ORIGINAL ARTICLE

Structural response of human corneal and scleral tissuesto collagen cross-linking treatment with riboflavinand ultraviolet A light

Samjin Choi & Seung-Chan Lee & Hui-Jae Lee &

Youjin Cheong & Gyeong-Bok Jung & Kyung-Hyun Jin &

Hun-Kuk Park

Received: 13 August 2012 /Accepted: 12 November 2012 /Published online: 23 November 2012# Springer-Verlag London 2012

Abstract High success rates in clinical trials on keratoconiccorneas suggest the possibility of efficient treatment againstmyopic progression. This study quantitatively investigatedthe in vitro ultrastructural effects of a photooxidative colla-gen cross-linking treatment with photosensitizer riboflavinand UVA light in human corneo-scleral collagen fibrils. Atotal of 30.8×2 mm corneo-scleral strips from donor tissuewere sagittally dissected using a scalpel. The five analyticparameters namely fibril density, fibril area, corneo-scleralthickness, fibril diameter, and fibril arrangement were in-vestigated before and after riboflavin–UVA-catalyzed colla-gen cross-linking treatment. Collagen cross-linking effectswere measured at the corneo-scleral stroma and were basedon clinical corneal cross-linking procedures. The structuralresponse levels were assessed by histology, digital mechan-ical caliper measurement, scanning electron microscopy,and atomic force microscopy. Riboflavin–UVA-catalyzedcollagen cross-linking treatment led to an increase in the

area, density, and diameters of both corneal (110, 112, and103 %) and scleral (133, 133, and 127 %) stromal collagens.It also led to increases in corneal (107 %) and scleral(105 %) thickness. Collagen cross-linking treatment throughriboflavin-sensitized photoreaction may cause structuralproperty changes in the collagen fibril network of the corneaand sclera due to stromal edema and interfibrillar spacingnarrowing. These changes were particularly prominent inthe sclera. This technique can be used to treat progressivekeratoconus in the cornea as well as progressive myopia inthe sclera. Long-term collagen cross-linking treatment ofkeratoconic and myopic progression dramatically improvesweakened corneo-scleral tissues.

Keywords Collagen cross-linking treatment . Riboflavinand ultraviolet A light . Keratoconus . Myopia . Cornealand scleral collagen fibrils

Introduction

Keratoconus is an ocular disorder characterized by cornealdegeneration due to corneal thinning and bilateral conicalprotrusion [1]. Collagen cross-linking techniques using thephotosensitizer riboflavin and ultraviolet A (UVA) lightwith a 370-nm wavelength were first introduced to manageprogressive keratoconus. The basic principle of this thera-peutic method is that the riboflavin–UVA-catalyzed cornealcollagen cross-linking reaction achieves additional covalentbonding between collagen molecules. This leads to in-creased corneal stiffness and enhanced resistance [1–10].

Myopia is one of the most prevalent ocular disorders andis characterized by a mismatch between the power and axiallength of the eye. Myopia is due to scleral thinning andlocalized ectasia of the posterior sclera [11–13]. Various

Samjin Choi and Seung-Chan Lee contributed equally to this paper.

S. Choi :Y. Cheong :G.-B. Jung :H.-K. Park (*)Department of Biomedical Engineering and Healthcare IndustryResearch Institute, College of Medicine, Kyung Hee University,1 Hoegi-dong, Dongdaemun-gu,Seoul 130-701, Republic of Koreae-mail: sigmoidus@khu.ac.kr

S.-C. LeeDepartment of Ophthalmology, Kangwon National University,Gangwon-do, Republic of Korea

H.-J. Lee :K.-H. JinDepartment of Ophthalmology, Kyung Hee University,Seoul, Republic of Korea

H.-K. ParkDepartment of Medical Engineering, Kyung Hee University,Seoul, Republic of Korea

Lasers Med Sci (2013) 28:1289–1296DOI 10.1007/s10103-012-1237-6

clinical treatments, including pharmaceutical agents, pro-gressive additional lenses, rigid gas-permeable contactlenses, orthokeratotic lens, and scleral reinforcement sur-gery, have been proposed for reducing myopic progression,but they have not achieved outstanding results in clinicaltrials [14–17]. An efficient treatment for progressive myopiahas not yet been found [11]. As with keratoconus, the exactetiology of myopia is still controversial. However, a combi-nation of genetic, environmental, and hormonal factors isthought to contribute to its development. It is generallyaccepted that keratoconus and myopia progression is asso-ciated with a lack of the synthesis of extracellular matrixcomponents during corneal and scleral development [13, 18,19].

High success rates in clinical trials on keratoconic cor-neas suggest the possibility of efficient treatment againstmyopic progression. Several studies [11–13] have reportedthe effects of collagen cross-linking treatment throughriboflavin-sensitized UVA photoreaction on the sclera.However, to the best of the authors' knowledge, there areno reports on the ultrastructural effects of simultaneousriboflavin–UVA-catalyzed collagen cross-linking treatmentof the cornea and sclera. This study quantitatively examinedand compared the immediate structural response of collagencross-linking treatment with photosensitizer riboflavin and370 nm UVA irradiation on human corneo-scleral speci-mens using histological, thickness, scanning electron mi-croscopy (SEM), and atomic force microscopy assessments.

Materials and methods

Strip preparation

Five human corneo-scleral tissues (37±9 years, male) werecollected from the Eye Bank of the Kyung Hee UniversityMedical Center in Seoul, Republic of Korea. Informedconsent for the use of human tissue for research wasobtained from the eye bank. Serologic tests of the donortissues were negative for hepatitis, syphilis, and humanimmune deficiency virus. The tissue was preserved for3 months in 90 % ethanol for sterilization after donation.Before photooxidative collagen cross-linking treatment, six8×2 mm strips including the cornea and sclera were sagit-tally dissected from the donor tissue using a scalpel. Allstrips (n030) were then removed from the ethanol andirrigated with balanced salt solution (BSS) at roomtemperature.

Collagen cross-linking treatment

Half of the prepared strips (n015) were immediately fixedin a 4 % buffered paraformaldehyde solution for 1 day, and

the other half (n015) were used in the following procedure.A 0.1 % riboflavin photosensitizer solution (10 mgriboflavin-5-phosphate in 10 ml 20 % dextran-T-500) wasinstilled into the strips for 10 min before UVA irradiation.UVA irradiation at 370 nm was applied using a double UVAsource (UV-XTM Illumination System, IROC AG, Zurich,Switzerland), with a surface irradiance of 4.2 mW/cm2 at adistance of 30 mm from the strips for 30 min. The collagencross-linked strips were irrigated with BSS and fixed in a4 % buffered paraformaldehyde solution for 1 day.

Histological examination

After fixation the specimens were embedded in paraffin.They were sectioned on the sagittal plane and stained withMasson's trichrome to semi-quantify the collagen fibrils andinterfibrillar spacing. Two professional pathologists whowere blinded to the group assignment used a ScanningProbe Image Processor version 4.8 (SPIP, Image Metrology,Lyngby, Denmark) to perform morphometric analysis of thecorneo-sclera, including fibril density and area.

Thickness measurement

Corneo-scleral thickness was examined using a digital me-chanical caliper in BSS solution. The thickness was mea-sured f ive t imes for each str ip at the onset ofexperimentation. The thickness values of the corneo-sclerawere determined by an average of the five replicates.

Atomic force microscopy measurement

After measuring the thickness, specimens were dehydratedwith increasing concentrations of ethanol and immobilizedon mica with double-sided tape. Two areas of each stripwere scanned using a NANOS N8 NEOS (Bruker, Herzo-genrath, Germany) in non-contact mode. Atomic force mi-croscopy tapping mode corneo-scleral images were acquiredusing a silicon cantilever with an integral pyramidal-shapedtip in air at 35 % relative humidity, at a resolution of 256×256 pixels, a scan speed of 0.8 line/s, and a size of 5,000×5,000 nm (SICONG, Santa Clara, CA, USA) [20]. A total of150 atomic force microscopy corneo-scleral images weretaken. Ten atomic force microscopy tapping mode imagesper group were used for morphometric analysis with SPIPsoftware by two observers blinded to the experimentalgroup assignment.

Scanning electron microscopy measurement

SEM (Hitachi S-4700, Hitachi Co. Ltd., Tokyo, Japan) wasperformed to examine strip surfaces on the micrometerscale. Each strip was vacuum-coated with a layer of carbon

1290 Lasers Med Sci (2013) 28:1289–1296

followed by a layer of gold. The surfaces of all strips werethen investigated with a beam voltage of 5 kV and a mag-nification×15,000. Two observers blinded to the group as-signment measured the collagen fibril arrangement.

Statistics

Fibril density, fibril area, corneo-scleral thickness, fibrildiameter, and fibril arrangement before and after photooxi-dative collagen cross-linking treatment were expressed asmeans ± standard deviations. Statistical analysis was per-formed to compare the structural changes in the collagenfibrils of cross-linked corneo-scleral stromal tissues with thecontrol using a two-tailed Student's t test. P values<0.05were considered significant.

Results

Histological changes

Figure 1 shows a representative result of the cross-sectionalhistopathology examination (Masson's trichrome staining,×300) of human corneo-scleral tissues before and after collagencross-linking treatment with riboflavin and UVA light. Thenormal (CO) cornea and sclera showed collagen lamellae andinterfibrillar spacing (Fig. 1a). The collagen cross-linked (CxL)corneal and scleral stroma showed mild inflammatory infiltra-tion, stromal swelling, and narrowing of the interfibrillar spac-ing. Interestingly, RGB histograms of histological images forthe corneal stroma showed different patterns than those of thescleral stroma (Fig. 1b). Although there was no differencebefore and after the photooxidative collagen cross-linking treat-ments, the scleral R histogram showed a wider distribution thanthe corneal R histogram, contributing to high contrast. Theseimages indicate that collagen cross-linking treatment signifi-cantly increased (Fig. 1c, p00.0004) the area of collagen fibrilsby 110.07±5.24 % for the cornea and 133.48±7.36 % for thesclera. In addition, there was a significant decrease (p<0.0001)in the area of interfibrillar spacing by 37.26±3.83 % for thecornea and 15.85±3.40 % for the sclera. Riboflavin–UVA-catalyzed collagen cross-linking treatment led to a significantincrease (Fig. 1d, p00.0038) in the density of collagen fibrils,112.94±6.50 % and 133.48±7.36 % for the cornea and sclera,respectively. Treatment also led to a significant decrease (p00.0001) in the density of interfibrillar spacing by 36.87±6.12%for the cornea and 14.83±2.66 % for the sclera.

Thickness changes

The thickness of normal corneo-scleral tissues was 0.56±0.04 mm for the cornea and 0.75±0.04 mm for the sclera(Fig. 2). Cross-linked corneo-scleral thickness was 0.60±

0.06 mm for the cornea and 0.79±0.05 mm for the sclera.Scleral thickness was greater than corneal thickness, with anincrease of >130 %. However, there was no statisticallysignificant change in corneal (p00.2870) or scleral (p00.1879) thickness after riboflavin–UVA-catalyzed collagencross-linking treatment.

Structural changes

Figure 3 shows representative SEM images of human corneo-scleral stromal surfaces before and after riboflavin–UVA(RFUVA)-catalyzed collagen cross-linking treatment. SEMtopographical images showed that typical corneal (91.35±26.66 nm) and scleral (165.16±59.06 nm) surfaces have aregular parallel arrangement of collagen fibrils with clear axialperiodicity (Fig. 3a). However, corneal (91.90±20.56 nm) andscleral (200.82±97.97 nm) surfaces after collagen cross-linking treatments had tangled fibrils and fibrils running indifferent directions. Fibril diameters of the cornea weresmaller than those of the sclera by about 2.19 times for normaltissues and 1.79 times for cross-linked tissues. Riboflavin–UVA-catalyzed collagen cross-linking treatment led to in-creased collagen fibril diameters for the cornea (101 %, p00.3621, n0100) and the sclera (122 %, p00.9617, n0100),but it was not a significant increase (Fig. 3b).

Figure 4 shows the representative atomic force microscopy(AFM) tapping mode topographical images of human corneo-scleral stromal surfaces through RF-sensitized photoreaction.These images (5×5 μm) show that the typical normal corneal(Fig. 3a, c, 94.03±16.51 nm) and scleral (153.35±32.55 nm)surfaces had a regular parallel arrangement of collagen fibrilswith clear axial periodicity. RFUVA-catalyzed collagen cross-linking treatment resulted in an irregular parallel arrangementof collagen fibrils with increased diameters. These changeswere more pronounced in the sclera than in the cornea, withdiameters of 97.46±10.76 nm for the cornea (104 %, p00.6072, n0100) and 201.76±55.61 nm for the sclera(132%, p00.0369, n0100) following treatment. Atomic forcemicroscopy tapping mode 3D (AFM3D) images were used toinspect the details of the elaborate structures of corneo-scleralcollagen composites (Fig. 4b). Collagen cross-linking treat-ment led to a wide distribution of collagen fibril diameters forthe sclera rather than for the cornea (Fig. 3c). Overall, the fibrildiameters of the sclera were larger than those of the cornea(Fig. 4d).

Discussion

This study quantitatively compared the ultrastructuraleffects of RFUVA-catalyzed collagen cross-linking treat-ment on the morphology of collagen fibrils in human cor-neal and scleral tissues. Measurements of fibril density, fibril

Lasers Med Sci (2013) 28:1289–1296 1291

area, corneo-scleral thickness, fibril diameter, and fibrilarrangement were examined to determine the effects ofcollagen cross-linking treatment through RF-sensitizedUVA photoreaction. Histology, digital mechanical calipermeasurements, SEM, and atomic force microscopy wereused to assess morphological changes in collagen fibrils.

The following four methodology-based findings resultedfrom this study:

1. Histological analysis showed that RFUVA-catalyzedcollagen cross-linking treatment led to an increase inboth the area and density of corneo-scleral stromal

collagen composites. This change was more pro-nounced in the sclera than in the cornea.

2. Mechanical caliper assessment showed that RFUVA-catalyzed collagen cross-linking treatment led to an in-crease in corneo-scleral thickness, but it was not signifi-cant. Scleral thickness was greater than corneal thickness.

3. SEM assessment showed that RFUVA-catalyzed colla-gen cross-linking treatment led to morphologicalchanges such as tangled fibrils and fibrils running indifferent directions, but these changes were not signifi-cant. The fibril diameters of the sclera were bigger thanthose of the cornea.

Fig. 1 Representative cross-sectional histological images andmorphomet-ric analysis of human corneal and scleral stromal tissues before and aftercollagen cross-linking treatment through riboflavin-sensitized UVA irradi-ation. (a) Normal (CO), cross-linking (CxL), Masson's trichrome (MT)(scale bars0100 μm). (b) RGB histograms of histological images for thecorneal stroma showed different patterns than those of the scleral stroma

regardless of the photooxidative collagen cross-linking treatments. (c) Arearatio. (d) Density ratio. The collagen cross-linked cornea and sclera showstromal edema and narrowing of the interfibrillar spacing compared tonormal cornea and sclera. These effects were more prominent in the sclera.Single asterisk p<0.005, double asterisk p<0.0005, triple asteriskp00.0001 vs. corneal tissues, quadruple asterisk p<0.0001

1292 Lasers Med Sci (2013) 28:1289–1296

4. Atomic force microscopy assessment showed that over-all results were similar to SEM findings. RFUVA-catalyzed collagen cross-linking treatment led to a sig-nificantly increased fibril diameter for the sclera, but notfor the cornea.

Various clinical procedures have been proposed for treat-ing patients with keratoconus according to their diseasestage and progression [1–3, 8]. Deep anterior lamellar ker-atoplasty can cause increased surgery time and surgicaldifficulty, resulting in poor vision [9, 21]. Contact lensesand intra-stromal corneal ring segments lead only to changes

in the corneal shape, whereas collagen cross-linking leads tochanges in the structure and/or composition of the cornea[3].

Corneal collagen cross-linking is a minimally invasiveprocedure in which a chemical agent is applied to theresidual cornea after epithelium removal. This chemicalagent, either by itself or when exposed to UVA light, ini-tiates the formation of new molecular bonds between colla-gen fibrils and lamellae. New bonds likely increase themechanical strength of the cornea because they can physi-cally link individual collagen fibrils and adjacent lamellae ofthe corneal stroma.

Raiskup-Wolf et al. [1] revealed that patients who haveprogressive keratoconus for more than 3 years show long-term stabilization and improvement after collagen cross-linking. McCall et al. [2] explained the mechanism of col-lagen cross-linking with riboflavin–UVA, in whichriboflavin-sensitized UVA irradiation generates free radicalsand reactive oxygen species such as superoxide anion (O2

−),hydroxyl radical (OH), and hydrogen peroxide (H2O2). Thismainly occurs by the type-I pathway of photosensitizedoxidation. Wittig-Silva et al. [7] reported that randomizedcontrolled trials showed a temporary stabilization of alltreated eyes after collagen cross-linking. Most studies aboutcollagen cross-linking treatment of keratoconic progressionsuggest that collagen cross-linking treatment throughriboflavin-sensitized UVA irradiation increases corneal ri-gidity. However, there are no reports on the ultrastructuraleffects of riboflavin–UVA-catalyzed collagen cross-linkingtreatment on corneal tissues. This study showed that ribo-flavin–UVA-catalyzed collagen cross-linking treatment ledto structural improvements in corneal tissue as well asincreases in thickness (107 %), fibril density (112 %), area(110 %), diameter (103 %), and parallel arrangement. This

Fig. 2 Thickness changes in corneo-scleral tissues after riboflavin–UVA-catalyzed collagen cross-linking treatment. Normal (CO), cross-linking (CxL). Triple asterisk p<0.001 vs. cross-linked corneal tissues,quadruple asterisk p<0.0005 vs. normal corneal tissues. Riboflavin–UVA-catalyzed collagen collagen cross-linking treatment led to no sig-nificant change in corneal (p00.2870) or scleral (p00.1879) thickness

Fig. 3 Representative SEM topographical images (×15,000) of normaland collagen cross-linked corneo-scleral stromal surfaces. (a) Typicalnormal corneal and scleral surfaces showed a regular parallel arrangementof collagen fibrils with clear axial periodicity. Normal (CO), cross-linking(CxL). Collagen cross-linking treatment through RF-sensitized

photoreaction led to tangled fibrils and fibrils running in different direc-tions. (b) There were increased fibril diameters in both the cornea andsclera, but they were not significant (p00.3621 and 0.9617 for cornealand scleral tissues). Single asterisk p<0.05 vs. cross-linked cornealtissues, double asterisk p<0.01 vs. normal corneal tissues

Lasers Med Sci (2013) 28:1289–1296 1293

structural finding suggests that collagen cross-linking treat-ment through riboflavin-sensitized UVA irradiation mightincrease corneal stiffness because of the additional covalentbinding between collagen molecules.

Few studies have reported the effect of scleral collagencross-linking treatment through riboflavin-sensitized photo-reaction [11–13], because the stromal thinning process ofprogressive myopia is similar to that of progressive kerato-conus. This similarity suggests the possibility of a novel

therapy against progressive myopia. Riboflavin–UVA-cata-lyzed collagen cross-linking treatment leads to significantincreases in the biomechanical rigidity efficiency of scleralstromal collagens, which may prevent myopic progression.However, the studies also provided no information aboutstructural changes in scleral stromal collagen. A previousstudy [22] revealed through Raman spectroscopy that ribo-flavin–UVA-catalyzed collagen cross-linking treatment ledto changes in the structure and chemical composition of

Fig. 4 Representative atomic force microscopy tapping mode topo-graphical images and fibril distributions of normal and collagen cross-linked corneo-scleral surfaces. (a) Typical normal corneal and scleralsurfaces showed a regular parallel arrangement of collagen fibrils withclear axial periodicity. RFUVA-catalyzed collagen cross-linking treat-ment resulted in an irregular parallel arrangement of collagen fibrils withsignificantly increased diameters in both the cornea and sclera. Normal(CO), cross-linking (CxL), atomic force microscopy (AFM). (b) Atomic

force microscopy tapping mode 3D images showed the details of theelaborate structures of corneo-scleral collagen composites. Atomic forcemicroscopy 3D (AFM3D). (c) RFUVA-catalyzed collagen cross-linkingtreatment led to a wide distribution of collagen fibril diameters for thesclera rather than for the cornea. (d) The fibril diameters of the sclera werelarger than those of the cornea. Triple asterisk p<0.001 vs. normal cornealtissues, quadruple asterisk p<0.0001 vs. cross-linked corneal tissues,dagger p<0.05 vs. normal scleral tissues

1294 Lasers Med Sci (2013) 28:1289–1296

sclera. However, the study did not provide clear qualitativeinformation by histology or atomic force microscopy aboutthe structural effect of collagen cross-linking treatment onthe sclera. This study investigated the immediate structuralresponses of weakened or thinned corneo-scleral tissues toriboflavin–UVA-catalyzed corneal collagen cross-linkingtreatment. Fortunately, the findings of this study suggestthat collagen cross-linking treatment through riboflavin-sensitized UVA irradiation is an effective treatment againstprogressive myopic sclera. Due to stromal edema and inter-fibrillar space narrowing, this treatment leads to increases inthickness (105 %), fibril density (133 %), area (133 %),diameter (127 %), and parallel arrangement.

The interaction of collagen catalyzed by riboflavin–UVAfrom a human cornea or sclera leads to the formation of astable cross-link by increasing the number of cross-linkedcollagens. This phenomenon is verified by the findings ofthis study that this treatment led to thicker and more rigidcorneo-scleral tissues. Under the experimental conditions ofthis study, collagen cross-linking treatment resulted in re-markable structural changes in the sclera and stiffnesschanges in the cornea.

Collagen fibrils of the corneo-scleral stroma were exam-ined using four investigation methods in this study. Histo-logical analysis was used to examine the cross-sectionalstructural response of riboflavin–UVA-catalyzed corneo-scleral collagen cross-linking treatment on human tissues.Conventional vernier calipers were used to measure tissuethickness. Conventional SEM and novel atomic force mi-croscopy analyses were used to examine the ultrastructuralresponse of corneo-scleral collagen cross-linking treatmentwith photosensitizer riboflavin and 370-nm UVA irradia-tion. Histology, thickness, SEM, and atomic force micros-copy assessments provided satisfactory results, as expected.

Conclusions

This study compared the immediate quantitative effects ofcollagen cross-linking treatments with photosensitizer ribo-flavin and 370-nm UVA irradiation on human corneo-scleraltissues through structural changes in stromal collagen com-ponents. The results suggest that although there was not adramatic improvement in corneo-scleral tissues, the collagencross-linking technique leads to increases in the density,area, diameter, and thickness of human corneo-scleral tis-sues. This technique can be used to treat progressive kera-toconus in the cornea as well as progressive myopia in thesclera. Long-term collagen cross-linking treatment of kera-toconic and myopic progression dramatically improvesweakened corneo-scleral tissues. There are some limitationsto this study, including small sample size and that theexaminations were only performed immediately following

treatment and not over time. In addition, no biologicalinformation was collected. Further studies are required toevaluate the long-term effects of various clinically basedapplications and analyses in riboflavin–UVA-catalyzed col-lagen cross-linking treatment.

Acknowledgments This study was supported by a grant of the KoreanHealth Technology R&DProject, Ministry of Health &Welfare, Republicof Korea (A110216).

References

1. Raiskup-Wolf F, Hoyer A, Spoerl E, Pillunat LE (2008) Collagencrosslinking with riboflavin and ultraviolet-A light in keratoconus:long-term results. J Cataract Refract Surg 34:796–801

2. McCall AS, Kraft S, Edelhauser HF, Kidder GW, LundquistRR, Bradshaw HE, Dedeic Z, Dionne MJ, Clement EM, Con-rad GW (2010) Mechanisms of corneal tissue cross-linking inresponse to treatment with topical riboflavin and long-wavelength ultraviolet radiation (UVA). Invest Ophthalmol VisSci 51:129–138

3. Gefen A, Shalom R, Elad D, Mandel Y (2009) Biomechanicalanalysis of the keratoconic cornea. J Mech Behav Biomed Mater2:224–236

4. Agrawal V (2009) Corneal collagen cross-linking with riboflavinand ultraviolet-A light for keratoconus: results in Indian eyes.Indian J Ophthalmol 57:111–114

5. Vinciguerra P, Albè E, Trazza S, Rosetta P, Vinciguerra R, Seiler T,Epstein D (2009) Refractive, topographic, tomographic, and aber-rometric analysis of keratoconic eyes undergoing corneal cross-linking. Ophthalmology 116:369–378

6. Kymionis GD, Diakonis VF, Kalyvianaki M, Portaliou D, SiganosC, Kozobolis VP, Pallikaris AI (2009) One-year follow-up ofcorneal confocal microscopy after corneal cross-linking in patientswith post laser in situ keratosmileusis ectasia and keratoconus. AmJ Ophthalmol 147:774–778

7. Wittig-Silva C, Whiting M, Lamoureux E, Lindsay RG, SullivanLJ, Snibson GR (2008) A randomized controlled trial of cornealcollagen cross-linking in progressive keratoconus: preliminaryresults. J Refract Surg 34:S720–S725

8. Wollensak G, Redl B (2008) Gel electrophoretic analysis of cor-neal collagen after photodynamic cross-linking treatment. Cornea27:353–356

9. Al-Torbak AA, Al-Motowa S, Al-Assiri A, Al-Kharashi S, Al-Shahwan S, Al-Mezaine H, Teichmann K (2006) Deep anteriorlamellar keratoplasty for keratoconus. Cornea 25:408–412

10. Spoerl E, Wollensak G, Seiler T (2004) Increased resistance ofcrosslinked cornea against enzymatic digestion. Curr Eye Res29:35–40

11. Wollensak G, Iomdina E (2009) Long-term biomechanical proper-ties of rabbit sclera after collagen crosslinking using riboflavin andultraviolet A (UVA). Acta Ophthalmol 87:193–198

12. Wollensak G, Iomdina E, Dittert DD, Salamatina O, Stoltenburg G(2005) Cross-linking of scleral collagen in the rabbit using ribo-flavin and UVA. Acta Ophthalmol Scand 83:477–482

13. Wollensak G, Spoerl E (2004) Collagen crosslinking of human andporcine sclera. J Cataract Refract Surg 30:689–695

14. Siatkowski RM, Cotter SA, Crockett RS, Miller JM, Novack GD,Zadnik K (2008) Two-year multicenter, randomized, double-masked, placebo-controlled, parallel safety and efficacy study of

Lasers Med Sci (2013) 28:1289–1296 1295

2 % pirenzepine ophthalmic gel in children with myopia. J AAPOS12:332–339

15. Siatkowski RM, Cotter S, Miller JM, Scher CA, Crockett RS, NovackGD (2004) Safety and efficacy of 2 % pirenzepine ophthalmic gel inchildren with myopia: a 1-year, multicenter, double-masked, placebo-controlled parallel study. Arch Ophthalmol 122:1667–1674

16. Gwiazda J, Hyman L, Hussein M, Everett D, Norton TT, Kurtz D,Leske MC, Manny R, Marsh-Tootle W, Scheiman M (2003) Arandomized clinical trial of progressive addition lenses versussingle vision lenses on the progression of myopia in children.Invest Ophthalmol Vis Sci 44:1492–1500

17. Katz J, Schein OD, Levy B, Cruiscullo T, Saw SM, Rajan U, ChanTK, Yew Khoo C, Chew SJ (2003) A randomized trial of rigid gaspermeable contact lenses to reduce progression of children's myo-pia. Am J Ophthalmol 136:82–90

18. Mao JR, Bristow J (2001) The Ehlers–Danlos syndrome: on be-yond collagens. J Clin Invest 107:1063–1069

19. Bell GR (1993) Biomechanical considerations of high myopia:part I–physiological characteristics. J Am Optom Assoc 64:332–338

20. Choi S, Cheong Y, Lee HJ, Lee SJ, Jin KH, Park HK (2011) AFMstudy for morphological and material property of human sclerasurface. J Nanosci Nanotechnol 11:6382–6388

21. Terry MA, Ousley PJ (2005) Deep lamellar endothelial keratoplasty?Visual acuity, astigmatism, in a large prospective series. Ophthalmol-ogy 112:1541–1548

22. Jung GB, Lee HJ, Kim JH, Lim JI, Choi S, Jin KH, Park HK(2011) Effect of cross-linking with riboflavin and ultraviolet A onthe chemical bonds and ultrastructure of human sclera. J BiomedOpt 16:125004

1296 Lasers Med Sci (2013) 28:1289–1296