, begin to expand away from the tooth surface to form columns that move outwardly in long chains of pallisading bacteria. These parallel columns of bacteria are separated by uniformly narrow spaces. Plaque growth proceeds by deposition of new species into these open spaces (Listgarten et al, 1975). Figure 3 illustrates a cross section of such columns and open spaces.

These newly deposited species attach to pioneer species in a specific, molecular locking manner. Expansion of existing species in a lateral direction causes the interbacterial spaces to merge. It is hypothesized that, when the spaces are close enough, a starter substance is secreted by bacteria within the plaque matrix, signaling the surrounding bacteria to undergo a growth spurt. Within a short time, the tooth surface adjacent to the gingiva is covered by intermeshed bacteria. New bacteria derived from saliva or surrounding mucous membranes now sense only the bacteria-laden landscape of the tooth surface and attach by a bonding interaction to bacteria already attached to the plaque. These associations, called intergeneric coaggregations, are mediated by specific attachment proteins that occur between two partner cells (Di Renzo et al, 1985; Kolenbrander, 1988).

All this activity occurs within the first 2 days of plaque development and, for descriptive purposes, is called phase I of plaque formation (Theilade et al, 1976). After 24 to 48 hours, continuous plaque has formed along the gingival margin (Fig 4). The plaque is dominated by cocci and a few rods.

In phase II of plaque development, the remaining interstices are occupied by increasing levels of gram-positive rods, such as Actinomyces viscosus, and gram-negative cocci, including Neisseria and Veillonella species (Fig 5). The outer surface of the gingival plaque is covered by tall rods. Figure 6 illustrates the thickness of freely accumulated gingival plaque after 2, 3, and 4 days. There is a dramatic increase in plaque thickness after 3 and 4 days compared to the first 2 days. Now the gingival plaque is mature, and so-called homeostasis is established among the different

microorganisms.

In phase III, 5 to 7 days after initiation, plaque begins to migrate subgingivally, and bacteria and their products permeate and circulate in the pocket. In phase IV, 7 to 11 days after initiation, the diversity of the flora increases to comprise motile bacteria, including spirochetes and vibrios as well as fusiforms. Attached gingival plaque fills the gingival sulcus, while spirochetes and vibrios move around in the outer and more apical regions of the sulcus (Fig 7).

Fig 3 Cross section of 2-day-old gingival plaque growth. (From Listgarten et al, 1975. Reprinted with permission.)

Fig 4 Scanning electron micrograph showing 24 to 48 hours of continuous plaque formation along the gingival margin (arrow). Plaque is dominated by cocci (right). (Courtesy A. Saxton)

Fig 5 Surface of 3- to 4-day-old gingival plaque. (Courtesy A. Saxton)

Fig 6 Thickness of gingival plaque: 2, 3, and 4 days old. (From Listgarten, et al, 1975. Reprinted with permission.)

Fig 7 Gingival plaque and motile microflora filling the gingival sulcus. (From Listgarten, 1976. Reprinted with permission.)

Measurement of plaque

Amount of accumulation

Several indices for recording supragingival plaque have been developed. The two most frequently used are the Plaque Index (PI), developed by Silness and Loe (1964), and O'Leary's Plaque Index (O'Leary et al, 1972).

The Silness and Loe Plaque Index has a four-point scale:

 Score 0 = The tooth surface is clean.

 Score 1 = The tooth surface appears clean, but dental plaque can be removed from the gingival third with a sharp explorer.

 Score 2 = Plaque is visible along the gingival margin.

 Score 3 = The tooth surface is covered with abundant plaque.

O'Leary's Plaque Index is based on the visible continuous plaque along the gingival margin after staining. Four or six sites per tooth are examined, and the percentage of tooth surfaces exhibiting stained plaque is calculated. Unlike Silness and Loe's PI, no attempts are made to evaluate the area of tooth surface covered by plaque. O'Leary's

Plaque Index is most commonly used for evaluation of the oral hygiene standard of the individual patient and for patient motivation, based on self-diagnosis.

Used in dental practice, the Plaque Index is capable only of revealing areas that the patient has failed to clean effectively, even though he or she may have made a special effort on the day of the dental appointment; it does not indicate the rate at which plaque forms in the individual or the oral hygiene status 1 week before or after the dental appointment. This accounts for the failure by clinicians as well as examiners in clinical studies to observe the correlation between on one hand, the amount and location of plaque and, on the other, the sites of carious lesions.

Despite these limitations, disclosure of plaque by staining is the fastest and most efficient method for self-diagnosis by the patient. The technique also allows the clinician to locate remaining plaque and to demonstrate the close relationship between the localization of plaque and the presence of gingivitis and dental caries (Figs 8 and 9). Prevention of dental caries and gingivitis must, therefore, be based on plaque control.

The telemetric method, developed by Graf and Muhlemann (1966), allows in vivo measurement of the "true" pH on the tooth surface beneath the undisturbed plaque. The importance of the age, amount, and composition of plaque, as well as different concentrations of sugar, can thereby be evaluated. Using the telemetric method, Imfeld (1978a) showed that rinsing with a 10% sucrose solution causes a dramatic drop in pH to below 4 in 3-day-old interdental plaque. Such plaque is typical for the approximal surfaces of the molars and premolars in a toothbrushing population. In contrast, the fall in pH in immature lingual plaque (12 hours old) is very limited (Fig 10).

Firestone et al (1987) used the same telemetric test in vivo, measuring the pH drop after subjects rinsed with a 10% sucrose solution. Four different sites on molars with approximal plaque (Fig 11) were compared to plaque-free approximal surfaces. The authors concluded: "removing plaque from interdental surfaces significantly reduced the exposure of the surfaces to plaque acids following sucrose rinse. This further supports mechanical removal of plaque from interdental surfaces as a means of reducing dental caries."

In toothbrushing populations, that is, those who have an established habit of using a toothbrush and fluoride toothpaste daily, dental plaque more than 2 days old is located mainly on the approximal surfaces of the molars and premolars, partly subgingivally. Access with a toothbrush to the wide approximal surfaces is limited by the buccal and lingual papillae. At least in European countries, although daily toothbrushing with a fluoride dentifrice is an established oral hygiene habit, special aids to approximal oral hygienesuch as dental floss, dental tape, toothpicks, and interdental brushesare used daily by fewer than 10% of the population. These conditions explain why caries, gingivitis, and marginal periodontitis are much more prevalent on the approximal surfaces of the molars and premolars than on the buccal and lingual surfaces of the dentition.

Rate of accumulation (Plaque Formation Rate Index)

The quantity of plaque that forms on clean tooth surfaces during a given time represents the net result of interactions among etiologic factors, many internal and external risk indicators and risk factors, and protective factors:

 The total oral bacterial population

 The quality of the oral bacterial flora

 The anatomy and surface morphology of the dentition

 The wettability and surface tension of the tooth surfaces

 The salivary secretion rate and other properties of saliva

 The intake of fermentable carbohydrates

 The mobility of the tongue and lips

 The exposure to chewing forces and abrasion from foods

 The eruption stage of the teeth

 The degree of gingival inflammation and volume of gingival exudate

 The individual oral hygiene habits

 The use of fluorides and other preventive products, such as chemical plaque control agents

This observation has been the rationale for the development of the Plaque Formation Rate Index (PFRI) by Axelsson (1989, 1991). The index includes all but the occlusal tooth surfaces and is based on the amount of disclosed plaque freely accumulated (de novo) in the 24 hours following professional mechanical toothcleaning (PMTC), during which period subjects refrain from all oral hygiene practices. In a pilot study on 50 adult subjects, adherent plaque was disclosed on 5% to 65 % of the total number of tooth surfaces (for details on materials and methods, see Axelsson, 1989, 1991). On the basis of this study, the following five-point scale was constructed for the PFRI.

 Score 1 = 1% to 10% of surfaces affected: very low

 Score 2 = 11% to 20% of surfaces affected: low

 Score 3 = 21% to 30% of surfaces affected: moderate

 Score 4 = 31% to 40% of surfaces affected: high

 Score 5 = More than 40% of surfaces affected: very high

The PFRI was evaluated in a large-scale cross-sectional study of 14-year-old

schoolchildren (n = 667) in the city of Karlstad, Sweden, in 1984. The subjects were followed over a 5-year period, up to the age of 19 years (Axelsson, 1989, 1991).

Many indicators and factors possibly related to PFRI were also evaluated, including (1) caries prevalence and caries incidence; (2) gingival inflammation; (3) Plaque Index; (4) dietary intake during the 24 hours of free plaque accumulation; (5) salivary levels of Streptococcus mutans and glucosyl transferase; (6) agglutinin levels in resting saliva; and (7) oral hygiene, dietary, and fluoride habits.

Figure 12 shows the frequency distribution of PFRI scores 1 to 5 among the 14-year-old schoolchildren. The majority were low (score 2 = 48%) or moderate (score 3 = 27%) plaque formers. However, the standard of oral hygiene is very high among schoolchildren in Karlstad, and, as a consequence, the caries prevalence is low.

The relationship between caries prevalence (the mean number of decayed or filled surfaces) and different scores is presented in Fig 13. These results indicated a threshold for caries risk between PFRI scores 2 and 3, and this was subsequently confirmed in the longitudinal part of the study over 5 years (Axelsson, 1989, 1991).

Among other observations from the study were:

1. Individuals with a PFRI score of 4 or 5 had considerably higher scores for gingival bleeding than did individuals with a PFRI score of 1 or 2.

2. An initially high Plaque Index usually correlated with PFRI scores 3 to 5.

3. There was no significant correlation between different salivary S mutans levels and PFRI scores.

4. The level of salivary glucosyl transferase was lower in individuals with a PFRI score of 4 or 5 than in those with a score of 1 or 2, probably because glucosyl transferase had already accumulated in the matrix of the plaque in the high and very high plaque formers.

5. The scores for individuals with a very low and low PFRI (scores 1 and 2, respectively) tended to remain constant over the 5-year period, while the scores of some individuals with PFRI scores of 3 to 5 tended to vary, increasing or decreasing by 1 unit.

This final observation indicates that plaque formation rates in individuals with a PFRI score of 4 or 5 can be reduced; such individuals should, therefore, be thoroughly evaluated to identify the factors that contribute to their rapid plaque formation. Needs-related preventive measures could then be introduced.

For example, there is a strong correlation among plaque formation rate, the severity of gingival inflammation, and the volume of gingival exudate (Axelsson, 1989; Quirynen et al, 1986a; Ramberg et al, 1994a,b, 1995; Saxton 1973, 1975). Initially intensive and frequent mechanical and chemical plaque control, both self-care and professional, is indicated in individuals with PFRI scores of 4 and 5 and high gingival index scores to heal all inflamed sites as quickly as possible and thereby reduce the

plaque formation rate.

If the high plaque formation rate is associated with inadequate salivary secretion, frequent plaque control measures (before every meal) should be supplemented with salivary stimulation, provided by the use of fluoridated chewing gum immediately after every meal.

A high intake of fermentable carbohydratesparticularly sucrosewill result in sticky plaque, rich in polysaccharides, and an increased plaque formation rate (Carlsson and Egelberg, 1965). Needs-related prevention for individuals with a PFRI score of 4 or 5 and a frequent intake of sugar-containing products should, therefore, emphasize not only frequent plaque control but also a reduction in the frequency of sugar intake. In the above study, it was observed that some individuals with a PFRI score of 5 had consumed several bananas during the 24-hour period of free plaque accumulation.

Many other factors are also related to plaque formation rate. For example, antimicrobial proteins of human whole saliva may influence plaque formation rate (Box 1).

The PFRI has recently been applied in studies on different populations and age groups. From more than 1,000 17 to 19 year olds in the city of Karlstad, Sweden, 30% with the highest gingival index score were selected to participate in a 4-month double-blind mouthrinse study. At baseline, most of the subjects had PFRI scores of 3 (more than 40%) or 4 (about 25%) (Axelsson et al, 1994). Subjects with the highest gingival index scores also had the highest PFRI scores. In addition, sites with gingival inflammation had significantly higher plaque formation rates than healthy gingival sites (Rahmberg et al, 1995).

Brazil has the highest caries prevalence in the world. In Sao Paulo, a 3-year caries-preventive study based on self-diagnosis and self-care was carried out in 12- to 15-year-old schoolchildren; the PFRI was used as a tool for self-diagnosis and establishment of needs-related oral hygiene habits. At baseline, almost 100% of the 12-year-old schoolchildren had a PFRI score of 5. The mean percentage of surfaces with reaccumulated plaque was more than 70%, probably because of the extremely high caries prevalence, a high gingival index, and the presence of erupting permanent teeth. At reexamination of the subjects 3 years later, the PFRI had dropped significantly: most of the 15 year olds had scores of 3 or 4. The main contributing factors were an improvement in oral hygiene habits and gingival health and the fact that all teeth were now fully erupted (Albander et al, 1995; Axelsson et al, 1994; Buischi et al, 1994).

In Duisburg, Germany, the PFRI was evaluated in different age groups of children; preschool children, children with mixed dentitions and erupting permanent teeth, and children with fully erupted teeth. Children with erupting teeth had the highest PFRI scores (Fig 14). However, the German children generally had higher PFRI scores than did Swedish children of comparable age with very low caries prevalence, excellent gingival conditions, and good oral hygiene habits (Fig 15) (Axelsson, 1991; Cunea and Axelsson, 1997). According to the World Health Organization's Data Bank (1993), caries prevalence is high among 12-year-old German children.

Pattern of plaque reaccumulation

As discussed earlier, plaque formation rate is influenced by such factors as (1) the anatomy and surface morphology of the teeth; (2) the stage of eruption and functional status of the teeth; (3) the wettability and surface tension of the tooth surfaces (both intact and restored surfaces); and (4) gingival health and volume of gingival exudate. The pattern of plaque reaccumulation will also be influenced by these factors, but may differ somewhat between, on the one hand, tooth surfaces exposed to chewing forces, abrasion from foods, and friction from the dorsum of the tongue, the lips, and the cheeks, and, on the other hand, less accessible areas, such as approximal sites, along the gingival margin, and in irregularities such as occlusal fissures. These less accessible areas are often designated "stagnation areas" for plaque.

In a 6-week study by Lang et al (1973), plaque reaccumulation was registered in four groups of dental students who carried out oral hygiene procedures (mechanical toothcleaning by self-care) with different frequencies: twice daily or every second, third, or fourth day. Figure 16 shows the pattern of reaccumulated plaque according to the Silness and Loe (1964) Plaque Index (scores 0 to 3) on the distal, mesial, facial, and lingual surfaces of the maxillary and mandibular teeth. After only 12 hours of free plaque reaccumulation, there was visible plaque on some of the approximal surfaces of the molars and the lingual surfaces of the mandibular molars (score 2). After 48 hours, almost 100% of these surfaces and most of the remaining approximal surfaces had scores of 2 or 3. The pattern of visible plaque after 2 and 3 days seems to be similar, except on the facial surfaces.

According to Listgarten (1976), freely accumulated plaque is about five times thicker after 3 days than after 2 days (see Fig 6). This explains why gingivitis developed in the group of students cleaning only every third or fourth day but not in those who cleaned at least every second day. It also explains the striking difference in response to rinsing with 10% sucrose solution shown by Imfeld (1978): a dramatic fall in pH on approximal surfaces covered by 3-day-old plaque compared to the pH on lingual surfaces covered by immature (12-hour) plaque (see Fig 10).

Figure 17 presents the percentage of freely reaccumulated (de novo) plaque, 24 hours after PMTC, in 667 14-year-old children in the city of Karlstad (Axelsson, 1989, 1991). Plaque reaccumulation was greatest on the mesiolingual and distolingual mandibular surfaces (33%), particularly on the molars, followed by the mesiobuccal and distobuccal surfaces of both maxillary and mandibular teeth, particularly on the molars. There was almost no plaque reaccumulation (3%) on the palatal surfaces of the maxillary teeth, mainly because of friction from the rough dorsum of the tongue.

Figures 18 and 19 illustrate the percentage of de novo plaque on maxillary and mandibular tooth surfaces, respectively, 24 hours after PMTC, in young German subjects (Cunea and Axelsson, 1997). The highest percentages are found in 6 to 14 year olds with many erupting teeth on distobuccal and mesiobuccal surfaces of molars, and on distolingual and mesiolingual surfaces of mandibular molars.

Carvalho et al (1989) studied the pattern and amount of de novo plaque, 48 hours after PMTC, on the occlusal surfaces of partly and fully erupted first molars. Figure 20 illustrates the heavy plaque reaccumulation, particularly in the distal and central

fossae, in the eruptin maxillary and mandibular molars, in contrast to the reaccumulation in the fully erupted molars, which are subjected to normal chewing friction. Abrasion from normal mastication significantly limits plaque formation; this explains why almost 100% of occlusal caries in molars begins in the distal and central fossae during the eruption period of 14 to 18 months.

It is important to differentiate between plaque indices and plaque reaccumulation rate (PFRI). For successful primary and secondary prevention of dental caries and periodontal diseases, an understanding of plaque formation rates and patterns is essential. Mechanical removal of dental plaque according to the nonspecific plaque hypothesis is a rational method for prevention and control of periodontal diseases as well as dental caries, because it is directed toward the cause (etiology) of these diseases. However, for cost-effectiveness, the program should be related to the pattern of plaque reaccumulation, PFRI, and predicted risk. (For reviews on plaque formation rate and the role of needs-related plaque control see Axelsson 1994, 1998.)

Fig 8 The anterior teeth of a 12-year-old boy with gingivitis at the following sites: 13 mesiobuccal (mb); 12 distobuccal (db); 21 db; 22 mb, db; 23 mb; 43 mb; 42 db; 32 db; and 33 mb. Enamel caries is found at 13 mb, 43 mb, 42 b, 32 db, 33 mb, and 34 mb. There is a cavity on 22 d. Fig 9 A disclosing agent was used to visualize plaque.Observe the close relationship between the localization of gingivitis, carious lesions (Fig 8), and dental plaque.

Fig 10 Limited drop in pH in 12-hour-old lingual plaque compared to critical drop in pH beneath 3-day-old (3d) interdental plaque after rinsing with 10% sucrose solution. (Courtesy T. Imfeld.)

Fig 11 pH drop in molars with approximal plaque at four different sites after rinsing with sucrose solution. (From Firestone et al, 1987.)

Fig 12 Frequency distribution of PFRI scores (1 to 5) among 14-year-old schoolchildren. (From Axelsson, 1989, 1991.)

Fig 13 Relationship of caries prevalence and PFRI scores. (From Axelsson, 1989, 1991.)

Fig 14 Plaque accumulation, by age. (From Cunea and Axelsson, 1997.)

Fig 15 PFRI scores in German and Swedish schoolchildren. (From Cunea and Axelsson, 1997.)

Fig 16 Pattern of reaccumulated plaque according to the Silness and Loe Plaque Index. (Modified from Lang et al, 1973.)

Fig 17 Percentage of freely accumulated plaque, 24 hours after PMTC, in 667 14-year-old children in the city of Karlstad. (From Axelsson, 1989, 1991.)

Fig 18 Percentage of freely accumulated plaque on maxillary tooth surfaces 24 hours after PMTC. (From Cunea and Axelsson, 1997.)

Fig 19 Percentage of freely accumulated plaque on mandibular tooth surfaces 24 hours after PMTC. (From Cunea and Axelsson, 1997.)

Fig 20 Pattern of freely accumulated plaque on the occlusal surfaces of partly and fully erupted first molars 48 hours after PMTC. (Modified from Carvalho et al, 1989.)

Role of the Oral Environment

Introduction

In certain aspects, the oral cavity may be regarded as a single microbial ecosystem. A major regulatory factor is the flow rate of saliva, which decreases to almost 0.0 mL/min during sleep, is approximately 0.4 mL/min at rest, and increases to 2.0 mL/min after stimulation.

Although saliva is not a good medium for supporting the growth of many bacteria, 1.0 mL of whole saliva may contain more than 200 million microorganisms, representing more than 300 different species. Most originate from local environments in the oral cavity, but a minority belong to the so-called normal microflora of saliva and obtain nutrients from salivary proteins.

All the surfaces of the oral cavity are colonized by microorganisms. The facultatively anaerobic streptococci constitute an essential part of the microflora that constantly colonize the mucous membranes and the teeth. Microorganisms are regularly swallowed with saliva and the amount within the oral cavity fluctuates, simply because the microbial deposits building up on mucous membranes and, in particular, on tooth surfaces grow and multiply, thus providing a reservoir for the oral environment. Fluctuations also occur during sleeping and waking hours, and also as a result of such activities as eating and drinking and oral hygiene procedures.

Because the composition of the microflora in mixed saliva is mainly a result of the microorganisms that colonize oral surfaces, the salivary microflora to some extent reflects the gross composition of the microbial deposits on the various oral surfaces.

The oral cavity presents two types of surface for colonization by bacteria, the soft tissues and the hard tooth surfaces, modified to some extent by a coating of saliva, or, in the case of the hard surfaces, by a pellicle formed by adsorption of salivary components. A distinct and important difference between the two types of surface is that the soft tissue surfaces are lost when the cells are shed; thus, readherence of bacteria is essential for survival. In contrast to the hard surfaces which will support heavy deposits of bacteria in dental plaque, the soft tissue surfaces do not support formation of complex layers of bacteria (biofilms).

It is also likely that microbes attached to desquamated epithelial cells spread, via saliva, to different tooth surfaces and typically colonize sheltered regions: interproximal spaces, gingival margins, and occlusal fissures (Saxton, 1975).

Colonization of microenvironments

The oral cavity consists of several major and minor compartments, each constituting a separate microenvironment not easily affected by major events in the oral cavity. Examples of major compartments are the tongue, the oral mucosa, and the tonsils. The different approximal tooth surfaces, occlusal fissures, and gingival sulci are regarded as minor compartments.

A specific area that supports a bacterial flora is termed a habitat. The flora of a habitat develops through a series of stages, collectively called colonization. Colonization is a complex process, because it involves not only interaction between bacteria and their environment but also interactions among different bacteria. The first important prerequisite for colonization is access. The organisms must be able to enter the habitat and consequently they must be able to be transmitted from one habitat to another. For example, mothers can serve as reservoirs for oral bacteria, which they transmit to their children. Within a single host, bacterial reservoirs can aid survival of the organism.

In the human mouth, not only the oral mucosa but also the tongue and tonsils may serve as as reservoirs for bacteria, which, under favorable conditions, may colonize the teeth as well as the periodontal pockets. It is well known that the dorsum of the tongue is the main reservoir for Streptococcus salivarius, which is a very potent, cariogenic (acidogenic) bacteria. In one study, however, higher numbers of S mutans were repeatedly found on the dorsum of the tongue after five thorough scrapings with a tongue scraper than prior to scraping, indicating this to be an important reservoir (Axelsson et al, 1987). Lindquist et al (1989) found a significant correlation between the prevalence of S mutans in saliva and its prevalence on the dorsum of the tongue. These data support the inclusion of the dorsum of the tongue in oral hygiene procedures, at least in patients highly infected by periopathogens and/or cariogenic bacteria, such as S mutans.

Although there are general definitions of habitats, studies of the oral microflora should always include careful definition of the habitats being examined. It is important to recognize that the physical dimensions of a habitat do not fall within specific limits: The whole oral cavity, an occlusal tooth surface, or even a defined area on the occlusal surface may be considered a habitat. In oral microbiology, changes in the flora of a habitat such as the saliva may indicate, for example, patients at risk of developing caries, while changes in tooth surface microenvironments can

identify a surface at risk of disease.

Effect of plaque ecology

Owing to differences in local environmental conditions, the microflora of mucosal surfaces differs in composition from that of dental plaque. Similarly, the plaque microflora varies in composition at distinct anatomic sites on the toothfor example, in fissures, on approximal surfaces, and in the gingival crevice. The resident microflora of a site acts as part of the host defenses by preventing colonization by exogenous (and often pathogenic) microorganisms.

The early colonizers of the tooth surface include members of the genera Streptococcus, Actinomyces, Haemophilus, Neisseria, and Veillonella (Liljemark et al, 1986; Nyvad and Kilian, 1987). These bacteria adhere to the acquired enamel pellicle by specific and nonspecific molecular interactions between adhesions on the cell and receptors on the surface (Busscher et al, 1992; Gibbons, 1989). Once established, the microflora at a site remains relatively stable over time, despite regular minor disturbances in the oral environment (Marsh, 1989). This stability (termed microbial homeostasis) stems not from any metabolic indifference among the components of the microflora but rather from a dynamic balance of microbial interactions, including both synergism and antagonism (Sanders and Sanders, 1984).

It has been proposed that the ability to maintain homeostasis within a microbial community increases with its species diversity (Alexander, 1971). In dental plaque, diversity is enhanced by the development of food chains between bacterial species and their use of complementary metabolic strategies for the catabolism of endogenous nutrients, such as glycoproteins and proteins. Individual species possess different but overlapping patterns of enzyme activity, so that certain mixed cultures of oral bacteria can synergistically degrade complex host molecules (van der Hoeven and Camp, 1991). Antagonism is also a major mechanism in maintaining microbial homeostasis in plaque (James and Tagg, 1988; Marsh, 1989).

Unless removed by diligent oral hygiene, plaque accumulates preferentially at stagnant or retentive sites, such as the posterior approximal surfaces, the fissures of erupting molars, and along the gingival margin. Igarashi et al (1989), studying the effect of rinsing with sucrose solution on 4-day-old plaque, showed that the fall in pH in plaque was significantly greater on the approximal surfaces of the molars than in the fissures.

The ecological plaque hypothesis, introduced by Marsh (1991), proposes that a change in a key environmental factor (or factors) will trigger a shift in the balance of the resident plaque microflora, and this might predispose a site to disease. The occurrence of potentially pathogenic species as minor members of the resident plaque microflora would be consistent with this proposal. Under the conditions that prevail in health, these organisms would be only weakly competitive and might also be suppressed by intermicrobial antagonisms, thus constituting only a small percentage of the plaque microflora, without clinical effect. Microbial specificity in disease would result from the fact that only certain species are competitive under the new (changed) environmental conditions.

It is a basic tenet of microbial ecology that a major change in an ecosystem produces a corresponding disturbance in the stability of the resident microbial community (Alexander, 1971; Brock, 1966; Fletcher et al, 1987). An increasing mass of plaque impedes penetration by saliva to protect the enamel. Microbial homeostasis can break down, and major shifts in the composition of the microflora can occur.

For example, frequent consumption of fermentable dietary carbohydrates is associated with an increased risk of dental caries (Loesche, 1986). Such diets lead to a rise in the proportions of mutans streptococci (MS) and lactobacilli, with a concomitant fall in levels of other streptococci, especially members of the Streptococcus oralis group, which include S sanguis, S oralis, and S mitis (Dennis et al, 1975; de Stoppelaar et al, 1970; Minah et al, 1985; Staat et al, 1975). The metabolism of plaque also changes from a heterofermentative pattern to one in which sugars are converted primarily to lactic acid.

Studies by Bradshaw et al (1989a) have shown that low pH, rather than the availability of carbohydrates per se, is the factor driving the selection of potentially cariogenic species. This selection is at the expense of acid-sensitive species, some of which are associated more with oral health. The experiment was repeated to determine if there were a "critical pH" at which this breakdown in homeostasis would occur. Plaque microorganisms were pulsed with glucose in three replicate experiments in which the pH was allowed to fall only to fixed values of pH 5.5, 5.0, or 4.5. The microbial community was disrupted irreversibly only when the pH fell regularly below 5.0 (Bradshaw et al, 1989b).

The predominant species in these experiments always became Streptococcus mutans, Lactobacillus casei, and Veillonella dispar (Bradshaw et al, 1989b). These three species have been associated with nursing caries (Milnes and Bowden, 1985) and progressing caries (Boyar et al, 1989) in humans. Pure culture studies have also shown that the growth of these three species is less sensitive to low pH than is that of other oral bacteria (Bradshaw et al, 1989a; Harper and Loesche, 1986). Furthermore, mouthrinsing with acidic buffers (pH 3.9) was found to increase the proportions of mutans streptococci in human fissure plaque (Svanberg, 1980). Studies by van Houte et al (1991) have shown that streptococci other than mutans streptococci will increase the acidogenic potential of plaque at low pH. Collectively, these findings show that the selection of cariogenic species following regular sugar consumption is likely to be a consequence of their aciduric physiology, which enables them to compete successfully at low pH.

On the other hand, in subjects with a conventional low-sugar diet, the composition of plaque microflora would be stable, only small amounts of acid would be produced at main meals, and the processes of demineralization and remineralization would be in equilibrium. If the frequency of fermentable dietary carbohydrate intake were to increase, however, there would be longer periods of low plaque pH (Loesche, 1986a). Such conditions would favor the proliferation of mutans streptococci and lactobacilli at the expense of less acid-tolerant species, tipping the equilibrium toward demineralization (Fig 21).

Factors reducing the flow of saliva (eg, xerostomia) would lead to similar shifts in the microflora. Greater numbers of mutans streptococci and lactobacilli would lead to

even faster rates of acid production from sugars, enhancing demineralization still further, while the elevated levels of lactic acid in plaque would also select for Veillonella spp. Acid-sensitive species, such as members of the S oralis group (eg, S sanguis, S oralis, and S mitis), would decline in proportion, thereby accounting for the widely reported inverse relationship between S sanguis and mutans streptococci seen in plaque. Other bacteria could also produce significant amounts of acid under similar conditions, at slower rates (van Houte, 1993), but nevertheless providing an explanation for demineralization in the absence of mutans streptococci.

Fig 21 Ecological plaque hypothesis and prevention of caries. (Modified from Marsh, 1994.)

Strategies for prevention and control of caries based on plaque ecology hypothesis

According to the plaque ecology hypothesis, low pH (less than 5) will promote overgrowth of aciduric microorganisms, such as the cariogenic mutans streptococci and lactobacilli, at the expense of less acid-tolerant plaque microorganisms, such as S oralis, which are associated with healthy tooth surfaces.

Therefore the treatment strategy would be to increase plaque pH and thereby promote reestablishment of the harmless normal microflora of the tooth surfaces (see Fig 21). Increased pH can successfully be achieved by a combination of the following preventive measures:

1. Frequent mechanical removal of the plaque from all tooth surfaces: "Clean teeth never decay," and frequent removal of plaque (once or twice a day) limits the thickness of reaccumulated plaque, ensuring that saliva has accessibility for diluting and buffering the acid that is formed.

2. Reduction of sugar clearance time by reducing the intake of sticky sugar-containing products.

3. Use of sugarless chewing gums containing fluoride and chlorhexidine as a "dessert" for 15 to 20 minutes directly after every meal (including between-meal snacks).

Use of this kind of gum has several beneficial effects:

1. Salivary flow is increased and the acid formed during the meal is diluted and buffered.

2. Fluoride will reduce acid formation by the acidogenic microorganisms at low pH.

3. Chlorhexidine has not only a nonspecific antiplaque effect but also a specific effect on mutans streptococci and acid formation by acidogenic microorganisms.

4. Fluoride ions and minerals from the increased salivary flow will accelerate remineralization directly after the acid attack during the meal.

This recommendation is very important for caries-susceptible patients, particularly

those with xerostomia (for reviews on plaque ecology related to caries etiology, see Bowden, 1997; Bowden and Edwardsson, 1994; Marsh, 1993.)

Role of Specific Cariogenic Microflora

Introduction

Microorganisms implicated in the etiology of dental caries must be acidogenic as well as aciduric. To initiate carious lesions in enamel, the microorganisms must also be able to colonize the tooth surface and survive in competition with less harmful species, forming biofilmsthe so-called dental plaque. As early as 1960, Fitzgerald and Keyes showed that certain microorganisms isolated from human dental plaque, when inoculated in germ-free rodents on a high-sucrose diet, resulted in the spread of rampant caries. Therefore, dental caries should be regarded as an infectious, transmissible disease.

There is abundant support for the so-called specific plaque hypothesis, introduced by Loesche (1982, 1986), which proposes that some specific species of the plaque flora be regarded as major pathogens in the etiology of dental caries. Included in the major pathogens are those bacteria associated with caries in humans and also able to induce carious lesions in experimental animals. The most important are the mutans streptococci: there are seven species, of which two, S mutans and S sobrinus, are closely associated with caries in humans. The remainder are found in animals or, if present in humans, are not highly cariogenic. The relationship of S sobrinus to caries in humans is not as well understood as that between S mutans and caries, and only recent studies identify the species separately.

The second genus closely associated with caries is Lactobacillus, commonly isolated from carious dentin (Edwardsson, 1974), thought to be its main habitat in the mouth. Compared to the extensive research into identification of the mutans streptococci, much less attention has been paid to speciating lactobacilli isolated from carious lesions.

Also associated with the etiology of dental caries, but considered to be less cariogenic than S mutans, S sobrinus, and Lactobacillus, are Actinomyces odontologica, Actinomyces naeslundii, and some other species of MS. To clarify the role of S mutans in the etiology of dental caries, many cross-sectional and longitudinal human studies have been conducted over the past two to three decades, particularly by the Krasse and Bratthall research groups in Sweden.

Cariogenicity of mutans streptococci

Mutans streptococci are acidogenic as well as aciduric and can adhere to tooth surfaces (Gibbons et al, 1986). Mutans streptococci can produce extracellular and intracellular polysaccharides from sucrose. Intracellular polysaccharides in particular can be degraded during periods of low nutrient supply, indicating that these polysaccharides increase the virulence of some MS species (S mutans and S sobrinus). Because the microbial ecology of the mouth is highly complex, strains of the same species could vary considerably in virulence (Bowden and Edwardsson, 1994). In other words, MS fulfill all the requirements of caries-inducing bacteria.

Colonization of the teeth by mutans streptococci is highly localized; some tooth surfaces are colonized but not others. The amount of mutans streptococci in saliva is related to the number of colonized surfaces (Lindquist et al, 1989). This is the basis for saliva tests for MS. A high count in saliva (more than 1 million colony-forming units [CFUs] per 1 mL of saliva) indicates that most teeth are colonized by these bacteria, ie, that many tooth surfaces are subject to increased caries risk. However, a salivary MS count does not provide information about the origin of the bacteria, ie, the specific tooth surfaces which are colonized.

The most common types of mutans streptococci, Streptococcus mutans (serotypes c, e, and f) and Streptococcus sobrinus (serotypes d and g), are present worldwide. Their prevalence differs among populations. Test values also differ, depending on the method of detection (Axelsson et al, 1987b; Bratthall et al, 1986; Beighton et al, 1989; Buischi et al, 1989; for reviews, see Bratthall, 1991; Carlsson, 1988). About 10% to 30% of a population may have little or no MS, 0 to 100,000 CFUs/mL of saliva. The percentage of individuals with very high levels of MS ( 1 million CFUs/mL of saliva) in a population may vary considerably, depending on age, caries prevalence, dietary habits, and so on.

Evidence

Several cross-sectional studies in human populations with relatively high caries prevalence have shown a correlation between very high salivary MS levels and very high caries prevalence (Axelsson et al, 1999b; Buischi et al, 1989; Klock and Krasse, 1977; Salonen et al, 1990; Zickert et al, 1982). This is exemplified in Fig 22, from a study of 12-year-old Brazilian children (Buischi et al, 1989). However, in populations with relatively low caries prevalence and high standards of oral hygiene, the threshold value of more than 1 million CFUs of MS/mL of saliva no longer seems to apply, as exemplified in Fig 23, from a study of 13- to 14-year-old schoolchildren in Karlstad, Sweden (Kristoffersson et al, 1986). In this population, the critical difference was between MS-negative and MS-positive subjects.

As also shown in Fig 23, it was impossible to find any correlation between intake of sticky sugar products (estimated point scale) and caries prevalence, highlighting the multifactorial nature of dental caries: The lower the prevalence and incidence of caries in a population, the more difficult it is to demonstrate a significant correlation for one single etiologic or modifying factor.

Transmission. Because MS require a hard, nondesquamating surface for colonization (Berkowitz et al, 1975; Carlsson et al, 1975; Catalanotto et al, 1975; Stiles et al, 1976), infants do not harbor MS until some time after tooth eruption: The major source of the infection is thought to be maternal. Evidence for this comes from several studies showing that isolates of MS harbored by mothers and their children exhibit similar or identical bacteriocin profiles (Berkowitz and Jordan, 1975; Berkowitz and Jones, 1985; Davey and Rogers, 1984) and identical plasmid or chromosomal DNA patterns (Caufield et al, 1985, 1986, 1988; Caufield and Walker, 1989; Hagan et al, 1989; Kulkarni et al, 1989).

Several studies have suggested that the extent of MS colonization and, to some

degree, subsequent carious activity experienced by a child may be correlated with the mother's salivary level of MS: Mothers with high levels of MS tend to have children with high levels and vice versa (Caufield et al, 1988; Kohler et al, 1984; Kohler and Bratthall, 1978; van Houte et al, 1981). While correlations between caries or MS levels in mothers and those in their children may be explained in part by common genetic or environmental factors, others have suggested that a child's degree of colonization or disease may be dictated by the mother's levels of MS at the time of transmission. In a landmark study, Kohler and coworkers (1983, 1984) selected mothers with initially high levels of MS in saliva and determined the effects of various preventive and treatment regimens aimed at reducing MS below a predetermined threshold level. Children of these mothers were monitored for initial acquisition of MS and, subsequently, for carious activity over a 3-year period. A statistically significant difference was observed between control and experimental groups in terms of when a child acquired MS, the levels of MS harbored by the mother and child, and the child's caries outcome. Figure 24, from the longitudinal study by Kohler et al (1988), illustrates that the earlier the colonization by MS, the higher the caries prevalence at 4 years of age.

In a recent study by Caufield et al (1993), oral bacterial levels of 46 mother-child pairs were monitored from the birth of the child to 5 years of age to study the acquisition of MS by the children. In 38 children, initial acquisition occurred at the median age of 26 months, during a discrete period that was designated as the "window of infectivity." In the remaining eight children (17%), MS was undetectable throughout the study (median age 56 months).

No significant differences were found in salivary levels of MS or lactobacilli of mothers of children with and without MS. Comparisons between a caries-active cohort colonized by MS (9 of 38) and children without detectable MS revealed similar histories in terms of antibiotic usage, gestational age, and birth weight. Interestingly, half the children who were MS negative between the ages of 1 and 2 years were minded by caregivers other than the mother, while all the children who were caries active during this age interval were cared for by their mothers; the difference was statistically significant. This study by Caufield et al (1993) was the first to present evidence that MS is acquired during a defined period in the ontogeny of a child. Support for the notion of a discrete window of infectivity comes from other sources, including animal models. By studying mother-child and father-child pairs, Alaluusua et al (1991) found a strong correlation between teenagers and mothers with high numbers of decayed, missing, or filled surfaces and high salivary MS levels, but no such correlation in father-child pairs.

The above studies in humans confirm the earlier animal studies by Fitzgerald and Keyes (1960) that dental caries is an infectious disease, transmissible by MS. Several experimental and clinical studies have also confirmed that MS can be isolated from dental plaque covering active carious lesions in enamel (Axelsson et al, 1987; Kristoffersson et al, 1985) and at the root (van Houte et al, 1990) as well as secondary carious lesions (Gonzales et al, 1995) and the margins of restorations (Wallman and Krasse, 1992; for reviews, see Bowden and Edwardsson, 1994; Loesche, 1986a).

Caries incidence. Many longitudinal human clinical studies have shown correlations between high salivary MS counts and high caries incidence. In preschool-aged

children (primary dentition), correlations between salivary MS counts and caries incidence have been shown by Alaluusua et al, 1990; Kohler et al, 1988; Roeters et al, 1995; Thibodeau and O'Sullivan, 1996; and Twetman et al, 1996. In the permanent dentition, Klock and Krasse (1978, 1979) showed that 9 to 12 year olds with high salivary MS levels developed significantly more new carious surfaces than did children with low levels of MS during a 2-year period. However, when a test group of children with more than 1 million CFUs of MS/mL of saliva received high-quality plaque control by frequent professional mechanical toothcleaning, they developed fewer new carious surfaces than did the control groups with high and low salivary MS levels (0.9 new carious surfaces versus 2.2 and 4.3 new carious surfaces, respectively). In Molndal, Sweden, Zickert et al (1982b) also attained a significant correlation between the prevalence of MS in saliva and the incidence of new carious lesions. During a 3-year period, children with high levels of salivary MS ( 106 CFUs/mL) developed about three times as many new carious lesions as did control groups with lower levels of MS. Subjects in test groups on a treatment program including chlorhexidine developed significantly fewer cavities.

In US adolescents, Kingman et al (1988a) also showed that subjects with high salivary MS levels developed more new carious surfaces than did subjects with lower MS levels.

In particular, controlled intraindividual longitudinal studies monitoring the microflora at the tooth surface level have clarified the cariogenic potential of MS (Axelsson et al, 1987b; Kristoffersson et al, 1985; MacPherson et al, 1990). An advantage of such studies is that several other external and internal modifying factors such as diet, fluoride, and saliva are equal for test and control sites. These studies have clearly shown that, in the same mouth, a tooth surface colonized by MS is at greater risk for caries than is a similar surface without MS.

During a 30-month period, S mutans on all the approximal surfaces was studied in a selected group of 13 year olds with more than 1 million CFUs of MS/mL of saliva. From a population of 720 13 year olds, subjects with more than 1 million CFUs/mL were selected (n = 187). Every 6 months, S mutans was sampled from saliva, the dorsum of the tongue, and every approximal tooth surface. Interproximal samples were obtained with a sterile wooden toothpick, as described by Kristoffersson and Bratthall (1982) (Fig 25). The contaminated sides of the toothpick were then pressed directly against selective (mitis-salivarius-bacitracin) agar plates (Fig 26). After incubation, the number of colonies formed (CFUs) was evaluated for every approximal surface.

In 17 subjects who consistently had a minimum of one surface highly colonized with MS and a minimum of one MS-negative or lightly colonized surface, about 60% of the highly colonized surfaces developed carious lesions (Fig 27, left), but only 3% of the MS negative or sparsely colonized surfaces did (Fig 27, right) (Axelsson et al, 1987b).

A prior study showed the surfaces most heavily colonized with MS to be the approximal surfaces of the molars and the second premolars (Fig 28) (Kristoffersson et al, 1984). In fact, a previously mentioned study of more than 600 14 year olds (Axelsson, 1989, 1991) showed that the same surfaces also had the highest PFRI

scores. These observations explain why, in toothbrushing populations, these approximal surfaces have the highest prevalence of decayed, missing, or filled surfaces (Fig 29). For optimal caries prevention in such populations, plaque control and topical application of fluorides should target these key-risk surfaces.

Methods of sampling

As mentioned earlier, the correlation between salivary MS counts and the number of MS-colonized tooth surfaces is relatively good (Lindquist et al, 1989), and simple salivary sampling methods are a more convenient and realistic means of assessing the severity of MS infection than sampling from individual tooth surfaces.

Laboratory methods. Saliva is collected, mixed with a proper transport medium, and forwarded to a microbiologic laboratory. After incubation using a selective medium, mutans colonies are counted and the results are expressed as the number of colony-forming units per milliliter of saliva.

Several selective media are available, and their properties are not identical: This fact must be taken into consideration in assessing the results. A common type of selective medium for plating mutans streptococci is the mitis-salivarius-bacitracin agar (Gold et al, 1973). With the exception of the rare serotype a, all types of mutans streptococci grow on this medium. Bacitracin is the main selective ingredient. Because the plates have a shelf life of only about 1 week, they are not convenient for chairside tests.

For screening surveys using agar plates, the following simplified method has been described, eliminating the need for transportation, dilution, and plating of saliva (Kohler and Bratthall, 1979; Newbrun et al, 1984) (Fig 30): Wooden spatulas are contaminated with saliva and immediately pressed against selective agar plates. After incubation, the number of colonies on a predetermined area of the plate is calculated.

Chairside method. The so-called Strip Mutans test (Fig 31), described by Jensen and Bratthall (1989), is based on the ability of mutans streptococci to grow on hard surfaces and the use of a selective broth (high sucrose concentration in combination with bacitracin). Because the bacitracin can be added to the broth just before use, the shelf life of the test can be prolonged considerably compared to that of agar plates.

The test set is used as follows: A bacitracin disc is taken from the vial with forceps or a needle. The cap is reclosed tightly. The bacitracin disc is put in the culture broth vial and allowed to stand for at least 15 minutes (Fig 32). The vial is shaken gently after 15 minutes. When more than one Strip Mutans test is to be run, the bacitracin discs can be added to the vials beforehand. However, only one Strip Mutans test can be performed in each vial, and vials to which bacitracin has been added must be used on the same day.

The patient is given a Dentocult paraffin pellet and instructed to chew it for 1 minute. The stimulated saliva should be swallowed or spat out (Fig 33).

A test strip is removed from the Strip Mutans container, so that only the square end is touched. About two-thirds of the strip is placed in the patient's mouth and rotated on the surface of the tongue about 10 times. The strip is removed from the mouth, pulled

between the patient's closed lips to remove any excess saliva (Fig 34). The strip is placed in the culture vial containing the well-mixed bacitracin-broth solution. The cap is reclosed tightly (Fig 35). The patient data is written on the label, and the label is attached to the vial. The strip is incubated for 2 days at 35C to 37C (95F to 99F).

The strip is removed from the culture vial and allowed to air dry. The treated side, which is marked with a line, can be examined immediately or later (Fig 36). After it is dried, the test strip can be stored for future reference in a plastic bag, plastic wrap, autoclave plastic, or other similar material.

The bacitracin and the higher sugar content inhibit the growth of practically all microorganisms, other than S mutans, that grow in mitis-salivarius medium. In proportion to their actual amount in saliva, S mutans in the specimen will adhere to the treated side of the strip, and grow as small, dark or light blue colonies, 1 mm in diameter, or considerably less, when growth is very dense. The amount of S mutans per milliliter of saliva is estimated by comparing the colony density on the strip with the standard charts included in the instructions.

If the number of S mutans is very high, the treated side of the test strip will turn blue, and separate colonies will be indistinguishable. A test strip without S mutans may have a blue shade as a result of precipitation of the color indicator present in the culture medium. A magnifying glass or a microscope should be used to verify questionable cases.

The Strip Mutans method should not be used within 12 hours of an antibacterial mouthwash (eg, chlorhexidine) or 2 weeks of a course of antibiotics to avoid false-negative results.

Fig 22 A study of 12-year-old Brazilian children shows a correlation between very high salivary MS levels and very high caries prevalence. (From Buischi et al, 1989.)

Fig 23 A study of 13- to 14-year-old schoolchildren in Karlstad with good oral hygiene and low caries prevalence shows no correlation between caries prevalence and different dietary scores and different levels of salivary S. mutan. The cutoff is S mutans negative or positive. (From Kristoffersson et al, 1986.)

Fig 24 Dental caries in relation to colonization of MS. (From Kohler et al, 1988.)

Fig 25 Interproximal samples of S mutans are obtained with a sterile wooden toothpick.

Fig 26 Slides containing S mutans are pressed on agar plates.

Fig 27 Development of carious lesions on surfaces highly colonized (left) and sparsely colonized (right) with MS. (From Axelsson et al, 1987b.)

Fig 28 The approximal surfaces of molars and second premolars have been shown to be the most highly colonized with MS according to approximal MS scores 0-3. (From Kristoffersson et al, 1984.)

Fig 29 Caries prevalence in 12 year olds in the county of Vormland, Sweden, 1964-1994. (From Axelsson, 1998.)

Fig 30 Wooden spatula contaminated with saliva and pressed on an agar substrate (left). Typical S. mutans colonies on the surface of the agar substrate (right). Fig 31 Strip Mutans test. (Courtesy D. Brathall.)

Fig 32 Placement of a bacitricin disc in the culture broth vial in which Strip Mutans test is being performed. (Courtesy D. Brathall.)

Fig 33 Dentocult paraffin being chewed to stimulate saliva. Chewing friction loosens S mutans from the tooth surface. (Courtesy D. Brathall.)

Fig 34 The test strip is removed from the patient's mouth and placed in the vial containing bacitracin solution. (Courtesy D. Brathall.)

Fig 35 The test strip is removed from the patient's mouth and placed in the vial containing bacitracin solution. (Courtesy D. Brathall.)

Fig 36 Examination of test strip. (Courtesy D. Brathall.)

Cariogenicity of lactobacilli

According to the specific plaque hypothesis, some strains of lactobacilli are considered to be major caries pathogens along with S mutans and S sobrinus. Lactobacilli are acidogenic and even more aciduric than MS. Mutans streptococci are strongly correlated to the etiology of initial enamel and root surface lesions, because they can adhere to and colonize the tooth surfaces. Lactobacilli are more dependent on retentive sites for heavy colonization: Mutans streptococci are regarded as the pioneers, followed by lactobacilli in the succession toward more cariogenic plaque. This has been shown in a study on the development of so-called nursing caries (Milnes and Bowden, 1985) and by Mac Pherson et al (1990) in another study on plaque flora associated with early enamel demineralization. Lactobacilli are most often found in the deepest part of the lesion (dentin), an environment with prolonged periods of very low pH.

Lactobacilli are highly influenced by the dietary carbohydrate content and intake frequency, in addition to reflecting an acidogenic environment by their very presence, because they are so aciduric. They also indicate the presence of substrate for other bacteria, such as mutans streptococci. Persistently high levels of lactobacilli after elimination of retention sites such as cavitated carious lesions indicate a diet rich in carbohydrates.

Evidence

Lactobacillus counts have been used to predict the incidence of new carious lesions. Crossner (1981) studied a group of children, who had been given dental treatment at baseline so that no open lesions were present at the bacterial sampling. Two subgroups in this material are of special interest: those with very low or very high lactobacillus counts. Very few individuals in the low lactobacillus group developed new carious lesions over a 64-week period. In the high lactobacillus group, many, but not all, developed new lesions.

Some studies have shown that caries incidence is significantly increased when MS and lactobacilli occur in the same individual: During a 3-year longitudinal study, Alaluusua et al (1990) showed that teenagers with high salivary values for both MS and lactobacilli developed several times more new carious surfaces than did those with lower counts of either. The percentages of children who developed 0, 1 to 3, and more than 3 new carious surfaces were correlated to the total scores of MS and lactobacilli (Fig 37). Similar findings were reported by Stecksen-Blicks et al (1985) (Fig 38).

In a 2-year study by Crossner et al (1989), samples from saliva, the tongue, and 276 interdental spaces were obtained from 23 7 year olds to (1) relate the presence of lactobacilli at various oral sites to the occurrence of lactobacilli in saliva, and (2) relate the presence of mutans streptococci and various types of lactobacilli interdentally to the development of proximal carious lesions. The results showed that the number of interdental samples containing lactobacilli increased as the number of salivary lactobacilli increased. Furthermore, lactobacilli were never found interdentally without the presence of mutans streptococci, and lactobacilli proved to be the more suitable microorganism for prediction of proximal carious lesions. Neither the number nor the differentiation into different species of interdental lactobacilli seemed to be of importance, merely their presence or absence. The presence of lactobacilli probably reflects a caries-inducing environment (etiologic microflora plus fermentable carbohydrates), thus explaining their high predictive ability despite their rather limited etiologic importance in the initiation of caries.

Lactobacilli are also implicated in root surface caries. Fure and Zickert (1990a) studied a group of 208 randomly selected 55, 65, and 75 year olds. To estimate the number of root caries lesions, an index, DFS%(R), was used, indicating the number of decayed and filled root surfaces as a percentage of the exposed root surfaces. The mean DFS%(R) was 13 for the 100 subjects with low lactobacillus counts (less than104), and 23 for the 52 subjects with high counts (more than 105). In cross-sectional and longitudinal studies, it has been demonstrated that lactobacillus counts are among the factors with a significant correlation to the development of root caries lesions (Ravald and Birkhed, 1991, 1994; van Houte et al, 1990). Ellen et al (1985) also showed that root surfaces harboring both MS and lactobacilli are most likely to develop root caries.

A number of studies have tried to clarify the prevalence of lactobacilli in various populations, eg, two Swedish studies, one comprising 646 9 to 12 year olds (Klock and Krasse, 1977) and the other a group of 101 13 to 14 year olds (Zickert et al, 1982b). About 50% showed low salivary counts (less than 10,000 CFUs/mL of saliva)

and 10% to 20% had high counts (100,000 to 1 million CFUs/mL of saliva). Generally the MS counts were 10 times greater than the lactobacillus counts. Most children with low mutans streptococci levels also had low numbers of lactobacilli, but there were situations where one type of bacterium was high and the other low.

In a dentate Swedish population of 80 and 85 year olds, Kohler and Persson (1991) found that 95% of the subjects had detectable salivary lactobacilli counts and 35% very high levels (more than 100,000 CFUs/mL of saliva). Almost 90% were MS positive, and 30% had more than 1 million CFUs/mL of saliva.

Methods of sampling

Laboratory method. The standard method for determining the number of lactobacilli includes the use of a selective medium, Rogosa selective Lactobacillus (SL) agar. Saliva obtained by chewing a piece of paraffin is shaken with glass beads to break up aggregates of bacteria. The saliva is then mixed with a buffer solution, and 1 mL of the dilutions 10-2 and 10-3 is mixed with 10 mL of melted SL agar. Another 10 mL is then poured into the Petri dish, and the plates are incubated at 37C for 2 days. The plates are then ready for a colony count.

Chairside method. A chairside dip-slide method (Dentocult LB) is also available for simplified evaluation of salivary lactobacilli counts (Fig 39). After aerobic incubation for 4 days at 37C, the number of lactobacilli is estimated by comparing the slides with a chart supplied by the manufacturer. An advantage of this method is that the results can be shown directly to the patient.

Figure 40 shows examples of low (less than 10,000 CFUs/mL of saliva) and high salivary lactobacillus counts. The high levels indicate a cariogenic enviroment (low pH) in the oral cavity. The number of lactobacilli in saliva seems to be fairly stable during the daytime. Significantly higher levels are often obtained if saliva is collected in the early morning, before breakfast and toothbrushing, than if samples are obtained during the rest of the day, especially for subjects with high numbers of lactobacilli.

Fig 37 Distribution of caries increment in children with different combined levels of mutans streptococci and lactobacilli scores. Score systems based on Dentocult SM for mutans streptococci and Dentocult LB for lactobacilli, both systems ranging from score 0 (low levels of bacteria) to score 3. (From Alaluusua et al, 1990. Reprinted with permission.) Fig 38 Lactobacilli and S mutans among 13-year-old children with different net caries increment. (From Stecksen-Blicks, 1985. Reprinted with permission.)

Fig 39 Dip-slide used for simple evaluation of salivary lactobacilli counts.

Fig 40 Examples of low (left) and high (right) salivary lactobaccilli counts using the Dentocult method.

Cariogenicity of other bacteria

There are overwhelming data from experimental and clinical studies in humans showing that S mutans and S sobrinus and lactobacilli are strongly correlated to caries etiology. However, the use of selective substrates in most of these studies may have introduced some bias. For example, Sansone et al (1993) found that plaque samples with and without MS and lactobacilli were equally acidogenic when cultured at low pH and in the presence of excess glucose. Borgstrom et al (1997) evaluated the pH-lowering effect of plaque from carious lesions in enamel as a virulent variable and found a relatively weak correlation among lactobacilli, MS, and dental caries.

Root caries is considered to develop at higher pH than enamel caries. Normally, demineralization precedes the breakdown of the organic part (about 40% by volume) of the root surface. Schupbach et al (1995), using a new sampling technique and unselective, anaerobic culturing methods, evaluated the composition of the plaque microbiota covering sound root surfaces, actively carious root surfaces, or arrested lesions.

On all surfaces Actinomyces spp predominated, and streptococci and lactobacilli formed a minor part (less than 1%) of the microbiota. With respect to the detected proportions of anaerobes, microaerophiles, Actinomyces naeslundii, Prevotella buccae, and Selenomonas dianae, significant differences were observed among the three categories of root surfaces. The total numbers of CFUs were significantly higher on both caries-free and caries-active surfaces than on arrested lesions. In general, the results supported a polymicrobial etiology for caries initiation on root surfaces, during which A naeslundii, Capnocytophaga spp, and Prevotella spp make specific contributions to the processes of cementum and dentin breakdown. In other words, the roles of bacteria-producing proteolytic enzymes and of acidogenic bacteria other than MS and lactobacilli in the development of root surface caries warrant further investigation (for reviews on specific microflora related to the etiology of dental caries, see Bowden, 1997; Bowden and Edwardsson, 1994; Bratthall and Ericsson, 1994; Emilsson and Krasse, 1985; Loesche, 1986).

Prediction of Caries Risk

Principles of risk prediction

Some basic principles have to be followed for successful and cost-effective caries prediction, caries prevention, and caries control:

1. The higher the risk of developing caries for most of the population, the more significant the effects of one single preventive measure and the stronger the correlations between one single etiologic or modifying risk factor and the risk for caries development.

2. In populations in which only a minority of the people will develop new carious lesions, it is necessary to use accurate risk predictive measures to select at-risk individuals and introduce needs-related combinations of caries-preventive measures, in other words, a "high-risk strategy."

In most child populations today, caries incidence is skewed. The majority of the children in most age groups develop no or very few new carious surfaces, while a small minority, 5% to 10%, develop several new carious surfaces each year. Therefore, accurate caries risk predictors are useful. However, the dilemma is that no method will guarantee that 100% of the selected high-risk individuals are "true" high-risk individuals. The percentage of "true" high-risk individuals among the selected group of high-risk individuals is termed sensitivity of the risk predictive method. Similarly, methods are used to select nonrisk or low-risk individuals. The percentage of "true" nonrisk individuals is termed specificity of the predictive method used.

These principles may be exemplified in Fig 41 (a to c) showing the outline of a typical study with the aim of evaluating the predictive power of a risk marker of dental caries. In the beginning, the baseline caries status and the level of the selected risk marker are assessed. Caries recordings at the end of the follow-up period make it possible to assess the true caries incidence during the period.

Prediction studies deal with two dichotomies: (1) individuals for whom it is believed that the risk is high or low, and (2) the ones for whom true high or low caries incidence is observed. Thus group a in Fig 41 (a) consists of correctly classified individuals, true-positives, for whom it was believed that the risk was high and whose actual caries incidence was high. Correspondingly, group d represents correctly classified true-negatives. For individuals falling into groups b and c, misclassification has occurred. For the false positives, in group "b," a high risk was assumed, but the true caries incidence was low. Correspondingly the false-negatives in group c were believed to have a low risk, but their actual caries incidence was high.

This design is only usable for one predictor at a time. In practice, several predictors are often regarded in prediction studies. In the case of multiple predictors, each of them can be considered separately, which leads to predictor-specific numbers of true- or false-positives and true- or false-negatives. Alternatively, the information of many predictors can be condensed into a single variable on the level of which the prediction of high or low risk is based. The techniques for such condensing range from simple summaries to sophisticated regression-based multivariable models.

Even in the case of one predictor, the risk markers are seldom natural dichotomies. To generate the four groups of interesttrue- and false-positives, and true- and false-negativesit is necessary to artificially dichotomize both multi- ple-level predictors and the outcome, true caries incidence, which, in the data collection, is usually regarded as the number of new carious tooth surfaces, not as a dichotomy.

The dichotomization can be done in different ways. Each threshold level for believed high risk and for observed high true caries incidence leads to a different distribution of the study subjects into the four groups of interest (true- and false-positives, and true- and false-negatives). Thus, when the results of a prediction study are evaluated, it is of utmost importance to consider the threshold levels that have been used. To

estimate the accuracy of the classification of the four groups in Fig 41 (a), the quantities a, b, c, and d are organized in the form of a 2 x 2 contingency table (Fig 41 (b)).

Six different measures of accuracy and their estimators are given in Fig 41 (c). As mentioned earlier, sensitivity is the proportion of those who were believed to have a high risk among the individuals whose actual caries incidence during the follow-up was high. Specificity is the proportion of those who were believed to have a low risk among the patients whose actual caries incidence during the follow-up was low.

False-positive rate and false-negative rate carry exactly the same information as sensitivity and specificity but, in contrast, reveal proportions of misclassified subjects. False-positive rate is the proportion of those who were believed to have a high risk among the subjects whose actual caries incidence during the follow-up was low. False-negative rate is the proportion of subjects who were believed to have a low risk among those individuals whose actual caries incidence was high.

Positive predictive value is the proportion of those whose actual caries incidence was high among the subjects who were believed to have a high risk. Negative predictive value is the proportion of subjects whose actual caries incidence was low among the patients for whom a low risk was predicted.

All six of these measures should always be examined pairwise. For instance, sensitivity has no meaning if the specificity is not known. This is important from a cost-effectiveness point of view during screening of populations of children with relatively low caries prevalence and incidence before needs-related caries preventive programs are designed. However, for the individual patient the consequences of being a false nonrisk or low-risk individual are very different from those of being selected as a false-positive high-risk individual.

Fig 41 (a and b) A study of the evaluation of the predictive power of a risk marker of dental caries. (c) Selected measures for evaluating the accuracy of predictions. N = study cohort; BHR = believed to have a high risk; BLR = believed to have a low risk; TCI = true caries incidence; Group a = believed to have a high caries risk, actual high caries incidence (true-positive); Group b = believed to have high caries risk; actual low caries incidence (false-positives); Group c = believed to have low caries risk; actual high caries incidence (false-negatives); Group d = believed to have low caries risk; actual low caries incidence (true-negatives). (Modified from Hausen, 1997. Reprinted with permission.)

Accuracy of risk assessments in practice

A perfect risk marker would have a sensitivity of 100% and a specificity of 100%, implying no errors in risk assessment. Consequently, the false-positive and false-negative rates would be 0%, and positive and negative predictive values would be 100%. Having perfect accuracy means that the predicted high-risk group would consist of only true high-risk individuals and that only true low-risk individuals would be included in the predicted low-risk group. Unfortunately, no such marker is available for the assessment of caries risk. A certain proportion of errors have to be

accepted. However, there are no generally accepted rules of what the acceptable level of error might be.

It has been suggested that, in a risk model, the sum of sensitivity and specificity be at least 160% before a caries risk marker can be considered a legitimate candidate for targeting individualized prevention (Kingman, 1990). This is in agreement with an alternative suggestion that a sensitivity and specificity of 80% would be acceptable for practical use in the community. Although neither of these suggestions takes into account the fact that errors related to poor sensitivity do have consequences that are different from those related to poor specificity, both proposals can be used as a starting point for evaluating the performance of proposed markers for high caries risk.

What would a combined sensitivity and specificity of 160% mean in practice? If both the sensitivity and specificity were 80%, every fifth individual with a true high risk would remain undetected in a risk assessment and thus fail to receive the intensified protection against caries that he or she needs. Correspondingly, every fifth individual with a true low risk would erroneously be included in the high-risk group and receive preventive measures to no or little purpose. Thus, even the proposed minimum acceptable level of accuracy would result in an uninvitingly high rate of misclassifications.

If the proportion of caries-risk individuals in a population is close to half or more, this clearly implies that the occurrence of caries is not low enough to justify the effort and expense of identifying key-risk individuals. In such a situation, the preventive efforts should be targeted to the whole population.

The proportion of the target population that can be given individual protection against further caries development naturally varies from one setting to another. In most cases, risk groups of a size exceeding 30% seem to be unworkable. In a thorough review by Hausen et al (1994), an effort was made to compare the predictive power of risk markers in a situation where the aim was to select the 30% of the target population with the highest risk of developing new lesions. For none of the markers aimed at assessing the risk for coronal caries did the predictive power reach the proposed combined sensitivity and specificity of 160% (Kingman, 1990). This level was surpassed in one study only, where a combination of several predictors had been used for assessing the risk of root caries (Scheinin et al, 1992).

The difficulty of predicting caries is not unexpected. The multifactorial etiological and modifying factors of dental caries make it likely that even the most sophisticated risk models will be of limited value in predicting future caries development very accurately. Furthermore, even a perfect marker is capable of predicting a person 's future caries experience only if the conditions on which the prediction is based remain stable. In most industrialized countries, where virtually all the prediction studies have been conducted, the populations are exposed to a variety of professional prevention and treatment regimens as well as self-care, which, if applied selectively, most probably reduce the observed power of such studies. The living conditions and oral health behaviors may change over time, thus modifying a person's caries risk in either direction. In addition, the rational and ethical consequence of risk prediction in clinical practice is to introduce needs-related measures for caries prevention and caries control. The optimal outcome therefore should be no new carious lesions. For

these reasons, it is not likely that, even in the future, caries risk can be assessed accurately by using one single risk marker.

Past caries experience (caries prevalencethe number of decayed, missing, or filled teeth and surfacesand caries incidencethe number of new carious teeth and surfaces in a year) has so far been the most powerful single predictor for future caries incidence, at least in children and young adults. That is because carious lesions represent the sum result of all the etiologic and modifying risk factors to which the individual has been exposed.

For example, in a recent 3-year longitudinal study, Bjarnason and Kohler (1997) achieved 89% sensitivity value in a group of Swedish adolescents by comparing the prevalence of non-cavitated enamel caries and DFS at the baseline as predictors. Together, the sensitivity and specificity values reached 160% or more. High salivary MS and lactobacilli scores resulted in 71% sensitivity and 75% specificity, respectively (cutoff level for high caries risk was 5 or more new carious surfaces in 3 years). However, only baseline values of incipient enamel caries were significantly correlated to the caries incidence.

The use of past caries experience as an indicator of future incidence has justly been criticized by the argument that the aim should be to determine the high-risk individuals before there are any signs of past caries experience. In other words, efforts should focus on primary prevention instead of secondary prevention. In particular this is important in infants and children with erupting permanent teeth. Wendt et al (1994) found that the caries incidence in infants and toddlers aged 1 to 3 years was strongly correlated to the plaque scores and oral hygiene regimens even at 1 year of age.

Selection of caries-risk patients

Inability of a sole salivary MS test to predict caries risk

As already mentioned in this chapter, numerous cross-sectional as well as longitudinal studies have shown significant correlations between salivary MS levels and caries prevalence and caries incidence (for review, see Bratthall, 1991; Bratthall and Ericsson, 1994; Beighton et al, 1989). At the surface level, even more significant correlations between MS colonization and caries incidence have been found (Axelsson et al, 1987b; Kristoffersson et al, 1985).

Most of the early salivary MS studies were carried out in child populations with relatively high caries prevalence (Sweden in the 1970s), and at that time more than 1 million CFUs of MS/mL of saliva was shown to be a good predictor of caries risk (Klock and Krasse, 1977; Zickert et al, 1982). However, since then, caries prevalence in Sweden and many other industrialized countries has decreased significantly. The correlation between one single etiologic factor, such as salivary MS levels, and caries prevalence and caries incidence tends to be weaker in such populations, because dental caries is a multifactorial disease.

In a more recent 2-year longitudinal Swedish study in children (5 to 7 years and 12 to 14 years), Sullivan et al (1989) found that the correlation between caries incidence and both salivary MS and lactobacilli was weak at the individual level, particularly

after correction for confounding factors, such as oral hygiene status. In another study, Sullivan et al (1996) found that MS and lactobacilli, whether in saliva or in plaque, was not a powerful enough tool for caries prediction in a group of 14 to to 15 year olds. Kingman et al (1988a) also found that the predictive values for salivary MS and lactobacilli on caries incidence in 10- to 15-year-old US schoolchildren was low (31% and 39%, respectively). The moderate-to-low predictive value of salivary MS may partly be explained by differences in virulence, not only among species of MS but also among individual clones of S mutans and S sobrinus (Bowden, 1997). Even in relation to root caries, the important role of MS and lactobacilli has recently been questioned (Schupbach et al, 1995; for recent reviews on the importance of the specific microflora for prediction of caries risk, see Bowden, 1997; Bowden and Edwardsson, 1994; Bratthall and Ericsson, 1994; Hausen et al, 1994; Hausen, 1997.

More recent cross-sectional studies in Swedish schoolchildren have repeatedly found that the cutoff for correlation between salivary MS counts and caries prevalence is MS negative or MS positive rather than 1 million CFUs of MS/mL of saliva (see Fig 23) (Kristoffersson et al, 1986). However, the dilemma is that only 10% to 30% of the individuals are MS negative in most populations (higher in young children and lower in elderly). The question is how to select 5% to 25% high- and very high-caries risk individuals from among the 70% to 90% MS-positive subjects.

Rationale for combining salivary MS tests and PFRI for prediction of caries risk

Like the inflammation induced in gingival soft tissues adjacent to dental plaque, carious lesions that develop on the individual enamel surface beneath bacterial plaque should be regarded as the net result of an extraordinarily complex interplay between harmless and harmful bacteria, antagonistic and synergistic bacterial species, their metabolic products, and their interaction with the many other external (fermentable carbohydrates etc) and internal (saliva and other host factors) modifying factors, which are discussed in more detail in chapters 2 and 3. In other words, enamel carious lesions develop only on the specific tooth surfaces where thick plaque with a high percentage of acidogenic and aciduric bacteria remains too longits "acid slag products" demineralize the underlying tooth surface.

As discussed earlier in the chapter, the quantity of plaque that forms on clean tooth surfaces during a given time represents the net result of interactions among etiologic factors, many internal and external risk indicators and risk factors, and protective factors. This observation was the rationale for the development of the Plaque Formation Rate Index by Axelsson (1984, 1989, 1991). The index, based on the amount of plaque freely accumulated (de novo) 24 hours after PMTC, is described in more detail earlier in the chapter.

An earlier 30-month longitudinal study showed a very strong correlation between development of approximal carious lesions and the level of MS colonization (Axelsson et al, 1987). Other studies have shown that salivary MS counts are correlated to the number of tooth surfaces that are colonized by MS (Lindquist et al, 1989; for review, see Bowden, 1997; Bowden and Edwardsson, 1994; Bratthall and Ericsson, 1994). Therefore, it seems reasonable that MS-positive individuals with high and very high PFRI scores (4 and 5, respectively) should be more caries susceptible than MS-negative individuals or MS-positive individuals with very low or

low PFRI scores (1 and 2, respectively). That is because the total number of the most cariogenic bacteria (MS) should be significantly higher on tooth surfaces in subjects with a PFRI score of 4 or 5 than in subjects with a PFRI score of 1 or 2, if the percentage of MS in their plaque is the same.

Prediction study. In 1984, a large-scale, combined cross-sectional and longitudinal study was initiated with the following objectives:

1. To determine the distribution of the PFRI in a large number of schoolchildren and the distribution characteristics of plaque formation on the individual tooth surfaces in the dentition.

2. To determine whether there is a correlation among the salivary S mutans level, a Cariostat test, and the PFRI score, separately or in combination, and the prevalence of smooth-surface caries.

3. To determine whether a combination of the SM level and the PFRI score is more closely related to caries prevalence than are the variables individually.

4. To determine whether there is any association between the SM level and the PFRI score.

5. To determine the influence of individual factors on the PFRI score (these data were not available at the time of writing and are not included in this chapter).

6. To determine whether caries development can be predicted by a combination of salivary S mutans levels and the PFRI.

All 716 14 year olds in Karlstad, Sweden, were recruited to participate in the study. Each was given two dental appointments, precisely 24 hours apart. The first appointment comprised the following procedures (for details on methods and materials, see Axelsson, 1989, 1991):

1. Measurement of salivary secretion rate.

2. Salivary S mutans test using the spatula method described by Kohler and Bratthall (1979).

3. Cariostat test (based on a sample of approximal plaque). The acidogenic capacity of the sample was estimated from the colorimetric indicator in the test tube showing different pH values.

4. Gingival index.

5. Plaque index based on disclosed plaque (O'Leary et al, 1972).

6. Professional mechanical tooth cleaning. The subject was instructed to refrain from all oral hygiene until the appointment scheduled for the following day.

7. Examination of all smooth surfaces for caries, with the following diagnosis: sound

surface, enamel caries, dentin caries, or filled surface. This was recorded as the subject's mean decayed or filled surface (DFS) score.

On day 2, 24 hours later, the appointment began with plaque disclosure. The presence of adherent plaque mesiobuccally, buccally, distobuccally, mesiolingually, lingually, and distolingually was noted for each tooth. The percentage of tooth surfaces with plaque was calculated according to the following formula:

Total number of surfaces with plaque x 100                 Number of teeth x 6

Each subject's PFRI was then scored according to the scale described earlier in the chapter.

At the same visit, a thorough 24-hour dietary history was recorded for subsequent evaluation of the influence of dietary factors. Many other indicators and factors possibly related to the PFRI were also evaluated, including gingival inflammation, salivary levels of glucosyl transferase, agglutinin levels in resting saliva, and oral hygiene, dietary, and fluoride habits.

Of the 716 children aged 14 years, 667 participated in the PFRI study. Figure 12 presents the frequency distribution of PFRI scores in the population. Figure 17 presents the plaque distribution on various surfaces.

Six hundred fifty-four children, 333 boys and 321 girls, formed the population for further analysis; for each of these, complete examinations were available. The other 62 children among the 716 originally selected for the study were excluded from the statistical analysis mainly because of incomplete examinations, antibiotic treatment, orthodontic bands, refusal to participate, or illness.

Results of prediction study. The examination for caries showed that 70% of the children had no dentin caries or restorations on smooth surfaces. Of the total number of lesions on these surfaces, enamel caries constituted more than 80%. Figure 42 shows the mean number of approximal carious lesions per individual in the extreme groups in relation to the PFRI score, salivary S mutans level, and Cariostat test. The group with a PFRI score of 5 had, on average, twice as many carious or restored surfaces as did the group with score 1. The difference between S mutans counts of fewer than 100,000 CFUs/mL of saliva and more than 1 million CFUs/mL of saliva was much less marked and of the same order as the differences between Cariostat blue (low acidity) and greenish yellow (high acidity).

An analysis of caries prevalence (mean DFS) related to different PFRI scores indicated a thres- hold for caries risk between PFRI scores 2 and 3 (see fig 13), and this was subsequently confirmed in the longitudinal part of the study, over 5 years (Axelsson 1989, 1991). For S mutans, this critical level was between 0 and 100,000 CFUs/mL. The Cariostat test offered no additional diagnostic advantages over S mutans counts.

Table 1 presents the mean values for caries prevalence per individual in relation to different scores of salivary SM and PFRI. The mean values were low in the

approximately 20% of subjects with 0 SM, irrespective of the PFRI score. On the other hand, individuals among the 80% of the SM-positive subjects with a PFRI score 3 showed markedly higher caries prevalence than did others. The level of salivary S mutans appeared to lack significance in this context.

These results were are also confirmed by the lack of correlations between salivary S mutans levels and PFRI scores (Table 2).

In the cross-sectional part of the study, a score of 0.0 DFS was regarded as a true-negative value for low caries risk and 10.0 DFS was chosen as a true-positive cutoff value for high caries risk because the mean value for the whole group was only 3.5 DFS. The results in Table 2 indicate that the SM level and PFRI score were independent, ie, there was no association between the variables. The contingency coefficient was low: 0.16 (upper limit = 0.87).

From the data in Table 3, it can be calculated that a combination of the PFRI and S mutans gave values of 92.1, 60.9, and 67.3%, for sensitivity, specificity, and diagnostic power, respectively; For these extremely low- or high-risk groups, the values obtained were better than were those for sensitivity, specificity, and diagnostic power for any of the variables alone (PFRI, Cariostat, SM 0, and SM 106 CFUs/mL).

Five years later, the children were reexamined. Table 4 shows the mean caries incidence on the approximal surfaces in the predicted nonrisk (SM-negative) group and the predicted risk group (SM-positive subjects with PFRI scores 3 to 5). The risk group developed five times more new approximal carious lesions in dentin per individual per 5 years than did the no-risk group, in spite of ongoing preventive programs. The question remains, without answer, how big the difference would have been without any preventive program.

The 14-year-old age group was selected as particularly appropriate for this investigation. At this age, smooth-surface cariesparticularly approximal lesionswould have developed within the previous 2 years; that is, the caries prevalence on these surfaces would largely correspond to the caries incidence over this period, and the subjects would still have a suboptimal number of intact approximal surfaces at risk. The occlusal surfaces of the molar teeth were deliberately excluded from the investigation because indications for restoration of these surfaces vary widely.

Recommendations derived from prediction study. For caries prevention, S mutans-positive subjects with a PFRI score 3 should be encouraged to clean their teeth more frequently than other 14 year olds. These subjects should probably clean their teeth twice as often as others, that is, morning and evening with an efficient use of fluoride toothpaste. For patients at extreme risk, cleaning immediately before meals and the use of fluoride chewing gum as a "dessert" after every meal may be recommended. In lingual plaque up to 12 hours old, critically low pH values do not occur after rinsing with a 10% sucrose solution. However, in approximal 3-day-old plaque, there is a potentially dangerous drop in pH. Rinsing with a sucrose solution does not cause a critical drop in pH approximally if these surfaces have been cleaned just prior to rinsing (Imfeld, 1978). Wright et al (1979) demonstrated, in a split-mouth study, that

approximal cleaning once a day gave a reduction of slightly more than 50% in approximal caries.

Artificial cleaning of the palatal surfaces is, for caries control purposes, unnecessary. The extremely low plaque formation is probably attributable to the constant friction of the rough surface of the dorsum of the tongue on these surfaces. Needs-related tests of patients' oral hygiene have clearly shown that special efforts should be concentrated on the linguoapproximal surfaces of the mandibular molars and premolars and the buccoapproximal surfaces of the corresponding maxillary teeth. Experience has also shown that the recall visit on day 2 is the ideal occasion for successful introduction of such individual needs-related oral hygiene practices.

Traditionally, a salivary S mutans count of 1 million CFUs/mL has been regarded as a critical value in assessing caries risk. The results of the previously described investigation do not support this concept. Rather, the critical limit for salivary S mutans is 0 CFUs/mL. Similar findings were reported in another study of Karlstad schoolchildren of the same age (Kristoffersson et al, 1986). However, when the same material was analyzed using approximal tooth surfaces as the unit, a very clear association emerged between the different levels of S mutans colonization and caries risk (Axelsson et al, 1987b).

Table 1 could serve as a guideline for selecting nonrisk and risk individuals. A salivary S mutans test screens out SM-negative subjects (about 25%) as not being at risk. Of the remaining 75% or so (SM-positive subjects), those with a PFRI score 3 are selected as risk patients (approximately 20%). From these subjects, an extremely high-risk group may be further selected: those with a PFRI score of 4 or 5 and an SM score of 2 or 3 (around 5%). Such a guideline is illustrated in Fig 43.

In general, if the aim of screening is to direct intensive preventive treatment toward high-risk subjects, a screening procedure offering high sensitivity and predictive value is preferable both for individual patients and for community dental health planning. A false-negative diagnosis would deny a subject at risk the benefit of additional preventive measures. The occurrence of many false-positive diagnoses would make unnecessary demands on community dental health resources. In this study, Cariostat tests and high salivary S mutans counts were less reliable as predictors than were salivary S mutans-positive status and high PFRI (scores 3 to 5).

Evaluation of the influence of several individual risk factors on the PFRI is in progress. It is anticipated that the three or four main factors will be identified. Identification of which of these factors dominates in patients with a PFRI score 3, should make it possible to design an individual preventive program in which, where feasible, preventive measures are specifically directed toward minimizing the influence of these factors.

Individuals with a PFRI score of 1 or 2 were stable over 5 years, while scores in individuals with a PFRI score of 3 to 5 sometimes changed over time. This observation indicates that plaque formation rates in individuals with a PFRI score of 4 or 5 can be reduced: Such individuals should be thoroughly evaluated to identify the factors contributing to their rapid plaque formation. Needs-related preventive measures could then be introduced.

High-caries rate prediction study. This study was followed up by a 3-year longitudinal study in Polish children (Axelsson et al, 2000a). Selected 12-year-old schoolchildren in Warsaw were randomly assigned to a test or a control group. At the baseline examination, caries prevalence (DFSs), salivary SM counts (Strip-SM), the PFRI, the Plaque Index (O'Leary, 1967), etc, were recorded. Figures 44, 45, and 46 show the frequency distribution of DFSs, PFRI scores, and Strip-SM scores, respectively, among all the children at the baseline examination.

Based on the baseline examination, the subjects were assigned to low-risk, risk, and high-risk groups, according to the following criteria:

1. Low-risk groups: Streptococcus mutans-negative individuals with a PFRI score of 1 or 2 (test: n = 47; control: n = 43).

2. Risk groups: Streptococcus mutans-positive individuals with a PFRI score of 3 (test: n = 30; control: n = 32).

3. High-risk groups: Streptococcus mutans-positive individuals with a PFRI score of 4 or 5 (test: n = 14; control: n = 13).

During the following 3 years, the children in the test group participated in a preventive program based on professional mechanical toothcleaning at needs-related intervals but with very simplified methods. For ethical reasons, the children in the control group were maintained in the regular, simple, school-based preventive program, based on oral hygiene instructions and topical fluoride administration.

After 3 years, the children in the low-risk test group had developed significantly fewer new DFSs per individual per 3 years than had the children in the risk test group, who also had developed significantly fewer new DFSs than both high-risk groups. However, all the test groups had developed fewer new DFSs than had the control groups (Fig 47). This study showed that future caries development can be predicted, even in populations with very high caries incidence, by a combination of salivary S mutans counts and the PFRI (Axelsson et al, 2000a).

Fig 42 Correlation of PFRI, S mutans, plaque pH, and caries prevalence interproximally. (From Axelsson, 1989.)

Fig 43 Four-point scale for prediction of caries risk based on S mutans and PFRI. (From Axelsson, 1991.)

Fig 44 Caries prevalence in 12-year-old Polish schoolchildren: Frequency distribution of DFSs. (From Axelsson et al, 2000a.)

Fig 45 Plaque formation rate index in Polish schoolchildren. (From Axelsson et al, 2000a.)

Fig 46 Salivary strip S mutans in Polish schoolchildren. (From Axelsson et al, 2000a.)

Fig 47 Effect of a needs-related preventive program for dental caries in Polish schoolchildren: Results after 3 years. (From Axelsson et al, 2000a.)

Conclusions

Etiology of caries

The clinical carious lesion that develops on the tooth surface beneath undisturbed

bacterial plaque represents the net result of an extraordinarily complex interplay among harmless and harmful bacteria, antagonistic and synergistic bacterial species, their metabolic products, and their interaction with the many salivary and other host factors. In other words, dental caries not only is a multifactorial disease but also has a complicated etiology. It is more difficult to demonstrate a correlation between one single species of cariogenic bacteria and future caries development in populations with low caries prevalence than it is in populations with high caries prevalence.

Three different theories for the etiology of dental caries have been proposed: the nonspecific plaque hypothesis, the ecological plaque hypothesis, and the specific plaque hypothesis. However, the true etiology is none of these, but a complex combination of all three processes. The criteria for cariogenic bacteria is that they must be acidogenic; that is, organic acids are formed as waste products of the fermentation of carbohydrates. In addition, the bacteria must be aciduric to survive in the resultant acidic environment (low pH) in the plaque and the carious lesion. Even bacteria-producing enzymes, which destroy the organic components of root cementum and dentin, may be involved in the development of root caries and dentin caries.

The basic principle of the nonspecific plaque hypothesis is that thick plaque on the tooth surface, if left undisturbed for long periods, allows the total amount of acid produced within this plaque to initiate the development of a carious lesion. Accordingly, very high plaque formers (PFRI scores 4 and 5) would be expected to develop more carious lesions than low or very low plaque formers (PFRI scores 1 and 2), if the standards of oral hygiene and the composition of the microflora were the same in the two groups.

In addition, carious lesions tend to develop on the particular tooth surfaces on which most plaque reaccumulates between toothcleaning procedures (mesiolingual and distolingual surfaces of the mandibular molars and mesiobuccal and distobuccal surfaces of the maxillary molars), and, in a toothbrushing population, where the toothbrush has limited access (the approximal surfaces of the molars and premolars). This is confirmed in studies on the pattern of caries prevalence in different populations.

Not only the frequency but also the main target of needs-related oral hygiene procedures should be based on the score and the pattern of the PFRI. Because the quantity of plaque that forms on clean tooth surfaces during a given time represents the net result of interactions among etiologic factors, many internal and external risk factors, and protective factors, future research should be directed toward methods of identifying the major factors causing rapid plaque formation in the individual patient. If possible, these factors should be reduced or eliminated.

The ecological plaque hypothesis is based on the principle that a change in a key environmental factor (or factors) will trigger a shift in the balance of the resident plaque microflora and this might predispose a site to disease. For example, the thicker the plaque, the less accessibility there is for the saliva to dilute and buffer the organic acids formed by fermentation of carbohydrates by acidogenic plaque bacteria. As a consequence, the pH will continuously decrease the more fermentable carbohydrates (for example, sucrose) are supplemented and the longer the plaque remains undisturbed. The lowered pH ( 5) in the plaque will promote a shift of the

composition of the plaque bacteria toward an increased number and assortment of acidogenic and aciduric species such as the cariogenic mutans streptococci and lactobacilli.

According to this hypothesis, the strategy for caries prevention should be to maintain a high pH on all tooth surfaces (microenvironments) by frequent removal of plaque, thereby limiting the thickness of undisturbed plaque; and by reduction of the "sugar clearance time," through a diet that stimulates saliva, and use of supplementary fluoride chewing gum as a dessert directly after every meal, particularly in patients with reduced salivary flow. Future research should focus on efficient methods for achieving homeostasis of dental plaque, maintaining high pH ( 6).

There is abundant support for the so-called specific plaque hypothesis, which proposes that some specific species of the plaque flora should be regarded as major pathogens in the etiology of dental caries. The most significant of these bacteria are some of the mutans streptococci. This group includes seven species, although two, S mutans and S sobrinus, are most closely associated with dental caries in humans. In longitudinal studies, particularly at surface level, a very strong correlation has been shown between S mutans and development of caries lesions on smooth surfaces. However, in populations with low caries prevalence, a correlation between various levels of salivary S mutans counts and caries incidence seems to be less significant: The threshold would appear to be S mutans-negative or S mutans-positive status.

The second genus closely associated with caries etiology is Lactobacillus, commonly isolated from the dentin in both coronal and root caries lesions, its main habitat in the mouth. Compared to Streptococcus, Lactobacillus has been less extensively studied.

Actinomyces odontologica, Actinomyces naeslundii, and other species of the MS group are also associated with the etiology of dental caries but are considered to be less cariogenic than S mutans, S sobrinius, and Lactobacillus.

Future research involving DNA probes and so-called genetic-fingerprinting techniques will result in tools for evaluation of (1) how S mutans is transmitted among individuals; (2) how stable the oral population is; (3) how many types of S mutans an individual carries; (4) if particular clonal lines of S mutans are more virulent (cariogenic) than others; and (5) if individuals with higher levels of carious activity carry particular types of S mutans.

Prediction and prevention of caries

The younger the population and the lower the caries prevalence in the population, the higher the percentage of caries-free subjects. In these populations, it is necessary to focus on "high-risk strategy" and primary prevention, rather than secondary prevention.

For practicing primary prevention according to the high-risk strategy, the etiologic factors used for caries prediction must be as sensitive as possible, that is, optimizing the percentage of true high-risk individuals for cost effectiveness. Because dental caries is a multifactorial disease with a complicated etiology, it is necessary to combine as many etiologic factors as possible to predict caries risk in children with

low caries prevalence, which is the situation among most children in the world.

In this approach high and very high plaque formers (PFRI scores 4 and 5, respectively) with a high percentage of cariogenic bacteria such as S mutans would be expected to develop significantly more new carious surfaces than would those with a very low or low plaque formation rate (PFRI scores 1 and 2, respectively) and little or no S mutans in the plaque.

There is a correlation between salivary S mutans counts and the number of tooth surfaces colonized with S mutans. Therefore, the combination of salivary S mutans counts and Plaque Formation Rate Index (PFRI scores 1 to 5) is recommended for caries risk prediction, according to the following scale:

1. No caries risk: Streptococcus mutans-negative individual

2. Low caries risk: Streptococcus mutans-positive individual with a PFRI score of 1 or 2

3. Caries risk: Streptococcus mutans-positive individual with a PFRI score of 3

4. High caries risk: Individual with high S mutans counts and a PFRI score of 4 or 5

Chapter 2. External Modifying Factors Involved in Dental Caries

Introduction

Awareness of the multifactorial nature of dental caries is of fundamental importance. Figure 48 illustrates the interdependence of most of the determinate variables associated with dental caries. Besides etiologic, preventive, and control factors, many other factors may modify the prevalence, onset, and progression of dental caries. Such factors may be divided into external (environmental) and internal (endogenous) factors (to be discussed in chapter 3).

Factors that have proved, in cross-sectional studies, to be significantly associated with increased prevalence of a specific disease are termed risk indicators (RIs). Factors that have proved, in well-controlled prospective studies, to increase significantly the risk for onset or progression of a specific disease are termed risk factors (RFs) and prognostic risk factors (PRFs), respectively. The RF and PRF are often expressed as the odds ratio for the onset or progression of a specific disease.

Among external modifying RIs, RFs, and PRFs for dental caries are fermentable carbohydrates, poor socioeconomic status, systemic disease, medication that impairs salivary function, and irregular dental care attendance habits.

Fig 48 Relationship between the etiologic factor in dental caries (plaque) and determinants (inner yellow circle) and cofounders (outer blue circle) in dental caries.(Modified from Fejerskov and Manji, 1990.)

Role of Dietary Factors

Role of fermentable carbohydrates (sugar and starch)

A diet rich in fermentable carbohydrates (frequent sugar intake) is indisputably a very powerful external RF and PRF for dental caries in populations with poor oral hygiene habits and an associated lack of regular topical fluoride exposure from toothpaste. However, in populations with good oral hygiene and daily use of fluoride toothpaste, sugar is a very weak RF and PRF, because clean teeth never decay, and fluoride is a unique preventive factor. The biochemical role of fermentable carbohydrates such as sucrose in the development of an enamel caries lesion on a plaque-covered tooth surface is illustrated in Fig 2 (see chapter 1).

Figure 49 depicts the most important variables and the interactions determining an eventual acid attack on enamel after dietary intake. The final (eventually cumulative) effect of single dietary intakes is dependent on their frequency. During and following consumption, depending on the quality of salivary gland function, a certain amount of saliva is stimulated by particular characteristics of the food, such as taste, acid content, and surface texture, and by the intensity of mastication. Together with the volume of saliva secreted, other substrate qualities, such as solubility, stickiness, and an individual food- and host-dependent intraoral distribution, will determine the specific oral clearance of the item in question.

Thus, after each intake, distinct amounts of dietary fermentable carbohydrates, acids, and neutralizing agents will be present and capable of influencing the pH of the surface of the tongue, of plaque, and of saliva for a given time. The resolution of the interactions among these three factors, which is greatly influenced by the thickness and diffusion characteristics of the dental plaque on the specific tooth surface, determines the severity (fall in pH) and duration of the acid attack on the tooth surface. The "true" plaque acid production, potentially dangerous to the tooth, should therefore be measured at the enamel surface beneath the undisturbed plaque.

Methods for measuring the pH of plaque will be discussed later in this chapter.

Categories of fermentable carbohydrates

Box 2 shows the fermentable carbohydrates, ranked in order of complexity. All can be fermented to acids by the plaque bacteria. In addition, the sugars may influence the quantity and quality, and thus the cariogenicity, of microbial plaque on the teeth.

For several reasons, sucrose is regarded as the arch-criminal in dental caries. Sucrose refined from sugar cane or beet is the most common dietary sugar and is largely responsible for the above-described effects. Apart from familiar sweet products, such as candy, cakes, desserts, jam, dried fruits, and soft drinks, a surprisingly large variety of other everyday foods contains added sucrose: most breakfast cereals, many milk products, some meat and fish products, salad dressings, ketchup, etc. Sucrose also occurs naturally in fruit.

The dietary sugars all diffuse rapidly into the plaque and are fermented to lactic and

other acids or can be stored as intracellular polysaccharides by the bacteria, prolonging the fall in pH and promoting a suitable environment for other aciduric and acidogenic bacteria (see chapter 1). Sucrose, however, is unique because it is the substrate for production of extracellular polysaccharides (fructan and glucan) and insoluble matrix polysaccharides (mutan). Thus, sucrose favors colonization by oral microorganisms and increases the stickiness of the plaque, allowing it to adhere in larger quantities to the teeth.

Figure 50 (a and b) shows free plaque accumulation on the same tooth during a week on a "sugar-free" diet and a week of high sugar intake, respectively. The sugar-free diet is associated with a thick pellicle and a homogenous, relatively thin plaque, while frequent sugar intake promotes the development of a thick, sticky, sugar plaque with a high percentage of extracellular polysaccharides. Because of this effect on the quality of plaque, sucrose is considered to be somewhat more cariogenic than other sugars.

Regardless of possible minor differences between the caries-inducing potential of sucrose and that of other sugars, for practical purposes all dietary monosaccharides and disaccharides are regarded as powerful risk factors: All are rapidly fermented on plaque-covered tooth surfaces. Glucose, fructose, maltose, and sucrose give identical curves for falls in the pH of plaque; for lactose, the fall in pH is somewhat smaller (Neff, 1967).

Sucrose constitutes the bulk of dietary sugar. Lactose is present in milk, and maltose is derived mainly from hydrolysis of starch. Glucose and fructose occur naturally in fruit and honey and are also formed by acid hydrolysis of sucrose during the manufacture and storage of soft drinks, marmalade, and other acidic products. Some foods (Swedish baby food) are produced with invert sugar, which is hydrolyzed sucrose. In industrial food processing, the use of glucose is increasing, produced by hydrolysis of starch from cereals or potatoes and declared in the contents as dextrose, corn syrup, or glucose syrup. Therefore, a decrease in national sucrose consumption may not necessarily reflect a drop in sugar consumption.

Starch, the major storage polysaccharide of plants, is the major dietary carbohydrate. In many countries, cereals, such as wheat, rice, maize, oats, and rye, provide about 70% of the calories; in the United States and Western Europe, the corresponding figure is 25%. Other important sources of starch are root vegetables (potato, sweet potato, cassava, yams, taro) and pulses (beans, lentils, and peas). Starch is a polysaccharide of glucose.

The starch granules in plants are only slowly attacked by salivary amylase because the starch is in an insoluble form and protected by cellulose membranes. Heating at temperatures used in cooking and baking, however, causes partial degradation to a soluble form, which can be further broken down by salivary and bacterial amylases to maltose, maltotriose, dextrins, and small amounts of glucose. Although polysaccharide molecules are too large to diffuse into the plaque, low-molecular weight carbohydrates (released in saliva or at the plaque surface) become available for bacterial fermentation.

Consumption of raw starch has little effect on the pH of plaque. The fall in pH following consumption of soluble (cooked) starch and starch-containing foods, such

as bread or crackers, while not as pronounced as for sugars, may easily reach pH 5.5 to 6.0, levels which may be critical for initiation of root caries. A combination of soluble starch and sucrose would be expected to be a more powerful caries risk factor than sucrose alone, because the increased retention of the food on the tooth surfaces would prolong sugar clearance time.

Evidence from epidemiologic studies

Numerous worldwide epidemiologic studies during the 20th century have shown that caries prevalence is low in developing countries or populations living on a local, carbohydrate-rich diet, based on starch instead of sucrose. Figure 51 shows sugar consumption in 1977 in a number of countries worldwide. Consumption is extremely low in China, and caries prevalence among 12 year olds is very low. On the other hand, sugar consumption in Japan is only about half that of other industrialized countries, but caries prevalence is moderate to high.

In contrast, for the last 30 to 40 years, sugar consumption in Sweden has remained persistently high, at about 120 g/per day (Fig 52). At the same time, caries prevalence has decreased from very high to low. Since the early 1950s, it has been "common knowledge" in Sweden that caries is "caused" by frequent intake of sweets. Despite this, over the last 30 years, indirect sugar consumption in the form of sticky sweets, cakes, and so on has increased from about 30% to more than 60% of total sugar consumption (see Fig 52). The dramatic reduction in caries prevalence is therefore attributable not to a reduction in dietary sugar but to a marked improvement in oral hygiene habits, an associated widespread, regular use of fluoride toothpaste, and needs-related professional preventive measures.

However, comparison of international data discloses an association between sugar consumption and caries development. Using information on sugar supplies in various countries, obtained from food balance sheet data prepared by the FAO, and data on caries prevalence from the World Health Organization for 6 year olds in 23 nations and 12 year olds in 47 nations, Sreebny (1982) demonstrated a significant positive correlation between the quantity of sugar available per capita in a country and caries prevalence in 12 year olds, but not in 6 year olds. In both age groups, the availability of less than 50 g sugar per person per day in a country was always associated with decayed, missing, or filled teeth scores of less than 3. However, this type of epidemiologic comparison is flawed: Sugar availability cannot directly be extrapolated to consumption specifically by 6 or 12 year olds. Both caries prevalence and sugar consumption vary among different age groups within each country.

In wartime, the availability of sugar is usually restricted. In Japan, annual sugar consumption fell from 15 kg per person prior to World War II to 0.2 kg in 1946. Many attempts have been made to relate the level of sugar consumption before, during, and after World War II to caries prevalence in the children: In Norway, Finland, and Denmark there was a clear relationship between sugar consumption and caries development in permanent first molars in children.

One of the most thorough literature surveys was made by Sognnaes (1948), who reviewed 27 wartime studies from 11 European countries, involving 750,000 children. Reductions in caries prevalence and severity were observed in all studies. Because of

the high prevalence of caries in Europeans, reductions in severity were usually greater than reductions in prevalence. Sognnaes observed that, in many of the studies, there appeared to be a delay of about 3 years between the reduction (or increase) in sugar consumption and a reduction (or increase) in caries severity.

Evidence from cross-sectional studies

Numerous cross-sectional observational studies in children have used dietary interview and questionnaire methods to study the relationship between caries prevalence and consumption of sugar and sweets. The results are somewhat conflicting (Rugg-Gunn, 1989): A significant, but not very strong, correlation between caries and the total quantity of sugar consumed has been found in some studies but not in others. A closer relationship has been demonstrated between caries and the quantities of sweets and confectionery consumed, probably because these products are consumed in ways that enhance cariogenicitybetween meals and over long periodswhereas consumption of even large quantities of sugar at meals seems to do little harm.

There may be several reasons for these findings. Most of the studies have shown a weaker correlation between sugar intake and caries than might, in theory, be expected. A general methodologic weakness is that dietary data obtained by questionnaires, 24-hour recall, or diet history interviews cover a very limited period, from only 1 day to some months, while caries data express total caries experience over the years (Birkhed, 1990). Furthermore, some people, such as the obese, are known to underreport their intake of sugar.

Interstudy comparison is difficult, because the studies have been carried out with a large range of variables: with different age groups of subjects, at different times, in different countries, and in specific populations. For example, dietary information has been collected in a variety of ways: Some reports subdivided confectionery into types of sweets, only some of which were significantly related to caries experience. The term sugary food was seldom defined, which frequently made interpretation of the correlation between caries and frequency of sugar intake difficult. Some studies were limited to only one aspect of sugar consumption, such as bedtime eating habits. In most of the studies, children were not selected for inclusion on the basis of their level of caries experience, but some studies compared only the eating habits of children at the two extremes of the range of caries experience.

Absolute figures for caries experience were not reported: In many studies only correlation coefficients were reported. In some studies, although significant correlations were found, the absolute differences in caries prevalence were small. In other cases large differences in sugar consumption habits were observed but insufficient data were presented to allow their inclusion.

Several studies have investigated the effect of infant-feeding practices on caries, particularly "rampant caries" (or labial incisor caries) in the very young. Five British studies (Goose, 1967; Goose and Gittus, 1968; James et al, 1957; Winter et al, 1966; Winter et al, 1971) have all shown a strong relationship between labial incisor caries and sugared infant pacifiers, especially nursing bottles. One study which did not show such a relationship was reported by Richardson et al (1981a) in South Africa. The

worldwide use of comforters and their effect on oral health has been reviewed by Winter (1980).

Two studies by Granath et al (1976, 1978) are of particular interest: Not only was the level of consumption of sugary foods compared with caries severity, but also two other important confounding factors, fluoride supplementation and oral hygiene practices, were taken into account. The first study, involving 6 year olds, was small (179 children) and the higher level of caries found in the children who consumed larger amounts of sugary foods between meals was not statistically significant. However, the second study, involving 4 year olds, was larger (515 children) and differences between the dietary groups were highly significant. When the effects of oral hygiene and fluoride were kept constant, the children with low between-meal sugar intake had 86% fewer buccal and lingual carious lesions and 68% fewer approximal carious lesions than did children with high between-meal sugar intake.

Hausen et al (1981), in a study involving more than 2,000 Finnish children, aged 7 to 16 years, reported that water fluoride level, toothbrushing frequency, and sugar exposure were all important determinants of caries prevalence, but least important was sugar exposure. Similarly, in another study in Finland, involving 543 children in three age groups (5, 9, and 13 years), Kleemola-Kujala and Rasanen (1982) found a stronger relationship between poor oral hygiene and caries than between high sugar consumption and caries, although both relationships were important. However, the combination of poor oral hygiene and poor dietary habits seemed to be synergistic. Very similar results were reported among 159 12- to 16-year-old French Canadians (Lachapelle-Harvey and Sevigny, 1985).

Holund et al (1985) reported more frequent consumption of sugary drinks in caries-active than caries-inactive 14-year-old Danes. Continuing the work begun by Granath in the 1970s, Schroeder and Granath (1983) found that poor dietary habits and poor oral hygiene were both good predictors of caries in 3-year-old Swedish children. A few years later, Schroeder and Edwardsson (1987) reported that the predictive potential of diet and oral hygiene can be enhanced by the addition of salivary Lactobacillus and Streptococcus mutans counts.

In another group of Swedish 13-year-old schoolchildren, positive salivary S mutans values were found to be a significant but weak risk indicator for caries, but evaluation of the intake of sticky sugar products according to an estimated point scale disclosed no correlation with caries prevalence (Kristoffersson et al, 1986).

Stecksen-Blicks et al (1985) conducted a large survey of the relationship between dietary and toothbrushing habits and caries prevalence in children of three age groups (4, 8, and 13 years) living in two northern communities and one southern community in Sweden. Children from the south had considerably more carious lesions in both primary and permanent teeth. This was attributed to differences in toothbrushing frequency and the age at which dental care started. The lack of observed differences in diet between north and south indicated that diet was not an important factor.

A large cross-sectional study in the US specifically investigated the relationship between consumption of soft drinks and caries prevalence (Ismail et al, 1984). Analyses of data from 3,194 Americans, aged 9 to 29 years, revealed significant

positive associations between frequency of between-meal consumption of soft drinks and high decayed, missing, or filled teeth scores. These associations remained even after the researchers allowed for the reported concurrent consumption of other sugary foods and other confounding variables.

In some studies, caries experience has been correlated with the subjects' dietary habits some years previously. Persson et al (1985) reported a positive correlation between the consumption of sucrose-rich foods at 12 months of age and the presence of caries at 3 years of age in 275 Swedish children. Both factors were linked to the educational status of the mother. The importance of social factors as determinants of eating habits and caries experience of young children has been highlighted in a number of studies; for example, Blinkhorn (1982) reported caries and sugar consumption in Edinburgh, Scotland, to be much higher in children from socially deprived backgrounds. The role of socioeconomic factors will be discussed later in this chapter.

In Hertford, England, Silver (1987) collected data on infant feeding and caries status in children aged 3 years and compared the data to dietary habits and caries status when the subjects were aged 8 to 10 years. "Poor infant feeding" (including the use of sugared foods and drinks) was positively correlated with the subjects' caries experience at 3 years and at 8 to 10 years. Children who in infancy had been bottle-fed with sweet drinks were more likely to be consuming sugar-containing snacks at the age of 8 to 10 years, supporting the concept that a taste for sweet food, acquired in infancy, persists in later childhood.

The importance of establishing good oral health habits as early as possible and postponing bad habits for as long as possible has recently been highlighted in a 2-year prospective study by Wendt (1995). Almost 700 infants were examined at the age of 1 year and reexamined after 1 and 2 years. At the baseline examination, the amount of plaque, gingival conditions, caries prevalence (decayed or filled surfaces), and salivary S mutans levels were recorded. At the annual examinations, the accompanying parent was interviewed about the child's oral hygiene and dietary habits. The percentage of caries-free children decreased from 99.5% to 71.7% at the age of 3 years. Among children (n = 61) of immigrant parents, only 35% were cariesfree (Wendt et al, 1992).

Children who were regularly bottle-fed with sweet drinks at night or breast-fed for more than 12 months (mostly at night, when salivary function is at resting level) developed significantly more new carious lesions than did children with more disciplined dietary habits. Bottle-feeding with sweet drinks was common among children of immigrants (Wendt and Birkhed, 1995). Because this was a prospective study, it confirmed that regular bottle-feeding with sweet drinks, and prolonged breast-feeding at night, should be regarded as risk factors for caries development in infants and toddlers.

Children who were caries free at 3 years of age had had their teeth brushed more regularly and frequently: At 1 and 2 years of age, these children already had less visible plaque than did children with caries. Immigrant children had had their teeth brushed less frequently, used fluoride toothpaste less frequently, and, at 1 year of age, already had a higher prevalence of visible plaque than did nonimmigrant children (Wendt et al, 1994).

If dietary risk behavior was already apparent at 1 year of age, the chance of remaining caries free until 3 years of age was highest if good oral hygiene habits were established by the age of 2 years. Caries-related behavioral patterns established during infancy, such as oral hygiene and dietary habits, persisted throughout early childhood (Wendt et al, 1996).

In this context, it is of interest to note that in the county of Varmland, Sweden, large-scale preventive programs at maternal and child welfare centers emphasize early establishment of good oral hygiene and dietary habits: As a result, from 1973 to 1993, the percentage of caries-free 3 year olds increased from 35% to 97%.

Although a few studies (eg, those investigating sugar intake in infant feeding) have attempted to assess lifelong habits of sugar consumption, nearly all cross-sectional studies have attempted to relate current caries prevalence to current consumption of sugar or sweets or, at the most, consumption over the previous 3 to 7 days. As discussed earlier, this approach may be valid in young children, whose teeth have erupted and developed caries within the preceding few years and whose sugar consumption habits may have remained relatively constant since the time of tooth eruption; for older groups, its validity is questionable. In a child of 12 years, caries experience is typically confined mainly to the permanent first molars, which erupted 6 years previously and may have developed caries quite early. It cannot be assumed that there has been no change in sugar consumption habits between the ages of 6 and 12 years.

It is therefore not valid to relate the dietary habits at one point in time (eg, at 12 years) to caries experience over a very much longer period (eg, 6 to 12 years). However, most cross-sectional studies have attempted to do just this. A typical example is the study by Mansbridge (1960), reporting that caries prevalence was 13% greater in 12 to 14 year olds who admitted consuming more than 8 oz (227 g) of sweets per week than it was in those claiming to consume less. The difference, although statistically significant, was modest. First molar caries prevalence was similar in the two groups, but the difference was pronounced for premolar and second molar caries. The first molars had erupted about 6 to 8 years before the sweet-eating habits were assessed, compared to fewer than 4 years for the premolars and second molars.

The many cross-sectional studies conducted several decades ago showed that, at the time, frequent intake of sugar-containing products was often a risk indicator for caries in very young individuals with relatively high caries prevalence. However, recent studies of populations older than 12 years, with good oral hygiene, including regular daily use of fluoride toothpaste, generally show very weak or no correlation between intake of sugar-containing products and caries prevalence. However, a combination of poor oral hygiene and a high frequency of sugar intake seems to have a synergistic cariogenic effect.

Evidence from human longitudinal, interventional, and experimental studies

There are many reasons why there are so few planned interventional human studies of diet and dental cariesfor example, the problem of persuading groups of people to maintain rigid dietary regimens for long periods of time. Although most of such

studies involved providing daily sugar supplements to subjectsa practice that would be considered unethical todaythese studies made an important contribution to dental knowledge. However, 25 to 50 years ago, at the time of these studies, the standard of oral hygiene was very poor and fluoride toothpaste was unavailable: In most industrialized countries, both caries incidence and caries prevalence were high.

Vipeholm study. The Vipeholm study is indisputably unique in the annals of caries research. All previous and most subsequent human studies of the diet-caries relationship have been either epidemiologic or cross-sectional surveys, based on dietary recall. Because such studies are noninterventional, the investigator has no control over either the amount or the frequency of sugar ingestion. Hence, the importance of the Vipeholm findings is matchless.

The study was conducted in Sweden, over a 5-year period (1946 to 1951). The aim was to clarify the relationship between sugar intake and caries incidence (Gustavsson et al, 1954). The subjects comprised 436 institutionalized, mentally handicapped, or retarded adults. At the baseline examination in 1946, their mean age was 32 years.

Because of their poor oral hygiene (fewer than 20% brushed their teeth regularly), they had abundant amounts of plaque, an important prerequisite for caries development. The subjects were therefore not representative of the general population. The stated objective was to study the relationship between sugar consumption and caries activity by varying the cariogenic substrate (present or absent), the amount of sugar (less than, equal to, or double the normal intake), the form of sugar (nonsticky or sticky), and the frequency of sugar intake (only at meals or at and between meals).

The chronology of the study falls broadly into three categories. During the preparatory and vitamin period (1945 to 1947), all subjects received a diet relatively low in sugar (about half the normal intake) and no additional sugar at meals. The baseline caries incidence was low, about 0.34 new carious surfaces per patient per year. During the next 2 years (1947 to 1949), carbohydrate study I, most groups consumed about twice the normal amount of sugar, but only at meals. During the final 2 years (1949 to 1951), carbohydrate study II, most groups ate normal amounts of sugar, some only at meals and others both at and between meals. Seven distinct study groups were established:

1. Control group: continued on a low-sugar diet, only at meals

2. Sucrose group: received a high-sugar diet, mostly in drinks with meals

3. Bread group: received sugar intake either half that or equal to normal, but only in sweetened bread at meals

4. Caramel group: given 22 sticky candies, either in two portions at meals (carbohydrate study I) or in four portions between meals (carbohydrate study II)

5. Eight-toffee group: given eight toffees in two portions at meals (carbohydrate study I) or in four portions between meals (carbohydrate study II)

6. Twenty-four-toffee group: allowed to eat 24 toffees, at their pleasure throughout the day, with about twice the normal total intake of sugar

7. Chocolate group: given milk chocolate in four portions between meals (carbohydrate study II)

Sugar consumption at meals in a nonsticky form, over a wide range of total daily intake, from 30 to 300 g (carbohydrate study I), had very little influence on the baseline caries rate of 0.3 to 0.5 new carious surfaces per year. The addition of sugar to the diet resulted in an increased caries incidence, but the increase varied depending on the manner of consumption (carbohydrate study II). Sugar consumed in sweet drinks with meals or in bread eaten at meals had little effect. The group receiving chocolate four times daily between meals showed a moderate increase in caries. However, a dramatic increase occurred in groups receiving 22 caramels, eight toffees, or 24 toffees between and after meals. Thus, caries risk was greatest if the sugar was consumed between meals, in a form that was retained in the mouth for a long time and provided high concentrations of sugar (Fig 53). However, there were wide individual variations. In fact, approximately 20% of the patients did not develop any caries, even after consuming 24 toffees daily.

In the Vipeholm study the total sugar consumption by the subjects was about twice that of the normal Swedish diet, and their plaque accumulation was far heavier than normally is found today. Other caries-modifying variables were also different 50 years ago. The results, therefore, should not be extrapolated directly to modern societies. The main conclusions from the Vipeholm study were:

1. Consumption of sugar, even in large quantities, is associated with only a small increase in caries incidence, provided that ingestion is limited to mealtimes, at most four times a day.

2. In subjects with poor oral hygiene, consumption of sugar both between meals and at meals is associated with a marked increase in caries incidence.

3. Under uniform experimental conditions, the increase in caries incidence varies widely from person to person.

4. Caries activity subsides once sugar-rich foods are withdrawn from the diet.

5. In subjects with poor oral hygiene, carious lesions occur despite the avoidance of sugar.

Turku sugar studies. By 1970, there was considerable evidence of variation in the rate of acid production from different sugars by plaque microorganisms: For example, the sweet polyalcohols (sorbitol, xylitol, and mannitol) produced virtually no acid. Animal experiments had also demonstrated that sugars differed in their cariogenicity. To test whether the same difference applied to humans, a clinical study was conducted in Turku, Finland, from 1972 to 1974 (Scheinin and Makinen, 1975). The objective was to study the effect on dental caries incidence of almost total substitution of sucrose, with either fructose or xylitol, in a normal diet.

Because the study required the full cooperation of the subjects, including undergoing a wide range of biochemical and microbiologic tests, the study was restricted mainly to adults, most of whom were associated with the Turku dental or medical schools. Of the original 125 subjects, 115 remained after 2 years. There were three groups of subjects: sucrose (S), fructose (F), and xylitol (X). Because full cooperation was essential, the subjects were invited to choose which group they wished to join. All clinical caries examinations were conducted blindly by the observer throughout the study. Two standardized bitewing radiographs were taken of each side of the mouth. Precavitational and cavitational lesions were recorded, for both primary and secondary caries.

The organization of the dietary regimens for the subjects in the three groups was very complex, requiring virtually all foods that normally contain sucrose to be manufactured with fructose or xylitol instead of sucrose. The cumulative development of caries, diagnosed both clinically and radiographically, is presented in Fig 54 (a). These results include both precavitational and cavitational lesions. The 24-month caries incidences in the S, F, and X groups were 7.2, 3.8, and 0.0, respectively. These refer to primary caries only. Inclusion of secondary caries (Fig 54 (b)) gives a considerable increase in the magnitude of caries incidence (10.5, 6.1, and 0.9 decayed, missing, or filled surfaces in the S, F, and X groups, respectively at 24 months), indicating the importance of secondary caries in adults.

In 1987, to quantify changes in the size of approximal carious lesions, Rekola reexamined the radiographs from the Turku study using a planimetric method. At baseline, the mean size of the lesions was similar for the X and S groups, but at the end of the 2-year study the mean size was significantly smaller in the X group (P 0.01): Lesion size increased almost linearly by 0.12 mm2/year in the S group but remained virtually unchanged in the X group.

Analyses of the caries data over the 2-year period showed that substitution of xylitol for sucrose in a normal Finnish (high-sucrose) diet resulted in a markedly lower caries incidence for both initial and manifest lesions. Although subjects in the S group developed more initial lesions than did those in the F group, more lesions in the F group progressed to cavitation. The X diet was clearly less cariogenic than either the S or F diet, but it cannot be concluded that the F diet was less cariogenic than the S diet.

Comprehensive biochemical and microbiologic tests were carried out parallel to the caries assessments. Although a very slight fall in plaque weight was observed in the S and F groups, a much greater decrease was recorded in the X group (P 0.005). Substitution of dietary sucrose with xylitol did not affect the proportion of major bacterial groups in dental plaque but did reduce the number of most organisms, especially the acidogenic and aciduric flora, including S mutans. Plaque from X group subjects showed a reduced rate of sucrose hydrolysis. No adaptation by plaque organisms to produce acid from xylitol was observed.

Experimental caries study. In experimental human studies (Von der Fehr et al, 1970), development of buccogingival enamel caries was evaluated by the use of a dissection microscope. Over a period of 23 days, dental students rinsing nine times daily with 10 mL of a 50% sucrose solution developed a higher Caries Index and more early lesions

than did the control group. Both groups abstained from oral hygiene. After 30 days of oral hygiene and daily fluoride rinses, the Caries Index returned to preexperimental levels (Fig 55).

This experiment demonstrated the rapid cariogenic effect of sucrose in combination with dental plaque. The fact that sugar in solution proved highly cariogenic suggests that the critical factor is the duration and frequency of sugar exposure rather than the physical form of the sugar-containing food. However, during the 23 days without oral hygiene, the controls also developed enamel caries.

Subsequently, the sucrose rinsing experiment was repeated for 3 weeks. This time, the subjects employed chemical plaque control by rinsing twice a day with 0.2% chlorhexidine solution but used no fluoride; no caries developed (Loe et al, 1972). These two short-term human experimental studies showed that:

1. Sugar is not an etiologic factor for caries development, but it is a modifying risk factor.

2. Dental plaque is an etiologic factor for caries development.

3. Despite frequent sugar intake, clean teeth do not develop caries, even in the absence of fluoride.

These early longitudinal, interventional, and experimental studies in Scandinavian adults clearly showed sugar to be an external modifying risk factor for caries development. However, 25 to 50 years ago, both caries incidence and caries prevalence in Scandinavia were very high. Moreover, in the designs of both the Vipeholm study and the experimental caries study, the frequency of sugar intake (eight to 24 times per day) was extreme, and the major modifying preventive factors (plaque control and fluoride administration) were absent. In other words, the etiologic factor (thick, undisturbed dental plaque) was continuously in situ on most tooth surfaces and there was no intermittent supply of fluoride to modify the fall in plaque pH.

For ethical reasons, under the Helsinki Declaration, such interventional human studies would no longer be permitted. Therefore, the relative role of sugar as an external modifying risk factor for caries development under present conditions in Scandinavia is unknown. Because of the excellent standard of plaque control and associated use of topical fluorides, particularly in toothpaste, both caries incidence and caries prevalence in children are very low, despite an increase in the consumption of sticky, sugar-containing products over the past 30 years.

Observational studies. Although interventional human longitudinal studies with frequent sugar administration are no longer permitted, longitudinal observational human studies are still allowed, and a few have been conducted during the past two decades. The studies by Wendt et al (1992), Wendt (1995), Wendt and Birkhed (1996), and Wendt et al (1996), described earlier, documented the relationship between oral hygiene and dietary habits and caries development from the ages of 1 to 3 years.

In a region of Egypt where the water fluoride concentration was higher than 1 mg/L, Axelsson and El Tabakk (2000b) followed caries incidence in relation to dietary habits in a 2-year study of 685 12 year olds with very poor oral hygiene habits (fewer than 10% brushed their teeth daily). The diet was evaluated according to a cariogenicity point scale. The results showed that a diet rich in sugar was a risk factor for caries development, albeit a weak one. The caries incidence per individual over 2 years, related to cariogenicity scores 1 to 8, 9 to 13, and 14 to 17, was 0.8, 1.0, and 1.9 new carious surfaces, respectively.

Two other important large-scale observational longitudinal studies were conducted in schoolchildren in Northumberland, England, by Rugg-Gunn et al (1984) and in Michigan by Burt et al (1988). Some data from the two studies are compared in Table 5. To avoid confounding effects, both investigations were conducted in communities with low concentrations of fluoride in water. In the English study, from 1979 to 1981, the subjects were initially aged 11.5 years; in the American study, from 1982 to 1985, the subjects were initially aged 11 to 15 years. Dietary analyses differed: Rugg-Gunn et al used 3-day diet diaries on five separate occasions, each followed by an interview, using models to assess portion size. Burt et al used 24-hour recall interviews, conducted with the aid of food models, on three or four occasions. To assess the importance of frequency of eating, the timing and grouping of food intakes were noted in both studies. Caries was scored by clinical and partial radiographic examination (Rugg-Gunn et al) or clinically only (Burt et al). In both studies, pit and fissure caries and approximal caries were scored separately.

In the Rugg-Gunn et al study, caries incidence was related to a wide range of dietary variables and by examining groups of children with extremes (high versus low) of sugar intake and caries incidence (0 versus more than 7 new decayed, missing, or filled surfaces). For total daily intake of sugars and total caries incidence, the coefficient of correlation was low, but increased when the incidence of fissure caries was considered alone. The overall incidence for smooth-surface caries may have been too low to give statistically significant results.

In a later (1987) analysis of the data, Rugg-Gunn et al examined the possible interaction between starch and sugar in the development of caries. The subjects were divided into a high-sucrose/low-starch group, and a low-sucrose/high-starch group. The former developed more new carious lesions than did the latter, but only the difference for fissure caries approached significance. No significant correlations were found between starch intake and any measure of caries incidence. The high-sugar/low-starch group ate more frequently than did the low-sucrose/high-starch group (7.8 versus 5.7 times a day).

Burt et al (1988) did not report correlation coefficients between caries incidence and dietary variables but divided the subjects into groups corresponding to contrasting caries incidences or dietary practices. Table 6 shows the selected dietary variables for their 0-increment group compared to those from children with 2 or more new carious approximal surfaces during the 3-year study. Only the comparisons of energy derived from carbohydrate (or sugar) in snacks and the percentage of total energy from sugars in snacks reached conventional levels of significance. In contrast to the conclusions of the English study, no difference was observed in the energy intakes from total sugars or from meals and snacks with one or more high-sugar foods. These differences were

unaffected when baseline age was taken into account. Social factors had a highly significant relationship with caries incidence but did not confound any of the relationships with dietary factors.

When the subjects were grouped on the basis of high and low sugar intake, those consuming high levels of total sugars developed more new carious lesions, but the difference was not pronounced and approached significance only for approximal surfaces. There was no change in significance when the subjects were grouped according to frequency of eating or high intake of sugary snacks. However, significantly more new approximal carious lesions developed in those eating large amounts of sugars between meals, and here the differences in total caries incidence approached significance. In contrast to the findings of Rugg-Gunn et al (1984), none of the analyses disclosed any differences with respect to fissure caries.

Differences in methods of collecting and analyzing the data preclude any direct comparison of these two well-conducted longitudinal trials. Nevertheless, in general, both studies confirmed that, while there is a relationship between caries development and aspects of dietary sugar consumption, it does not explain intersubject variance. Both studies also cast considerable doubt on the importance of the frequency of eating as a caries predictor.

Differences between the findings have already been indicated. The English study found that the fissures were more sensitive than the approximal surfaces to intake of sugars and other dietary influences, although the number of fissures attacked was greater and would be expected to show significance more readily. In the American study, although more than 80% of the new carious lesions developed in pits and fissures, caries at these sites was apparently unaffected by sugar intake or eating frequency.

Furthermore, the American study found a relationship between caries and the amount of sugar in snacks but not between caries and total sugar intake or intakes with one or more high-sugar items. The English data revealed high degrees of correlation between caries and total sugar intake and a significant relationship between caries and high-sugar items. In both reports, the findings were interpreted in the light of the relatively low caries incidence observed ( 1 new carious surface per individual per year).

Rugg-Gunn et al (1987) attributed the unimpressive correlation coefficients between caries and many of the sugar-related variables to problems of data collection (especially dietary data), intrasubject variation, and the low caries incidence and suggested, in hindsight, that a longer study and intergroup comparisons would have been preferable. These recommendations were adopted by the American workers, without an apparent increase in the sensitivity of the trial. In this context, it is of interest that comprehensive epidemiologic data collected annually from the total population in the county of Varmland, Sweden, disclosed that in the corresponding age groups, caries incidence decreased from 1.2 new DS in 1979 to 0.1 new DS in 1997 (Axelsson, 1998), even though there was no concurrent reduction in consumption of sugar-containing products.

Both animal experiments and human intra- oral biochemical tests (plaque pH, which will be discussed further) strongly suggest that the most critical predictor of dietary

cariogenicity is frequency of intake and that foods containing sugars generally have the greatest potential to raise the acidogenicity of plaque and thus the cariogenic challenge. The two aforementioned clinical studies, in which the methods were optimized, indicate that these two factors (especially frequency) are not of overriding importance, and this brings into question the likely efficacy of current dietary advice to patients. For example, in the Rugg-Gunn study (1987), the low-sugar/high-starch group (the subjects presumably complying with recommendations to reduce sucrose and increase dietary starch intake) achieved only a 31.7% reduction in cariesa modest result compared to the groups whose dietary practices would certainly not be approved by dental health educators.

The most significant predictors of caries risk identified in the study by Burt et al (1988) were social factors: parental education and income. The role of educational level as an external modifying factor will be discussed later in this chapter.

Fig 2 Development of noncavitated enamel caries. (Modified from Fejerskov and Clarkson, 1996.)

Fig 49 The variables and interactions that determine an eventual acid attack on enamel after eating. (Modified from Imfeld, 1983.)

Fig 50 Free plaque accumulation on the same tooth during a week of a sugar-free diet (left) and a week of frequent, high sugar consumption (right). (From Egelberg, 1965. Reprinted with permission.)

Fig 51 Sugar consumption in selected countries, 1977.

Fig 52 Sugar consumption in Sweden, 1960-1990.

Fig 53 Results of the Vipeholm study, which clarified the relationship between sugar intake and caries incidence. - DMFT = decayed, missing, and filled teeth - Vit = vitamin period - CHI1 = carbohydrate study I, part 1 - CHI2 = carbohydrate study I, part 2 - CHII1 = carbohydrate study II, part 1 - CHII2 = carbohydrate study II, part 2 - (From Gustafsson et al, 1954.)

Fig 54 (a) Cumulative development of primary caries, including precavitational and cavitational lesions, diagnosed clinically and radiographically. (b) Caries incidence including primary and secondary caries. (From Scheinin and Mokinen, 1975.)

Fig 55 Experimental study showing the rapid cariogenic effect of sucrose and dental plaque. (From Von der Fehr et al, 1970. Reprinted with permission.)

Influence of hydrogen ion concentration (pH) of plaque

It is generally accepted that enamel caries is the result of a disturbance in the equilibrium between enamel hydroxyapatite and the calcium and phosphate ion concentrations of the dental plaque covering the enamel surface. At neutral pH, plaque seems to be supersaturated with these ions. A fall in pH, however, caused by intraplaque bacterial fermentation of carbohydrates, leads to a shift in the equilibrium of concentrations and to dissolution of enamel. The "critical pH" for enamel dissolution ranges from 4.5 to 5.5, depending on such conditions as the presence of fluoride in the plaque and enamel crystal fluids.

Because dental caries is a multifactorial disease, many factors influence the pH of plaque:

1. The amount, thickness, age, site, and composition of the plaque

2. The amount, concentration, composition, clearance time, and permeability into the plaque of fermentable carbohydrates in retentive microenvironments of the dentition, and in saliva and gingival exudate

3. The amount and quality of saliva, as well as its access to and ability to permeate the plaque

4. The concentration of fluoride, calcium, and phosphate ions in the plaque

If the acidogenic theory of caries etiology is accepted, measurement of plaque pH before, during, and after a food is eaten should be a guide to its cariogenic potential. As a basis for counseling patients on the potential cariogenicity of their diet, the acidogenicicty of various foods, drinks, and meal patterns can be compared under standardized conditions. Although acidogenicity is measured, not cariogenicity, there should be a strong correlation between the two, modified only by the possible presence of protective factors, such as fluoride, which may protect the enamel against dissolution, even at low pH.

Measurement of pH

Three main methods have been used for measuring plaque pH. The original method, still in use, is the scraping, or harvesting, method developed by Fosdick et al (1941) and subsequently used in Sweden (Frostell, 1969), the United States (Edgar et al, 1975), and the United Kingdom (Rugg-Gunn et al, 1975, 1978). Small samples of plaque are obtained from representative tooth surfaces and pooled. The pH is measured in the laboratory with a pH meter.

The touch-on/microtouch method was originally developed by Stephan (1940, 1943,

1944) in his renowned "Stephan curve" experiments. Microelectrode metal probes or glass probes are inserted in plaque in situ. The method was commonly adopted and later improved by the introduction of new thin palladium oxide microelectrodes, providing increased accessibility through the entire thickness of the plaque with less disturbance. A disadvantage is that the plaque is penetrated from the outer surface, and the "true" pH between the tooth surface and the deepest part of the plaque may be altered, for example, by saliva.

The telemetric indwelling electrode method, developed by Graf and Muhlemann (1966), is the most technologically advanced and expensive, but also the most accurate method for measuring the true pH beneath undisturbed plaque. A glass electrode tip is built into either the crown of an extracted tooth or a denture tooth in a partial prosthesis, in such a way that the tip is positioned, for example, in the approximal space. Plaque is allowed to accumulate on the tip of the electrode. Wires or radiotransmitters can be used to relay readings from the mouth (Figs 56a, 56b, and 56c).

Figure 57 is a detail of plaque, freely accumulated over 7 days, on the tip of an indwelling electrode inserted in the approximal surface of an extracted natural tooth crown fixed in a partial denture. This telemetric indwelling electrode method allows continuous readings of pH at the undisturbed glass-tooth surface-plaque interface, even in the least accessible areas interproximally, where metal or glass touch-on electrodes cannot be applied.

The new microelectrodes for the touch-on method have partly overcome this problem, but will disturb the microflora each time they are inserted at the site of measurement, with a possible effect on plaque permeability. On the other hand, microelectrodes allow studies on large representative samples of individuals at any site in the mouth and can be used under field conditions.

Recent comparative studies of the three different methods (for review, see Nyvad and Fejerskov, 1996) for measuring plaque pH have indicated that the microtouch and telemetric methods give more pronounced pH responses than does the sampling method and are therefore more appropriate for differentiation of the acidogenic potential of different foods. However, irrespective of the method used, the original observations by Stephan (1944) have been confirmed: When microbial deposits are exposed to a fermentable carbohydrate, such as sucrose, for a short period of time (1 to 2 minutes), pH falls rapidly within the ensuing minutes. The pH then gradually rises, although not as rapidly, and the baseline level is resumed within 30 to 60 minutes. The severity and duration of the fall in pH will depend somewhat on the developmental stage and age of the plaque covering the tooth surface.

However, when the telemetric and Stephan methods are used concurrently on the same plaque-covered surface, the telemetric pH curve is more individualized and sensitive than the standard Stephan curve for different food items tested in sequence. The telemetric method is most frequently used on posterior approximal surfaces, which, in toothbrushing populations are the most caries susceptible. As a reference, a lingual "plaque-free" surface is used. A 10% sucrose solution is usually used as a positive control, after the subject has chewed a piece of paraffin wax for a few minutes. In plaque more than 3 days old, a 10% sucrose solution results in optimal pH

fall. Most of the telemetric studies have been conducted by Imfeld (1977, 1983).

Relationship of plaque location to pH

Stephan curves from approximal plaque show significant intraoral regional differences: Mandibular plaque has a less pronounced pH response than does maxillary plaque (Fig 58). Even within the maxilla, there are local differences in the Stephan response, attributable partly to variations in accessibility to saliva. The lowest pH values are recorded for anterior sites. The gradual resumption of baseline pH values probably results from diffusion of acids out of the plaque and the neutralizing effect of buffers within the plaque and in the salivary film covering the plaque surface (Fejerskov et al, 1992).

Accessibility of saliva is influenced by tooth morphology and location and variations in the flow of saliva from the different salivary glands. Access to pits and fissures and approximal surfaces is poor, favoring plaque acidity (Kleinberg and Jenkins, 1964); other tooth surfaces in the vicinity of these sites will be more accessible to saliva and as a result, the plaque will be much less acidic (see Fig 11).

Some teeth, such as the mandibular incisors, are located in regions of the mouth where saliva is abundant. The plaque on the maxillary incisors is less alkaline than the corresponding mandibular plaque and favors the development of caries, whereas there is a greater tendency to calculus formation on the mandibular incisors. The volume of saliva secreted by the major salivary glands varies considerably: The greatest flow is from the submandibular and sublingual glands, which have duct orifices in the floor of the mouth just lingual to the mandibular incisors.

There is an important difference between the intraoral distribution of sugar ingested in solution and that of sugar in solid foods. Sugar in solution flows over the same tooth surfaces as the saliva and will rapidly be cleared from the oral cavity, except under certain conditions, eg, if ingested in high concentrations or if salivary secretion is seriously impaired. The sugar in solid foods that have to be chewed will enter pits and fissures and be retained in approximal embrasures, the stagnation sites in the dentition.

By using the wire telemetric method, Igarashi et al (1989) showed that, after a 1-minute rinse with 10% sucrose solution (Fig 59), the pH was much lower in 4-day-old approximal plaque than in the corresponding fissure plaque.

Relationship of plaque age and composition to pH

The telemetric method has been used to evaluate the influence of plaque age on pH, following a 2-minute rinse with 10% sucrose solution. Figure 60a shows the pH fall in 2-, 3-, 5-, and 6-day-old interdental plaque in a 14-year-old boy. Irrespective of the subject's age, and in experiments in the same test subject over a 2-year period, it seems that a critical fall in pH (to below 5) occurs only in 3-day-old plaque. Figure 60b shows the fall in pH in 3-day-old plaque in a 52-year-old woman, a 7-year-old girl, and a 7-year-old boy after they rinsed with sucrose (Imfeld, 1978, 1983).

In a toothbrushing population, such mature plaque would be found, if at all, only on

the approximal surfaces of the molars and premolars. This explains why, in such a population, these surfaces are the most susceptible to caries.

Plaque composition also may influence the pH of plaque. The fall in pH after a sucrose rinse would be expected to be more severe in cariogenic plaque with a high percentage of acidogenic bacteria than in noncariogenic plaque. This is illustrated in Fig 61. In a group of 14 year olds, the fall in plaque pH (Stephan curves) after a sucrose rinse was measured in intact occlusal surfaces, inactive occlusal carious lesions, and active occlusal carious lesions (Fejerskov et al, 1992). The age of the plaque, however, may also have varied from subject to subject.

Relationship of different carbohydrates and sugar concentrations to pH and clearance time

Neff (1967) used the Stephan method for evaluation of plaque pH changes associated with different fermentable carbohydrates. Figure 62 (a) shows the drop in pH for lactose, glucose, maltose, fructose, and sucrose. Figure 62 (b) shows the effect of raw starch, cooked starch, maltose, and sucrose. These experiments indicated that raw starch can be regarded as noncariogenic. However, under certain conditions, lactose and cooked starch may cause a drop in pH to critical values for initiation of root caries. In cariogenic plaque, glucose, maltose, fructose, and sucrose all seem to have the potential to cause a fall in pH to the critical value for the development of enamel caries.

In a series of telemetric experiments, Imfeld (1978, 1983) measured the pH beneath 4-day-old interdental plaque after 2-minute rinses with 0.025%, 1.25%, 2.5%, 5%, and 10% sucrose solutions. At the beginning and end of each session, as well as between treatments, the acidified plaque was neutralized by salivary flow, which was stimulated by the chewing of neutral paraffin. If very low plaque pH values were attained by glycolysis, rinsing with a 3% carbamide solution greatly improved the neutralization of plaque acids, through intraplaque ammonia formation. Although carbamide is cleared from plaque very rapidly, a further paraffin chewing phase was introduced to ensure its removal, and this always resulted in physiologic plaque pH values.

Figure 63 shows the telemetrically recorded pH of 4-day-old interdental plaque in one subject during and after rinsing with increasing concentrations of sucrose solutions. The test solutions were always spit out after the subject rinsed.

Sucrose is rapidly fermented in plaque. Regardless of the concentration of sucrose, intraplaque pH drops immediately on sucrose intake and throughout the entire 2-minute rinsing period. The very small quantity of sucrose remaining after the 15-mL 0.025% sucrose solution has been spit out is sufficient to depress plaque pH below 5.7. Up to a certain limit (15 mL, 10%), the amount of fermentable substrate is negatively correlated with the lowest pH value reached. Higher sucrose concentrations do not further depress pH. This is the rationale for using a 10% sucrose solution as a positive control in most telemetric experiments.

Figure 64 shows the falls in pH occurring on a plaque-free lingual surface and on 4-day-old interdental plaque after rinsing with 0.1%, 0.5%, 1%, and 5% sucrose

solutions (Imfeld, 1978, 1983). Even weak sucrose solutions (2.5% to 5%) yield suboptimal drops in pH (to the level of pH 4.2 to 4.5), well within the critical pH range for enamel caries.

For reference, Table 7 shows the concentrations of glucose, fructose, and sucrose in some common Swedish food products. It is clearly unrealistic to try to exclude sugar or reduce dietary concentrations to levels low enough to eliminate the risk of inducing critical plaque pH values in mature, cariogenic plaque. A more realistic approach to caries prevention and control would involve:

1. Removal of dental plaque from all tooth surfaces once or twice a day, with concurrent use of fluoride toothpaste.

2. Restriction of the total number of food intakes, including snacks, to four to six per day and exclusion of sticky sugar-containing products: This will reduce the total daily sugar clearance time.

In addition to the chemical composition of foods, physical and organoleptic properties (particle size, solubility, adhesiveness, texture, and taste) are important for cariogenicity, because they influence eating patterns and intraoral retention of foods. The oral carbohydrate concentration and the length of time carbohydrates remain in the mouth during and after eating are important characteristics.

Foods are eliminated during and after mastication by the flushing action of saliva and by the activities of the masticatory muscles, tongue, lips, and cheeks. Clearance times may be prolonged by retentive factors in the dentition (carious lesions, poor restorations, fixed partial dentures, and removable partial dentures), by low secretion rates, or by high viscosity of saliva. According to the telemetric method, initial oral carbohydrate concentrations and clearance times show large individual variations (Imfeld, 1983) (Figs 65, 66, 67, 68, and 69), and slow clearance increases caries risk.

Different foods also vary greatly in initial oral carbohydrate concentration and clearance times. The carbohydrates in fruits with a high acid content, such as apples and oranges (see Figs 66 and 67), vegetables, and various drinks are eliminated within 5 minutes. Sweets, such as sugar-containing chewing gum, caramels, toffees, chocolates, and lozenges, generally result in high oral sucrose concentrations and clearance times ranging from 40 minutes for chewing gum to 15 to 20 minutes for other sweets.

On the other hand, concluding each meal with sugar-free fluoride chewing gum is an excellent caries-preventive measure, particularly for high-risk xerostomic patients. Clearance times for bread and crackers may be reduced because the rough texture requires vigorous chewing, which stimulates a high salivary flow. The high secretion rate of saliva induced by vigorous chewing not only has a mechanical rinsing effect but also increases the buffering capacity of saliva, which accelerates neutralization of plaque acids.

Increasing the contact time between dental plaque and sucrose leads to a continuously declining interdental plaque pH, thereby increasing its cariogenicity (see, for example, the effect of bananas in Fig 65). The effect of other products, such as some dried fruits

and cakes, is probably even greater. In marked contrast is the effect of cheese (see Fig 68).

Bananas, which are pasty and contain 15% sugar (sucrose, fructose, glucose), have the potential to reduce the pH of 4-day-old interdental plaque to almost the same level as rinsing with 10% sucrose solution over a prolonged period (1.5 hours); apples (10% sugar) and oranges seem to be noncariogenic but tend to be erosive because of their acid content (see Figs 65, 66, and 67). This tendency may be counterbalanced by the increased salivary flow stimulated by the acids in fruit.

Relationship of eating patterns to pH

Studies of the effect of meal patterns on intraplaque acid formation have shown that the fall in plaque pH after consumption of sugary foods may be considerably modified by the consumption of less fermentable foods before, concurrently, or afterward. Imfeld (1983), using the telemetric method, demonstrated the pronounced influence of the last course of a meal on the duration of the postprandial fall in plaque pH. Eating 30 g of Camembert cheese after lunch (see Fig 68) raised the pH of 5-day-old interdental plaque, which had fallen during the meal, but eating chocolate cream as a dessert prolonged and exacerbated the low pH of the interdental plaque (see Fig 69).

The observation of the effect of cheese is in agreement with other studies (Schachtele et al, 1982). Animal studies have shown that cheese reduces caries incidence in rats (Edgar et al, 1981). Eating cheese not only stimulates the flow of saliva but also releases calcium and phosphate, which enhance the buffer capacity and remineralization potential.

Relationship of sugar substitutes to pH

For thousands of years, humans have craved sweet food. Infants rapidly become accustomed to a sweet taste, and this is sometimes acquired prenatally.

In frequently consumed snack foods such, as sweets and drinks, less fermentable and noncariogenic sweeteners are increasingly being used as substitutes for potentially cariogenic sugars (monosaccharides and disaccharides). These sugar substitutes are often classified as caloric or noncaloric sweeteners. Among the caloric sugar substitutes are the sugar alcohols (sorbitol, xylitol, and mannitol) and hydrogenated glucose (Lycasin). Examples of common noncaloric sugar substitutes are saccharin, cyclamate, and aspartame. Table 8 from Rugg-Gunn (1989) shows the sweetness of different sugars and sugar substitutes relative to sucrose.

Most of the sugar substitutes have been tested by the telemetric method (Imfeld, 1983; Imfeld and Muhlemann, 1978). Figure 70 shows the telemetric pH of 5-day-old interdental plaque in one subject during and after rinsing with 10% aqueous solutions of Lycasin 80/55, xylitol, sorbitol, sorbose, and sucrose.

The sugar substitutes Lycasin 80/55, xylitol, and sorbitol, and the sugar sorbose have been declared safe for teeth according to the criteria applied by the Swiss Office of Health. Sucrose, used as a positive control, and administered in the same way as the sugar substitutes and sorbose, resulted in a prolonged fall in pH to below 4.5. Among

others, the longitudinal clinical Turku study (Scheinin et al, 1975) described earlier, as well as the following chewing gum study, have shown that xylitol is noncariogenic. All noncaloric sweeteners are also noncariogenic: They cannot be fermented at all by the acidogenic plaque bacteria.

Fig 11 pH drop in molars with approximal plaque at four different sites after rinsing with sucrose solution. (From Firestone et al, 1987.)

Fig 56a Volunteer sitting in the Faraday cage during a test to measure the response of interdental plaque pH to the consumption of a European breakfast. (Courtesy T. Imfeld.)

Fig 56b Oral wire telemetric prosthesis inserted with cable connection during a recording session. (Courtesy T. Imfeld.)

Fig 56c Tip of an interdental pH electrode covered with 6-day-old plaque. (Courtesy T. Imfeld.)

Fig 57 A close up of plaque, freely accumulated over 7 days, on the tip of an electrode. (Courtesy T. Imfeld.)

Fig 58 Stephan response curves obtained after a sucrose rinse from approximal spaces in the maxilla and the mandible. In the 7 year olds, pH was measured between deciduous molars, and in the 14 year olds, between premolars. Note that mandibular sites exhibit a less pronounced pH

drop. Bars indicate standard errors. (From Fejerskov et al, 1992. Reprinted with permission.)

Fig 59 Results of the wire telemetric method show the low level of pH in approximal plaque compared to fissure plaque. (From Igarashi et al, 1989. Reprinted with permission.)

Fig 60a Telemetrically recorded pH of 2-, 3-, 5-, and 6-day-old interdental plaque in a 14-year-old boy during and 15 minutes after a sucrose rinse (15mL, 10%). PC = 3 minutes paraffin chewing. Fig 60b Comparsion of telemetrically recorded pH values of 3-day-old interdental plaque in a 52 year old and two 1 year olds during and 15 minutes after a 2-minute sucrose rinse (15 mL, 10%). (From Imfeld. Reprinted with permission.)

Fig 61 Stephan response curves obtained from sound occlusal surfaces, inactive occlusal caries lesions, and deep, active occlusal carious cavities following a sucrose rinse in a group of 14 year olds. (From Fejerskov et al, 1992. Reprinted with permission.)

Fig 62 The Stephan method evaluates plaque pH changes associated with particular carbohydrates. (From Neff, 1967. Reprinted with permission.)

Fig 63 Telemetrically recorded pH of 4-day-old interdental plaque after rinsing with increasing concentrations of sucrose solution. U = 2 min 3% urea rinse. PC = 3 min paraffin chewing. (Courtesy T. Imfeld.) Fig 64 Fall in pH on a plaque-free lingual surface (lg) and on 4-day-old interdental plaque (id) after rinsing with different concentrations of sucrose solutions. PC = 3 min paraffin chewing. (Courtesy T. Imfeld.) Fig 65 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.) Fig 66 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.) Fig 67 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.) Fig 68 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.)

Fig 69 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.) Fig 70 The telemetrically recorded pH of 5-day-old interdental plaque during and after rinsing with water solution of some sugar substitutes (Lycasin, xylitol, sorbitol, sorbose) and 10% sucrose. U = 2 min 3% urea rinse. PC = paraffin chewing. (Courtesy T. Imfeld.)

Evaluation of dietary factors

The human longitudinal studies described earlier showed that, in individuals with little or no plaque control and no use of fluoride, frequent intake of sugar-containing products is a significant risk factor or prognostic risk factor for dental caries.

In addition, in vivo plaque pH measurements have shown that the drop in pH and sugar clearance time in undisturbed plaque (more than 2 days old) is related to the sugar concentration and consistency of the food item being evaluated (see Figs 63, 64, 65, 66, 67, 68, and 69). Frequent intake of sticky sugar-containing products results in prolonged sugar clearance time, which further prolongs the drop in pH on all tooth surfaces covered with undisturbed cariogenic plaque.

Because a prolonged drop in plaque pH eventually results in demineralization of the enamel and development of a carious lesion, the sugar clearance time, based on an evaluation of dietary habits and eating patterns, would seem to be essential information to obtain for all caries-active individuals. Data obtained from evaluation of dietary habits not only provides background material for caries risk assessment but also aids dietary counseling in needs-related caries control and the encouragement of good dietary habits in general health promotion.

Dietary assessment in dental practice is aimed at estimating the cariogenic challenge caused by carbohydrates and assessing the general nutritive value of a diet. This means that information on eating patterns and intake of fermentable carbohydrates, as well as energy and other nutrients, should be collected and evaluated. The goal is to establish the absolute magnitude of these variables with the least degree of measurement error. These objectives form the basis for selecting a method. Of several available methods, the following are suitable for dental practice: dietary history, 24-hour recall, dietary record, and food frequency questionnaires.

Dietary history

All methods can be used in dental practice, but, in its original version, the dietary history method takes 1 to 2 hours. This method is considered accurate when validated with nitrogen excretion in urine, but is generally too time consuming for dental practice. However, modified forms may be combined with one of the other interview methods.

24-hour recall

This method is widely used. A trained member of the dental team interviews the

patient about the intake of food and beverages during the latest 24-hour period. Consistency in the technique and the skill of the interviewer are important factors influencing communication and patient cooperation and thereby the result. Food models or life-sized illustrations are recommended by most researchers as an aid in estimating quantities. The portion size can also be given in household measures, such as glasses, cups, tablespoons, ounces, and pounds.

To reduce bias, the 24-hour recall is done without prior notice to the patient. It should then be repeated for at least 4 days to establish the eating pattern and intake of energy-providing nutrients. For nutrients with large day-to-day variations, the number of days is increased. For example, the time required to estimate the true intake of vitamin A is reported to be approximately 40 days. The days should be selected to represent ordinary days and include weekdays as well as a weekend. Boxes 3 and 4 present examples of 24-hour recalls of a highly cariogenic diet and a noncariogenic diet, respectively.

Dietary record

In dietary records, also called food diaries, the patient records the type and quantity of all food and drink consumed over a prescribed period, usually 3 to 7 days. Estimates of portion sizes and selection of days are the same as for the 24-hour recall. The patient is given the following detailed instructions:

1. To make the evaluation as accurate as possible, ordinary dietary habits should be kept. Record carefully and precisely. For example:

a. How many slices and what kind of bread is used for sandwicheswhat kind of spread is used and what filling?

b. What is drunk with or between meals?

c. Is jam or sugar used with milk, buttermilk, or yogurt?

d. How many lumps of sugar in tea or coffee?

e. Are vegetables raw or boiled?

2. Include all snacks: soft drinks, sweet rolls, fruit puree, milk with a sandwich, fruit or sweets, chewing gum, and throat lozenges.

This prospective strategy may increase measurement error because of incomplete registration or deliberate or inadvertent changes in diet. Both the dietary record and the 24-hour recall method are reported to underestimate intake slightly compared to the dietary history method and excretion of urinary nitrogen.

Food-frequency questionnaires

A food-frequency questionnaire contains a list of food items, usually 50 to 150 items, selected to illustrate the whole diet or a specific nutrient, eg, sucrose. The patient marks his or her consumption on a scale, ranging from never to several times per day.

An example of this is a questionnaire aiming to measure intake of food items (Fig 71). The patients mark with a cross the most appropriate square. Figure 72 shows another questionnaire with special reference to the frequency of sugar-containing products.

The frequency questionnaire also can be used to estimate nutrient intake. There is a strong correlation between consumption frequency and intake of energy and nutrients. The frequency questionnaire method is uncomplicated and inexpensive and may be useful as a screening instrument or for obtaining dietary data at a group level.

Analysis of dietary data

When the advantages and disadvantages of these methods are assessed, use of the repeated 24-hour recall and the food record method for 4 to 7 days seems to be the most appropriate for dental practice. The 24-hour recall method is preferable for adolescents, for the elderly, and when communication is poor. The length of the study period is decided according to demands for precision of micronutrients, such as vitamins and minerals. For caries, a 4-day record usually meets the requirements.

After completion of data collection and a check on the plausibility of the reported consumption, the intake is evaluated. Evaluation of the cariogenic potential includes an estimation of factors such as the number of intakes containing fermentable carbohydrates, the consumption of snacks and sugar-containing drinks at night, and the retentiveness of the cariogenic products. In children and adolescents with an uncomplicated pattern of caries, simply scoring sucrose intake is often adequate.

Several inexpensive computer software programs for evaluation of energy and nutrients in the diet are available in Scandinavia, and computer-based analysis of dietary registrations is common. This is a convenient way to evaluate the nutritive value of the intake. The results of changes to the diet are readily demonstrated, which is of great educational value for patients.

Another way to estimate nutritive value of the diet is to score the number of intakes representing six specific food groups (Fig 73). Samples of charts that can be used in such a food group-based evaluation are shown in Tables 9 and 10.

When the cariogenic potential and nutritive value are assessed, other properties of the food, known to modify the carious process, for example, food that requires chewing, should also be considered. The ensuing stimulation of salivary secretion and distribution reduces the duration of a drop in plaque pH.

Fig 71 Dietary questionnaire for adults. (From Holm et al, 1983.)

Fig 72 Dietary habits questionnaire. (Courtesy D. Birkhed.)

Fig 73 The six main food groups arranged in a food circle with recommended daily intake for the average adult. (From the Danish Goverment Home Economics Council, Copenhagen, 1989.)

Dietary recommendations for general health promotion

General recommendations

General guidelines for energy and nutrient intake are given in the Nordic recommendations from 1989, and in recommendations specific for each Scandinavian country. They give age- and sex-specific recommendations for daily energy and nutrient intake as well as minimal daily required amounts for healthy individuals older than 3 years. It is recommended that energy intake be at a level that does not cause obesity and that there be five or six daily intakes of food at even intervals throughout the day. Recent studies indicate that a more frequent eating schedule would offer physiologic benefits, for example, a decrease in total serum cholesterol concentration. However, other aspects of such recommendations must be considered before they are generally adopted.

The daily recommended energy intake originates mainly from carbohydrates (55% to 60% of the total energy). Fat should provide a minimum of 20% and a maximum of 30% of the energy. A fat intake below 20% to 25% of total energy may lead to deficiency of essential fatty acids. Protein provides the remaining 10% to 15% of daily energy. Specific recommendations are also given for fiber, salt, alcohol, and micronutrients.

To fulfill these recommendations, the average diet in all Scandinavian countries would need modifications described in Box 5. These recommendations are useful guidelines for medical as well as dental practice and can also be applied to reduce caries risk. The dietary recommendations for diabetics are also in general agreement with these guidelines.

Individual recommendations

After assessment of the dietary information, the advised plan for the individual is formulated. A useful tool may be "sugar clocks," demonstrating the high caries risk associated with frequent eating (Fig 74).

In some patients, carious activity may be attributable to a single habit, eg, frequent consumption of sugar-containing lozenges or snacking or drinking soft drinks at night. Such habits are readily identified and usually easily rectified. In other patients, eating habits may be more complex, comprising snacks only and no main meals. In such cases, a change in basic behavior is required.

This process is complicated by the fact that humans dislike change. Therefore, enforced dietary changes will not succeed unless the benefit accrues rapidly and is of demonstrable advantage. This can be seen in some weight-reducing programs or, for example, when uremic patients adopt a protein-reduced diet. Otherwise, a successful change in dietary behavior requires a program of repeated, small steps. This applies to the introduction of new food items and habits in small children as well as in adults.

It is also important that the advice be compatible with possible disease conditions or medication in the individual patient and that the proposed changes be acceptable to the patient. Of further importance is that dietary counseling take into account the patient's social situation. The basis for designing advice sheets on proper energy and nutrient intake is beyond the scope of this book. For further information the reader is referred to textbooks on nutrition.

The objective of dietary evaluations and recommendations related to dental caries should be to reduce the total sugar clearance time per day. However, because root caries can develop at a pH as high as 6, the intake of sticky, starch-containing products must also be regarded as a powerful modifying risk factor in elderly people with exposed root surfaces and impaired salivary function. High salivary levels of lactobacilli indicate a high sugar intake and low intraoral pH. The Lactobacillus test is therefore a valuable objective supplement to the dietary questionnaire.

For caries prevention and control, compliance with the following dietary recommendations is essential (Box 6).

Fig 74 Sugar clocks. (left) Frequent eating results in many periods of acid formation in dental plaque (rod areas). (right) Eating occurs five times a day, resulting in long periods (green area) with no acid formation. (Modified from Johansson and Birkhed, 1994. Reprinted with permission.)

Influence of other risk factors on diet-related caries

Certain conditions may predispose people to risk for diet-related dental caries. Systemic diseases and regular medication may affect caries risk. The disease or medication per se might increase caries risk, but sometimes the increased risk is related to treatment. The increased need for energy and nutrients during a disease episode is often not met, and the patient may be undernourished. Intake of medicines containing sucrose must be noted, eg, fiber supplements for constipation, cough mixtures, and antibiotics.

Further, the intake of soft drinks and sweets is found to be high in hospitalized patients. In some diseases, dietary treatment relieves disease symptoms. Thus, a reduced-fat diet eases diarrhea associated with Crohn's disease or irradiation of the abdominal tract. A low-protein diet defers the need for dialysis in patients with uremia. To compensate for the reduced fat or protein intake, carbohydrate intake is increased, and this increases caries risk. Monosaccharides and disaccharides are used generously; otherwise, the meals would be too large.

Dental caries in patients with psychiatric disorders may be complex to explain.

Carbohydrates favor the uptake of tryptophan to the brain, and serotonin production is enhanced. Thus, carbohydrates can have a sedative effect, and frequent eating may induce relaxation. Caries resistance may be lowered by concurrent medication with psychiatric drugs which often impair salivary secretion, as will be discussed in chapter 3.

Abuse of recreational drugs, such as hashish, may be associated with a craving for sweets. These patients frequently have high caries activity, typically with smooth-surface lesions.

A few decades ago, pregnancy was regarded as a cause of tooth loss resulting from dental caries. Although this is no longer the case, pregnancy may be associated with increased caries risk in some women. During the first trimester, problems with oral hygiene may result from nausea. Pregnancy is often associated with cravings for sweets and more frequent eating. Hormonal changes will also reduce the amount and quality of saliva during the final months of pregnancy.

Studies have shown an association between obesity and caries prevalence. However, the association with diet has not been clear. Several studies have shown that the obese underreport total energy, fat, and sucrose intake, but overreport vitamin C and fiber. It could, therefore, be assumed that the sucrose intake in obese individuals with a caries problem is higher than is disclosed by the patient during the dietary registration.

Occupations in which frequent food sampling is possible, or even a necessary aspect of work, are associated with an increased risk for dental caries. Examples of such occupations are workers in the confectionery industry and restaurant personnel. Bakery workers were also once considered to be at higher risk for caries (for reviews on dietary factors related to dental caries, see Imfeld, 1983; Rugg-Gunn, 1989, in Murray, 1989; Edgar and Higham, 1991, Geddes, 1991, Bowen, 1994, Geddes, 1994, Imfeld, 1994a,b, Marsh, 1994, Johansson and Birkhed, 1994, Nyvad and Fejerskov, 1994, Carlsson and Hamilton, 1994, Rugg-Gunn, 1994).

Role of Socioeconomic and Behavioral Factors

Introduction

At group and population level, socioeconomic factors, particularly educational levels, are emerging as the most important external factors related to dental caries today.

History has clearly shown a relationship between social characteristics and dental disease patterns and, in particular, how social changes have influenced those patterns. Wartime, urbanization, and industrialization, to mention a few, have affected caries prevalence. Most often, we think of social class when we talk about social factors. There are various classifications of social class, usually based on the income of the head of the household and the length and type of education.

The links between social class and dental caries have been demonstrated in many studies (Antoft et al, 1988; Beal, 1989; Holm et al, 1975; Koch and Martinsson, 1970; Milen, 1987; Schwarz, 1985; Zadik, 1978). Throughout the 20th century, in temperate and industrialized countries, caries prevalence in primary teeth has been found to

increase with decreasing socioeconomic status. In contrast, during the first half of the century, caries experience in permanent teeth was more prevalent in the highest social class, but the situation is now reversed (Milen and Tala, 1986). In most tropical and developing countries, on the other hand, caries prevalence has been reported to increase with increasing socioeconomic status (for review, see Enwonwu, 1981).

Social factors are closely linked to behavioral factors, and a great number of behaviors, particularly health behaviors, are characteristic and distinctive for each social class. Other indicators have also been used, for example, which newspaper the household reads, whether the family has a car, the number of households with no bath, the absence of an inside toilet, shared toilet facilities, residents per room, and similar factors. With such information, geographic areas can be ranked separately for each variable. A combination of variables gives a clear indication of the most and least advantaged areas: these appear at opposite ends of the scale. Such a system is valid for extremes, such as high-risk groups, but less reliable for the middle ranges.

Palmer and Pitter (1988) used such a classification and clearly demonstrated wide variations in caries status and treatment levels in 8-year-old English children from different social backgrounds. Socially disadvantaged children had a much higher level of dental caries than did their more socially advantaged contemporaries. The potential number of social and behavioral indicators of deprived or disadvantaged groups or individuals is enormous; such indicators must be chosen with care, with special reference not only to relevance to a given society but also to the changing nature of these indicators with time.

The Korner Report (Department of Health and Social Security [DHSS], 1982) recently questioned the validity of social class as a health-related variable and set up an inquiry to study alternatives. Sarll et al (1984) have studied the advantages and limitations of a composite indicator, A Classification Of Residential Neighborhoods (ACORN), a system based on census statistics, in terms of its use in planning dental services. In the industrial area in the north of England they found that the socioeconomic ACORN analyses effectively identified differences in caries prevalence. Data collection was simple, and a high proportion of subjects could be classified. In addition, the ACORN classification, relying on postal address, avoided the need for questions about occupation and economic circumstances, which may be particularly difficult in studies of children.

The national survey of children's dental health in the United Kingdom (Todd and Dodd, 1985) disclosed regional inequalities in dental health: Children living in England had the least dental caries, and those in Northern Ireland had the most. The findings showed that, at the national level, where a child lives is a more important factor than social class in determining caries experience.

Influence of socioeconomic status

Social class

The relationship between parents' social status and children's dental health has been demonstrated in numerous studies. Many studies in Western industrialized countries have also shown a relationship between on the one hand, the parents' dental health

status, dental knowledge, and dental care habits and on the other, the prevalence and incidence of dental caries in their children (Martinsson, 1973; Martinsson and Petersson, 1972). For example, Martinsson and Petersson (1972) found a much higher percentage of edentulous parents among children with high caries experience than children with low caries experience. Asher et al (1986) reported a significant correlation between parents' dietary carbohydrate intake and the oral health of the dependent child.

Beal (1989) has detailed risk factors contributing to higher caries prevalence in children of low socioeconomic background. These include infant-feeding practices conducive to nursing bottle caries, lesser parental involvement in hygiene practices, and a much lesser parental knowledge of, and involvement in, topical and supplementary fluoride regimens.

In the US caries prevalence in schoolchildren in relation to the educational level of the children's mothers was evaluated. Because the area had fluoridated drinking water, caries prevalence in the children was generally low; nevertheless, there was still significantly less caries in children with well-educated mothers than in those whose mothers had less education. In the 3-year longitudinal study in almost 500 US schoolchildren, discussed earlier, Burt et al (1988) failed to find any correlation between frequency of intake of sugary products and children who developed 0 or more than 2 approximal carious lesions. Social factors (parents' income and educational level) had a highly significant relationship with caries incidence, but these factors did not confound any of the relationships with dietary factors.

In a longitudinal study, Grytten et al (1988c) examined the influence of various social and behavioral variables on, and the predictability of, caries experience in early childhood. Data were collected when the children were 6, 18, and 36 months old, through parental questionnaires and, at 36 months of age, clinical examination. Caries experience at 36 months showed a statistically significant association with the child's sugar consumption as well as with the mother's dental health, dental care attendance pattern, and level of education. However, when a multivariate model was constructed of predictors that bivariately had shown a statistically significant association with caries experience, only the number of missing teeth in the mother was significantly associated with caries experience, and the explained variance of the dependent variable was low.

Primosch (1982) investigated the effect of family structure on dental caries experience of children, in an attempt to identify those at greatest risk. Multiple linear regression analysis showed that none of the selected variables in family structure was sensitive enough to predict children at greatest risk. Maternal age at marriage and family size, however, seemed to show the most promise for predictive value. Comparison of the family structure of children with high and low caries experience disclosed the following:

1. Children of parents who married young (mother younger than 20 years and father younger than 22 years) had significantly greater caries prevalence.

2. Children born to mothers younger than 23 years and fathers younger than 28 years were also more susceptible to caries.

3. Children with birth ranks or family size at either extreme (one child or more than three children) were significantly more susceptible to caries.

4. Age-span differences between siblings had little effect on the caries experience of the subject.

The predictive power of a number of sociologic and behavioral variables was investigated by Poulsen (1988) in a study of the public child dental service in Denmark. A multivariate logistic regression analysis, expressing caries risk by the odds ratio, showed high risk of caries (Table 11) in the following cases:

1. Learning disability in the child

2. A high level of pocket money spent on sweets

3. Little support from family

4. Little or no discussion about dental health

5. Negative attitudes toward dental health

6. Negative parental attitudes toward a healthy diet

7. Low educational level

8. Economic pressures in the family

The sensitivity was 66%, the specificity 80%, and the predictive power 71%. However, when sociologic variables as well as epidemiologic variables were included in the analysis, sensitivity increased to 95%, specificity to 91%, and the predictive power to 91%. This study clearly shows the value of integrating family health support and living conditions in caries-predictive models. The validity and practical application of these promising findings warrant testing on another pediatric population.

Ethnicity

Several studies have shown highly significant caries differences between racial groups (eg, Clerehugh and Lennon, 1986). In the English city of Coventry, Paul and Bradnock (1986) found the dental health of Asian children to be considerably poorer overall than that of indigenous children. In Sweden, Widstrom and Nilsson (1986) found that the proportion of each immigrant group who visited a dentist was significantly smaller than the corresponding proportion of Swedes, and extractions, endodontic procedures, and dentures were more common in all the immigrant groups.

In Britain, the Dental Strategy Review Group (DHSS, 1981) recommended that the community dental service look to the requirements of "special needs groups." Gelbier and Taylor (1985) stated that young Asian children, and possibly children of other ethnic minorities, are dentally disadvantaged through language, primary socialization,

and the lack of appreciation of minority cultures and needs among the ethnic majority. There is extensive evidence of dietary differences between Asians and other groups within the community, not only resulting from different cultural backgrounds but also associated with social deprivation and communication problems.

In a study of 5-year-old Asian schoolchildren in an area of Britain with multiple deprivation (Bedi, 1989), three distinct dental high-risk groups were identified: (1) children of Muslim, English-speaking mothers; (2) children of Muslim, non-English-speaking mothers; and (3) children of non-Muslim, non-English-speaking mothers. In West Birmingham, where Asian children were shown to have a rate of decayed, missing, or filled teeth nearly twice as high as that of white children, special programs, tailored to meet the needs of special groups, have been recommended (Bradnock et al, 1988).

Apart from language and cultural problems and an often low standard of education among immigrants, emigration disrupts traditional eating habits and leads to exposure to new foods. Studies consistently show that breakfast and snacks, the meals with the least symbolic importance, are the first to change. Therefore, immigrants with poor standards of oral hygiene and an associated irregular use of fluoride toothpaste are at high risk of developing caries when they come to Western countries, and this can partly be attributed to dietary changes.

The role of the parents' immigrant background on caries development in infants and toddlers was recently highlighted in the longitudinal studies by Wendt et al (1994) and Wendt and Birkhed (1995), mentioned earlier. The aim of the initial studies was to describe oral hygiene factors in infants and toddlers living in Sweden, with special reference to caries prevalence at 2 and 3 years of age and to immigrant status. The study was designed as a prospective, longitudinal study starting with 671 children, aged 1 year. At 3 years, all the children were offered a further examination. A total of 298 children, randomly selected from the original group, were also examined at 2 years. The accompanying parent was interviewed about the child's oral health habits.

Compared to the children with caries at age 3 years, the caries-free children had had their teeth brushed more frequently at 1 and 2 years of age, had used fluoride toothpaste more often at 2 years of age, and had a lower prevalence of visible plaque at 1 and 2 years of age. Immigrant children had had their teeth brushed less frequently, had used fluoride toothpaste less often, and had a higher prevalence of visible plaque at 1 year of age than did nonimmigrant children. Seventy-eight percent of 3-year-old nonimmigrant children were caries free, compared to only 50% of the children of immigrant parents. The authors concluded that early establishment of good oral hygiene habits and regular use of fluoride toothpaste seem to be important for achieving good oral health in preschool children. These goals are achieved less commonly in children of parents with an immigrant background (Wendt et al, 1994).

The purpose of the second study was to describe dietary habits in infants and toddlers living in Sweden, with special reference to caries prevalence at 2 and 3 years of age and to immigrant status. The study was designed as a prospective, longitudinal study starting with children aged 1 year. At 3 years, all children were offered a further examination. The accompanying parent was interviewed about the child's dietary habits.

Children with caries at 2 and 3 years of age and immigrant children, at the age of 1 year, had consumed caries-risk products, had been fed at night, and had been bottle-fed with sweet drinks more often than caries-free 2 and 3 year olds and nonimmigrant children. Although a great variation in dietary habits in infants and toddlers was recorded, the use of sugar-containing products is widespread in Sweden even in early childhood (Wendt and Birkhed, 1996). In contrast to many immigrant parents, however, almost all nonimmigrant parents of today's infants and toddlers are educated at least to matriculation level and have had access to regular preventive programs since birth.

With respect to ethnic minorities, the main problems are therefore not the prediction and identification of high risk but the lack of programs tailored to meet their special needs. Few studies of this kind have been conducted in developing countries; these would be of great interest, because the particular parental characteristics associated with children's caries experience are bound to differ in different cultures.

Influence of social and behavioral variables

As discussed earlier in this chapter, the development of dental caries is a complex interaction of etiologic factors and many modifying risk and protective factors. Social factors influence behavior directly related to dental caries, such as oral hygiene, dietary habits, and dental care habits. Besides the influence of social and sociobehavioral factors on, and interaction with, sugar intake, one behavior in particular influences the caries-promoting effect of sugar intake, namely oral hygiene. It is generally accepted that caries occurs only after plaque has accumulated on susceptible tooth surfaces in individuals who eat sugar frequently.

One reason for the difficulty in proving the direct relationship between oral cleanliness and dental caries in point prevalence surveys, as well as in longitudinal retrospective or even prospective studies, is the complex interaction of a number of factors. Several studies have shown an interaction between sugar intake and oral cleanliness. Kleemola-Kujala and Rasanen (1982) found, in a study of 543 Finnish children, a significant relationship between the amount of plaque and dental caries at all levels of sugar consumption. With increasing total sugar consumption, the risk of caries increased significantly only when oral hygiene was also poor. Further analysis showed that the effect estimates for the two factors in combination were always greater than the sums of the separate effects, indicating a synergistic interaction between the two caries determinants.

Granath et al (1976) also found an interaction between oral hygiene and dietary habits, but the significance was low in individuals with low caries prevalence. Rajala et al (1980) found, in a study of male adults, that caries experience was consistently higher for sporadic toothbrushers. Their findings indicated that the positive association between reported daily toothbrushing and low caries experience may be more pronounced in groups with higher overall risk status, for example, in the strata where education and income are low, frequency of dental visits is irregular, use of sucrose is high, and fluoride exposure is low.

However, in a well-controlled longitudinal 3-year study in 12-year-old Brazilians, it

was recently shown that oral hygiene habits could be improved, and caries incidence thereby reduced by more than 50%, even though test and control subjects all lived in an area with fluoridated water and were supplied with fluoride toothpaste once a month (Axelsson et al, 1994).

In most of the aforementioned, the most commonly investigated social factor related to dental caries is social class or socioeconomic status, and the most commonly studied behavioral factors are oral hygiene or dietary habits. This is not surprising, because diet and oral hygiene are the factors most obviously and directly related to caries development. Socioeconomic status also has been recognized for years as one of the main factors influencing equality, or rather inequality, in both general and dental health. Because few studies have addressed the question, little is known of the caries-predictive value of other social and behavioral factors.

However, there is some indirect evidence of a relationship with other social and behavioral factors. In a major questionnaire survey in Scotland, toothbrushing was studied in relation to a number of other health-related behaviors in 4,935 11, 13, and 15 year olds (Schou et al, 1990). Toothbrushing was shown to be significantly related to the subjects' health perception, smoking habits, alcohol consumption, breakfast habits, bedtime, sweet consumption, fruit consumption, and video watching. The minority of children who reported low toothbrushing frequency also reported unfavorable behavior in all the other areas.

Awareness that many health-related behaviors may interact with each other and with other environmental factors to determine individual health outcome has led to the so-called lifestyle approach in health promotion. Dental health factors are seldom included in analyses of the influence of lifestyle factors on health, and conversely, a person's lifestyle is seldom taken into account in studies of determinants and predictors of dental health. In recent analytic epidemiologic studies in adults we found higher caries prevalence in 35- and 50-year-old smokers than nonsmokers (Axelsson et al, 1998). In another randomized study in almost 600 50 to 55 year olds, we found that subjects who seldom or never exercised had significantly greater tooth loss than those who exercised regularly (Axelsson and Paulander, 1994).

The role of social class and educational level of the parents in the dental status of young children has already been discussed. In a randomized analytic epidemiologic study in 35, 50, 65, and 75 year olds in the county of Varmland, Sweden, one of the factors evaluated was the relationship between dental status and educational level. The following clinical data were collected: the percentage of edentulous subjects, the number of remaining teeth, masticatory function according to the modified Eichner Index, prevalence of removable and fixed prostheses, probing attachment level, furcation involvement, Community Periodontal Index of Treatment Needs, caries prevalence (decayed, missing, or filled surfaces and root caries), prevalence of endodontics and apical periodontitis, oral mucosal lesions, and Plaque Index (O'Leary et al, 1972). The subjects also filled in a questionnaire about educational level, other socioeconomic conditions, diseases, use of drugs, body mass index, dental care, and oral hygiene and dietary habits. For evaluation of educational level, the subjects were randomized into elementary school level (low) and more than elementary school level (high) (Axelsson et al, 1990).

Figure 75, from the data collected in 1988, shows the percentage of subjects with low and high educational levels in the four age groups. Among 35 year olds only, 22% had a low level of education, in contrast to 69% and 72% among the 65 and 75 year olds, respectively. However, today almost all 35 year olds are educated to matriculation or tertiary level. For the last 20 years, Sweden has had compulsory education to the end of secondary school (at least 12 to 14 years of education). It is also estimated that, of the current 50 year olds, only 25% have low educational levels, because of the continuous improvement in every cohort of age groups, and the availability of adult education programs.

Figure 76 illustrates the dental care habits among all the subjects related to educational level: Irregular dental attendance is much more common among subjects with low educational levels (82.5%) than among those with higher education (17.5%).

Figure 77, from the 1988 data, shows the percentage of edentulous subjects in the four age groups in relation to educational level: Among the well-educated subjects, except the 75 year olds, edentulousness was extremely rare. However, the percentage of edentulousness has declined dramatically, even among those with lower educational standards. This can be attributed mainly to changes in indications for extraction of teeth and the introduction in 1973 of a national dental insurance scheme that covers all residents of Sweden. Figure 78 shows the mean numbers of teeth (excluding third molars) in persons with low and high educational levels.

Figure 79 shows the percentage of sound and decayed, missing, or filled surfaces in 50 year olds in relation to elementary school (low), secondary school (middle), or tertiary (high) educational level. Subjects with higher levels of education have a greater percentage of intact surfaces and a lower percentage of missing surfaces than do those with less education. However, the percentage of carious surfaces is almost negligible, indicating that the available resources for provision of dental care are adequate, at least with respect to treatment of caries.

The results of this large-scale, analytic, cross-sectional study show that low educational level is a very significant risk indicator for tooth loss, dental caries, and periodontal diseases, not necessarily because highly educated people are more intelligent or wealthier. (In Sweden, there are very limited differences in net income, after tax, between occupational categories such as poorly educated laborers and well-educated teachers.) The difference in dental health status is attributable to the fact that highly educated people know how to learn from written information, to seek information about health promotion, and to apply theoretical information, for example, to self-care.

Fig 75 Age and level of education. - Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.)

Fig 76 Level of education and regularity of dental care visits. - Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.)

Fig 77 Percentage of edentulous patients in relation to level of education. - Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.) Fig 78 Mean number of teeth in relation to level of education. - Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.)

Fig 79 Percentage of intact and DMFSs in 50 year olds in relation to level of education. DSs = decayed surfaces; - Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.)

Conclusions

Introduction

The most important external modifying factors related to dental caries are frequent intake of fermentable carbohydrates and socioeconomic factors.

Dietary factors

The fermentable carbohydrates may be ranked in order of complexity, as monosaccharides (glucose and fructose), disaccharides (sucrose, maltose, and lactose), polysaccharides (glucan, fructan, and mutan) and starch.

If there is undisturbed cariogenic plaque on an accessible tooth surface, intake of any of the fermentable carbohydrates will result in a drop in pH in the plaque and on the underlying tooth surface, where some demineralization may occur (see Fig 2). The most precipitous fall in pH is induced by sucrose, closely followed by glucose, and fructose, while the effect of raw starch is negligible. Sucrose, glucose, and fructose are therefore considered to be highly cariogenic.

"Sugar" (sucrose) is used universally as a sweetener and an inexpensive source of energy. Excluding China and some other developing countries, the average annual consumption is about 50 kg per individual. In Sweden, for example, daily consumption has remained persistently high (about 120 g per individual) for 40 years, although the proportion of indirect consumption, in the form of drinks and sticky sweets, has doubled, increasing from about 30% to more than 60%. Nevertheless, during the same period, a dramatic decrease in caries has been achieved in Sweden.

Experimental studies have shown that, in germ-free animals, frequent intake of sugar does not result in caries (Orland et al, 1954). However, if cariogenic human bacteria (mutans streptococci) are inoculated into the mouth of one animal in a group being fed on fermentable carbohydrates, rampant caries develops in the whole group (Fitzgerald and Keyes, 1960). In other words, dental caries is an infectious, transmissible, but multifactorial disease. Frequent sugar intake is not an etiologic factor, but an external (environmental) modifying risk factor for development of caries on tooth surfaces covered with cariogenic plaque.

Conflicting results are reported from the numerous cross-sectional human clinical studies investigating the correlation between sugar consumption and caries prevalence. Most of the early studies, conducted in populations with high caries prevalence, showed that high intake of sugar-containing products was a significant risk indicator for dental caries. In more recent studies, in populations where caries prevalence is low, because of high standards of oral hygiene and regular use of fluoride toothpaste (for example, in Scandinavia), little or no such correlation has been found, because "clean teeth never decay," and caries prevalence (experience) expresses the cumulative caries incidence (increment), since eruption of the tooth.

A few human longitudinal interventional or observational studies have been designed to evaluate possible correlations between intake of sugar-containing products and caries incidence. Experimental interventional human studies have been carried out in the absence of plaque control and fluoride (Gustavsson et al, 1954; Scheinin and Makinen, 1975; von der Vehr et al, 1970). These early Scandinavian interventional studies in adults demonstrated the following:

1. In the absence of plaque control and fluoride, frequent intake of sugar-containing products is a significant risk factor and prognostic risk factor for dental caries.

2. If sugar is substituted with nonfermentable sweeteners, a significant reduction in caries may be achieved.

Recent longitudinal observations in children, however, have shown little or no correlation between the intake of sugar-containing products and caries incidence.

Because there is a strong correlation between the in vivo fall in the pH of the plaque and demineralization of the underlying tooth surface, the effect on plaque pH of dietary products containing different fermentable carbohydrates has been investigated extensively. The cariogenic outcome of falls in plaque pH is influenced, however, by the concentrations of fluoride, calcium, and phosphate ions in the plaque fluids and by the microbial composition of the plaque.

In vivo plaque pH measurements have shown the following:

1. Plaque pH after rinsing with a sucrose solution is related to plaque age and site. The lowest values are recorded in the maxillary teeth and on the most central part of the approximal surfaces of molars. The pH drops below 5 in interdental plaque more than 3 days old but not in less mature plaque. In a toothbrushing population, interdental plaque more than 3 days old, if present at all, should be located only between molars and premolars.

2. Of the fermentable carbohydrates, the lowest plaque pH is induced by sucrose, closely followed by glucose, fructose, and maltose. The fall in pH associated with lactose and cooked starch (to pH 5.5 to 6.0) is not as severe but is critical for initiation of root caries.

3. Plaque pH is related to the sugar concentration. Rinsing with even a weak sucrose solution (2.5% to 5.0%) results in a suboptimal pH drop (below 5) in interdental plaque that is more than 3 days old. The optimal pH drop occurs with 10% sucrose

solution; concentrations greater than 10% do not further depress plaque pH. Many dietary products, such as mustard, ketchup, salad dressing, soft drinks, and ice cream, contain 8% to 13% sucrose. While it is therefore unrealistic to exclude all products containing more than 2% sucrose, the daily number of intakes should be restricted.

4. Plaque pH is correlated not only to the sugar concentration of the product but also to the consistency (texture) and the pattern of consumption. For example, eating cheese directly after sugar-containing products will rapidly raise the plaque pH, in contrast to pasty bananas or sugary desserts. A habit that could be recommended for caries prevention is a combination of the Southern European custom of finishing a meal with a cheese platter followed by the new Scandinavian recommendation of using sugarless fluoride chewing gum after meals.

5. Neither caloric sugar substitutes (sorbitol, xylitol, lycasine, and sorbose) nor noncaloric sugar substitutes (saccharin, cyclamate, aspartame, etc) induce critical falls in plaque pH, even to levels critical for root caries development. While these are now widely used as sweeteners in products frequently consumed between meals, it is unrealistic, nutritionally and economically, to recommend sugar substitutes in food consumed mainly at mealtimes.

Clinical cross-sectional studies, longitudinal interventional and observational studies in humans, and animal experiments, as well as in vivo plaque pH measurements indicate a synergistic cariogenic effect of dental plaque and fermentable carbohydrates (particularly sucrose) on plaque-covered tooth surfaces.

Evaluation of dietary habits is important, particularly in caries-susceptible individuals. Because caries is a multifactorial disease, dietary data complement clinical and case history data used to compile the patient's riskprofile (see chapter 4). The most common methods for evaluation of dietary habits in relation to dental caries are the dietary history and the 24-hour recall. Emphasis is on the frequency of intake of sticky, sugar-containing products, which prolong sugar clearance time.

Dietary recommendations for caries control, while emphasizing noncariogenic or low-cariogenic food habits, should also meet nutritional requirements and recommendations for general health: fortunately, a healthy diet is not cariogenic. The diet recommended for diabetics is in general agreement with such recommendations: a high intake of fresh vegetables and fruits, carbohydrate intake from starch instead of sucrose, and a low intake of fat.

For caries prevention and control, there are five major dietary recommendations:

1. Breakfast should be a balanced composition of dairy products, grains, and fruits.

2. The total daily number of intakes, including snacks, should be limited to about four.

3. Sticky sugar-containing products, which prolong sugar clearance time, should be eliminated. Sugarless sweets and soft drinks are available as substitutes.

4. Each meal should include fiber-rich products, which stimulate chewing and salivary flow. Cheese is recommended at the end of the meal.

5. Certain caries-susceptible individuals, particularly subjects with reduced salivary flow, should use sugarless fluoride chewing gum for 20 minutes after every meal.

In future, refinement of the intraoral wire telemetric and the microtouch methods for in vivo plaque pH measurements is expected. This will allow a more systematic classification of the cariogenicity of food, for example, a scoring system from 1 to 5. Several years ago, a similar system for sweets, assessed by the intraoral wire telemetric method, was introduced in Switzerland. For ethical reasons, human interventional longitudinal clinical studies are no longer allowed. Therefore, in vivo plaque pH measurement is the only available method for evaluation of cariogenicity of dietary products in humans.

Further improvements may also be expected in sugar substitutes, noncaloric as well as caloric, with respect to taste and side effects. Use of such sweeteners will become more widespread in snack foods such as sweets, confectionery, and soft drinks.

A concerted effort should be made to prevent infants from acquiring a taste for sweet foods. Animal experiments have shown that, by frequent intake of sucrose during pregnancy, a sweet taste can be acquired prenatally. The studies by Wendt and Birkhed (1996) clearly showed that bottle-feeding of sweet drinks, particularly at night, resulted in significantly increased caries development from the age of 1 to 3 years.

Finally, how do international experts perceive the past and present relationship between dietary sugar and dental caries? Bratthall et al (1996), in a recent questionnaire, sought the opinions of 55 international experts on dental caries and preventive dentistry as to the main reasons for the caries decline in many Western countries during the last three decades, specifically in 20 to 25 year olds. The respondents were asked to rank reduced sugar consumption, reduced sugar frequency, fluoride toothpaste, school fluoride programs, reduced amount of plaque, and fissure sealants. The respondents ranked these in the following order:

1. Fluoride toothpaste

2. Reduced amount of plaque

3. School fluoride programs

4. Reduced sugar frequency

5. Fissure sealants

6. Reduced sugar consumption

However, it should be noted that fewer than 10% of people worldwide use fluoride toothpaste, and the Western industrialized countries represent 30% to 40% of the world's population. In addition, fewer than 1% of school-aged children worldwide have access to school-based fluoride programs.

Socioeconomic and behavioral factors

Early establishment of good oral hygiene and dietary habits and regular use of fluoride toothpaste are of utmost importance. Several studies in infants and toddlers have clearly shown that such habits, as well as dental status, are strongly correlated to the parents' social class (particularly educational level), dental status, regularity of dental care (particularly preventive programs), and ethnic background (immigrants). Organized oral health education programs at maternal and child welfare centers are therefore important strategies for reducing such inequalities. In particular, especially disadvantaged parents, such as some immigrant groups, should be identified and offered special oral health promotion programs tailored to their ethnic background, language, culture, dietary customs, oral hygiene habits, and educational level.

It has also been shown that the socioeconomic and educational level of the parents is much more significantly related to caries incidence in children than, for example, the frequency of intake of sugar-containing products. On the other hand, health-related behavior that influences dental caries development, eg, dietary, oral hygiene and dental care habits, and the use of fluorides, is strongly correlated to parental socioeconomic class and particularly to educational level. Prediction of caries risk in early childhood and in schoolchildren might therefore be improved by combining data on behavioral and social factors with clinical examination, rather than analysis of behavioral or parental social variables only.

Social class, oral hygiene and dietary habits, and the use of fluorides are the variables conventionally related to caries prevalence. However, many other social and behavioral variables may also influence oral health status. The role of parental educational level on children's dental health status has already been discussed. Even more important is the role of educational level on oral health status in the adult population. Generally, the trend in the industrialized countries is toward an acceleration in the percentage of well-educated adults, particularly among 20 to 50 year olds. There is increasing exposure to information and education about self-care, self-diagnosis, and so on from departments of health, oral health personnel, the media, and others. Such conditions favor a positive outcome for oral health promotion and a consequent improvement in oral health status in all age groups.

Other behavioral factors that have also been shown to correlate with oral health status are socalled lifestyle behaviors, such as smoking habits, regular or irregular exercise, and a vegetarian diet.

Conflicting results have been reported from studies of caries in mentally and physically handicapped people: Although prevalence is often no greater and sometimes lower than in normal children or adults, more of the caries present in handicapped people remains untreated, and more teeth are extracted. For mentally retarded children, the most important determinant of caries risk is the poor standard of oral hygiene. A mental or physical handicap does not in itself seem to be a predictor of high risk, but handicapped people need special care, and this is not always as readily available as is routine care for the nonhandicapped population.

Multivariate predictive methods are superior to single analysis of any social and behavioral variables. Models that include not only sociologic and behavioral but also

clinical variables are superior to those based only on sociologic or epidemiologic variables. However, despite the relatively high sensitivity and specificity of models, few studies have analyzed their practical application.

The decision to initiate high-risk programs is not merely an academic question: the impact, politically and philosophically, is of far greater consequence. If a philosophy of equality in resource allocation prevails, equality in health may never be achieved. The high-risk strategy will probably require unequal allocation of resources to achieve equality in health. It is hoped that application of current knowledge and the results of ongoing research about prediction of risk groups and risk individuals will help to advance equality in health. It is, however, more difficult to predict caries risk at that individual level than to identify groups in the population at high caries risk. Social and behavioral markers, although not perfect, are the best available markers for identification of groups but less satisfactory at the individual level. Based on current knowledge of dental disease patterns, public dental health strategies should specifically target those in need, rather than the whole population, irrespective of need.

Chapter 3. Internal Modifying Factors Involved in Dental Caries

Introduction

As discussed in chapter 2, many factors modify the prevalence, onset, and progression of dental caries. The major internal (endogenous) modifying risk indicators, risk factors, and prognostic risk factors related to dental caries are reduced salivary secretion rate (SSR), poor salivary quality, impaired host factors, chronic diseases, unfavorable macroanatomy and microanatomy of the teeth, and the stage of eruption, all of which favor plaque retention, poor quality and maturation of enamel, and exposed root cementum or dentin. Impaired salivary function, particularly an inadequate SSR, is of utmost importance.

Role of Saliva

Introduction

The secretion rate and quality of saliva are important not only in caries development but also for remineralization.

Function of saliva

Saliva serves as a first line of both nonspecific and specific defense in the oral cavity against infectious diseases, erosion, attrition, and traumatic lesions of the oral mucosa.

Saliva is vital to the integrity of the mineralized tissues (teeth) as well as the soft tissues; to the selection, ingestion, and preparation of food for digestion; and to the ability to communicate. Maintenance of the integrity of the oral tissues is primarily a function of the unstimulated (resting) basal secretions; the functions related to digestion are served by salivary flow stimulated by the intake of food.

Saliva has manifold functions in protecting the integrity of the oral cavity from food residue, debris, and bacteria:

1. Saliva has some buffering effect against strong acids and bases.

2. Saliva provides the ions needed to remineralize the teeth.

3. Saliva has antibacterial, antifungal, and antiviral capacities.

Components of saliva also facilitate the motor functions of chewing, swallowing, and speaking, as well as sensory and chemosensory functions in the oral cavity. These functions are summarized in Table 12.

Secretion of saliva

The normal daily volume produced by the salivary glands is about 0.5 to 1.0 L, of which only about 2% to 10% is produced during sleep. About 80%, stimulated by chewing, is produced during meals; that is, the mechanisms of salivary production are capable of rapid response to physiologic demand. About 90% of the total volume is produced by three major pairs of symmetrically located salivary glands: glandulae parotidea, glandulae sublingualis, and glandulae submandibularis (Fig 80).

In humans, salivary glands are classified, according to the nature of their secretion, as serous, mucous, or mixed. Serous glands, for example, the parotids, produce a thin, watery secretion rich in enzymes. Mucous glands, for example, the minor glands of the soft palate, produce a viscid secretion. In mixed salivary glands, such as the submandibular and sublingual glands, the secretory product varies, depending on the proportion of mucous to serous cells within the gland. The submandibular glands are mainly serous, and the sublingual glands are mainly mucous.

Salivary glands can also be classified as simple or compound, according to their duct system. The glands comprise mainly ducts and acini (Fig 81). The duct system of the submandibular and parotid glands is well developed and branched, containing intercalated, striated, and excretory ducts. In sublingual glands, the intercalated and striated ducts are sparsely distributed. The minor salivary glands are classified as simple branched tubular glands.

Of the major salivary glands (see Fig 80), the parotids are the largest, weighing 20 to 30 g each. The parotid duct (Stensen's duct) is about 5 cm long and opens into the oral cavity opposite the buccal surface of the maxillary second molar (the parotid papilla). The submandibular glands are smaller than the parotids and are surrounded by a well-defined capsule. The main duct (Wharton's duct) is about 5 cm long and opens at the summit of the sublingual papilla just lateral to the frenulum of the tongue. The sublingual gland is composed of several smaller glands; the main duct (Bartholin's duct) opens close to the duct of the submandibular gland.

The minor salivary glands, numbering between 200 and 400, produce about 10% of the total volume of saliva. They occur throughout the oral mucosa, with the exception of the gingivae and anterior part of the hard palate. They are named, according to their

location, as labial, buccal, palatine, lingual, glossopalatine, and minor sublingual glands.

Salivary secretion from major and minor glands is controlled by both parasympathetic and sympathetic stimuli. Depending on the nature of the stimulus, this also affects the composition of saliva. In general, parasympathetic stimuli increase the output of water and electrolytes, whereas sympathetic stimuli enhance protein synthesis and secretion. Clinically, this difference may have relevance, because both the volume of fluid and the concentration and nature of salivary proteins are important for protection against microbial diseases, such as dental caries: In the clearance process, the water-electrolyte fraction is important, and the actual antimicrobial activity is determined by the protein fraction.

Saliva is secreted in response to neurotransmitter stimuli. For most of the day, neurotransmitter release is low and salivary flow is basal, or unstimulated. During food ingestion, in response to gustatory and masticatory stimuli (via mechanical stimulation of the nerves in the periodontal ligaments), there is a pronounced increase in neurotransmitter release, and secretion is stimulated. Resting secretion is considered to be mainly protective, while the larger volume of stimulated saliva is needed to facilitate ingestion (formation and swallowing of a food bolus) and communication. The bulk of the stimulated saliva is secreted by the parotid gland, which is estimated to contribute about 10% of unstimulated and more than 50% of stimulated whole saliva.

Salivary secretion rate (SSR)

"Normal" values and thresholds. Of the many studies of SSRs in presumably healthy individuals in different countries, the most remarkable finding is the enormous variability; the SSR ranges between 0.08 and 1.83 mL/min, a 23-fold range, for resting whole saliva and between 0.2 and 5.7 mL/min, an almost 30-fold range, for stimulated saliva. Throughout these vast ranges, individuals are generally free of subjective complaints and objective signs of salivary gland dysfunction. These studies show that normal oral function can be maintained with wide individual ranges in saliva production.

Because of this heterogeneity, it is difficult to assess the status of a patient's salivary gland function from a single measurement of SSR. In the absence of complaints or signs, it is difficult to determine the presence of a salivary gland disorder. Furthermore, it is evident that caution should be exercised in comparing a single SSR with a population standard. Changes in a patient's SSR over time are probably a more reliable indicator of oral health. If clinicians routinely assessed saliva production in all their patients, they would be able to establish a patient`s normal SSR and be alert to any decline. This would allow early intervention to prevent or limit the deleterious consequences of salivary gland dysfunction.

"Whole saliva," the fluid present in the mouth, comprises not only pure secretions from the major and minor salivary glands but also gingival exudate, microorganisms and their products, epithelial cells, food debris, and, to some extent, nasal exudate. Whole saliva is of clinical relevance for susceptibility to caries and carious activity.

However, there is no linear association between SSR and carious activity, but rather a "threshold effect." Although, for clinical purposes, there is consensus on the thresholds presented in Table 13, this is clearly an oversimplification, particularly at the individual level.

The so-called normal values for unstimulated and stimulated SSR exhibit a large biologic variation, which should be considered in relation to sex, body weight, and age. For example, 3- to 4-year-old children, with limited experience of different tastes, seem to have an extremely high SSR per kilogram of body weight, about five times as high as that of 10 year olds. On the other hand, in healthy adults, there is only a limited decline in stimulated SSR with age.

Studies by Heintze et al (1983) established the gender-related ranges of unstimulated whole SSR (Fig 82) and stimulated SSR (Fig 83) in adults, with peaks at about 0.3 to 0.4 mL/min and 1.5 mL/min, respectively. However, the secretion rates for both unstimulated and stimulated saliva were significantly lower for females than for males. In most studies, the reported mean value for stimulated SSR is about 1.5 mL/min in females and 2.0 mL/min in males; the difference is attributed mainly to the greater body weight of males. When the secretion rate is evaluated, body weight (reflecting glandular size) should be taken into account. Although Heintze et al (1983) found a significant correlation between unstimulated (resting) and stimulated SSRs, the individual variations were so large that one type of SSR could not readily be predicted from the other.

A study by Percival et al (1994) compared the SSR of unstimulated whole saliva and stimulated saliva from the parotid gland in relation to age and gender in "healthy" adults (without medication). The mean values were lower in females than in males. However, while the unstimulated whole SSR (Fig 84) was significantly lower in the older age groups (80 or older) than in the younger groups (20 to 39 years), there was no correspondingly significant difference for the stimulated parotid SSR (Fig 85).

Randomized samples of adults will, however, include both healthy and unhealthy individuals. Particularly among the elderly, there is widespread regular use of pharmaceuticals that have systemic depressive effects on SSR as well as the quality of the saliva. Loss of teeth is also strictly age related; because of total or partial edentulousness, chewing stimulation is reduced in a relatively high percentage of the elderly (about 20% to 50% of 65 to 90 year olds).

In a randomized sample of about 1,000 50, 65, and 75 year olds, one of many clinical and anamnestic variables evaluated was stimulated whole SSR (Axelsson et al, 1990). Figure 86 shows the percentage of individuals with 0.0 to 0.7 mL/min, 0.8 to 1.4 mL/min, and more than 1.5 mL/min in the three age groups.

Link to caries risk. The relationship between stimulated SSR and the development of carious lesions has been studied extensively. Although caries risk is extreme in the absence of saliva or in the presence of very low secretion rates, there does not seem to be a strictly linear correlation. An inverse relationship between stimulated SSR and caries incidence, for both enamel and root caries, is found in most studies, and statistical significance has also been demonstrated in some cross-sectional investigations. Stimulated SSR values of less than 0.7 mL/min are regarded as a

threshold for considerably increased risk of further caries development. Therefore, it is interesting that in randomized samples of 50, 65, and 75 year olds, such low values were recorded in as many as 15%, 20%, and 25%, respectively, about one subject in five over the age of 50 years (see Fig 86) (Axelsson et al, 1990).

However, SSR cannot be assessed qualitatively. As the most important clinical variable of saliva affecting susceptibility to dental caries, simple quantitative assessment of stimulated whole saliva should be a routine clinical procedure in the adult population. The same saliva sample can also be used to measure salivary buffering capacity and the levels of salivary mutans streptococci and lactobacilli.

In clinical practice, measurement of saliva (sialometry) is particularly indicated:

1. As part of the initial examination of a new patient to be treated for dental caries.

2. During evaluation of preventive and restorative treatment of dental caries, to assess how the overall treatment has affected oral health.

3. In elderly patients who take regular medication, and/or have exposed root surfaces.

4. As part of the investigative procedures for suspected hyposalivation associated with, for example, regular use of medicines with systemic depressive effects on SSR, Sjogren's syndrome and other diseases associated with reduced SSR, or irradiation to the head and neck region.

The data gathered by Axelsson et al (1990) were used to analyze the relationship between SSR and dental health. Figure 87 shows the mean values of stimulated whole SSR in dentate and edentulous individuals. Figure 88 shows the mean stimulated whole SSRs related to sex, regular medication, and edentulous versus dentate. Figure 89 shows the mean number of teeth (third molars excluded) in 50, 65, and 75 year olds with low and high stimulated SSRs (0.0 to 0.7 and more than 1.5 mL/min, respectively). The results indicated that the stimulated SSR values may influence the number of teeth lost. Figure 90, from the same study (Axelsson et al, 1990), shows the percentage of intact, decayed, filled, and missing surfaces among 50-, 65-, and 75-year-old dentate individuals and all individuals with low and high stimulated SSRs.

In a more recent cross-sectional study of a randomized sample of more than 600 50 to 55 year olds, among many clinical and anamnestic variables caries prevalence was related to stimulated whole SSR and regular use of medicines with known systemic effects on SSR (Axelsson and Paulander, 1994). Twenty-nine percent of the subjects were taking medication regularly, and 22% used medicines that impair SSR. Figure 91 compares the frequency distribution of intact, decayed, missing, and filled surfaces in subjects with a stimulated SSR of less than 0.7 mL/min, versus subjects with an SSR of greater than 1.5 mL/min, and in subjects using drugs that impair salivary function versus subjects not taking any medication.

These data show conclusively that the SSR is an important factor in caries severity and should be considered when caries risk is assessed. Very low stimulated SSR (hyposialosis) (less than 0.7 mL/min, and particularly less than 0.4 mL/min) results in a high risk of caries. Clinically, it is therefore important to determine whether SSR is

normal or impaired.

Symptoms of salivary gland hypofunction resulting in hyposalivation

Apart from an increased susceptibility to caries, other oral and systemic disturbances may also be associated with hyposalivation (Box 7).

Hyposalivation, or reduced SSR, is not synonymous with xerostomia, which is a symptom reflecting the end result of the process of inflow of pure saliva, evaporation, adsorption to the oral mucosa, and outflow of saliva. Of the saliva that enters the mouth, as much as 0.20 to 0.25 mL/min may evaporate, causing a sensation of dryness, especially in mouth breathers. Smokers may also experience dryness. A study on dental health status related to smoking habits found that, although smokers had higher mean values for stimulated SSR than did nonsmokers, significantly more smokers reported symptoms of dry mouth (Axelsson et al, 1998).

Experiments by Dawes (1987) showed that subjects experienced a feeling of dry mouth when their normal SSR was temporarily reduced by 50% (eg, a reduction in stimulated SSR from 2 to 1 mL/min or in unstimulated SSR from 0.4 to 0.2 mL/min). Even healthy individuals with "normal" salivary flow rates can experience symptoms of dry mouth: Studies have reported that 20% of 30 year olds and 50% of 55 year olds are so discomforted by dry mouth that they resort to salivary stimulation or rinsing.

The sensation of dryness is usually attributable to hypofunction of the minor salivary glands, which produce a mucin-rich, high-viscosity secretion, rather than to hypofunction of the major glands. Because there is little, if any, relationship between subjective complaints of xerostomia and actual quantitative salivary flow, it is important to measure the SSR in each individual patient. The data in Table 13 show that proper individual diagnosis is almost impossible, although on a population level these numbers are relatively reliable. To assess susceptibility to caries or carious activity, the SSR should be monitored regularly in individual patients and not assessed as "normal" or "abnormal" on the basis of just one measurement.

A very low SSR, particularly for unstimulated saliva, results in clinical changes in the oral cavity (Box 8). The most conspicuous feature of salivary gland hypofunction is dryness of the lining oral tissues. The oral mucosa may appear thin and pale, lose its glossy sheen, and feel dry: a tongue blade or mirror drawn across the surface may adhere. Such dry, thin mucosa can also be diagnosed by optical measurements with infrared light (Fig 92) or by mechanical friction measurements (Fig 93).

Other clinical changes are increases in dental caries; oral infections, especially candidiasis; fissuring and lobulation of the dorsum of the tongue and occasionally the lips; angular cheilosis (Figs 94 and 95); and occasional swelling of the salivary glands. Milking of the salivary glands may not yield any saliva. New carious lesions are common, and develop rapidlywithin weeks or months rather than yearsand often at atypical sites: the mandibular anterior teeth, at the cervical margins of recent restorations (Fig 96), and on the incisal edges.

Candidiasis may appear as smooth red patches or as a diffuse area of intense redness (the erythematous or atrophic form); as white-to-ecru, removable plaques (the

pseudomembranous form or thrush); or as white plaques that cannot be removed by scraping (the hyperplastic form). These lesions often appear on the dorsum of the tongue and the palate. The presence of Candida on the mucosal surfaces and in the saliva can readily be determined by a simple dip-slide test.

Patients with xerostomia may also have a wide variety of nonoral clinical signs (see also Box 7). Ocular changes include xerophthalmia, keratoconjunctivitis, decreased lacrimation, and the accumulation of viscous secretions in the conjunctival sac. Involvement of the exocrine glands may lead to pharyngitis and laryngitis, persistent hoarseness, a dry cough, and difficulty with speech. Nasal dryness may induce scab formation, epistaxis, and loss of olfactory acuity. A decrease in the production of saliva, as well as in secretions from the gastrointestinal tract, may lead to reflux esophagitis, heartburn, and constipation.

Causes of hyposalivation and xerostomia

The salivary glands derive their fluid from the circulating blood. This fluid, with its electrolytes and small organic molecules, is modified by the glands and, together with the macromolecules synthesized by the gland cells, secreted into the oral cavity (see Figs 80 and 81). Secretion occurs in response to neural stimulation. Disturbances of the blood supply to the gland, of its secretory apparatus, or of the stimuli that elicit secretion may lead to a decrease in the production of saliva.

As mentioned earlier, a person experiences oral dryness when the volume of saliva decreases to about half the normal flow rate; in xerostomia, the most extreme form of dry mouth, the decrease is significantly greater. For the resting flow of saliva to fall to such a level, more than one gland must be affected. The loss of activity of a single gland, observed in patients with salivary gland tumors and of sialoliths, does not result in oral dryness. Thus, xerostomia is the result of multiglandular salivary hypofunction, frequently as a result of the intake of xerogenic drugs, therapeutic irradiation, or certain systemic conditions. Age and decreased mastication may also contribute to the feeling of oral dryness. The most common causes of salivary gland hypofuction and xerostomia are listed in Box 9.

Medicines. The most common cause of hyposalivation and xerostomia is the use of xerogenic drugs. It is estimated that more than 400 drugs, some commonly used, can cause oral dryness and induce salivary gland hypofunction. These include anticholinergics, anorectics, antihistamines, antidepressants, antipsychotics, antihypertensives, diuretics, and antiparkinsonian drugs (Box 10). Many physicians are still unaware of these side effects, and patients are not informed about the increased risk of caries.

The dentist should therefore always ask for detailed information about the patient's medication. When the medication has a known saliva-inhibiting effect, the physician should be consulted: Although management of systemic disease has first priority, and medication should not be changed for dental reasons alone, the physician may be able to suggest an alternative drug or modify the dose.

Irradiation. Patients irradiated for the treatment of oral, head, and neck carcinomas frequently experience severe hyposalivation (and even the absence of salivation),

xerostomia, mucositis, and dysgeusia. The effects of the radiation are dose, time, and gland dependent. As far as possible, radiation oncologists shield the glands from the full dose of radiation, but, when bilateral exposure of the salivary glands is unavoidable, xerostomia may be permanent.

Patients often experience oral dryness at an early stage of treatment, and this is exacerbated as the therapy proceeds. In one study, a 50% reduction in the resting flow rate of parotid saliva was recorded 24 hours after the administration of only 2.25 Gy; after 6 weeks of treatment (60.00 Gy/fraction), the flow rate had declined by more than 75% (Sreebny et al, 1992). In most patients, the disturbance of salivary gland function and the associated xerostomia are irreversible. A more than 95% reduction in salivary secretion has been found to persist 3 years after treatment.

The mechanisms underlying the acute effects of radiation on salivary function are not known; the early effects may result from damage to the blood supply or interference with transmission of nerve impulses. The later effects are due to destruction of the secretory apparatus and its subsequent replacement by fibrous connective tissue and to specific vascular damage (endarteritis). The secretory cells, the blood supply and the nerves may all be affected by ionizing radiation. Serous cells are more sensitive to radiation than are mucous cells: the parotid gland, which contributes the bulk of the serous component of saliva, is therefore most vulnerable to damage, while the minor salivary glands may still function normally. Thus there is not only a pronounced reduction in salivary flow rate but also a change in salivary composition. The saliva becomes a viscous, white, yellow, or brownish fluid with a reduced pH, reduced buffering capacity, and altered electrolyte and protein content (see Fig 94).

Among the irradiation-induced changes in the mouth is also a pronounced increase in the numbers of acidogenic, cariogenic microorganisms at the expense of noncariogenic bacteria. Clinically, the most significant changes are increases in mutans streptococci, lactobacilli, and Candida species. The quantitative and qualitative salivary changes predispose the irradiated patient to a variety of oral problems, such as extreme xerostomia and all its symptoms, listed in Box 7, and rapid and extensive dental caries (see Fig 96), unless intensive caries-preventive measures are stringently followed. In addition to having rapid onset and progression, radiation caries characteristically occurs at sites normally relatively resistant to dental caries (lingual and incisal surfaces). The areas just below the contact points, usually the sites most susceptible to caries, are often the last to be affected.

Because therapeutic doses of radiation cause no direct damage to the tooth structure, the associated enormous increase in the cariogenic challenge is attributable to hyposalivation-related alterations in microbial, chemical, immunologic, and dietary variables. The salivary glands are usually located within the treatment portals for head and neck cancer, and, at present, there is no proper clinically acceptable radioprotection. Treatment of the resulting severe hyposalivation is therefore partly palliative and partly directed toward caries prevention: mechanical plaque control, use of antimicrobial and slow-release fluoride agents, stimulation of residual salivary gland tissue by masticatory and gustatory stimuli, such as fluoride chewing gums, and symptomatic relief of oral dryness with artificial saliva that contains fluoride. These measures are not limited to irradiated patients but are appropriate for all patients with severe hyposalivation and xerostomia.

Systemic diseases. Systemic diseases, and the drugs used in their management, frequently cause a marked reduction in salivary secretion. Xerostomia and salivary gland hypofunction are intimately associated with a number of systemic diseases and conditions, some of which cause progressive destruction, usually irreversible, of the gland parenchyma. Others may have vascular or neural effects that are transient and reversible. Included among the diseases are rheumatoid conditions (sometimes referred to as collagen-vascular, connective tissue, or autoimmune disorders); hyposecretory states; certain common diseases (eg, hypertension and diabetes mellitus); cystic fibrosis; certain neurologic conditions; depression and dehydration, anorexia nervosa, and hormonal changes (see Box 9).

The classic example of the rheumatoid conditions is Sjogren's syndrome. The primary form is characterized by salivary and lacrimal gland involvement, usually presenting as dry mouth and dry eyes. The secondary form involves at least one of these organs and, in addition, is an associated collagen disorder, most commonly rheumatoid arthritis. Systemic lupus erythematosus, scleroderma, dermatomyositis, and primary biliary cirrhosis may also be associated with secondary Sjogren's syndrome.

In the early stages, there may be little change in the SSR; as the disease progresses, there is a corresponding decrease in SSR, resulting from the gradual destruction of the gland parenchyma by a lymphoreticular cell infiltrate. There is massive irreversible acinar cell degeneration and atrophy. Changes resulting from Sjogren's syndrome are not restricted to the mouth and eyes; there may be extraglandular manifestations in the gastrointestinal, renal, genitourinary, and pulmonary systems. The condition is associated with increased risk of pseudolymphoma and malignant lymphoreticular disease.

The diagnosis of Sjogren's syndrome is usually made several years after onset, and, in many patients, the dentition is severely damaged before the definitive diagnosis is made. Meticulous assessment of salivary secretion rate by the dentist is an aid to early identification of these patients. The presence of Sjogren's syndrome is often confirmed by biopsy of the minor salivary glands of the lower lip.

Sjogren's syndrome is believed to occur in about 1% of the population. It affects eight times as many women as men, most above the age of 45 years. Although it is the cause of profound distress in older women, it is a condition that has been largely overlooked, and surprisingly little information is available in the medical and dental literature.

There is increasing evidence that xerostomia is associated with a number of common disorders and diseases such as hypertension and diabetes mellitus. The evidence for the link between diabetes mellitus and xerostomia is of two types. First, there is a greater incidence of diabetes mellitus in xerostomic subjects than in nonxerostomic controls. Second, diabetic subjects with no other diseases and taking no drugs other than insulin have a far greater prevalence of xerostomia than do matched nondiabetic controls (Sreebny et al, 1992). However, insulin-dependent (type 1) diabetes mellitus, as such, does not damage salivary glands to such an extent that hyposalivation can be regarded as a common complication. Reduced salivary secretion is typical only during periods of diabetic instability or during the onset of the disease: During these periods,

increased glucose levels in salivary secretions are common, heightening caries risk. This is one reason why such diabetic patients are at least as caries susceptible as nondiabetic patients.

Other conditions associated with reduced SSR. The changes in SSR associated with chronic depression are generally persistent. When oral dryness cannot be attributed to organic change, the patient should be advised to consult a psychologist or psychiatrist to explore possible psychogenic factors. Although psychic states can induce oral dryness, the underlying mechanisms are not well understood. Depression is frequently treated with tricyclic antidepressants, which tend to aggravate the severity of oral dryness.

In patients suffering from severe or prolonged malnutrition or anorexia nervosa, deterioration in SSR and the quality of the saliva may lead to oral symptoms, for example, increased susceptibility to dental caries, erosion, and dry mouth. Although short-term fasting also reduces the SSR significantly, it does not lead to true hyposalivation, and the flow rate is restored to normal values soon after the fasting period ends.

Hormonal changes may also affect the SSR and the composition of human saliva. The most profound changes are postmenopausal, many studies confirming that the salivary secretion rate is lower in postmenopausal women than in younger women. However, there is a widely varying individual response: Some postmenopausal women experience no detectable change in SSR, while others experience distressing oral dryness (with a number of symptoms, such as "burning mouth," sore tongue, difficulties with speech and swallowing, and fungal infections). Estrogen supplementation therapy has little, if any, effect on SSR after menopause.

Age as such does not have a clinically obvious effect on SSR: In healthy individuals, the stimulated whole SSR does not decline. Age-related decreases in secretions from both minor and submandibular glands have been observed, similar decreases in SSR from the parotid glands have not been found. [au: Reference?]Comparative studies of stimulated minor gland secretion in older and younger adults have reported an age-related decline of more than 50% in secretion, a functional decrease consistent with morphologic studies showing a 40% to 50% reduction in acinar volume with age (Percival et al, 1994).

These physiologic changes may explain why many old people experience discomfort from dry mouth, even though the secretion rate for paraffin-stimulated whole saliva is normal. Alterations in submandibular and sublingual gland functions have the greatest impact on the sensation of oral dryness. These changes will also increase the risk for development of denture stomatitis and reduced denture retention in edentulous elderly.

Both human and animal studies (Dawes, 1987) have shown salivary gland atrophy to be associated with decreased masticatory activity. In humans, this has been reported in subjects on liquid diets and in patients whose jaws were wired together after orthognathic surgery. The extent to which varying degrees of decreased masticatory activity contribute to salivary gland hypofunction and xerostomia in humans is not known. On the other hand, as will be discussed later in this chapter, studies have

shown that introduction of agents to stimulate chewing may increase the SSR in patients with hyposalivation (Axelsson et al, 1997a).

Evaluation of hyposalivation

Based on the principle that "status is determined by clinical examination but explained by the case history," the following points should be considered for proper diagnosis of hyposalivation:

1. Stimulated salivary secretion rate

2. Resting SSR

3. Anamnestic data: possible side effects of medication; systemic diseases known to cause salivary gland hypofunction; difficulty in swallowing dry food; difficulty in speaking; soreness of the oral mucosa; frequent episodes of sore throat; difficulty in tolerating removable dentures

4. Tenderness of salivary glands to palpation or swelling of the glands

5. Inflammatory changes in the oral mucosa or tongue

6. Indicative test: Does the examination mirror stick to the cheek mucosa?

7. Atypical pattern of caries (lesions on smooth surfaces or on the tips of incisors or cusps)

If most of these above features are present and the resting SSR is low, the diagnosis almost certainly includes hyposalivation, and the patient should be considered at high risk of developing caries. Apart from measurement of SSR, the other anamnestic and clinical variables have already been discussed in detail. The importance of repeated measurements of SSR in the individual patient, and particularly in caries-susceptible individuals, cannot be overemphasized.

There are noninvasive, painless techniques for sampling not only whole saliva but also saliva from the individual major and minor salivary glands. Whole saliva is easily obtained and in most cases a good indicator of whole-mouth dryness. Disease in a major salivary gland can often be diagnosed from secretion collected directly from the gland.

The purpose and method of the collection procedures should be explained to the patient beforehand. Saliva should be collected about 1 1/2 to 2 hours after eating or after an overnight fast. Patients should be instructed to do nothing that might stimulate the SSR prior to the collection, including mastication of food, chewing gum, or candy; smoking; toothbrushing; mouthwashing; or drinking. The test should be conducted in quiet surroundings. Detailed instructions for standardized measurements of SSR are shown in Box 11.

To obtain a mean SSR, sampling should be repeated at least once, at about the same time of day. If the patient's baseline has been established earlier, the values obtained

can be used as a comparative indicator of the patient's present salivary status. If the baseline is not known, as is usually the case, the SSR is compared with the relevant standards for the population (see Table 13). As with any test, the results should be interpreted in light of the patient's history, the presence of any signs of disease, and the results of other tests.

Whole saliva may be collected and measured by a variety of volumetric and gravimetric techniques: draining (drooling), spitting, suction, and swab. The volumetric methods to be described, especially a combination of the drooling and spitting techniques, are easily carried out in the dental or medical office.

Either of two measuring devices may be used: a Sialometer or any finely calibrated measuring cylinder. The Sialometer is a specially constructed, reusable device that allows collection of both resting and stimulated saliva in a single vessel. Alternatively, the following equipment can be obtained from chemical supply houses: two funnels and two measuring cylinders, each with a volume of about 12 mL, calibrated to no less than 0.1 mL.

Measurement of stimulated whole SSR. In the clinic, the usual procedure is to measure SSR during masticatory stimulation (ie, while the patient chews a piece of paraffin). To achieve reliable, standardized results, the patient should be given detailed instructions (see Box 11). The patient is instructed to chew the 1-g piece of paraffin for 1 minute to soften it and then to swallow or spit out all saliva. The patient then chews the softened bolus of paraffin for a fixed time (5 minutes), spitting the saliva into the graduated cylinder. Foam can be avoided or reduced by using an ice-chilled beaker or by the addition of a drop of octanol. The secretion rate is calculated in milliliters per minute.

As an alternative to mechanical stimulation by chewing, saliva may be stimulated chemically by a 2% solution of citric acid (made up at a local pharmacy), applied to the laterodorsal surface of the tongue at 30-second intervals, for 2 minutes. The patient then spits the saliva into the receiving vessel. The procedure is repeated twice more, for a total collection time of 6 minutes. As before, the SSR is expressed as milliliters per minute.

Measurement of unstimulated (resting) SSR. It is impossible to sample true "resting" saliva, because the SSR is always influenced by some kind of stimulus during consciousness. Nevertheless, a sample collected by passive drooling, without any deliberate physical or chemical stimulation, is more reliable than stimulated saliva as an indicator of reduced SSR and hyposalivation.

When resting secretion is collected, the patient is instructed to sit in a relaxed position, with the elbows resting on the knees and the head lowered between the arms, the so-called coachman's position. This position is also good for the collection of stimulated saliva. Even slight movements of the tongue, cheeks, jaws, or lips should be avoided. The lips are only slightly apart, and the patient allows the saliva to drool passively over the lower lip into the measuring cylinder, avoiding actively spitting.

For healthy adults, the resting SSR should exceed 0.1 mL/min. In patients suspected to have hyposalivation, the sampling period should be 15 minutes, to avoid bias

caused by fluctuations in the SSR. For clarity, the results should be expressed in milliliters per minute and in milliliters per 15 minutes.

Measurement of SSR from the major salivary glands. Parotid saliva is usually obtained in a modified, two-chambered Carlson-Crittenden collector). The inner chamber is placed over the orifice of Stensen's duct; the outer chamber is attached, via thin tubing, to a rubber bulb that, when compressed, creates a slight negative pressure and permits the device to adhere to the surrounding mucosa. This device makes it possible to collect pure parotid saliva in a noninvasive manner.

A simple method that makes it easy for the dentist to collect submandibular and sublingual saliva has also been reported. The region of Wharton's ducts is isolated with gauze, and the orifices of Stensen's ducts are covered. The saliva, resting or stimulated, that has collected during a known time is aspirated with a plastic micropipette. The flow rate is expressed as milliliters per minute per pair of submandibular or sublingual glands.

Measurement of SSR from minor salivary glands. Saliva may be obtained from the minor salivary glands of the lower lip or the palate. The minor glands are dried and isolated with gauze or cotton rolls. For resting saliva, the fluid that is present at the orifice of one or more of these glands after 2 minutes is adsorbed onto filter strips (Perio-Paper). The volume of fluid on each strip is read electronically in a special device (Periotron). For stimulated minor gland saliva, the tongue is swabbed with 2% citric acid solution, as described earlier. The results are expressed as microliters per minute. Because the number of glands and the area sampled vary, the SSR is semiquantitative.

Fig 80 Major salivary glands. (Modified from Tenovuo and Lagerlof, 1994. Reprinted with permission.)

Fig 81 Simplified diagram of a secretory unit in a major salivary gland. The ducts usually branch leading to several acini. (Modified from Tenovuo and Lagerlof, 1994. Reprinted with permission.)

Fig 82 Mean unstimulated whole salivary secretion rates in males and females. (From Heintze et al, 1983. Reprinted with permission.)

Fig 83 Mean stimulated salivary secretion rates in males and females. (From Heintze et al, 1983. Reprinted with permission.)

Fig 84 Mean unstimulated whole secretion rates in healthy (unmedicated) males and females, by age group. *Statistically significant difference between the 80+ females and the other groups. (From Percival et al, 1994.)

Fig 85 Mean stimulated parotid salivary secretion rates in healthy (unmedicated) males and females, by age group. (From Percival et al, 1994.)

Fig 86 Frequency distribution of stimulated whole salivary secretion rates, by age. (From Axelsson et al, 1990.)

Fig 87 Mean stimulated whole salivary secretion rates in dentate and edentulous individuals, by age. (From Axelsson et al, 1990.)

Fig 88 Mean stimulated whole salivary secretion rates, by sex; by use of drugsregular medication (Med) or no regular medication (No med); and by status of the dentitionedentulous (Edent) or dentate (Dent). (From Axelsson et al, 1990.) Fig 89 Mean numbers of teeth (excluding third molars) in individuals with low (0.0 to 0.7 mL/min) and high (1.5 mL/min) stimulated salivary secretion rates, by age. (From Axelsson et al, 1990.)

Fig 90 Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) in individuals with low (0.0 to 0.7 mL/min) and high (1.5 mL/min) stimulated salivary secretion rates, by age. (From Axelsson et al, 1990.)

Fig 91 Frequency distribution of missing, filled, decayed, and intact surfaces in individuals with low (0.7 mL/min) and high (1.5 mL/min) stimulated salivary secretion rates; and in individuals using drugs that impair salivary function (Med) and those not taking any medication (No med). (From Axelsson and Paulander, 1994.) Fig 92 Optical measurement of the mucosa with infrared light. (Courtesy of T. Axell.)

Fig 93 Mechanical friction measurement of dry, thin oral mucasa in a patient with hyposalivation. (Courtesy of T. Axell.)

Fig 94 Typical dorsum of the tongue in a patient with xerostomia. (Courtesy of T. Axell.)

Fig 95 Angular cheilosis and typical dorsum of the tongue in a patient with xerostomia. (Courtesy of T. Axell.)

Fig 96 Typical pattern of carious lesions and restorations in mandibular teeth in an older person with xerostomia. (Courtesy of T. Axell.)

Composition of saliva

Although composed mainly of water, saliva is a complex secretion. As discussed earlier, so-called whole saliva consists primarily of the secretions from the major and minor salivary glands. Whole saliva also contains a number of constituents of nonsalivary origin: crevicular fluid, serum, and blood cells; bacteria and bacterial products; desquamated epithelial cells and cellular components; viruses and fungi; food debris; fluoride; and some bronchial secretions. It is estimated that 1 mL of whole saliva contains about 700 million viable bacteria, 0.5 million leukocytes (more than 90% polymorphonuclear neutrophil leukocyte-cells), thousands of desquamated epithelial cells, 2 mg of proteins, 800 mg of lipids, 100 mg of immunoglobulins, and some inorganic electrolytes, such as calcium, phosphates, bicarbonate, sodium, chloride, and fluoride. Even pure secretions collected directly from the orifices of the main excretory ducts of the parotid, submandibular, or sublingual glands contain the saliva synthesized by the secretory cells, along with certain substances supplied by the circulation.

The composition of whole saliva is influenced by a number of physiologic factors. Important among these are the source, the method of collection, and the degree of stimulation. As described earlier, the major salivary glands are composed of different acinar cells, programmed to synthesize quite different secretions. The parotid glands

have serous acinar cells and produce a proteinaceous, watery secretion. The secretion from the sublingual glands is mucous and hence more viscous. The submandibular glands have both serous and mucous acinar cells and produce a saliva with lower protein content and higher viscosity than do the parotid glands. The minor salivary glands are purely mucous glands and produce a saliva that is particularly viscous and rich in secretory immunoglobulin A (IgA).

In response to stimulation, there may be a manifold increase in salivary output, with significant changes in consistency and in the concentration of many of its constituents. About 99% of the saliva is water. The remaining 1% consists mainly of large organic molecules, (eg, proteins, glycoproteins, and lipids); small organic molecules (eg, glucose and urea); and electrolytes (chiefly sodium, calcium, chloride, and phosphates). Most of the organic molecules are produced by the acinar cells; some are synthesized in the ducts, and some are transported into the saliva from the blood. A list of salivary constituents, subclassified as proteins, small organic molecules, and electrolytes, is presented in alphabetical order in Box 12.

The major proteins of the salivary glands are produced by the acinar cells and exist as families. Each family has a number of distinct but closely related members (genetic polymorphism). They include the proline-rich proteins (with at least 13 discrete members); the histatins (histidine-rich proteins with five related components); the cystatins (cystine-containing proteins); the tyrosine-rich proteins (statherin and others); mucins of high and low molecular weight; glycosylated and nonglycosylated amylases; and several salivary peroxidases.

Other salivary proteins exist in a single form, some produced by acinar and some by ductal cells. Among the acinar proteins are epidermal growth factor, secretory component, and lactoferrin. Lysozyme is known to be produced by duct cells, but, for many other constituents, the site of origin is unknown. Included among the compounds that are transported from the blood into the salivary secretion are the major electrolytes; albumin, IgA, immunoglobulin G (IgG), and immunoglobulin M (IgM); and vitamins, drugs, hormones, and water.

There is a good correlation between plasma and salivary levels of a number of hormones and medications. This correlation forms the basis for proposals to use saliva collection as a noninvasive means of monitoring hormones and both therapeutic and illicit drugs. Salivary sampling is currently being tested as a screening method for the presence of antibodies to human immunodeficiency virus 1. However, such methods are complicated by the fact that glandular inflammation results in a marked increase in the number and concentration of serum elements in saliva.

Role of saliva as a modifying factor in dental caries

Indisputably, an adequate secretion rate and saliva of good quality are essential for oral health. Saliva is well known to have specific protective effects against dental caries. The most direct evidence of this is the rampant caries that can occur following the loss of salivary function as a result of irradiation for head and neck tumors. Within a few weeks, tooth surfaces not normally susceptible to caries may be affected, leading to complete coronal destruction. The principal properties of saliva that protect the teeth against caries are:

1. Dilution and clearance of dietary sugars

2. Neutralization and buffering of the acids in plaque

3. Supply of ions for remineralization

4. Both endogenous and exogenous antiplaque and antimicrobial factors

The major functions of different salivary components have been presented in Table 12.

Dilution and clearance of food components and clearance of microorganisms

Of the many functions of saliva, the most important is the clearance of oral microorganisms and food components from the mouth to the gut. Therefore, an adequate volume to flush noxious (and also commensal) microorganisms out of the oral cavity is a prerequisite for a healthy balance between host defense and endogenous and exogenous microbial attack in the mouth.

This balance can be disturbed by either extensive growth of bacteriaas a consequence, for example, of poor oral hygiene, excessive dietary intake of fermentable carbohydrates, or some systemic diseasesor reduced SSR (hyposalivation). In the most highly caries-susceptible individuals, a combination of these factors is common. Although caries research has concentrated on salivary clearance of sucrose and fluoride, the principles that apply to sucrose clearance are valid for any substance introduced to the oral cavity. Besides sugars and fluoride, other substances are of relevance to the clinician: chemical plaque control agents (chlorhexidine, etc), chloride in relation to corrosion of amalgam, and citric acid and other acidic products that might be implicated in tooth erosion.

The study of sugar clearance was pioneered by Swenander-Lanke (1957), who found that, following the consumption of solid carbohydrate foods, the concentration of sugar in saliva fell exponentially, at first rapidly and then more slowly. Sreebny et al (1985) noted that sugar solutions were cleared in a two-stage pattern and that the rapid clearance rates over the first 6 minutes and the slower ones thereafter were proportional to the shifts in the SSR at those times. In 1983, Dawes developed a computer model for sugar clearance, based on the following postulate: that the important factors in clearance were (1) the volume of saliva just before and after swallowing and (2) the unstimulated SSR. The computer predictions based on this postulate were confirmed in studies using an "artificial mouth" system and in human experiments. The computer predicted that the clearance was rapid when both salivary volumes were low and the unstimulated SSR was high.

Figure 97 shows a physiologic model of the oral cavity that includes the most important properties for understanding the clearance process. The events after intake of sucrose may be described as follows: In the oral cavity, there is a minimum volume of saliva after swallowing, the residual volume. Spread out as a thin film, this volume has been estimated to be, on average, 0.8 mL, but the interindividual variation is large. Dissolution of a small amount of sucrose in this small volume of saliva will

give rise to a very high sucrose concentration. For example, dissolving one tenth (0.3 g) of a sugar lump in the residual volume will result in sucrose concentrations much higher than would be found in an ordinary sucrose-containing beverage. The taste of sucrose, together with optional flavoring agents, stimulates the salivary glands to respond in a few seconds with an increase in the flow rate. The volume of saliva will increase until a maximum value is reached. This maximum value is about 1.1 mL (ie, a normal swallow can be estimated to be 0.3 mL). The swallowing reflex is stimulated, and some of the sucrose is eliminated. The remaining sucrose is then progressively diluted by the saliva entering the mouth until the maximum volume is reached, triggering another swallow, and so on.

After some time, the concentrations of sucrose and the optional flavoring agent reach such low levels that the stimulation of the glands decreases to an unstimulated state, which results in a slower clearance process, dependent on the unstimulated SSR. The time it takes to reach a given detectable low level has been used as a measure of clearance rate. Several variables are important for the clearance rate, the most important being the SSR and the volumes of saliva in the mouth before and after swallowing. A high SSR will result in rapid clearance, compared to the slow clearance obtained at low SSR. From the great differences in clearance rates, it is clear that caries risk increases enormously with a low SSR.

The clearance rate is an individual property that is constant over time. However, if changes in health status cause a decrease in the SSR, a drastic change in clearance rate will ensue. The clearance rate also differs considerably at different sites, because of the complicated rheology of the oral cavity. The film overlying the mucous membrane and the teeth moves at varying rates, from 0.8 to 8.0 mm/min. In sites where the salivary film may be expected to move rapidly, for example, in proximity to the ductal orifices, the clearance rate is considerably greater than in sites where the saliva is stagnant (eg, the buccal areas of the maxillary anterior teeth and mandibular molars), which may explain in part the pattern of caries on different teeth and tooth surfaces.

Sucrose in saliva and in the salivary film diffuses readily into dental plaque. A few minutes after sugar intake, the plaque will be overloaded with sucrose, with a greater sugar concentration than is present in the saliva. Provided that the plaque is not too thick to impede accessibility of the saliva, the flow of sucrose will be reversed. Therefore, there is a correlation between pH changes in the plaque and the salivary clearance of sucrose. In contrast to rapid clearance, slow clearance resulting from limited salivary accessibility will cause steep Stephan curves (see chapter 2). After sucrose rinsing, the pH fall in approximal plaque on molars is much more severe in the center of the surface than it is lingually, because the central area is inaccessible to saliva for dilution and buffering of the plaque acids.

Neutralization and buffering of acids

Although while the effect of saliva in facilitating sugar clearance can partly explain why saliva reduces formation of plaque acids and therefore caries, the neutralizing and buffering actions of saliva are more dramatic. These are due predominantly to salivary bicarbonate, originating mainly from the parotid gland. In unstimulated saliva, the bicarbonate level is low; at the greater secretion rates of stimulated saliva, the concentration is higher, the pH rises, and the buffering power of saliva increases

dramatically. There are also other less important buffering systems in saliva, such as macromolecular proteins.

Ingestion of sugar causes a drop in plaque pH. When saliva is experimentally prevented from entering the mouth (by cannulating the excretory ducts and discharging the saliva extraorally), the fall in plaque pH after ingestion of sugar is greater and more prolonged than when salivary access is normal. If, after ingestion of sugar, flow is stimulated by chewing of paraffin or cheese, the plaque exhibits an immediate and dramatic rise in pH and a fall in lactic acid concentration, accompanied by a change in its amino acid spectrum. Similar effects are seen with sugar-free chewing gum and even with sucrose-sweetened gum, provided that this is chewed for longer than the time it takes for the sugar to be dissolved.

Although the plaque of caries-resistant patients and the plaque of caries-susceptible patients respond similarly to a sugar challenge, the levels at which these responses occur are quite different. In the plaque from a caries-resistant person, the presugar pH is higher and the fall in pH after the sugar challenge is smaller. Studies have also shown that the capacity to buffer plaque acids is greater in caries-resistant patients than it is in caries-susceptible patients.

The buffering effects of saliva are mostly measured in vitro by laboratory methods or chairside methods. In the laboratory, 1.0 mL of saliva is mixed with 3.0 mL of hydrochloric acid (0.0033 M for resting saliva; 0.005 M for stimulated saliva). A stream of air is then passed through the mixture for 20 minutes and the pH (the "final pH"), is measured. If the air stream step, which removes carbon dioxide, is excluded, about the same results are obtained for saliva with low buffering effect, final pH 5 or lower.

Chairside tests are available, allowing the clinician to evaluate the salivary buffering effect directly after sampling and to discuss the results with the patient. In the Dentobuff Strip system (Fig 98a to 98c), one drop of stimulated saliva is placed on a test strip containing an acid and a pH indicator. After the reaction between saliva and acid, the color of the test pad is compared to a chart, and the final pH value is obtained. This test is highly simplified and will discriminate among low, medium, and high buffering capacities. The method is particularly useful for identifying individuals with risk values, that is, low buffering capacity (final pH of 4 or less). As with secretion rates, there is a normal range of buffer capacity, with no apparent relation to caries risk. However, below a threshold value (final pH less than 4), the carious process seems to be facilitated.

Figure 99 shows the frequency distribution of the buffering effects in males and females, taken from the previously described salivary study in adults by Heinze et al (1983); more females had low values (pH less than 4.0) for both resting and stimulated saliva. Notably, other studies have shown a dramatic reduction in salivary buffering effects during the last months of pregnancy, which may explain in part why caries incidence seems to increase during pregnancy.

On a population basis, there is a positive correlation between SSR and buffering effect, but there are many individual exceptions. A low SSR combined with a low or moderate buffering effect clearly indicates poor salivary resistance to microbial

attack: Clearance of microorganisms is slow, and the residual saliva, ranging in various individuals from 0.5 to 1.0 mL, is spread as a thin film on the oral surfaces. Fermentable carbohydrates dissolved in this small volume of saliva would be neutralized only slowly, because of the low buffering effect.

The interpretation of salivary buffering tests in isolation is questionable. In most investigations, there is little or no correlation with variables measuring different aspects of dental caries. One important explanation is that the decisive events in a carious attack take place in the plaque and below the enamel surface. In these loci, the buffering mechanisms are very different from those found in saliva. It is unlikely that salivary buffering substances could significantly influence pH changes in the depth of the plaque, particularly in areas of limited accessibility, for example, the approximal surfaces of the molars. The buffering capacity of the plaque may have greater relevance, but test methods are as yet unavailable. On more accessible mandibular lingual surfaces covered with only a thin plaque, the salivary buffering effect may play a more significant role as a modifying factor in lesion development.

The human mouth is quite frequently exposed to agents that have a pH different from that of saliva (6.5 to 7.5) and are potentially damaging to the teeth (erosion) or to the mucosa. Under these conditions, the role of the buffering agents in saliva is to restore the pH to the normal range as quickly as possible.

Demineralization and remineralization of tooth surfaces

The physicochemical integrity of dental enamel in the oral environment is entirely dependent on the composition and chemical behavior of the surrounding fluids: saliva and plaque fluids. The main factors governing the stability of enamel apatite are pH and the free active concentrations of calcium, phosphate, and fluoride in solution, all of which can be derived from the saliva (see Box 12).

The development of a clinical carious lesion involves a complicated interplay between a number of factors in the oral environment and the dental hard tissues. The carious process is initiated by bacterial fermentation of carbohydrates, leading to the formation of a variety of organic acids and a fall in pH. Initially, H+ will be taken up by buffers in plaque and saliva; when the pH continues to fall (H+ increases), however, the fluid medium will be depleted of OH- and PO3

4-, which react with H+ to form H2O and HPO2

4-. On total depletion, the pH can fall below the critical value of 5.5, where the aqueous phase becomes undersaturated with respect to hydroxyapatite (HA). Therefore, whenever surface enamel is covered by a microbial deposit, the ongoing metabolic processes within this biomass result in fluctuations in pH and occasional steep falls in pH, which may result in dissolution of the mineralized surface. The role of the saliva in this process is highly dependent on accessibility, which is closely related to the thickness of the plaque (for review see Pearce, 1991; Tenovuo, 1997).

Caries versus erosion. As discussed earlier, dissolution of enamel can result in the development of either a carious lesion or an erosive lesion. Caries is defined as the result of chemical dissolution of the dental hard tissues, caused by bacterial degradation products, that is, acids produced by bacterial metabolism of low-molecular weight sugars in the diet. Erosion is defined as chemical dissolution of

tooth substance caused by any other acid-containing agent. Mixed lesions may well exist, particularly when the dentin has been exposed by erosion, causing hypersensitivity, which may lead to inadequate plaque control and, subsequently, to caries. This condition occurs frequently on exposed root surfaces.

The appearance of the two lesions differs. The carious lesion is characterized by a subsurface demineralized lesion body, covered by a rather well-mineralized surface layer. In erosion, the surface has been etched away layer by layer, and there is no subsurface demineralization. Under normal conditions, in the absence of thick undisturbed plaque and/or very high frequency of acidic dietary products, teeth do not dissolve in saliva, because it is supersaturated with calcium, phosphate, and hydroxyl ions, which constitute the mineral salts of the tooth. The degree of supersaturation is even greater in plaque, especially in its extracellular fluid phase, which is in direct contact with the tooth surface. In addition, in individuals who have a regular daily source of fluoride (eg, fluoride toothpaste), both the saliva and the plaque fluid should contain an abundance of fluoride ions. In the dynamic equilibrium of the carious process, the supersaturation of saliva provides a barrier for demineralization and a driving force for remineralization. This equilibrium is greatly affected by fluoride, which reduces demineralization and enhances remineralization. Salivary saturation is overcome only when the plaque pH falls so far that the hydroxyl and phosphate ion concentrations are reduced below a critical value (through conversion of PO4

3- to HPO4

2- and H2PO4-).

In principle, dental enamel can be dissolved under two different chemical conditions. When the surrounding aqueous phase is undersaturated with respect to hydroxyapatite and supersaturated with respect to fluorapatite (FA), HA dissolves and FA is formed. The resulting lesion is a carious lesion in which the dissolving HA originates from subsurface enamel and FA is formed in the surface enamel layers. The higher the supersaturation with respect to FA, the more fluoride is taken up in the enamel surface, the better mineralized the surface enamel layer becomes, and the less demineralized is the subsurface body of the lesion.

On the other hand, if there is undersaturation with respect to both HA and FA, both apatites dissolve concurrently, and layer after layer is removed. This will result in an erosive lesion. Fresh acidic fruit, fruit juices, acidic carbonated soft drinks, and some champagnes are all unsaturated with respect to both apatites and are able to cause erosive demineralization of the teeth. These mechanisms for enamel dissolution are illustrated in Figure 100.

Role of calcium. In these processes, the most important inorganic ions are calcium, phosphate, and fluoride. Calcium is a bivalent ion excreted, together with zymogen proteins, into the lumen of the acinus. Therefore, the concentration of calcium found in the saliva is dependent on the SSR. Going from an unstimulated state to a "somewhat stimulated" state, the calcium concentration decreases somewhat. However, the excretion pattern is complicated by the different calcium concentrations found in the secretions from different glands: the concentration in the submandibular or sublingual fluid is about twice as high as that in the parotid saliva. As the proportion of parotid secretion in the total volume of saliva increases with stimulation, the resulting flow pattern in whole saliva is a linear increase related to the calcium concentration.

Depending on the pH, calcium is distributed in saliva in ionized and bound forms. The free, ionized calcium is especially important in the carious process, because it participates in establishing the equilibrium between the calcium phosphates of the dental hard tissue and its surrounding liquid. At pH values close to normal, the ionized calcium constitutes approximately 50% of the total calcium concentration, but it increases if salivary pH is lowered. At pH values below 4, most of the salivary calcium is in ionized form.

The nonionized calcium is distributed on a diversity of ligands with association constants in a large range; that is, the calcium is more or less firmly bound to inorganic ions such as inorganic phosphate, bicarbonate, and fluoride (10% to 20% of the total calcium concentration, depending on pH and SSR), to small organic ions (less than 10%) such as citrate, and to many macromolecules (10% to 30%). Some salivary macromolecules have been attributed a special role in oral calcium homeostasis.

The influence of SSR on the distribution of calcium on free and bound fractions is complex. As already pointed out, the pH of saliva is strongly dependent on the SSR, as are the concentrations of most of the calcium-complexing substances. At low SSR, the bicarbonate concentration is very low, with a correspondingly low concentration of the calcium bicarbonate complex.

The tooth is usually separated from the saliva by an intermediate layer of integuments, in the form of a pellicle or a plaque. The total calcium concentration in these compartments is slightly higher, sometimes much higher, than in the saliva, because of a high concentration of binding sites for calcium and because of the presence of precipitated calcium salts. There is a strong correlation between both total and ionized calcium in saliva and dental plaque, showing a flow of calcium over the plaque-saliva interface following existing diffusion gradients in ionized calcium. This gradient will be large after sugar intake, liberating bound calcium; as the plaque pH slowly increases, the concentrations of ionized calcium in saliva, pellicle, and plaque will slowly reach an equilibrium.

Role of inorganic phosphate. The inorganic orthophosphate in saliva consists of phosphoric acid (H3PO4) and the primary (H2PO4

-), secondary (HPO42-), and tertiary

(PO43-) inorganic phosphate ions. The concentrations of these ions are dependent on

the pH of the saliva. The sum of the ions and the molecule constitutes the total phosphate concentration. The lower the pH, the less the concentration of the tertiary ion, implying that the ion product of hydroxyapatite decreases considerably with decreasing pH. This phenomenon is the main cause of demineralization of the teeth. As with calcium, it is evident that the content of inorganic phosphate in saliva is a prerequisite for the stability of the tooth mineral in the oral environment. The concentration of total inorganic phosphate decreases with increasing SSR. As is the case for calcium, the different glands differ in phosphate excretion: the phosphate concentration in the submandibular glands is only about one third that in parotid saliva, but is about six times higher than that in the minor mucous glands, the glands nearest the tooth surfaces. Therefore, it may be assumed that the inorganic phosphate concentration shows a large variation in the microenvironment.

About 10% to 25% of the inorganic phosphate, depending on pH, is complexed to inorganic ions such as calcium or is bound to proteins. A small part, less than 10%, is in the dimer form, pyrophosphate (H4P2O7), which is a potent inhibitor of the precipitation of calcium phosphate and influences the formation of calculus. This is the rationale for the inclusion of pyrophosphate in toothpastes intended to inhibit calculus formation. However, the prevalence and incidence of caries in individuals who exhibit rapid formation of salivary calculus tend to be less than average.

In a recent 3-year, longitudinal, double-blind study of fluoride toothpaste in more than 4,000 11 to 12 year olds, those with salivary-derived calculus developed significantly fewer new carious surfaces than did the others (Stephan et al, 1994). Among adult patients maintained for several years in a needs-related preventive program and with very high standards of oral hygiene, a subselection of participants with very rapid formation of salivary-derived calculus had more intact tooth surfaces than did patients with little or no calculus formation (Axelsson et al, 1995, unpublished).

The inorganic phosphate of saliva has several important biologic functions, the most important from a caries aspect being its contribution to solubility products with respect to calcium phosphates and thus its role in the maintenance of the tooth structure. Its minor role in salivary buffering has already been discussed.

Role of fluoride. Fluoride in the fluids surrounding the enamel crystals has been shown to have the potential to reduce the rate of demineralization. When present in the liquid phase of remineralization, fluoride will be incorporated into the enamel crystals and the enamel will become more resistant to demineralization (Fig 103). Fluoride has also been shown to reduce acid production in dental plaque. Therefore, in caries-preventive programs, the aim of fluoride administration should be to ensure that fluoride levels in the oral fluids are adequate to prevent and inhibit caries. The fluid bathing a plaque-covered tooth surface consists of saliva, plaque fluid, and the fluid surrounding the enamel crystals and is sometimes influenced by the crevicular fluid. These fluids constitute a continuous system, and ions will diffuse according to their concentration gradients. Fluoride introduced into the oral cavity will be distributed in saliva and thus influence the fluoride concentration in plaque fluid and enamel crystal fluid.

Fluoride is present in saliva in concentrations that depend on fluoride in the environment, especially in drinking water. Other important sources are fluoride toothpastes and other fluoride products used for caries prevention and control. In areas with low concentrations of fluoride in the drinking water (below 10 umol [0.2 ppm]), the basal concentration of fluoride in whole saliva is usually less than 1 uM. The concentration may be much higher in areas with higher water fluoride concentrations.

After an intake of fluoride, the levels of fluoride in the blood increase, reaching a peak after 30 minutes to 1 hour. The fluoride enters the saliva by simple diffusion over the membranes of the acinar cells. The concentration of fluoride in the duct saliva will therefore follow the plasma values, at a 30% to 40% lower level. This results in an increase of fluoride concentration in whole saliva, although only 0.1% to 0.2% of the ingested fluoride is excreted via the salivary glands.

In some foods and beverages, the fluoride is mainly in ionized form, which readily dissolves in the saliva; in others, fluoride may be firmly bound, making it difficult to predict the resulting fluoride concentration after exposure. This variable should be taken into account in the formulation of caries-preventive topical agents: For example, fluoride tablets and fluoride chewing gum have very different solubility rates. In addition, some slow-release fluoride agents, such as fluoride varnish, may contain up to 2% to 5% fluoride; glass-ionomer cements may be intermittently reloaded with fluoride, and as a consequence, release fluctuating amounts of fluoride.

Because the oral cavity contains only a small volume of saliva in a thin film, even if only very small amounts of fluoride dissolve in the residual saliva, the resulting concentration may be very high. For example, if a fluoride tablet of 0.25 mg is dissolved in 1 mL of saliva, the resulting fluoride concentration is about 13 mol (about 200 ppm), more than 10,000 times higher than the basal fluoride concentration. Even higher fluoride concentrations could be expected in loci close to the fluoride source. For example, if a fluoride tablet is placed on one side of the oral cavity, very large differences in salivary fluoride concentration are found between the exposed and unexposed sides of the mouth (Sjogren et al, 1993). Therefore, slowly dissolving fluoride tablets and fluoride chewing gum should be moved around the mouth continuously to distribute fluoride to as many microenvironments as possible, and slow-release fluoride agents should be applied to key-risk teeth and key-risk surfaces.

The high initial fluoride concentration in the salivary film after fluoride exposure will establish a concentration gradient between the dental integuments and the plaque. Fluoride will diffuse from saliva into the pellicle and the plaque, rapidly elevating the concentration of fluoride in the plaque fluid. Mineral calcium fluoride (CaF2) may form in saliva, in the pellicle, and in plaque fluid (see Fig 101) (Ogaard et al, 1983a, b).

The limiting factor for the formation of CaF2 is the calcium content of the oral fluids. Therefore, the use of fluoride chewing gum after every meal as a combined saliva-stimulating and fluoride agent, resulting in increased calcium release from the saliva, fluoride release, and increased buffering effect, offers a rational, self-administered measure for caries control during or just after the fall in pH. Calcium fluoride releases fluoride slowly (Fig 102). Fluoride diffusing into microorganisms also prevents participation of the enzyme enolase in the glycolytic pathway by binding magnesium, essential for optimal function of the enzyme. However, this will not occur on plaque-free tooth surfaces.

After the initial exposure to fluoride, the salivary concentration of fluoride decreases rapidly, by the same mechanisms involved in sugar clearance. The most important factor for the fluoride clearance rate is, as for sugar, the salivary secretion rate, which is dependent on the degree of stimulation. Fortunately, fluoride clearance is significantly slower in patients with hyposalivation than in individuals with normal SSR. Clearance varies markedly at different sites in the oral cavity, is generally more rapid from lingual than from buccal sites, and is most rapid beneath the tongue.

However, there are some important differences between salivary fluoride clearance and salivary sugar clearance. The saliva contains a certain basal level of fluoride, which results in a gradual, theoretically asymptomatic decrease of fluoride to the

basal level. This slow adaptation is often prolonged for several reasons. First, swallowed fluoride will partly reenter the saliva, increasing the amount of fluoride, but the effect on the fluoride concentration is probably minor. Second, after a few minutes, the fluoride concentration in pellicle and plaque fluids is higher than it is in the saliva, causing the concentration gradient to reverse direction. Some fluoride will therefore diffuse back from the pellicle into the saliva. Third, after the fluoride concentration in the pellicle has fallen to a level that makes the fluid undersaturated with respect to calcium fluoride, this salt may start to dissolve slowly, increasing the ionized fluoride concentration.

This last factor is complicated by its dependency on the pH of the pellicle, because at pH values in the normal range, calcium fluoride dissolution is inhibited by adsorbed phosphate ions. When pH approaches 5, this coating of the calcium fluoride particles vanishes (see Fig 102). When pH rises again, phosphate and protein-coated CaF2 is re-formed in the pellicle (Fig 103).

Antimicrobial and other protective properties

The saliva contains many different proteins and some other small organic proteins that together protect the oral cavity (the soft tissues as well as the teeth) from frictional wear, dryness, erosion, pathogenic bacteria, and so on (see Box 12).

Lubrication and other protective properties. Almost all salivary proteins are glycoproteins; that is, they contain variable amounts of carbohydrates linked to the protein core. Glycoproteins are often classified according to their cellular origin and subclassified on the basis of their biochemical properties. A characteristic feature is that many occur in multiple forms, constituting families; these families, may, however, exhibit remarkable functional differences.

Mucous glycoproteins, the mucins, are of acinar cell origin, have a high molecular weight, and contain more than 40% carbohydrate. The mucins are produced by the minor salivary glands in the palate and provide a nonfrictional, lubricant layer that protects the soft tissues from wear and tear and facilitates swallowing of food. Because the mucins have a strongly negative charge, other negatively charged molecules, such as those contained in the cell walls of many oral bacteria, are repelled from the mucin-coated oral mucosa. Among other properties, the mucins also bind water and thereby protect the oral mucosa from drying out.

Serous glycoproteins have a much lower molecular weight than mucins and contain less than 50% carbohydrate: Many belong to a group called proline-rich glycoproteins (PRPs), of which several are phosphorylated. These proteins are secreted from the parotid and submandibular glands.

The collective name glycoprotein refers to all carbohydrate-linked proteins, making this group very heterogenous and large. Most salivary proteins, such as secretory IgA, lactoferrin, peroxidases, and agglutinins, belong to this group. Because human saliva is supersaturated with respect to most calcium phosphate salts, some proteins are necessary to inhibit their spontaneous precipitation in the salivary glands and their secretions. Such proteins include statherin and PRPs. The resulting stable but supersaturated state of the saliva with respect to calcium phosphate salts constitutes a

protective and reparative environment of importance for the integrity of the teeth.

Statherin is present in both submandibular and parotid salivas. Proline-rich proteins form a complex group of proteins with large numbers of genetic variants, some of which also have the ability to inhibit spontaneous precipitation of calcium phosphate salts. The molecular size of PRPs ranges from 106 to 150 amino acid residues. Like statherin, PRPs are remarkable for their high degree of compositional and charge asymmetry. Proline-rich proteins are readily adsorbed from saliva to hydroxyapatite surfaces and it is most likely that these adsorbed PRPs inhibit the crystal growth of calcium phosphate salts. Although present in whole saliva, PRPs are also susceptible to proteolytic degradation by oral microorganisms.

-Amylase is one of the most important salivary enzymes, accounting for as much as 40% to 50% of the total salivary gland-produced protein. Most (80%) is synthesized in the parotid glands and the remainder in the submandibular glands. The biologic role of salivary amylase is to split starch into maltose, maltotriose, and dextrins. Maltose can be further fermented by oral bacteria. Therefore, although amylase in saliva clears starch-containing food debris from the mouth, acids are formed in this process. In this way, starch may have some cariogenic potential. Salivary -amylase is inactivated in the acidic parts of the gastrointestinal tract, and therefore its action is limited to the oral area.

Antimicrobial properties. As described earlier, saliva plays a significant role in maintaining an appropriate balance within the ecosystem associated with tooth surfaces. This balance is of great significance in the control of dental caries, because saliva will enhance the ability of some bacteria to survive and will reduce the competitiveness of others. Saliva achieves this control over the oral flora through its components, which can be constantly present or activated by a specific host response.

The major antimicrobial proteins are listed in Box 13. Many studies have shown that most of these proteins can inhibit the metabolism, adherence, or even the viability of cariogenic microorganisms in vitro (for review see Tenovuo, 1997). However, their role in vivo is largely unknown: It seems that they are important for the control of microbial overgrowth in the mouth, but their selectivity against pathogens has not been determined.

A newly proposed biologic function for PRPs is the ability of adsorbed acidic PRPs to selectively mediate bacterial adhesion on tooth surfaces. Recently it was shown that the negative charge of these acidic PRPs binds electrostatically to calcium on the tooth surfaces, while the outer ends, consisting of proline and glutamine amino acids, attract and bind very strongly to the harmless and protective normal microflora of the teeth (Streptococcus oralis, Streptococcus sangius, and Streptococcus mitis). This may explain early scanning electron micrographs obtained by Lie (1978), showing how a gram-positive "pioneer colonizer" attaches to the pellicle-covered tooth surface, in contrast to a gram-negative bacterium and the pellicle (Figs 104 and 105).

This primary colonization of the protective normal microflora occurs during the first 24 hours after cleaning. However, recent research has shown that the so-called secondary colonization by other, more pathogenic microorganisms (gram-positive as well as gram-negative) is strongly related to the binding between galactose amine

structures on the surfaces of the normal microflora as well as the secondary colonizers. The production and the individual structures of acidic PRPs and galactose amines are genetically related and may partly explain individual variations in plaque formation rates. This is a field of ongoing research (Stromberg, 1996).

The lysozyme in whole saliva is derived from the major and minor salivary glands, gingival crevicular fluid, and salivary leukocytes (polymorphonuclear neutrophil leukocytes). Salivary lysozyme is present in newborn babies at levels equal to those of adults, suggesting a preeruptive antimicrobial function. The classic concept of the antimicrobial action of lysozyme is based on its muramidase activity, ie, the ability to hydrolyze the bond between N-acetylmuramic acid and N-acetylglucosamine in the peptidoglycan layer of the bacterial cell wall. Gram-negative bacteria are more resistant to lysozyme because of the protective function of the outer lipopolysaccharide layer. In addition to its muramidase activity, lysozyme is strongly cationic, and can activate bacterial "autolysins," which can destroy the cell wall components.

Lactoferrin is an iron-binding glycoprotein secreted by the serous cells of the major and minor salivary glands. Polymorphonuclear leukocytes are also rich in lactoferrin and release it into gingival fluid and whole saliva. The biologic function of lactoferrin is attributed to its high affinity for iron and its consequent expropriation of this essential metal from pathogenic microorganisms. This bacteriostatic effect is lost if the lactoferrin molecule is saturated with iron, a factor that should be taken into account in areas where the drinking water is rich in iron. In its iron-free state (apolactoferrin), it has a bactericidal, irreversible effect against a variety of microorganisms, including mutans streptococci. Apolactoferrin can also agglutinate Streptococcus mutans cells.

Salivary peroxidase is produced in the acinar cells of the parotid and submandibular glands but not in the minor salivary glands. Salivary peroxidase systems have two major biologic functions: (1) antimicrobial activity and (2) protection of host proteins and cells from hydrogen peroxide toxicity.

Salivary agglutinins are glycoproteins that have the capacity to interact with unattached bacteria, resulting in clumping of bacteria into large aggregates that are more easily flushed away by saliva and swallowed: the term aggregation is therefore often used synonymously with agglutination. Listed in Box 13 are salivary proteins with agglutinating capacity. The most potent agglutinin is a high-molecular weight glycoprotein that has been isolated from human parotid saliva. Despite a concentration in parotid saliva of only 0.001%, it is very effective. Mucins are also able to agglutinate bacteria. In high-molecular weight glycoproteins, sugar residues and sialic acid are important for the interaction with bacteria.

The secretory immunoglobulins, most notably secretory IgA, act by aggregating bacteria. They target specific bacterial molecules, such as adhesins, or enzymes, such as glucosyl transferase. Studies of the correlation between secretory IgA levels and caries prevalence have reported conflicting results (Riviere and Papaginnoulis, 1987). The saliva also contains IgG and IgM from serum and local production in the gingival tissues.

The conflicting results of recent longitudinal clinical studies of the relative predictive values of antimicrobial salivary components for caries incidence (Tenovuo et al, 1997) may be attributed to the fact that dental caries is a multifactorial disease.

Fig 97 Physiologic model of the oral cavity, demonstrating the clearance of sucrose by saliva. C1= volume of saliva that has an upper level before swallowing (Vmax) and a lower level after swallowing (Vresid). C2= volume of integuments and plaque on teeth and mucosal membrane.(From Lagerlof and Oliveby, 1990. Reprinted with permission.)

Figs 98a to 98c Chairside testing of the salivary pH with the Dentobuff Strip system. One drop of stimulated saliva is placed on a test strip containing an acid and a pH indicator. After the reaction between the saliva and acid is completed, the color of the test pad is compared to a chart, and the pH value is determined. Fig 99 Frequency distribution of the buffering effects of saliva (final pH) in males and females. (From Heinze et al, 1983. Reprinted with permission.)

Fig 100 Mechanisms for enamel dissolution at different pH challenge and saturation of minerals.

Fig 101 Tooth surface after topical fluoride treatment (pH 7.0).

Fig 102 Tooth surface during cariogenic challenge (4.5 pH 5.5): Undersaturated with hydroxyapatite; supersaturated with fluorapatite.

Fig 103 Tooth surface after cariogenic challenge (pH 7.0).

Fig 104 Attachment of a gram-positive (G+) pioneer colonizer to the pellicle-covered tooth surface (white arrow) in contrast to the relationship between a gram-negative bacterium (G-) and the pellicle (black arrows). (From Lie, 1978. Reprinted with permission.)

Fig 105 Close-up of the attachment (arrows) between a gram-positive bacterium (B) and the pellicle (P). (From Lie, 1978. Reprinted with permission.)

Formation and functions of pellicle

Saliva is seldom in direct contact with the tooth surface but is separated from it by the acquired pellicle, defined as an acellular layer of salivary proteins and other macromolecules, approximately 10 um thick, adsorbed onto the enamel surface. It forms a base for subsequent adhesion of microorganisms, which under certain conditions may develop into dental plaque. The pellicle layer, although thin, has an important role in protecting the enamel from abrasion and attrition, but it also serves as a diffusion barrier.

Figure 106 shows that the undisturbed pellicle is formed in different layers. There are many nonattaching bacteria close to the outer surface of the pellicle. Because of abrasion, for example from toothbrushing, the thickness will vary between 2 and 10 um, depending on the toothbrushing intervals. Saxton (1976) showed that complete removal of the pellicle requires about 5 minutes' cleaning with pumice in a rotating rubber cup.

Figure 107 shows a groove made in the pellicle with a knife. Such grooves were earlier thought to be abrasive defects in the enamel surface resulting from the use of abrasive toothpaste. The pellicle shown in Figure 108 was removed by intensive cleaning for about 5 minutes. The pellicle-free enamel surface was partly covered with nail varnish, while the outer part was exposed to saliva in vivo for several hours. Figure 108 shows the thickness of the new pellicle compared to the naked enamel surface after removal of the nail varnish.

Movement of molecules by forces other than diffusion is less frequent in the pellicle than in most other parts of the salivary film. The relatively undisturbed layer of liquid in the pellicle influences the solubility behavior of the enamel surface. Adsorption to the enamel of macromo- lecules, usually originating from the saliva, is selective; certain macromolecules show a higher affinity for the mineral surface than do others.

In the normal oral pH range, the enamel surface has a negative net charge, because of the structure of hydroxyapatite, in which phosphate groups are arranged close to the surface. Count -erions (of opposite charge), for example, calcium, are attracted to the surface, forming a hydration layer of unevenly distributed charges. The exact composition of this layer will be determined by several factors (eg, pH, ionic strength, and the type of ions present in the saliva). Normally the hydration layer close to the enamel surface contains mainly calcium and phosphate ions in the proportion of 10:1, but other ions, such as sodium, potassium, fluoride, and chloride must also be present (see the formation of phosphate- and protein-coated CaF2 crystals in the pellicle; see Figs 101, 102, and 103). Because of the domination of calcium, the resulting net charge of the enamel surface with its hydration layer is positive, implying that the hydration layer will attract negatively charged macromolecules, as illustrated by Waerhaug more than 25 years ago (Figs 109 and 110).

Negative charges on macromolecules are found in acidic side chains with end groups of phosphate or sulfate. These side chains have a high affinity to the tooth surface. Recent research has shown that the bulk of the pellicle consists of salivary micelle-like structures of great importance for reducing diffusion through the pellicle and reducing friction between the teeth and other oral tissues.

Not all the proteins contributing to the pellicle are well defined. However, salivary proteins, such as amylase, lysozyme, peroxidase, IgA, IgG, and glycosyltransferase, participate in formation of the pellicle matrix, together with mucins and breakdown products from macromolecules of both salivary and bacterial origin. Of special interest are the acidic PRPs, mentioned earlier, which bind via their amino-terminal segments to the tooth surface, leaving their carboxy-terminal regions directed toward the oral cavity, where they may interact with oral microorganisms.

During the first hour, pellicle formation is rapid, and then decreases. It seems likely that adsorption of the first layer of molecules onto a clean surface is instantaneous. The formation rate varies among individuals, probably as a result of differences in salivary composition, the frequency of oral hygiene, and diet composition.

Fig 106 Undisturbed pellicle in cross section. It is formed in different layers. Many nonattaching bacteria (left) are close to the outer surface of the pellicle (right). (From Saxton, 1976. Reprinted with permission.)

Fig 107 Groove in the pellicle, down to the enamel surface. The groove was made with a knife. (From Saxton, 1976. Reprinted with permission.)

Fig 108 Thickness of new pellicle (right) compared with a naked enamel surface that had been protected by nail varnish (left). The groove shown in Fig 107 was removed by 5 minutes' intensive cleaning. The pellicle-free enamel surface was partly covered with nail varnish (left), while the other part (right) was exposed to saliva for several hours. The surface is shown after removal of the nail varnish. (From Saxton, 1976. Reprinted with permission.) Fig 109 Negatively charged salivary macromolecules. (Illustrated by J. Waerhaug. Courtesy Department of Periodontics, University of Oslo.)

Fig 110 Negatively charged salivary macromolecules attached to the positively charged enamel surface. (Illustrated by J. Waerhaug. Courtesy Department of Periodontics, University of Oslo.)

Salivary stimulation and substitution in patients with hyposalivation and xerostomia

Stimulation of saliva

Recognition of the key role of saliva in maintaining normal oral function has stimulated research on its protective properties against caries and on the treatment of xerostomia and salivary hypofunction. Salivary clearance, buffering power, and degree of saturation with respect to tooth mineral are the major protective properties (for review, see Sreebny et al, 1992; Tenuvuo, 1997), their effect increasing with salivary stimulation: The saliva stimulated by consumption of fermentable carbohydrates reduces the fall in plaque pH that could lead to demineralization and increases the potential for remineralization. When saliva is stimulated after carbohydrate intake, acids produced in the plaque are neutralized, and experimental lesions in enamel are remineralized. The pH-raising effects are more easily explained by the buffering action of stimulated saliva than by clearance of carbohydrates. Remineralization is dependent on the presence of fluoride.

These findings suggest that the protective properties of saliva can be potentiated by appropriate salivary stimulation. In addition to established procedures, such as diligent oral hygiene and fluoride regimens, general recommendations for caries prevention might therefore include eating patterns that stimulate secretion of saliva.

There are now a number of management options for protecting the oral cavity from the devastating effect of inadequate salivary function and for relieving the patient's discomfort. Treatment is determined by the degree of functional impairment. For the patient who has some remaining glandular function, stimulation of secretion is the optimal approach. Patients with negligible natural function can be offered symptomatic treatment to relieve oral dryness. For either patient category, selection of specific treatment measures is determined by a number of factors, including the patient's medical status. The practitioner must also be able to manage the complications of salivary hypofunction: increased incidence of caries, oral candidiasis, altered oral function, and pain. Stimulation of secretion, locally or systemically, has the great advantage of providing the benefits of natural saliva.

Systemic. There has been increasing interest in systemic pharmacologic stimulation of salivary function. Three agents have been studied in some detail: bromhexine hydrochloride, anethole trithione, and pilocarpine hydrochloride. All three should be used only under specialist supervision and following medical examination. Bromhexine is a mucolytic agent used in the management of chronic bronchitis. Its use in managing dryness of the eyes associated with Sjogren's syndrome is controversial. No beneficial effects on salivary dysfunction have been demonstrated.

Anethole trithione has been proposed as a treatment for salivary hypofunction caused by psychotropic drugs, radiation, and Sjogren's syndrome. Conflicting results have been reported on the efficiency of treatment. In one study, 74% of patients with Sjogren's syndrome had increases in the output of unstimulated whole saliva, whereas Swedish studies, on patients with more pronounced salivary dysfunction, have failed to show any improvement in salivary function. Patients with postradiation xerostomia showed no improvement following drug treatment, compared to controls. Further trials are necessary to delimit the ability of this drug to improve salivary function(for review see Pearce, 1991; Edgar et al, 1994).

Pilocarpine hydrocholoride is a parasympathomimetic drug that functions primarily as a muscarinic-cholinergic agonist with mild -adrenergic stimulatory properties. It has been used for more than 100 years as a potent stimulant of exocrine secretion. In the last decade, carefully controlled studies have shown that it can increase salivary output in normal volunteers and effectively relieve oral dryness in patients with salivary gland hypofunction. In a 6-month trial by Fox et al (1986) in patients with irradiation-induced salivary hypofunction and in patients with Sjogren's syndrome, 5 mg of pilocarpine, three times a day, was effective; side effects were well tolerated; and there were no significant alterations in heart rate, blood pressure, or electrocardiographic parameters. Greenspan and Daniels (1987) reported that pilocarpine treatment resulted in subjective and objective improvement in about 80% of patients with postradiation xerostomia. A synergistic effect on salivary stimulation from a combination of pilocarpine and anethole trithione was reported by Epstein and Schubert (1987).

Although pilocarpine appears to be the most effective systemic sialagogue currently available, it is of limited use in the management of salivary hypofunction. It is ineffective if insufficient functional tissue remains, as in advanced stages of Sjogren's syndrome or following head and neck radiotherapy. Possible interactions with other medications or potential adverse cardiovascular and pulmonary effects further limit patient eligibility. Further clinical studies are necessary to determine optimal doses, administration schedules, and systemic effects of pilocarpine. For the patient with an irreversible condition requiring long-term or lifelong management of dry mouth, a sustained-acting preparation would be ideal.

Local. Local stimulation is feasible, because the salivary glands are highly responsive to stimulation from taste, masticatory activity, and the sensory nerves of the mucosa and periodontium. Because the salivary secretion rate usually increases during meals (a physiologic response), an important first step to potentiate salivary flow should be a diet of fiber-rich, well-flavored aromatic food, such as fruit. In addition, finishing a meal with matured cheese (for instance, cheddar) has been shown to increase SSR and decrease plaque pH significantly (Imfeld, 1983).

In developed societies, the reduced masticatory activity required to chew highly processed modern foods may favor a measure of salivary hypofunction, because of disuse atrophy. Studies in both animals (Johnson and Sreebny, 1982) and humans (Axelsson et al, 1997a; Dodds et al, 1991; Jenkins and Edgar, 1989) have indicated that prolonged increases in levels of salivary stimulation after dietary changes result in greater salivary output. Adequate water intake is another prerequisite for normal salivation. For optimal prevention and control of dental caries, every meal containing easily fermented carbohydrates, particularly sucrose, should be followed immediately by supplementary local salivary stimulation to increase the sugar clearance, buffering effect, plaque pH, and the access of calcium and phosphate ions. As described earlier, demineralization is thereby decreased and remineralization is enhanced.

Local saliva-stimulating agents that contain fluoride will further enhance the remineralization potential significantly and should therefore be recommended in preference to similar agents without fluoride. Specially formulated lozenges and chewing gum, both with and without fluoride, are available. Because frequent intake is recommended (four to six times per day, directly after meals and snacks), it is important that these agents not contain sugar and not be potentially erosive. The sweetening agents commonly used in these products are sorbitol, xylitol, and saccharin, separately or in combination.

Fluoride lozenges containing 0.25, 0.50, 0.75, and 1.00 mg of fluoride are also commercially available. The 0.25-mg lozenges are recommended for children older than 5 years and 0.25 to 1.00 mg lozenges are recommended for selected young adults and adults, particularly in cases of hyposalivation, for combined salivary stimulation and fluoride delivery directly after meals. A recent study by Sjogren et al (1995) showed that in subjects with reduced SSR, fluoride lozenges and chewing gum not only improved the SSR, but also significantly prolonged fluoride clearance time, an important factor in prevention and control of caries in such patients, who are generally at high caries risk.

To date, the most promising system for local salivary stimulation is the recently introduced fluoride chewing gum (Fluorette and Fludent, used widely in Scandinavia). It is sugarless and sweetened with xylitol and sorbitol. Each piece contains 0.25 mg of fluoride. Chewing gum is unique because it is usually chewed for a prolonged periodaround 30 minutesbut its caloric value is almost negligible, both important characteristics in the context of salivary stimulation. Chewing gum has been shown to elicit a continued flow of saliva during prolonged mastication (Dawes and Macpherson, 1992), but the level of stimulus gradually declines: As the flavoring agents are released and swallowed, the gustatory stimulatory component is rapidly depleted, and the intensity of the masticatory stimulus abates, because of softening of the gum (Rosenhek et al, 1993). The effect of varying not only the duration of chewing, but also the interval elapsing between the caries challenge (drop in pH) and the subsequent gum chewing, has been studied. For maximum neutralization of the plaque pH, the gum must be chewed for at least 15 minutes, immediately following the caries challenge (Park et al, 1993).

Recently, Sjogren et al (1997) compared the effect of chewing fluoride gum for 5, 10, 15, 20, 30, or 45 minutes on approximal plaque pH and salivary fluoride concentration on the chewing and nonchewing sides. The subjects had undisturbed approximal plaque, 3 days old, and rinsed for 1 minute with 10 mL of 10% sucrose solution. The resultant fluoride concentrations were two to three times higher on the chewing side than on the nonchewing side (Fig 111). The best recovery in approximal plaque pH was also noted on the chewing side, but the difference was not as pronounced as for the salivary fluoride concentration. Significantly higher values of plaque pH were found during prolonged chewing (Fig 112), while variations in chewing duration caused only minor variations in salivary fluoride concentration. This study showed that to attain optimal fluoride- and plaque pH-raising effects throughout the entire dentition, the gum should be chewed for at least 20 minutes, using both sides of the mouth.

Studies in patients with low salivary flow rates (Abelson et al, 1990; Markovic et al, 1988) also showed that use of a sorbitol-sweetened gum raised the pH of the plaque on both enamel and root surfaces.

Two principal mechanisms are implicated in the plaque pH-raising effects of chewing gum or foods such as cheese: clearance of carbohydrates and buffering of plaque acids. For chewing gum, the relative contributions of these two factors were studied by Dawson (1993). After a sucrose mouthrinse, the test subjects chewed sugar-free or sucrose-sweetened gum; the control subjects did not chew any gum. Over 45 minutes, SSRs, pH, sugar concentrations, and bicarbonate levels were measured. The results were compared with plaque pH data noted under similar conditions (Manning and Edgar, 1993).

Both gums had pH-raising effects on plaque. The sugar-free gum accelerated clearance of the sucrose rinse, while the sugared gum released sugar, at potentially acidogenic concentrations, throughout the 45-minute period. The gums had a similar effect on salivary secretion rates, but the sugared gum resulted in lower levels for salivary pH and bicarbonate, indicating active participation of salivary bicarbonate in neutralizing and buffering acids produced in the mouth from the sugars released from the gum. Thus, the buffering effect was capable of overwhelming the acids formed

from sugars derived from sucrose-sweetened gum.

If the bicarbonate level of saliva is thus paramount in eliciting the pH-raising action of chewing gum (and, by analogy, other pH-raising foods), then it is important to note that the relationship between salivary flow rate and bicarbonate concentration is not linear, but approaches a maximum at an intermediate flow rate. Thus, increasing salivary flow above this rate would not be expected to result in a comparable increase in pH-raising action on plaque, contrary to what would be obtained if the major plaque pH-controlling factor were salivary clearance of sugars or acids. This point merits further study with various levels of stimulation of saliva and observation of the effects on plaque pH.

Sugar-free chewing gum containing urea (V6) is available in Europe and Scandinavia, and is claimed to exert plaque pH-raising effects superior to those of gum without added urea, presumably because of the increased synthesis of ammonia in plaque resulting from ureolysis. Recently, Imfeld et al (1995) used the telemetric technique to compare the effect of chewing xylitol or xylitol-carbamide (urea) gum on approximal plaque pH after a sucrose rinse. Figure 113 shows the extraordinary effect of the carbamide-containing gum compared to the xylitol gum and no chewing gum.

Together, these studies offer convincing evidence that, following consumption of fermentable carbohydrates, the chewing of sugarless gum rapidly elevates plaque pH toward resting levels, where it persists for the duration of the experiment. In addition to causing an increase in the buffering power by stimulating saliva, the chewing increases bicarbonate levels leading to an increase in salivary pH and thus to the degree of supersaturation of stimulated saliva with respect to calcium phosphate solids, including hydroxyapatite. The increase in the degree of supersaturation with calcium and phosphate leads to the conclusion that stimulated saliva can influence the equilibrium between demineralization and remineralization in the development of caries, not only by reducing the duration of demineralization resulting from the pH changes in plaque but also by enhancing the potential for remineralization.

This hypothesis was tested by Leach et al (1989). An intraoral device carrying a piece of partially demineralized human enamel, covered with gauze to encourage deposition of plaque, was attached to a mandibular first molar tooth in volunteers. The subjects used a sorbitol-sweetened chewing gum for 20 minutes after three meals and two sugary snacks each day. After 3 weeks, the enamel particle was replaced, and the subjects consumed the same meals and snacks but without using gum. The order (gum versus no gum) was reversed in half the subjects. The subjects continued to use their usual fluoride-containing dentifrice thoughout the study.

Analysis of the mineral content of the carieslike lesions after intraoral exposure showed a significantly greater increase in the mineral content after gum chewing than was found without gum chewing, indicating a potentially beneficial remineralizing effect of the stimulated saliva. The increase in remineralization after the use of gum could have occurred either because of a reduction in the degree of demineralization via the effect of gum chewing on plaque pH or because of an increase in remineralizing potential. These results should not be interpreted as indicating that early white-spot lesions in enamel can necessarily be completely remineralized through the use of sugar-free gum. Rather, the data indicate the possibility of

favorable alteration of the equilibrium between demineralization and remineralization, preventing the development of an initial carious lesion.

In such studies of salivary stimulation, the environment in which remineralization of the enamel lesion occurs is the fluid phase of plaque, which differs from saliva in pH and concentration of calcium and phosphate ions. While an increase in the supersaturation of saliva resulting from stimulation would be expected to have a direct effect on the supersaturation of plaque fluid, to date no such effect has been demonstrated. Sternberg et al (1992) found, however, that chewing of sorbitol and xylitol gum, besides reducing plaque accumulation and gingival inflammation, increased the concentration of acid-extractable calcium in plaque by more than one third. This would be expected to increase the remineralizing potential of the plaque, and it was suggested that the effect was due to complexing of calcium by both xylitol and sorbitol, leading to their retention in plaque. Equally, it may be that the elevation of plaque pH that would have followed gum chewing resulted in increased retention of calcium in plaque in the form of insoluble calcium phosphate deposits. Fluoride chewing gums should increase the reservoir of phosphate and protein-coated CaF2 crystals in the pellicle, as well as in any plaque that might remain.

In the remineralization studies already described, a therapeutic fluoride environment was provided by the use of a fluoride dentifrice. The essential role of fluoride in the remineralizing potential of sugared gum was shown in preliminary data. Subjects chewed sucrose gum for 20 minutes after meals and snacks for successive 21-day periods, during which the fluoride content of the dentifrice was varied between 0 and 1,000 ppm. In the presence of fluoride, some remineralization was observed, although it was statistically insignificant; with the nonfluoridated dentifrice, significant demineralization occurred (Manning and Edgar 1993). In caries-susceptible patients with impaired salivary secretion, use of the new sugarless fluoride chewing gum directly after meals is therefore a promising adjunctive measure. Compared to subjects with normal salivary flow, patients with reduced salivary flow have a fluoride clearance time that is fortuitously prolonged, as shown in the aforementioned study by Sjogren et al (1993).

The effect of a fluoride sugar-free chewing gum on stimulated salivary secretion rate, Plaque Index, Gingival Index, and Plaque Formation Rate Index, as well as on salivary mutans streptococci scores, was recently evaluated. The selected group of patients (n = 53) had less than 0.7 mL/min stimulated SSR (mean = 0.4 mL/min). The subjects were instructed to chew a stick of fluoride chewing gum (0.25 mg of fluoride) for 15 to 20 minutes after every meal (four to six times per day) for 6 months (Axelsson et al, 1997a).

The results showed an increase in the mean stimulated salivary secretion rate from 0.4 to 0.6 mL/min. This finding is important because it shows that through regular salivary stimulation with chewing gum after every meal, a gradual increase in salivary secretion is possible. An average reduction of about 35% was achieved for Plaque Index, Gingival Index, and Plaque Formation Rate Index (Fig 114). The percentage of subjects with salivary mutans streptococci scores of 0 to 3 at baseline and after 6 months is shown in Fig 115. There was a pronounced shift from high scores to low; in particular a decline in score 2 and an increase in score 1 (Axelsson et al, 1997a). These results indicate that regular use of a fluoride chewing gum after every meal

would have a very significant caries-controlling effect in patients with hyposalivation, over and above the primary effect of the fluoride release from the chewing gum.

In two recent studies on experimentally induced caries (enamel and root lesions), the remineralizating effect of fluoride chewing gum (0.1 mg of fluoride) used five times per day for 21 days was compared with an in situ slow-release fluoride device that released 0.5 mg of fluoride/day. A nonfluoride toothpaste was used for oral hygiene three times a day during the test period. The degree of remineralization of enamel lesions was 35.5% for the chewing gum and 34.0% for the slow-release device (Wang et al, 1993). On root lesions, De los Santos et al (1994) achieved similar results (36.0% and 35.8% for the gum and the device, respectively) for remineralization. However, the chewing gum resulted in higher stimulated SSR than did the fluoride-releasing device and the control (2.1, 1.8, and 1.7 mL/min, respectively) and higher mean salivary fluoride concentration during stimulation (3.0, 0.2, and less than 0.02 ppm of fluoride, respectively).

The potential clinical effect of saliva stimulation per se has not been tested in a clinical trial, although it is possible to interpret results from certain clinical studies as effects of salivary stimulation. Thus, the caries-preventive effects of xylitolincluding apparent reversals of carious lesions (remineralization) shown in the Turku chewing gum trial, when sucrose- or xylitol-sweetened gum was chewed ad libitum over a 12-month period (Scheinin et al, 1975)could be due to the enhanced remineralizing potential, although inhibition of plaque acidogenicity and other effects of xylitol cannot be excluded.

More conclusive indications of a beneficial effect of salivary stimulation are disclosed in studies by Moller and Poulsen (1973) in which chewing of sorbitol gum was associated with a small but significant reduction in caries incidence; by Isokangas et al (1989), in which significant long-term benefits of xylitol gum, used two or three times daily were shown; and by Kandelman and Gagnon (1990), in which a decrease of 65% in caries progression was found in children who chewed xylitol gum as part of a preventive program. It should be noted that the study designs did not stipulate routine postprandial chewing for 20 minutes, which would have maximized the influence on saliva.

The only direct comparison of xylitol and sorbitol gums appears to be the recently published 2-year study by Makinen et al (1996), in which South American children chewed sorbitol or xylitol gum daily. Compared with controls not using any gum, the observed reduction in caries incidence was greater for xylitol than for sorbitol gum, while an increase in caries incidence occurred in the children using sucrose gum. The caries onset risks for xylitol and sorbitol pellet chewing gum were 35% and 44%, respectively of that in the non-gum group. The effects were greater with pellet gums than with stick gums and with increasing frequency of gum chewing. Both xylitol and sorbitol mixtures in pellet form were associated with a caries onset rate comparable with that of the xylitol stick gum. The largest reduction in caries risk was observed in the group receiving xylitol pellet gum.

Thus, the protective action of saliva, essential for normal dental health, can be enhanced by stimulation through appropriate dietary manipulation and selection. Sugar-free chewing gum may be of particular value in stimulating salivation over a

prolonged period without increasing the energy content or acidogenicity of the diet. With the use of such gum after meals and snacks, for normal gum-chewing periods of 20 minutes or more, the effects of fluoride in favoring remineralization may be enhanced, and the benefits of salivary neutralization, buffering, and sugar clearance in opposing demineralization may be mobilized, as part of a program of prevention against caries.

The use of gum not only as a salivary stimulant but also as a vehicle for preventive agents may further extend its potential applications. Recently, a new chewing gum containing chlorhexidine (10 mg per stick, available in Scandinavia) has shown significant antiplaque effects (Smith et al, 1996; Tellefsen et al, 1996) comparable to 0.2% chlorhexidine mouthrinse. It seems possible that in the near future a combined chlorhexidine-fluoride chewing gum will become available. Meanwhile, concurrent chewing of one stick each of the fluoride and the chlorhexidine chewing gum for 20 minutes directly after every meal could be recommended for high-caries-risk patients with reduced salivary secretion rate, high or very high Plaque Formation Rate Index (score 4 or 5) and high salivary mutans streptococci levels. In this way, chemical plaque control, fluoride, and salivary stimulation will act synergistically at the crucial time; during the acid attack. In addition, because the stimulatory effect is related to the volume of the stimulating agent, an even greater effect on salivary stimulation should be achieved if twice the amount of gum is chewed.

Symptomatic therapy

In the absence of natural salivation, it is essential to try to protect the oral hard and soft tissues by salivary substitution. Saliva substitutes, also called artificial salivas, are frequently recommended for patients complaining of dry mouth (xerostomia).

Although many studies suggest that saliva substitutes are useful in the management of xerostomia, clinical experience has shown that these products are not well accepted by patients. Most patients do not continue to use the substitutes regularly, relying instead on water or other fluids to relieve their symptoms. One reason may be that most saliva substitutes are more viscous than natural saliva and may be uncomfortable for an individual with dry mucosal surfaces. Another reason may be that the need for frequent application to keep the mouth moist makes these substitutes inconvenient and expensive. Also, the artificial salivas fail to provide the broad spectrum of antimicrobial and other protective functions of natural saliva. There is a pressing need for more effective saliva substitutes and better delivery systems.

Meanwhile, frequent sips of water or other fluids for the relief of oral dryness are often as effective as saliva substitutes. Patients should be advised to carry fluids with them at all times. (The water bottles used by cyclists or plastic glasses with snap-on lids are convenient.) Often, this simple suggestion will bring substantial relief at minimal cost, will improve mucosal hydration, and ease swallowing and speaking. Individuals could (and should) be cautioned to avoid not only fluids containing sugar but also those containing alcohol or caffeine, as these too may worsen the xerostomia or increase the risk of caries.

A common complaint is dryness and cracking of the lips. If applied regularly, petroleum jelly-based compounds may be helpful. Patients may prefer lanolin-

containing creams, which help hydrate the tissues. Patients should be advised to use room humidifiers, especially at night, as an aid to relieving frequent symptoms of dryness of the throat and tongue. For institutionalized patients, demented, and other severely handicapped patients, a recently introduced aid is Saliswab (available in Europe), which acts as a combined salivary substitute and stimulating agent. In contrast to Lemon-Glycerin Swabs, it is not erosive.

The practitioner must be prepared to manage the complications of salivary hypofunction: increased caries, oral candidiasis, altered oral function, and pain. Initially, patients with xerostomia do not have extensive restorative treatment needs because it takes some time for clinical caries to develop. Therefore, it is important to diagnose impaired salivary function and xerostomia as early as possible and introduce intensive needs-related preventive programs before caries has developed.

In patients who have already developed several carious lesions, restorative treatment should be carried out in stages, beginning with excavation of caries and placement of provisional restorations using slow-release fluoride materials, such as glass-ionomer cements or resin-modified glass-ionomer cements, combined with an initially intensive, individually tailored preventive program. Once carious activity is under control, the next stage is definitive therapyrestorations in the form of complete crowns and fixed partial dentures. Most patients with severely impaired salivation and xerostomia should be regarded as lifelong high-caries-risk patients; they must continue on an intensive maintenance preventive program.

Fig 111 Mean fluoride concentrations on the chewing and nonchewing sides after use of fluoride chewing gum, by duration of chewing. *=P0.05; **=P0.01. (From Sjogren et al, 1997. Reprinted with permission.)

Fig 112 Mean plaque pH values on the chewing and nonchewing sides after use of fluoride chewing gum, by duration of chewing. *=P0.05. (From Sjogren et al, 1997. Reprinted with permission.) Fig 113 Approximal mean plaque pH values after a sucrose rinse, use of xylitol gum, and use of a combination xylitol-carbamide gum. (From Imfeld et al, 1995. Reprinted with permission.)

Fig 114 The percentage of tooth surfaces with disclosed plaque (PI), de novo reaccumulated plaque 24 hours after PMTC. PFRI and inflamed gingiva (GI) at baseline (BL) and after 6 months. (From Axelsson et al, 1997a.) Fig 115 The percentage of subjects with Strip mutans score 0, 1, 2, and 3 at baseline (BL) and after 6 months. (From Axelsson et al, 1997a.)

Preventive programs for patients with hyposalivation and xerostomia

For such high-caries-risk patients, most preventive measures, self-administered as well as professional, must be optimized. The following regimens are recommended.

Plaque control and self-administered fluoride

Patients with impaired salivation are extremely fast plaque formers (Plaque Formation Rate Index scores 4 and 5). Therefore, not only the frequency but also the quality of combined mechanical and chemical plaque control by self-care has to be optimized, and all tooth surfaces must be targeted. To reduce the risk of demineralization resulting from a fall in plaque pH during meals, mechanical removal of plaque, with a fluoride toothpaste that contains antiplaque agents, is recommended before every meal. Directly after the meal, one stick each of fluoride and chlorhexidine chewing gum should be chewed for 20 minutes. In patients with xerostomia and very sensitive oral mucosa, use of a fluoride toothpaste that does not contain sodium lauryl sulfate is recommended, as is application of a chlorhexidine-fluoride gel in customized trays for 5 minutes per day.

The diet should be mild, without spicy flavoring, and the patient should be advised to drink copious amounts of water. Sweets, sweet drinks, and confectionery should be sweetened with sugar substitutes.

Professional plaque control and use of fluorides

Needs-related intervals of professional mechanical toothcleaning are essential. After PMTC, the caries-susceptible tooth surfaces should be treated with slow-release fluoride and chlorhexidine varnishes. In a recent study, application of a 1:1 mixture of chlorhexidine varnish (Cervitec: 1% chlorhexidine and 1% thymol) and a fluoride varnish (Fluorprotector: 0.1% fluoride) significantly prolonged the depression of approximal mutans streptococci compared to the application of Cervitec alone (Twetman and Petersson, 1997). Although this effect is probably attributable to the superior retention of the fluoride varnish, for optimal prevention, the supplementary effect of slow fluoride release is essential (for reviews on saliva, see Edgar et al, 1994; Lagerlof and Oliveby, 1994; Pearce, 1991; Sreebny et al, 1992; Tenovuo, 1997; Tenovuo and Lagerlof, 1994; Tenovuo and Lumikari, 1991).

Role of Chronic Systemic Diseases and Impaired Host Factors

Chronic systemic disease

Of the systemic diseases, by far the greatest caries risk is associated with rheumatoid conditions, particularly Sjogren's syndrome, because of its severe depressive effect on the salivary secretion rate as well as the quality of the saliva. The most severe xerostomia is seen in patients with Sjogren's syndrome. Other systemic and chronic diseases that cause salivary gland hypofunction and xerostomia and are thereby regarded as risk factors and prognostic risk factors are listed in Box 9. Chronic diseases such as psychogenic disorders (eg, depression), allergies, hypertension, and so on, should also be regarded indirectly as risk factors and prognostic risk factors because of the associated long-term medication with significant depressive effects on SSR (see Box 10).

Some diseases impair the immune system generally, eg, leukemia or specifically acquired immunodeficiency syndrome (T cells). Other diseases, such as early-onset (juvenile) diabetes and Down's syndrome, are often associated with hypofunction of

the phagocytozing polymorphonuclear neutrophil leukocyte cells, which represent the first line of nonspecific defense in the gingival crevice.

Host resistance

In principle, dental caries, like other diseases caused by microorganisms, is characterized by microbial attack and host resistance; for dental caries, however, both processes are very complex and difficult to define. Thus, the microbial attack cannot be defined as the mere presence and activity of a certain pathogenic microorganism. Although there is substantial evidence that mutans streptococci (primarily Streptococcus mutans and Streptococcus sobrinus) are important etiologic agents of human dental caries, disease may develop without their presence. The number of cariogenic bacteria and the volume and microbial composition of the plaque in which they are present are also important factors. Furthermore, other factors, for example, the pattern of sugar intake, determine both the amount and nature of the acids released by mutans streptococci and other plaque bacteria, and thereby their cariogenic potential (see chapter 2).

Host resistance is even more difficult to define. Unlike resistance to many infectious diseases, resistance to caries is not determined solely by nonspecific antibacterial compounds and the alertness of the immune system at a given time. Nevertheless, an important conclusion can be drawn: Given the high potential for caries development in part of the population, the alertness developed by the immune system under natural conditions is inadequate for protection at the population level. That there are marked variations at the individual level, however, was shown in the Vipeholm study (Gustavsson et al, 1954); in the most extreme interventional test group, about 20% of the subjects failed to develop any new carious lesions (see chapter 2).

Several studies have shown (for review, see Kilian and Bratthall, 1994) that dental caries does not lead to acquired immunity. In this respect, dental caries is different from almost all other diseases caused by microorganisms. There may be two reasons. First, the mucosal immune system has evolved to maintain a natural balance with members of the commensal microflora of the body and not to eliminate them. Second, like some other members of the oral microflora, S mutans releases a protein with immunosuppressive properties.

Immune factors

The soft and hard tissues of the oral cavity are protected by both nonspecific and specific immune factors, which limit microbial colonization of the oral surfaces and prevent the penetration of noxious substances and ensuing damage to the underlying tissues. The nonspecific immune factors present in saliva include lysozyme, the lactoperoxidase system, lactoferrin, various little-known antibacterial compounds, and high-molecular weight glycoproteins and other salivary components, that may act as bacterial agglutinins (see Box 13). Unlike antibodies, these nonspecific factors lack immunologic memory and are not subject to specific stimulation. However, several of the nonspecific immune factors may interact with salivary immunoglobulins, resulting in a mutual amplification of their respective activities.

Infection and disease caused by cariogenic microorganisms occur in an environment

containing a variety of specific host immune factors, derived from several sources. The major and minor salivary glands constitute one such source, contributing essentially all the secretory IgA to whole saliva, together with lesser amounts of IgM and IgG. Secretory IgA mediates its protective effect mainly through primary binding of antigen. Binding can inactivate toxins, inhibit enzyme-based systems, and affect many other mechanisms involved in microbial colonization. Binding of several organisms results in their agglutination and consequent clearing from the mouth. Other immunoglobulin isotypes may also participate in these mechanisms.

An additional source of immune factors is the gingival crevicular fluid: This contributes most of the IgG, as well as some monomeric IgA. The crevicular fluid also contains many of the complement components and cell types that, together with IgG or IgM antibody, can inactivate or opsonize bacteria. Thus, the potential exists for several specific host immune mechanisms to intervene in the colonization and/or the pathogenic activity of cariogenic microorganisms. These specific host immune factors in whole saliva are assisted by the phagocytozing nonspecific polymorphonuclear neutrophil leukocyte cells migrating from the gingival crevice.

Numerous studies in animals and humans have shown that increased antibody levels, either secretory IgA or IgG, to S mutans can enhance its elimination and/or interfere with its cariogenic activity. However, the exact molecular mechanisms are only partly understood. Some of the potential mechanisms are summarized in Box 14.

On the other hand, the fact that people who lack a functioning secretory immune system have significantly more dental caries than do their age-matched counterparts seems to implicate salivary antibody in some aspect of the modification of the cariogenic potential of the oral microbiota. Therefore, individuals with complete secretory immunodeficiency (total absence of both secretory IgA and IgM) would constitute one group at greater risk of developing dental caries. There are currently no available analyses of caries experience in patients with IgG or selective IgG subclass deficiencies. Secretory IgA has a different molecular structure from serum IgA. While serum IgA occurs mainly in the classic monomeric form characteristic of other immunoglobulin classes, IgA in saliva and other exocrine secretions occurs as a larger complex, composed of dimeric IgA linked with a specific salivary glycoprotein, the secretory component. This complex is known as secretory IgA. Apart from providing active transport of the immunoglobulin molecule across secretory epithelia, the secretory component gives the secretory IgA molecule greater resistance to proteolytic enzymes than serum IgG, IgA, and IgM, thus enhancing its function in the enzymatically hostile environment of the oral cavity and other mucosal surfaces.

Two subclasses of IgAIgA1 and IgA2have been identified in human serum and secretions. The subclass IgA1 accounts for 80% to 90% of serum IgA, whereas in external secretions the proportions of IgA1 differ at different mucosal sites. Salivary secretory IgA usually contains 65% to 75% IgA1, but there are significant individual variations.

There is only limited information about differences in biologic functions of the two IgA subclasses. However, the distinction is important because IgA1 is susceptible to a group of bacterial enzymes, labeled IgA1 proteases, while IgA2 is resistant to such proteases. These enzymes are excreted by the streptococcal species that initiate dental

plaque formation: Streptococcus oralis, Streptococcus mitis, and Streptococcus sanguis.

The secretion of the IgA2 subclass is delayed in a significant number of infants, sometimes into the period of tooth eruption. The early deficit in the IgA2 subclass may limit the spectrum of antibody specificities available for reaction. In addition, S mitis and S sanguis each have strains that secrete IgA1 proteases; both streptococcal species are prominent colonizers of the mouth in the first year of life. If, because of poor oral hygiene, the early flora is rich in IgA1 protease-secreting strains, then the child could be at increased risk because IgA1-specific defense mechanisms would be susceptible to inactivation.

This may partly explain why, in the studies by Wendt (1995), discussed in chapter 2, children whose oral hygiene was poor at the age of 1 year developed multiple carious lesions during the following 2 years, while children whose teeth were clean at the age of 1 year because of regular cleaning by their parents remained caries free at the age of 3 years. However, in the first year of life, there is significant individual variation in the response to oral antigens. Longitudinal studies [au: Reference?]have also demonstrated that children develop responses to the same antigens at different rates. If responses are poor or of low affinity, or if responses to critical antigens are not mounted early enough, then a different sequence of colonization might occur, which could be detrimental to the host. Thus, the ability to identify the longitudinal development of immune responses to critical antigens could predict future caries risk.

Early immune responses to some antigens that may participate in plaque formation have been studied. Gahnberg et al (1985) measured the presence of salivary IgA antibody to glucosyltransferases from S sanguis and S mutans in the first 4 years of life. Glucosyltransferase was selected because it participates in the glucan-mediated processes of plaque formation by mutans streptococci. Streptococcus sanguis can be recovered from nearly all children by the end of their first year of life. Antibodies to S sanguis glucosyltransferase were detected by the end of the second year, but antibodies to S mutans glucosyltransferase could be detected in fewer than 15% of children as old as 4 years, even though more than 50% had been colonized by mutans streptococci.

The apparent delay in antibody formation to this potentially important antigen may diminish the immune protective effect, at least with respect to initial colonization. Longitudinal studies could reveal whether the children who had developed salivary antibody to this (or other) important antigens before significant colonization were also those who subsequently had little or no experience with caries. Such information could be vital in assessing future caries risk.

Because placental transfer of IgG antibodies may regulate the early immune responses of the offspring, elevated maternal levels of serum IgG (especially IgG1) antibody to critical colonization antigens may indicate decreased risk for dental caries in the primary dentition. Similarly, in contrast to bottle-feeding, breast-feeding allows the passive transfer of similar secretory IgA antibody specificities, which, if continued during the period of early challenges by cariogenic streptococci, may delay colonization and thus decrease the risk of disease.

Theoretically, it would seem that a vigorous set of immune responses, active during the period of initial colonization of newly erupted tooth surfaces, should influence the composition of the microbiota on those surfaces. A lag in the expansion of lymphocytes of a particular IgA subclass may delay the synthesis of antibody to certain types of important oral antigens and create an increased risk of caries. The early expansion of clones that will differentiate into plasma cells secreting salivary IgA or serum IgG antibody to bacterial components critical to some phase of pathogenesis may decrease eventual caries risk. Thus, early identification of these specificities may be predictive of future caries experience. The significance of the role of infective dose in early immune stimulation is unclear, but it is likely that artificial triggering of the secretory or systemic antibody compartment to synthesize appropriate antibody prior to infection would significantly modify the caries experience of the child.

While definitive immune predictors of caries risk are incompletely defined, Smith and Taubman (1991) have presented some guidelines (Table 14).

Future caries vaccine

To date there is no efficient vaccine against dental caries, particularly for early childhood, before colonization by the cariogenic microflora. The ideal determinant for use in caries vaccine would be one that induces antibodies that exert one or both of the following effects on S mutans and S sobrinus: (1) limit the colonization of the organisms in dental plaque; and (2) affect S mutans and S sobrinus in such a way that processes of importance for the development of caries (such as growth and production of acids and polysaccharides) are inhibited or reduced to a level not resulting in caries.

The most extensively studied bacterial components for development of a vaccine are the surface proteins that mediate contact with the pellclecoated tooth surface, and the glycosyltrans- ferase complex of enzymes that synthesize water-soluble and water-insoluble glucans from sucrose (for reviews on specific host immune response, see Brandtzaeg, 1989; Kilian and Bratthall, 1994; Kilian and Reinholdt, 1986; Smith and Taubman, 1991; Taubman and Smith, 1992).

Role of Tooth Size, Morphology, and Composition

Introduction

One approach to the prediction of future caries incidence is to study the tooth itself, allowing for the fact that the environment of the tooth will be equally important. The various aspects of tooth resistance then take on greater importance. With this approach, the individual or group of individuals showing resistance to caries can be identified.

Various aspects of the resistance of a tooth to dental caries can be described. The shape and size of the whole tooth may affect the degree of crowding and influence susceptibility to caries. However, other characteristics must also be considered. Because dental caries is initiated on the enamel surface, physical characteristics, such as a defective or rough enamel surface, and the chemistry of the enamel might also be

determinants of tooth resistance.

Physical characteristics of the tooth

Tooth size

Hunter (1967) studied the size of primary teeth in relation to past caries experience. He found that teeth that had been restored were significantly larger than untreated, caries-free teeth. This implies that the larger teeth were more caries susceptible, and indeed they were found to have been in the mouth for a shorter time because they erupted later than did smaller teeth. Grahnen and Ingervall (1963) had earlier noted a relationship between tooth width and caries resistance, smaller teeth being associated with lower caries incidence. Paynter and Gainger (1962) reported that the overall dimensions of teeth were smaller in areas with water fluoridation. Other studies have also reported an association between smaller teeth and lower caries prevalence: For example, Stern and Curzon (1975) found that size was related to geographic as well as individual caries prevalence. Similar findings have come from studies of naval recruits in the Untied States (Keene, 1971) and of stable island populations such as in Papua New Guinea (Schamschula et al, 1972a).

However, while on a population basis it can be shown that groups of individuals with low caries may have smaller teeth, this is hardly a useful tool for caries prediction in the individual patient because dental caries is such a multifactorial disease: The effect of tooth size would be negligible in comparison with the combined effect of other factors such as the quality of plaque control, use of fluoride, and dietary habits. However, tooth size may lead to extremely prolonged eruption time and subsequent rotation and tipping of teeth, which will increase plaque accumulation, hinder access for mechanical plaque control, and postpone occlusal contact and the beneficial frictional effect derived from chewing fiber-rich food.

Studies by Carvalho et al (1989) have indicated that de novo plaque reaccumulation over 48 hours is about five times greater in the fissures of erupting first molars than in fully erupted molars, particularly in the distal and central fossae. This explains why almost all fissure caries in molars is initiated during the long eruption time (12 to 18 months), while fissure caries in premolars, which have an eruption time of only 1 to 2 months, is rare. A further factor is the high susceptibility of the immature enamel to caries during eruption and the following year, until completion of secondary maturation.

Tooth morphology and cusp and fissure pattern

The pattern of carious and restored tooth surfaces varies significantly even in the primary dentition. In a toothbrushing population, caries susceptibility in the permanent dentition may be ranked in the following order:

1. Fissures of the molars.

2. Mesial and distal surfaces of the first molars.

3. Mesial surfaces of the second molars and distal surfaces of the second premolars.

4. Distal and mesial surfaces of the maxillary first premolars.

5. Distal surfaces of the canines and mesial surfaces of the mandibular first premolars.

6. Approximal surfaces of the maxillary incisors.

The pattern of carious and restored surfaces is related to the buccolingual width of the tooth crown, and to sites where dental plaque stagnates, that is, to toothbrush accessibility. In nontoothbrushing populations, or in individuals with poor and irregular oral hygiene habits (precluding regular use of fluoride toothpaste) and a high intake of sticky sugary food, cervical lesions may also develop on the buccal surfaces of the maxillary teeth and on the mandibular molars and premolars.

Although tooth morphology is basically similar among the races, some racial characteristics that predispose particular groups to dental caries have been identified. Plaque retention is enhanced by the presence of buccal pits, lingual pits on incisors, deep palatal grooves, or grooves within the Carabelli cusps. The dentition of the American Indians and Inuit (Eskimo) has shovel-shaped incisors, barrel-shaped incisors, and deep buccal pits on molars; plaque accumulation from a cariogenic diet will predispose these areas to caries (Mayhall, 1977). A similar feature predisposing to caries in the maxillary molars of whites is the higher prevalence of the Carabelli cusp (Dahlberg, 1961). The fissure pattern of permanent molars also varies with racial groups: Some teeth have deeply convoluted fissures prone to caries (Taylor, 1978), whereas in other teeth the fissures are shallow and almost imperceptible. The distribution of fissure patterns will vary among population groups, and even within a racial group, rendering teeth more or less susceptible to dental caries. The permanent first molar of the Inuit demonstrates this characteristic in particular (Mayhall, 1977).

One congenital condition resulting in characteristically small teeth and few molar fissures is Down's syndrome (Brown and Schodel, 1976).

The pattern of cusp form and fissure pattern is genetically determined but is probably of minor importance in caries resistance (or susceptibility). In an individual patient, tooth morphology may be an aid in deciding whether or not a tooth should receive a fissure sealant or not. Attempts to classify the shape of fissures so that they might be assessed as more or less caries prone were made many years ago (Bossert, 1937).

Furthermore, fissure pattern and its relation to structure within the depth of enamel is highly variable (Mortimer, 1964). However, based on the cross-sectional shape of the fissures of first molars, it has been found that most of the teeth (almost 90%) have so-called normal fissures, in cross section they have a relatively wide opening, followed by a narrow cleft, approximately 1.0 mm deep (width 0.1 mm), reaching almost to the dentinoenamel junction (Fig 116). The carious lesion usually starts as an enamel lesion on both sides of the entrance of the fissure, which is visible and accessible to a probe. However, some atypical fissures (fewer than 10%) with a narrow opening and a bulbous widening at the base should be regarded as risk fissures (Fig 117), because a lesion can start at the base as well as at the entrance to the fissure. Fortunately, from a diagnostic point of view, there is a strong correlation between steep cuspal inclination and such sticky risk fissures.

However, even extreme risk fissures with irregularities such as "horizontal tunnels" can be kept free of caries, as shown in Fig 118 (my daughter Eva at the age of 10 years). These first molar fissures were successfully maintained free of caries, without fissure sealants, by diligent oral hygiene and daily use of fluoride toothpaste (Fig 119). The many genetic variations in the shape of the tooth crown, vertical and horizontal, will also influence plaque retention.

Fig 116 Cross section of a molar fissure.

Fig 117 Atypical fissure with a narrow opening and a bulbous widening at the base. This type of fissure should be considered an at-risk fissure.

Fig 118 Fissures with horizontal tunnels in a 10-year-old girl. These irregularities put the first molar at extreme risk of developing caries.

Fig 119 Same individual as in Fig 118, caries and gingivitis free, at the age of 38 years. Meticulous toothcleaning has prevented oral disease, even in the absence of preventive measures such as fissure sealants.

Enamel structure

Enamel development is conventionally described in five histologically recognizable stages: secretion (matrix deposition and transition), cell organization, preabsorption, early maturation, and late maturation. In later work, only four stages are described, based on chemical composition. Developmental disturbances may occur at any stage. There is, however, no clear clinical evidence that aberrations in enamel structure affect resistance to dental caries, unless the defects are major and result in rough surfaces that enhance plaque retention. Trace elements may be essential for the development of normal enamel structure. For example, selenium is essential for proper matrix formation, but an excess may adversely affect calcification. Thus, either an excess or a deficiency of selenium, or other trace elements, may affect tooth resistance. This will be considered later, in more detail.

Opacities are isolated disturbances of the internal structure of enamel, appearing clinically as white patches, with or without an associated defect in surface contour. Opacities are common on the facial surfaces of permanent maxillary incisors and may be sequelae to trauma to the primary incisors. Some have demonstrated that opacities may be associated with water composition. Because opacities affect only small areas of a tooth, they will not influence caries resistance (Weatherell et al, 1977). On the contrary, whereas there is limited posteruptive uptake of fluoride in sound enamel, fluoride is found in relatively high concentrations in opacities, probably because of

the greater porosity.

So-called Turner's teeth, malformed premolars resulting from infection in the preceding primary molar, are caries susceptible. The condition is, however, too uncommon to be used in identifying susceptible individuals.

So-called mottled enamel involves most teeth in the mouth of an affected individual. It arises from a disturbance of enamel formation at one or several of the developmental stages and appears to be systemic rather than local in origin. The four stages of enamel development based on chemical composition are (1) the secretion of a partially mineralized matrix; (2) selective withdrawal of amelogenin components; (3) massive selective loss of amelogenins with maturation and selective mineralization; and (4) the production of hard, mature enamel. Mottling of enamel that is caused by high levels of ingested fluoride is related to the presence of less well-mineralized enamel on a per volume basis, resulting from a disturbance of stage 3. Incorporation of fluoride into the enamel may produce hypomineralized subsurface layers. Mottled enamel as a result of ingestion of relatively high levels of fluoride has been investigated intensively since Dean and Elvove (1935) showed that it was related to the concomitant presence of low dental caries experience.

Studies show that caries prevalence is lower in groups of individuals with fluorosis than it is in nonfluorotic groups (Axelsson and El Tabakk, 2000c), not because fluorosis prevents caries, but because individuals with fluorosis will continue to benefit from the posteruptive effect of fluoride as long as they continue to live in an area with high water fluoride concentration. Today it is well known that the caries-preventive effect of fluoride is almost 100% posteruptive. However, in communities in the United States with up to four times the optimal level of fluoride in the drinking water, caries protection from the fluoride is compromised by severe fluorosis. If the fluorosis becomes so severe as to cause substantial hypoplasia, possibly aggravated by posteruptive loss of enamel, then the prevalence of dental caries starts to increase again because of exposed dentin and increased plaque retention.

Although dental researchers have shown that mottling could be ascribed to the presence of dietary fluoride, other trace elements might also be responsible, and by extrapolation there might also be an association between other elements and caries susceptibility. Data on the trace element composition of various US water supplies, when matched with those for dental caries prevalence and enamel mottling in the areas supplied, show almost as significant a relationship between dental caries and the concentration of strontium as that between caries and fluoride. Other trace elements may also give rise to tooth resistance and enamel mottling. Elements such as strontium could give rise to enamel mottling identical to that produced by fluoride. Defects of enamel have been shown to occur with high levels of yttrium and lithium.

In humans, strontium is the only trace element other than fluoride shown to be associated with increased enamel mottling (Curzon and Spector, 1977). Children consuming high levels of strontium at 33 ppm (mg/L) had a significantly higher prevalence of enamel mottling than did those using drinking water with lower strontium concentrations. The level of mottling was greater than that expected from the concentration of fluoride in the drinking water (1.0 to 1.2 ppm). Dental caries was found to be exceptionally low in these communities (Curzon et al, 1978), and it was

suggested that this was the result of the combination of fluoride and strontium in the drinking water. The combination of the two trace elements could be said to be a marker of considerable tooth resistance to dental caries as well as associated with increased enamel mottling.

Although mild-to-moderate fluoride mottling of the enamel is associated with resistance to caries on a population basis, at the individual level it is at best a rough indicator of resistance. It cannot be overemphasized, however, that the overwhelming caries-preventive effect of systemic administration of fluoride from drinking water is posteruptive and that dental caries is a multifactorial disease. In most developing countries with natural fluoride and other trace elements in the drinking water, the climate is tropical or subtropical, and the frequency of water intake (topical effect) and the daily volume of drinking water consumed (daily systemic dose) are high compared to those in regions with temperate climates. In most developing countries, particularly Africa and Asia, the normal diet (vegetables, rice, and fish) has low cariogenicity. It is of interest to note that most studies on the relationship between mottled enamel and caries prevalence have been carried out in such regions.

Enamel chemistry

Enamel mottling apart, the fact that fluoride affects dental caries has been confirmed by many well-controlled studies of topical fluoride agents and studies of the posteruptive caries-preventive mechanisms of fluoride (for review, see Fejerskov et al, 1996a, b).

For many years, it was believed that incorporation of fluoride into enamel increased the resistance of the tooth to dissolution and that the surface enamel fluoride concentration could be a marker of tooth resistance or susceptibility to caries. This was determined by enamel biopsy; a small area of enamel was dissolved or abraded away and then analyzed for fluoride content. A number of variations on this technique have been described and used to associate enamel fluoride with tooth resistance.

For ease of access, the enamel biopsy was taken from buccal smooth surfaces, which are not susceptible to caries. Smooth-surface caries arises mainly on approximal areas below the contact point of the molars and premolars, which are impossible to biopsy in vivo. In studies on enamel surface fluoride, it is assumed the biopsy actually gives a reliable measure, but the fluoride concentration can vary markedly from one small area to the next, even within the same tooth (Weatherell et al, 1977). It would appear, therefore, that statistically significant relationships between surface enamel fluoride and caries cannot be demonstrated on an individual basis, and in studies conducted some decades ago it is not surprising that no consistent results were reported.

It is highly likely that in the populations studied, largely in the United States, other fluoride mechanisms have been far more important, for example, the action of fluoride on plaque and bacterial metabolism, coupled with its acknowledged role in remineralization. Fluoride in surface enamel may have a slow-release or reservoir function, releasing fluoride ions to retard enamel demineralization and enhance remineralization. Simple biopsy of enamel for fluoride remains, therefore, only a very weak marker of tooth resistance on both an individual and a population basis.

The relationship between other trace elements in whole enamel and caries prevalence has also been evaluated. In the largest of these studies (almost 500 enamel samples) more than 30 trace elements were evaluated. The results showed significant positive correlations with the caries prevalence of the tooth donor for manganese, copper, and cadmium; significant negative correlations were found for aluminum, iron, selenium, and strontium (Curzon and Croker, 1978). Studies based on whole enamel do not, however, serve as a good indicator of tooth resistance on an individual basis. Because caries is initiated at the enamel surface, surface trace element concentrations would, as with fluoride, be more appropriate.

It is now recognized this is a very dynamic process, with fluoride and other trace elements passing in and out of the surface enamel. The total concentration of an element in enamel is therefore less important than its availability at the tooth surface during periods of cariogenic challenge. Simple analysis of enamel, surface or whole, is therefore no longer considered relevant to tooth resistance.

Nor is solubility directly related to dental caries. In addition, the solubility of enamel is dependent not only on fluoride content but also on other factors, such as carbonate content, many trace elements, and various other inorganic components. For example, experiments have shown that test samples of enamel from sharks, consisting of almost 100% fluorapatite, are more easily decayed than is well-matured enamel from elderly humans (Ogaard et al, 1988).

Morphology of the cementoenamel junction

An exposed cementoenamel junction is a potential area of plaque retention (Figs 120a and 120b), and this is one reason why root caries tends to develop along the cementoenamel junction (Fig 121). However, wide variations in the morphology of the cementoenamel junction may also contribute to caries susceptibility (for review, see Bevenius, 1994).

Fig 120a Exposed root surface showing bacteria colonization along the gingival margin and the cementoenamel junction (arrow).

Fig 120b MS replica from the plaque retentive cementoenamel junction showing bacterial colonies. The colonies in the circle were identified as S sobrinus. (From Bevenius et al, 1994. Reprinted with permission.) Fig 121 Plaque-retentive active cavitated root caries along the cementoenamel junction. (Courtesy B. Nyvad and Munksgaard.)

Exposure of root surfaces

In the young, healthy adult, root surfaces, like the cementoenamel junctions, are not exposed to the oral cavity. At the population level, the prevalence of exposed root surfaces is strictly age related and is attributed to the long-term effects of trauma from toothbrushing (buccal surfaces) and gingival recession associated with periodontal

disease. With the decline in prevalence and severity of enamel caries, and hence the preservation of an intact dentition into old age, root caries is becoming an increasing problem in clinical practice. A discussion of tooth resistance must therefore include possible predictors of the development of root caries in the individual tooth.

Morphologically, the surface of intact cementum and the cementoenamel junction are very rough, compared to the enamel surface (Fig 122). Because the rough surface is highly retentive to plaqueboth supragingival and subgingival (Fig 123)so-called root planing is emphasized as an important phase of scaling procedures in periodontal therapy. The roughness of the intact root cementum may vary from individual to individual as well as between different tooth surfaces.

The prevalence of cementum hypoplasia may also differ, not only between populations and individuals, but also symmetrically between different teeth. For example, subjects with localized early-onset periodontitis have a high prevalence (Lindskog and Blomlof, 1983) of cementum hypoplasia in first molars and central incisors (Fig 124). Such disturbances of cementum formation should be symmetric, because the cementum of all first molars and central incisors forms during the same period. It may be speculated that excessive doses of, for example, fluoride result not only in enamel hypoplasia (fluorosis) but also in cementum hypoplasia.

If root dentin is exposed by cementum hypoplasia or by aggressive removal of root cementum by scaling, bacteria may migrate via the dentinal tubules into the pulp (Adriaens et al, 1986, 1988a, b). Other studies have shown that if the root cementum is absent, bacteria from infected root canals may migrate to the root surface, initiating or maintaining local periodontitis (Jansson et al, 1995; Ehnevid et al, 1995a).

It should also be noted that root cementum and dentin are quite different from enamel in chemical composition: organic components constitute less than 1% of enamel but 35% to 40% of the total volume of root cementum and dentin. Not only the etiology but also the histopathology of root caries is therefore different from enamel caries: A synergistic effect of acidogenic bacteria and bacteria that produce proteolytic enzymes (mainly collagenases) has been proposed.

Fig 122 Section demonstrating that the root cementum (C) is much rougher than the enamel (E). The cementoenamel junction (EC) also has great plaque-retentive potential.

Fig 123 Cross section demonstrating attachment of subgingival microflora to the rough outer surface of the root cementum. AC = acellular root cementum. (From Listgarten, 1976. Reprinted with permission.) Fig 124 Cementum hypoplasia (right) and intact root cementum (left) in an individual with localized early-onset periodontitis. (Courtesy S. Lindskog.)

Conclusions

Introduction

The most important internal modifying factors related to dental caries are salivary hypofunction, some chronic diseases, impaired host factors, and unfavorable macroanatomy and microanatomy and eruption stage of the teeth that favor plaque retention. Of utmost importance is impaired salivary function, particularly stimulated salivary secretion rate.

Salivary factors

Salivary secretion rate, the buffering effect, and possibly the in vivo concentrations of some salivary constituents, such as fluoride, hypothiocyanite, and agglutinins (possibly including IgA), seem to be the most important determinants of caries susceptibility and/or activity.

About 0.5 to 1.0 L of saliva is produced per day. Most (about 80%) is secreted by stimulation during meals and only a minimal amount is produced during sleep; changes in the SSR reflect physiologic demand. The normal ranges for unstimulated and stimulated whole saliva are 0.25 to 0.35 and 1.00 to 3.00 mL/min, respectively. A rate of less than 0.1 mL/min for unstimulated saliva and 0.7mL/min for stimulated saliva is regarded as hyposalivation and is associated with significantly increased caries risk. Epidemiologic studies have shown about 20% to 25% of the population older than 50 years of age to have stimulated SSR values of less than 0.7 mL/min and a higher caries prevalence than people with normal SSR.

Oral symptoms associated with salivary gland hypofunction include dry mouth (xerostomia), which is the most common symptom among elderly people with reduced unstimulated SSR. Other typical oral symptoms are difficulty in swallowing and eating dry foods and burning or tingling sensations, especially on the tongue. Extreme hyposalivation and xerostomia (in systemic disease such as Sjogren's syndrome) may be associated with extraoral symptoms, such as dryness of the eyes, the nose, and the vagina.

Among principal causes of salivary gland hypofunction and xerostomia, the most common is the long-term use of medicine or drugs, in particular psychotherapeutic agents, antihistamines, antihypertensive and diuretic agents, and some analgesics. The most severe hyposalivation and xerostomia is associated with therapeutic irradiation to the head and neck area and some systemic diseases, particularly Sjogren's syndrome.

The principle properties of saliva that protect the teeth against caries are the clearance and dilution of oral microorganisms, dietary sugars, and so on; the neutralizing and buffering of the acids in plaque; the supply of ions for remineralization; and endogenic as well as exogenic antiplaque and antimicrobial factors. The most important of these functions is the clearance of oral microorganisms and food components (particularly sugar) from the mouth to the gut; effective clearance is dependent on an abundant flow of stimulated saliva. The clearance and dilution effects of saliva on sugar reduce acid formation in plaque.

Provided that the plaque is not impenetrable, saliva also significantly reduces the risk of demineralization of the tooth surface by its neutralizing and buffering effect on

plaque acid. In individuals who clean all tooth surfaces once or twice a day, no such thick plaque should accumulate. The buffering effect is strongly correlated to the stimulated SSR, because the most important buffer is the bicarbonate system, mostly originating from stimulated saliva of the parotid gland.

The physicochemical integrity of dental enamel in the oral environment is entirely dependent on the composition and chemical behavior of the surrounding fluids (saliva and plaque fluids). The main factors governing the stability of enamel apatite are pH and the free active concentrations of calcium, phosphate, and fluoride in solution, all of which can be derived from saliva. Under normal conditions, in the absence of thick, undisturbed plaque and/or very high frequency of acidic dietary products, teeth do not dissolve in saliva, because saliva is supersaturated with calcium, phosphate, and hydroxyl ions, which constitute the mineral salts of the tooth. In individuals with a regular supply of fluoride, eg, daily use of fluoride toothpaste, both the saliva and the plaque fluid will also contain abundant fluoride ions. In the dynamic equilibrium of the carious process, the supersaturation of saliva provides a barrier for demineralization and a driving force for remineralization.

The saliva contains many different organic macromolecules and some other small organic proteins that together protect the oral cavity from infection as well as wear, dryness, and erosion. Among them are the antimicrobial lactoferrin, peroxidases, and agglutinins as well as secretory IgA. In isolation, the effect of any of the above components is weak, and future research should be directed toward combinations or clusters of salivary components or variables typical of caries-active and caries-inactive individuals and populations. Instead of individual salivary variables, it is likely that some functional measures of saliva, such as bacterial aggregation rate, promotion of bacterial adherence to saliva-coated hydroxyapatite, bacterial growth and enzyme inhibition, and cell killing are more closely related to carious activity. Each of these properties may be mediated by multiple proteins and each varies among human subjects.

Saliva is seldom in direct contact with the tooth surface but is separated from it by the acquired pellicle, an acellular layer of adsorbed salivary proteins and other macromolecules on the enamel surface. This thin layer forms the base for subsequent adhesion of microorganisms, which under certain conditions may develop into dental plaque. Although thin, the pellicle is important in protecting the enamel against abrasion and attrition and serves as a diffusion barrier.

The protective properties of saliva that increase on stimulation include salivary clearance, buffering power, and degree of saturation with respect to tooth mineral: The maximum effect occurs when saliva is stimulated after the consumption of fermentable carbohydrates, reducing the fall in plaque pH that normally leads to demineralization and increasing the potential for remineralization. Production of plaque acid is neutralized and experimental lesions in enamel are remineralized when saliva is stimulated after a carbohydrate intake. The pH-raising effects are more easily explained by the buffering action of the stimulated saliva than by clearance of carbohydrates. Remineralization is dependent on the presence of fluoride in the saliva. Because the protective effect of saliva can be mobilized by appropriate salivary stimulation, a combination of frequent topical application of fluoride and salivary stimulation is of great importance in patients with hyposalivation and xerostomia.

Therapeutic stimulation of secretion, whether systemic or local, has the great advantage of providing the benefits of natural saliva. Although pilocarpine hydrochloride appears to be the most effective systemic sialagogue presently available, its usefulness in the management of salivary hypofunction is limited. Frequent and adequate intake of water is also important in therapeutic stimulation.

Physiologically, saliva is stimulated by fiber-rich, well-flavored, aromatic food. The most promising local saliva-stimulating agents for caries prevention are the recently introduced fluoride chewing gums, to be chewed for 20 minutes after every meal. A chewing gum containing chlorhexidine is also now available. Combining fluoride and chlorhexidine chewing gums would not only improve salivary stimulation but also prolong fluoride clearance time and provide chemical plaque control, directly after the acid attack.

Patients without sufficient residual salivary function to benefit from attempts to stimulate natural secretion, are offered symptomatic treatment designed to relieve oral dryness. For such patients, artificial saliva that contains fluoride should be recommended as a supplement to frequent water intake.

Systemic and immunologic factors

Of the chronic systemic diseases, by far the most important risk factor and prognostic risk factor for dental caries is Sjogren's syndrome, because of its extremely depressive effect on both the salivary secretion rate and the quality of the saliva. Indirectly, reduced SSR is associated with other chronic diseases in which medical management involves regular use of drugs with side effects on the salivary system. Some other general chronic diseases, such as leukemia, acquired immunodeficiency syndrome, diabetes mellitus, and Down's syndrome, impair the immune system generally or specifically.

Several studies have shown that dental caries does not lead to acquired immunity. However, the soft and hard tissues of the oral cavity are protected by both nonspecific and specific immune factors. The nonspecific immune factors present in saliva include lysozyme, the lactoperoxidase system, lactoferrin, various little-known antibacterial compounds, high-molecular weight glycoproteins, and other salivary components that may act as bacterial agglutinins. Even phagocytozing polymorphonuclear neutrophil leukocyte cells derived from the gingival crevice represent the nonspecific immune system in the oral cavity. In whole saliva, the following specific immune factors are present: secretory IgA, IgG, IgM, and serum IgA. In both secretory and serum IgA two subclassesIgA1 and IgA2have been identified.

Because mutans streptococci, the most cariogenic bacteria, do not colonize the mouths of infants and toddlers until after significant maturation and expansion of the host immune networks, such colonization should be vulnerable to specific (and innate) host immune mechanisms. If there is delayed development of one or more of these mechanisms, then the child is at increased risk for early colonization by mutans streptococci.

Although definitive immune predictors of caries risk are incompletely defined, the absence of a functioning secretory immune system appears to be a primary indicator of increased risk. Because placental transfer of IgG antibody may regulate the early immune responses of the offspring, elevated maternal levels of serum IgG antibody (especially IgG1) to critical colonization antigens may indicate decreased risk for dental caries in the primary dentition. Similarly, the passive transfer of similar secretory IgA antibody specificities during breast-feeding, if continued during the period of early challenges by cariogenic streptococci, may delay colonization and thus decrease future caries risk.

Theoretically, a vigorous set of immune responses, active during the period of initial colonization of newly erupted tooth surfaces, should influence selection of the colonizing bacteria. A lag in the expansion of lymphocytes of a particular IgA subclass may delay the synthesis of antibody to certain types of important oral antigens. Because of poor oral hygiene in 1 to 3 year olds, a flora rich in organisms secreting IgA1 proteases (S sanguis and S mitis) could decrease the protective potential of this IgA subclass. Both phenomena might create an increased risk of disease.

The importance of the role of infective dose in early immune stimulation is still unclear, but it is likely that artificial triggering of the secretory or systemic antibody compartment to synthesize appropriate antibody prior to infection would significantly modify the caries experience of the child. Many questions remain unanswered in the understanding of initial colonization and the participation of the host in modulating this process:

1. What are the antigenic components of the oral streptococci critical for effective immune responses?

2. To what extent are these components immunogenic and do they favor the formation of antibody of a particular subclass?

3. What is the nature of the interplay between specific serum and salivary immune components and innate host factors in these processes?

Answers to these questions may allow us not only to assess risk more clearly but also to accelerate the processes resulting in caries prevention.

Tooth-related factors

Physical characteristics of the teeth may increase the risk for caries: tooth size, tooth morphology, cusp and fissure patterns, enamel structure (defects, opacities, mottling, and roughness of the surface), the morphology of the cementoenamel junction, and exposed root surfaces. In addition, the chemistry of the enamel, dentin, and root cementum may influence caries susceptibility.

Studies to date indicate that large teeth in crowded mouths are more likely to develop caries, but this cannot be predicted on an individual basis. Similarly, certain types of crown morphology (shovel-shaped incisors, deep convoluted fissures, buccal pits, and palatal grooves) render a tooth more caries prone if the diet is cariogenic, because

they allow accumulation of undisturbed plaque. Structural defects of enamel allow cariogenic bacteria to colonize undisturbed and will also predispose a tooth to caries. Mottling, caused by fluoride, is associated with caries resistance, except when severe enough to cause substantial hypoplasia or loss of enamel, creating areas of plaque retention. The above factors increase caries susceptibility mainly by creating microenvironments for retention and stagnation of plaque.

The concentrations of fluoride and other trace elements in whole and surface enamel have been shown to be related to tooth resistance. However, chemical analysis, particularly of surface enamel, is only a weak predictor of tooth resistance. For surface enamel fluoride, the predictive value reaches statistical significance only in large samples on a population basis and has limited practical application for identification and counseling of the high-risk patient. Far more important are the presence of fluoride in the plaque, pellicle, and enamel fluids and the presence of reservoirs of CaF2 in the pellicle. The enamel of erupting teeth is far more caries susceptible than enamel exposed to the oral environment for some years (after secondary maturation).

Root caries is an increasing problem of the aging dentition, and to date there is no reliable predictor of tooth resistance. Mandibular posterior teeth are the most susceptible, and the use of fluoride lessens the chance of dentin caries.

Finally, prediction based on tooth resistance on an individual or a population basis remains unsatisfactory, because of the multifactorial nature of the disease.

Chapter 4. Prediction of Caries Risk and Risk Profiles

Introduction

For successful prevention and control of dental caries in both children and adults, some basic principles must be adopted: For example, the higher the risk of developing caries (new carious surfaces) in most of the population, the greater the effect of one single preventive measure. This may be illustrated by the Swedish experience, where 30 to 35 years ago, caries prevalence was extremely high. Almost every child developed several new lesions every year, mainly because of very poor oral hygiene. Regular toothbrushing was not an established habit, and no effective fluoride toothpaste was available. Under the prevailing conditions, well-organized, school-based fluoride mouthrinse programs in which 0.2% sodium fluoride solution was provided once every 1 or 2 weeks resulted in caries reductions of 30% to 50% (Forsman, 1965; Torell and Ericsson, 1965).

Twenty years later, a 3-year double-blind study revealed no benefit from weekly rinsing with sodium fluoride solutions compared to rinsing with distilled water (Axelsson et al, 1987). There had been a dramatic fall in both caries prevalence and caries incidence from 1964 to 1984, and particularly after 1974, following the introduction of needs-related preventive programs. As an analogy, a raincoat is very cost effective for a week in London in November but not for a visit to the Sahara Desert.

Similarly, at the population level, it is easy to find a positive correlation between one single risk factor and caries incidence in populations with high caries prevalence and incidence, where almost 100% of individuals develop new carious lesions every year. In such populations, the so-called whole population strategy will be cost effective. However, in populations with low or moderate caries incidence, well-established self-care habits, and well-organized oral health care, administration of one single preventive measure to all subjects in the population, irrespective of predicted risk, will not be cost effective; individual risk prediction and needs-related combinations of preventive measures are necessary. To ensure high sensitivity of risk prediction, several etiologic and modifying risk factors must be combined. For cost effectiveness, the so-called high-risk strategy would be the method of choice.

These two conditions may be exemplified by the following: The Vipeholm study (Gustavsson et al, 1954) confirmed 40 years ago that, in the absence of oral hygiene and fluoride, prolonged sugar clearance time was an external modifying risk factor for caries development in mentally handicapped people with heavy plaque accumulation. The daily intake of sugar-containing products in Sweden has remained unaltered over the last 40 years (about 120 g per individual per day), and the percentage of sticky sugary products consumer, such as sweets and cakes, has actually increased. However, over the past two decades studies have repeatedly failed to find any correlation between the intake of sugar-containing products and caries prevalence in the population (Sundin et al, 1983; Kristoffersson et al, 1986)

Caries prevalence and caries incidence have declined dramatically as a result of integration of caries-preventive measures by self-care, supported by needs-related professional treatment. Particularly successful has been the integration of a self-care program of excellent plaque control and the use of fluoride toothpaste, with professional mechanical toothcleaning and fluoride varnish at needs-related intervals. In a totally integrated caries-preventive program, however, external modifying risk factors such as a high frequency of sugar intake should also be addressed.

The risk for caries development varies significantly for different age groups, individuals, teeth, and tooth surfaces. Therefore, caries-preventive measures must be integrated and must be based on predicted risk. As an analogy, a medium-sized suit would not fit all the men in the world; it would be a reasonable fit for at most 40%, but too small for 30% and too large for the remaining 30%.

In this chapter, prediction of caries risk at the group, individual, tooth, and surface levels will be discussed in the context of the high-risk strategy. Methods for prediction of caries risk based on etiologic factors by the combination of salivary mutans streptococci (MS) levels and Plaque Formation Rate Index (PFRI) as well as principles for evaluation of sensitivity (percentage of true risk individuals), specificity (percentage of true nonrisk individuals), and predictive values (positive and negative) are discussed in chapter 1.

Risk Groups

Risk age groups

Recent studies have shown that carious lesions are initiated more frequently at specific ages. This applies particularly to children but also to adults. In children, the key-risk periods for initiation of caries seem to be during eruption of the permanent molars and the period during which the enamel is undergoing secondary maturation. In adults, most root caries develops in the elderly, partly because of the higher prevalence of exposed root surfaces.

Key-risk age group 1: Ages 1 to 2 years

Studies by Kohler et al (1978, 1982) showed that mothers with high salivary MS levels frequently transmit MS to their babies as soon as the first primary teeth erupt, leading to greater development of caries. Other studies have shown that 1-year-old babies with plaque and gingivitis develop several dental carious lesions during the following years, while babies with clean teeth and healthy gingivae, maintained by regular daily cleaning by their parents, remain caries free (Wendt et al, 1994). It was also shown that the practice of giving infants sugar-containing drinks in nursing bottles at night increases the development of caries (Wendt and Birkhed, 1995).

In another investigation, Grindefjord et al (1995) studied the relative risk (odds ratio) that 1-year-old infants would develop caries by the age of 3.5 years: Those with poor oral hygiene, bad dietary habits, salivary MS, little or no exposure to fluoride, and parents with a low educational level or an immigrant background were at 32 times greater risk than were children without the corresponding etiologic and external risk factors. The importance of establishing good habits as early as possible, and of postponing or preventing bad habits, should not be underestimated.

In addition, the enamel of erupting and newly erupted primary and permanent teeth is at its most caries-susceptible stage until completion of secondary maturation (Kotsanos and Darling, 1991). In 1- to 3-year-old infants, the specific immune system, particularly immunoglobulins in saliva, is immature. Poor oral hygiene will therefore favor the establishment of cariogenic microflora such as MS.

On average, the permanent teeth in particular erupt 6 to 12 months earlier in girls than they do in boys (Teivens et al, 1996). On this basis, the first-priority risk age groups are expectant mothers and 1 to 2 year olds, starting with girls (Fig 125). To prevent postnatal transmission of cariogenic bacteria and poor dietary habits from mother to child, expectant mothers who are at risk should be offered a special preventive program comprising intensified plaque control (mechanical and chemical) and reduction of sugar intake, to reduce the number of cariogenic microflora.

Key-risk age group 2: Ages 5 to 7 years (eruption of first molars)

The pattern and amount of de novo plaque reaccumulation on the occlusal surfaces of the permanent first molars, 48 hours after professional mechanical toothcleaning, was studied in relation to eruption stage by Carvalho et al (1989). Plaque reaccumulation is heavy on the occlusal surfaces of erupting maxillary and mandibular molars, particularly in the distal and central fossae and related fissures. This is in sharp contrast to the fully erupted molars, which are subjected to normal chewing friction. Abrasion from normal mastication significantly limits plaque formation, and this explains why almost all occlusal caries in molars begins in the distal and central

fossae during the extremely long eruption period of 14 to 18 months. In contrast, fissure caries is very rare in premolars, which have a brief eruption period of only 1 to 2 months.

In addition, the enamel of erupting and newly erupted teeth is considerably more susceptible to caries until secondary maturation is completed, more than 2 years after eruption. However, the caries-reducing effect of fluoride is also about 50% greater in erupting and newly erupted teeth than it is in teeth that have undergone secondary maturation.

The next high-risk age is, therefore, from 5 to 7 years, during eruption of the first molars (the key-risk teeth), starting with girls (see fig 125). Intensified mechanical plaque control twice a day with fluoride toothpaste should be performed by the children's parents, particularly on the erupting first molars. Home care should be supplemented at needs-related intervals by professional mechanical toothcleaning and fluoride varnish. In the most caries-susceptible children, glass-ionomer cement should be used in the fissures, as a slow-release fluoride agent.

Key-risk age group 3: Ages 11 to 14 years (eruption of second molars)

Normally, the second molars start to erupt at the age of 11 to 11 1/2 years in girls and at around the age of 12 years in boys. The total eruption time is 16 to 18 months. During this period, the approximal surfaces of the newly erupted posterior teeth are undergoing secondary maturation of the enamel and are also at their most caries susceptible. Therefore, 11 to 14 year olds have not only, by far, the highest number of intact tooth surfaces, but also the greatest number of surfaces at risk.

Integrated plaque control measures and use of fluoride agents should therefore be intensified on the approximal surfaces of all the posterior teeth and the buccal surfaces of the second molars, starting with 11 to 11 1/2-year-old girls (see Fig 125), to protect intact tooth surfaces and to remineralize incipient lesions. If this program is maintained throughout the secondary maturation period, and needs-related self-care habits are established, there is a high probability that the remaining intact tooth surfaces will remain intact for the individual's entire life.

Key-risk age groups in young adults and adults

Under certain circumstances, young adults (19 to 22 year olds) may also be regarded as a risk age group. Most have erupting or newly erupted third molars without full chewing function but with highly caries-susceptible fissures and mesial surfaces until completion of secondary maturation of the enamel. In addition, many young adults leave home to study or work elsewhere, with ensuing changes not only in lifestyle but also in dietary and oral hygiene habits. They may also be exposed to peer pressure toward good or bad habits.

Another risk age group among adults is the dentate elderly, most of whom have multiple restorations with plaque-retentive margins as well as root surfaces exposed by periodontitis. Regular use of medication with depressive effects on the saliva and poor oral hygiene and dietary habits further increase the risk for development of secondary caries and root caries.

Fig 125 Timing of cost-effective caries-preventive programs, related to age and gender. (From Axelsson, 1998.)

Other risk groups

Chapters 2 and 3 discuss, in detail, some other caries risk groups:

1. Persons who work in occupations where frequent food sampling is required

2. Persons who are obese because of frequent eating

3. Persons who abuse recreational drugs

4. Persons who have systemic diseases and are taking regular medication

5. Women who are pregnant

6. Persons who have psychiatric disorders

7. Persons who have impaired salivary function or immune response

8. Persons who are poorly educated, particularly those of immigrant background

9. Persons who have poor dental care habits

10. Persons who live in rural areas

Figure 126 is from a randomized sample, taken in 1993, of more than 600 residents, aged 50 to 55 years, of the county of Varmland, Sweden. The mean number of remaining teeth (third molars excluded) was related to living area (urban or rural), gender, dental care habits, educational level, smoking habits, and snuffing habits. From these data it was estimated that, on average, a well-educated, nonsmoking, urban woman with regular dental care in private practice would have about 50% more remaining teeth than would a poorly educated, smoking, rural man with irregular dental care. Among the 50 to 55 year olds it is estimated that about 80% of extractions are attributable, directly or indirectly, to caries and only 10% to orthodontic treatment and 10% to periodontal disease.

Fig 126 Mean number of remaining teeth (excluding third molars) in 50 to 55 year olds, related to living area, gender, dental care habits (regular care in Public Dental Health Service [PDHS] or in a private practice [PP] or no regular dental care [None]), educational level (Low or High), smoking habits (smoker [Smo+] or nonsmoker [Smo-]), and tobacco-snuffing habits (snuffing [Snu+] or no snuffing [Snu-]). (From Axelsson and Paulander, 1994.)

Individual Risk

By combining etiologic factors, caries prevalence (experience), caries incidence

(increment), external and internal modifying risk indicators, risk factors, and prognostic risk factors, as well as preventive factors, caries risk may be evaluated at the individual level, as no risk (C0), low risk (C1), risk (C2), and high risk (C3). As discussed earlier in this chapter, these conditions may vary in different age groups. Therefore, the criteria for C0, C1, C2, and C3 should be defined for at least the following general groups: preschool children (primary teeth), schoolchildren (permanent teeth), adults, and the elderly.

Caries prevalence and caries incidence as well as socioeconomic conditions may vary considerably among different populations and countries. The criteria listed for caries prevalence and incidence in children (C0 to C3) apply to Scandinavian conditions (low caries prevalence). Boxes 15, 16, 17, and 18 exemplify criteria for C0 to C3 in preschool children, children, adults, and the elderly, respectively, starting with etiologic factors, followed by caries prevalence, caries incidence, external modifying risk factors, internal modifying risk factors, and preventive factors. The more factors that can be identified in the individual subject, the greater the validity of the predicted risk evaluation.

In adults, as in children, caries prevalence and incidence as well as treatment needs may also vary significantly among different populations and countries. For example, in many developing countries, caries prevalence (decayed or filled surfaces) and incidence are relatively low in adults, but treatment needs (new decayed surfaces) are very high. In industrialized countries, on the other hand, caries prevalence (decayed, missing, or filled teeth and surfaces) is high, while caries incidence is moderate and treatment needs (new carious surfaces) are limited, because of well-organized dental care systems. The caries prevalence and caries incidence values given for different risk categories (C0 to C3) in adults, based on figures from the county of Varmland, are representative for Swedish conditions. Extrapolation should be made with caution, and reevaluation may be necessary. Although it is difficult to identify all the criteria for risk prediction in one individual, the higher the number of criteria identified, the greater the validity of the risk prediction.

Recently, all the adult patients in the Public Dental Health Service of the county of Varmland were categorized by risk for dental caries as no risk, low risk, risk, or high risk (C0 to C3). Preliminary data show that more than 85% of the 20 to 50 year olds were categorized as C0 to C1 and less than 5% as C3.

Key-Risk Teeth and Surfaces

Introduction

The pattern of dental caries in the dentition, reflected in missing teeth and decayed, missing, and filled surfaces, is generally as unevenly distributed as caries prevalence among individuals. Therefore, needs-related preventive programs not only should be tailored to predicted individual risk, but also should focus on the key-risk teeth and surfaces in the dentition.

Key-risk teeth

The factors determining future tooth loss are related to age, dental caries, periodontal

diseases, iatrogenic root fractures, trauma, orthodontic therapy, and so on. Therefore, it may be argued that it is difficult to analyze the true reasons for tooth loss in the adult, particularly in the elderly. The reasons for tooth loss may vary not only among different age groups but also among different populations and countries, depending on differences in prevalence of dental caries and periodontal diseases as well as the availability of resources for dental care.

In Sweden, for example, almost all the teeth lost up to the age of 35 years are premolars, extracted for orthodontic indications (Fig 127) (Axelsson et al, 1990). From the same study, the pattern of remaining teeth (third molars excluded) in randomized samples of 50 and 65 year olds is shown in Figs 128 and 129, respectively. In 50 year olds, almost 100% of the mandibular incisors and about 90% of the first and second molars remain.

In 50 year olds it is estimated that about 80% of the tooth loss was attributable to caries directly, or indirectly (endodontic complications, apical periodontitis, or post-related root fractures). In 1948 to 1953, when these subjects were 10 to 15 years old, caries prevalence and incidence were very high, and school dental care was based on "drilling, filling, billing, and killing the pulp." Only about 10% of the missing teeth would have been extracted for orthodontic indications, and 10% because of periodontal diseases.

In 65 year olds about 75%, 20%, and 5% of the missing teeth are estimated to have been lost because of dental caries, periodontal diseases, and orthodontic therapy, respectively. In other words, dental caries is the major reason for lost teeth in the older adult population. In 65 year olds only 5% and 10% of the mandibular and maxillary first molars, respectively, and about 40% and 60% of the maxillary and mandibular incisors, respectively, remain.

Quite simplistically, the risk for tooth loss could be predicted by combining measurement of the buccolingual width of the tooth crown and its distance posteriorly from the lips. The molars (key-risk teeth) are the most posterior teeth. The first molars have the widest approximal surfaces and the mandibular incisors the narrowest. In a toothbrushing population, the posterior teeth therefore require supplementary approximal cleaning and topical application of fluoride.

Fig 127 Frequency distribution of remaining teeth (third molors excluded) in 35 year olds (Federation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.) Fig 128 Frequency distribution of remaining teeth (third molars excluded) in 50 year olds (Federation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.)

Fig 129 Frequency distribution of remaining teeth (third molars excluded) in 65 year olds (Federation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.)

Key-risk surfaces

As mentioned earlier, depending on the age and caries prevalence of the population,

there may be pronounced variations in the pattern of both lost teeth and decayed or filled surfaces. Figure 130 shows caries prevalence and the pattern of decayed or filled surfaces in 12-year-old children in the county of Varmland, Sweden, in 1964, 1974, 1984, and 1994. The molars are clearly the key-risk teeth. In a toothbrushing population, the key-risk surfaces are the fissures of the molars and the approximal surfaces, from the mesial aspect of the second molars to the distal aspect of the first premolars. Integration of mechanical plaque control by self-care and the use of fluoride toothpaste, supplemented at needs-related intervals by professional mechanical toothcleaning, fluoride varnish, and chlorhexidine varnish should therefore target these key-risk teeth and surfaces, according to the principles discussed earlier in this chapter.

As shown in Fig 130, the mean caries prevalence in 1964 was around 40.0 decayed or filled surfaces, generally involving all the approximal surfaces and occlusal surfaces of the molars and premolars, but also some buccal and lingual surfaces. One mandibular first molar was missing, extracted because of caries. During the following 10 years, toothbrushing and fluoride toothpaste were introduced. As a result, the number of decayed or filled surfaces decreased to about 25.0. The reduction was mainly in carious lesions on the approximal surfaces of the incisors and the buccal and lingual surfaces of the molars and premolars. The separate effects of the toothbrush versus fluoride toothpaste are difficult to estimate.

In 1975, a needs-related plaque control program (both professional and home care) combined with use of fluoride toothpaste and application of fluoride varnish was gradually introduced targeting the key-risk surfaces of schoolchildren. The number of decayed or filled surfaces decreased to 3.0: The reduction occurred on the approximal surfaces of the molars and the premolars. The remaining caries, it is suggested, represents mainly overtreatment of first molar fissures.

Our preventive program for the occlusal surfaces of the molars was initiated in 1984. In 1994, caries prevalence was less than 1.0 decayed or filled surface. It is predicted that, in 1999, the first group of 19 year olds to have followed the integrated preventive program from birth will have less than 1.0 approximal decayed or filled surface, of which the filled component should account for less than 0.3, because approximal carious lesions without cavitation into the dentin can be treated noninvasively, without restoration.

Figure 131 shows the mean pattern of manifest caries or restorations with or without initial caries (enamel caries) included on the posterior approximal surfaces of a randomized sample of 19 year olds from four counties in Sweden (Forsling et al, 1999). The distal surface of the mandibular right first molar is clearly the most frequently decayed. This is probably because most people are right-handed, and it is well known that in right-handed people the mandibular right linguoapproximal surfaces show the greatest tendency to plaque accumulation and gingivitis.

That the distal surfaces of the second premolars constitute a relatively high percentage of carious surfaces may be explained as follows: The wide mesial surfaces of the first molars are frequently carious and exposed to cariogenic microflora when the second premolars erupt. In caries-susceptible (C2 or C3) individuals, it is difficult to achieve successful arrest of such enamel lesions during the short period of eruption (1 to 2

months) of the second premolars, and lesions are sometimes unrestored. Until completion of secondary maturation of the enamel, the environment is extremely unfavorable for the newly erupted distal surfaces of the second premolars.

Figures 132, 133, 134, 135, and 136 show the pattern of intact, decayed, filled, and missing surfaces occlusally, mesially, distally, buccally, and lingually in a randomized sample of 50 year olds in 1988 in the county of Varmland, Sweden (Axelsson et al, 1988, 1990). While almost no intact occlusal surfaces exist (Fig 132), close to 100% of the lingual surfaces of the mandibular incisors are intact (Fig 136). The lingual surfaces constitute the highest percentage of intact surfaces, closely followed by the buccal surfaces (Fig 135). Of the approximal surfaces, the mesial and distal surfaces of the first molars and maxillary premolars have the lowest percentage of intact surfaces (5% to 10%), followed by the mesial surfaces of the second molars and the distal surfaces of the second mandibular premolars (Figs 133 and 134). The gracile mandibular incisors have by far the highest percentage of intact approximal surfaces (about 70%).

Fig 130 Changes in caries prevalence among 12 year olds living in the county of Vormland, Sweden, 1969 to 1994. (DFS) Decayed or filled surface. (From Axelsson, 1998.)

Fig 131 Mean pattern of manifest caries or restorations with or without initial caries (enamel caries) included on the posterior approximal surfaces of 19 year olds. (D) Dentin; (D1, D2) enamel caries lesion; (D3) dentin caries; (FS) filled surface; (p) posterior; (m) mesial; (d) distal. (From Forsling et al, 1999. Reprinted with permission.) Fig 132 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) occlusally (Federation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.) Fig 133 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) mesially (Feder- ation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.) Fig 134 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing sur- faces (MSs) distally (Federation Dentaire Internationale tooth-numbering system). Fig 135 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) buccally (Federation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.) Fig 136 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) lingually ((Federation Dentaire Internationale tooth-numbering system).

Risk Profiles

Introduction

By combining the symptoms of disease (prevalence, incidence, treatment needs, etc); etiologic factors; external modifying risk indicators, risk factors, and prognostic risk factors; internal modifying risk indicators, risk factors, and prognostic risk factors; and preventive factors, it is possible to present risk profiles for tooth loss, dental caries, and periodontal diseases in graphic form. This can be done manually or by computer. The degree of risk, 0, 1, 2, or 3, is visualized using green, blue, yellow, and red, respectively. The graphs are also appropriate tools for communication with the patient when discussing the details of case findings and treatment recommendations.

Combined risk profiles for dental caries and periodontal diseases

Because some patients may suffer from both dental caries and periodontal diseases, risk profiles for these diseases can be designed in combination or separately. Figure 137 illustrates a combined risk profile for a patient, who after the first detailed examination and history taking, was classified as a high-risk patient for both dental caries and periodontal diseases (C3P3), on the following basis:

1. The prevalence of caries and the prevalence of periodontitis were high.

2. The incidence of caries and periodontitis had been very high.

3. The patient had been exposed to many etiologic factors, both nonspecific (high plaque formation rate and plaque volume) and specific (caries-related pathogens and periopathogens).

4. The patient exhibited many external and internal modifying risk indicators, risk factors, and prognostic risk factors for dental caries as well as periodontal diseases.

a. For dental caries, the most important external factors were high frequency of intake of sticky, sugar-containing products and medication with salivary depressive side effects. For periodontal diseases, the most important external factor was regular smoking of 10 to 20 cigarettes per day.

b. Among internal factors, the most important for dental caries was reduced stimulated salivary secretion rate (0.6 mL/min). For periodontal diseases, it was diabetes mellitus.

5. The standard of oral hygiene was very low, and dietary habits were poor.

6. The patient had no preventive dental care habits and his dental care visits were irregular.

After presentation of the case findings and a session of self-diagnosis, the dentist and patient discussed a treatment strategy based on sharing of responsibilities between the patient (the owner of the oral cavity) and the oral health personnel. Two years later, he was classified as a low-risk patient for both dental caries and periodontal diseases

(C1P1), on the following basis:

1. The etiologic factors had been dramatically reduced (from red to green), by an initial intensive combination of mechanical and chemical plaque control (self-care and professional) and by maintenance of a high standard of plaque control, ie, a dramatic improvement in the most important preventive factors.

2. Treatment needs (excavation and restoration of open carious lesions, and scaling, root planing, and debridement of diseased periodontal pockets) and plaque-retentive factors were eliminated.

3. Important external modifying factors were reduced. The patient stopped smoking and reduced the estimated daily sugar clearance time by 80%. In addition, there was no further need of medicine with salivary depressive effects. As a consequence of this and regular use of fluoride chewing gum, the salivary secretion rate increased from 0.6 mL/min to 1.0 mL/min.

4. The use of fluorides was increased. A new fluoride toothpaste technique was introduced, and use of fluoride chewing gum was recommended after meals; this was supplemented by professional application of fluoride varnish.

As a consequence of these preventive measures and the healthier lifestyle, the patient developed no new carious lesions and experienced no further loss of periodontal support.

Fig 137 Risk profile for dental caries and peridontal diseases. (solid line) Initial presentation, C3P3; (dotted line) 2-year recall, C1P1.

Detailed risk profiles for dental caries

If a patient is at high risk predominantly for either caries or periodontal disease, a more detailed risk profile is available for the specific disease. Box 19 shows a list of abbreviations for the most important variables related to caries risk.

Figure 138 illustrates how a high-risk patient (C3) has been transformed to a low-risk patient (C1) by improved self-care supplemented by professional preventive measures. The greater the difference between the solid line and the dotted line, the greater the improvement. The absence of any change suggests that this particular factor cannot be influenced (eg, genetic factors, age, and some chronic diseases).

The patient in question was a 40-year-old woman with the following clinical diagnosis and anamnestic data at the first visit:

1. Caries prevalence was very high. All occlusal surfaces, most approximal surfaces, and some buccal surfaces were restored. There were several recurrent lesions.

2. Caries incidence was very high; the patient was developing more than three new carious surfaces (more than 85% recurrent caries) per year.

3. Values for etiologic factors were extremely high:

a. The plaque formation rate was very high (PFRI score 5).

b. The amount of plaque was excessive (PI = 93%).

c. The level of salivary mutans streptococci was very high ( 1 million CFU/mL).

d. The salivary lactobacillus level was very high ( 500,000 CFU/mL).

4. The external modifying risk indicators, risk factors, and prognostic risk factors were:

a. Ongoing infectious disease, requiring medication with salivary depressive effects.

b. Mild rheumatoid arthritis, which occasionally required medication with salivary depressive side effects.

c. To date, very irregular dental attendance habits.

d. A very high frequency of intake of sticky, sugar-containing products, which resulted in extremely prolonged sugar clearance time.

e. Poor dietary habits, with negligible intake of fiber-rich fresh vegetables and fruits, accounting for the low Dietary Hygiene Index.

5. The most important observations with respect to internal modifying risk indicators, risk factors, and prognostic risk factors were:

a. A chronic reduction of the immune response.

b. A reduced stimulated salivary secretion rate (0.5 mL/min).

c. A low salivary buffering effect (SBE).

6. The preventive factors were:

a. The absence of known genetic defects on tooth shape, saliva, etc.

b. A relatively high educational level.

c. An absence of preventive dental care habits.

d. A low level of cooperation.

e. A very low standard of oral hygiene.

f. A lack of fluoride toothpaste or other fluoride agents.

g. Very poor dietary control.

h. An absence of added salivary stimulation from chewing fiber-rich food.

During case presentation, the risk profile was used as a tool for communication with the patient. Concurrently, the patient was instructed in self-diagnosis, to confirm the diagnosis of her own oral health status and treatment needs. Thereafter, an agreement was reached with respect to a treatment strategy, in which responsibility for the patient's oral health was shared between the patient and the oral health personnel at the clinic.

This was followed by an initial intensive preventive period, including education in self-care based on self-diagnosis, elimination of plaque-retentive factors, semipermanent restoration of recurrent caries using resin-modified glass-ionomer material, so-called complete-mouth disinfection, comprising professional mechanical toothcleaning, tongue cleaning, and chlorhexidine therapy (varnish, gels, toothpaste, or mouthrinse), and fluoride varnish applications. The first reevaluation was carried out after 2 months. Thereafter, the patient began a maintenance program tailored to her individual requirements.

The first detailed reexamination was carried out after 1 year. Most important at this reexamination was that the patient was activated in self-evaluation. Again, the risk profile was used as a tool for communication with the patient, to supplement self-evaluation in the mouth and on radiographs. Figure 138 shows how successfully the patient and the dental personnel had carried out their responsibilities.

The etiologic factors were dramatically reduced by improved mechanical plaque control and intermittent use of chlorhexidine by self-care, supplemented by needs-related intervals of professional mechanical toothcleaning and chlorhexidine varnish:

1. The PFRI was reduced from score 5 to score 2.

2. The Plaque Index was reduced from 93% to 8%.

3. The MS count was reduced from 1 million to 10,000 CFU/mL.

4. The lactobacilli count was reduced from 500,000 to 10,000 CFU/mL.

Marked reductions in sugar clearance time and Dietary Hygiene Index were achieved by:

1. Elimination of sticky, sugar-containing products from the diet.

2. Reduction of the total number of meals and snacks from nine to four per day.

3. An increase in the intake of fiber-rich vegetables and fruits, to stimulate salivation by chewing.

4. An increased in the intake of vegetarian proteins and fat and a reduction in the intake of animal fat and proteins.

The salivary secretion rate was increased from 0.6 mL/min to 1.0 mL/min and the buffering effect of saliva was improved from low to normal by:

1. Use of fluoride chewing gum for 20 minutes after every meal.

2. An improvement in dietary habits, particularly an increased intake of fiber-rich products that require chewing, eg, fresh vegetables and fruits: The chewing stimulates salivation.

3. Use of cheese and fresh fruits as dessert.

4. Elimination of medicines with salivary depressive effects.

Fluoride supplementation, a modifying caries-preventive factor intended to retard demineralization, enhance remineralization, and modify falls in plaque pH, was achieved by:

1. Regular use of fluoride toothpaste.

2. Use of fluoride chewing gum for 20 minutes after every meal.

3. Application of fluoride varnish after professional mechanical toothcleaning, at needs-related intervals.

4. Placement of glass-ionomer restorations, which function as slow-release agents for fluoride and can be recharged with fluoride.

As an effect of the above improvements by self-care and dental visits at needs-related intervals, for professional preventive measures and self-evaluation, the caries incidence (CI) was 0 after 1 year; no new lesions had developed. If there are no new lesions after a further 2 years of excellent self-care habits in combination with the needs-related maintenance program, the patient will be classified as low risk (C1).

Fig 138 Risk profile for dental caries. (solid line) Initial presentation, C3; (dotted line) 1-year recall, C1 (for explanation of abbreviations, see Box 19).

Cariogram Model

A new model, the Cariogram, was presented in 1996 by Bratthall for illustration of the interactions of caries-related factors. The model makes it possible to single out individual risk or resistance factors. A special interactive version for the estimation of caries risk has been developed.

The original Cariogram was a circle divided into three sectors, each representing factors strongly influencing carious activity: diet, bacteria, and susceptibility. The development of the model was based on a need to explain why, in certain individuals, carious activity could be low in spite of, for example, high sucrose intake, poor oral hygiene, high mutans streptococci load, or nonuse of fluorides.

Examples of Cariograms are presented in Fig 139. They can represent a situation at a single tooth surface, in a particular individual, or for a whole population. A closed circle, as in Fig 139 (a), describes a situation where demineralization occurs, meaning that caries will develop over a given time. The point is that there are enough bacteria, there is a caries-inducing diet, and there is a susceptible host.

An open circle, as in Fig 139 (b to h), illustrates a situation where no carious lesions will occur, the reason being that something necessary to the development of demineralization is missing. For each component, a large sector thus indicates an unfavorable situation, while a small sector means favorable conditions. Each sector can be very large or small, but none of them can disappear totally. Bratthall (1996) explained the various Cariogram examples in the following way:

In Fig 139 (a), there are three components responsible for closing the circle, bacteria, diet, and susceptibility to disease. The term bacteria is here understood to include type and amount of bacteria, bacterial adhesion, plaque formation rate, acid producing capacity, and all other factors which make the dental plaque more or less cariogenic. Similarly, diet describes all factors making diet more or less suitable for bacterial growth and acid formation. That means that contents of fermentable carbohydrates and frequency of food consumption are included, as well as possible antibacterial components in the food. In susceptibility, all factors are included that reflect the resistance to disease, such as mineralization of teeth, fluorides, saliva secretion and buffer capacity, salivary antibodies, and all other salivary or "host" components affecting demineralization and remineralization.

The bacteria sector becomes larger if there is abundant plaque, if there is a high portion of mutans streptococci and lactobacilli, if the plaque is very sticky and fast-growing, etc. The diet sector becomes larger if there is a high and frequent intake of fermentable carbohydrates, in particular sucrose, if sugar substitutes are seldom being consumed, if diet is deficient from other crucial aspects. The susceptibility sector becomes larger if there is a lack of fluoride, if fluoride toothpastes are seldom being used, if saliva secretion is low, if saliva buffer capacity is low, if there are other important saliva factors missing, etc.

The three sectors represent factors with an immediate action on a tooth surface, at a site where a cavity may or may not occur. However, behind each sector there are several factors that determine why the sector in a particular case is large or small. For example, a disease may explain why saliva secretion is low, a troublesome social situation can result in a minimal interest to clean the teeth and thus explain why plaque is abundant and fluoride from toothpaste absent, etc.

In Fig 139 (b), the circle is open, thus describing a situation where caries will not develop"something is missing" for cavity formation. In this particular case, the reason is that all sectors have "improved." The bacterial sector has been reduced, for example, due to reduced amount of plaque or a change to less cariogenic microorganisms. The diet sector is reduced, perhaps because of reduced sugar intake. Also, the susceptibility to disease has been reduced; in other words, the resistance has increased, perhaps because of implementation of a fluoride program.

Based on the Cariogram concept, an interactive version for caries risk estimation was

developed (Bratthall et al, 1997). There are a few fundamental differences between this program and the original version. First, the risk for future carious activity varies on a scale from 0% to 100%, but it cannot be more than 100% (Fig 140). Thus, the sectors cannot overlap each other. Second, a further sector, circumstances, was included. This sector includes factors such as caries experience and systemic diseases-factors to consider when the risk is calculated, in spite of the fact that these factors themselves do not participate directly in the development of the lesion.

The purpose of the program is educational, and it illustrates a possible risk evaluation: It does not replace the responsibility of the dentist, but it may help in making proper decisions. The program operates in such a way that individual data for a patient regarding bacteria, diet, saliva, and fluoride are entered into the program, together with information regarding circumstances. The values entered are based on specific criteria. The score 0 is the most favorable value, and the maximum score, 3, indicates a high, unfavorable risk value.

According to a formula, the program calculates the caries risk and indicates the "chance to avoid new cavities." A large chance to avoid caries thus means a low caries risk. The formula is based on weighted figures, meaning that factors believed to have a high impact on the caries risk will influence the risk to a higher extent. Figures 141, 142, 143, and 144 illustrate some examples of Cariograms for selected patients.

Thus, the Cariogram model is a simple way to illustrate how various caries-related factors can interact. It is useful in various situations when there is a need to discuss the importance of etiologic factors. In its interactive version, it is possible to demonstrate how the risk may change as a result of various actions. Also, the program will accept the influence of the "clinical feeling" of the operator.

Fig 139 The Cariogram: A graphic map of the interactions of factors that determine carious activity and caries risk. (A) A closed circle illustrates a situation in which carious lesions will develop over a given time. Sufficient bacteria, a caries-inducing diet, and a susceptible host are present. In this example, all sectors are equal in size. If one factor is extremely unfavorable, it can occupy more than one third of the circle. (B) Open circles indicate a situation in which carious lesions will not develop over a given time. In this case, all sectors have been reduced, indicating, for example, sugar discipline, plaque control, and increased resistance to disease. The result is that the risk for caries is reduced: No carious lesions will develop over time. (C) The three gaps in (B) have been placed together. The size of the gap indicates the safety sector. A large gap represents

a favorable situation. (D) The light blue sector is smaller, indicating that the susceptibility to disease has been reduced. Proper use of fluoride is one example of a way to increase resistance to caries. The result is that demineralization is slower and remineralization is more efficient: No lesions will develop over a given time. (E) The dark blue sector is smaller, indicating a more favorable situation for the diet. For example, the frequency of sugar intake may have been reduced and use of sugar substitutes may have been introduced. The result is that acid attacks are less frequent and fewer acids are formed: No lesions will develop over a given time. (F) The red sector is smaller, indicating that the number of cariogenic bacteria has been reduced. Institution of proper oral hygiene and reduction of mutans streptococci and lactobacilli levels are examples of

ways to achieve this change. The result is that fewer acids are formed and demineralization is slower. (G) A small gap indicates a high-risk situation. A slight change is enough to result in demineralization, followed by formation of a carious lesion. For example, a slight decrease in salivary secretion, an increase in sugar consumption, or a decrease in the level of oral hygiene will close the gap. (H) A large gap indicates a very low risk for caries. With all factors under proper control, the caries risk will be very low. In the situation illustrated, there would be no detrimental result from, for example, increasing sugar consumption to a certain degree. There is a clear safety sector before carious lesions would develop. (I) This represents an extremely unfavorable situation. No factor is favorable, and one factor is so prominent that it would have needed more space. A slight improvement in any

sector is not enough to stop carious activity; more radical improvements are needed. The dietary habits are poor, and the plaque is abundant and contains high proportions of mutans streptococci and lactobacilli. The host is also susceptible. The result is that carious activity is high (several new carious lesions each year). (From Bratthall, 1996. Reprinted with permission.)

Fig 140 Interactive Cariogram program for estimation of individual caries risk.

Fig 141 Cariogram for a patient with very high caries risk. The patient has a normal caries experience (score 2) for his age group and a disease (handicap) that is considered somewhat relevant (score 1) to carious activity. The dietary content of

sugars is fairly high (score 2), with a frequency of seven intakes per day, including between-meal snacks (score 2). Oral hygiene (plaque amount) is fairly good (score 1), but the level of mutans streptococci is very high (Strip Mutans score 2). Fluoride exposure consists of fluoride from toothpaste only (score 2), without extra supplements. Salivary secretion is very low (xerostomia; score 3), and the salivary buffering capacity is somewhat reduced (Dentobuff Green; score 1).

The combination of factors indicates that the risk for development of new carious lesions in the coming year is very high. The "chance to avoid caries in the future" is only 2%. The low salivary secretion rate is combination with the cariogenic diet and the high levels of mutans streptococci make it imperative that preventive measures be introduced. The low salivary secretion rate has a heavy impactthat is why the light blue sector is so large. (Courtesy D. Bratthall.)

Fig 142 Cariogram for a patient with high caries risk. The patient has a normal caries experience (score 2) for his age group and no disease (handicap) that is relevant (score 0) to carious activity. The dietary content of sugars is fairly high (score 2), with more than seven intakes a day, including between-meal snacks (score 3). Oral hygiene (plaque amount) is fairly poor (score 2), and the level of mutans streptococci is fairly high (Strip Mutans score 2). Fluoride exposure consists of fluoride from toothpaste only (score 2), without extra

supplements. Salivary secretion is normal (score 0), and the salivary buffering capacity is also normal (Dentobuff Blue; score 0). The combination of factors indicates that the risk for development of new carious lesions in the coming year is very high, although not as high as that shown in Fig 141. The cariogenic diet and the high levels of mutans streptococci, in combination with the relatively poor oral hygiene, make it imperative that preventive measures be introduced. (Courtesy D. Bratthall.)

Fig 143 Cariogram for a patient with low caries risk. The patient has a normal caries experience (score 2) for his age group and no disease (handicap) that is relevant (score 0) to carious activity. The dietary content of sugars is fairly high (score 2), but the frequency of intake is rather low, a maximum of five intakes a day, including between-meal

snacks (score 1). Oral hygiene (plaque amount) is poor (score 2), but the level of mutans streptococci is rather low (Strip Mutans score 1). Fluoride exposure consists of fluoride from toothpaste only (score 2), without extra supplements. Salivary secretion (score 0) and salivary buffering capacity (Dentobuff Blue; score 0) are both normal. The combination of factors indicates that the risk for development of new carious lesions in the coming year is rather low. The fairly low dietary frequency is important, as are the low mutans streptococci levels. Some actions to further decrease the caries risk are recommended. (Courtesy D. Bratthall.) Fig 144 Cariogram for a patient with very low caries risk. The patient has less caries experience (score 0) than is normal for his age group and no disease that is relevant (score 0) to carious activity. The dietary content of sugars is fairly low (score 0), and the frequency of intakes is also rather low, a maximum of five intakes a day, including between-meal snacks (score 1). Oral hygiene (plaque amount) is good (score 1), and the level of mutans streptococci is low (Strip Mutans score 1). Fluoride exposure consists of fluoride from toothpaste, plus extra supplements (score 1). Salivary secretion (score 0) and salivary buffering capacity (Dentobuff Blue; score 0) are both normal. The combination of factors indicates that the risk for development of new carious lesions in the coming year is very low. The diet is fine from a cariologic viewpoint, oral hygiene is good, mutans streptococci levels are low, and a supplementary fluoride program is being followed. Actions to further decrease the caries risk should not be necessary. (Courtesy D. Bratthall.)

Conclusions

Caries risk

From a cost-effectiveness aspect caries-preventive measures should be applied strictly according to predicted caries risk. In populations with very high caries prevalence and caries incidence (where almost everyone develops new lesions every year) the traditional whole population strategy would be cost effective. The number of such populations is dwindling, however, particularly in the industrialized countries where caries prevalence was high 20 to 30 years ago. Most of the world's populations have low or moderate caries incidence. In such populations, particularly those with well-established self-care habits and access to well-organized oral health services, the so-called high-risk strategy would be very cost effective; caries-preventive measures should target key-risk age groups and other risk groups, key-risk individuals, key-risk teeth, and key-risk tooth surfaces.

Preventive programs should target the following key-risk age groups in children:

1. One to two year olds, to establish good oral health habits as early as possible and prevent bad habits for as long as possible

2. Five to seven year olds, to prevent fissure caries in the erupting permanent first molars

3. Eleven to fourteen year olds, to prevent fissure caries in the erupting second molars and the approximal surfaces of the posterior teeth, until secondary maturation of the enamel surfaces is completed

Other age groups are at risk:

1. Young adults who leave home to study or work elsewhere, often changing their lifestyle and dietary habits

2. Elderly dentate people with exposed root surfaces, reduced salivary function, and other risk factors

Other risk groups include:

1. Persons in dietary-related occupations.

2. Individuals taking medication that impairs salivary function.

3. Poorly educated people, particularly those of immigrant background.

A combination of etiologic factors, caries prevalence (experience), caries incidence (increment), external and internal modifying risk indicators, risk factors, and prognostic risk factors, as well as preventive factors, may be used to assess the individual caries risk as no risk, low risk, risk, or high risk.

The pattern of dental caries in the dentition, reflected in terms of missing teeth, and decayed, missing, or filled surfaces, is generally as unevenly distributed as caries prevalence among individuals. Caries-preventive measures, therefore, not only should be tailored to predicted individual risk but also should target the key-risk teeth and surfaces in the dentition. The molars are clearly the key-risk teeth. Related to age group and the caries prevalence of the population, the key-risk surfaces could be ranked in the following order:

1. The fissures of the molars

2. The approximal surfaces of the posterior teeth, from the mesial surfaces of the second molars to the distal surfaces of the first premolars.

3. The approximal surfaces of the maxillary incisors, the buccal surfaces of the molars, and the lingual surfaces of the mandibular molars

In elderly people with reduced salivary function, exposed root surfaces should be regarded as key-risk surfaces, particularly buccally and approximally.

Risk profiles

Risk profiles for tooth loss, dental caries, and periodontal diseases can be visualized graphically using manual or computer-aided methods. The graphs should also be used as an interactive tool for communication with the patient during discussion of the oral

health status, etiology, modifying factors, prevention, possibilities, responsibilities, reevaluations, and results.

The Cariogram was developed to illustrate the interaction of caries-related factors. An interactive version for estimation of individual caries risk has been developed.

Chapter 5. Development and Diagnosis of Carious Lesions

Introduction

A carious lesion should be regarded not as a disease entity, but as tissue damage or a wound caused by the disease dental caries. The coronal lesion begins as clinically undetectable subsurface demineralization of enamel, visible only at microscopic level, and gradually progresses, first to visible demineralization of the enamel surface and to cavitation of the dentin, and finally to complete destruction of the tooth crown despite restoration, but without prevention (Fig 145).

On the tooth crown, primary carious lesions are usually supragingival and particularly common on the occlusal surfaces of the molars and the approximal surfaces of the posterior teeth. In highly caries-active individuals, lesions may also develop on the approximal surfaces of the incisors, the buccal surfaces of the posterior teeth, and the lingual surfaces of the mandibular molars.

In elderly people and other adult caries-risk patients with root surfaces exposed by periodontal disease, root caries may also develop. In most industrialized countries with well-organized dental care, primary caries accounts for almost all lesions up to the age of 20 years. In adults older than 40 years, about 90% of lesions are secondary caries.

According to the World Health Organization (WHO) system, the shape and the depth of the carious lesion can be scored on a four-point scale (D1 to D4):

D1: clinically detectable enamel lesions with intact (noncavitated) surfaces

D2: clinically detectable "cavities" limited to the enamel

D3: clinically detectable lesions in dentin (with and without cavitation of dentin)

D4: lesions into pulp

For diagnosis and assessment of treatment need, it is important to note that enamel, dentin, and root caries may be detected clinically at the noncavitated stage, as well as with cavitation. In state-of-the-art dental practice, all noncavitated lesions can and should be arrested; ie, a preventive, noninvasive approach is required.

It is also important to determine whether the lesion is active or inactive. This is of particular importance with respect to visible enamel and root surface lesions. Table 15

shows the clinical diagnosis related to the type, localization, size, depth, and shape of the carious lesion.

Fig 145 Development, in chronologic order of coronal carious lesions: intact tooth (tooth 43), primary enamel caries (tooth 42), primary dentin caries with cavitation (tooth 41), secondary caries with cavitation (tooth 31), advanced secondary caries (tooth 32), and complete destruction of the crown (tooth 33). (Courtesy D. Bratthall.)

Development of Carious Lesions

Enamel caries

Development

The physicochemical integrity of dental enamel in the oral environment is entirely dependent on the composition and chemical behavior of the surrounding fluids: saliva and plaque fluids. The main factors governing the stability of enamel apatite are pH and the free active concentrations of calcium, phosphate, and fluoride in solution.

The development of a carious lesion in enamel involves a complicated interplay among a number of factors in the oral environment and the dental hard tissues. The carious process is initiated by bacterial fermentation of carbohydrates, leading to the formation of a variety of organic acids and a fall in pH. The pH can fall below the critical value of 5.5, where the aqueous phase becomes undersaturated with respect to hydroxyapatite. This process is described in more detail in chapter 3.

While the typical carious lesion in enamel is the result of chemical dissolution of the dental hard tissues caused by bacterial degradation products, ie, acids produced by bacterial metabolism of low-molecular weight sugars in the diet, a lesion resulting from chemical dissolution by any other acid-containing agent is defined as erosion. The typical carious lesion is characterized by a subsurface demineralized lesion body, covered by a rather well-mineralized surface layer. Whereas in the erosive lesion, however, the surface has been etched away layer by layer, and there is no subsurface demineralization.

Microradiograms by Haikel et al (1983) have shown that early typical noncavitated carious lesions in enamel consist of a porous surface layer, about 20 to 40 um thick, which acts as a "micropore filter," and a so-called lesion body, where there is more pronounced loss of mineral (Fig 146). A detail of the boundary between the intact enamel and the lesion body reveals the presence of every single enamel prism, even in the lesion body. However, compared to the inner part of the prisms in the intact enamel, the prisms in the lesion body are somewhat demineralized (Fig 147).

On bitewing radiographs, radiolucencies in approximal enamel represent the sum total of this mineral loss from each single enamel prism, all of which are still present, and not a cavity in the enamel. For the clinician, this is of fundamental importance in treatment planning, because all noncavitated lesions can and should be arrested, rather than treated invasively. Enamel carious lesions on the buccal and lingual surfaces, as well as in the fissures, can easily be detected by direct visual inspection, after

mechanical removal of plaque. When the carious attack rate is high (a very active lesion associated with very low plaque pH), the surface of the enamel is rough, resembling unglazed china or chalk. Such lesions may develop on the buccocervical surfaces of the incisors during orthodontic treatment of patients with poor oral hygiene (Fig 148).

Scanning electron microscopic (SEM) studies by Haikel et al (1983) have shown that under such conditions (very low plaque pH), most of the mineral loss on the enamel surface is intraprismatic, as in erosion (Fig 149a). On the other hand, when the carious attack rate is slow because of moderate falls in plaque pH, there is limited localized loss of interprismatic minerals (Fig 149b). The outer and most caries-resistant part of the enamel forms a micropore filter through which minerals may be released, not only from the body of an enamel lesion but also from noncavitated lesions of dentin, as long as hydrogen ions (H+) penetrate the enamel surface.

In children with poor and irregular oral hygiene, primary enamel lesions develop during and soon after eruption, in so-called stagnant areas of the tooth surface, where bacteria can colonize, developing thick cariogenic plaque that is protected and undisturbed from mechanical chewing forces. Typical stagnation areas are the interproximal embrasures, along the gingival margin and the fissures of erupting teeth, and teeth without normal chewing function. Figure 150 shows the plaque accumulation in such stagnant areas after 3 days without toothbrushing.

Carvalho et al (1989) showed that most occlusal lesions in molars are initiated during eruption, in the distal and central fossae, because plaque reaccumulates much more rapidly in the fissures of erupting molars without normal chewing function, than it does on fully erupted teeth.

In addition, susceptibility to caries is strongly correlated to the posteruptive age of the enamel. The enamel is most susceptible to dental caries during and just after eruption, until secondary maturation is completed, after some years' exposure to the oral environment (Kotsanos and Darling, 1991). During this risk period, plaque control and topical application of fluoride should be intensified.

As early as 1966, Backer-Dirks evaluated the development and arrest of enamel lesions in relation to stage of eruption and the posteruptive age of the enamel. He examined about 90 boys and girls at ages 7 to 15 years, monitoring pits and fissures, approximal surfaces, and free smooth surfaces. Buccal surfaces were examined in more detail than the other surfaces: Each surface was cleaned with a toothbrush and dried with compressed air, and then classified as: sound, white-spot caries (if the surface showed a white opaque lesion, with or without partial loss of surface gloss), and caries with cavitation (if a break in the continuity of the enamel was perceptible with the explorer).

Table 16 summarizes the fate of the same buccal surface of maxillary first molars from age 8 to 15 years. Only nine of 72 opaque spots progressed to cavitation. More than half the surfaces with white-spot opaque lesions were classified as sound at age 15 years. Reversion from a white-spot lesion to a clinically sound surface occurred at all ages: from 8 to 9, 10 to 11, 12 to 13, and 14 to 15 years, on 11, 10, 12, and 4 surfaces, respectively (37). Because 48% of white-spot lesions were noted within 6

months after eruption, and 84% within 18 months after eruption, it was concluded that, on buccal surfaces, white-spot lesions develop soon after eruption (Backer-Dirks, 1966).

The disappearance of the opacities on the buccal surfaces was attributed to either remineralization or surface abrasion, or both. Figures 151 and 152 illustrate the marked changes in the gingival level of the buccal surface of maxillary first molars from the ages of 7 to 15 years. During this period, there is gradual physiologic detachment of the gingival from the surface of the tooth and continuing exposure of the clinical crown. During this period, the maxillary second molar erupts, leading to a further repositioning of the gingival attachment on the distal aspect of the first molar. Thus, the physiologic, passive exposure of the crown leads to a change in local conditions for plaque accumulation.

It may therefore be concluded from this study that conditions favoring plaque accumulation along the gingival margin of erupting maxillary first molars lead to early development of white-spot lesions. Further eruption leads to changes in the local environment that favor mechanical removal or suppression of cariogenic plaque, causing either arrest of lesion progression or complete disappearance of lesions. Similar conditions are frequently found on the buccal and the lingual surfaces of erupting and newly erupted mandibular first molars.

Another approach to studying the impact of oral mechanical forces on caries development is to withdraw toothbrushing for controlled periods. In the experimental caries study by von der Fehr et al (1970), discussed in chapter 2, the subjects were volunteer dental students. During a preexperimental period, sound gingival conditions were established. The tooth surfaces were carefully cleaned and thoroughly dried and scored according to the following Caries Index: 0 = surface appears intact; 1 = limited grayish tinge, with and without accentuated perikymata; 2 = well-accentuated perikymata, in some areas forming confluent grayish white spots; and 3 = pronounced white decalcification. The recordings were made using a binocular dissection microscope fitted with two spotlights.

All participants then refrained from oral hygiene procedures for 23 days. One group of subjects was assigned to a sucrose group, which rinsed with 10 mL of dilute sucrose solution for 2 minutes, 9 times a day, between meals. At the end of the no hygiene period, the teeth were carefully cleaned, polished, and reexamined for caries. Oral hygiene was resumed, and the participants were instructed to use 0.2% sodium fluoride mouthrinse daily. After 1 month, the teeth were cleaned, polished, and examined for caries. Following a further month of oral hygiene and fluoride rinsing, the experiment was terminated with a final caries examination of the cleaned and polished teeth.

At the end of the no hygiene period, the mean Caries Index had increased in both groups but was considerably higher for the sucrose group than for the control group. At the end of the two periods with resumption of oral hygiene and fluoride rinsing, the mean index returned to preexperimental levels. Figure 153 illustrates the typical appearance of an experimental subject after refraining from oral hygiene for 3 weeks. On the left side, the plaque has been disclosed: during the first weeks, most plaque accumulated on the so-called stagnant areas. On the right side, plaque has been

removed, revealing the development of noncavitated enamel lesions (white spots) on the buccocervical surfaces, where plaque accumulation was greatest.

The study showed that in the absence of daily mechanical removal or disturbance of bacterial accumulation on the teeth, formation of cariogenic plaque led to the development of early signs of buccocervical enamel demineralization, and that this process was accelerated by daily sucrose rinsing between meals. When daily mechanical plaque control was resumed, and supplemented by daily fluoride rinsing and professional mechanical toothcleaning (PMTC) on three occasions, there was not only arrest of caries progression, but also a reversal of the clinical signs of enamel dissolution.

In principle, identical results were obtained in the study by Loe et al (1972), using a similar experimental design. In a preliminary reappraisal of the experimental caries method proposed by von der Fehr et al (1970), Jenkins et al (1973) were unable to confirm the need for frequent sucrose rinses in inducing carieslike changes in enamel. Controls who did not rinse with sucrose showed an equal rise in Caries Index scores, suggesting that, with increased plaque accumulation, dietary carbohydrates dit not produce a maximal change.

Geddes et al (1978) therefore repeated the experiment, with minor technical modifications, and used a 14-day experimental period, which gave an adequate change in Caries Index (Edgar et al, 1978). The results of the new study confirmed the original findings of von der Fehr et al (1970); the mean Caries Index scores rose during the period without dental hygiene, and the rise was highest in the group rinsing with sucrose nine times a day. One month after resumption of oral hygiene, the mean Caries Index values reverted to preexperimental levels.

These experimental caries studies showed that withdrawal of oral hygiene caused development of caries, with intraindividual and interindividual variations in lesion progression. Frequent rinsing with sucrose during the experimental period of undisturbed plaque accumulation seemed to increase the caries progression rate, but with large individual variations. The deposition of plaque along the gingival margin is clinically visible less than 24 hours after cessation of toothbrushing (Axelsson, 1989, 1991; Lang et al, 1973). After this initial establishment, plaque rapidly accumulates in the coronal direction until, after approximately 1 week, the thickness and clinical extension of the plaque on different teeth and tooth surfaces have reached their maximum (Listgarten, 1976; Loe et al, 1965).

Whereas there are no major differences in the thickness of the gingivally located plaque, the occlusal and incisal extension of plaque may vary in different groups of teeth as well as on the various surfaces, presumably reflecting individual masticatory patterns. While friction through mastication has an effect on incisal and occlusal growth of plaque (Carvalho et al, 1989, 1991; Ekstrand et al, 1993), examination of plaque development (Loe et al, 1965) as well as experimental studies (Lindhe and Wicen, 1969; Wilcox and Everett, 1963) indicate that the gingival margins and the cervical areas of the teeth are not subjected to physical stress from food particles in the modern diet. In the aforementioned studies, developing caries was also observed along the gingival margin, demonstrating that visible signs of caries develop where bacterial plaque has been protected from oral mechanical disturbance for the longest

period of time (Thylstrup et al, 1994).

However, compared to individuals with poor and irregular oral hygiene habits, people who brush meticulously every day exhibit quite a different and limited pattern of undisturbed plaque. In a toothbrushing population, undisturbed dental plaque is most likely to persist on the posterior approximal surfaces, where toothbrush accessibility is limited, and these are the surfaces most frequently decayed. In the above experimental studies, meticulous PMTC before caries reexamination, in conjunction with resumption of oral hygiene procedures, resulted not only in arrest of further lesion progression but also in regression of the superficial enamel lesions to a stage at which they were no longer readily discernible clinically.

The most extreme model for human experimental caries ensures total elimination of mechanical forces on tooth surfaces, thereby allowing undisturbed plaque accumulation. The first such model, by Nygaard Ostby et al (1957), used a gold plate, retained on the tooth by two pinledges. Opaque spots developed on the enamel over periods of 4 to 6 weeks.

Von der Fehr (1965) used the same method to examine histologic features of enamel caries, induced over periods varying from a few weeks to several months. Corresponding to the area protected by the gold plate, there was macroscopic loss of enamel translucency, the changes ranging from slight accentuation of the perikymata to distinct white spots. Microradiographic examination revealed a radiographically dense surface zone overlying zones with low x-ray absorption ("inner spots") running parallel to the outer surface.

Hals and Simonsen (1972) modified the technique, and, in studies of caries around amalgam restorations in vivo, used a preformed orthodontic band with two metal posts, 0.3 or 0.5 mm thick, welded to the inner surface to create a space between the band and the buccal surface of the tooth. This model was applied by Holmen et al (1988) in 15 children undergoing orthodontic treatment, to investigate the effect of regular disturbance or removal of dental plaque. Two homologous premolars were banded for 5 weeks. One tooth in each pair served as a control, to which the band remained cemented for the entire test period. The other band was removed weekly, and the buccal surface was cleaned, either by careful pumicing with a nonfluoride toothpaste or simply by wiping with a cotton pellet. No fluoride of any kind was added during the entire test period.

The teeth were examined macroscopically, in polarized light, and by SEM. The enamel changes in the control teeth ranged from slightly accentuated overlapping of the perikymata to pronounced white opaque lesions. By contrast, all the experimental teeth appeared clinically sound. In polarized light, the control teeth showed classic subsurface lesions of varying severity; no subsurface dissolution could be discerned in any of the experimental teeth, regardless of cleaning procedure (Fig 154).

In SEM, the control teeth showed signs of active carious dissolution. The pumiced surfaces of the test teeth were characterized by a general smoothing out of surface irregularities and the presence of microscratches. The appearance of the surfaces cleaned with cotton pellets was very similar but with less microwear. This study convincingly demonstrated the importance of intraoral mechanical forces for caries

initiation and progression: complete elimination of mechanical forces (undisturbed plaque) caused development of caries in all individuals, even without excluding the complex interplay of other individual factors, as indicated by variations noted in the rate at which the lesion advanced (Fig 155). The determining factor is mechanical suppression of bacterial activity, even in the absence of fluoride. The fact that none of the experimental teeth showed any visible evidence of caries offers further support for the principle that "clean teeth do not decay."

These in vivo studies convincingly demonstrate that partial or total elimination of intraoral mechanical forces leads to evolution of cariogenic plaque, resulting in carious dissolution of enamel. In addition, they show that reexposure to mechanical forces not only arrests further progression of the lesion, but also results in partial regression of the lesion. In all studies, the localized loss of normal enamel translucency, clinically discerned as white opacities or white spots, served as an indicator of carious dissolution.

Arrest

Fluoride and plaque control

Arrest of enamel carious lesions is a reality, as shown in the studies by Backer-Dirks (1966) and von der Fehr et al (1970). In vitro as well as in vivo studies have shown that carious lesions in enamel can successfully be arrested by plaque control or topical use of fluoride. The most efficient means is a combination of both, as exemplified in Fig 156. On the left is an active, noncavitated enamel lesion on the mesiolingual surface of a mandibular second molar. Fluoride accumulates in the plaque fluid and as calcium fluoride on the enamel surface. During the acid challenge, calcium fluoride is dissolved. The enamel surface acts as a micropore filter and F- and H+- ions (HF) diffuse into the subsurface lesion, increasing the amount of fluoride in the active lesion compared to the surrounding intact enamel. Within the lesion, the F- ions retard demineralization of the enamel crystals during acid challenge and enhance remineralization by crystal growth and accumulation of fluorapatite on the crystal surfaces when the pH rises. Such a lesion can successfully be arrested if the patient maintains a high standard of approximal plaque control and uses fluoride toothpaste. Remineralization of the lesion is usually incomplete. "Continuous" access to a low concentration of fluoride results in more complete remineralization than does a high concentration of fluoride, which induces more rapid remineralization of the outer surface of the lesion (sealing of the micropore filter). As a result, the remineralized enamel surface will be less caries prone than the original intact surface. The total amount of fluoride is greater in the arrested lesion.

At the subclinical, microscopic level, repeated cycles of acid challenge, followed by a rise in pH, combined with frequent (daily) access to low concentrations of fluoride from water and toothpaste, will result in so-called secondary maturation, and the tooth enamel will gradually become more caries resistant. In vivo studies on the development of experimental carieslike lesions have shown that lesions in extracted, unerupted teeth with mature enamel could be induced to depths about 1.5, 2, and 3 times greater than in extracted teeth that had been exposed to the oral environment for 0 to 3 years, 4 to 10 years, and more than 30 years, respectively (Kotsanos and Darling, 1991).

Figure 157 shows a cross section of an in vitro experimentally developed, noncavitated, enamel carious lesion in polarized light. There is an outer surface zone (the micropore filter), and the inner so-called lesion body has more extensive mineral loss. Figure 158 illustrates successful arrest of such a lesion in vitro, through the use of low concentrations of fluoride. Only at the base of the lesion body is there some residual net loss of minerals. Clinically such an arrested enamel lesion would have a smooth, hard, translucent surface like that of intact enamel, but the inner part of the enamel would tend to be somewhat whiter.

Applying the orthodontic band technique by Hals and Simonsen (1972) described earlier, Holmen et al (1987) created areas of undisturbed plaque on the buccal surfaces of premolars that were scheduled for extraction for orthodontic indications, inducing active carious lesions in enamel over a period of 4 weeks. Figure 159 shows the typical white-spot appearance of such an active enamel lesion on removal of the orthodontic band after 4 weeks of undisturbed plaque accumulation. After only 1 week of regular plaque control and exposure to the oral environment, including fluoride from fluoride toothpaste, the inactive or arrested lesion appears less whitish, because of remineralization, wear, and polishing of the external, partly dissolved surface (Fig 160).

Figure 161 shows another active enamel lesion, less white on removal of the orthodontic band after plaque accumulation. After 2 weeks of plaque control and exposure to the oral environment, the arrested lesion is not as readily discernible clinically. The surface has a glossy appearance (Fig 162).

On the basis of these experiments, it was concluded that local elimination of oral mechanical disturbance of plaque accumulation enhanced lesion progression and that reexposure to mechanical disturbance, including oral hygiene procedures, not only arrested further progression of lesions, but also resulted in partial regression of lesions. The gradual enhancement of microwear in relation to time supports the concept that mechanical removal of the cariogenic biomass has been responsible for the observed arrest (Holmen and Thylstrup, 1986).

In orthodontic patients with poor and irregular oral hygiene and limited use of topical fluoride, caries is a frequent complication. The lesions typically associated with the direct-bonding technique develop on the cervical enamel because of plaque retention. In children who had undergone routine orthodontic treatment for about 2 years, Artun and Thylstrup (1986, 1989) monitored such lesions at the time of debonding, at 1, 2, 3, 4, 8, and 12 weeks.

The demineralized areas on the maxillary incisors, adjacent to the bonded brackets, were examined in detail: (1) clinical examination of plaque distribution, the extent and surface texture of the lesions, and clinical estimation of enamel opacity; (2) color slides at each appointment, before and after removal of microbial deposits, the latter preceded by air-drying for 20 seconds; and (3) SEM examination of replicas made at each appointment, after the tooth surfaces had been pumiced and washed with a solution of 5% vol/vol hypochlorite for 30 seconds. As a reference for SEM examinations, at the time of debonding, a furrow was made in the bonding area with a sharp hand instrument.

Heavy accumulations of dental plaque were observed in all areas gingival to the brackets at the time of debonding. After cleaning, the cervical region of the labial enamel showed the characteristic chalky white appearance of active enamel caries. The border between the lesion and the sound enamel that had been covered by the bonding material was very distinct (Fig 163).

From being chalky and soft at the time of debonding, a gradual change was noted during the ensuing 2 to 8 weeks: The surface of the lesions became glossy, and gentle probing disclosed a gradual improvement in hardness, to the level of the adjacent sound enamel. Concomitantly, the pronounced whiteness of the lesion at the time of the debonding changed to a more diffuse opacity (Fig 164).

At the end of the 3-month observation period, only some residual enamel opacity persisted. In some cases, small surface microcavities had developed during the first weeks of observation: These areas regained normal enamel translucency relatively rapidly. In the SEM, there was a marked ledge at the border between the surface of the active lesion and the adjacent sound enamel (Fig 165).

After 3 months, the difference in levels between the lesion surface and the sound surface became more marked, indicating somewhat greater wear of the "soft" lesion surface than of the sound enamel (Fig 166). The furrow in the bonding area could not be discerned, but the ledge at the border between the lesion and the bonding area was still distinct. The 3-month observations generally revealed marked wear, and the surface microcavities were either leveled out or barely discernible.

From this study, it was concluded that long periods of undisturbed plaque, associated with orthodontic banding, result in more pronounced carious dissolution than was revealed in previous short-term studies. The direct dissolution of the outer surface became much more evident, with a visible difference in surface levels between the lesion and the sound enamel. The clinical impression of a less white, arrested lesion was therefore predominantly a result of wear and polishing of the partly dissolved, "chalky" surface of the active lesion. This phenomenon also explained the clinical impression of regained surface hardness. The study again demonstrated that removal of cariogenic plaque resulted in arrest of lesion progression and that the clinical impression of lesion regression is related to remineralization and intraoral wear, including oral hygiene procedures.

Progression can be arrested by plaque control and either short-term use of high fluoride concentration or long-term use of low fluoride concentration. To date, the long-term effect of rapid arrest at the enamel surface has not been compared with a slow but more complete arrest throughout the entire enamel lesion.

Access

In toothbrushing populations of children and young adults, the approximal surfaces of the posterior teeth, particularly from the mesial surfaces of the second molars to the distal surfaces of the second premolars, are the sites commonly susceptible to caries. In the permanent dentition, the mesial surfaces of the first molars are usually the first approximal surfaces to decay, because of the relatively wide contact with the distal

surfaces of the second primary molars and limited accessibility for the toothbrush. Figure 167 shows an active enamel lesion on the mesial surface of a maxillary first molar. The surface has been exposed by exfoliation of the primary second molar. Note the location of the gingival margin at the cervical border of the opaque lesion. The contact area is located above the lesion.

Figure 168 shows an SEM of an initial surface dissolution (a micropore filter) cervical to the contact facet in such an active enamel lesion. Details of the surface dissolution patterns are shown in Fig 169. These figures clearly illustrate that approximal enamel lesions of the molars are initiated apical to, and sometimes subgingival to, the contact facet. To prevent the development of such lesions, or to arrest early lesions, cariogenic plaque apical to the contact facet must therefore be removed.

During the relatively short interval of about 2 months, from exfoliation of the primary second molars to full eruption of the second premolars, the mesial surfaces of the permanent first molars are readily accessible for plaque control as well as for topical fluoride application, offering an excellent opportunity for arrest of carious lesions in enamel (Figs 170 and 171).

If there is caries on the distal surface of the primary second molar, the grinding technique (Figs 172 and 173) is recommended, to prolong exposure of the mesial surface of the permanent first molars, allowing prevention of enamel lesions and arrest of established lesions in enamel and noncavitated lesions in dentin. The exposed dentin of the primary second molar (with caries excavated where necessary) should be semipermanently restored with glass-ionomer cement. This material acts as a slow-release fluoride reservoir, which can be recharged by application of topical fluoride agents, such as fluoride varnish.

Carvalho et al (1989) reported that, 48 hours after PMTC, plaque reaccumulation is almost five times greater on the occlusal surfaces of erupting permanent first molars than it is on fully erupted molars. This explains why almost all fissure caries in molars is initiated during the extremely long eruption time (12 to 14 months for first molars and 14 to 18 months for second molars) and why fissure caries seldom occurs in premolars, which have an eruption time of only 1 to 2 months.

Carvalho et al (1992) taught the parents of children with erupting or recently erupted permanent first molars to brush the fissures with a special technique. In a selected high-risk group, this was supplemented, at needs-related intervals, by PMTC and application of sodium fluoride solution until the molars were fully erupted and in chewing function. Figure 174 shows the number of enamel lesions at the baseline examination that were subsequently arrested in the different sites of the occlusal surfaces, and the very limited number of new noncavitated enamel lesions that developed during the 3-year longitudinal study. This study showed that even occlusal enamel lesions could be successfully arrested by plaque control and topical use of fluoride.

In another study, similar results have been achieved in first and second molars (Kuzmina, 1997). These studies support the contention that application of fissure sealants to fully erupted, caries-free teeth is costly overtreatment. However, in populations with high caries prevalence, the earliest possible use of fluoride-releasing

fissure sealents (glass-ionomers) in the erupting permanent first and second molars is still a very efficient method of fissure caries control.

Fig 146 Cross section of a carious lesion in enamel (E) without cavitation. (From Haikel et al, 1983. Reprinted with permission.)

Fig 147 Closeup of the deeper part of the lesion shown in Fig 146. Every separate enamel prism (E) remains, but there is some mineral loss from every enamel prism (P) and small interprismatic spaces (S). (From Haikel et al, 1983. Reprinted with permission.)

Fig 148 Active caries, resulting in rough enamel resembling unglazed china or chalk. (Courtesy A. Thylstrup.)

Fig 149a Intraprismatic mineral loss from resulting from a rapid caries attack rate (very low pH). Observe that most demineralization is in the center of the individual enamel prisms (P) and (i) not interprismatic. (From Haikel et al, 1983. Reprinted with permission.)

Fig 149b Limited localized loss of interprismatic mineral resulting from a slow caries attack rate with small interprismatic spaces (S). (From Haikel et al, 1983. Reprinted with permission.)

Fig 150 Plaque accumulation in "stagnant" areas of the teeth after 3 days without toothbrushing. (Courtesy A. Thylstrup.)

Fig 151 Plaque accumulation on the occlusal surfaces of a partially erupted maxillary first molar without friction from chewing function. (Courtesy A. Thylstrup.)

Fig 152 Plaque accumulation of the occlusal surfaces of the same molars, now completely erupted. There has been gradual physiologic detachment of the gingiva from the surface of the tooth, increasing the exposure of the clinical crown. These changes favor mechanical removal or suppression of cariogenic plaque. (Courtesy A. Thylstrup.) Fig 153 Typical appearance of an experimental subject after refraining from oral hygiene procedures for 3 weeks. On the left side, plaque has been disclosed. On the right side, plaque has been removed, revealing the development of white-spot lesions on the buccocervical surfaces, where plaque accumulation was greatest. (From von der Fehr et al, 1970. Reprinted with permission.) Fig 154 (A) Classic subsurface lesions in control tooth banded for 5 weeks of orthodontic treatment. (B) Unchanged experimental tooth, debanded for weekly cleaning during 5-week period of orthodontic treatment. No subsurface dissolution is visible. (C) Polarized light view of the control tooth. (D) Polarized light view of the experimental tooth. (From Holmen et al, 1988. Reprinted with permission.) Fig 155 Caries progression (clinical and miscroscopic signs) in control teeth, with undisturbed plaque, and experimental teeth, with weekly plaque removal. (From Holmen et al, 1988. Reprinted with permission.)

Fig 156 (left) Active, noncavitated enamel carious lesion on the mesiolingual surface of a mandibular second molar. (right) Lesion arrested by plaque control and administration of topical fluorides. (Modified from Weatherell et al, 1977.) Fig 157 Cross section of an in vitro experimentally developed noncavitated enamel lesion in polarized light. Note the outer surface zone (the micropore filter) and the inner lesion body, with more extensive mineral loss. (From Silverstone, 1973. Reprinted with permission.)

Fig 158 Successful arrest of an experimentally developed enamel lesion in vitro, through application of low concentrations of fluoride. Only at the base of the lesion body is there some residual net loss of minerals. (From Silverstone, 1973. Reprinted with permission.)

Fig 159 White-spot lesion developed after 4 weeks of undisturbed plaque accumulation under an orthodontic band. (From Holmen et al, 1987. Reprinted with permission.)

Fig 160 Same lesion shown in Fig 159, inactivated or arrested after only 1 week of regular plaque control and exposure to the oral environment, including fluoride toothpaste. The lesion appears less white, because of reminera- lization, wear, and polishing of the external, partly dissolved surface. (From Holmen et al, 1987.)

Fig 161 Active enamel lesion, less white than that in Fig 159, after removal of an orthodontic band, indicating a less advanced caries attack rate. (From Holmen et al, 1987. Reprinted with permission.)

Fig 162 Same lesion shown in Fig 161, after 2 weeks of plaque control and exposure to the oral environment. The arrested lesion is not as readily discernible. Note the glossy appearance of the surface. (From Holmen et al, 1987. Reprinted with permission.)

Fig 163 Distinct border between sound enamel (which had been covered by bonding material for 2 years because of orthodontic appliances) and active enamel lesions after removal of orthodontic brackets. The lesions correspond to areas of heavy plaque accumulation, gingival to the brackets. (Courtesy A. Thylstrup.) Fig 164 Same lesions shown in Fig 163, 3 month after debonding of orthodontic brackets. The active lesion has been completely arrested. The white appearance and rough surface of the lesion have diminished due to remineralization and polishing of the enamel surface. (Courtesy A. Thylstrup.)

10 additional images not shown.Dentin caries

Whether or not an active, noncavitated carious lesion in enamel will progress into the dentin and the rate of progression are determined by many factors:

1. The overall estimated caries risk (C1 to C3) of the individual

2. The rate at which the enamel lesion has developed

3. The size, depth, and site of the enamel lesion

4. The posteruptive age of the enamel

5. The future efficacy of self-care and supplementary needs-related preventive programs

On the approximal surfaces of the posterior teeth, the progression of a carious lesion through the enamel into the dentin can easily be followed on serial bitewing radiographs. However, the radiographs do not disclose whether the lesions are cavitated or noncavitated. In an early study, Backer-Dirks (1966) showed that 50% of enamel carious lesions on the mesial surfaces of the permanent first molars progressed into the dentin from ages 11 to 15 years, 67% from 9 to 15 years, and 74% from 7 to 15 years. However, more recent studies on premolars and molars have shown that only 14% of the approximal enamel carious lesions progressed into the dentin from ages 13 to 15 years (Bille and Carsten, 1989), 30% to 40% from 14 to 18 years (Lervik et al, 1990), and as low as 5.4% from 11 to 22 years in the most recent study, which was based on prevention and remineralization rather then restoration (Meja`re et al, 1999).

Development

In this context, it is important to recognize that the two mineralized tissues of the tooth crownthe enamel and the dentinare different not only in origin but also in composition. They therefore respond differently to stimuli such as chemicals (acids), attrition, temperature, and so on.

The enamel is derived from the ectodermal component of the tooth germ, while the pulpodentinal organ is developed from the mesenchymal component. The enamel is avascular and acellular and cannot respond to injuries, whereas the dentin and the dentinal cells, the odontoblasts, are integral parts of the pulpodentinal organ and are vital tissues, with specific defense reactions to external insults. Because the enamel is a microporous solid, stimuli from the oral cavity can permeate to the dentin and the pulp. Changes in dentin during caries progression cannot be understood, therefore, without taking into account the spread of the enamel lesion.

The most common defense reaction by the pulpodentinal organ is tubular sclerosis, ie, deposition of mineral within the dentinal tubules. The earliest dentinal response to the enamel lesion that can be detected by light microscopy is tubular sclerosis, at the site where the central travers (CT) crosses the dentinoenamel junction (Fig 175). Enamel demineralization increases enamel porosity and hence the permeability of the enamel, and the first mild stimuli initiating the defense reaction reach the dentin underlying the most porous part of the enamel lesion. The light microscope provides relatively low magnification, and much earlier dentinal reactions have been detected using biochemical and histochemical methods.

Initial tubular sclerosis can be detected before the advancing front of the enamel lesion reaches the dentinoenamel junction. When contact is established between the enamel lesion and the dentinoenamel junction, the first sign of dentinal demineralization can be seen along the junction (Figs 175 and 176) as yellow or brownish discoloration, depending on the aggressiveness of lesion formation. For many years, it has been thought that demineralization of dentin progresses laterally along the dentinoenamel junction, on the assumption that the anatomic discontinuity between the two tissues facilitates penetration of destructive agents.

However, recent systematic investigations have concluded that discolored dentinal demineralization never extends beyond the limits of the enamel lesion at the dentinoenamel junction (Bjorndal, 1991). If the active enamel lesion is regarded as a multitude of microlesions at different stages of progression, then the dentinal sclerosis lateral to the demineralization may be interpreted as a reaction to stimuli in the direction of the rods from the less advanced parts of the enamel lesion approaching the dentinoenamel junction (see Figs 175 and 176).

At this stage of lesion progress, therefore, the lesion in dentin should not be considered as an entity in itself, with a central and spreading focus of destruction, as conventionally assumed. The dentinal changes merely represent a continuum of pulpodentinal reactions to variations in acid challenge at the enamel surface, with transmission of the stimulus through the enamel in the directions of the rods.

The implication of this approach is that when acid production at the surface ceases, because of regular disturbance or removal of the cariogenic microbial biomass, then

demineralization also ceases, arresting further progression of the lesion.

Figure 177 shows a cross section of an active carious lesion with a cavity into the enamel and some spread of the lesion into the dentin along the dentinoenamel junction. To the right is a detail (SEM) of the base of the enamel cavity, which is covered by cariogenic plaque, predominantly cocci. Every single enamel prism can still be seen, albeit the outer structure of some prisms is less mineralized.

A cross section of another carious lesion with microcavitation in the enamel and a noncavitated lesion in the underlying dentin is shown in Fig 178. However, after arrest of such a lesion, there is only very limited mineral uptake from saliva by the enamel and the dentin, and both the demineralized enamel and the demineralized dentin therefore persist, as scars in the tissue. Figure 179 shows a microradiograph of a ground section through an inactive approximal carious lesion, with cavitation in the outer part of the enamel and involvement of dentin; the caries has been arrested for several years (Thylstrup and Fejerskov, 1994). In the enamel, some mineral redeposition can be seen corresponding to the base of the cavity, whereas the peripheral demineralized dentin remains unchanged after arrest of the lesion.

Intervention

Conventionally, involvement of dentin has been regarded as the stage in caries progression at which operative intervention becomes necessary to arrest further destruction, and there have been many studies attempting to improve the radiographic detection of this stage. The term dentinal involvement is, however, too vague to define the continuum of changes occurring in the pulpodentinal organ during caries progression and therefore serves no useful purpose as an indicator for operative intervention.

To understand the gradual exposure of the pulpodentinal organ during progressive lesion formation, it is important to be aware that, although there has been some mineral loss from the enamel and the lesion is thus characterized as porous, adequate mineral remains to preserve the structural composition of the enamel (see Figs 146 and 147). The area below the surface zone is not an empty space, but highly mineralized tissue, despite some degree of mineral loss. The first signs of surface breakdown are therefore limited to the outermost enamel and are presumably caused by mechanical trauma during mastication, microtrauma during interdental wear, or iatrogenic damage from careless probing of approximal surfaces (see Fig 177).

If such areas are not kept relatively free of dental plaque, the process will continue, because, all other matters being equal, the bacteria harbored in the microcavity will be more sheltered than those on the surface, and this will favor an ecological shift toward anaerobic and acidogenic bacteria (see chapter 1). The progressive destruction of the enamel or the gradual enlargement of the cavity is therefore the combined result of continued acid production in the protected microbial biomass and mechanical microtrauma.

As long as the base of the cavity is still confined to the enamel, massive bacterial invasion of the dentin is unlikely. At this stage, the lesion can successfully be arrested (see Fig 179). However, a persistently undisturbed active carious lesion will

eventually progress to cavitation of the dentin: Exposure of the dentin to the masses of bacteria in the cavity will soon lead to decomposition of the most superficial part of the dentin, as a result of the action of acids and proteolytic enzymes. This zone is referred to as the zone of destruction (see Fig 175). Beneath this zone, tubular invasion of bacteria is frequently seen (Fig 180).

If lesion progression is very rapid, it is not uncommon to see so-called dead tracts in the dentin, indicating destruction of the odontoblastic processes without tubular sclerosis. Such empty dentinal tubules are readily invaded by bacteria (Fig 181). Between the zone of bacterial penetration and the sclerotic dentin, the translucent zone, there is a zone of demineralization, caused by acids produced in the biomass of anaerobic and aciduric bacteria in the cavity.

There is still some uncertainty concerning the degree of pulpal response to various stages of caries development. Reactionary (tertiary) dentin may form even before bacterial invasion of the dentin: Reactionary dentin is less mineralized and contains irregular dentinal tubules. When demineralization of the dentin approaches within 0.5 to 1.0 mm of the pulp, an inflammatory response may be seen in the subodontoblastic region. This does not constitute true infection of the pulp, however; the inflammatory cell reactions are believed to occur in response to bacterial toxins.

From a treatment perspective, it should be noted that there is no true indication for invasive intervention until the cavity has reached the dentin: until then, the lesion can be arrested, and even softened dentin is not infected by bacteria. These principles apply particularly to buccal and lingual surfaces but should also be valid for most approximal and occlusal surfaces.

Actively progressing dentin caries is soft and yellowish in color. The purpose of excavation is to remove infected and necrotic tissue but to avoid cutting into the underlying sound tissue, destroying thousands of odontoblastic processes. The most appropriate method of excavation is the use of hand instruments or slowly rotating burs, allowing the operator to recognize the interface between the relatively hard translucent zone and the demineralized dentin. After excavation, although there may be no marked discoloration and the dentin is hard on probing, some microorganisms may still remain in open dentinal tubules (see Fig 180). Microbiologic investigations indicate that about 25% of excavated teeth still harbor bacteria, and histologic techniques have demonstrated microorganisms (Reeves and Stanley, 1966) in one or more tubules in 30% to 50% of teeth. On average, there are 45,000 dentinal tubules per 1 mm2 and some may be dead tracts with remaining microorganisms (see Fig 181). However, under a well-sealed, retentive restoration, they are denied access to substrate and will do no harm.

Fig 165 SEM micrograph of replica of the active lesion (Fig 163). Note the distinct step between the eroded surface of the active lesion and the adjacent sound enamel, where a furrow has been made. (Courtesy A. Thylstrup.) Fig 166 SEM micrograph of replica of the arrested lesion (Fig 164). After 3 months, the furrow has almost disappeared, and the step between the sound and the arrested surface is slightly enhanced. (Courtesy A. Thylstrup.)

Fig 167 Active enamel lesion on the mesial surface of a maxillary first molar. The surface has been exposed by exfoliation of the primary second molar. Note the location of the gingival margin at the cervical border of the opaque lesion. The contact area is located above the lesion (arrows). (Courtesy A. Thylstrup.)

Fig 168 Initial surface dissolution cervical to the contact facet (CF) in an active enamel lesion. (From Thylstrup and Fejerskov, 1981. Reprinted with permission.) Fig 169 Increased magnification of the surface dissolution patterns shown in Fig 168. (From Thylstrup and Fejerskov, 1981. Reprinted with permission.)

Fig 170 Enamel lesion (arrow) on the mesial surface of the first molar, adjacent to the primary second molar, which has a lesion in dentin.

Fig 171 Remineralization of the enamel (arrow) on the first molar shown in Fig 170. Six years later, there are no radiographic signs of the lesion. Remineralization occurred between exfoliation of the primary second molar and eruption of the second premolar (in about 2 month). Fig 172 Grinding technique to prolong exposure of the mesial surface of the permanent first molars. The exposed dentin of the primary second molar should be restored with glass-ionomer cement. (From Ek and Forsberg, 1994. Reprinted with permission.) Fig 173 Radiograph demonstrating use of the grinding teeth to expose the mesial surface of the permanent first molars. (Courtesy H. Forsberg.)

Fig 174 Development of occlusal enamel caries in children in a noninvasive preventive program. (green [R]) Enamel lesions at baseline that were subsequently arrested; (red [P]) limited number of new noncavitated enamel lesions. (Modified from Carvalho et al, 1992.)

Fig 175 Schematic illustration of progressive stages of lesion formation. (1) Reactive dentin (2) Sclerotic reactive or translucent zone(3) Zone of demineralization (4) Zone of bacterial invasion (5)

Peripheral rod direction;(CT) Central traverse(Modified from Bjorndal, 1991.)

Fig 176 Cross section of a noncavitated approximal dentin caries lesion in a molar. The dark color along the dentinoenamel junction indicates an inactive, arrested lesion. (Courtesy I. Espelid.)

Fig 177 (left) Cross section of an active carious lesion with a cavity into the enamel and some spread of the lesion into the dentin along the dentinoenamel junction. (right) Scanning electron micrograph of the base of the enamel cavity, which is covered by cariogenic plaque, particularly cocci. (Courtesy A. Thylstrup.)

Fig 178 Ground section of active approximal lesion, which reaches the dentinoenamel junction with demineralization of the outer dentin (ZD) and sclerotic reactions (TZ).(Courtesy A. Thylstrup.)

Fig 179 Microradiograph of a ground section through an inactive approximal carious lesion with cavitation in the outer part of the enamel and dentinal involvement. The lesion has been arrested for several years. (Courtesy A. Thylstrup.)

Fig 180 Dentin lesion with cavitation, superficial zone of destruction, and zone of bacterial invasion into the dentin tubules (ZP). Bacteria have already reached the pulp. (From Tylstrup and Fejerskov, 1994. Reprinted with permission.) Fig 181 Clusters of bacteria penetrating dentin tubules. (From Tylstrup and Fejerskov, 1994. Reprinted with permission.)

Root caries

According to Hix and O'Leary (1976), root surface caries is defined as "a cavitation or softened area in the root surface which might or might not involve adjacent enamel or existing restorations (primary and recurrent lesions)." Nyvad and Fejerskov (1987) introduced the definitions of active and inactive carious lesions of the root. Root caries may be classified as primary or secondary, cementum or dentin, active or inactive, and with or without cavitation (see Table 15). The lesions can also be classified according to the texture (soft, leathery, or hard) and the color (yellow, light brown, dark brown, or black).

Development

The initial conditions necessary for development of carious lesions at the root are:

1. A root surface accessible to a cariogenic microflora

2. Exposure of the root surface to cariogenic plaque (biomass) until the lesions develop

Root surfaces inevitably are exposed by gingival recession resulting from poor oral hygiene and gradual loss of periodontal attachment with age. Even in populations with adequate oral hygiene, some recession occurs, with a highly characteristic pattern of distribution in elderly populations. Another typical pattern is frequently seen on the buccal surfaces of several teeth in adolescents, attributed to inappropriate plaque control procedures, ie, iatrogenic effects of horizontal toothbrushing.

When root surfaces are exposed to the oral environment as a result of gingival recession, there is an increase in areas of potential plaque retention, particularly in the large interproximal areas and along the gingival margin and the cementoenamel junctions, which represent stagnant areas for accumulation of plaque (Figs 182 and 183). In addition, compared to the enamel surface, the intact root surface is very rough and retentive to plaque. Therefore, the primary carious lesion of the root has a greater horizontal than vertical dimension, because of greater plaque buildup along the gingival margin.

Initial active root lesions are soft on probing, have a leathery consistency, and are

normally covered with plaque. The color is yellow or light brown. However, Lynch and Beighton (1994) reported that, irrespective of the color, soft active lesions are closest to the gingival margin and hard inactive lesions are the most distant; leathery lesions occupy an intermediate position.

Experimental studies have shown that carious lesions develop very rapidly on root surfaces continuously covered with plaque (Nyvad et al, 1989). Figure 184 shows a microradiogram, illustrating the depth and shape of the root carious lesion after 1, 2, and 3 months. There is a gradual increase in mineral content in the surface zone, despite progressive loss of subsurface mineral.

The surface of the radicular carious lesion is infiltrated by microorganisms at a very early stage. The bacteria seem to split the collagen fibers of the cementodentinal junction, and microorganisms may be found in the many exposed dentinal tubules even in initial root lesions. The response by the dentin is similar to that described for coronal caries; ie, the pulpodentinal organ corresponding to the involved areas of the root responds with increased mineralization deep within the tissue, resulting in a zone of higher mineral content in the involved areas. Likewise, tertiary, reactive dentin is frequently observed at the pulpal surface of the dentin, corresponding to the involved tubules.

If the exposed root surface is still covered by a layer of cementum, the early stages in caries development normally involve a haphazard demineralization of this layer (Fig 185), because of acid formed by the acidogenic bacteria colonizing the root surface. Because of the much higher organic content in the hard tissue of the root, species other than those initiating enamel caries may sometimes be involved in the development of root lesions.

For example, in teeth with periodontitis, scaling and root planing may in places completely remove the hypermineralized cementum. In such areas, the combination of acidogenic and proteolytic bacteria will result in wide root dentin lesions beneath the remaining root cementum (Fig 186). The root cementum is only 0.03 to 0.10 mm thick on the coronal third of the root: 10 to 20 strokes with a sharp curette or a 0.10-second application of a 15-um diamond-coated rotating tip can completely remove such a thin layer, exposing the dentinal tubules of the root, which will immediately be invaded by microorganisms. A nonaggressive approach to scaling and debridement is therefore important.

Although root lesions do not usually exhibit cavities, it is very important that the surface zone of active radicular carious lesions not be iatrogenically damaged by gentle probing or by scaling. As with the surface zone of active carious lesions in enamel, damage to the surface of a root lesion may initiate the development of localized cavities.

Arrest

Despite the invasion of microorganisms into dentinal tubules in most carious lesions of the root, it is possible to arrest active lesions through improvement in oral hygiene and use of fluoride toothpaste. Studies by Nyvad and Fejerskov (1986) in elderly patients with active root lesions showed that improved oral hygiene converted active

lesions to inactive lesions. Figure 187 illustrates how, with improved oral hygiene, an active, plaque-covered root lesion on the buccal surface of a maxillary left canine gradually underwent changes in color and surface structure as the lesion became inactive. In a more recent study by Nyvad et al (1997), carious lesions were experimentally developed in situ on root specimens borne in partial dentures and then arrested by cleaning with fluoride toothpaste for 3 months.

Even root lesions with deep cavities can successfully be converted to inactive lesions by improved plaque control and use of fluoride. Figure 188 by Nyvad and Fejerskov (1997) shows active, plaque-covered root caries with a typical yellow color and a soft surface. After improved plaque control and topical use of fluoride, 10 years later the cavities were inactive, with a typical dark brown to black color and a semihard surface (Fig 189).

Fig 182 Plaque accumulation along the cementoenamel junction seen with stain solution. (From Thylstrup and Fejerskov, 1994. Reprinted with permision.) Fig 183 Plaque accumulation along the cementoenamel junction seen with stain solution. (From Thylstrup and Fejerskov, 1994. Reprinted with permision.) Fig 184 Microradiograms of sections of root caries lesions, developed experimentally in the oral cavity over 1, 2, and 3 months (left to right). Note that the mineral content in the surface zone gradually increases despite a growing subsurface mineral loss.(From Nyvad et al, 1989. Reprinted with permission.)

Fig 185 Demineralization of the cementum layer of an exposed root surface. (From Nyvad and Fejerskov, 1986. Reprinted with permission.)

Fig 186 Lesions in dentin (D) beneath the remaining root cementum after removal of hypermineralized cementum (C), most probably by aggressive scaling and root planing. (From Nyvad and Fejerskov, 1986. Reprinted with permission.)

Fig 187 Changes in color and surface structure in an active, plaque-covered root lesion after improvement in oral hygiene. (A) Active lesion. (B) Lesion after 2 months of improved oral hygiene. (C) Lesion at 6 months. (D) Lesion at 18 months. (From Nyvad and Fejerskov, 1986. Reprinted with permission.) Fig 188 Active, plaque-covered root caries with a typical yellow color and a soft surface. (From Nyvad, 1997. Reprinted with permission.)

Fig 189 Same cavities shown in Fig 188, 10 years later. With improved plaque control and topical use of fluoride, the cavities are inactive. (From Nyvad, 1997. Reprinted with permission.)

Diagnosis and Registration of Carious Lesions

Introduction

The coronal carious lesion starts as a clinically undetectable subsurface demineralization. With further progression, it will eventually become clinically detectable, and can then be classified according to type, localization, size, depth, and shape (see Table 15).

Apart from for the occult fissure lesion penetrating deeply into the dentin, dilemmas in clinical detection and registration arise not with the advanced lesion, but primarily with the early lesion (confined to the outer enamel), the noncavitated lesion of dentin, recurrent caries (around the margins of restorations), and subgingival root caries. This is further complicated by the fact that diagnostic methods and criteria may vary, depending on the purpose of the examination. For the epidemiologist, measuring caries prevalence or assessing treatment needs at the community level, or for the dental researcher, measuring caries increments in relation to the efficacy of an anticaries agent, decisions about the presence of a lesion are not complicated by the obligation to consider treatment options for the individual patient. In general, the epidemiologist surveying a population confines a positive diagnosis of caries to unequivocal cavitation, to reduce variability. Most guidelines for surveys specifically state that questionable lesions should be coded as sound.

On the other hand, for the clinician, detection of a lesion, whether confined to enamel and therefore potentially reversible, or frank cavitation, raises the question of appropriate treatment. In addition, different tooth surfaces present different problems for correct diagnosis of the questionable lesion. In other words the diagnostic method of choice depends on the purpose of the examination.

According to Pitts (1997), the ideal method or tool for diagnosis of carious lesions would be noninvasive and provide simple, reliable, valid, sensitive, specific, and robust measurements of lesion size and activity, and be based on biologic processes directly related to the carious process. It should also be affordable, acceptable to dentists and patients, and allow early implementation in both clinical practice and research settings. Its use should promote informed and appropriate preventive treatment decisions, enhancing long-term oral health. Unfortunately, there is at present no single, all-embracing method that fulfills these requirements. While awaiting further technologic development, dentists and researchers have to select the combination of methods that is most appropriate to the particular diagnostic task at hand.

Diagnostic tools

Some decades ago, visual diagnosis (light and mirror) and probing, supplemented by bitewing radiographs, were the only tools available for clinical diagnosis of caries. For epidemiologic surveys and for examination of most patients, these are still useful tools. However, the last 10 years have seen a considerable increase in the assortment of diagnostic tools based on new technology. The following methods are now available:

1. The visual method used by many general practitioners

2. The visual-tactile method with light, mirror, and gentle probing

3. The conventional visual method used in European epidemiologic surveys

4. The meticulous clinical visual method

5. The visual method with temporary elective tooth separation

6. The visual method with temporary elective tooth separation and impression of the approximal lesion

7. The conventional bitewing radiographic method

8. The digital radiographic method

9. The computer-aided radiographic method

10. The fiber-optic transillumination (FOTI) method

11. The electrical conductance (fixed frequency) method

12. The alternating current impedance spectroscopy technique (ACIST)

13. The endoscopic filtered fluorescence (EFF) method

14. The quantitative laser (light) fluorescence (QLF) method

The accuracy (sensitivity and specificity), usefulness, and cost effectiveness of these methods vary considerably. Some are very quick and inexpensive, but subjective, and are therefore useful for large-scale epidemiologic surveys (the visual-tactile and European epidemiologic methods), while others are objective and offer quantitative diagnosis but are time consuming and require costly equipment (ACIST, EFF, and QLF). At present, the latter methods are restricted to research projects.

Visual method used in general practice

The visual method, a combination of light, mirror, and the probe for detailed examination of every tooth surface, is by far the most commonly applied method in general practice worldwide. Although sensitivity is low and specificity is high, it may be possible to detect:

1. Noncavitated enamel lesions (D1) on the free smooth surfaces (buccal and lingual), most anterior approximal surfaces, and the opening of some fissures

2. Clinically detected "cavities" limited to the enamel (D1, D2)

3. Dentin lesions (D3) with cavitation into the dentin on the buccal and lingual

surfaces and the anterior approximal surfaces, but limited detection of posterior approximal and occlusal lesions

4. Secondary lesions with cavitation

5. Active and inactive root lesions, with or without cavitation

A major shortcoming is that this method is very limited for detecting noncavitated lesions in dentin on the posterior approximal and occlusal surfaces.

Clinical visual-tactile method

This method is based on a combination of light, mirror, and gentle probing and is used in most epidemiologic surveys in the United States. Caries is diagnosed if the tooth meets the American Dental Association criteria of softened enamel that catches an explorer and resists its removal (the so-called sticky fissure) or allows the explorer to penetrate proximal surfaces under moderate-to-firm probing pressure. Lighting is usually adequate, but the teeth are neither cleaned nor dried. The examination takes about 3 minutes per subject. The method is also used frequently in general practice in the United States.

Visual method used in European epidemiologic surveys

Probing has been criticized for several reasons: It may allow transmission of cariogenic bacteria from infected sites, it can irreversibly traumatize potentially remineralizable noncavitated lesions of enamel and dentin, and it may provide no more accuracy in diagnosis than visual inspection alone, particularly in the fissures and on the posterior approximal surfaces.

Accordingly, a so-called European system of examination for surveys, based primarily on detailed visual examination, has been adopted by many epidemiologists. The subjects clean their teeth before examination, the tooth surfaces are dried with compressed air, and the examination requires about 10 minutes per subject. It has been argued that under the prevailing disease conditions (low caries prevalence and slow progression) a visual technique that emphasizes specificity at the expense of some loss of sensitivity (for dentinal involvement) is to be preferred (for review, see Pitts, 1997).

Meticulous clinical visual method

Meticulous clinical examination following cleaning (including flossing) of all surfaces and thorough drying will disclose more lesions than the aforementioned rapid clinical examinations.

Visual method with temporary elective tooth separation

The once popular technique of temporary elective tooth separation as an aid to diagnosis of caries in approximal smooth surfaces is now regaining popularity, albeit with more humane and less traumatic methods that seem acceptable to most patients and dentists. This method permits a more definite assessment of whether

radiographically detectable approximal enamel (D1, D2) and dentin lesions (D3) are cavitated (Pitts and Longbottom, 1987; Pitts and Rimmer, 1992; Rimmer and Pitts, 1990). Figures 190, 191, and 192 illustrate the use of a regular orthodontic elastomeric separator for temporary tooth separation.

Visual method with temporary elective tooth separation and impression of the approximal lesion

Temporary elective tooth separation, complemented by a localized impression of the opened interproximal space, allows a more sensitive diagnosis of cavitation than does the purely visual separation method. This also has the advantage of providing a replica as a reference for visual monitoring of changes in size or even measurement of serial impressions (Neilson and Pitts, 1993; Seddon, 1989).

This method is currently gaining international acceptance, and a recently demonstrated association between cavitation status and carious activity may be an important step forward in aiding appropriate decision making (Bjarnason, 1996; Danielsen et al, 1996; Lunder and von der Fehr, 1996).

Conventional bitewing radiographic method

Several factors have contributed to the general adoption of radiographic examination as an aid to the detection and subsequent treatment of caries:

1. It discloses sites inaccessible to other diagnostic methods. Radiography facilitates detection of carious lesions at an earlier, potentially reversible stage. Usually, more approximal and occlusal lesions are recorded when clinical examinations are supplemented by radiography.

2. The depth of the lesion can be evaluated and scored, eg, by the radiographic index by Grondahl et al (1977), modified from Moller and Poulsen (1973): 0 = no radiographic changes in enamel; 1 = radiographic changes in enamel; 2 = radiolucency extending to the dentinoenamel junction; 3 = radiolucency penetrating approximately halfway through dentin; and 4 = radiolucency close to the pulp.

3. Because the radiograph provides a permanent record, recall examinations allow assessment of progression or regression of lesions, evaluation of disease activity, and the efficacy of preventive and therapeutic measures.

4. Radiography is noninvasive, whereas gentle probing may cause iatrogenic damage to the surface of noncavitated enamel and dentin lesions.

Radiographs have, however, some limitations:

1. For accurate reproducibility, standardized geometric angulation, exposure time, processing procedures, and analyzing facilities are necessary. A bitewing film holder fixed to a radiographic long cone facilitates standardized geometric angulation.

2. Radiography does not disclose the earliest stages of lesion development.

3. Radiography does not unequivocally distinguish among approximal surfaces that are sound, have subsurface lesions, or are cavitated.

4. To some degree, radiographs underestimate the extent of demineralization, but overestimations may also occur, as a result of projection errors.

5. Radiographic diagnosis is subjective, and the interpretation of radiographic findings is subject to interobserver and intraobserver variation.

6. Approximal secondary caries on the more apical part of a restoration may not be detected.

7. Noncavitated carious lesions on the root are difficult to diagnose.

There is a wealth of data relating to conventional radiographic techniques that are used in general practice, research, and clinical trials, but studies predating the recent changes in the pattern of the disease process should be extrapolated with caution to present conditions.

Radiographic results are best considered by site. For approximal surfaces, recent studies show moderate levels of sensitivity at the D1 threshold, disclosing many more relatively small approximal lesions that may be amenable to preventive care than are disclosed by most other techniques. Specificity is generally high, although not quite as high as for the clinical methods. At the D3 threshold, sensitivity is also moderate and specificity is high. For occlusal surfaces, newer findings have changed perceptions of performance and the applicability of radiographic methods (for review, see Pitts, 1997).

While the intrinsic image geometry of the bitewing projection, with superimposition of large volumes of sound enamel, precludes sensitive radiographic diagnosis of enamel lesions, the method is now highly applicable, with moderate sensitivity, for detecting extensive dentinal lesions which may be undetected at clinical examination.

Digital radiographic method

Digital, filmless, techniques for intraoral radiography have been developed for several important reasons:

1. Conventional film absorbs only a few percent of the x-rays that reach it, utilizing very little of the radiation to which the patient has been exposed.

2. Poor darkroom procedure can lead to both unnecessarily high doses of radiation and loss of diagnostic information.

3. Development of films is time consuming, and the developer and fixing solutions are hazardous to the environment.

For intraoral radiography, digital techniques with direct image acquisition have been available only since the end of the 1980s. Research and development in indirect digital radiography paved the way for direct digital (filmless) techniques. The first to

become available, in 1989, was based upon a charged-coupled device (CCD) chip similar to that found in digital videocameras. In most CCD systems, no more than one molar or two premolars and a small amount of periapical bone can be visualized in one image. The small image areas, the relatively bulky sensors, and the need for a wire connection between the sensor and the computer make the clinical use of these systems cumbersome. In addition, these systems have a relatively narrow dynamic range, which means that best image quality can be attained only within a limited exposure range.

The Digora image plate system is an alternative, with fundamentally different digital image acquisition from that of CCD systems: The radiographic information is captured on a phosphorus storage screen, or image plate (Fig 193). The essential components are the image plate and the readout devicethe scanner, which is connected to a personal computer. The outer dimensions of the scanning unit are 483 x 452 x 135 mm. After exposure, the image plate is placed in the scanner, where the laser beam is deflected across the phosphorus screen. The released light energy is collected in a photomultiplier and converted into an analog signal, which is then digitized. With the Digora system, the anatomic area displayed is almost the same as that shown in modern film-based technology (Fig 194).

Hence, apart from the scanning of the image plate, which corresponds to development of the film, all procedures necessary to obtain an intraoral radiograph are identical to those of conventional radiography. Readout of the image plate takes less than 30 seconds, during which time the image gradually appears on the computer monitor. The exposure range of the image plate is wide and linear. Because of the wide exposure range, the high sensitivity of the image plate, and the high quality of modern photomultiplier tubes, the image plate system can acquire data over many orders of magnitude in exposure compared to CCD or film systems.

As with other digital images, the Digora images can be altered after exposure to enable task-specific image characteristics. The system works in a Microsoft Windows environment, which simplifies all operating procedures. Image brightness and contrast can be changed by moving and angulating, respectively, a line displayed in a coordinate system where the gray-level values in the original image and the altered image are seen on the x-axis and the y-axis, respectively. The image-processing software also allows edge enhancement and gray-scale inversion.

In addition, different types of measurements, such as measurements of linear distances (in tenths of millimeters) and angles, can be performed. All values are displayed on the screen. It is possible to display a histogram of the distribution of gray levels within a chosen area, the mean gray level value and the deviation around the mean. The gray-level values may be displayed graphically along a line of selected position and angulation.

No significant differences have been found among the CCD systems (Durr Vista Ray Trophy RVG, Sens-A-Ray, and Visualix/Vixa) and the Digora system with respect to the accuracy with which approximal and occlusal carious lesions can be detected. However, the recently introduced CMDs/APS sensors have not been compared to the CCD and Digora systems.

Digital radiography linked to the dental unit offers an attractive design, because the flat screen is adapted to the bracket table of the dental unit, directly in front of the patient, facilitating discussion with the patient about findings from the radiographs as well as from an intraoral camera (Fig 195). Another alternative is Durr`s mobile system including Vista Ray digital radiography equipment, Vista Cam2 intraoral camera, flat screen, computer, and printer on a very stable mobile trolley, which can be easily moved around (Figs 196 (a and b)).

In general, the new digital systems are comparable to conventional radiography, although contrast enhancement may boost sensitivity at the expense of some loss of specificity.

Computer-aided radiographic method

Computer-aided radiographic methods exploit the measurement potential of computers in assessing and recording lesion size. In the new Trophy 97 system an artificial intelligence software (Logicon Caries Detector) is integrated: Approximal carious lesions are diagnosed and evaluated with the aid of a unique histologic database, allowing graphic visualization of the size and progression of the lesion (Figs 197a to 197c).

At both the D1 and D3 thresholds, computer-aided methods offer high levels of sensitivity for approximal lesions. Earlier software paid the penalty of some trade-off with specificity, but newer methods also have high values for this measure.

The trend is toward integration of several computer-aided tools and services in a local network with a powerful personal computer (Fig 198): Radiographs (conventional and digitized as well as computer aided) and clinical slides can be transmitted for consultations and information.

Fiber-optic transillumination method

Fiber-optic transillumination is a development of a classic diagnostic aid, advocated some 20 years ago, which has never gained wide acceptance. However, it should be a regularly used tool for diagnosis of caries, in the incisor and premolar regions at least, to supplement clinical examination and bitewing radiographs. Fiber-optic transillumination has enjoyed variable success in studies evaluating its performance, possibly because of failure to appreciate that the technique, like any other, requires an extended learning phase.

However, in a recent experimental study, Vaarkamp et al (1997) showed that use of wave length-dependent FOTI allowed quantitative diagnosis of early enamel lesions. In an earlier study, Verdonschot et al (1991) found that FOTI was more useful than bitewing radiographs for detection of enamel lesions. In another study, Peers et al (1993) evaluated in vitro the validity and reproducibility of clinical examination, FOTI, and bitewing radiographs for the diagnosis of small approximal carious lesions. The results showed that the validity of FOTI was at least as high as that of bitewing radiographs, and both diagnostic tools were superior to unaided clinical diagnosis.

Although FOTI cannot sensitively detect approximal lesions at the D1 level, it is

specific, and its success compared to clinical methods at the D3 level means that FOTI should seriously be considered as an adjunct and used in situations where radiography is not appropriate or feasible. Traditionally advocated for approximal lesions in dentin, FOTI may also have some application in the diagnosis of occlusal lesions. The results to date are equivocal, but Wenzel et al (1992) showed that, compared to visual inspection and various radiographic image modalities, the FOTI method gave, on average, the most accurate diagnosis of noncavitated, occlusal carious lesions of dentin in extracted teeth.

Electrical conductance (fixed frequency) method

Electrical methods of caries diagnosis are not new. There has been recent revival of interest in fixed frequency electrical devices, which show considerable promise for detection of occlusal and approximal lesions. A device is now commercially available in The Netherlands; similar machines were produced in the United States and in Japan some years ago.

The electrical detection methods are seen by many as having the greatest potential for significantly improving diagnostic performance in the years to come. For occlusal surfaces, the commercially available fixed frequency devices have repeatedly shown high sensitivity and specificity for enamel lesions. For lesions in dentin, sensitivity is high, although specificity is only moderate. There are special difficulties in applying this method to approximal surfaces, but recent data show that these can be overcome and that sensitivity is high at both the D1 and D3 levels. Specificity appears to be high at the D1 level, but may be a little lower (moderate) for D3 (Pitts et al, 1995).

Alternating current impedance spectroscopy technique

A more sophisticated approach to lesion detection and measurement is to characterize the electrical properties of the tooth and lesion by using the ACIST, which scans multiple frequencies. The ACIST is new and has been evaluated only to a limited extent on whole carious teeth. However, the results to date are extremely encouraging, indicating 100% sensitivity and specificity at the D1 level and only a marginal decrease in specificity at the D3 level (Longbottom et al, 1996).

Endoscopic filtered fluorescence method

Pitts and Longbottom (1987) explored the use of EFF for the clinical diagnosis of carious lesions and compared results with conventional alternatives on occlusal and approximal sites. This work developed to include the use of an intraoral video system for caries detection, the prototype "videoscope." Now that commercial intraoral cameras are increasingly available in practices, this may prove to be of practical clinical importance.

The EFF method has been shown to be highly sensitive for occlusal caries in enamel, but sensitivity is poor for occlusal caries in dentin (D3) (Ten Cate et al, 1996). Specificity is poor for occlusal surfaces but high for approximal lesions at both thresholds. The method is reasonably good at detecting approximal lesions in enamel but not lesions in dentin.

Quantitative laser (light) fluorescence method

A method that is related to EFF and is attracting considerable interest is the quantitative laser fluorescence technique. At present, QLF can assess only accessible smooth surfaces and is limited to part of the enamel thickness.

The principle for the QLF method is shown in Fig 199. The excitation is performed with blue-green light (488 nm) from an argon ion laser. The fluorescence in the enamel, occurring in the yellow region (approximately 540 nm), is observed through a yellow high-pass filter (520 nm) to exclude the tooth-scattered blue laser light. Dark regions characteristic of demineralization are registered visually or on a photographic film.

For the initial smooth-surface enamel lesions it can detect, the QLF method has a sophisticated computer-based method for measuring changes in lesion size, which is valuable for some applications.

Recently, a commercial laser fluorescence system, Kavo-Diagnodent, has been introduced. This system seems to be efficient for the diagnosis of noncavitated enamel and dentin lesions on buccal, lingual, and occlusal surfaces (Lussi et al, 1999). In particular, this system should be useful in longitudinal caries-preventive studies.

Selection

Table 17 reveals that most of the caries-diagnosing methods discussed are subjective, and this compromises the potential to make accurate measurements of disease activity over time. The potentially objective tools are the separation method used with a local impression, computer-aided radiographic diagnosis, the electrical methods, and QLF. These tools also have the potential for quantitative measurement, which can be used to aid the diagnosis and determination of carious activity and thus the prognosis. In the past, many of the diagnostic tools have been used only to support a single, dichotomous decision of the presence or absence of disease. Most methods can be used at either the D1 or the D3 threshold. The two exceptions are conventional epidemiologic examinations, which are undertaken at the D3 (dentin caries) level, and the QLF method, which can detect only enamel lesions.

Figure 200 summarizes the performance of the previously discussed methods, in terms of sensitivity and specificity for occlusal and approximal surfaces at diagnostic thresholds D1 (noncavitated enamel lesion) and D3 (dentin lesion). Clearly, no single method will be sufficient for accurate diagnosis of all kinds of carious lesions (see Table 15).

The relatively poor performance and widespread limitations of the available methods requires clinicians to seek better and more intelligent ways of using existing methods, while developing new, more accurate, more appropriate devices.

A target for future investigations would be to explore multiple methods in both supplementary and adjunctive combinations. To meet the clinical requirements of most general practitioners, a combination of meticulous clinical visual examination,

radiographs (conventional bitewing radiographs or digitized radiographs), and fiber-optic transillumination would be adequate.

Fig 190 Orthodontic elastomeric separators for temporary tooth separation. Different sizes are available for molars and premolars.

Fig 191 The application is facilitated by stretching the elastomeric separator with two surgical forceps.

Fig 192 The elastomeric separator in situ. (It remains for about 3 days in premolars and 5 days in molars.)

Fig 193 Digora image plate system. Radiographic information is captured on a phosphorus storage screen

Fig 194 Display of information in the Digora system.

Fig 195 New Kavo unit.

Fig 196 Durr's Vista Ray mobile system.

Fig 197a to 197c Size and progression of a carious lesion evaluated with the Trophy 97 system.

Fig 198 Components of an integrated computer-aided network for diagnosis and information sharing.

Fig 199 (a) Principle for the quantitative laser (or light) fluorescence method. (From Hafstrom and Bjorkman, 1992.) (b) the Kavo Diagnodent laser fluorescence instrument. (c) Diagnodent offers the advantage of fluorescence measurement in fissure areas where the laser is reflected through minute access routes. (d) Schematic of the Kavo Diagnodent system.

Fig 200 Summary comparison of performance of caries diagnosis tools by threshold (D1= noncavitated enamel lesion; D3= dentin lesion) and surface (approximal and occlusal) for (A) clinical; (B) separation, radiographic, and transillumination; and (C) electrical and optical methods. (GP) General practice; (EP) European epidemiology; (ECM) electrical conductance method; (ACIST) alternating current impedence spectroscopy technique; (EFF) endoscopically viewed filtered fluorescence; (QLF) quantitative laser (light) fluorescence. (From Pitts, 1997. Reprinted with permission.)

Diagnosis of occlusal caries

It might be expected that occlusal carious lesions would be fairly easy to diagnose, because unlike approximal and subgingival root surfaces, these surfaces are readily accessible for visual inspection. However, clinically (visual or visual-tactile by probing) or radiographically, diagnosis of occlusal lesions is a delicate problem, because of the complicated three-dimensional shape of the occlusal surfaces, incorporating fossae and grooves with a great range of individual variations.

Disease progression

It is a common clinical observation that caries on occlusal surfaces does not involve the entire fissure system with the same intensity but is a localized occurrence. Viewed in a stereomicroscope or SEM, the occlusal surface of a permanent molar appears as a convoluted landscape, with high mountains separated by valleys, some that are deep rifts and others that resemble open river beds (Fig 201).

Each tooth type in the dentition has its own specific occlusal surface anatomy, and caries is usually detected in relation to the same specific anatomic configuration in identical tooth types. In maxillary molars, for example, the central and the distal fossae are sites that typically accumulate plaque and hence also the sites at which caries most often occurs. In general, occlusal caries is initiated at sites where bacterial accumulations are well protected against functional wear (see Figs 20 and 174).

Thus, two factors have been considered of importance for plaque accumulation and caries initiation on occlusal surfaces: (1) stage of eruption or functional usage of teeth, and (2) tooth-specific anatomy. This explains why almost all molar occlusal caries is initiated during the extremely long eruption period (12 to 18 months) and why occlusal caries is uncommon in premolars, with an eruption time of only 1 to 2 months. This was confirmed in a 2.5-year longitudinal study by Mansson (1977), who examined first molars every 3 months from the start of eruption and found that development of caries occurred, on average, within 11 months of the start of eruption, ie, during eruption (most had decayed within 3 to 9 months). On the other hand, there was virtually no further initiation of occlusal caries beginning 15 months after the start of eruption.

Monitoring for and measures to prevent the development of occlusal caries should be intensified during the eruption period (high-risk period). If the teeth have erupted into natural chewing function without developing occlusal caries, then the risk is over; examinations can be more cursory and less frequent. Nor is there any indication for application of fissure sealant.

In cross section, most molar fissures have a relatively wide opening (entrance) followed by a narrow cleft, approximately 1.0 mm deep and 0.1 mm wide, almost to the dentinoenamel junction (see Fig 116). The carious lesion usually starts in the enamel on either side of the entrance to the fissures and is visible as a noncavitated white-spot enamel lesion. Gentle probing with a sharp explorer will damage the surface zone of such a lesion and initiate cavitation to the lesion body. A rule of thumb is to use sharp eyes and a blunt probe (or no probe at all) and to arrest the lesion by plaque control and fluoride.

Most clinical and scientific concern with respect to occlusal caries has been over the possible events in deep and inaccessible fissures. However, caries always starts in the surface enamel, from the metabolic activity of bacterial accumulations on the surface. It is reasonable to assume that evolution of plaque with cariogenic potential requires space that, in this context, is available only above the entrance to the narrow fissures, the grooves. This assumption is supported by ultrastructural studies indicating that, in contrast to the vital bacteria found at the entrance, nonvital bacteria or different stages of calculus formation are usually harbored by the depths of the fissures (Ekstrand, 1988; Theilade et al, 1976).

Fewer than 10% of fissures are atypically flask shaped, with a narrow neck and a bulbous base: The carious lesion can start at the entrance as well as at the base of the fissure (see Fig 117). These fissures should be regarded as at risk. Fortunately, from a diagnostic point of view, there is a strong correlation between steep cuspal inclination and such sticky risk fissures.

Figure 202 shows an unusually wide, shallow fissure, full of stagnant, cariogenic plaque and an associated noncavitated enamel lesion around the entire fissure. Figure 203 shows a so-called risk fissure with stagnant, cariogenic plaque in the entrance as well as at the base of the fissure. In this case, localized, noncavitated enamel lesions have developed on both sides of the entrance and around the bulbous base of the fissure. However, even extreme risk fissures, with irregularities such as horizontal tunnels, can be maintained free of caries (see Fig 118).

Progressive destruction of the occlusal surface thus begins as a local process in the deepest part of the groove-fossa system, as a result of accumulation of bacterial plaque. In this area, which is already sheltered from physical wear, the formation of a microcavity further improves the potential for bacterial attachment and colonization. This accelerates demineralization and destruction, further enhancing local conditions for bacterial growth.

Figures 204a, 204b, and 204c show different stages of localized progressive occlusal lesions in a mandibular molar with a discolored, cavitated lesion in the distal fossa. The cross section of the lesion in the fossa shows superficial enamel breakdown with cavitation into about 50% of the enamel but no cavitation into the dentin; ie, there is no bacterial invasion of the dentinal tubules, and the lesion could be arrested (Fig 204b). However, the enamel lesion (demineralized area) is approaching the dentinoenamel junction and there is demineralized dentin in the contact area, corresponding to the direction of the rods. The anterior fissure in Fig 204a is shown in cross section in Fig 204c. There is no progressive demineralization.

Figure 205a shows a localized cavity in the central fossa of a maxillary first molar. The cross section of the lesion shows that the cavity is truncated and that the superficial zone of destruction and the zone of dentinal demineralization are confined to the involved enamel (Fig 205b). The cross section of the fissures shows that these are filled with calcified material, indicating total absence of cariogenic plaque (Fig 205c). It is therefore assumed that the dark brown cavity is arrested or stagnant.

In people living in communities without dental health care, the natural progression of occlusal caries is rapid, because of the particular anatomic configuration of the occlusal surface where caries is initiated. Occlusal caries usually begins in a fossa, ie, a depression where two or more interlobal grooves meet. Several surfaces are involved in the initial dissolution, and the process is therefore three-dimensional. Because enamel demineralization always follows the rods, the enamel lesion initiated in a fossa gradually assumes the shape of a cone, with its base toward the dentinoenamel junction. The response by the dentin corresponds to the direction of the involved enamel rods. A section through such a lesion has the two-dimensional appearance of two separate, independent lesions, but the lesion is three-dimensional and actually cone shaped. Although textbooks have traditionally emphasized the undermining character of occlusal or so-called hidden caries, the pattern of lesion growth in these areas is not particularly surprising in the context of the structural arrangement of rods in the occlusal groove-fossa system.

As enamel destruction proceeds, a true cavity forms, the outline reflecting the arrangement of rods in the areas: The cavity has the shape of a truncated cone. The particular anatomic configuration of the occlusal surface at the site of caries initiation explains why the openings of occlusal cavities are always smaller than the base. The "closed" nature of the process obviously favors undisturbed growth of bacteria and hence accelerated destruction of the tissue. Occlusal enamel breakdown is the result of further demineralization from an initially established focus, rather than general demineralization involving the entire fissure system.

Figure 206 illustrates the progressive stages of lesion formation in an occlusal fossa, from the earliest noncavitated enamel lesion to cavitation into the dentin with a zone of bacterial invasion and dentin destruction, where excavation and restoration are indicated. However, at the second-to-last stage (E), no such invasive intervention is indicated, despite a considerable zone of demineralized dentin and a sclerotic and translucent zone into the pulp. The method of choice would be placement of a fissure sealant or a minimally invasive sealant restoration, using a resin-based glass-ionomer material or compomer.

Diagnostic methods

In typical fissures, and particularly in atypical sticky fissures (see Fig 203), most of the early stages of the lesion are hidden from the naked eye, although in a clean, dry fissure, it might be possible to observe active noncavitated white-spot lesions on the walls. Soon after eruption, most of these lesions are arrested (see Figs 174, 204a-c, 205c) and take up a brown stain from items in the diet. This diagnostic problem was recognized many years ago by GV Black (1908) who wrote:

Very many pits and fissures show evidence of some slight softening in early youth, which is stopped by the coming of immunity or some change of local conditions. These become dark in color and so remain without further change. These should not be interfered with, as they are just as safe without any filling whatever.

Dentists continue to this day to have difficulty in differentiating active from arrested lesions and usually base their decision on "clinical judgment," which should include an assessment of a patient's past caries experience, oral hygiene and dietary habits, salivary function, and likelihood of compliance with a preventive regimen.

None of the aforementioned diagnostic methods has gained universal acceptance, particularly for detection of early occlusal caries. Because occlusal caries today constitutes a major portion of all new lesions in children, its diagnosis assumes considerable importance. Several studies have reported that clinically undetected occlusal lesions, extending well into the dentin, are a serious problem in many communities. In one comparative study of molar lesions that had been clinically undetected (by mirror and probe), 10% more lesions were found with the aid of bitewing radiographs in 1974, whereas 32% more were found in 1982 (Sawle and Andlaw, 1988).

Radiographs, although better than visual inspection alone, are inaccurate in estimating the extent of the carious lesion or in detecting enamel occlusal lesions (D1). Radiographically, occlusal lesions appear as large, subtle, diffuse dark radiolucent areas in dentin, centrally located under the fissure. Figure 207 shows a limited lesion in dentin (D3) in the mandibular left second molar, and Fig 208 shows a very advanced lesion in dentin, probably involving the pulp (D4), in the mandibular right first molar.

The sensitivity of both visual and clinical-tactile (probing) diagnosis for noncavitated occlusal enamel and dentin lesions is low, but the sensitivity of meticulous clinical visual examination is somewhat better. However, all the clinical diagnostic methods have high specificity (see Fig 200).

Noncavitated occlusal carious lesions are by far the most prevalent lesions in children and young adults; traditionally, these have been restored three times more frequently than have similar lesions on buccal and lingual surfaces. Ideally, noncavitated lesions should be arrested instead of restored. Of great importance is differential diagnosis between active and nonactive lesions and between noncavitated and cavitated lesions.

Probing with explorers

During the past 10 years, the role of explorers in caries detection has become a controversial issue. Historically, an explorer was essential; if the tip caught in a pit or fissure, or a cavity, a restoration was indicated. There is no place for such a procedure in modern caries management: today a noncavitated lesion is managed by remineralization or by minimally invasive techniques, such as sealants or microrestorations.

There is also consistent evidence that explorers do not improve the accuracy of caries diagnosis. Applied with slight force, an explorer could damage a tooth surface,

converting a white-spot lesion into a cavity. Noncavitated lesions in narrow pits are particularly vulnerable (Figs 209a and 209b). Explorers should not be used to probe any pit or fissure, or any other tooth surface: their use in stained or noncavitated pits and fissures is unethical. When required, a blunt periodontal probe may be used, to remove plaque and debris from the tooth surface prior to examination and to check the surface texture of a lesion atraumatically.

Lussi (1991) demonstrated the limitations of the explorer as an aid to accurate diagnosis of occlusal lesions and treatment decisions in an in vitro study. Thirty-four dentists were instructed to provide diagnoses for 61 teeth and recommend treatment. The teeth were then histologically prepared and assessed (the gold standard). The agreement between histologic and clinical diagnoses was determined.

The results showed no difference in diagnostic accuracy between explorer and visual inspection only. Sensitivity (62%) and specificity (84%) showed that the dentists were more likely not to treat carious teeth than they were to restore sound teeth. Approximately 42% of teeth were diagnosed correctly, but the percentage of "clinically" correct treatment decisions was 73%. It was concluded that, compared to visual inspection alone, the use of an explorer does not improve the validity of the diagnosis of fissure caries.

Figures 210a, 210b, 210c, 211a, 211b, 212a, and 212b show the clinical appearance of the occlusal lesions and the histologic sections of the lesions. Because most of the extracted teeth were third molars that were only partly erupted or without chewing function, they exhibited active noncavitated enamel lesions in the fossae surrounding the fissures.

Radiographs

Although conventional bitewing radiography will significantly improve the potential for detecting noncavitated occlusal lesions in dentin (D3) (see Figs 200 and 207), there is some risk of a false-positive diagnosis.

In a Norwegian study (Espelid et al, 1994) 640 dentists were asked to diagnose the occlusal surface of an intact molar based on a color photograph of the occlusal surface (Fig 213a) and a radiograph (Fig 213b). Of the participants, 15% diagnosed a dentin lesion, 53% considered the occlusal surface to be intact, and 32% were uncertain. They were also asked to suggest therapy appropriate for a 20-year-old patient with average oral hygiene and dietary habits, presenting with such a lesion: for this individual, 22% proposed no treatment at all, 23% proposed fluoride treatment, 19% proposed fissure sealant, and 36% proposed restoration.

In a similar study in vivo (Elderton and Nutall, 1983) 18 subjects underwent clinical and (bitewing) radiographic diagnosis by 15 dentists. The suggestions for treatment were wide-ranging; from 20 to 153 occlusal restorations. Only two occlusal surfaces received an identical diagnosis by all 15 dentists. This study highlighted the need for considerable improvement in the competence of clinicians, not only with respect to diagnostic skills, but also in the management of occlusal caries.

To standardize the criteria for diagnosis based on meticulous clinical visual

examination and radiographs, a five-point scale has been proposed (Espelid et al, 1994; Tveit et al, 1994) (Fig 214).

Grade 1: noncavitated white spot or slightly discolored carious lesion in enamel and no lesion detected on the radiograph

Grade 2: some superficial cavitation in the entrance of the fissures, some noncavitated mineral loss in the surfaces of the enamel surrounding the fissures, and/ or a carious lesion in enamel, detected on the radiograph

Grade 3: moderate mineral loss with limited cavitation in the entrance of the fissure and/or a lesion into the outer third of the dentin, detected on the radiograph

Grade 4: considerable mineral loss with cavitation and/or a lesion into the middle third of the dentin, detected on the radiograph

Grade 5: advanced cavitation and/or a lesion into the inner third of the dentin, detected on the radiograph

New methods

Of the new diagnostic methods, an increase in electrical resistance measurement is a better predictor of occlusal carious lesions than are meticulous clinical visual examination, fissure morphology and discoloration, radiographs, and FOTI examination. The electrical conductance (fixed frequency) method has high sensitivity and specificity on occlusal enamel (D1) as well as dentin (D3) lesions.

In a recent in vitro study, Ekstrand et al (1997) compared the reproducibility and accuracy of a new visual scoring system, a new electrical conduction tool (the Electronic Caries Meter), and conventional radiographs for assessment of the depth of demineralization on the occlusal surface. After hemisectioning of the teeth, histologic evaluation was used as the gold standard.

The new visual system appears promising, but takes time to learn and requires the teeth to be clean. The reproducibility and accuracy of the electric conduction method was acceptable, while early occlusal lesions were not detectable on radiographs. The results of this study indicate that both the visual method and the electronic conduction method, used in conjunction with other relevant clinical observations, can improve the accuracy of diagnosis of occlusal caries.

Fig 20 Pattern of freely accumulated plaque on the occlusal surfaces of partly and fully erupted first molars 48 hours after PMTC. (Modified from Carvalho et al, 1989.)

Fig 116 Cross section of a molar fissure.

Fig 117 Atypical fissure with a narrow opening and a bulbous widening at the base. This type of fissure should be considered an at-risk fissure.

Fig 118 Fissures with horizontal tunnels in a 10-year-old girl. These irregularities put the first molar at extreme risk of developing caries.

Fig 201 Occlusal morphology of a permanent molar.

Fig 202 Polarized light micrograph showing, in cross section, an unusually wide, shallow fissure, full of stagnant, cariogenic plaque. An associated noncavitated enamel lesion is located around the entire fissure.

Fig 203 Risk fissure with stagnant, cariogenic plaque located at both the entrance and the base of the fissure. Localized, noncavitated enamel lesions have developed on both sides of the entrance and around the base of the fissure.

Fig 204a Localized occlusal lesion (arrows) in a mandibular molar with a discolored, cavitated lesion in the distal fossa. (From Thylstrup et al, 1989. Reprinted with permission.)

Fig 204b Cross section of the lesion in the distal fossa. There is superficial enamel breakdown with cavitation into about 50% of the enamel but no cavitation into dentin. Thus the cavity could have been treated with glass ionomer without drilling into the dentin, despite obvious demineralization of dentin (D) corresponding to the direction of the enamel rods of the lesion. (From Thylstrup et al, 1989. Reprinted with permission.) Fig 204c Cross section of the anterior fissure (arrowheads) indicated in Fig 204a. There is no progressive deminera- lization. (From Thylstrup et al, 1989. Reprinted with permission.)

Fig 205a Localized cavity (arrows) in the central fossa of a maxillary first molar. (arrowheads) Fissures. (From Thylstrup et al, 1989. Reprinted with permission.)

Fig 205b Cross section of the lesion revealing a truncated cavity. The superficial zone of destruction and the zone of dentin demineralization (D) are confined to the involved enamel. (From Thylstrup et al, 1989. Reprinted with permission.)

Fig 205c Cross section of the fissures indicated in Fig 205a. They are filled with calcified material, indicating the total absence of cariogenic plaque. Thus the superficial dark brown cavity of the fissure should be regarded as inactive and arrested. (From Thylstrup et al, 1989. Reprinted with permission.)

Fig 206 Progressive stages of lesion formation in an occlusal fossa. (A) Noncavitated enamel lesion. (B) Superficial cavitated enamel lesion. (C) Superficial cavitated enamel lesion with some reaction of dentin. (D) Superficial cavitated enamel lesion with noncavitated dentin lesion. (E) Superficial cavitated enamel lesion with noncavitated dentin lesion and some pulp reaction. (F) Cavitated dentin lesion and more chemical reaction of the pulp. (1) Translucent zone into the pulp; (2) zone of demineralized dentin; (3) bacterial invasion and destruction of dentin. (Modified from Ekstrand et al, 1995). Fig 207 Limited lesion in dentin (D3) in the mandibular left second molar.

Fig 208 Advanced lesion in dentin, probably involving the pulp (D4), in the mandibular right first molar.

Fig 209a Noncavitated white-spot lesion. (Courtesy I. Espelid.)

Fig 209b Possible creation of a cavity with pressure from an explorer. (Courtesy I. Espelid.)

Fig 210a Noncavitated occlusal lesion (arrow) in the distal fossa of an extracted third molar. This lesion should have been detectable on a conventional bitewing radiograph. (From Lussi, 1991. Reprinted with permission.)

Fig 210b Polarized light micrograph of a cross section of the lesion. Note the superficial demineralization (red) around the entrance of the fissure, down to the dentinoenamel junction, and extending into the outer third of dentin. Even at this stage of lesion development, invasive therapy is contraindicated. PMTC and fissure sealant with slow release of fluoride should be appropriate. At most, a pointed diamond tip might be used to open the entrance of the fissure, to confirm the diagnosis. In the absence of cavitation at the base of the fissure, a minimally invasive adhesive restoration would be placed. (From Lussi, 1991. Reprinted with permission.)

8 additional images not shown.Diagnosis of approximal caries

The issues to be considered by the clinician with respect to caries of the approximal surfaces are similar to those considered at other sites: Is the surface sound, or is there a lesion? If so, how advanced is the lesioninvolvement of enamel only, enamel and dentinal involvement, or pulpal exposure? Finally, is there cavitation?

Diagnostic methods

Meticulous visual examination

In the thin anterior teeth, both noncavitated and cavitated approximal lesions are readily detectable by meticulous clinical visual examination. The diagnostic potential is improved by fiber-optic transillumination.

Temporary tooth separation

The approximal surfaces of the posterior teeth cannot usually be viewed directly, and temporary elective tooth separation has been recommended as an inexpensive, reversible method of clinical examination (Pitts and Longbottom, 1987) (see Figs 190, 191, and 192). The main disadvantage is the need for two appointments.

Supplemented by impressions, this method offers the highest sensitivity and specificity both for enamel (D1) and dentin (D3) lesions (see Fig 200). It also allows the clinician to determine whether the lesion is noncavitated or cavitated and to estimate the size and depth of the cavity. The impression itself, or a cast replica, can be stored for future reference, to allow monitoring of progression of cavitation.

Pitts and Rimmer (1992) applied this method in a study of more than 200 schoolchildren and correlated the findings with the appearance of the lesions on bitewing radiographs: None of the radiographic lesions in the outer half of the enamel exhibited cavitation, 10% of the lesions in the inner half of the enamel and 40% of lesions penetrating to half the depth of the dentin were cavitated, and all the lesions involving more than half the depth of the dentin exhibited cavitation. However, no data were provided on cavity depth, ie, the numbers of cavities located in the enamel and in the dentin.

Lunder and von der Fehr (1996) applied the same technique to investigate the relationship between cavitation and carious activity, by comparing the prevalence of cavitation in radiographic enamel lesions extending to the dentinoenamel junction (D2) and into the dentin (D3) in 17 to 18 year olds with high or moderate carious activity. Cavitation of D2 lesions was very rare in moderately caries-active subjects, while there was usually cavitation of both D2 and D3 lesions in highly caries active subjects. However, the depth of the cavitieswhether involving only superficial enamel or extending further into the enamelwas not evaluated.

All noncavitated approximal lesions, not only in enamel (D1 and D2), but also in dentin (D3), can be arrested by intensified, targeted plaque control and use of fluoride. It is unlikely that cavitated lesions will be arrested, because plaque will be retained in the inaccessible cavity. The risk of overtreatment by operative intervention can be minimized by applying the temporary tooth separation method to differentiate between noncavitated and cavitated approximal lesions. The method warrants more widespread application in developed countries with well-organized dental care and declining caries prevalence.

Radiographs

To date, the bitewing radiograph has been the standard diagnostic method for detecting approximal lesions, and its value is beyond dispute. Radiographs usually disclose 50% more (30% to 69%) small approximal lesions than does clinical examination alone. As discussed earlier, there are three possible outcomes for such small approximal lesions: progression, arrest, or regression (Meja`re et al, 1999).

In Swedish preschool children (4 to 6 year olds), Moberg-Skold et al (1997) detected 3.8 approximal carious lesions per child by using bitewing radiographs, compared to only 2.0 lesions per child by meticulous clinical visual examination. However, in schoolchildren with declining caries prevalence, the efficiency of bitewing radiographs in diagnosing approximal lesions warrants critical reassessment. De Vries et al (1990) concluded that, for diagnosis of approximal caries, there would not be substantial loss of information without radiographs in children younger than 12 years, but that radiographs should be included in examinations of older children.

The frequency of monitoring of these age groups with periodic bitewing radiographs is related to the risk category: annually for high-risk patients and every 2 to 4 years for low-risk patients (the risk category should be reassessed at each visit before radiographs are taken). In children at the mixed dentition stage, a 1- to 2-year interval is recommended. In Western industrial societies, progression of approximal enamel lesions to the stage at which there is radiographic evidence of extension into dentin is now only gradual (3 to 9 years), allowing ample time to monitor lesion progression, arrest, or regression (see Meja`re et al, 1999).

Bitewing radiographs used for this purpose must be standardized with respect to positioning, exposure, and development of the films, and overlapping of the teeth must be avoided. Film-holding beam-aiming devices are useful for standardizing the position of the film and the angulation of the beam. In addition, films should be read with proper lighting and magnification.

The use of ionizing radiation is not without hazard, and all efforts must be made to minimize radiation exposure, consistent with achieving a diagnostically useful image. The radiation dose can be reduced by collimation of the x-ray beam, ensuring that the timer is functioning accurately, avoiding technical errors and attendant retakes, using fast E-speed film, and protecting the patient with a lead apron and a thyroid collar.

For systematic longitudinal monitoring of approximal caries, several scores or indices have been proposed. The D1 to D4 scale was described earlier in this chapter. The following three-point scale for radiographically detectable lesions was introduced some years ago by the Swedish Board of Health and Welfare:

D1: lesion in the outer half of the enamel

D2: lesion in the inner half of the enamel

D3: lesion in dentin

Figure 215 illustrates the five-point Norwegian system for scoring radiographs (Espelid and Tveit, 1984):

Grade 1: lesion in the outer half of the enamel

Grade 2: lesion in the inner half of the enamel but not into the dentin

Grade 3: lesion into the outer third of the dentin

Grade 4: lesion into the middle third of the dentin

Grade 5: lesion into the inner third of the dentin

Treatment decisions: Prevention versus extension

An unresolved question is not the detection of approximal caries, but at what stage restorative treatment is indicated. Some dentists believe that all lesions should be restored, irrespective of extent. In regions where carious activity remains relatively high, operative intervention for an approximal lesion with a radiolucency confined to enamel is not unusual; elsewhere, it is considered unethical to restore such early lesions. To restore lesions that are not active and progressively extending through the enamel is tantamount to malpractice. A majority of dentists consider that an appropriate end point for nonoperative intervention is radiographic evidence that the lesion has extended into the dentin. However, provided that there is no cavitation in the dentin, the dentinal tubules have not been invaded by microorganisms, and the lesion can still be arrested. Operative intervention for noncavitated lesions of dentin should therefore also be regarded as malpractice.

The dentist should also be aware that restoration of the approximal surfaces is not without its hazards. One recent study reported a high frequency of iatrogenic damage (about 66%) associated with cavity preparation for restoration of approximal lesions (Qvist et al, 1992). A so-called proxitector should be used to protect the neighboring approximal surface when rotating instruments are used. Such a surface may have a noncavitated lesion that could be arrested.

In teenagers and young adults, the cervical margins of approximal restorations are frequently placed subgingivally, because the interproximal "space" beneath the contact area is completely occupied by the buccal and lingual papillae and the col. The cervical margins of most such restorations are inadequately finished, some have overhangs, and composite materials are particularly plaque retentive. By enhancing plaque retention, these factors contribute to increased risk for recurrent caries and development of periodontitis.

In determining whether invasive therapy is indicated, the clinician must address the following questions:

1. How rapidly is the lesion progressing?

2. What is the size or depth of the enamel or dentin lesion (see the Norwegian five-grade scale)?

3. Is the lesion noncavitated or cavitated? This may be determined by temporary separation if no cavity is found by meticulous clinical examination.

4. What is the patient's predicted caries risk and risk profile?

5. Why have the patient's self-care and the professional preventive measures failed to prevent development of the lesion?

6. How can combined preventive efforts be improved to arrest the lesion?

7. How much time should elapse before the outcome of preventive efforts is evaluated?

The second and third points in this list were evaluated by Bille and Thylstrup (1982) in a study of 8- to 15-year-old children. The subjects were examined clinically and radiographically before undergoing regular dental care, including restorations. Approximal lesions were scored on the radiographs according to the following system (Fig 216), which is similar to the diagnostic scales proposed by Moller and Poulsen (1973), and Grondahl et al (1977):

Score 0: enamel with no radiographic changes

Score 1: radiographic changes in enamel

Score 2: radiolucent lesion reaching the dentinoenamel junction

Score 3: radiolucent lesion penetrating approximately halfway through dentin

Score 4: radiolucent lesion close to the pulp

During cavity preparation, drilling was discontinued when the maximal extent of the lesion could be seen on the base of the approximal box, cervical to the interproximal contact area. With the help of an intraoral mirror and probe in normal clinical lighting, the tissue changes observed at the base of the approximal box were classified according to the six-point clinical scoring system (see Fig 216). This became the gold standard for diagnosis of the approximal lesions:

Scores 1 and 2: progressive changes in the enamel

Score 3: changes in dentin, without cavitation in the enamel

Scores 4 and 5: changes in dentin and progressive cavitation in the enamel (ie, at this stage no bacterial invasion of the dentinal tubules has occurred, and there is no indication for operative intervention)

Score 6: cavitation involving dentin (possible indication for "drilling, filling, and billing")

Figure 217 shows a bitewing radiograph from the study, showing radiolucent enamel lesions, radiographic score 1, on the distal surface of the mandibular right first molar and the mesial surface of the mandibular right second molar. The radiolucency of an approximal enamel lesion represents the sum of the loss of mineral from every single

enamel prism: At this stage, there has been no destruction of prisms. A cavity prepared by drilling from the occlusal surfaces to the depth of the lesions observed on the radiographs disclosed that these are enamel lesions without cavitation (clinical score 1), limited to the distolingual surface of the first molar and the mesiolingual surface of the second molar. Note the location of the lesions and the close proximity to the inflamed gingival margin and the etiologic factor: the supragingival and subgingival plaque (Fig 218).

Such enamel lesions should be arrested by use of a fluoridated wooden toothpick twice a day from a lingual direction, supplemented at needs-related intervals by PMTC with reciprocating pointed triangular tips and application of fluoride varnish. As discussed earlier, in vivo studies by von der Fehr et al (1970) and Holmen et al (1987) (see Figs 159, 160, 161, and 162) and in vitro studies by Silverstone, (1973) (see Figs 157 and 158) have shown that experimentally induced enamel lesions can successfully be arrested.

If the approximal surfaces can be maintained as sound, or with noncavitated lesions, through the period of secondary maturation, the risk for future cavitation is diminished (Kotsanos and Darling, 1991).

The relationship between radiographic and clinical scores is shown as a cross-tabulation in Fig 216. Of 158 lesions, radiolucencies penetrating approximately halfway through the dentin (radiographic score 3) were found in only 58, and none of these exhibited clinical cavitation into the dentin (clinical score 6). Only two of nine lesions with a radiographic score of 4 had a clinical score of 6 (Bille and Thylstrup, 1982). Other studies have shown similar results (Mejare and Malmgren, 1986; Pitts and Rimmer, 1992).

These studies confirm the importance of not interpreting radiographic findings in isolation, to ensure that appropriate treatment decisions are made about approximal lesions. Rather, radiographic findings must be assessed in the context of data from other sources: temporary tooth separation and meticulous clinical visual examination as well as careful consideration of each of the seven questions listed earlier.

Of the 158 carious lesions scheduled for restoration, cavity preparation disclosed that 66% were without macroscopic cavitation. The changes observed by direct clinical inspection of the tissues during cavity preparation correlated poorly with accepted standardized radiographic criteria. Thus cavitation was observed in only 20% of lesions with radiolucencies extending to the dentinoenamel junction and in 50% of lesions with radiolucencies involving the dentin, and all cavities were confined to the enamel.

Bille and Thylstrup (1982) concluded that: "Assuming a macroscopical cavity into the dentin to be indicative of restorative treatment, the present results indicate that a more individualized treatment decision strategy than hitherto is warranted in populations attending comprehensive dental health care." Although this study was published in 1982, it seems to have had little impact on treatment decision making as taught in undergraduate dental education or as practiced by most general practitioners.

In a Swedish study, dental hygienists and general practitioners examined approximal

lesions on radiographs of 100 extracted teeth and conducted meticulous clinical visual examinations of more than 200 patients. No statistically significant differences were found between dental hygienists and general practitioners in the inaccuracy in detecting and recording lesions. It was concluded that, by undergoing examination by a dental hygienist instead of a general practitioner, no patient with restorative treatment need would have been overlooked, and a more accurate nonrestorative (preventive) treatment need might have been addressed (Ohrn et al, 1996).

In a later large-scale study, Thylstrup et al (1986) reaffirmed their earlier findings (Bille and Thylstrup, 1982) by using the same diagnostic scoring systems. At the time of operative treatment, 263 Danish dentists recorded clinical tissue changes in 1,080 approximal lesions. The findings were related to type of tooth, age of the patient, and information available from radiographs. The tissue changes observed varied considerably, ranging from barely discernible white-spot lesions to frank cavitation. This variation occurred irrespective of information available from radiographs. The most frequently observed stage of lesion formation was a stage preceding cavitation. Of 78 lesions with a radiographic score of 3, true cavitation was observed in fewer than 10%. Of primary molars, only three exhibited cavitation into the dentin. Cavitation into dentin was observed in only three of 330 lesions with radiographic scores of 3, and in 27 of 102 lesions with radiographic scores of 4. With the current potential for arresting caries through noninvasive methods, the findings indicate the need to review the rationale for operative treatment of approximal lesions.

In another in vivo study, Meja`re et al (1985) examined 63 teenagers who were to have premolars extracted for orthodontic indications. Comparison between direct clinical examination after extraction and radiographs showed that noncavitated lesions were the most frequent finding on the tooth surfaces examined (305 noncavitated versus 28 cavitated). Of the noncavitated lesions, about 66.6% (n = 203) were not detected by the bitewing radiographs. When the findings from direct inspection were compared with those from clinical examination with an explorer, the accuracy of diagnosing the absence of cavitation was 98.0% for incipient lesions and 100.0% for sound tooth surfaces. Only 28.5% of the cavitated surfaces observed by direct inspection were identified by probing.

In a supplementary study by Meja`re and Malmgren (1986), clinical tissue changes were recorded during restorative treatment of approximal lesions in young premolars and molars. Sixty approximal surfaces with radiographic radiolucencies in the inner half of the enamel or the outer half of the dentin were treated. The extent and character of the tissue changes were documented by magnification of photographs taken during cavity preparation. The maximum extent of each lesion was correlated to the extent of the radiographically observed lesion.

Cavitation was observed in 78% of the lesions with radiolucencies in the outer half of the dentin, which is considerably higher than was found in the previously described studies by Bille and Thylstrup (1982), Thylstrup et al (1986), and Pitts and Rimmer (1992). However, these cavities were mainly limited to the enamel. In all cases, there was discoloration of the enamel; the dentin was soft and discolored in 83% of the lesions. In 12%, the tooth substance was severely damaged. Figure 219 shows examples of the lesions. With the preventive materials and methods available today, in theory the lesions that had not progressed to cavitation of dentin could have been

arrested. However, the plaque-retentive cavities, although limited to the enamel, had a less favorable prognosis for caries arrest.

These studies confirm the importance of supplementing radiographic examination of approximal carious lesions with clinical inspection prior to treatment decisions. For ethical reasons, in the study by Pitts and Rimmer (1992), visual inspection was achieved by using orthodontic elastometric separators for temporary tooth separation. Bjarnason (1996) used a similar technique for temporary separation, not only for clinical inspection, but also for minimally invasive operative intervention. The restorative technique combined an external approximal composite filling, placed under rubber dam isolation, and so-called partial tunneling.

Atraumatic restorative technique

Based on the current knowledge the following example illustrates recommended procedures for noncavitated approximal lesions of enamel and dentin and for lesions with limited cavitation: In a new patient, bitewing radiographs reveal one or more posterior approximal lesions in the outer half of the dentin. Meticulous clinical visual examination discloses no cavitation.

At the same or the following appointment, plaque is mechanically removed by PMTC, in which reciprocating tips (Eva-Profin) are used on the affected approximal areas (Figs 220 and 221). A slow-release chemical plaque control agent (Cervitec 1% chlorhexidine-thymol varnish) is applied to these surfaces to destroy the remaining cariogenic microorganisms (Fig 222). An orthodontic elastometric separator is then inserted for temporary elective tooth separation to allow clinical visual inspection after 5 days (Figs 223 and 224).

At the second appointment, after PMTC, the approximal surfaces are inspected directly. Elastomeric impressions may be taken, to allow replication as records for future reference. Lesions showing no cavitation are coated with chlorhexidine and fluoride varnish (Cervitec + Fluor Protector) to arrest and seal the outer micropore surface of the lesions (see Fig 223). If a limited cavity is diagnosed, it is mechanically cleaned with a small ball-shaped finishing bur and filled with light-cured glass-ionomer cement, placed under pressure from a translucent matrix band. After they are finished with an extrathin tungsten-coated reciprocating tip, these surfaces are coated with chlorhexidine and fluoride varnish to arrest and "seal" the surface of the surrounding, noncavitated enamel lesions (see Fig 224).

Fig 210c Close up of the dentinal involvement of the lesion. Note the dentinal tubules in the inner part of the lesion. (From Lussi, 1991. Reprinted with permission.)

Fig 211a Noncavitated superficial lesion (arrow) surrounding the entrance of the fissure. It is unlikely that lesion would have been detectable on a conventional bitewing radiograph. (From Lussi, 1991. Reprinted with permission.)

Fig 211b Polarized light micrograph of a cross section of the fissure shows noncavitated as well as dentin lesions. (From Lussi, 1991. Reprinted with permission.)

Fig 212a Obvious cavity (arrow) in the distal fossa. (From Lussi, 1991. Reprinted with permission.)

Fig 212b Polarized light micrograph of a cross section of the lesion. The cavitation was limited to the outer surface of the enamel around the wide, shallow fissure. The demineralized zone of enamel has not progressed to the dentinoenamel junction. (From Lussi, 1991. Reprinted with permission.) Fig 213a Occlusal surface of an intact molar, used in a study of diagnostic accuracy. (From Espelid et al, 1994. Reprinted with permission.)

Fig 213b Radiograph of the same molar. (From Espelid et al, 1994. Reprinted with permission.)

Fig 214 Five-point scale for diagnosis of occlusal caries on radiographs. (left to right) Grade 1 = noncavitated white spot or slightly discolored carious lesion in enamel, and no lesion detected on the radiograph; grade 2 = some superficial cavitation in the entrance of the fissures, some noncavitated mineral loss in the surfaces of the enamel surrounding the fissures, and/or a carious lesion in enamel, detected on the radiograph; grade 3 = moderate mineral loss with limited cavitation in the entrance of the fissure and/or a lesion into the outer third of the dentin, detected on the radiograph; grade 4 = considerable mineral loss with cavitation and/or a lesion into the middle third of the dentin, detected on the radiograph; grade 5 = advanced cavitation and/or a lesion into the inner third of the dentin, detected on the radiograph. (From Espelid et al, 1994. Reprinted with permission.) Fig 215 Five-point Norwegian scale for diagnosis of approximal caries on radiographs. (left to right) Grade 1 = lesion in the outer half of the enamel; grade 2 = lesion in the inner half of the enamel but not into the dentin; grade 3 = lesion into the outer third of the dentin; grade 4 = lesion into the middle third of the dentin; grade 5 = lesion into the inner third of the dentin. (From Espelid and Tveit, 1984. Reprinted with permission.) Fig 216 Comparison of radiographic and clinical scoring of approximal carious lesions. Radiographic scoring (Moller and Poulsen, 1973; Grondahl et al, 1977): (0) no radiographic changes in enamel; (1) radiographic changes in enamel; (2) radiolucent lesion that has reached the dentinoenamel junction; (3) radiolucent lesion

penetrating approximately halfway through dentin; (4) radiolucent lesion close to the pulp. Clinical scoring: (1, 2) progressive changes in enamel; (3) changes in dentin without cavitation in the enamel; (4, 5) changes in dentin and progressive cavitation in the enamel, ie, still no bacterial invasion of the dentinal tubules and no indication for invasive tooth preparation; (6) cavitation involving dentinpossible indication for tooth preparation and restoration. (Modified from Bille and Thylstrup, 1982.) Fig 217 Radiolucent carious lesions in enamel on the distal surface of the mandibular right first molar and the mesial surface of the mandibular right second molar. (From Bille and Thylstrup, 1982. Reprinted with permission.)

Fig 218 Cavity preparations to the depths of the lesions, as indicated on the radiograph, revealing that these are enamel lesions without cavitation, localized on the distolingual surface of the first molar and the mesiolingual surface of the second molar. Note the close relationship among the localization of the lesions, the inflamed gingival margin, and the supragingival and subgingival plaque (the etiologic factor). (From Bille and Thylstrup, 1982. Reprinted with permission.)

Fig 219 Clinical tissue changes in approximal lesions in young premolars and molars. In theory, with the preventive materials and methods available today, lesions that have not progressed to cavitation of dentin (A, B, C, D, E, and H) could have been arrested. However, the plaque-retentive cavities (B, D, and E), although limited to enamel, would have had a less favorable prognosis for caries arrest. Normally, lesions F and G have to be restored. (From Meja`re and Malmgren, 1986. Reprinted with permission.)

Fig 220 Professional mechanical toothcleaning.

Fig 221 Professional mechanical toothcleaning.

Fig 222 Application of a slow-release chemical plaque control agent (Cervitec) to destroy cariogenic organisms.

Fig 223 Prevention instead of extension. (PMTC) Professional mechanical toothcleaning; (CHX) chlorhexidine; (F) fluoride. (From Axelsson, 1995.) Fig 224 Atraumatic restorative treatment. (PMTC) Professional mechanical toothcleaning; (CHX) chlorhexidine; (F) fluoride. (From Axelsson, 1995.)

Diagnosis of root caries

Definition and classification

Root caries usually appears as a shallow area, less than 2 mm deep, a mostly noncavitated, ill-defined, softened, and often discolored lesion, characterized by destruction of cementum and penetration of dentin. Several definitions and classifications have been proposed; Hix and O'Leary (1976) defined root caries as "a cavitation or softened area in the root surface which might or might not involve adjacent enamel or existing restorations (primary and recurrent lesions)."

Billings (1986) proposed the classification presented in Box 20, based on four grades of severity: incipient; shallow surface defect (less than 0.5 mm deep), some pigmentation; deep lesions (more than 0.5 mm deep) with cavitation; and pulpal involvement.

Nyvad and Fejerskov (1982, 1987) proposed differentiation between active and inactive lesions, using the following criteria:

1. Active root surface lesion. Any area that is well-defined and shows a yellowish or light brown discoloration. The lesion is most likely covered by visible plaque and/or presents a softening or a leathery consistency on probing with moderate pressure (Fig 225).

2. Inactive (arrested) root surface lesion. Any root surface area that shows a well-defined, dark brown or black discoloration. The surface of the lesion is smooth and shiny and appears hard on probing with moderate pressure (Fig 226).

Figure 226 illustrates how the active lesion in Fig 225 has been arrested and inactived by improved plaque control and use of fluoride toothpaste (Nyvad and Fejerskov, 1986). Both active and inactive lesions may exhibit cavitation, but in the latter the margins are smooth (see Fig 226).

Root caries may be classified as primary or secondary, cementum or dentin, active or inactive, with or without cavitation. The lesions can also be classified according to the texture: soft, leathery, or hard, and the color: yellow, light brown, dark brown, or black (see Table 15). When root surfaces are exposed to the oral environment as a result of gingival recession, the areas of potential plaque retention increase, particularly in the large interproximal areas and along the cementoenamel junctions. The primary carious lesion of the root has a greater horizontal than vertical dimension because of the greater thickness of supragingival plaque along the gingivocervical margin. Initial active root lesions are soft on probing, have a leathery consistency, and are normally covered with plaque. The color is yellow or light brown but, with longer exposure to the oral environment, changes to dark brown and black. This change results from extrinsic factors, such as staining from dietary components and smoking, and possibly from chromogenic bacteria present in the lesion.

Figure 227 illustrates various root lesions.

Diagnostic methods

The problems associated with diagnosis of root caries are somewhat different from

those associated with coronal caries. On the buccal and lingual surfaces, meticulous clinical visual examination will be sufficient for differential diagnosis of a lesion as cavitated or noncavitated or active or inactive, according to the aforementioned criteria. In borderline cases between inactive and active lesions, the texture of the surface (soft, leathery, or hard) is more important than the color (yellow, tan, brown, dark brown, or black). Lynch and Beighton (1994) reported that, irrespective of color, soft lesions are closest to the gingival margin, and hard lesions are the most distant; leathery lesions occupy an intermediate position.

For clinical examination, sharp eyes and a blunt probe are recommended, and even gentle probing with a sharp explorer is contraindicated

The most difficult type of lesion to diagnose is on approximal surfaces where there has been loss of attachment but no recessionin other words, lesions within deep periodontal pockets that are hidden from view. Vertical (standing) bitewing radiographs are essential for diagnosis. These lesions appear to progress more rapidly than enamel lesions, and failure to detect them at an early stage may result not only in pulpal involvement but also in a tooth that cannot be treated endodontically. Clinicians are cautioned to differentiate such true carious lesions from cervical radiolucency, which appears as a dark shadow on approximal root surfaces as a result of the contrast between adjacent parts of the image.

Treatment

As with active enamel lesions, treatment of active root caries initially should be preventive and noninvasive and directed toward arresting and converting the lesion from active to inactive. As shown in the study by Nyvad and Fejerskov (1986), discussed earlier, it is possible to arrest even cavitated carious lesions of the root simply by improving mechanical plaque control by self-care and instituting the use of fluoride toothpaste (see Figs 187, 188, 189 and 227 (c and f)). Repeated PMTC in combination with, for example, slow-release chlorhexidine and fluoride varnish further improves the potential for arresting active root lesions.

However, restoration of black, cavitated, but inactive root lesions may be indicated, not only because of plaque retention but also for esthetics. The restorative materials of choice are tooth-colored, fluoride-releasing materials such as glass-ionomer materials, resin-based glass-ionomer materials, or compomers.

Fig 225 Active root surface lesion.

Fig 226 Inactive root surface lesion, arrested by improved plaque control and use of fluoride toothpaste.

Fig 227 Various radicular carious lesions: (a) active buccal lesion; (b) inactive buccal lesions; (c) active lesions covered with plaque; (d) inactive lesions and proper oral hygiene; (e) active buccal lesion; (f) same lesion converted to an inactive state after 12 months of improved oral hygiene and application of fluoride. (From Ravald, 1992. Reprinted with

permission.)

Diagnosis of secondary caries

Definition and prevalence

Secondary caries has been reported to be eight times more common than primary lesions in adults, particularly in those older than 50 years (Goldberg et al, 1981). However, prevalence may vary markedly in different countries, depending on the total caries prevalence in the population and the level of development of the dental care system. In developing countries with low caries prevalence in the adult population and poor dental care resources, secondary caries may be almost negligible. On the other hand, the prevalence of secondary caries is very high in industrialized countries with high caries prevalence among adults (particularly many filled surfaces) and a dental care system with emphasis on restorative rather than preventive dentistry.

In a cross-sectional study in randomized samples of 35, 50, 65, and 75 year olds in the county of Varmland, Sweden, examined by a combination of meticulous clinical visual examination and complete-mouth radiographs, 1.0 secondary lesion per individual was found, on average, in the 50, 65, and 75 year olds. The number of primary coronal lesions was 0.4, 0.3, and 0.2, respectively, in the three older age groups. The number of root lesions was 0.3 in 50, 65, and 75 year olds. The 35 year olds exhibited 0.7 secondary lesions and 0.7 primary coronal lesions, but no root lesions (Axelsson et al, 1990). In a 15-year longitudinal preventive study based on mechanical plaque control, the subjects developed only one new carious lesion per individual per 15 years, even though the oldest age group was 65 to 85 years old at reexamination. However, 90% of the lesions were secondary caries (Axelsson et al, 1991).

With the exception of the occlusal surfaces, secondary caries occurs most frequently on the surfaces most frequently restored, ie, the approximal surfaces of the posterior teeth (mostly subgingivally), followed by the buccal surfaces of the posterior teeth and the lingual surfaces of the mandibular molars. According to a recent review, secondary caries is the major reason for the failure of restorations (Kidd et al, 1992).

Diagnostic methods

The diagnosis of secondary caries is usually based on clinical examination, including gentle probing, and may not be correct. For example, it is generally considered that the wider the gap at defective margins of restorations, the greater the likelihood of recurrent caries. Two studies in which extracted teeth were examined visually and by probing revealed these methods to have poor validity, predictive value, and specificity for detection of actual secondary caries, as determined after the removal of the restoration (Soderholm et al, 1989) or sectioning of the tooth. In fact, in more than 50% of cases in which replacement of the filling was recommended, this was not justified on the basis of true secondary caries.

There are, however, other indications for replacing restorations with faulty margins. In a study attempting long-term elimination of mutans streptococci from the mouths of adults by use of chlorhexidine varnish, treatment was unsuccessful in one third of

the subjects, because of retentive areas that served as a reservoir for the organisms (Sandham et al, 1988).

The difficulties involved in diagnosis of secondary caries are only in some respects similar to those in primary lesions. As with primary lesions, there is the problem of differentiating active caries from a chronic, static lesion. If a lesion is not progressing, it may not require operative intervention. Unfortunately there are currently no clinical variables with which to differentiate active from inactive secondary caries. Secondary caries may be located on the crown as well as on the root and is frequently located cervically, involving both crown and root along the cementoenamel junction. It may be noncavitated or cavitated (see Table 15).

The diagnosis of secondary caries is also associated with other problems. First, lesions on the occlusal surface, between the restoration and the enamel (the so-called wall lesion) cannot be detected until they have reached an advanced stage. Such lesions spread more in the dentin than in the enamel. The color next to amalgam is not always predictive, because gray or blue discoloration could be due to corrosion products as well as to secondary caries.

Second, lesions at the cervical approximal margin, the most frequent site of secondary lesions (about 94% of amalgam and 62% of composite restorations) (Mjor, 1985), can be detected only on radiographs, by careful comparison with previous bitewing radiographs. Dark shadows at the margin of the restoration indicate secondary caries (Fig 228). However, the lesion may not always be detectable on the radiograph: For example, a limited secondary lesion on the mesiolingual margin of a restoration may be obscured by a more apically located mesiobuccal margin. Radiographs should, therefore, be used in combination with meticulous clinical examination, including probing, to determine whether the lesion is cavitated or not. Accessibility for probing the cervical margins of approximal restorations is not a problem.

Figure 229 shows, in chronologic order, a Norwegian five-grade system for scoring secondary caries on the approximal surfaces (Espelid and Tveit, 1986):

Grade 1: noncavitated white spot or light, discolored lesion and/or radiographically detectable lesion in the outer half of the enamel

Grade 2: superficial cavitation and/or radiographic lesion in the inner half of the enamel

Grade 3: small cavity and/or radiographic lesion in the outer third of the dentin

Grade 4: substantial cavity and/or radiographic lesion into the middle third of the dentin

Grade 5: advanced cavitation and/or radiographic lesion into the inner third of the dentin

Fig 228 Dark shadows at the margin of the restorations, indicating secondary caries. The black line indicates the localization of the gingival margin and papillae.

Fig 229 Five-point Norwegian scale for diagnosis of secondary approximal caries on radiographs. (left to right) Grade 1 = noncavitated white spot or light, discolored lesion and/or radiographically detectable lesion in the outer half of the enamel; grade 2 = superficial cavitation and/or radiographic lesion in the inner half of the enamel; grade 3 = small cavity and/or radiographic lesion in the outer third of the dentin; grade 4 = substantial cavity and/or radiographic lesion into the middle third of the dentin; grade 5 = advanced cavitation and/or radiographic lesion into the inner third of the dentin. (From Espelid and Tveit, 1986. Reprinted with permission.)

Conclusions

Development

A carious lesion should be regarded as damage resulting from the infectious disease dental caries. The coronal lesion starts as clinically undetectable demineralization of enamel, visible only at the microscopic level, and proceeds gradually to visible, noncavitated demineralization of first the enamel surface and then the dentin, and finally to cavitation of the dentin. Primary carious lesions are most frequently located supragingivally on the crowns and particularly on the occlusal surfaces of the molars and the approximal surfaces of the posterior teeth. Root caries may occur in elderly people and other adult caries-risk patients with root surfaces exposed by periodontal disease.

Carious lesions may be classified according to type (primary or secondary caries), location (crown, root, and surfaces), size and depth (enamel, dentin, or root cementum), and shape (noncavitated, smooth, rough or soft surface, cavitation, etc). From a treatment needs aspect, it is important to evaluate whether the lesions are active or inactive and noncavitated or cavitated.

Enamel lesions develop mainly where cariogenic plaque accumulates and remains undisturbed for lengthy periods. This plaque, together with accessible fermentable carbohydrates, results in prolonged periods of low pH. In toothbrushing populations, these conditions occur most frequently on the approximal surfaces of the posterior teeth (particularly from the mesial surfaces of the second molars to the distal surfaces of the second premolars) and the occlusal surfaces of erupting molars (particularly the distal fossae).

Moderate pH levels result in a noncavitated enamel lesion with a surface zone that is like a micropore filter, and a so-called lesion body where there is greater loss of minerals from every single enamel prism. Such a noncavitated lesion can be arrested without any loss of surface minerals. At very low pH (4.5 to 5.0), an erosive-like enamel lesion will develop with intraprismatic dissolution of the enamel with microcavities. Such a lesion can also be arrested, but with some loss of the enamel surface compared with the surrounding intact enamel. In other words, all noncavitated as well as more active erosive-like enamel lesions should be diagnosed as early as possible and arrested to prevent extension of the lesion into the dentin, with eventual cavitation, which would generally require operative intervention.

In populations with low or moderate caries prevalence and access to well-organized preventive programs, most enamel lesions will be arrested and therefore never progress into the dentin. On the other hand, in highly caries-active individuals with poor oral hygiene and no daily use of fluoride, enamel lesions can progress to cavitation of the dentin very rapidly (within 6 to 12 months). The demineralization of the dentin at the dentinoenamel junction corresponds to the width of the outer surface of the enamel lesion and progresses in the same direction as the dentinal tubules. Very advanced demineralization of the dentin may occur without enamel cavitation; under such conditions there is no microbial invasion of the dentinal tubules. All noncavitated lesions of dentin can and should, therefore, be arrested and not be treated invasively.

Root caries develops only on root surfaces that are accessible to cariogenic microflora and exposed to the cariogenic plaque (biofilms). The primary lesions develop predominantly at plaque-retentive areas, particularly in the large interproximal areas and along the gingival margin and the cementoenamel junction. These represent stagnant areas where plaque accumulates. The primary root lesion has greater horizontal than vertical dimensions because of the greater thickness of supragingival plaque along the gingivocervical margin. If the exposed root surface is still covered by a layer of cementum, the early stages in caries development involve a haphazard demineralization of this layer, because of acid formation by the acidogenic bacteria colonizing the root surface. Such initial lesions result in a soft, yellow, but noncavitated root surface that can be arrested.

However, where root cementum has been removed in patches as an iatrogenic effect of aggressive scaling and root planing, the exposed root dentin may be destroyed very rapidly; cavity formation is a combined effect of acidogenic microorganisms and collagenase-producing microorganisms.

Even cavitated active root lesions can be converted to nonactive brown-black lesions by improved plaque control and use of fluoride. Restoration of such arrested lesions may still be indicated, however, not only to prevent plaque retention, but also to improve esthetics.

Diagnosis

For several decades, the accepted method for detecting carious lesions in patients, as well as in clinical trials, has been a combination of clinical visual-tactile (light, mirror, and probing) examination and bitewing radiographs. For most patients, these techniques are still appropriate. However, over the last decade there has been a considerable increase in materials and methods available for this purpose:

1. The visual method, still used by many general practictioners

2. The visual-tactile method with light, mirror, and gentle probing

3. The conventional visual method in use in European epidemiologic surveys

4. The meticulous clinical visual method based on mechanical cleaning and drying of

the tooth surfaces before examination

5. The visual method with temporary elective tooth separation

6. The visual method with temporary elective tooth separation and the use of elastomeric impression material for evaluation of the size and depth of cavitated lesions

7. The conventional (bitewing) radiographic technique

8. The digital radiographic method, which minimizes radiation exposure compared to conventional radiographs

9. The computer-aided radiographic method, which utilizes the measurement potential of computers in assessing and recording lesion size

10. The fiber-optic transillumination method (FOTI)

11. The electric conductance (fixed frequency) method

12. The endoscopic filtered fluorescence method (EFF)

13. The alternating current impedance spectroscopic technique (ACIST)

14. The quantitative laser (light) fluorescence method (QLF)

The accuracy (sensitivity and specificity) and applicability of these methods vary considerably. Some, eg, the visual-tactile and European methods, are very rapid and inexpensive, but subjective, and are therefore useful for large-scale epidemiologic surveys. Others are objective and offer quantitative diagnosis but are very time consuming and require costly equipment (ACIST, EFF, and QLF), and to date are being applied only in limited research projects.

The diagnostic method of choice depends on the purpose of the examination. Apart from the occult fissure lesion penetrating deeply into the dentin, difficulties in clinical detection and registration arise not with the advanced lesion but primarily with the early lesion (confined to the outer enamel), the noncavitated lesion of dentin, recurrent caries (around the margins of restorations), and subgingival root caries.

The general trend in clinical examination is away from reliance on gentle probing with a sharp explorer, toward meticulous visual inspection (sharp eyes and a blunt probe) for early detection of noncavitated lesions: Treatment is directed toward arresting the lesion and preserving the surface zone, to prevent the initiation of cavitation.

In general practice, the aforementioned method, in combination with radiographs (conventional bitewings or digital) and FOTI should be adequate for diagnosis of carious lesions in patients. Temporary elective tooth separation should be used to determine whether posterior approximal lesions of dentin are noncavitated or cavitated.

For diagnosis of occlusal caries, probing offers no advantage in accuracy over visual inspection after mechanical cleaning and drying. Noncavitated active enamel lesions should be detectable as white spots on either side of the entrance of the fissures or pits. However, even noncavitated lesions of dentin occur frequently.

Conventional bitewing radiographs improve the potential for detecting both noncavitated and cavitated occlusal lesions of dentin. However, there is some risk of false-positive diagnoses.

A new aid, Diagnodent, based on QLF technique, appears to be advantageous for detection of occlusal enamel as well as dentin caries lesions.

All noncavitated occlusal lesions, in enamel as well as in dentin, can be arrested by plaque control and fluoride or treated with noninvasive methods such as fissure sealants. In a radiographically detected lesion of dentin, or by QLF aid, explorative opening of the entrance to the fissures may be indicated, to allow direct visual inspection of possible cavitation through the base of the fissure into the dentin. In the absence of cavitation, the fissure should be sealed with a fluoride-releasing material (resin-based glass-ionomer materials or compomers).

Of the recent innovations, the QLF technique followed by the electrical conductance (fixed frequency) method seems to be the most promising for diagnosis of occlusal lesions of enamel as well as dentin.

In the anterior teeth, approximal lesions of enamel and dentin, noncavitated as well as cavitated, are readily detectable by a combination of meticulous clinical visual examination and FOTI. In the posterior teeth, radiographs (bitewings, digital, or computer aided) are excellent for detection of approximal lesions in enamel and dentin, as well as for monitoring progression, arrest, or regression. In this area, meticulous clinical examination is of limited value in the detection of early lesions.

However, neither radiography nor meticulous clinical examination is useful for differentiating between cavitated and noncavitated approximal lesions of dentin. Temporary elective tooth separation is therefore recommended; for visual inspection with or without a pouring of an impression for future reference. Studies have found no dentin cavitation in the majority of radiographically detected lesions in the outer half of the dentin. Such lesions should be arrested rather than restored.

For differential diagnosis and for assessment of treatment needs, root caries should be classified as active or nonactive and noncavitated or cavitated. This can usually be done by meticulous clinical examination. A rough, soft plaque-covered surface and yellow color indicate an active lesion, while a smooth, dark brown or black surface that is hard to moderate probing pressure indicates an inactive (arrested) lesion.

Cavity formation may be associated with both active and nonactive lesions, but in the latter the margins appear smooth. On the buccal and lingual surfaces, meticulous clinical examination should be adequate for diagnosis according to the aforementioned criteria. Probing with a sharp explorer is contraindicated.

The most difficult type of root lesion to diagnose is located subgingivally on approximal surfaces. Meticulous clinical examination should be supplemented by vertical (standing) bitewing radiographs.

As with active enamel lesions, the initial treatment of active root lesions should be preventive and noninvasive (improved plaque control and use of fluoride), aimed at arresting the lesion and converting it from active to inactive.

The major emphasis is on the early detection of the noncavitated active lesion. Treatment is preventive and needs related, aimed at arresting the lesion as soon as possible. Because the arrested lesion does not progress to cavity formation, there is no indication for operative intervention (for reviews on the development of carious lesions, see Thylstrup et al, 1994; Thylstrup and Fejerskov, 1994; for reviews on diagnosis of carious lesions, see Grondahl, 1994; Ismail, 1997; Pitts, 1997).

Chapter 6. Epidemiology of Dental Caries

Introduction

An important function of the World Health Organization (WHO) Oral Health Unit is the collection and analysis of global epidemiologic data on oral diseases, recorded in national, computer-aided studies. Goals for the level of oral health status are set and revised at certain intervals. Because epidemiologic studies measure dental caries in groups or populations, some care must be taken to ensure that the same diagnostic criteria are applied to each individual examined.

Dental caries presents interesting challenges for epidemiologists. For example, the signs of the disease (lesions) may be found on several sites and/or several teeth in the individual and frequently vary in severity. Carious lesions exhibit a broad spectrum of clinical features, depending on how far destruction has progressed on a particular surface. Early demineralization may be detected only with the aid of sophisticated techniques, such as radiography, or after careful cleaning and drying and meticulous examination of the surface. In the more advanced lesion, cavitation is readily detected. At the intermediate stages, the broad range of clinical signs represent past or current carious attack.

For the results of a particular study to be meaningful, the researchers must establish certain criteria:

1. To fulfill the purpose of the investigation.

2. To allow consistency in application by the examiner over the period of the study (reproducibility).

3. To allow consistency between examiners (if more than one is involved).

4. To establish external validity (ie, measure what they are supposed to measure).

5. To provide a pathobiologic rationale (ie, actually reflect the disease).

There is no gold standard for caries in epidemiologic studies. The most important determinant should be the purpose of the study.

Limitations of Epidemiologic Surveys

Caries thresholds

Simple oral health surveys usually apply criteria from WHO guidelines, recording the signs of disease only at an advanced stage, on a dichotomous principle (yes or no); that is, the surface can be recorded only as either sound or carious (caries is recorded as present when a lesion in a pit or fissure, or on a smooth tooth surface, has a detectably softened floor, undermined enamel, or a softened wall. A tooth with a provisional restoration should also be included in this category. On approximal surfaces, the examiner must be certain that the explorer has entered a lesion. Where any doubt exists, the surface is recorded as sound.

In surveys, some compromise is inevitable, and this factor is not inconsequential. Because the criteria are established for practicality and convenience under specific conditions, a surface that does not fulfill the minimum criterion for a positive diagnosis is not necessarily sound. For convenience and practicality, a number of false-negative results will be accepted; along with the truly sound surfaces, surfaces with some degree of caries will also be denoted as sound. In other words, the true number of carious teeth and surfaces is considerably underestimated in epidemiologic studies conducted according to WHO criteria.

Pitts (1997) has pointed out that noncavitated enamel lesions (D1 and D2) are about three times more common than are lesions in dentin (D3 and D4), particularly those with cavitation into the dentin. The precision of caries diagnosis is illustrated as an iceberg in Fig 231; the level at which the iceberg "floats" will depend on the selected threshold. In this figure, the water level is at the threshold used in classic dental epidemiologic studies; caries into dentin (D3); ie, the examiner ignores all signs of lesions less severe than clinically detectable lesions in dentin and records such surfaces as "caries free." The iceberg has been stratified into discrete levels, or diagnostic thresholds, from the most severe D4 (lesions extending into the pulp chamber) to subclinical lesions, less advanced than even clinically detectable D1 lesions (enamel lesions with apparently intact surfaces).

The D1 to D4 terminology has formerly been widely used by the WHO. The two most commonly selected thresholds are D3 (dentin caries, comprising D3 and D4 lesions only) and D1 (enamel caries, comprising lesions at D1 + D2 + D3 + D4). Figure 231 shows clearly that examinations in clinical practice will detect more lesions than will examinations using the same methods at a different threshold in a survey. Similarly, the use of diagnostic aids will also result in the detection of more lesions.

For example, in contrast to national epidemiologic surveys according to WHO criteria, surveys in Sweden routinely record approximal caries on the basis of bitewing radiographs, and enamel lesions (D1, D2) as well as noncavitated and cavitated lesions in dentin are detected. Compared to other national surveys,

epidemiologic data from Sweden, which include noncavitated approximal lesions in dentin, are therefore overestimated. Figure 232 from the county of Jonkoping, Sweden, illustrates the proportion of approximal enamel (D1, D2) and dentin lesions detected on bitewing radiographs in 3-, 5-, 10-, 15-, and 20-year-old children and young adults in 1973, 1983, and 1993 (Hugoson et al, 1999).

There are, moreover, additional problems when caries is measured as a dichotomous variable, because mineral loss from the surface, leading to cavitation, represents a continuum of changes as a result of the carious process. To dichotomize this continuous variable inevitably results in some loss of information (just as it would, for example, if the same were done when the height of a growing child was measured). Unfortunately, there are no methods of measuring the lesion as a continuous variable (like height or weight). There is, however, no a priori reason to classify lesions only as either one of two categories (present or absent). Alternatives have been proposed, eg, the Norwegian five-point scores for occlusal, approximal, and secondary caries (Espelid and Tveit, 1986; Espelid et al, 1994; Tveit et al, 1994; see chapter 5).

With all methods of measuring caries, two additional descriptive dichotomous categories are always included: filled (presumably because there had, at some time, been a carious lesion), and missing (for teeth extracted because of caries). An additional problem arises in epidemiologic studies of dental caries. The unit of attack of a lesion is usually the surface of a given tooth, eg, occlusal, mesial, buccal, distal, or lingual. Depending on the purpose of the study, these surfaces may constitute the unit of diagnosis: The worst lesion present on the surface determines the classification. It may sometimes be necessary to classify the surfaces differently: for example, when information on caries affecting different morphologic types (eg, pits, fissures, and smooth surfaces) is required. The unit of diagnosis is not fixed. For rapid surveys, it may be appropriate to classify each tooth, rather than the surface, based on the worst condition on any surface. Epidemiologic studies in children deal almost exclusively with primary coronal caries. In adults, however, coronal and root caries are usually considered separately, and from a treatment needs aspect, secondary caries is also included. Each method has its own strengths and limitations, and some compromise may have to be made regarding what loss of information is tolerable for the specific purpose of the study.

Fig 231 Levels of caries diagnosis in epidemiologic surveys, represented by an iceberg. The threshold that is selected determines the level at which the iceberg "floats." In this illustration, the threshold traditionally used in epidemiologic studies, caries into dentin (D3), is the water level. The examiner ignores all signs of lesions less severe than clinically detectable lesions and records such surfaces as "caries free." (FOTI) Fiber-optic transillumination; (BWs) bitewing radiographs. (From Pitts, 1997. Reprinted with permission.) Fig 232 Proportion of approximal enamel and dentin lesions detected on bitewing radiographs of children and young adults in the county of Jonkoping, Sweden, from 1973 to 1993. (From Hugosson et al, 1999. Reprinted with permission.)

Reproducibility

It is also important that the measurements be reproducible, to allow some general conclusions about the study population or the sample examined. The following are necessary to ensure reliability of the measurements:

1. A well-defined set of diagnostic criteria

2. Examiners who have been adequately trained in using the criteria

3. Conditions that are conducive to conducting the examinations

4. Measurement of the degree to which each examiner is internally consistent (intraexaminer reproducibility)

5. Measurement of the degree to which each examiner applies the same criteria (interexaminer reproducibility), if more than one examiner is involved

To assess intraexaminer consistency, a subsample is usually reexamined by the same examiner, and the results of the two examinations are compared. To determine interexaminer reliability, each examiner examines the same group of subjects.

Individual variation

Irrespective of the purpose of the study, in dental epidemiology the unit of analysis is always the individual subject, for two reasons. First, teeth do not exist independently of the individual: It is not possible to define a population of teeth that is independent of the individuals in whom they occur. It is not even possible to draw a sample of teeth (at least, not without extracting them). A sample of individuals is drawn from a population comprising individuals. Second, the teeth of an individual are not independent of each other: They share the same biologic environment, and the risk factors that influence one tooth are likely to influence the others. However, there are also different ecosystems or microenvironments at the tooth and surface levels, and this explains why the range of caries prevalence among different tooth surfaces is sometimes greater than that among individuals (eg, the difference between the occlusal surfaces of the mandibular first molars and the lingual surfaces of the mandibular incisors).

Recording systems

Decayed, missing, or filled (DMF) scores are used to collect epidemiologic data about the prevalence of coronal caries in permanent teeth or surfaces. (In the primary dentition, caries prevalence is expressed in decayed, extracted, and filled teeth [DEFT] or surfaces [DEFS].) Once the caries status of the individuals has been recorded, the next step is to assign some score that expresses their accumulated caries experience, for example, simply by counting the number of surfaces (or teeth) that are decayed (D), filled (F), and missing (M). The sum of these provides a score for the individual. If surfaces have been counted, the score is termed DMFS; or if the teeth have been counted, the score is termed DMFT. The DMF system is often referred to as an index. This is not strictly correct, because the term index implies the use of the same diagnostic criteria, and these scores may be derived from a variety of different diagnostic criteria; therefore, the index score from one study may not necessarily be

equivalent to the index score of another.

As discussed already, most national epidemiologic surveys of caries prevalence include only carious lesions with visible cavitation, while in Sweden even noncavitated dentin lesions detected on bitewing radiographs are included. In addition, an unknown number of noncavitated enamel lesions (D1 and D2) are not included in the national surveys based on the DMF score. Thus, the D component of the DMF score, particularly at the surface level, is considerably underestimated.

Until recently, in most industrialized countries with well-organized dental care, the "extension for prevention" principle has been practiced for several decades; noncavitated occlusal carious lesions and sticky fissures, not only in molars but also in premolars, have routinely been filled, and restoration frequently has included the fissure systems, extending to the buccal surfaces of the mandibular molars and the palatal surfaces of the maxillary molars. In addition, approximal enamel lesions detected on bitewing radiographs have been restored, with the inclusion, for retention, of many intact occlusal surfaces (particularly in premolars). Complete-coverage crowns of ceramic or porcelain fused to gold have also been used for esthetics, and many missing teeth have been replaced with fixed partial dentures, sometimes requiring abutment preparation of intact teeth. In other words, in these countries, the F component is considerably overestimated in the DMF scores (particularly DMFS scores) of adults of all ages.

In many developing countries with limited dental resources, the M component, particularly in DMFS scores, is overestimated, because lesions involving only one surfacefor example, the occlusal surfaces of the molarsmay progress undisturbed into the pulp or until the crown of the tooth is completely destroyed. In most industrialized countries, the occlusal surface would have been sealed or restored at an early stage.

In the permanent teeth of children and young adults, caries prevalence should be based on a DFS score, rather than DMFT: In these age groups, fewer teeth are lost to caries, and increasingly more because of orthodontic indications or trauma. The DFS score is also useful for estimating current treatment needs. For example in 12 year olds (the WHO indicator age group for caries prevalence), a DMFT score of 1 in Scandinavia is often also a DMFS score of 1, representing overtreatment by restoration of the occlusal fissure of a first molar. In many developing countries, the same DMFT score may represent one MT (= 5 MSs?) or 1 to 5 DSs with cavitation. In other words, although the DMF scores, particularly DMFT, are used worldwide to express caries prevalence (accumulated caries experience), there is a risk for generalization and misinterpretation.

Prevalence of Caries

Coronal caries

Prevalence in children and young adults

During the last two decades, caries prevalence has decreased significantly among children and young adults in most industrialized countries. Every year since 1969, the

WHO has compiled a world map of caries prevalence among 12 year olds, expressed in decayed, missing, and filled teeth. At age 12 years, the five-level scale varies from 0.0 to more than 6.5 DMFT (Table 18).

The WHO's goals for caries prevalence in 12 year olds in the years of 2000, 2010, and 2025 are 3, 2, and 1 DMFT, respectively. Figures 233, 234a, and 234b show global maps of caries prevalence in 12 year olds in 1969, 1993, and 1997. In 1969, there were sharp contrasts: The number of DMFT was very high, high, or moderate in the industrialized countries and generally very low, low, or occasionally moderate in the developing countries.

During the following 20 years, there was a downward trend in caries prevalence in most of the industrialized countries, with particularly dramatic improvement (from very high to low prevalence) in the Scandinavian countries, Australia, and New Zealand. From 1993 to 1997, the following decreases in levels of DMFT were achieved: Canada, France, Spain, Italy, Greece, and Iceland from moderate to low; Brazil, Peru, and Paraguay from very high to high; Germany and the Balkan countries from high to moderate; and Australia and Finland from low to very low. On the other hand, during the same period, prevalence increased from very low to low in most of China and from low to moderate in South Africa.

Regionally, some areas improved even more dramatically; for example, in the county of Varmland, in southwestern Sweden, prevalence declined from among the highest in the world (40 DFSs in 1964) to very low ( 1 DFS in 1994).

Caries prevalence has decreased from very high or high levels to moderate or low levels in most industrialized countries, while it has increased from very low or low levels to low or moderate levels in many developing countries. Figure 235 shows the mean level of DMFT in 12 year olds in industrialized and developing countries compared to the global mean from 1980 to 1991. The trend in developing countries is similar to the global trend, because the developing countries represent most of the world's population, with about 40% living in China, India, and Indochina. In the developing countries, the general trend is for caries prevalence to increase, except where preventive programs have been established.

What lies behind the spectacular fall in caries prevalence in some countries? How can it be prevented from rising again? How can the deteriorating situation in other countries be halted? The answer to all three questions is one and the same: prevention, more prevention and still more prevention. The factors underlying the unprecedented public health success story in the industrialized world are the promotion of oral hygiene, the widespread use of fluoride toothpaste, the fluoridation of water (or, in some countries, the introduction of fluoridated salt), nutritional counseling (no sweets between meals, etc), and the establishment of well-organized, school-based preventive programs (particularly in Scandinavia).

Apart from the fluoridation of water, salt, and milk, which requires more advanced technology and supervised central administration, the aforementioned methods use simple techniques, cost little, and are perfectly suited to implementation at the primary health care level. As a result of the progress made in the last 25 years, the developing countries now have the knowledge and means of prevention to enable

them to avoid the costly problems that the industrialized countries have had to face, and indeed are still facing.

Dental care, as currently practiced, is categorized as follows:

1. "Primary" primary prevention: Preventive measures for all expectant mothers, to prevent postnatal transmission from mother to child of cariogenic microbes and poor dietary habits

2. Primary prevention: Maintenance of the intact dentition, ie, prevention of dental caries, gingivitis, and periodontitis in a completely healthy mouth

3. Secondary prevention: Prevention of recurrence of disease (dental caries, gingivitis, and periodontitis) after successful symptomatic treatment

4. Tertiary prevention: Symptomatic treatment of dental caries, gingivitis, and periodontitis, ie, restorations, scaling, and periodontal surgery

5. Relief of pain: Extractions and endodontic treatment

Table 19 shows the distribution in percentage of the different forms of dental care for children and young adults in the county of Varmland, Sweden, from 1900 to 1990, underlying the dramatic reduction of DFSs in 12 year olds from 1964 (40 DFSs) to 1994 ( 1 DFS).

As an effect of primary primary prevention and primary prevention, available to expectant mothers and 1 to 3 year olds through prenatal and maternal and child welfare centers, the percentage of caries-free 3 year olds in Sweden has increased from only 35% in 1973 to 97% in 1993.

Since 1979, a computer-aided epidemiologic system has been used to evaluate caries prevalence and annual incidence in all 3- to 19-year-old individuals in the county of Varmland, Sweden. Figure 236 compares the mean DFT scores in 7 to 19 year olds in the county of Varmland in 1995, 1996, and 1997 to the mean Swedish national scores in 12 and 19 year olds in 1996. Caries prevalence in Varmland has been steadily declining, at least from the age of 12 years, and is considerably below the national mean values. The frequency distribution of individuals in relation to different DFT values in 1997 is shown in Fig 237.

Every year, the Swedish Board of Health and Welfare collects data from every county on DFT in 12 and 19 year olds and approximal DFSs in 19 year olds. The mean values and frequency distributions of DFT in 12 year olds from 1985 to 1994 are shown in Figs 238 and 239. The mean values of DFT and approximal DFSs and frequency distributions of approximal DFSs in 19 year olds from 1985 to 1994 are shown in Figs 240, 241, and 242.

The changes from 1973 to 1993 in frequency distribution of DFS scores in 5 year olds and DFS scores in 15 year olds in the county of Jonkoping, Sweden, are shown in Figs 243 and 244.

Prevalence in adults and the elderly

There are few representative national surveys on caries prevalence in adults. However, recently, the first WHO global data on caries prevalence in 35 to 44 year olds were presented, expressed in DMFT (Fig 245).

Caries prevalence is high in most of the industrialized countries and moderate in the others (the United States, Argentina, Spain, Portugal, Eastern European countries, and Japan). The currently high prevalence in 35 to 44 year olds in the Scandinavian countries, Western Europe, Canada, Australia, and New Zealand is in sharp contrast to the low to very low (Australia and Finland) prevalence in 12 year olds (see Figs 234a and 234b). However, these are the countries in which caries prevalence has declined dramatically. In 1969, prevalence in 12 year olds was very high. The 12 year olds of 1969 have become the 40 year olds of 1997, and this explains the high caries prevalence in 35 to 44 year olds: Their caries experience has accumulated from an already very high level at the age of 12 years. On the other hand, the low or very low prevalence in 12 year olds of 1997 should be reflected in similarly low prevalence when they become the 40 year olds of 2025.

Caries prevalence is, of course, strongly correlated with age. Figure 246 shows the frequency distribution of intact, decayed or filled, and missing teeth in randomized samples of 35, 50, 65, and 75 year olds in the county of Varmland, Sweden. Among the 65 and 75 year olds, about 75% of the MT component is estimated to be due to caries and 25% to periodontal diseases. Figure 247 shows the prevalence of decayed and filled surfaces from the age of 3 to 80 years in the county of Jonkoping in 1973, 1983, and 1993. The number of DFSs increases up to the age of 50 years and decreases thereafter, because of increasing numbers of missing surfaces. From 1973 to 1993, the DFS score has decreased from the ages of 3 to 40 years but has increased from the ages of 50 to 80 years, because there are considerably more remaining teeth in 1993 than there were in 1973. Because of well-organized dental care, the decayed surfaces component is almost negligible.

For comparison, Fig 248 provides data from Kenya and China, two developing countries with limited resources for dental care. The figure shows the proportion of decayed and filled teeth with coronal carious lesions in enamel, dentin, and pulp and root caries lesions in different age groups. Although a very low proportion of decayed teeth are restored, the mean DFT score is also low compared to that found in industrialized countries, eg, Australia (Fig 249).

Key-risk teeth and surfaces

The pattern of dental caries in the dentition, reflected in missing teeth and decayed, missing, and filled surfaces, is generally as unevenly distributed as caries prevalence is among individuals (see chapter 4). The factors determining future tooth loss are related to age, dental caries, periodontal diseases, iatrogenic root fractures, trauma, and orthodontic therapy, among other factors. Therefore, it may be argued that it is difficult to analyze the true reasons for tooth loss in the adult, particularly in the elderly. The reasons may vary, not only in different age groups, but also in different populations and countries, depending on differences in prevalence of dental caries and periodontal diseases as well as the availability of resources for dental care.

In Sweden, for example, almost all the teeth lost up to the age of 35 years are premolars, extracted for orthodontic indications. In 50 year olds, it is estimated that about 80% of the tooth loss was the result of caries, directly or indirectly (endodontic complications, apical periodontitis, or post-related root fractures). In 1948 to 1953, when these subjects were 10 to 15 year olds caries prevalence and caries incidence were very high, and school dental care was based on "drilling, filling, billing, and killing the pulp." Only about 10% of the missing teeth would have been extracted for orthodontic indications and 10% because of periodontal diseases. In 65 year olds, about 75%, 20%, and 5% of the missing teeth are estimated to have been lost because of dental caries, periodontal diseases, and orthodontic therapy, respectively. In other words, dental caries is the major reason for tooth loss in the Swedish adult population. In 65 year olds only 5% and 10% of the mandibular and maxillary first molars, respectively, and about 40% and 60% of the maxillary and mandibular incisors, respectively, remain (see Figs 128 and 129).

Quite simplistically, the risk for tooth loss could be predicted by combining measurement of the buccolingual width of the tooth crown and its distance posteriorly from the lips. The molars (key-risk teeth) are the most posterior teeth. The first molars have the widest approximal surfaces and the mandibular incisors the narrowest. In a toothbrushing population, the posterior teeth therefore require supplementary approximal cleaning and topical application of fluoride.

As mentioned earlier, depending on the age and caries prevalence of the population, there may be pronounced variations in the patterns of both lost teeth and DFSs.

In a toothbrushing population, the key-risk surfaces are the fissures of the molars and the approximal surfaces, from the mesial surface of the second molars to the distal surface of the first premolars (see Fig 130).

Figure 250, from a study of 20-year-old Finnish men (recruits) in 1970, shows the mean number of DMFSs, according to individual tooth type. The first molars were the most frequently decayed, followed by the second molars and second premolars. Forsling et al (1999) reported that, in a randomized sample of 19 year olds (1996) from four counties in Sweden, the distal surface of the mandibular right first molar was the most frequently decayed. This is probably because most people are right handed, and in right-handed people the mandibular right linguoapproximal surfaces show the greatest tendency to plaque accumulation and gingivitis. The distal surfaces of the second premolars also constituted a relatively high percentage of carious surfaces (see Fig 131).

Figure 251, from a study of an adult Chinese population, illustrates the proportion of teeth, related to toothtype and age groups, with carious lesions in the enamel, dentin, and pulp and with restorations and root caries. The molars, both maxillary and mandibular, are the most frequently carious teeth in all age groups. Among premolars and incisors, the maxillary teeth are carious more frequently than are the mandibular teeth. Of the carious teeth, the proportion of restored teeth is low.

Figure 252 shows the number and proportion of decayed, filled, and sound teeth in 88- and 92-year-old Swedish adults related to tooth type. Even among these elderly

people, the proportion of carious teeth is very low. However, very few maxillary teeth remain, and most sound teeth are found among the mandibular incisors.

Dental caries is obviously distributed in a characteristic pattern within the oral cavity (see Figs 130, 131, 132, 133, 134, 135, and 136). First, there is a certain symmetry between the arches (see Figs 251 and 252). Moreover, it is evident that there is bilaterality in the occurrence of dental caries (see Fig 250).

The clinical implication of this bilaterality is that, if a lesion is detected on a surface, the risk of a lesion on the bilateral surface is greater than on any other surface on the same side of that mouth. The mandibular molars are the most severely affected teeth in the entire dentitionslightly more than are the maxillary molarsbut the maxillary premolars and anterior teeth are otherwise generally more severely affected than the corresponding mandibular teeth.

However, it should be noted that the apparent symmetry found when population-based data are pooled does not imply that the disease occurs symmetrically in individuals. Rather, the symmetry reflects the fact that the probability of developing caries on the left side is equivalent to that on the right. The same is not, of course, the case with respect to the maxillary and mandibular arches, in which the probability of developing caries in homologous tooth types is very different.

Pits and fissures in molars are the most susceptible environments followed by the approximal surfaces of molars, premolars, and maxillary anterior teeth. Caries on the free smooth surfaces is more rare; if it occurs, it is usually regarded as being indicative of a highly caries-active individual. Lingual surfaces rarely develop carious lesions except on the mandibular molars (see Figs 130, 131, 132, 133, 134, 135, and 136). Such variations in surface susceptibility reflect variations in the intraoral environment and not any known variations in composition of the tooth surfaces. The presently known intraoral factors that most likely favor this particular pattern of caries distribution include the pattern and rate of plaque formation, the amount of plaque left undisturbed (easier access to the buccal surfaces and the anterior part of the oral cavity), openings of salivary glands in relation to composition of saliva (secreted versus mixed saliva), the velocity of salivary flow, differences in the rate of elimination of sugars from various parts of the oral cavity, and variations in plaque pH values in various parts of the oral cavity.

Fig 128 Frequency distribution of remaining teeth (third molars excluded) in 50 year olds (Federation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.)

Fig 129 Frequency distribution of remaining teeth (third molars excluded) in 65 year olds (Federation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.)

Fig 130 Changes in caries prevalence among 12 year olds living in the county of Vormland, Sweden, 1969 to 1994. (DFS) Decayed or filled surface. (From Axelsson, 1998.)

Fig 131 Mean pattern of manifest caries or restorations with or without initial caries (enamel caries) included on the posterior approximal surfaces of 19 year olds. (D) Dentin; (D1, D2) enamel

caries lesion; (D3) dentin caries; (FS) filled surface; (p) posterior; (m) mesial; (d) distal. (From Forsling et al, 1999. Reprinted with permission.) Fig 132 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) occlusally (Federation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.) Fig 133 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) mesially (Feder- ation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.) Fig 134 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing sur- faces (MSs) distally (Federation Dentaire Internationale tooth-numbering system). Fig 135 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) buccally (Federation Dentaire Internationale tooth-numbering system). (From Axelsson et al, 1988, 1990.) Fig 136 Caries prevalence in 50 year olds: Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) lingually ((Federation Dentaire Internationale tooth-numbering system). Fig 233 Worldwide caries prevalence among 12 year olds in 1969. (From WHO, 1994. Reprinted with permission.)

Fig 234a Worldwide caries prevalence among 12 year olds in 1993. (From WHO, 1994. Reprinted with permission.)

Fig 234b Worldwide caries prevalence among 12 year olds in 1997. (From WHO, 1998. Reprinted with permission.)

Fig 235 Mean level of decayed, missing, or filled teeth (DMFT) among 12 year olds in developing and industrialized countries, as well as globally, from 1980 to 1991. (From WHO, 1994. Reprinted with permission.)

Fig 236 Caries prevalence (decayed or filled teeth [DFT]) among 7 to 19 year olds in the county of Vormland, Sweden, from1995 to 1997. For comparison, the 1996 Swedish national norms for 12 and 19 year olds are given. (From Axelsson, 1998.) Fig 237 Frequency distribution of decayed or filled teeth (DFT) among 7 to 19 year olds in the county of Vormland, Sweden, in 1997. (From Axelsson, 1998.)

Fig 238 Mean numbers of decayed or filled teeth (DFT) among Swedish 12 year olds from 1985 to 1994. (From Sundberg et al, 1996. Reprinted with permission.)

Fig 239 Frequency distribution of decayed or filled teeth (DFT) among Swedish 12 year olds in 1985, 1990, and 1994. (From Sundberg et al, 1996. Reprinted with permission.)

Fig 240 Mean numbers of decayed or filled teeth (DFT) among Swedish 19 year olds from 1985 to 1994. (From Sundberg et al, 1996. Reprinted with permission.)

Fig 241 Mean numbers of approximal decayed or filled surfaces (DFSs) among Swedish 19 year olds from 1985 to 1994. (From Sundberg et al, 1996. Reprinted with permission.)

Fig 242 Frequency distribution of decayed or filled teeth (DFT) among Swedish 19 year olds in 1985, 1990, and 1994. (From Sundberg et al, 1996. Reprinted with permission.)

10 additional images not shown.Root caries

The prevalence of root caries is usually evaluated according to the Root Caries Index (RCI) described by Katz (1980), ie, the number of decayed (R-D) and filled (R-F) root surfaces with gingival recession in proportion to the number of decayed, filled, and sound (R-N) root surfaces with recession:

(R-D) + (R-F)______________________ x 100 = RCI(R-D) + (R-F) + (R-N)

A decayed or filled root surface is registered accordingly, irrespective of visible gingival recession. The denominator includes all exposed root surfaces. In this index, root surfaces with a prosthetic crown restoration are counted as sound unless they are also affected by caries or restorations. However, the reason for restored root surfaces should be evaluated. In a recent study, Walls et al (2000) found that 55% of the restorations on buccal root surfaces were due to cervical wear.

Figure 253 shows the frequency distribution of root caries prevalence according to the RCI, in patients referred for periodontal treatment (Ravald and Birkhed, 1991). Fure and Zickert (1990b) examined the prevalence of root caries in 55, 65, and 75 year olds. Figure 254 shows the frequency distribution of carious root surfaces in each age group. The distribution of mean RCI values, according to tooth group and type of tooth surface in each jaw and age group, is shown in Fig 255. In all age groups, the molars had the highest RCI, followed by the premolars (particularly the buccal surfaces of the mandibular teeth).

Three large national surveys, recording only carious cavities and/or restorations confined to root surfaces, gave similar results: in Denmark (Table 20), Finland (Table 21), and the United States (Table 22). The prevalence of cavitated carious lesions of

the root is rather low up to about 60 years of age, affecting about one third or fewer of the population. Among people older than 60 years, prevalence rates range from about 40% to 63%, which is in accordance with other studies using random samples, although it is striking that the rates did not exceed 33% in Finland, even in the oldest cohorts. On the other hand, the percentage of edentulousness among elderly people in Finland is very high.

Vehkalahti and Paunio (1994) analyzed the occurrence of root caries in relation to periodontal status in a representative sample of Finnish adults (N = 4, 777). Only primary root caries was considered, and secondary caries and restored surfaces were excluded. Subjects with a healthy periodontium seldom (4%) had caries on the root surfaces; in contrast, 15% with gingival inflammation had root caries and 17% with pocketing had root caries. There was a strong association between subgingival plaque retention and root caries (Figs 256 and 257). Because the study population was representative of Finnish adults, the results can be generalized with reasonable confidence.

Recently it was proposed that active and inactive root lesions be differentiated. Fejerskov et al (1991) reported the prevalence of active and inactive root surface caries lesions in a selected group of 60- to 80-year-old Danes (N = 90). The researchers recorded 1,092 root surface lesions. Of these, 156 were diagnosed as active lesions and 509 as inactive. An additional 427 were filled; ie, only 25% to 30% of the detected root lesions needed to be "inactivated" by improved prevention (Figs 258, 259, 260, and 261).

Fig 243 Frequency distribution of decayed or filled surface (DFS) scores among 5 year olds in the county of Jonkoping, Sweden, from 1973 to 1993. (From Hugosson et al, 1999. Reprinted with permission.) Fig 244 Frequency distribution of decayed or filled surfaces (DFSs) among 15 year olds in the county of Jonkoping, Sweden, from 1973 to 1993. (From Hugosson et al, 1999. Reprinted with permission.) Fig 245 Worldwide caries prevalence among 35 to 44 year olds in 1997. (From WHO, 1998. Reprinted with permission.)

Fig 246 Frequency distribution of intact teeth, decayed or filled teeth (DFT), and missing teeth (MT) among adults in the county of Vormland, Sweden. (From Axelsson et al, 1990. Reprinted with permission.) Fig 247 Prevalence of decayed and filled surfaces among various age groups in the county of Jonkoping, Sweden, from 1973 to 1993. The number of filled surfaces among adults aged 50 years and older has increased over the years, because older adults now retain greater numbers of teeth. Separation of the numbers of decayed and filled surfaces reveals that, because of well-organized dental care, the decayed surfaces component is negligible. (From Hugosson et al, 1995b. Reprinted with permission.) Fig 248 Frequency distribution of decayed or filled teeth (DFT), with carious lesions involving the enamel, dentin, pulp, and root, among

adults in Kenya and China. (From Fejerskov et al, 1994. Reprinted with permission.)

Fig 249 Frequency distribution of decayed, missing, or filled teeth among Australians. (From Spencer et al, 1994. Reprinted with permission.)

Fig 250 Mean numbers of decayed, missing, or filled (DMF) surfaces among 20-year-old Finnish men, by tooth type. (I) Incisor; (P) premolar; (M) molar. (From Ainamo, 1970. Reprinted with permission.) Fig 251 Frequency distribution of carious lesions in enamel, dentin, pulp, and root, as well as restorations according to various tooth and age groups, among Chinese adults. (From Luan et al, 1989. Reprinted with permission.) Fig 252 Frequency distribution of decayed, filled, and sound teeth among 88- and 92-year-old Swedish adults, by tooth type and arch. (From Lundgren, 1997. Reprinted with permission.)

Fig 253 Frequency distribution of decayed and filled root surfaces (DFS %), according to the Root Caries Index, among patients referred for periodontal treatment. (From Ravald and Birkhed, 1991. Reprinted with permission.)

Fig 254 Frequency distribution of carious root surfaces (primary and secondary) among 55, 65, and 75 year olds. (From Fure and Zickert, 1990. Reprinted with permission.)

Fig 255 Frequency distribution of mean Root Caries Index values, by type of tooth and tooth surface, among 55, 65, and 75 year olds. (From Fure and Zickert, 1990. Reprinted with permission.)

Fig 256 Percentage of Finnish adults with root caries, by periodontal conditions and age. (From Vehkalahti and Paunio, 1994. Reprinted with permission.)

Fig 257 Percentage of Finnish adults with root caries, by periodontal conditions and number of remaining teeth. (From Vehkalahti and Paunio, 1994. Reprinted with permission.)

Fig 258 Frequency distribution of root surfaces with active caries, inactive caries, or fillings among 60- to 80-year-old Danes. (From Fejerskov et al, 1991. Reprinted with permission.)

Fig 259 Frequency distribution of facial root surfaces with active caries, inactive caries, or fillings among 60- to 80-year-old Danes. (From Fejerskov et al, 1991. Reprinted with permission.)

Fig 260 Frequency distribution of proximal root surfaces with active caries, inactive caries, or fillings among 60- to 80-year-old Danes. (From Fejerskov et al, 1991. Reprinted with permission.)

Fig 261 Frequency distribution of lingual root surfaces with active caries, inactive caries, or fillings among 60- to 80-year-old Danes. (From Fejerskov et al, 1991. Reprinted with permission.)

Incidence of Caries

Coronal caries

Caries incidence is usually expressed as the number of new decayed surfaces per year in an individual, group, or population. Synonymous terms for caries incidence are caries increment, caries attack rate, caries progression, and carious activity. In children and young adults with only primary caries, prevalence (experience) thus represents the accumulated sum of caries incidence since the time of tooth eruption. However, in adults and particularly in the elderly with multiple restorations, new secondary lesions predominate.

In most longitudinal epidemiologic and clinical studies, the threshold for new DSs is at the D3 level. In epidemiologic surveys, this generally comprises cavitated lesions of dentin, but in controlled clinical studies D3 includes even noncavitated approximal and occlusal lesions, detected on bitewing radiographs. In well-controlled experimental clinical studies, new enamel lesions (D1 and D2) are also included.

Caries incidence is most frequently presented as the mean number of new DSs per individual per year and the frequency distribution of individuals with given numbers of new DSs. Because national epidemiologic surveys on caries incidence are lacking, it is impossible to compare caries incidence, for example, in 12 year olds in different countries, as is done for caries prevalence.

Caries incidence in children and young adults

As early as 1979, a computer-aided epidemiologic program was introduced in the county of Varmland, Sweden, to evaluate the effect of needs-related preventive programs in the Public Dental Health Service. Caries prevalence and incidence are diagnosed once a year in almost 100% of all 3 to 19 year olds in the county. Between 1979 and 1996, in almost every age group among 7 to 19 year olds, a reduction in caries incidence of about 80% was achieved, which accounts for the considerable reduction in caries prevalence during the same period.

Caries incidence in adults and the elderly

There are few data on caries incidence from controlled longitudinal clinical studies in adults and the elderly, because most new lesions in adults and the elderly with multiple restorations, particularly in industrialized countries, are secondary caries on the crown and root caries. In such populations, longitudinal epidemiologic surveys for evaluation of caries incidence could not be conducted without detailed information about when and why the teeth were restored.

During a 6-year period, caries incidence was evaluated in a test group of adults (N = 375), who visited a dental hygienist for education in self-care and needs-related professional mechanical toothcleaning four to six times per year, and in a control group (N = 180), who received regular dental care once a year (Axelsson and Lindhe, 1981). The subjects in the control group developed an average of 14.0 new DSs (90% secondary caries) over the 6 years, while the corresponding figure for subjects in the test group was only 0.2.

For ethical reasons, after the 6-year period, the control group was also offered needs-related preventive dentistry, and many of them accepted. For the following 9 years, all test subjects received an individualized secondary preventive program, supervised by the same dental hygienist. To maximize the cost effectiveness, the intervals as well as the preventive measures were based strictly on individual needs: About 65% visited the hygienist only once a year, 30% twice a year, and 5% (the high-risk individuals) three to six times a year.

Figure 262 shows the mean values of new primary, secondary, and all carious lesions per subject per 15 years, by age group. About 90% of the new lesions were secondary caries. However, the caries-preventive effect was as high in the oldest age group as in the youngest. Figure 263 shows the frequency distribution of all new primary and secondary lesions among the test subjects for the entire 15-year period. Of 317 subjects, 165 developed no new lesions at all, and only two subjects developed more than 10 (about 1 DS per year) (Axelsson et al, 1991).

Fig 262 Incidence (per 15 years) of primary and secondary (recurrent) caries, by age group: group 1 (36 to 50 year olds); group 2 (51 to 65 year olds); group 3 (66 to 85 year olds). (DSs) Decayed surfaces. (From Axelsson et al, 1991. Reprinted with permission.) Fig 263 Frequency distribution of primary and secondary decayed surfaces (DSs) over a 15-year period. (From Axelsson et al, 1991. Reprinted with permission.)

Root caries

In elderly people with many exposed root surfaces, and impaired salivary function as a side effect of medication, new root lesions constitute a considerable proportion of total caries incidence.

Incidence of root caries is usually expressed as new percentage of DFSs. This is calculated as the new DF root surfaces, over a stipulated period, in relation to the number of exposed root surfaces. As proposed by Ravald (1992), it can also be divided into newly decayed active lesion surfaces, newly decayed inactive lesion surfaces, and newly filled surfaces.

A 36-month incidence study of randomly selected elderly Americans revealed a mean annual root caries attack rate of 2.2 per 100.0 susceptible root surfaces in subjects aged 80 years and older. This was significantly higher than the rate in 70 to 74 year olds; 1.4 per 100.0 (Hand et al, 1988). In another incidence study from Boston, a mean annualized attack rate of 2.3 decayed and filled root surfaces was found for the 45- to 59-year age group, while the corresponding figure for those aged 70 and older was 3.5 (Joshi et al, 1993).

Ravald and Birkhed (1992) evaluated root caries incidence in 99 subjects involved in a fluoride program for 2 years after periodontal treatment (Fig 264). The distal surfaces of the maxillary incisors, canines, and premolars, and the distobuccal surfaces of the mandibular molars were more frequently decayed.

Caries incidence over 5 years was recently studied by Fure (1997) in a random sample of 60, 70, and 80 year olds in Gothenburg, Sweden. One aim of the study was to introduce a root caries index (DMFRS%) in which the missing root surfaces are taken into account. This study shows that dental caries was the main reason for tooth extraction. The study also revealed that coronal and root caries occurs more frequently in elderly than in younger people and that the incidence of root caries is positively correlated with coronal caries and negatively correlated with the number of remaining teeth. The 5-year DMFRS% increment values increased with advancing age, from 2.7 involved root surfaces per 100.0 susceptible surfaces in 60 year olds, to 4.8 in 70 year olds, to 10.7 in 80 year olds.

Figure 265 shows the mean percentage of coronal and exposed root surfaces affected by carious lesions and restorations during the 5-year period. The distribution of carious lesions and fillings in relation to exposed root surfaces during the 5-year period is shown in Fig 266. The frequent utilization of dental care among the participants was reflected in the finding of several new restorations and crowns. Most

of the very elderly people with carious lesions at baseline, however, had developed new lesions at the margins of the recent restorations.

Fig 264 Frequency distribution of newly decayed surfaces with inactive root lesions (DiSs), newly decayed surfaces with active root lesions (DaSs), and filled root surfaces (Fss), by tooth type and surface, among patients in a post-periodontal therapy fluoride program. (From Ravald and Birkhed, 1992. Reprinted with permission.) Fig 265 Frequency distribution of carious lesions and restorations among 60-, 70-, and 80-year-old Swedish adults over a 5-year period. (From Fure, 1997. Reprinted with permission.)

Fig 266 Frequency distribution of carious lesions and restorations, by tooth type and surface, among 60-, 70-, and 80-year-old Swedish adults over a 5-year period. (From Fure, 1997. Reprinted with permission.)

Caries Treatment Needs

For oral health planners, it is important to evaluate not only changes in caries prevalence among the population but also current treatment needs, so that resources can be adapted to "true" caries treatment need. In 1988, a computer-aided analytic oral epidemiologic system was designed and tested in a randomized sample of 35, 50, 65, and 75 year olds in the county of Varmland (Axelsson et al, 1990). Among other things, a new Community Caries Index of Treatment Needs (CCITN), analogous to the Community Periodontal Index of Treatment Needs, was designed for estimating caries treatment needs.

The rationale underlying the CCITN is that estimation of treatment need should encompass more than the restorative need: Emphasis should be on prevention. Active carious lesions in enamel should be arrested. The studies by Bille and Thylstrup (1982), Pitts and Rimmer (1992), and Lussi (1991), described in chapter 5, showed that very few carious lesions on approximal surfaces or in fissures exhibit cavitation into the dentin. Most such noncavitated lesions can be arrested. In addition, Nyvad and Fejerskov (1986) showed that, in response to improved oral hygiene, active root lesions also can be successfully converted to inactive lesions. These findings further emphasize the importance of "prevention instead of extension" or "prevention before extension," in contrast to G. V. Black's traditional concept of "extension for prevention."

Table 23 shows diagnosis and treatment needs at different levels. Restoration of lesions that are not actively and progressively penetrating the enamel should be classified as malpractice. Most clinicians consider radiographic evidence of dentinal involvement to be an appropriate threshold for operative intervention. However, as long as there is no cavitation of dentin, there will be no microbial invasion of the dentinal tubules, and the lesion can still be arrested. Therefore, restoration of noncavitated dentin lesions should also be regarded as malpractice (see chapter 5).

Of course caries treatment needs may vary considerably in different populations and among age groups and individuals. However, even in 1988, in an adult population,

caries treatment needs disclosed by analytic epidemiologic data were low (Axelsson et al, 1988, 1990). For example, Fig 267 shows the frequency distribution of all manifest carious lesions (CCITN scores 2:1, 2:2, 3:1, 3:2, 4:2, 5:1, and 5:2) in a randomized sample of 50 year olds (N = 448) in the county of Varmland, Sweden. Subjects were stratified as attending Public Dental Health Service (40%) or a private practice (55%), or as nonpatients (5%). No dentin caries, recurrent caries, or root caries was recorded in almost 50% of the patients receiving treatment through the Public Dental Health Service and patients attending a private practice. Those with four or more carious lesions were mainly nonpatients (Axelsson et al, 1990). Cavitated lesions are almost nonexistent in the permanent teeth of children and young adults in the county of Varmland.

Figures 268 and 269 show the mean index of treatment need and treatment done (restored), in relation to social class of the family in samples of 5-year-old and 12-year-old children, respectively, from eight different European countries. In developing countries, where dental care resources are very limited, caries treatment needs are enormous.

Recently, Pitts (1997) used the iceberg model to illustrate caries treatment needs in relation to progressive, cavitated, and stable, noncavitated coronal lesions, at the clinically detectable or subclinical level (Fig 270).

Fig 267 Frequency distribution of all manifest carious lesions, by type of dental care, among a randomized sample of 50 year olds in the county of Vormland, Sweden. (From Axelsson et al, 1988, 1990. Reprinted with permission.)

Fig 268 Mean index of treatment need (TNEED) and treatment done (TDONE) among 5 year olds in eight European countries, in relation to the social class of the family (high [H] or low [L]). (From Bolin et al, 1997. Reprinted with permission.)

Fig 269 Mean index of treatment need (TNEED) and treatment done (TDONE) among 12 year olds in eight European countries, in relation to the social class of the family (high [H] or low [L]). (From

Bolin et al, 1997. Reprinted with permission.)

Fig 270 Caries treatment needs, represented by an iceberg. (FOTI) Fiber-optic transillumination; (BWs) bitewing radiographs. (From Pitts, 1997. Reprinted with permission.)

Reasons for Changes in Caries Prevalence

Caries prevalence in 12 year olds in several industrialized countries has declined, from very high levels in 1969 to low levels in 1997. The trend in many developing countries in Asia and Africa has been an increase in caries prevalence, from very low levels in 1969 to low or moderate levels in 1997.

During these 20 years of declining caries prevalence in industrialized countries, several measures to control carious disease have been introduced, and numerous studies have evaluated the impact of one or a combination of these measures. Because the background is complex, and so many factors may have been involved, directly and indirectly, it is difficult to obtain a complete overview. In fact, no single experimental study has addressed the issue of the relative impact of all possible factors, and it is unlikely that such a study could ever be conducted. Furthermore, even if it were possible to describe conditions in a specific population, these may not apply to other populations living under different conditions. Nevertheless, as a foundation for planning future preventive strategies, the major factors should be ranked in order of importance. Guidelines for effective disease control would also be valuable for those developing countries facing a potential increase in caries prevalence following industrialization and improved economy.

In a recent questionnaire, Bratthall et al (1996) asked the following question of 55 internationally acknowledged authorities on caries research: "What are the main reasons 20 to 25 year olds in westernized industrialized countries have less caries nowadays, compared to 30 years ago?" Each expert was asked to think specifically of a country or area and to specify whether the chosen area had water fluoridation. Figure 271 shows the countries and areas represented by the 55 experts.

The questionnaire contained 25 specific items, grouped under the following main

headings: diet, fluorides, plaque, saliva, dentists/dental materials, and other factors. Each respondent was instructed to indicate the importance of the role of each item in the decline in caries prevalence, according to the following scale:

1. Very important: More than 40% of the caries reduction can be attributed to the item in question.

2. Important: Twenty-one percent to 40% of the caries reduction can be attributed to the item in question.

3. Less important: Five percent to 20% of the caries reduction can be attributed to the item in question.

4. Minor importance: Less than 5% of the caries reduction can be attributed to the item in question.

5. No importance at all: None (0%) of the caries reduction can be attributed to the item in question.

In addition, the respondents were asked to indicate the item that, in their view, was the single most important factor contributing to the reduction in caries.

The factor that most respondents considered to be very important, ie, accounting for more than 40% of the total caries reduction, was the use of fluoride toothpaste (Fig 272). However, some respondents considered it to be less important (accounting for 5% to 20% of the caries reduction). No one indicated that it was of zero importance.

The importance attributed to improved oral hygiene (excluding possible effects of fluoride) ranged over the whole scale. A less important or minor role was attributed to dietary changes, including possible changes in total sugar consumption, frequency of sugar consumption, or use of sugar substitutes. Similar results were also found for most measures carried out by oral health personnel.

The decline in caries in 20 to 25 year olds in industrialized countries is most probably attributable to the synergistic effects of improved plaque control (mainly by self-care), topical use of fluorides (mainly fluoride toothpaste), and well-organized, school-based preventive programs.

The following evidence is presented in support of this argument:

1. Caries prevalence declined in Scandinavian countries, Switzerland, and Australia, from very high levels in 1969 to low levels in 1993, but remained high in Germany. For the last 20 to 30 years, nearly all children and young adults in these countries have been using fluoride toothpaste daily. With the exception of Germany, these countries have also established comprehensive preventive programs, through the school systems or the public dental health services. Such programs normally include education in self-care and needs-related professional preventive measures.

2. In the county of Varmland, Sweden, caries prevalence in 12 year olds declined from 25 DFSs in 1974 (the highest in Sweden) to less than 1 DFS in 1994 (the lowest

in Sweden), despite the fact that for the past 20 to 30 years, almost all Swedish children have not only used fluoride toothpaste daily, but also participated in a nationwide, school-based fluoride rinsing program. From 1979 to 1999, the caries prevalence in 19 year olds in the county declined from 22 to 2 DFSs, the lowest in Sweden.

3. Fluoride toothpaste is by far the most frequently used fluoride agent in the world, but it is used regularly by fewer than 10% of the world's population, ie, about a third of the population of the industrialized countries.

4. The relative effect of toothbrushing and fluoride toothpaste has never been properly evaluated in a well-controlled 3-year longitudinal study. Such a study would have to be conducted in a nontoothbrushing population of 12 year olds with high caries incidence and prevalence and the highest number of newly erupted permanent tooth surfaces at risk. Two test groups and a true-negative control group should be selected.

The following study design should be used:

Test group 1: Once a day, at school, a dental assistant would brush the subjects' teeth according to the Bass method, using a placebo toothpaste. Quality control would be achieved by plaque disclosure after regular brushing.

Test group 2: Same as test group 1, but with fluoride toothpaste.

Control group: True-negative control group without any intervention.

Comparison of test group 1 and the control group, would indicate the caries-preventive effect of toothbrushing on the surfaces readily accessible to the toothbrush, ie, all surfaces of the anterior teeth and the buccal and lingual surfaces of the posterior teeth. Comparison of these tooth surfaces in test groups 1 and 2 would indicate the additional caries-preventive effect of fluoride toothpaste.

Unfortunately, it is probably too late to run such a study, because toothbrushing with fluoride toothpaste is such a widespread, well-established caries-preventive measure that it could not ethically be withheld from caries-susceptible subjects for such a long experimental period.

5. Finally, the daily intake of sugar in Sweden has remained persistently high (about 120 g/d) since 1960, and the proportion of indirect sugar consumption (in sweets, cakes, and drinks) has increased. The dramatic decline in caries over the last 30 years cannot, therefore, be attributed to a reduction in sugar consumption.

6. The increasing prevalence in, for example, China and India, is attributable to a combination of poor oral hygiene (plaque control), limited resources for dental care, and changes in lifestyle and dietary habits following industrialization and improvement in the economy. The high prevalence of periodontal diseases and high level of treatment needs in these two countries confirm a low standard of oral hygiene and limited dental care resources (about one dentist per 100,000 inhabitants and no dental hygienists). However, the daily use of fluoride toothpaste is steadily increasing.

Fig 271 Countries represented by 55 caries research experts who were questioned about reasons for the current decline in caries prevalence in westernized countries. (From Bratthall et al, 1996. Reprinted with permission.) Fig 272 Importance of selected factors to the current decline in caries prevalence, according to surveyed experts in the field of caries research. (From Bratthall et al, 1996. Reprinted with permission.)

Conclusions

Caries prevalence

In epidemiologic surveys using WHO criteria, caries prevalence represents only decayed teeth with clinically detected cavitation in the dentin (D3) and filled or missing teeth, the so-called DMFT score. Prevalence is considerably underestimated by exclusion of noncavitated dentin lesions, which could have been detected on bitewing radiographs, and enamel lesions. In countries with well-organized dental care systems, the M component is often excluded in children and young adults, because permanent teeth are extracted almost exclusively for orthodontic indications and not because of caries.

The WHO recommends regular national epidemiologic surveys for collection of caries prevalence data (DMFT) in 12 year olds. On the first WHO global map in 1969, caries prevalence was very high in most industrialized countries, and very low in developing countries, except for the Latin American countries. The most recent WHO global map, from 1997, shows that caries prevalence has decreased to low and even very low levels in the industrialized countries, except for Germany, Japan, Poland, and the Baltic countries. In most developing countries, caries prevalence has increased from very low to low levels; prevalence remains high in most Latin American countries and the Philippines.

The WHO also recommends collection of national survey data on coronal caries prevalence in 35 to 44 year olds. Recent data from the WHO show that, in 1997, prevalence was high in 35 to 44 year olds in many industrialized countries. The high caries prevalence in 35 to 44 year olds in Scandinavia, Canada, Australia, and New Zealand has its origins in the very high caries prevalence recorded in 12 year olds in these countries in 1969: The 12 year olds of 1969 have become the 40 year olds of 1997. On the other hand, the low or very low caries prevalence in the 12 year olds in 1997 should be reflected in a similarly low prevalence in these subjects when they reach the age of 40 years in 2025.

The pattern of dental caries in the dentition, reflected in decayed, missing, and filled surfaces, is generally as unevenly distributed as caries prevalence is among individuals. Depending on the age and caries prevalence of the population, there may be pronounced variations in the pattern of both lost teeth and carious and restored surfaces. In 5 year olds, the primary second molars are the most frequently decayed, followed by the mandibular first molars and the maxillary central incisors. In the permanent dentition of toothbrushing populations of children and young adults, the key-risk surfaces are the fissures of the molars and the posterior approximal surfaces,

from the mesial surfaces of the second molars to the distal surfaces of the first premolars. This pattern reflects areas where the toothbrush has limited access for removal of cariogenic plaque.

The prevalence of root caries is usually evaluated according to the Root Caries Index : The number of decayed and filled root surfaces with gingival recession in proportion to the total number of decayed, filled, and sound root surfaces with recession. The highest RCI score is found in the molars; this is followed by the premolars, particularly the buccal surfaces of the mandibular teeth. Root caries lesions should also be differentiated as active or inactive.

Caries incidence

Coronal caries incidence is usually expressed as the number of new decayed surfaces per year in an individual, group, or population. Synonymous terms are caries increment, caries attack rate, caries progression, and carious activity. In most longitudinal, epidemiologic, and clinical studies, the threshold for new DSs is at the D3 level. In epidemiologic surveys, this generally includes only new cavitated lesions. In controlled clinical studies, noncavitated approximal and occlusal lesions detected on bitewing radiographs are usually included. Almost all new carious surfaces in children and young adults are primary lesions, but in industrialized countries, secondary (recurrent) caries predominates in persons older than 40 years, because of the high prevalence of restored surfaces.

Root caries incidence is usually expressed as the number of new decayed or filled root surfaces, over a specified time, in relation to the number of exposed root surfaces. It can also be stratified as newly decayed active or inactive lesions.

Treatment needs

Enamel lesions and noncavitated lesions of dentin can be arrested, and active noncavitated and cavitated root lesions can be converted to inactive lesions. A Community Caries Index of Treatment Needs has been proposed; the index emphasizes prevention rather than operative intervention.

The proportion of cavitated carious lesions that need restoration differs considerably among different populations, depending not only on caries prevalence but also on availability of resources. In Sweden, for example, open cavitated lesions are very rare in adults as well as children, despite relatively high caries prevalence among adults and the elderly, because almost 90% of the adults follow a regular maintenance program. Almost all children and young adults are enrolled in well-organized preventive programs from the age of 1 year to 20 years: Caries prevalence is therefore very low, and the need for restorative treatment is negligible. In many developing countries, on the other hand, there are almost no restorations, but many untreated cavitated lesions.

Caries decline

A recent questionnaire sent to 55 internationally acknowledged caries researchers disclosed that, in their opinion, regular use of fluoride toothpaste and improved oral

hygiene are the main reasons for the significant decline in caries prevalence among 20 to 25 year olds in the westernized industrialized countries over the past 30 years. The most significant reduction has been achieved in the Scandinavian countries, Switzerland, and Australia, probably as a synergistic effect of the use of fluoride (particularly fluoride toothpaste), improved plaque control (particularly by self-care), and well-organized preventive programs for children and young adults in school-based or public dental health services (for reviews on caries epidemiology, see Bratthall et al, 1996; Burt, 1997; Manji and Fejerskov, 1994; Petersson and Bratthall, 1996; and Spencer, 1997).

References

Abelson DC, Barton JE, Maietti GM, Cowherd MG (1981). Evaluation of interproximal cleaning by two types of dental floss. Clin Prev Dent 3:19-21.

Adriaens P, Loesche W, de Bover, J (1986). Bacteriological study of the microbial flora invading the radicular dentin of periodontally diseased caries-free human teeth. T Lehner and G Cimasoni (eds). Borderland Between Caries and Periodontal Disease. 3rd ed. pp. 383. Geneva: Editions Medicine et hygiene.

Adriaens P, de Bover J, Loesche W (1988a). Bacterial invasion in root cementum and radicular dentine of periodontally diseased teeth in humans: a reservoir of periodontopathic bacteria. J Periodontol 59:222-230.

Adriaens P, Edwards C, de Bover J, Loesche W (1988b). Ultrastructural observations on bacterial invasion in cementum and radicular dentine of periodontally diseased human teeth. J Periodontol 55:493-503.

Ainamo J (1970). Concomitant periodontal disease and dental caries in young adult males. Proc Finn Dent Soc 66:303-366.

Alaluusua S, Myllarniemi S, Kallio M (1989). Streptococcus mutans infection level and caries in a group of 5-year-olds. Caries Res 23:190-194.

Alaluusua S, Kleemola-Kujala E, Gronros L, Evalathi M (1990). Salivary caries-related tests as predictors of future caries increment in teenagers. A 3-year longitudinal study. Oral Microbiol Immunol 5:77-81.

Alexander M (1971). Microbial Ecology. New York: John Wiley.

Antoft P, Gadegaard E, Jepsen P (1988). Caries experience, dental health behaviour and social status. A comparative study among Danish military recruits in 1972 and 1982. Comm Dent Health 5:255-264.

Artun J, Thylstrup A (1986). Clinical and scanning electron microscopic study of surface changes of incipient enamel caries lesions after debonding. Scand J Dent Res 94:193-210.

Artun J, Thylstrup A (1989). A 3-year clinical and SEM study of surface changes of carious enamel lesions after inactivation. Am J Dentofac Orthop 95:27-33.

Asher R, Kuster C, Moyer I, Boyd D (1986). Parents' carbohydrate intake and the oral health of their child. J Pedod 10:156-163.

Axelsson P, Lindhe J (1987). Efficacy of mouthrinses in inhibiting dental plaque and gingivitis in man. J Clin Periodontol 14:205-212.

Axelsson P, Paulander J, Nordqvist K, Karlsson R (1987a). The effect of fluoride containing dentifrice, rinsing and varnish on interproximal dental cariesa 3-year clinical trial. Comm Dent Oral Epidemiol 15:177-180.

Axelsson P, Kristoffersson K, Karlsson R, Bratthall D (1987b). A 30-month longitudinal study of the effects of some oral hygiene measures on Streptococcus mutans and approximal dental caries. J Dent Res 66:761-765.

Axelsson P (1988). The Community Caries Index of Treatment Needs (CCITN).

Axelsson P (1989). Placknybildningsindex PFRI - Indikator for karies- och parodontitprevention, munhygien-frekvens och ytrelaterad munhygien. Tandlak tidn 79:7:387-391.

Axelsson P, Paulander J, Tollskog G (1990). A new computer-based oral epidemiology system. The 2nd International Conference on Preventive Dentistry and Epidemiology, Karlstad, Sweden, 1990 [manuscript].

Axelsson P (1991). A four-point scale for selection of caries risk patients, based on salivary S. mutans levels and plaque formation rate index. In: Johnson N (ed). Risk Markers for Oral Diseases. Vol 1. Dental Caries. Cambridge: Cambridge University Press.

Axelsson P, Lindhe J, Nystrom B (1991). On the prevention of caries and periodontal disease. Results of a 15-year longitudinal study in adults. J Clin Periodontol 13:182-189.

Axelsson P, Paulander J (1994). The oral health status in 50-55-year-olds in the county of Varmland [manuscript].

Axelsson P (1994). Mechanical plaque control. In: Lang N, Karring T (eds). Proceedings of the 1st European Workshop on Periodontology. Chicago: Quintessence.

Axelsson P, Buischi Y, Barbosa M, Karlsson R, Pradi M (1994). The effect of a new oral hygiene training program on approximal caries in 12- to 15-year-old Brazilian children: Results after 3 years. Adv Dent Res 8:278-284.

Axelsson P (1997). Plackbildungsrateindex bei 3- bis 19-jahrigen. Phillip J 7/8:237-239.

Axelsson P, Paulander J, Hontwedt M, Ostlund L, Engstrom A (1997a). The effect of F-chewing gum on salivary secretion rate, plaque (PI), plaque formation rate (PFRI), salivary mutans streptococci (MS) and oral mucosa in subjects with reduced salivary secretion ratea 6-month longitudinal study [abstract]. Presented at the 5th World Congress on Preventive Dentistry, Cape Town, South Africa.

Axelsson P (1998). Needs-related plaque control measures based on risk prediction. In: Lang PN, Attstrom, Loe H (eds). Proceedings of the European Workshop on Mechanical Plaque Control. Chicago: Quintessence.

Axelsson P, Paulander J, Lindhe J (1998). Relationship between smoking and dental status in 35-, 50- 65-, and 75-year-old individuals. J Clin Periodontol 25; 297-305.

Axelsson P, Struzycka I, Wojcieszek D, Wierzbicka M (2000a). Prediction of caries risk based on salivary mutans streptococci (MS) levels and plaque formation rate index (PFRI) [manuscript].

Axelsson P, El Tabakk S (2000b). Caries incidence in Egyptian schoolchildren related to prevalence of fluorosis, salivary mutans streptococci levels and dietary habits [abstract]. Dept. of Periodontology, Goteborg University, Sweden.

Axelsson P, El Tabakk S (2000c). Caries prevalence in 12-year-old Egyptian schoolchildren related to prevalence of fluorosis, salivary mutans streptococci levels and dietary habits [abstract]. Dept. of Periodontology, Goteborg University, Sweden.

Backer-Dirks O (1966). Posteruptive changes in dental enamel. J Dent Res 45:503-511.

Beal J (1989). Social factors and preventive dentistry. In: Murray J (ed). The Prevention of Dental Disease. London: Oxford University Press.

Bedi R (1989). Ethnic indicators of dental health for young Asian schoolchildren resident in areas of multiple deprivation. Br Dent J 166:331-334.

Beighton D, Manji F, Baelum V, Fejerskov O, Johnson NW, Wilton J (1989). Associations between salivary levels of Streptococcus sobrinus, lactobacilli and caries experience in Kenyan adolescents. J Dent Res 68:1242-1246.

Berkowitz R, Jordan H (1975). Similarity of bacteriocins of Streptococcus mutans from between mother and infant. Arch Oral Biol 20:725-730.

Berkowitz R, Jones P (1985). Mouth-to-mouth transmission of the bacterium Streptococcus mutans between mother and child. Arch Oral Biol 30:377-379.

Bevenius J, Linder L, Hultenby K (1994). Site-related streptococcal attachment to buccocervical tooth surfaces. A correlative micromorphologic study. Acta Odontol Scand 52:294-302.

Bille J, Thylstrup A (1982). Radiographic diagnosis and clinical tissue changes in

relation to treatment of approximal carious lesions. Caries Res 16:1-6.

Bille J, Carstens K (1989). Approximal caries progression in 13- to 15-year-old Danish children. Acta Odontol Scand 47:347-354.

Billings R (1986). Restoration of carious lesions of the root. Gerodont 5:43-49.

Birkhed D (1990). Behavioural aspects of dietary habits and dental caries. Caries Res 24:27-35.

Bjarnason S (1996). Temporary tooth separation in the treatment of approximal carious lesions. Quintessence Int 27:4.

Bjarnason S, Kohler B (1997). Caries risk assessment in adolescents. Swed Dent J 21:41-48.

Bjorndal L (1991). Carieslaesionens tidlige udvikling i emalje og pulpa-dentinorganet [dissertation]. Copenhagen: University of Copenhagen.

Black G (1908). Operative Dentistry. Chicago: Medico-Dental Publishing.

Blinkhorn A (1982). The caries experience and dietary habits of Edinburgh nursery schoolchildren. Br Dent J 152:227-230.

Bolin A, Bolin A, Jansson L, Calltorp J (1997). Children's dental health in Europe. Swed Dent J 21:25-40.

Borgstrom M, Sullivan A, Granath L, Nilsson G (1997). On the pH lowering potential of lactobacilli and mutans streptococci from dental plaque related to the prevalence of caries. Comm Dent Oral Epidemiol 25:165-169.

Bowden G, Edwardsson S (1994). Oral ecology and dental caries. In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Cariology. Copenhagen: Munksgaard.

Bowden G (1997). Does assessment of microbial composition of plaque/saliva allow for diagnosis of disease activity of individuals? Comm Dent Oral Epidemiol 25:76-81.

Bowen W (1994). Food components and caries. Adv Dent Res 8:215-220.

Boyar R, Thylstrup A, Holmen L, Bowden G (1989). The microflora associated with the development of initial enamel calcification below orthodontic bands in vivo in children living in a water fluoridated area. J Dent Res 68:1734-1738.

Bradnock G, Jadoua I, Hamburger R (1988). The dental health of indigenous and non-indigenous infant schoolchildren in West Birmingham. Comm Dent Health 5:139-150.

Bradshaw D, McKee A, Marsh P (1989a). Effects of carbohydrate pulses and pH on population shifts within oral microbial communities in vitro. J Dent Res 68:1298-1302.

Bradshaw D, McKee A, Marsh P (1989b). Effect of pH on the stability of a microbial community [abstract]. J Dent Res 68:930.

Brandtzaeg P (1989). Salivary immunoglobulins. In: Tenovou J (ed). Human Saliva: Clinical Chemistry and Microbiology. Boca Raton, FL: CRC Press.

Bratthall D (1991). The global epidemiology of mutans streptococci. In: Johnson NW (ed). Risk Markers for Oral Diseases. Vol 1. Dental Caries. Cambridge: Cambridge University Press.

Bratthall D, Ericsson D (1994). Tests for assessment of caries risk. In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Cariology. 333-353. Copenhagen: Munksgaard.

Bratthall D (1996). Dental caries: intervenedinterruptedinterpreted. Concluding remarks and cariography. Euro J Oral Sci 104:486-491.

Bratthall D, Hansel-Petersson G, Sundberg H (1996). Reasons for the caries decline: What do the experts believe? Euro J Oral Sci 104:416-422.

Brock T (1966). Principles of Microbial Ecology. Englewood Cliffs, NJ: Prentice-Hall.

Buischi Y, Axelsson P, Zulske Barbosa M, Mayer M, Carmen M, de Oliviera L (1989). Salivary S mutans and caries prevalence in Brazilian schoolchildren. Comm Dent Oral Epidemiol 17:20-30.

Burt B, Eklund S, Morgan K (1988). The effect of sugar intake and frequency of ingestion on dental caries increment in a 3-year longitudinal study. J Dent Res 67:1422-1429.

Burt B (1997). How useful are cross-sectional data from surveys of dental caries? Comm Dent Oral Epidemiol 25:36-41.

Busscher H, Cowan M, van der Mei H (1992). On the relative importance of specific and non-specific approaches to oral microbial adhesion. FEMS Microbiol Rev 88:199-210.

Carlsson J, Egelberg J (1965). Effect of diet on early plaque formation in man. Odontol Revy 16:112-125.

Carlsson J, Grahnen H, Jonsson G (1975). Lactobacilli and streptococci in the mouth of children. Caries Res 9:333-339.

Carlsson P (1988). On the epidemiology of mutans streptococci [dissertation]. Malmo: University of Lund.

Carlsson and Hamilton (1994). Metabolic activity of oral bacteria. In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Cariology. Copenhagen: Munksgaard.

Carvalho J, Ekstrand K, Thylstrup A (1989). Dental plaque and caries on occlusal surfaces of first permanent molar in relation to stage of eruption. J Dent Res 68:773-779.

Carvalho J, Ekstrand K, Thylstrup A (1991). Results of 1 year of non-operative caries treatment of erupting permanent first molars. Comm Dent Oral Epidemiol 19:23-28.

Carvalho J, Thylstrup A, Ekstrand K (1992). Results after 3 years of non-operative occlusal caries treatment of erupting first permanent molars. Comm Dent Oral Epidemiol 20:187-192.

Catalanotto F, Shklair I, Keene H (1975). Prevalence and localization of Streptococcus mutans in infants and children. J Am Dent Assoc 91:606-609.

Caufield P, Childers N, Allen D, Hansen J (1985). Distinct bacteriocins correlate with different groups of Streptococcus mutans plasmids. Infect Immun 48:51-56.

Caufield P, Childers N, Allen D, Hansen J, Ratanapridakul K, Crabb D, et al (1986). Plasmids in Streptococcus mutans: Usefulness as epidemiological markers and association with mutacins. In: Hamada S, Michalek S, Kiyono H, Menaker L, McGhee J (eds). Proceedings of an International Conference on Cellular, Molecular, and Clinical Aspects of Streptococcus mutans, September 18-20, Birmingham, AL. New York: Elsevier.

Caufield P, Ratanapridakul K, Allen D, Cutter G (1988). Plasma-containing strains of Streptococcus mutans cluster within family and racial cohorts: Implications for natural transmission. Infect Immun 56:3216-3220.

Caufield P, Walker T (1989). Genetic diversity within Streptococcus mutans evident by chromosomal DNA restriction fragment length polymorphisms. J Clin Microbiol 27:274-278.

Caufield P, Cutter P, Dasanayake A (1993). Initial acquisition of mutans streptococci by infants: Evidence for a discrete window of infectivity. J Dent Res 721:37-45.

Clerehugh V, Lennon M (1986). A 2-year longitudinal study of early periodontitis in 14- to 16-year-old schoolchildren. Comm Dent Health 3:135-141.

Crossner C (1981). Salivary lactobacillus counts in the prediction of caries activity. Comm Dent Oral Epidemiol 9:182-190.

Crossner C, Claesson R, Johansson T (1989). Presence of mutans streptococci and various types of lactobacilli in interdental spaces related to development of proximal carious lesions. Scand J Dent Res 97:307-315.

Curzon M, Spector P (1977). Enamel mottling in a high strontium area of the USA. Comm Dent Oral Epidemiol 5:243-247.

Curzon M, Spector P, Iker H (1978). An association between strontium in drinking

water supplies and low caries prevalence in man. Arch Oral Biol 23:317-321.

Curzon M, Croker D (1978). Realtionship of trace elements in human tooth enamel to dental caries. Arch Oral Biol 23:647-653.

Dahlberg A (1961). Relationship of tooth size to cusp number and groove conformation of occlusal surface patterns of lower molar teeth. J Dent Res 40:34-36.

Danielsen B, Wenzel A, Hintze H, Nyvad B (1996). Temporary tooth separation as an aid to the diagnosis of cavitation in approximal surfaces. Caries Res 30:271.

Davey A, Rogers A (1984). Multiple types of the bacterium Streptococcus mutans in the human mouth and their intra-family transmission. Arch Oral Biol 29:453-460.

Dawes C, Jenkins G, Tonge C (1963). The nomenclature of the integuments of the enamel surface of teeth. Br Dent J 16:65-68.

Dawes C (1983). A mathematical model of salivary clearance of sugar from the oral cavity. Caries Res 17:321-334.

Dawes C (1987). Physiological factors affecting salivary flow rate, oral sugar clearance and the sensation of dry mouth in man. J Dent Res 66:648-653.

Dawes C, Macpherson L (1992). Effects on 9 different chewing gums and lozenges on salivary flow rate and pH. Caries Res 26:176-182.

Dawson L (1993). Oral sugar clearance and salivary buffering effects in the control of plaque pH. J Dent Res 72:691.

Dean H, Elvove E (1935). Studies on the minimal threshold of the dental sign of chronic endemic fluorosis (mottled enamel). Public Health Rep 50:1719-29.

Dennis D, Gawronski T, Sudo S, Harris R, Folke L (1975). Variations in microbial and biochemical components of four-day plaque during a four-week controlled diet period. J Dent Res 54:716-722.

De Stoppelaar JD, van Houte J, Backer-Dirks O (1970). The effect of carbohydrate restriction on the presence of Streptococcus mutans, Streptococcus sanguis, and iodophilic polysaccharide-producing bacteria in human dental plaque. Caries Res 4:114-123.

Department of Health and Social Security (DHSS) (1981). Towards Better Dental Healthguidelines for the Future. The Report of the Dental Strategy Review Group. London: HMSO.

De Vries H, Ruiken H, Konig K, Van't Hof M (1990). Radiographic versus clinical diagnosis of approximal carious lesions. Caries Res 24:364-370.

Di Renzo JM, Slots J (1990). Genetic approach to the study of epidemiology and pathogenesis of Actinobacillus actinomycetemcomitans in localized juvenile

periodontitis. Arch Oral Biol 35:79-84.

Dodds M, Hsieh S, Johnson D (1991). The effect of increased mastication by daily gum-chewing on salivary gland output and dental plaque acidogenicity. J Dent Res 70:1474-1478.

Downer M (1970). Dental caries and periodontal disease in girls of different ethnic groups. A comparison in a London secondary school. Br Dent J 128:379-385.

Edgar W, Bibby B, Mundorff S, Rowley J (1975). Acid production in plaques after eating snacks: modifying factors in foods. J Am Dent Assoc 90:418-425.

Edgar W, Rugg-Gunn A, Jenkins G, Geddes D (1978). Photographic and direct visual recording of experimental caries-like changes in human enamel. Arch Oral Biol 23:667-673.

Edgar W, Bowen W, Amsbaugh S, Monell-Torrens E (1981). Effect of eating pattern on dental caries in rodents [abstract]. Caries Res 15:179.

Edgar WM, O'Mullane DM, eds (1990). Saliva and dental health. Br Dent J 168:173-176.

Edgar WM, Higham SM (1991). Diet as a determinant of caries risk. In: Johnson N (ed). Risk Markers for Oral Diseases. Vol 1. Dental Caries. Cambridge: Cambridge Univiversity Press.

Edgar WM, Higham S, Manning R (1994). Saliva stimulation and caries prevention. Adv Dent Res 8:239-245.

Edwardsson S (1974). Bacteriological studies on deep areas of carious dentine. Odont Revy 25(suppl):32.

Ek PG, Forsberg H (1994). Hygienslipning i det primara bettet. Tandlakartidn 10:612.

Ekstrand K (1988). Strukturel undersogelse af det organiske vaev og den bakterielle plaque i okklusalfladers fure-fossa-system i relation til emaljeforandringer. En lysmikroskopisk og ultrastrukturel undersogelse foretaget pa ikke-frembrudte og delvis frembrudte visdomstaender [dissertation]. Copenhagen: Royal Dental College.

Ekstrand K, Nielsen L, Carvalho J, Thylstrup A (1993). Dental plaque and caries on permanent first molar occlusal surfaces in relation to sagittal occlusion. Scand J Dent Res 101:9-15.

Ekstrand K, Kuzmina I, Bjorndal L, Thylstrup A (1995). Relationship between external and histologic features of progressive stages of caries in the occlusal fossa. Caries Res 29:243-250.

Ekstrand K, Ricketts D, Kidd E (1997). Reproducibility and accuracy of three methods for assessment of demineralization depth on the occlusal surface: An in vitro examination. Caries Res 31:224-231.

Ellen RP, Banting DW, Fillery ED (1985). Streptococcus mutans and lactobacillus detection in the assessment of dental root surface risk. J Dent Res 64:1245-1249.

Emilson CG, Krasse B (1985). Support for and implications of the specific plaque hypothesis. Scand J Dent Res 93:96-104.

Enwonwu C (1981). Review of oral disease in Africa and the influence of socio-economic factors. Int Dent J 31:29-39.

Epstein J, Schubert M (1987). Synergistic effect of sialogogues in management of xerostomia after radiation therapy.Oral Surg Oral Med Oral Pathol 64:179-182.

Espelid I, Tveit A (1984). Radiographic diagnosis of mineral loss in approximal enamel. Caries Res 18:141-148.

Espelid I, Tveit A (1986). Clinical and radiographic assessment of approximal carious lesions. Acta Odontol Scand 44:31-37.

Espelid I, Tveit AB, Fjelltveit A (1994). Variations among dentists in radiographic detection of occlusal caries. Caries Res 28:169-175.

Fejerskov O, Manji F (1990). Risk assessment in dental caries. In: Bader J (ed). Risk Assessment in Dentistry. Chapel Hill: University of North Carolina.

Fejerskov O, Luan W, Nyvad B, Budtz-Jorgensen E, Holm-Pederson P (1991). Active and inactive root surface caries lesions in a selected group of 60- to 80-year-old Danes. Caries Res 25:385-391.

Fejerskov O, Scheie A, Manji F (1992). The effect of sucrose on plaque pH in the primary and permanent dentition of caries-inactive and active Kenyan children. J Dent Res 71:25-31.

Fejerskov O, Baelum V, Luan W, Manji F (1994). Caries prevalence in Africa and the People's Republic of China. Int Dent J 44:425-433.

Fejerskov O, Clarkson B (1996). Dynamics of caries lesion formation. In: Fejerskov O, Ekstrand J, Burt B. Fluoride in Dentistry. Copenhagen: Munksgaard.

Fejerskov O, Baelum V, Richards A (1996a). Dose-response and dental fluorosis. In: Fejerskov O, Ekstrand J, Burt B. Fluoride in Dentistry. Copenhagen: Munksgaard.

Fejerskov O, Richards A, Den Besten P (1996b). The effect of fluoride on tooth mineralization. In: Fejerskov O, Ekstrand J, Burt B. Fluoride in Dentistry. Copenhagen: Munksgaard.

Firestone A, Imfeld T, Schiffer S, Lutz F (1987). Measurement of interdental plaque pH in humans with an indwelling glass pH electrode following a sucrose rinse. A long-term retrospective study. Caries Res 21:555-558.

Fitzgerald R, Keyes P (1960). Demonstration of hte etiologic role of streptococci in experimental caries in the hamster. J Am Dent Assoc 61:9-19.

Fletcher M, Gray T, Jones J, eds (1987). Ecology of Microbial Communities. Cambridge: Cambridge University Press.

Forsling J, Halling A, Lundin S, Paulander J, Svenson B, Unell L, Wendt L (1999). Proximal caries prevalence in 19-year-olds living in Sweden. A radiographic study in 4 counties. Swed Dent J 23:59-70.

Forsman B (1965). Effect of mouthrinses with sodium fluoride in schools at Vaxjo, Sweden. Tandlak Forb Tidn 57:705-709.

Fosdick L, Campaigne E, Francher O (1941). Rate of acid formation in caries areas: the etiology of dental caries. Ill Dent J 10:85-95.

Fox P, van der Ven P, Baum B, Mandel I (1986). Pilocarpine for the treatment of xerostomia associated with salivary gland dysfunction. Oral Surg 61:243-248.

Frostell G (1969). Dental plaque pH in relation to intake of carbohydrate products. Acta Odontol Scand 27:3-29.

Fure S, Zickert I (1990). Root surface caries and associated factors. Scand J Dent Res 98:391-400.

Fure S (1997). Five-year incidence of coronal and root caries in 60-, 70- and 80-year-old Swedish individuals. Caries Res 31:249-258.

Geddes D, Cooke J, Edgar W, Jenkins G (1978). The effect of frequent sucrose mouthrinsing on the induction in vivo of caries-like changes in human dental enamel. Arch Oral Biol 23:663-665

Gahnberg L, Smith D, Taubman M, Ebersole J (1985). Salivary IgA antibody to glucosyltransferase of oral microbial origin in children. Arch Oral Biol 30:551-556.

Geddes D (1991). Methods for determining the cariogenicity of foodstuffs and their use in risk determination. In: Johnson N (ed). Risk Markers for Oral Diseases. Vol 1. Dental Caries. Cambridge: Cambridge University Press.

Geddes DA (1994). Diet patterns and caries. Adv Dent Res 8:221-224.

Gelbier S, Taylor S (1985). Some Asian communities in the UK and their culture. Br Dent J 158:416-418.

Goldberg A, Tanzer J, Munster E, Amara J, Thal F, Birkhed D (1981). Cross-sectional clinical evaluation of recurrent enamel caries, restorational marginal integrity, and oral hygiene status. J Am Dent Assoc 102:635-641.

Gonzales-Cabezas C, Li Y, Noblitt T, Gregory R, Kafrawy A, Stookey G (1995). Detection of mutans streptococci in secondary carious lesions using

immunofluorescent techniques and confocal laser scanning microscopy. Caries Res

Gibbons R, Van Houte J (1980). Bacterial adherence and the formation of dental plaques. In: Beachy E (ed). Bacterial Adherence. Receptors and Recognition. Series B, vol 6. London: Chapman.

Gibbons R (1989). Bacterial adhesion to oral tissues: A model for infectious diseases. J Dent Res 68:750-760.

Goose DH (1967). Infant feeding and caries of the incisors: An epidemiological approach. Caries Res 1:167-173.

Goose DH (1968). Infant feeding methods and dental caries. Pub Health Lond 82:72-76.

Graf H, Muhlemann H (1966). Telemetry of plaque pH from interdental area. Helv Odont Acta 19:94-101.

Grahnen H, Ingervall B (1963). Tooth width and morphology of the dentition in a group of caries resistant men. Ondont Revy 14:70-75.

Granath K, Rootzen H, Liljegren F, Holst K (1976). Variation in caries prevalence related to combination of dietary and oral hygiene habits in 6-year-olds. Caries Res 10:308-317.

Granath K, Rootzen H, Liljegren F, Holst K, Kohler L (1978). Variation in caries pevalence relaed to combinations of dietary and oral hygiene habits and chewing fluoride tablets in 4-year-old children. Caries Res 12:83-92.

Greenspan D, Daniels T (1987). Effectiveness of pilocarpine in post-radiation xerostomia. Cancer 59:1123-1125.

Grindefjord M, Dahllof G, Nilsson B, Modeer T (1995). Prediction of dental caries development in 1-year-old children. Caries Res 29:343-348.

Grondahl H, Hollender L, Malmcrona E, Sundquist B (1977). Dental caries and restorations in teenagers. I. Index and score system for radiographic studies of proximal surfaces. Swed Dent J 1:45-50.

Grondahl (1994). Radiologic diagnosis in caries management. In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Cariology. Copenhagen: Munskgaard.

Grytten J, Scheie A, Giertsen E (1988a). Synergistic antibacterial effects of copper and hexetidine against Streptococcus sobrinus and Streptococcus sanguis. Acta Odontol Scand 46:181-183.

Grytten J, Aamdal-Scheie A, Afseth J (1988b). Effect of a combination of copper and hexetidine on the acidogenicity and copper accumulation in dental plaque in vivo. Caries Res 22:371-374.

Gustafsson B, Quensel C, Lanke L, Lundqvist C, Grahnen H, Bonow B, Krasse B (1954). The Vipholm dental caries study. The effect of different levels of carbohydrate intake on caries activity in 436 individuals observed for five years. Acta Odontol Scand 11:232.

Hafstrom-Bjorkman U (1992). Laser fluorescence in the early diagnosis of enamel caries [thesis]. Kongl Carolinska Medico Chirurgiska Institutet.

Hagan T, Shaw G, Caufield P (1989). DNA fingerprinting for studying transmission of Streptococcus mutans [abstract]. J Dent Res 64:407.

Haikel Y, Frank RM, Voegel JC (1983). Scanning electron microscopy of human enamel surface layers of incipient carious lesions. Caries Res 17:1-13.

Hals E, Simonsen T (1972). Histopathology of experimental in vivo caries around silver amalgam fillings. Caries Res 6:16-33.

Hamilton I, Boyar R, Bowden G (1985). Influence of pH and fluoride on the properties of an oral strain of Lactobacillus casei grown in continuous culture. Infect Immun 48:664-670.

Hand J, Hunt R, Beck J (1988). Coronal and root caries in old Iowans: 36-month incidence. Gerodontics 4:136-139.

Harper D, Loesche W (1986). Inhibition of acid production from oral bacteria by fluorapatite-derived fluoride. J Dent Res 65:30-33.

Hausen H, Heinomen I, Paunio I (1981). Modification of occurrence of caries in children by toothbrushing and sugar exposure in fluoridated and non-flouridated areas. Comm Dent Oral Epidemiol 9:103-107.

Hausen H, Seppa L, Fejerskov O (1994). Can caries be predicted? In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Cariology. Copenhagen: Munksgaard.

Hausen H (1997). Caries predictionstate of the art. Comm Dent Oral Epidemiol 25:87-96.

Heintze U, Birkhed D, Bjorn H (1983). Secretion rate and buffer effect of resting and stimulated whole saliva as a function of age. Swed Dent J 7:227-238.

Hix J, O'Leary T (1976). The relationship between cemental caries, oral hygiene status and fermentable carbohydrate intake. J Periodontol 47:398-404.

Holm A, K'son-Blomquist H, Crossner C, Grahnen H, Samuelson G (1975). A comparative study of oral health as related to general health, food habits and socioeconomic conditions of 4-year-old Swedish children. Comm Dent Oral Epidemiol 3:34-39.

Holmen L, Thylstrup A (1986). Natural caries development and its arrestment. In: Leach S (ed). Factors relating to demineralization of the teeth. London: IRL Press.

Holmen L, Thylstrup A, Artun J (1987). Clinical and histological features observed during arrestment of active enamel carious lesions in vivo. Caries Res 21:546-554.

Holmen L, Mejare J, Malmgren B, Thylstrup A (1988). The effect of regular professional plaque removal on dental caries in vivo. A polarized light and scanning microscope study. Caries Res 22:250-256.

Holund U, Theilade E, Poulsen S (1985). Validity of a dietary interviewing method for use in caries prevention. Comm Dent Oral Epidemiol 13:219-221.

Hugoson A, Koch G, Bergendal T, Hallonsten A, Slotte C, Thorstensson B, Thorstensson H (1995b). Oral health of individuals aged 3-80 years in Jonkoping, Sweden in 1973, 1983 and 1993. II. A review of clinical and radiographic findings. Swed Dent J 19:243-260.

Hugoson A, Norderyd O, Slotte C, Thorstensson H (1998). Oral hygiene and gingivitis in a Swedish adult population 1973, 1983 and 1993. J Clin Periodontol 25:807-812.

Hugoson A, Koch G, Hallonsten A, Norderyd J, Aberg A (1999). Caries prevalence and distribution in individuals 3-20 years of age, Jonkoping, Sweden; 1973, 1978, 1983 and 1993. Comm Dent Oral Epidemiol (in press).

Igarashi K, Lee I, Schachtele C (1989). Comparison of in vivo human dental plaque pH changes within artificial fissures and at interproximal sites. Department of Oral Sciences and Department of Rehabilitative Sciences, University of Minnesota.

Imfeld T (1977). Evaluation of the cariogenicity of confectionary by intraoral wire telemetry. Helv Odont Acta 21:1-28.

Imfeld T (1978). In vivo assessment of plaque acid production. A long-term retrospective study. In: Guggenheim (ed). Proceedings of ERGOB Conference on Health and Sugar Substitutes. Basel: Karger.

Imfeld T (1983). Identification of low caries risk dietary components. In: Meyers H (ed). Monographs in Oral Science. Basel: Karger.

Imfeld T (1994a). Clinical caries studies with polyalcohols. A literature review. Schweiz Monatsschr Zahnmed 104:941-945.

Imfeld T (1994b). Cariogenicity tests. Adv Dent Res 8:225-228.

Ismail A, Burt B, Eklund S (1984). The cariogenicity of soft drinks in the US. J Am Dent Assoc 109:241-245.

Ismail AI (1997). Clinical diagnosis of precavitated carious lesions. Comm Dent Oral Epidemiol 25:13-23.

Isokangas P, Tiekso J, Alanen P, Makinen K (1989). Long-term effect of xylitol

chewing gum on dental caries. Comm Dent Oral Epidemiol 17:444-448.

Jansson L, Ehnevid H, Blomlof L, Weintraub A, Lindskog S (1995). Endodontic pathogens in periodontal disease augmentation. J Clin Periodontol 22:598-602.

James P, Parfitt G, Faulkner F (1957). A study of the aetiology of labial caries of the deciduous incisor teeth in small children. Br Dent J 103:37-40.

James S, Tagg J (1988). A search within the genera Streptococcus, Enterococcus, and Lactobacillus for organisms inhibitary to mutans streptococci. Microb Ecol Health Dis 1:153-162.

Jenkins G, Geddes D, Cooke J (1973). Reinvestigation of experimental caries in man [abstract]. J Dent Res 52:967.

Jenkins G, Edgar W (1989). The effect of daily gum-chewing on salivary flow rates in man. J Dent Res 68:786-790.

Johansson I, Birkhed D (1994). Diet and the caries process. In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Cariology. Copenhagen: Munksgaard.

Johnson D, Sreebny L (1982). Effect of increasing the bulk content of the diet on the rat parotid gland and saliva. J Dent Res 61:691-696.

Joshi A, Papas A, Giunta J (1993). Root caries incidence and associated risk factors in middle-aged and older adults. Gerodontol 10:83-89.

Katz R (1980). Assessing root caries in populations: The evolution of the root caries index. J Public Health Dent 40:7-16.

Kandelman D, Gagnon G (1990). A 24-month clinical sudy of the incidence and progression of dental caries in relation to consumption of chewing gum containing xylitol in school programs. J Dent Res 69:1771-1775.

Keene H (1971). Epidemiologic study of tooth size variability in caries free naval recruits. J Dent Res 50:1331-1345.

Kidd E, Toffenetti F, Mjor I (1992). Secondary caries. Int Dent J 42:127-138.

Kilian M, Reinholdt J (1986).Interference with IgA defence mechanisms by extracellular bacterial enzymes. In: Easmon D, Jeljaszewiczj (eds). Medical Microbiology. Vol 5. London: Academic Press.

Kilian M, Bratthall D (1994). Caries immunology. In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Cariology. Copenhagen: Munksgaard.

Kingmann A, Morrison E, Loe H, Smith J (1988). Systemic errors in estimating prevalence and severity of periodontal disease. J Periodontol 59:707-713.

Kingmann A (1990). Statistical issues in risk models for caries. In: Bader J (ed). Risk

Assessment in Dentistry. 193-200. Chapel Hill: University of North Carolina Dental Ecology.

Kirkegaard E, Borgnakke V, Gronbaek L (1985). Oral health status, dental treatment need and dental care habits in a representative sample of the adult Danish population. Survey of oral health of Danish adults [thesis]. Aarhus: Royal Dental College.

Kleemola-Kujala E, Rasanen L (1982). Relationship of oral hygiene and sugar consumption to risk of caries in children. Comm Dent Oral Epidemiol 10:224-233.

Kleinberg I, Jenkins G (1964). The pH of dental plaques in the different areas of the mouth before and after meals and their relationship to the pH and rate of flow of resting saliva. Arch Oral Biol 9:493-516.

Klock B, Krasse B (1977). Microbial and salivary conditions in 9- to 12-year-olds. Scand J Dent Res 85:56-63.

Klock B, Krasse B (1978). Effect of caries-preventive measures in children with high numbers of S. mutans and lactobacilli. Scand J Dent Res 86:221-230.

Klock B, Krasse B (1979). A comparison between different methods for prediction of caries activity. Scand J Dent Res 87:129-139.

Koch G, Martinsson T (1970). Socio-odontologic investigation of school children with high and low caries frequency. I. Socio-economic background. Odontol Revy 21:207-228.

Kohler B, Bratthall D (1978). Intrafamilial levels of Streptococcus mutans and some aspects of the bacterial transmission. Scand J Dent Res 86:35-42.

Kohler B, Bratthall D (1979). Practical method to facilitate estimation of Streptococcus mutans levels in saliva. J Clin Microbiol 584-588.

Kohler B, Andreen I, Jonsson B, Hultqvist E (1982). Effect of caries preventive measures on Streptococcus mutans and lactobacilli in selected mothers. Scand J Dent Res 90:102-108.

Kohler B, Bratthall D, Krasse D (1983). Preventive measures in mothers influence the establishment of the bacterium Streptococcus mutans in their infants. Arch Oral Biol 28:225-231.

Kohler B, Andreen I, Jonsson B (1984). The effect of caries preventive measures in mothers on dental caries and the oral presence of the bacteria Streptococcus mutans and lactobacilli in their children. Arch Oral Biol 29:879-883.

Kohler B, Andreen I, Jonsson B (1988). The earlier the colonization of mutans streptococci, the higher the caries prevelance. Oral Microbiol Immunol 3:14-17.

Kohler B, Persson M (1991). Salivary levels of mutans streptococci and lactobacilli in dentate 80- to 85-year-old Swedish men and women. Comm Dent Oral Epidemiol

19:352-356.

Kolenbrander P (1988). Intergeneric coaggregation among human oral bacteria and ecology of dental plaque. Ann Rev Microbiol 42:627-656.

Kotsanos N, Darling A (1991). Influence of posteruptive age of enamel on its susceptibility to artificial caries. Caries Res 25:241-250.

Kristoffersson K, Bratthall D (1982). Transient reduction of Streptococcus mutans interdentally by chlorhexidine gel. Scand J Dent Res 90: 417-422.

Kristoffersson K, Axelsson P, Bratthall D (1984). The effect of a professional tooth-cleaning program on interdentally localized Streptococcus mutans. Caries Res 18:385-390.

Kristoffersson K, Axelsson P, Birkhed D, Bratthall D (1986). Caries prevalence, salivary Streptococcus mutans, and dietary habits in 13-year-old schoolchildren. Comm Dent Oral Epidemiol 14:202-205.

Kulkarni G, Chan K, Sandham H (1989). An investigation into the use of restriction endonuclease analysis for the study of transmission of mutans streptococci. J Dent Res 68:1155-1161.

Kuzmina I (1997). A caries preventive program among children in a district of Moscow [thesis]. Department of Cariology and Endodontics, School of Dentistry, Faculty of Health Sciences, University of Copenhagen, Denmark.

Lachapelle-Harvey D, Sevigny J (1985). Multiple regression analysis of dental status and related food behaviour of French Canadian adolescents. Comm Dent Oral Epidemiol 13:226-229.

Lagerlof F, Oliveby A (1990). Computer simulation of oral fluoride clearance. Comput Meth Prog Biomed 31:97-104.

Lagerlof F, Oliveby A (1994). Caries-protective factors in saliva. Adv Dent Res 8:229-238.

Lang N, Cumming B, Loe H (1973). Toothbrushing frequency as it relates to plaque development and gingival health. J Periodontol 7:396-405.

Leach S, Lee G, Edgar W (1989). Remineralization of artificial caries-like lesions in human enamel in situ by chewing sorbitol gum. J Dent Res 68:1064-1068.

Lervik T, Haugejorden O, Aas C (1990). Progression of posterior approximal carious lesions in Norwegian teenagers from 1982 to 1986. Acta Odontol Scand 48:223-227.

Lie T (1978). Ultrastructural study of early dental plaque formation. J Periodontal Res 13:391-409.

Lindhe J, Wicen P (1969). The effect on the gingivae of chewing fibrous foods. J

Periodontal Res 4:193-201.

Liljemark W, Fenner L, Bloomquist C (1986). In vivo colonization of salivary pellicle by Haemophilus, Actinomyces and Streptococcus species. Caries Res 20:481.

Lindquist B, Emilson C, Wennerholm K (1989). Relationship between mutans streptococci in saliva and their colonization of tooth surfaces. Oral Microbiol Immunol 4:71-76.

Listgarten M, Mayo H, Tremblay R (1975). Development of dental plaque on epoxy resin crowns in man. A light and electron microscopic study. J Periodontol 46:10-26.

Listgarten M (1976). Structure of the microbial flora associated with periodontal health and disease in man. A light and electron microscopic study. J Periodontol 47:1-18.

Loe H, Theilade E, Jensen S (1965). Experimental gingivitis in man. J Periodontol 36:177-187.

Loe H, von der Fehr F, Schiott C (1972). Inhibition of experimental caries by plaque prevention. The effect of chlorhexidine mouthrinses. Scand J Dent 80:1-9.

Loesche WJ (1982). Dental Carie1s: A Treatable Infection. Springfield: Thomas.

Loesche WJ (1986). Role of Streptococcus mutans in human dental decay. Microbiol Rev 50:353-380.

Longbottom C, Hysmans M, Pitts N, Los P, Bruce P (1996). Detection of dental decay and its extent using AC impedance spectroscopy. Nature Med 235-237.

Luan WM, Baelum V, Chen X, Fejerskov O (1989). Tooth mortality and prosthetic treatment patterns in urban and rural Chinese aged 20-80 years. Comm Dent Oral Epidemiol 17:221-226.

Lunder N, von der Fehr F (1996). Approximal cavitation related to bite-wing image and caries activity in adolescents. Caries Res 30:143-147.

Lundgren M (1997). On dental caries and related factors in old age [thesis]. Departments of Cariology and Geriatric Medicine, University of Gothenburg, Sweden.

Lussi A (1991). Validity of diagnostic and treatment decisions of fissure caries. Caries Res 25:296-303.

Lynch E, Beighton D (1994). A comparison of primary root caries lesions classified according to colour. Caries Res 28:233-239.

MacPherson L, MacFarlane T, Stephen K (1990). An intra-oral appliance study of the plaque microflora associated with early enamel demineralization. J Dent Res 69:1712-1716.

Manji F, Fejerskov O (1994). An epidemiological approach to dental caries. In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Cariology. Copenhagen: Munksgaard.

Makinen K, Hujoel P, Bennett C, Isotupa K, Makinen P, Allen P (1996). Polyol chewing gums and caries rates in primary dentition: a 24-month cohort study. Caries Res 30:408-417.

Manning R, Edgar W (1993). PH changes in plaque after eating snacks and meals, and their modification by chewing sugared or sugar-free gum. Br Dent J 174:241-244.

Mansbridge J (1960). The effects of oral hygiene and sweet consumption on the prevalence of dental caries. Br Dent J 109:343-348.

Mansson B (1977). Caries progression in the first permanent molars. A longitudinal study. Swed Dent J 1:185-191.

Markovic N, Abelson D, Mandel I (1988). Sorbitol gum in xerostomics: the effects on dental plaque pH and salivary flow rates. Gerodontol 7:71-75.

Marsh P (1989). Host defenses and microbial homeostasis: Role of microbial interactions. J Dent Res 68:1567-1575.

Marsh P (1991). Sugar, fluoride, pH and microbia homeostasis in dental plaque. Proc Finn Dent Soc 87:515-525.

Marsh P (1993). The role of chemostats in the evaluation of antimicrobial agents for use in dental products. Mirobiol Ecol Dis 6:147-149.

Marsh P (1994). Microbial ecology of dental plaque and its significance in health and disease. Adv Dent Res 8:263-271.

Martinsson T, Petersson A (1972). Socio-odontologic investigation of schoolchildren with high and low caries frequency. IV. Dental condition in the parents. Odontol Revy 23:371-388.

Martinsson T (1973). Socio-odontologic investigation of schoolchildren with high and low caries frequency. V. Socio-economic factors and dental condition in the parents. Odontol Revy 24:59-74.

Mayhall J (1977). The oral health of the Canadian Inuit community. An anthropological approach. International Conference on Oral Biology. J Dent Res (special issue C55-61).

Mejare I, Grondahl HG, Carlstedt K, Grewer AC, Ohosson E (1985). Accuracy of radiography and probing for the diagnosis of proximal caries. Scand J Dent Res 93:178-184.

Mejare I, Malmgren B (1986). Clinical and radiographic appearance of proximal

carious lesions at the time of operative treatment in young permanent teeth. Scand J Dent Res 94:19-26.

Mejare I, Kollestal C, Stenlund H (1999). Incidence and progression of approximal caries from 11 to 22 years of age in Sweden: A prospective radiographic study. Caries Res 33:93-100.

Milen A, Tala H (1986). Social inequity in oral healtha newly awakened problem. Proc Finn Dent Soc 82-260-266.

Milen A (1987). Role of social class in caries occurrence in primary teeth. Int J Epidemiol 16:252-256.

Miller A, Brunelle J, Carlos J, Brown L, Loe H (1987). Oral health of United States adults. National findings 1985-86. Bethesda, MD: NIH Pub 87-2868.

Milnes A, Bowden G (1985). The microflora associated with the developing lesions of nursing caries. Caries Res 19:289-297.

Minah G, Solomon E, Chu K (1985). The association between dietary sucrose consumption and microbial population shifts at six oral sites in man. Arch Oral Biol 30:397-401.

Mjor I (1985). Frequency of secondary caries at various anatomical locations. Oper Dent 10:88-92.

Moller I, Poulsen (1973). A standardized system for diagnosing, recording and analyzing dental caries data. Scand J Dent Res 81:1-11.

Moberg-Skold U, Klock B, Lindvall A (1997). Differences in caries recording with and without bitewing radiographs. A study on 5-year-old children in the county of Bohuslan, Sweden. Swed Dent J 21:69-75.

Mortimer K (1964). Some histological features of fissure caries in enamel. Proc Euro Caries Res 2:85-95.

Murray J (1989). The Prevention of Dental Diseases. Ed 2. London: Oxford University Press.

Neff D (1967). Acid production from different carbohydrate sources in human plaque in situ. Caries Res 1:78-87.

Neilson A, Pitts N (1993). Development and application of a quantitative method of monitoring macroscopic cavitation in smooth surface lesions in vivo. Caries Res 27:140-146.

Newbrun E, Matsukubo T, Hoover C, Graves R, Brown A, Disney J, Bohannan H (1984). Comparison of two screening tests for Streptococcus mutans and evaluation of their suitability for mass screenings and private practice. Comm Dent Oral Epidemiol 12:325-331.

Nygaard-Ostby B, Morch T, Hals E (1957). A method for caries production in selected tooth surfaces in vivo-employed in a preliminary study of the caries-inhibiting effect of topically applied agents. Acta Odontol Scand 15:357-363.

Nyvad B, Fejerskov O (1982). Root surface caries: Clinical, histopathological and microbiological features and clinical implications. Int Dent J 32:312-326.

Nyvad B, Fejerskov O (1986). Active root surface caries converted into inactive caries as a response to oral hygiene. Scand J Dent Res 94:281-284.

Nyvad B, Fejerskov O (1987). Transmission electron microscopy of early microbial colonization of human enamel and root surface in vivo. Scand J Dent Res 95:297-307.

Nyvad B, Killian M (1987). Microbiology of the early colonization of human enamel and root surfaces in vivo. Scand J Dent Res 95:369-380.

Nyvad B, ten Cate J, Fejerskov O (1989). Microradiography of experimental root surface caries in man. Caries Res 23:218-223.

Nyvad B, Fejerskov O (1994). Development, structure and pH of dental plaque. In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Caries. Copenhagen: Munskgaard.

Nyvad B, Fejerskov O (1997). Assessing the stage of caries lesion activity on the basis of clinical and microbiological examination. Comm Dent Oral Epidemiol 25:69-75.

Nyvad B, ten Cate J, Fejerskov O (1997). Arrest of root surface caries in situ. J Dent Res 76:1845-1853.

Ohrn K, Crossner CG, Borgesson I, Taube A (1996). Accuracy of dental hygienists in diagnosing dental decay. Comm Dent Oral Epidemiol 24:182-186.

O'Leary T, Drake R, Naylor J (1972). The plaque control record. J Periodontol 43:38-39.

Orland F, Blayney J, Wendell-Harrison R (1954). Use of the germ-free animal technique in the study of experimental dental caries. J Dent Res 33:147-174.

Palmer J, Pitter A (1988). Differences in dental caries levels between 8-year-olds in Bath from different socioeconomic groups. Comm Dent Health 5:363-368.

Park K, Schemehorn B, Stookey G (1993). Effect of time and duration of sorbitol gum chewing on plaque acidogenicity. Pediatr Dent 15:197-202.

Paul P, Bradnock G (1986). The dental health of Asian and Causasion 4- and 5-year-olds resident in Coventry. Comm Dent Health 3:275-286.

Paynter K, Grainger R (1962). Relationship of morphology and size of teeth to caries.

Int Dent J 12:147.

Pearce E (1991). Salivary inorganic and physical factors in the aetiology of dental caries, and their role in prediction. In: Johnson N (ed). Risk Markers for Oral Diseases. Vol 1. Dental Caries. Cambridge: Cambridge University Press.

Peers A, Hill F, Mitropoulos C, Holloway P (1993). Validity and reproducibility of clinical examination, fibre-optic transillumination, and bite-wing radiology for the diagnosis of small approximal carious lesions: An in vitro study. Caries Res 27:307-311.

Percival R, Challacombe S, Marsh P (1994). Flow rates of resting whole and stimulated parotid saliva in relation to age and gender. J Dent Res 73(8):1416-1420.

Persson L, Holm A, Arvidsson S, Samuelson G (1985). Infant feeding and dental caries, a longitudinal study of Swedish children. Swed Dent J 9:201-206.

Petersson H, Bratthall D (1996). The caries decline: A review of the reviews. Euro J Oral Sci 104:436-443.

Pitts N, Rimmer P (1992). An in vivo comparison of radiographic and directly assessed clinical caries status of posterior approximal surfaces in primary and permanent teeth. Caries Res 26:146-152.

Pitts N, Longbottom C (1987). Temporary separation with special reference to the diagnosis and preventive management of equivocal approximal carious lesions. Quintessence Int 18:563-573.

Pitts N, Longbottom C, Hysmans M, Los P, Bruce P (1995). Assessing approximal caries depth with an AC impedance spectroscopy technique. J Dent Res 74:463.

Pitts N (1997). Diagnostic tools and measurementsimpact on appropriate care. Comm Dent Oral Epidemiol 25:24-35.

Poulsen V (1988). Cariesrisikogrupper I bornetandplejen, en sundhedssociologisk og epidemiologisk evaluering [thesis]. Copenhagen: Royal Dental College.

Primosch R (1982). Effect of family structure on the dental caries experience of children. J Public Health Dent 42:155-168.

Quirynen M, Marechal M, Busscher H, Weerkamp A, Darius P, van Steenberghe D (1990a). The influence of surface free energy and surface roughness on early plaque formation. An in vivo study in man. J Clin Periodontol 17:138-144.

Qvist V, Johannessen L, Bruun M (1992). Progression of approximal caries in relation to iatrogenic preparation damage. J Dent Res 71:1370-1373.

Ramberg P, Lindhe J, Dahlen G, Volpe A (1994a). The influence of gingival inflammation on de novo plaque formation. J Clin Periodontol 21:51-56.

Ramberg P, Lindhe J, Gaffar A (1994b). Plaque and gingivitis in the deciduous and permanent dentition. J Clin Periodontol 21:490-496.

Ramberg P, Axelsson P, Lindhe J (1995). Plaque formation at healthy and inflamed gingival sites in young individuals. J Clin Periodontol 22:85-88.

Ravald N, Birkhed D (1991). Factors associated with active and inactive root caries in patients with periodontal disease. Caries Res 25:377-384.

Ravald N (1992). Studies on root surface caries in patients with periodontal disease [thesis]. Department of Cariology, University of Goteborg, Sweden.

Ravald N, Birkhed D (1992). Prediction of root caries in periodontally treated patients maintained with different fluoride programs. Caries Res 26:450-458.

Reeves R, Stanley H (1966). The relationship of bacterial penetration and pulpal pathosis in carious teeth. Oral Surg Oral Med Oral Pathol 22:59-65.

Rekola M (1987). Approximal caries development during 2-year total substitution of dietary sucrose and xylitol. Caries Res 21:87-94.

Richardson B, Cleaton-Jones P, McInnes P, Rantsho J (1981a). Infant feeding practices and nursing bottle caries. J Dent Child 48:423-429.

Rimmer P, Pitts N (1990). Temporary elective tooth separation as a diagnostic aid in general dental practice. Br Dent J 169:87-92.

Riviere G, Papagiannoulis L (1987). Antibodies to indogenous and laboratory strains of Streptococcus mutans in saliva from children with dental caries and from caries-free children. Ped Dent 9:216-220.

Roeters F, Van de Hoeven J, Burgersdijk R, Schaeken M (1995). Lactobacilli, mutans streptococci and dental caries: A longitudinal study in 2-year-olds up to the age of 5 years. Caries Res 29:272-279.

Rosenhek M, Macpherson L, Dawes C (1993). The cause of the inital peak in salivary flow rate (FR) with unflavoured gum base. J Dent Res 72:375.

Rugg-Gunn A (1989). Diet and dental caries. In: Murray (ed). The Prevention of Dental Diseases. Oxford: Oxford Medical Publishers.

Rugg-Gunn A, Edgar W, Geddes D, Jenkins G (1975). The effect of different meal patterns upon plaque pH in human subjects. Br Dent J 139:351-356.

Rugg-Gunn A, MacGregor I (1978). A survey of toothbrushing behaviour in children and young adults. J Periodontal Res 13:383-389.

Rugg-Gunn A, Hackett A, Appleton D, Jenkins G, Eastoe J (1984). Correlations of dietary intakes of calcium phosphorus and Ca P ratio with caries data in children. Caries Res 18:149-152.

Rugg-Gunn A, Hackett A, Appleton D (1987). Relative cariogenicity of starch and sugars in a 2-year longitudinal study of 405 English schoolchildren. Caries Res 21:464-473.

Rugg-Gunn A, Carmichael C, Ferrel R (1997). Effect of fluoridation and secular trend in caries in 5-year-old chldren living in Newcastle and Northumberland. Br Dent J 165:359-364.

Rugg-Gunn AJ (1994). Sugarlesstowards the year 2000. Department of Child Health, University of New Castle Upon Tyne, UK.

Salonen L, Allander L, Bratthall D, Hellden L (1990). Mutans streptococci, oral hygiene and caries in an adult Swedish population. J Dent Res 69:1469-1475.

Sanders W, Sanders C (1984). Modification of normal flora by antibiotics: Effects on individuals and the environment. In: Koot R, Sande M (eds). New Dimensions in Antimicrobial Therapy. 217-241. New York: Churchill Livingstone.

Sandham H, Brown J, Phillips H, Chan K (1988). A preliminary report of long-term elimination of detectable mutans streptococci in man. J Dent Res 67:9-14.

Sansone J, Van Houte K, Joshipura K, Margolis H (1993). The association of mutans streptococci and non-mutans streptococci capable of acidogenesis at a low pH with dental caries on enamel and root surfaces. J Dent Res 72:508-516.

Sarll D, Whittle J, Mackie I (1984). The use of classification of residential neighborhoods (ACORN) as a health-related variable in service planning for dentistry. Comm Dent Health 1:115-124.

Sawle RF, Andlaw RJ (1988). Has occlusal caries become more difficult to diagnose? A study comparing clinically undetected lesions in molar teeth of 14- to 16-year-olds in 1974 and 1982. Br Dent J 164:209-211.

Saxton CA (1973). Scanning electron microscope study of the formation of dental plaque. Caries Res 7:102-119.

Saxton CA (1975). The formation of human dental plaque: A study by scanning electron microscopy [thesis]. London: University of London.

Saxton CA (1976). The effects of dentifrices on the appearance of the tooth surface observed with the scanning electron microscope. J Periodontal Res11:74-85.

Schachtele C, Jensen M, Harlander S, Halambeck S, Morris H (1982). Cheese as a paradigm for evaluating human plaque pH changes. J Dent Res 61(special issue A):266.

Schamschula R, Barnes D, Adkins B (1972a). Caries etiology in Papua-New Guinea. Associations of tooth size and dental arch width. Australian Dent J 17:188-195.

Scheinin A, Makinen K (1975). Turku sugar studies I-XXI. Acta Odontol Scand 33:307-320.

Scheinin A, Makinen K, Tammisalo E, Rekola M (1975). Turku sugar studies XVII. In relation to 1-year consumption of xylitol chewing gum. Acta Odontol Scand 33:269-278.

Scheinin A, Pienihakkinen K, Tiekso J, Holmberg S (1992). Multifactorial modeling for root caries prediction. Comm Dent Oral Epidemiol 20:35-37.

Schou L, Currie C, McQueen D (1990). Using a lifestyle perspective to understand toothbrushing behaviour in Scottish. Comm Dent Oral Epidemiol 18:230-234.

Schroeder U, Granath L (1983). Predictive value of dietary habits and oral hygiene for the occurrence of caries in 3-year-olds. Comm Dent Oral Epidemiol 11:308-311.

Schroeder U, Edwardsson S (1987). Dietary habits, gingival status and occurrence of Streptococcus mutans and lactobacilli as predictors of caries in 3-year-olds in Sweden. Comm Dent Oral Epidemiol 15:320-324.

Schupbach P, Osterwalder V, Guggenheim B (1995). Human root caries: Microbiota in plaque covering sound carious and arrested carious root surfaces. Caries Res 29:382-385.

Schwarz E (1985). Dental programmes for children and young adults in Denmark in a social perspective. Scand J Prim Health 3:113-120.

Seddon R (1989). The detection of cavitation in carious approximal surfaces in vivo by tooth separation, impression and scanning electron microscopy. J Dent 17:117-120.

Silness J, Loe H (1964). Periodontal disease in pregnancy. II. Correlation between oral hygiene and periodontal condition. Acta Odontol Scand 22:121-135.

Silver D (1987). A Longitudinal study of infant feeding practice, diet and caries related to social class in children aged 3 and 8-10 years. Br Dent J 163:296-300.

Silverstone L (1973). Structure of caries enamel including the early lesion. Dent Update 1:101-105.

Sjogren K, Birkhed D, Persson L, Noren J (1993). Salivary fluoride clearance after a single intake of fluoride tablets and chewing gums in children, adults and dry mouth patients. Scand J Dent Res 5:274-278.

Sjogren K, Birkhed D, Rangmar B (1995). Effect of a modified toothpaste technique on approximal caries in preschool children. Caries Res 28:127-131.

Sjogren K, Lingstrom P, Lundberg A, Birkhed D (1997). Salivary fluoride concentration and plaque pH after using a fluoride-containing chewing gum. Caries Res 31:366-372.

Smith A, Moran J, Dangler L, Lieght R, Addy M (1996). The efficacy of an antigingivitis chewing gum. J Clin Periodontol 23:19-23.

Smith D, Taubman M (1991). Association of specific host immune factors with dental caries experience. In: Johnson N (ed). Risk Markers for Oral Diseases. Vol 1. Dental Caries. Cambridge: Cambridge University Press.

Soderholm K, Antonson D, Fischlweiger W (1989). Correlation between marginal discrepancies at the amalgam/tooth interface and recurrent caries. In: Anusavice K (ed). Quality Evaluation of Dental Restorations: Criteria for Placement and Replacement. Chicago: Quintessence.

Sognnaes R (1948). Analysis of wartime reduction of dental caries in European children. Am J Dis Child 75:792-821.

Spencer A, Davies M, Slade G, Brennan D (1994). Caries prevalence in Australia. Int Dent J 44:415-423.

Spencer A (1997). Skewed distributionsnew outcome measures. Comm Dent Oral Epidemiol 25:52-59.

Sreebny L (1982). Sugar availability, sugar consumption and dental caries. Comm Dent Oral Epidemiol 10:1-7.

Sreebny L Chatterjee R, Kleinberg I (1985). Clearance of glucose and sucrose from the saliva of human subjects. Arch Oral Biol 30:369-274.

Sreebny L, Baum B, Edgar W, Epstein J, Fox P, Larmas M (1992). Saliva: Its role in health and disease. Int Dent J 42:291-304.

Staat R, Gawronski T, Cressey D, Harris R, Folke L (1975). Effects of dietary sucrose levels on the quantity and microbial composition of human dental plaque. J Dent Res 54:872-880.

Stephan R (1940). Changes in hydrogen-ion concentration on tooth surfaces and in carious lesions. J Am Dent Assoc 27:718-723.

Stephan R (1943). The effect of urea on conteracting the influence of carbohydrates on the pH of dental plaques. J Dent Res 22:63-71.

Stephan R (1944). Intra-oral hydrogen-ion concentration associated with dental caries. J Dent Res 23:257-266.

Stephan K, Chestnut I, Jacobson A, et al (1994). The effect of NaF and SMFP toothpaste on 3 years' caries increments in adolescents. Int Dent J 44:287-295.

Stern S, Curzon M (1975). Variations in tooth dimension and geographic caries prevalence. J Dent Res 54:910.

Stiles H, Meyers R, Brunelle J, Wittig A (1976). Occurrence of Streptococcus mutans and Streptococcus sanguis in the oral cavity and feces of young children. In: Stiles H, Loesche W, O'Brien T (eds). Proceedings, Microbial Aspects of Dental Caries. Vol 1 [spec suppl, microbiol abtracts].

Stecksen-Blicks C (1985). Salivary counts of lactobacilli and Streptococcus mutans in caries prediction. Scand J Dent Res 93:201-212.

Stecksen-Blicks C, Arvidsson S, Holm A (1985). Dental health, dental care, and dietary habits in children in different parts of Sweden. Acta Odontol Scand 43:59-67.

Stromberg N (1996). Salivens fingeravtryck avslojar risk for tandlossning. Tandl Tidn 88:138-141.

Sullivan A, Granath L, Widenheim J (1989). Correlation between child caries incidence and S mutans/lactobacilli in saliva after correction for confounding factors. Comm Dent Oral Epidemiol 17:240-244.

Sullivan A, Borgstrom M, Granath L, Nilsson G (1996). Number of mutans streptococci or lactobacilli in a total dental plaque sample does not explain the variation in caries better than the numbers in stimulated whole saliva. Comm Dent Oral Epidemiol 24:159-163.

Svanberg M (1980). Streptococcus mutans in plaque after mouthrinsing with buffers in varying pH value. Scand J Dent Res 88:76-78.

Swenander-Lanke L (1957). Influence on salivary sugar of certain properties of foodstuffs and individual oral conditions. Acta Odontol Scand 15:153-156.

Taubman M, Smith D (1992). Significance of immune responses to oral antigens in dental diseases. In: Ciardi J, McGhee J, Keith J (eds). Genetically Engineered Vaccines. New York: Plenum.

Taylor R (1978). Variations of Morphology of Teeth. Springfield: Charles C. Thomas.

Teivens A, Mornstad H, Reventlid M (1996). Individual variation of tooth development in Swedish children. Swed Dent J 20:87-93.

Ten Cate J, De Josselin de Jong E, Extercate R, Sundstrom G, Angmar Mansson B (1996). Quantification of enamel demineralisation with a new portable fluorescence device, validated by microradiology. Caries Res 30:299.

Tenovuo J, Lumikari M (1991). Organic factors in human saliva in relation to dental caries. In: Johnson N (ed). Risk Markers for Oral Diseases. Vol 1. Dental Caries. Cambridge: Cambridge University Press.

Tenovuo J, Lagerlof F. Saliva; in Thylstrup A, Fejerskov O (1994). Textbook of Clinical Cariology. Copenhagen: Munksgaard.

Tenovuo J (1997). Salivary parameters of relevance for assessing caries activity in

individuals and populations. Comm Dent Oral Epidemiol 25:82-86.

Tenovuo J, Kirstila, Hakkinen, Jentsch H, Vilja P (1997). Longitudinal analysis of the association of human salivary antimicrobial agents to 2-year caries increment [abstract]. Caries Res 31:305.

Theilade E, Fejerskov O, Horsted M (1976). A transmission electron microscopic study of 7-day-old bacterial plaque in human tooth fissures. Arch Oral Biol 21:587-598.

Thibodeau E, O'Sullivan D (1996). Salivary mutans streptococci and dental caries patterns in prechool children. Comm Dent Oral Epidemiol 24:164-168.

Thylstrup A (1990). Clinical evidence of the role of pre-eruptive fluoride in caries prevention. J Dent Res 69:742-750.

Thylstrup A, Fejerskov O (1981). Surface features of early carious enamel at various stages of activity. In: Rolla G, Sonju T, Embery G (eds). Proceedings of a Workshop on Tooth Surface Interactions and Preventive Dentistry. London: IRL Press.

Thylstrup A, Bille J, Qvist V (1986). Radiographic and observed tissue changes in approximal carious lesions at the time of operative treatment. Caries Res 20:75-84.

Thylstrup A, Chironga L, Carvalho J, Ekstrand K (1989). The occurrence of dental calculus in occlusal fissures as an indication of caries activity. In: Ten Cate J (ed). Recent Advances in the Study of Dental Calculus. Oxford: IRL Press.

Thylstrup A, Fejerskov O (1994). Clinical and pathological features of dental caries. In: Thylstrup A, Fejerskov O (eds). Textbook of Clinical Cariology. Copenhagen: Munksgaard.

Todd J, Dodd T (1985). Children's Dental Health in the UK, 1983. London: HMSO.

Torell P, Ericsson Y (1965). Two-year test with different methods of local caries-preventive fluoride application in Swedish schoolchildren. Acta Odontol Scand 23:287-322.

Tveit A, Espelid I, Fjellteviet A (1994). Clinical diagnosis of occlusal dentin caries. Caries Res 28:368-372.

Twetman S, Petersson L, Pakhomov G (1996). Caries incidence in relation to salivary mutans streptococci and fluoride varnish applications in preschool children from low- and optimal-fluoride areas. Caries Res 30:347-353.

Twetman S, Petersson L (1997). Effect of different chlorhexidine varnish regimens on mutans streptococci levels in interdental plaque and saliva. Caries Res 31:189-193.

Von der Fehr, Loe J, Theilade E (1970). Experimental caries in man. Caries Res 4:131.

Vaarkamp J, Ten Bosch J, Verdonschot E, Tranaeus S (1997). Quantitative diagnosis of small approximal caries lesions utilizing wavelength-dependent fiber-optic transillumination. J Dent Res 76:875-882.

Van der Hoeven J, Camp P (1991). Synergistic degredation of mucin by Streptococcus oralis and Streptococcus sanguis in mixed chemostat cultures. J Dent Res 70:1041-1044.

Van Houte H, Yanover L, Brecher S (1981). Relationship of levels of the bacterium Streptococcus mutans in saliva of children and their parents. Arch Oral Biol 26:381-386.

Van Houte J, Jordan H, Laraway R, Kent R, Soparkar P, Depaola P (1990). Association of the microbial flora of dental plaque and saliva with human root surface caries. J Dent Res 69:1463-1468.

Van Houte J, Sansone C, Joshipura K, Kent R (1991). Mutans streptococci and non-mutans streptococci acidogenic at low pH and in vitro acidogenic potential of dental plaque in two different areas of the human dentition. J Dent Res 73:1727-1734.

Van Houte J (1993). Microbiological predictors of caries risk. Adv Dent Res 7:87-96.

Vehkalahti M (1987). Occurrence of root caries and factors related to this [thesis]. Helsinki: University of Helsinki.

Vehkalahti M, Paunio I (1994). Association between root caries occurrence and periodontal state. Caries Res 28:301-306.

Verdonschot E, Bronkhorst E, Wenzel A (1991). Approximal caries diagnosis using fibre-optic transillumination: A mathematical adjustment to improve validity. Comm Dent Oral Epidemiol 19:329-332.

Von der Fehr F (1965). Maturation and remineralization of enamel. Adv Fluorine Res 3:83-95.

Von der Fehr F, Loe J, Theilade E (1970). Experimental caries in man. Caries Res 4:131.

Walls A, Silver P, Steele J (2000). Impact of treatment provision on the epidemiological recording of root caries. Euro J Oral Sci 108:3-8.

Wallman C, Krasse B (1992). Mutans streptococci in margins of fillings and crowns. J Dent 25:174-178.

Weatherell J, Deutsch D, Robinson C, Hallsworth A (1977). Assimilation of fluoride by enamel throughout the life of the tooth. Caries Res 11:85-115.

Wendt L, Hallonsten A, Koch G (1992). Oral health in preschool children living in Sweden. Part IIA longitudinal study. Findings at 3 years of age. Swed Dent J 16:41-49.

Wendt L, Hallonsten A, Koch G, Birkhed D (1994). Oral hygiene in relation to caries development and immigrant status in infants and toddlers. Scand J Dent Res 102:269-273.

Wendt L (1995). On oral health in infants and toddlers. Swed Dent J 106(suppl).

Wendt L, Birkhed D (1995). Dietary habits related to caries development and immigrant status in infants and toddlers living in Sweden. Acta Odontol Scand 53:339-344.

Wendt L, Hallonsten A, Koch G, Birkhed D (1996). Analysis of caries-related factors in infants and toddlers living in Sweden. Acta Odontol Scand 54:131-137.

Wenzel A, Pitts N, Verdonschot E, Kalsbeek H (1992). Diagnosis, structure and behaviour of small carious lesions. Diagnostic techniques: Radiology. Caries Res

Widstrom E, Nilsson B (1986). Immigrants and dental treatment in Sweden. Comm Dent Health 2:207-212.

Wilcox C, Everett F (1963). Friction on the teeth and the gingiva during mastication. J Am Dent Assoc 66:513-520.

Winter G, Hamilton M, James P (1966). Role of the comforter as an aetiological factor in rampant caries of the deciduous dentition. Arch Dis Child 41:207-212.

Winter G, Rule D, Mailer G, James P, Gordon P (1971). The prevalence of dental caries in preschool children aged 1-4 years. Br Dent J 130:434-436.

Winter G (1980). Problems involved with the use of comforters. Int Dent J 3028-3038.

Wright G, Banting D, Feasby W (1979). The Dorchester dental flossing study: Final report. Clin Prev Dent 1:23-26.

Zadik D (1978). Epidemiology of dental caries in 5-year-olds in Israel. Comm Dent Oral Epidemiol 6:91-96.

Zickert I, Emilson C, Krasse B (1982a). The effect of caries preventive measures in children highly infected with the bacterium Streptococcus mutans. Arch Oral Biol 27:861-868.

List of Abbreviations

ACISTalternating current impedance spectroscopy technique

ACORNA Classification Of Residential Neighborhoods

CaF2calcium fluoride

CCDcharged-coupled device

CCITNCommunity Caries Index of Treatment Needs

CFUcolony-forming units

CIcaries incidence

DMFSdecayed, missing, or filled surface

DMFTdecayed, missing, or filled teeth

DHSSDepartment of Health and Social Security

EFFendoscopic filtered fluorescence

FAfluorapatite

FOTIfiber-optic transillumination

H4P2O7pyrophosphate

HAhydroxyapatite

IgAimmunoglobulin A

IgGimmunoglobulin G

IgMimmunoglobulin M

MSmutans streptococci

PFRIPlaque Formation Rate Index

PIPlaque Index

PMTCprofessional mechanical toothcleaning

PRFprognostic risk factor

PRPproline-rich glycoprotein

QLFquantitative laser (light) fluorescence

RFrisk factor

RIrisk indicator

SBEsalivary buffering effect

SLselective Lactobacillus

SSRsalivary secretion rate

WHOWorld Health Organization