Current Eye Research
ISSN: 0271-3683 (Print) 1460-2202 (Online) Journal homepage: http://www.tandfonline.com/loi/icey20
Biomechanical Changes of Collagen Cross-Linking on Human Keratoconic Corneas Using Scanning Acoustic Microscopy Ithar M. Beshtawi, Riaz Akhtar, M. Chantal Hillarby, Clare O’Donnell, Xuegen Zhao, Arun Brahma, Fiona Carley, Brian Derby & Hema Radhakrishnan To cite this article: Ithar M. Beshtawi, Riaz Akhtar, M. Chantal Hillarby, Clare O’Donnell, Xuegen Zhao, Arun Brahma, Fiona Carley, Brian Derby & Hema Radhakrishnan (2015): Biomechanical Changes of Collagen Cross-Linking on Human Keratoconic Corneas Using Scanning Acoustic Microscopy, Current Eye Research To link to this article: http://dx.doi.org/10.3109/02713683.2015.1042545
Published online: 30 Jun 2015.
Submit your article to this journal
Article views: 20
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=icey20 Download by: [Karolinska Institutet, University Library]
Date: 24 September 2015, At: 12:07
Current Eye Research, Early Online, 1–7, 2015 ! Informa Healthcare USA, Inc. ISSN: 0271-3683 print / 1460-2202 online DOI: 10.3109/02713683.2015.1042545
ORIGINAL ARTICLE
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
Biomechanical Changes of Collagen Cross-Linking on Human Keratoconic Corneas Using Scanning Acoustic Microscopy Ithar M. Beshtawi1, Riaz Akhtar2, M. Chantal Hillarby3, Clare O’Donnell4, Xuegen Zhao5, Arun Brahma6, Fiona Carley6, Brian Derby5, and Hema Radhakrishnan7 1
Optometry Department, Faculty of Medicine and Health Sciences, An-Najah National University, Nablus, State of Palestine, 2Centre for Materials and Structures, School of Engineering, University of Liverpool, Liverpool, UK, 3Stopford Building, Centre For Tissue Injury and Repair, Institute of Inflammation and Repair, University of Manchester, Manchester, UK, 4Optegra Eye Sciences, Optegra Manchester Eye Hospital, Manchester, UK, 5Manchester Materials Science Centre, School of Materials, The University of Manchester, Manchester, UK, 6Manchester Royal Eye Hospital, Central Manchester University Hospitals NHS Foundation Trust, Manchester Academic Health Science Centre, Manchester, UK and 7Faculty of Life Sciences, The University of Manchester, Manchester, UK
ABSTRACT Purpose: To assess the biomechanical changes of collagen cross-linking on keratoconic corneas in vitro. Methods: Six keratoconic corneal buttons were included in this study. Each cornea was divided into two halves, where one half was cross-linked and the other half was treated with riboflavin only and served as control. The biomechanical changes of the corneal tissue were measured across the stroma using scanning acoustic microscopy (SAM). Results: In the cross-linked corneas, there was a steady decrease in the magnitude of speed of sound from the anterior region through to the posterior regions of the stroma. The speed of sound was found to decrease slightly across the corneal thickness in the control corneas. The increase in speed of sound between the crosslinked and control corneas in the anterior region was by a factor of 1.039. Conclusion: A higher speed of sound was detected in cross-linked keratoconic corneal tissue when compared with their controls, using SAM. This in vitro model can be used to compare to the cross-linking results obtained in vivo, as well as comparing the results obtained with different protocols. Keywords: Collagen cross-linking, keratoconus, scanning acoustic microscopy
normal cornea,2 despite the non-significant difference in the total collagen content in the normal and diseased corneas.2 Corneal collagen cross-linking using a combination of Ultraviolet-A (UV-A) and riboflavin as a photosensitizer has been used to treat or halt the progression of keratoconus.3 This treatment aims to increase the corneal biomechanical resistance,
INTRODUCTION Keratoconus is a bilateral non-inflammatory corneal ectasia where reduced biomechanical strength plays an important role in the bulging of the cornea.1 The uniaxial tensile strength of the keratoconic cornea was found to be only 60% of that of the
Received 28 November 2013; revised 14 March 2015; accepted 12 April 2015; published online 29 June 2015 Correspondence: Hema Radhakrishnan, University of Manchester, Faculty of Life Sciences, Moffat Building, P.O. Box 88, Manchester, M60 1QD, UK. E-mail: [email protected]
1
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
2
I. M. Beshtawi et al.
in a photopolymerization process, by inducing new covalent bonds between the collagen molecules and between proteoglycan core proteins.3,4 To evaluate the effects of collagen cross-linking on corneal structure and function, studies have been carried out in-vivo and in-vitro highlighting the clinical, structural, histological and biomechanical effects of the treatment. Additionally, the safety of UV-A irradiation, treatment penetration depth and the efficacy of combined treatment (e.g. with intracorneal rings) have been documented.5 The majority of in-vitro studies that have been carried out have used animal corneas and there have been relatively few which have utilized normal human corneas to assess the reinforcement of corneal tissue after treatment. Stiffness has been reported to be increased after cross-linking, by 330% in human corneas and by 72% in porcine corneas. However in this earlier study the control group was not given a placebo treatment. This might have led to a reduction in stiffness in the control corneas, due to the time that these corneas would have been left in culture medium, as opposed to the treatment corneas.6 In the rabbit cornea, stress measurements increased between 69.7% and 106% in a study that lasted for eight months.7 However, differences between animal and human corneal structure have been documented, e.g. in axial thickness6 and collagen fibre organization making direct comparisons difficult.8 Furthermore, the structure9 and the biomechanical2 properties of the cornea have been shown to be altered in keratoconus, arguably making it difficult to extrapolate the in vitro findings to the keratoconic in-vivo situation. The Young’s modulus values for cross-linked corneas have largely been based on strip-extensometry measurements due to its simplicity.10 Previous studies have addressed the limitations of using the extensometry method to measure the biomechanical properties of corneal strips.11 Some limitations of this approach are that it provides destructive measurements due to cutting the lamellae and that only one strip from each cornea can be used.11 We have previously tested the feasibility of applying SAM on normal human corneas after crosslinking,12 and we have measured the speed of sound of human corneas after low and high intensity cross linking treatments.13 SAM provides non-destructive, pixel-by-pixel, qualitative and quantitative assessment of the corneal biomechanical changes.14 SAM uses ultra-high frequency acoustic waves that propagate deeply within tissue and reflect back with different velocities according to the stiffness of the tested tissue with a high spatial resolution (around 1 mm at 1 GHz).14 This technique has been found to be fast, reliable and accurate in measuring and mapping the biomechanical properties of hard tissues such as teeth15 and bones16 and soft tissues such as blood vessels17 and cornea.12,13
To our knowledge, this study presents for the first time SAM data from keratoconic human corneas following collagen cross-linking treatment.
MATERIALS AND METHODS Corneal Tissue This study was approved by the National Research Ethics Service (NRES) Committee North West – Greater Manchester West (ethics reference no. R01849). The corneas were collected from keratoconic patients who had already decided to undergo fullthickness corneal transplantation at Manchester Royal Eye Hospital, UK. Informed consent was obtained from the keratoconic patients to use their corneal tissue for research before they underwent corneal transplantation surgery. Six keratoconic human corneas were used in this study, from four male and two female donors. Donors were aged between 17 and 37 years. Corneas were removed from the eye and placed in organ culture media18 and cross-linked on the same day.
Cross-Linking Procedure The central 8 mm of the epithelium diameter was removed mechanically by gently scraping with a blade. The cornea was then bisected. One half was assigned to be cross-linked and the other half was treated in the same way without the UVA exposure, and used as a control. Each half was placed on a small amount of healon (1.2% Sodium Hyaluronate, BD Medical, Warwickshire, UK) to maintain corneal stability, and the anterior surface was treated with 1–2 drops of 0.1% riboflavin (10 mg riboflavin-5-phosphate in 20% dextran-T-500 10 mL solution) at three minute intervals for 30 min. In the case of the corneal half assigned for cross-linking, this was followed by a 30 minute exposure to UV-A irradiation (370 nm, 3 mW/cm2) from the double UVA diode of a VEGA CBM X7 LINKERÕ (CSO, Florence, Italy). Further riboflavin was applied at 5 minute intervals during this procedure. The control half simply received the riboflavin treatment. Each corneal half was then rinsed with saline and placed in a fresh bottle of media containing 5% dextran, to limit the corneal swelling, and incubated at 34 C for 24 h prior to assessment.
Tissue Preparation Both the corneal halves were embedded in OCTTM resin (Optimal Cutting Temperature, Tissue-Tek, CellPath, Powys, UK). The OCT corneal blocks were Current Eye Research
SAM of Cross-Linked Keratoconic Corneas In-vitro clamped onto a chuck in a cryostat (The Leica CM3050-S Cryostat, Leica Microsystems) and sectioned at a thickness of 7 mm. Then, the sections were mounted on clean glass slides and stored at 20 C until needed. The corneal sections were then imaged with SAM.
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
Scanning Acoustic Microscopy The principles of SAM testing on soft tissue are explained in detail elsewhere.19 The SAM (SAM200 PVA TePla, Heborn, Germany) utilized in this study is composed of an ultrasonic transducer, an acoustic lens, and an xy-scanner and a z-stage. A highfrequency acoustic wave pulse is generated by the transducer, and then focused by the acoustic lens on to the specimen through coupling fluid (distilled water in this study). The z-stage allows the lenssample distance to be varied at increments down to 0.1 mm. In this study, 0.1 mm increment steps were used14 The lens is scanned horizontally in the x- and y-directions by a pair of oscillator coil drivers to produce a C-mode two-dimensional (2-D) image.19 The fast xy scanner can generate these C-scan 512 512 pixel images with a scan area of 200 200 mm in around 10 s. As described by Beshtawi et al.12, when a thin biological specimen is mounted on a glass substrate and immersed in the acoustic coupling fluid, reflections are generated at each interface in the system: lens/fluid, fluid/specimen and specimen/substrate. Further, Rayleigh waves may radiate along the substrate surface and these leaky Rayleigh waves radiate acoustic energy from the substrate toward the lens. Thus, the signal received at the lens results from the interference between the reflections and the amplitude determined by the intensities and phase of each wave. The MLPA (multi-layer phase analysis)19 method utilized in this study exploits the phase information that is preserved in the interference between the acoustic wave reflected from the substrate surface and internal reflections from the acoustic lens. This information can then be used to extract quantitative data from SAM images. 19 By using MATSAM software (Julius Wolff Institut & Berlin-Brandenburg School for Regenerative Therapies, Berlin, Germany) and MLPA (multi-layer phase analysis),19 the SAM images for the corneal sections were collected and the speed of sound data were extracted from the images using the methodology as described in detail by Zhao et al.19 In this study, the series of 200 200 mm C-scan images were taken at different z-positions starting from 4 mm above the substrate surface of the corneal tissue. The images were processed off-line with custom software using the phase analysis method developed by Zhao et al.19 Briefly, the gray scale value !
2015 Informa Healthcare USA, Inc.
3
for every pixel (x, y position) is extracted from all the images at each z position to form a V(z) curve. The V(z) data are then filtered and processed by Fast Fourier Transformation (FFT).20 A 2D phase array is then obtained for the dataset which is then processed and converted to a speed of sound map.12,13,19 Variations of the speed of sound values were measured in the central part of the corneal section from the anterior stroma towards the posterior stroma by averaging the values within any selected region. Quantitative measurements of the speed of sound were obtained using a series of 1D line profiles so that the variation in speed of sound could be recorded as a function of position. Any artificial gaps created due to the cryosectioning procedure were automatically excluded with our custom software routine. The variation of the speed of sound is plotted using a 0–255 level gray scale. The speed of sound for isotropic materials is proportional to the square root of the Young’s modulus and hence can be considered an indicator of stiffness. A 2D phase array is then obtained for the dataset which is then processed and converted to a speed of sound map as shown in Figure 1.
Statistical Analysis All data are reported as mean ± SE. Kolmogrov– Smirnov test was used to determine whether the data were normally distributed. Data were found to be normally distributed (p40.05). The paired sample t-test was used to test whether the mean values of the measurements are significantly different from each other. Data analysis was conducted using SPSS 19.0 (SPSS, Chicago, IL) and a p value of 50.05 was considered to be statistically significant.
RESULTS Keratoconic corneal tissue sections imaged by SAM are illustrated in Figure 2 which shows images taken from corneal tissue that underwent CXL treatment (Figure 2A and B) and their untreated controls (Figure 2C and D). The speed of sound of the anterior (200 200 mm) region of the cross-linked corneas (Figure 2A) was higher than the posterior part (Figure 2B). Additionally, the speed of sound of the cross-linked halves (Figure 2A and B) was higher than the control group (Figure 2C and D). SAM imaging of the cross-linked and control corneas was done for the central part of the corneal section in both groups. Figure 3A and B shows the local variation in speed of sound across a line profile. The speed of sound of each sample was compared between the anterior (200 200 mm) region and the posterior (200 200 mm) region. In the CXL treatment
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
4
I. M. Beshtawi et al.
FIGURE 1 Speed of sound map for keratoconic corneal section. The brightness of the tissue represents the different speed of sound of the corneal tissue.
FIGURE 2 SAM images of the anterior and posterior 200 200 mm portion of the cross-linked keratoconic tissue (A and B, respectively) and untreated (C and D) cornea tissue sections.
group, the mean ± SE of speed of sound of the crosslinked corneas was 1551.40 ± 7.50 ms 1 anteriorly and 1521.80 ± 6.80 ms 1 posteriorly, which represents an increase of 1.019 x. This increase was found to be statistically significant (p = 0.015). The speed of sound in the control corneas of this group was 1492.80 ± 5.14 ms 1 anteriorly and 1457.92 ± 5.10 ms 1 posteriorly, which represents an increase of 1.0239. The increase
in speed of sound between the cross-linked and control corneas in the anterior 200 200 mm region was by a factor of 1.039. This increase was found to be statistically significant (p = 0.005). Furthermore, the speed of sound across the stromal regions of the cornea was assessed for the CXL and their untreated controls (Figure 4). In the cross-linked group, there was a steady decrease in the magnitude Current Eye Research
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
SAM of Cross-Linked Keratoconic Corneas In-vitro
5
FIGURE 3 Line profile (A) across the cross-linked keratoconic corneal tissue and (B) the corresponding speed of sound variations.
FIGURE 4 Speed of sound measurements of the cross-linked keratoconic tissue and their untreated control from the anterior to the posterior regions of the stroma.
of speed of sound from the anterior region through to the posterior regions of the stroma. The change in the speed of sound was found to fit a cubic function for the CXL corneal halves (y = 0.0018x3 0.1755x2 + 2 0.5598x + 1562.9; R = 0.96915) and for the control !
2015 Informa Healthcare USA, Inc.
halves (y = 0.0084x3 0.2652x2 + 0.5119x + 1455.4; R2 = 0.99531), where y is the speed of sound and x is the corneal depth. Therefore the increase in speed of sound after cross-linking treatment could be calculated at any depth.
6
I. M. Beshtawi et al.
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
DISCUSSION Riboflavin UVA collagen cross-linking is a relatively new promising surgical treatment used to stabilize and strengthen the corneal tissue in keratoconus and other corneal ectasias.3 The new cross-links are induced through the production of oxygen radicals due to activating the riboflavin by UVA. These stiffen the corneal tissue.3 The increase in the corneal biomechanics is thought to improve the visual acuity and topographic outcomes,5 delay the progression of the disease3 and minimize the need for corneal transplantation. Several in-vitro studies reported an improvement in corneal biomechanics after cross-linking on postmortem human12,21 and animal6,7 corneas. In the present study, the increase in the speed of sound in corneal tissue after cross-linking was studied on donor keratoconic corneas from patients who underwent full keratoplasty, using SAM. We used SAM as it provides local analysis of the increase in speed of sound (which is proportional to the square root of stiffness) pixel by pixel.14 This is useful since CXL is thought to occur in the anterior one third of the stromal tissue only.21 The outcomes of this investigation revealed that the speed of sound measurements increased by a factor of 1.039 in the cross-linked corneal tissue when compared with the controls. In other studies where different techniques were used for measuring the increase in corneal stiffness, higher levels of induced stiffness were found with human (1.5) and porcine corneas (1.33)20 and (1.6).21 The differences in results may be due to the different techniques used i.e. strip extensometry as compared to SAM in the present study. Until now, clinical studies have not demonstrated significant changes in corneal biomechanics after crosslinking.22 However, using the Ocular Response Analyser, an increase of 35% was documented in the area under the second curve, but no change was noticed in the corneal hysteresis (CH) and corneal resistance factor (CRF) measurements.23 When comparing the speed of sound values between the anterior and posterior parts of the cross-linked tissue, the increase was by a factor of 1.019 which is lower than what we found previously using normal human postmortem corneas (1.046).13 This may be due to the different corneal thicknesses of the normal and keratoconic corneas. The thinner keratoconic corneas have a reduced proportion of posterior corneal tissue with no cross-linking effects, hence reducing this ratio. In most cases the corneal thickness was less than 400 mm. Additionally, a marginal drop in the speed of sound was recorded around 250 mm in the cross-linked keratoconic corneas, while it was recorded around 200 mm in the cross-linked normal human corneas in our previous study, which suggests more penetration of cross-linking effect
occurs in keratoconic corneas due to different collagen architecture and the differences in keratocyte density. The speed of sound of the control keratoconic tissue was lower than those from the control group (normal human corneas) in our previous study.13 This may be related to the same reasons mentioned above. Differences could also be due to ages of the older postmortem corneas (previous study) and the younger keratoconic corneal donors due to the natural crosslinking that occurs with age.24 Alternative protocols for treating thinner corneas are now available, for safety purposes, such as using hypo-osmolar riboflavin which swells the cornea to at least 400 mm prior to exposing it to UVA irradiation.25 The corneal thickness was not measured prior crosslinking in this study and therefore it could have been lower than 400 mm, therefore, CXL using the isotonic riboflavin and UVA may not be a safe option for these corneas. In a future study, corneal thickness will be measured before CXL, and hypo-osmolar riboflavin will be used when necessary.
ACKNOWLEDGMENTS The authors would like to thank Dr. Sebastian Brand (Fraunhofer Institute of Material Mechanics, Germany) and Professor Kay Raum (Julius Wolff Institut & Berlin-Brandenburg School for Regenerative Therapies, Germany) who developed the MATSAM software utilized in this study.
DECLARATION OF INTEREST Development of the SAM was funded by Wellcome Trust (WT085981AIA). Ithar Beshtawi’s PhD was funded by An-Najah National University, Nablus, Palestine.
REFERENCES 1. Ambekar R, Toussaint KC, Johnson AW. The effect of keratoconus on the structural, mechanical, and optical properties of the cornea. J Mech Behav Biomed Mater 2011; 4:223–236. 2. Andreassen TT, Simonsen AH, Oxlund H. Biomechanical properties of keratoconus and normal corneas. Exp Eye Res 1980;31:435–441. 3. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-ainduced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol 2003;135:620–627. 4. Zhang YT, Conrad AH, Conrad GW. Effects of ultravioletA and riboflavin on the interaction of collagen and proteoglycans during corneal cross-linking. J Biol Chem 2011;286:13011–13022. 5. Saelens IEY, Bartels MC, Bleyen I, Van Rij G. Refractive, topographic, and visual outcomes of same-day corneal cross-linking with ferrara intracorneal ring segments in Current Eye Research
SAM of Cross-Linked Keratoconic Corneas In-vitro
6.
7.
8.
9.
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
10.
11.
12.
13.
14.
15.
!
patients with progressive keratoconus. Cornea 2011;30: 1406–1408. Wollensak G, Spoerl E, Seiler T. Stress-strain measurements of human and porcine corneas after riboflavin-ultravioletA-induced cross-linking. J Cataract Refract Surg 2003;29: 1780–1785. Wollensak G, Iomdina E. Long-term biomechanical properties of rabbit cornea after photodynamic collagen crosslinking. Acta Ophthalmol 2009;87:48–51. Muller LJ, Pels E, Vrensen GF. The specific architecture of the anterior stroma accounts for maintenance of corneal curvature. Br J Ophthalmol 2001;85:437–443. Meek KM, Tuft SJ, Huang Y, Gill PS, Hayes S, Newton RH, et al. Changes in collagen orientation and distribution in keratoconus corneas. Invest Ophthalmol Vis Sci 2005;46: 1948–1956. Beshtawi IM, O’Donnell C, Radhakrishnan H. Biomechanical properties of corneal tissue after ultraviolet-A-riboflavin crosslinking. J Cataract Refract Surg 2013; 39:451–462. Ahearne M, Yang Y, Then KY, Liu KK. Non-destructive mechanical characterisation of UVA/riboflavin crosslinked collagen hydrogels. Br J Ophthalmol 2008;92:268–271. Beshtawi IM, Akhtar R, Hillarby MC, O’Donnell C, Zhao X, Brahma A, et al. Scanning acoustic microscopy for mapping the microelastic properties of human corneal tissue. Curr Eye Res 2013;38:437–444. Beshtawi IM, Akhtar R, Hillarby CM, O’Donnell C, Zhao X, Brahma A, et al. Biomechanical properties of human corneas following low and high intensity collagen cross-linking determined with scanning acoustic microscopy. Invest Ophthalmol Vis Sci 2013;54: 5273–5280. Akhtar R, Sherratt MJ, Watson RE, Kundu T, Derby B. Mapping the micromechanical properties of cryo-sectioned aortic tissue with scanning acoustic microscopy. Mater Res Soc Symp Proc 2009;1132:27262. Raum K, Kempf K, Hein HJ, Schuber J, Maurer P. Preservation of microelastic properties of dentin and tooth enamel in vitro – a scanning acoustic microscopy study. Dent Mater 2007;23:1221–1228.
2015 Informa Healthcare USA, Inc.
7
16. Hasegawa K, Turner CH, Recker RR, Wu E, Burr DB. Elastic properties of osteoporotic bone measured by scanning acoustic microscopy. Bone 1995;16:85–90. 17. Akhtar R, Sherratt MJ, Watson RE, Kundu T, Derby B. Mapping the micromechanical properties of cryo-sectioned aortic tissue with scanning acoustic microscopy. Mater Res Soc Symp Proc 2009;1132E:1132-Z1103-1107. 18. Armitage WJ, Easty DL. Factors influencing the suitability of organ-cultured corneas for transplantation. Invest Ophthalmol Vis Sci 1997;38:16–24. 19. Zhao XG, Akhtar R, Nijenhuis N, Wilkinson SJ, Murphy L, Ballestrem C, et al. Multi-layer phase analysis: quantifying the elastic properties of soft tissues and live cells with ultra-high-frequency scanning acoustic microscopy. IEEE Trans Ultrason Ferroelectr Freq Control 2012;59:610–620. 20. Schumacher S, Oeftiger L, Mrochen M. Equivalence of biomechanical changes induced by rapid and standard corneal cross-linking, using riboflavin and ultraviolet radiation. Invest Ophthalmol Vis Sci 2011;52:9048–9052. 21. Kohlhaas M, Spoerl E, Schilde T, Unger G, Wittig C, Pillunat LE. Biomechanical evidence of the distribution of cross-links in corneas treated with riboflavin and ultraviolet A light. J Cataract Refract Surg 2006;32:279–283. 22. Sedaghat M, Naderi M, Zarei-Ghanavati M. Biomechanical parameters of the cornea after collagen crosslinking measured by waveform analysis. J Cataract Refract Surg 2010;36:1728–1731. 23. Spoerl E, Terai N, Scholz F, Raiskup F, Pillunat LE. Detection of biomechanical changes after corneal crosslinking using Ocular Response Analyzer software. J Refract Surg 2011;27:452–457. 24. Elsheikh A, Geraghty B, Rama P, Campanelli M, Meek KM. Characterization of age-related variation in corneal biomechanical properties. J R Soc Interface 2010;7:1475–1485. 25. Hafezi F, Mrochen M, Iseli HP, Seiler T. Collagen crosslinking with ultraviolet-A and hypoosmolar riboflavin solution in thin corneas. J Cataract Refract Surg 2009;35: 621–624. 26. Luce DA. Determining in vivo biomechanical properties of the cornea with an ocular response analyzer. J Cataract Refrac Surg 2005;31:156–162.
ISSN: 0271-3683 (Print) 1460-2202 (Online) Journal homepage: http://www.tandfonline.com/loi/icey20
Biomechanical Changes of Collagen Cross-Linking on Human Keratoconic Corneas Using Scanning Acoustic Microscopy Ithar M. Beshtawi, Riaz Akhtar, M. Chantal Hillarby, Clare O’Donnell, Xuegen Zhao, Arun Brahma, Fiona Carley, Brian Derby & Hema Radhakrishnan To cite this article: Ithar M. Beshtawi, Riaz Akhtar, M. Chantal Hillarby, Clare O’Donnell, Xuegen Zhao, Arun Brahma, Fiona Carley, Brian Derby & Hema Radhakrishnan (2015): Biomechanical Changes of Collagen Cross-Linking on Human Keratoconic Corneas Using Scanning Acoustic Microscopy, Current Eye Research To link to this article: http://dx.doi.org/10.3109/02713683.2015.1042545
Published online: 30 Jun 2015.
Submit your article to this journal
Article views: 20
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=icey20 Download by: [Karolinska Institutet, University Library]
Date: 24 September 2015, At: 12:07
Current Eye Research, Early Online, 1–7, 2015 ! Informa Healthcare USA, Inc. ISSN: 0271-3683 print / 1460-2202 online DOI: 10.3109/02713683.2015.1042545
ORIGINAL ARTICLE
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
Biomechanical Changes of Collagen Cross-Linking on Human Keratoconic Corneas Using Scanning Acoustic Microscopy Ithar M. Beshtawi1, Riaz Akhtar2, M. Chantal Hillarby3, Clare O’Donnell4, Xuegen Zhao5, Arun Brahma6, Fiona Carley6, Brian Derby5, and Hema Radhakrishnan7 1
Optometry Department, Faculty of Medicine and Health Sciences, An-Najah National University, Nablus, State of Palestine, 2Centre for Materials and Structures, School of Engineering, University of Liverpool, Liverpool, UK, 3Stopford Building, Centre For Tissue Injury and Repair, Institute of Inflammation and Repair, University of Manchester, Manchester, UK, 4Optegra Eye Sciences, Optegra Manchester Eye Hospital, Manchester, UK, 5Manchester Materials Science Centre, School of Materials, The University of Manchester, Manchester, UK, 6Manchester Royal Eye Hospital, Central Manchester University Hospitals NHS Foundation Trust, Manchester Academic Health Science Centre, Manchester, UK and 7Faculty of Life Sciences, The University of Manchester, Manchester, UK
ABSTRACT Purpose: To assess the biomechanical changes of collagen cross-linking on keratoconic corneas in vitro. Methods: Six keratoconic corneal buttons were included in this study. Each cornea was divided into two halves, where one half was cross-linked and the other half was treated with riboflavin only and served as control. The biomechanical changes of the corneal tissue were measured across the stroma using scanning acoustic microscopy (SAM). Results: In the cross-linked corneas, there was a steady decrease in the magnitude of speed of sound from the anterior region through to the posterior regions of the stroma. The speed of sound was found to decrease slightly across the corneal thickness in the control corneas. The increase in speed of sound between the crosslinked and control corneas in the anterior region was by a factor of 1.039. Conclusion: A higher speed of sound was detected in cross-linked keratoconic corneal tissue when compared with their controls, using SAM. This in vitro model can be used to compare to the cross-linking results obtained in vivo, as well as comparing the results obtained with different protocols. Keywords: Collagen cross-linking, keratoconus, scanning acoustic microscopy
normal cornea,2 despite the non-significant difference in the total collagen content in the normal and diseased corneas.2 Corneal collagen cross-linking using a combination of Ultraviolet-A (UV-A) and riboflavin as a photosensitizer has been used to treat or halt the progression of keratoconus.3 This treatment aims to increase the corneal biomechanical resistance,
INTRODUCTION Keratoconus is a bilateral non-inflammatory corneal ectasia where reduced biomechanical strength plays an important role in the bulging of the cornea.1 The uniaxial tensile strength of the keratoconic cornea was found to be only 60% of that of the
Received 28 November 2013; revised 14 March 2015; accepted 12 April 2015; published online 29 June 2015 Correspondence: Hema Radhakrishnan, University of Manchester, Faculty of Life Sciences, Moffat Building, P.O. Box 88, Manchester, M60 1QD, UK. E-mail: [email protected]
1
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
2
I. M. Beshtawi et al.
in a photopolymerization process, by inducing new covalent bonds between the collagen molecules and between proteoglycan core proteins.3,4 To evaluate the effects of collagen cross-linking on corneal structure and function, studies have been carried out in-vivo and in-vitro highlighting the clinical, structural, histological and biomechanical effects of the treatment. Additionally, the safety of UV-A irradiation, treatment penetration depth and the efficacy of combined treatment (e.g. with intracorneal rings) have been documented.5 The majority of in-vitro studies that have been carried out have used animal corneas and there have been relatively few which have utilized normal human corneas to assess the reinforcement of corneal tissue after treatment. Stiffness has been reported to be increased after cross-linking, by 330% in human corneas and by 72% in porcine corneas. However in this earlier study the control group was not given a placebo treatment. This might have led to a reduction in stiffness in the control corneas, due to the time that these corneas would have been left in culture medium, as opposed to the treatment corneas.6 In the rabbit cornea, stress measurements increased between 69.7% and 106% in a study that lasted for eight months.7 However, differences between animal and human corneal structure have been documented, e.g. in axial thickness6 and collagen fibre organization making direct comparisons difficult.8 Furthermore, the structure9 and the biomechanical2 properties of the cornea have been shown to be altered in keratoconus, arguably making it difficult to extrapolate the in vitro findings to the keratoconic in-vivo situation. The Young’s modulus values for cross-linked corneas have largely been based on strip-extensometry measurements due to its simplicity.10 Previous studies have addressed the limitations of using the extensometry method to measure the biomechanical properties of corneal strips.11 Some limitations of this approach are that it provides destructive measurements due to cutting the lamellae and that only one strip from each cornea can be used.11 We have previously tested the feasibility of applying SAM on normal human corneas after crosslinking,12 and we have measured the speed of sound of human corneas after low and high intensity cross linking treatments.13 SAM provides non-destructive, pixel-by-pixel, qualitative and quantitative assessment of the corneal biomechanical changes.14 SAM uses ultra-high frequency acoustic waves that propagate deeply within tissue and reflect back with different velocities according to the stiffness of the tested tissue with a high spatial resolution (around 1 mm at 1 GHz).14 This technique has been found to be fast, reliable and accurate in measuring and mapping the biomechanical properties of hard tissues such as teeth15 and bones16 and soft tissues such as blood vessels17 and cornea.12,13
To our knowledge, this study presents for the first time SAM data from keratoconic human corneas following collagen cross-linking treatment.
MATERIALS AND METHODS Corneal Tissue This study was approved by the National Research Ethics Service (NRES) Committee North West – Greater Manchester West (ethics reference no. R01849). The corneas were collected from keratoconic patients who had already decided to undergo fullthickness corneal transplantation at Manchester Royal Eye Hospital, UK. Informed consent was obtained from the keratoconic patients to use their corneal tissue for research before they underwent corneal transplantation surgery. Six keratoconic human corneas were used in this study, from four male and two female donors. Donors were aged between 17 and 37 years. Corneas were removed from the eye and placed in organ culture media18 and cross-linked on the same day.
Cross-Linking Procedure The central 8 mm of the epithelium diameter was removed mechanically by gently scraping with a blade. The cornea was then bisected. One half was assigned to be cross-linked and the other half was treated in the same way without the UVA exposure, and used as a control. Each half was placed on a small amount of healon (1.2% Sodium Hyaluronate, BD Medical, Warwickshire, UK) to maintain corneal stability, and the anterior surface was treated with 1–2 drops of 0.1% riboflavin (10 mg riboflavin-5-phosphate in 20% dextran-T-500 10 mL solution) at three minute intervals for 30 min. In the case of the corneal half assigned for cross-linking, this was followed by a 30 minute exposure to UV-A irradiation (370 nm, 3 mW/cm2) from the double UVA diode of a VEGA CBM X7 LINKERÕ (CSO, Florence, Italy). Further riboflavin was applied at 5 minute intervals during this procedure. The control half simply received the riboflavin treatment. Each corneal half was then rinsed with saline and placed in a fresh bottle of media containing 5% dextran, to limit the corneal swelling, and incubated at 34 C for 24 h prior to assessment.
Tissue Preparation Both the corneal halves were embedded in OCTTM resin (Optimal Cutting Temperature, Tissue-Tek, CellPath, Powys, UK). The OCT corneal blocks were Current Eye Research
SAM of Cross-Linked Keratoconic Corneas In-vitro clamped onto a chuck in a cryostat (The Leica CM3050-S Cryostat, Leica Microsystems) and sectioned at a thickness of 7 mm. Then, the sections were mounted on clean glass slides and stored at 20 C until needed. The corneal sections were then imaged with SAM.
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
Scanning Acoustic Microscopy The principles of SAM testing on soft tissue are explained in detail elsewhere.19 The SAM (SAM200 PVA TePla, Heborn, Germany) utilized in this study is composed of an ultrasonic transducer, an acoustic lens, and an xy-scanner and a z-stage. A highfrequency acoustic wave pulse is generated by the transducer, and then focused by the acoustic lens on to the specimen through coupling fluid (distilled water in this study). The z-stage allows the lenssample distance to be varied at increments down to 0.1 mm. In this study, 0.1 mm increment steps were used14 The lens is scanned horizontally in the x- and y-directions by a pair of oscillator coil drivers to produce a C-mode two-dimensional (2-D) image.19 The fast xy scanner can generate these C-scan 512 512 pixel images with a scan area of 200 200 mm in around 10 s. As described by Beshtawi et al.12, when a thin biological specimen is mounted on a glass substrate and immersed in the acoustic coupling fluid, reflections are generated at each interface in the system: lens/fluid, fluid/specimen and specimen/substrate. Further, Rayleigh waves may radiate along the substrate surface and these leaky Rayleigh waves radiate acoustic energy from the substrate toward the lens. Thus, the signal received at the lens results from the interference between the reflections and the amplitude determined by the intensities and phase of each wave. The MLPA (multi-layer phase analysis)19 method utilized in this study exploits the phase information that is preserved in the interference between the acoustic wave reflected from the substrate surface and internal reflections from the acoustic lens. This information can then be used to extract quantitative data from SAM images. 19 By using MATSAM software (Julius Wolff Institut & Berlin-Brandenburg School for Regenerative Therapies, Berlin, Germany) and MLPA (multi-layer phase analysis),19 the SAM images for the corneal sections were collected and the speed of sound data were extracted from the images using the methodology as described in detail by Zhao et al.19 In this study, the series of 200 200 mm C-scan images were taken at different z-positions starting from 4 mm above the substrate surface of the corneal tissue. The images were processed off-line with custom software using the phase analysis method developed by Zhao et al.19 Briefly, the gray scale value !
2015 Informa Healthcare USA, Inc.
3
for every pixel (x, y position) is extracted from all the images at each z position to form a V(z) curve. The V(z) data are then filtered and processed by Fast Fourier Transformation (FFT).20 A 2D phase array is then obtained for the dataset which is then processed and converted to a speed of sound map.12,13,19 Variations of the speed of sound values were measured in the central part of the corneal section from the anterior stroma towards the posterior stroma by averaging the values within any selected region. Quantitative measurements of the speed of sound were obtained using a series of 1D line profiles so that the variation in speed of sound could be recorded as a function of position. Any artificial gaps created due to the cryosectioning procedure were automatically excluded with our custom software routine. The variation of the speed of sound is plotted using a 0–255 level gray scale. The speed of sound for isotropic materials is proportional to the square root of the Young’s modulus and hence can be considered an indicator of stiffness. A 2D phase array is then obtained for the dataset which is then processed and converted to a speed of sound map as shown in Figure 1.
Statistical Analysis All data are reported as mean ± SE. Kolmogrov– Smirnov test was used to determine whether the data were normally distributed. Data were found to be normally distributed (p40.05). The paired sample t-test was used to test whether the mean values of the measurements are significantly different from each other. Data analysis was conducted using SPSS 19.0 (SPSS, Chicago, IL) and a p value of 50.05 was considered to be statistically significant.
RESULTS Keratoconic corneal tissue sections imaged by SAM are illustrated in Figure 2 which shows images taken from corneal tissue that underwent CXL treatment (Figure 2A and B) and their untreated controls (Figure 2C and D). The speed of sound of the anterior (200 200 mm) region of the cross-linked corneas (Figure 2A) was higher than the posterior part (Figure 2B). Additionally, the speed of sound of the cross-linked halves (Figure 2A and B) was higher than the control group (Figure 2C and D). SAM imaging of the cross-linked and control corneas was done for the central part of the corneal section in both groups. Figure 3A and B shows the local variation in speed of sound across a line profile. The speed of sound of each sample was compared between the anterior (200 200 mm) region and the posterior (200 200 mm) region. In the CXL treatment
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
4
I. M. Beshtawi et al.
FIGURE 1 Speed of sound map for keratoconic corneal section. The brightness of the tissue represents the different speed of sound of the corneal tissue.
FIGURE 2 SAM images of the anterior and posterior 200 200 mm portion of the cross-linked keratoconic tissue (A and B, respectively) and untreated (C and D) cornea tissue sections.
group, the mean ± SE of speed of sound of the crosslinked corneas was 1551.40 ± 7.50 ms 1 anteriorly and 1521.80 ± 6.80 ms 1 posteriorly, which represents an increase of 1.019 x. This increase was found to be statistically significant (p = 0.015). The speed of sound in the control corneas of this group was 1492.80 ± 5.14 ms 1 anteriorly and 1457.92 ± 5.10 ms 1 posteriorly, which represents an increase of 1.0239. The increase
in speed of sound between the cross-linked and control corneas in the anterior 200 200 mm region was by a factor of 1.039. This increase was found to be statistically significant (p = 0.005). Furthermore, the speed of sound across the stromal regions of the cornea was assessed for the CXL and their untreated controls (Figure 4). In the cross-linked group, there was a steady decrease in the magnitude Current Eye Research
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
SAM of Cross-Linked Keratoconic Corneas In-vitro
5
FIGURE 3 Line profile (A) across the cross-linked keratoconic corneal tissue and (B) the corresponding speed of sound variations.
FIGURE 4 Speed of sound measurements of the cross-linked keratoconic tissue and their untreated control from the anterior to the posterior regions of the stroma.
of speed of sound from the anterior region through to the posterior regions of the stroma. The change in the speed of sound was found to fit a cubic function for the CXL corneal halves (y = 0.0018x3 0.1755x2 + 2 0.5598x + 1562.9; R = 0.96915) and for the control !
2015 Informa Healthcare USA, Inc.
halves (y = 0.0084x3 0.2652x2 + 0.5119x + 1455.4; R2 = 0.99531), where y is the speed of sound and x is the corneal depth. Therefore the increase in speed of sound after cross-linking treatment could be calculated at any depth.
6
I. M. Beshtawi et al.
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
DISCUSSION Riboflavin UVA collagen cross-linking is a relatively new promising surgical treatment used to stabilize and strengthen the corneal tissue in keratoconus and other corneal ectasias.3 The new cross-links are induced through the production of oxygen radicals due to activating the riboflavin by UVA. These stiffen the corneal tissue.3 The increase in the corneal biomechanics is thought to improve the visual acuity and topographic outcomes,5 delay the progression of the disease3 and minimize the need for corneal transplantation. Several in-vitro studies reported an improvement in corneal biomechanics after cross-linking on postmortem human12,21 and animal6,7 corneas. In the present study, the increase in the speed of sound in corneal tissue after cross-linking was studied on donor keratoconic corneas from patients who underwent full keratoplasty, using SAM. We used SAM as it provides local analysis of the increase in speed of sound (which is proportional to the square root of stiffness) pixel by pixel.14 This is useful since CXL is thought to occur in the anterior one third of the stromal tissue only.21 The outcomes of this investigation revealed that the speed of sound measurements increased by a factor of 1.039 in the cross-linked corneal tissue when compared with the controls. In other studies where different techniques were used for measuring the increase in corneal stiffness, higher levels of induced stiffness were found with human (1.5) and porcine corneas (1.33)20 and (1.6).21 The differences in results may be due to the different techniques used i.e. strip extensometry as compared to SAM in the present study. Until now, clinical studies have not demonstrated significant changes in corneal biomechanics after crosslinking.22 However, using the Ocular Response Analyser, an increase of 35% was documented in the area under the second curve, but no change was noticed in the corneal hysteresis (CH) and corneal resistance factor (CRF) measurements.23 When comparing the speed of sound values between the anterior and posterior parts of the cross-linked tissue, the increase was by a factor of 1.019 which is lower than what we found previously using normal human postmortem corneas (1.046).13 This may be due to the different corneal thicknesses of the normal and keratoconic corneas. The thinner keratoconic corneas have a reduced proportion of posterior corneal tissue with no cross-linking effects, hence reducing this ratio. In most cases the corneal thickness was less than 400 mm. Additionally, a marginal drop in the speed of sound was recorded around 250 mm in the cross-linked keratoconic corneas, while it was recorded around 200 mm in the cross-linked normal human corneas in our previous study, which suggests more penetration of cross-linking effect
occurs in keratoconic corneas due to different collagen architecture and the differences in keratocyte density. The speed of sound of the control keratoconic tissue was lower than those from the control group (normal human corneas) in our previous study.13 This may be related to the same reasons mentioned above. Differences could also be due to ages of the older postmortem corneas (previous study) and the younger keratoconic corneal donors due to the natural crosslinking that occurs with age.24 Alternative protocols for treating thinner corneas are now available, for safety purposes, such as using hypo-osmolar riboflavin which swells the cornea to at least 400 mm prior to exposing it to UVA irradiation.25 The corneal thickness was not measured prior crosslinking in this study and therefore it could have been lower than 400 mm, therefore, CXL using the isotonic riboflavin and UVA may not be a safe option for these corneas. In a future study, corneal thickness will be measured before CXL, and hypo-osmolar riboflavin will be used when necessary.
ACKNOWLEDGMENTS The authors would like to thank Dr. Sebastian Brand (Fraunhofer Institute of Material Mechanics, Germany) and Professor Kay Raum (Julius Wolff Institut & Berlin-Brandenburg School for Regenerative Therapies, Germany) who developed the MATSAM software utilized in this study.
DECLARATION OF INTEREST Development of the SAM was funded by Wellcome Trust (WT085981AIA). Ithar Beshtawi’s PhD was funded by An-Najah National University, Nablus, Palestine.
REFERENCES 1. Ambekar R, Toussaint KC, Johnson AW. The effect of keratoconus on the structural, mechanical, and optical properties of the cornea. J Mech Behav Biomed Mater 2011; 4:223–236. 2. Andreassen TT, Simonsen AH, Oxlund H. Biomechanical properties of keratoconus and normal corneas. Exp Eye Res 1980;31:435–441. 3. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-ainduced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol 2003;135:620–627. 4. Zhang YT, Conrad AH, Conrad GW. Effects of ultravioletA and riboflavin on the interaction of collagen and proteoglycans during corneal cross-linking. J Biol Chem 2011;286:13011–13022. 5. Saelens IEY, Bartels MC, Bleyen I, Van Rij G. Refractive, topographic, and visual outcomes of same-day corneal cross-linking with ferrara intracorneal ring segments in Current Eye Research
SAM of Cross-Linked Keratoconic Corneas In-vitro
6.
7.
8.
9.
Downloaded by [Karolinska Institutet, University Library] at 12:07 24 September 2015
10.
11.
12.
13.
14.
15.
!
patients with progressive keratoconus. Cornea 2011;30: 1406–1408. Wollensak G, Spoerl E, Seiler T. Stress-strain measurements of human and porcine corneas after riboflavin-ultravioletA-induced cross-linking. J Cataract Refract Surg 2003;29: 1780–1785. Wollensak G, Iomdina E. Long-term biomechanical properties of rabbit cornea after photodynamic collagen crosslinking. Acta Ophthalmol 2009;87:48–51. Muller LJ, Pels E, Vrensen GF. The specific architecture of the anterior stroma accounts for maintenance of corneal curvature. Br J Ophthalmol 2001;85:437–443. Meek KM, Tuft SJ, Huang Y, Gill PS, Hayes S, Newton RH, et al. Changes in collagen orientation and distribution in keratoconus corneas. Invest Ophthalmol Vis Sci 2005;46: 1948–1956. Beshtawi IM, O’Donnell C, Radhakrishnan H. Biomechanical properties of corneal tissue after ultraviolet-A-riboflavin crosslinking. J Cataract Refract Surg 2013; 39:451–462. Ahearne M, Yang Y, Then KY, Liu KK. Non-destructive mechanical characterisation of UVA/riboflavin crosslinked collagen hydrogels. Br J Ophthalmol 2008;92:268–271. Beshtawi IM, Akhtar R, Hillarby MC, O’Donnell C, Zhao X, Brahma A, et al. Scanning acoustic microscopy for mapping the microelastic properties of human corneal tissue. Curr Eye Res 2013;38:437–444. Beshtawi IM, Akhtar R, Hillarby CM, O’Donnell C, Zhao X, Brahma A, et al. Biomechanical properties of human corneas following low and high intensity collagen cross-linking determined with scanning acoustic microscopy. Invest Ophthalmol Vis Sci 2013;54: 5273–5280. Akhtar R, Sherratt MJ, Watson RE, Kundu T, Derby B. Mapping the micromechanical properties of cryo-sectioned aortic tissue with scanning acoustic microscopy. Mater Res Soc Symp Proc 2009;1132:27262. Raum K, Kempf K, Hein HJ, Schuber J, Maurer P. Preservation of microelastic properties of dentin and tooth enamel in vitro – a scanning acoustic microscopy study. Dent Mater 2007;23:1221–1228.
2015 Informa Healthcare USA, Inc.
7
16. Hasegawa K, Turner CH, Recker RR, Wu E, Burr DB. Elastic properties of osteoporotic bone measured by scanning acoustic microscopy. Bone 1995;16:85–90. 17. Akhtar R, Sherratt MJ, Watson RE, Kundu T, Derby B. Mapping the micromechanical properties of cryo-sectioned aortic tissue with scanning acoustic microscopy. Mater Res Soc Symp Proc 2009;1132E:1132-Z1103-1107. 18. Armitage WJ, Easty DL. Factors influencing the suitability of organ-cultured corneas for transplantation. Invest Ophthalmol Vis Sci 1997;38:16–24. 19. Zhao XG, Akhtar R, Nijenhuis N, Wilkinson SJ, Murphy L, Ballestrem C, et al. Multi-layer phase analysis: quantifying the elastic properties of soft tissues and live cells with ultra-high-frequency scanning acoustic microscopy. IEEE Trans Ultrason Ferroelectr Freq Control 2012;59:610–620. 20. Schumacher S, Oeftiger L, Mrochen M. Equivalence of biomechanical changes induced by rapid and standard corneal cross-linking, using riboflavin and ultraviolet radiation. Invest Ophthalmol Vis Sci 2011;52:9048–9052. 21. Kohlhaas M, Spoerl E, Schilde T, Unger G, Wittig C, Pillunat LE. Biomechanical evidence of the distribution of cross-links in corneas treated with riboflavin and ultraviolet A light. J Cataract Refract Surg 2006;32:279–283. 22. Sedaghat M, Naderi M, Zarei-Ghanavati M. Biomechanical parameters of the cornea after collagen crosslinking measured by waveform analysis. J Cataract Refract Surg 2010;36:1728–1731. 23. Spoerl E, Terai N, Scholz F, Raiskup F, Pillunat LE. Detection of biomechanical changes after corneal crosslinking using Ocular Response Analyzer software. J Refract Surg 2011;27:452–457. 24. Elsheikh A, Geraghty B, Rama P, Campanelli M, Meek KM. Characterization of age-related variation in corneal biomechanical properties. J R Soc Interface 2010;7:1475–1485. 25. Hafezi F, Mrochen M, Iseli HP, Seiler T. Collagen crosslinking with ultraviolet-A and hypoosmolar riboflavin solution in thin corneas. J Cataract Refract Surg 2009;35: 621–624. 26. Luce DA. Determining in vivo biomechanical properties of the cornea with an ocular response analyzer. J Cataract Refrac Surg 2005;31:156–162.
