|Year : 2018 | Volume
| Issue : 4 | Page : 228-236
Standard versus transepithelial collagen cross-linking in the management of keratoconus
Abd El-Magid M Tag El-Din1, Magdy E.E Tawakol1, Amr M.M Sheta2
1 Department of Ophthalmology, Al Azhar faculty of Medicin, Al Zagazig University, Cairo, Egypt
2 Professor of ophthalmology, ophthalmology department, Al Azhar faculty of Medicin, Al Zagazig University, Cairo, Egypt
|Date of Submission||16-Jun-2018|
|Date of Acceptance||21-Sep-2018|
|Date of Web Publication||20-Dec-2018|
Abd El-Magid M Tag El-Din
Flat 602, Building 13, Omar Zaafan Street, First District, Nasr City, Cairo 11765
Source of Support: None, Conflict of Interest: None
Objective The aim of this study was to compare corneal collagen cross-linking (CXL) using either standard or transepithelial CXL.
Patients and methods The study included 30 eyes of 18 patients with a mean age of 27.08 years. They were seven women and 11 men. Eyes were divided into two groups. Group I: included 15 eyes that underwent standard CXL after removal of the epithelium. Group II: included 15 eyes that underwent CXL with the epithelium intact. The patients were followed up postoperatively at 1 week, 1 month, 3 months, and 6 months for their uncorrected visual acuity (UCVA), best spectacle-corrected visual acuity (BSCVA), pentacam, ocular response analyzer, and specular microscopy.
Results The preoperative mean keratometry (K) max was 47.9 diopters (D). Preoperative mean astigmatism was −3.75 D and mean pachymetry of the corneal center was 467.1 µm. The mean preoperative UCVA for group I was 0.118±0.05 and the mean postoperative UCVA at the 6-month visit was 0.26±0.18. In group II the mean preoperative UCVA was 0.278±0.14 and the postoperative UCVA at 6 months was 0.408±0.3. The mean preoperative BSCVA for group I was 0.218±0.204, while the mean postoperative BSCVA at the 6-month visit was 0.48±0.18. The mean preoperative BSCVA for group II was 0.6±0.305 while the mean postoperative BSCVA at 6 months was 0.745±0.21, There was a statistically significant difference between the two groups regarding UCVA improvement. However, there was a statistically nonsignificant difference between the two groups regarding BSCVA improvement. Comparing the change in mean spherical equivalent between the two groups, it was found statistically significant (P=0.04). On the other hand, comparing that of corneal astigmatism, it was statistically nonsignificant. Pentacam revealed a mean preoperative Kmax of 48.06±5.05 D. The mean postoperative Kmax at 1 month was 47.1±5.6 D. The mean postoperative Kmax at 3 months was 45.75±5.1 D and at 6 months was 45.725±4.49 D. There was a statistically significant difference between mean preoperative and mean postoperative Kmax. A nonsignificant corneal thinning with a mean pachymetry of 433.78±46.71 µm was observed.
Conclusion There was significant difference between standard and transepithelial cross-linking with better results with the standard cross-linking, but with satisfactory results of both methods.
Keywords: collagen cross-linking, ectasia, keratoconus
|How to cite this article:|
Tag El-Din AMM, Tawakol ME, Sheta AM. Standard versus transepithelial collagen cross-linking in the management of keratoconus. Delta J Ophthalmol 2018;19:228-36
|How to cite this URL:|
Tag El-Din AMM, Tawakol ME, Sheta AM. Standard versus transepithelial collagen cross-linking in the management of keratoconus. Delta J Ophthalmol [serial online] 2018 [cited 2021 Dec 3];19:228-36. Available from: http://www.djo.eg.net/text.asp?2018/19/4/228/248087
| Introduction|| |
Keratectasia is a condition of progressive thinning and steepening of the cornea, which results in the production of myopia and irregular astigmatism. The most common ectatic disorders of the cornea associated with corneal biomechanical instability are keratoconus, pellucid marginal degeneration and, especially in the last decade, postlaser in-situ keratomileusis ectasia ,.
Therapeutic solutions proposed for the management of these ectatic disorders are spectacle correction, rigid gas-permeable contact lenses, and intrastromal corneal ring segments, in order to achieve adequate visual rehabilitation. If the disorder continues to advance, lamellar or penetrating keratoplasty are considered the ultimate treatment options .
Interest in induction of corneal collagen cross-linking (CXL) by riboflavin and ultraviolet A (UVA) light has grown steadily since. Wollensak et al.  first used it successfully to treat keratoconus in the early 21st century. New findings even suggest that a variety of other conditions can be treated with this relatively simple and inexpensive procedure as pellucid marginal degeneration and postlaser in-situ keratomileusis ectasia .
Cross-linking of human collagen is a physiological process. CXL is a new approach to increase the mechanical and chemical stability of the corneal tissue. The primary aim of this treatment was to create additional chemical bonds inside the corneal stroma by means of highly localized photopolymerization while minimizing exposure to the surrounding structures of the eye. The mechanism works by the action of free radicals and the end result is an increase in covalent bonding between collagen molecules in the fibrils and also between collagen fibrils. This results in several biomechanical, biochemical, and cellular changes in the corneal stroma such as increased corneal stromal stiffness, increase in the collagen fibril diameter, more resistance to enzymatic digestion and more resistance to hydrothermal shrinkage denoting greater stability ,.
While the conventional treatment options such as glasses and hard contact lenses mostly address the patient’s visual acuity, CXL addresses the pathology itself. CXL is the only treatment aiming at slowing the progression of corneal ectasia and thus reducing the demand for penetrating keratoplasty which is a far more invasive procedure ,.
The aim of this study was to compare corneal biomechanics, endothelial count, visual refractive outcomes, and pentacam findings following CXL using either standard or transepithelial CXL.
| Patients and methods|| |
This is a single-center prospective interventional study. It was conducted between March 2016 and December 2016 and included 30 eyes of 18 patients (seven women and 11 men) with a mean age of 27.08 years (range: 17–30 years). They were 13 female eyes and 17 male eyes ([Table 1]). All eyes were subjected to CXL using riboflavin and UVA light at Kobry Al Kobba Military Specialized Eye Hospital. The study was approved by the Local Ethics Committee of Al Azhar University. A written informed consent was signed by all participants in the study.
The patients were subdivided into two groups:
- Group I: included 15 eyes that underwent CXL after removal of the epithelium (epi-off).
- Group II: included 15 eyes that underwent CXL with the epithelium intact (epi-on).
Progressive keratoconus in cases over 20 years of age, keratoconus in patients under 20 years of age (Amsler-Krumeich grades I and II), pachymetric reading greater than 400 µm at the thinnest point of the cornea as measured by Pentacam and steep keratometry (Km) reading under 60.0 D as measured by Pentacam.
Patients who do not fulfill the inclusion criteria above, any other previous surgical interference or pervious CXL, corneal opacity of any kind, nursing or pregnant patients, age over 35 years, background systemic disease such as diabetes mellitus or collagen vascular disease and patients who refuse to sign the consent.
Following detailed medical and ophthalmic history, a complete ophthalmic examination was performed including: uncorrected and best corrected visual acuity, anterior segment examination by slit lamp, posterior segment examination using a +20 D lens for indirect ophthalmoscopy, a +90 D lens for slit lamp fundus biomicroscopy, and intraocular pressure (IOP) measurement by Goldmann applanation tonometer.
All patients were assessed for corneal curvature and astigmatism as follows.
- Refraction: objective refraction by Topcon automated refractometer (Topcon Medical System, Oakland, New Jersey, USA).
- Pentacam: Pentacam (Oculus, Wetzlar, Germany) was used to assess patients’ corneal topography, keratometry, and pachymetry. The value of K1, K2, and Kmax were determined together with the size, site, and centralization of the corneal cone. The amount of corneal astigmatism was measured. Notation of the astigmatism was done using the difference between the steep and flat corneal meridia. The corneal thickness was measured and the thickest and thinnest locations were determined.
All patients were assessed for corneal biomechanics as follows.
- Ocular response analyzer (ORA): corneal hysteresis (CH) and corneal resistance factor (CRF) were measured for each eye using the Reichert ORA (Reichert Corp., Depew, New York, USA).
- Specular microscopy: counting endothelial cells was performed by Konan Cellcheck apparatus (Konan Medical USA Inc., Irvine, Califronia, USA).
Group I: epi-off
This group included 15 eyes of 12 patients, seven eyes of male patients and eight eyes of female patients. The preoperative Kmax ranged between 42 and 51.6 D and astigmatism ranged between −1.2 and −8.0 D.
All eyes underwent photooxidative CXL using riboflavin and UV-A light after epithelial debridement of the central 8–9 mm of the cornea (epi-off). The same CXL machine (UV-X illumination system version 1000; IROC innocross AG, Bahnhofstrasse, Switzerland) was used in all cases.
- Anesthesia: all cases were conducted under topical anesthesia (Benoxinate HCL drops, Eipico, Cairo, Egypt) that was instilled twice for 2 min before the procedure.
- Operative details: after applying the eyelid speculum, an 8-mm diameter marker was used to mark the corneal epithelium in a central circle. Epithelium was removed in the central 8–9 mm with a blunt metal spatula, de-epithelialization was followed by instillation of riboflavin (0.1% solution 10 mg riboflavin-5-phosphate in 10 ml Dextran-T-500 20% solution; Avedro Inc., Waltham, Massachusetts, USA) every 2 min for 20–30 min until the stroma was completely filled with riboflavin. After confirming the presence of riboflavin in the corneal tissue and anterior chamber (by slit-lamp biomicroscopy), UVA irradiation was applied. The UVA irradiation was performed using an optical system (UV-X illumination system version 1000; IROC AG) with a light source consisting of an array of UV diodes (365 nm) in conjunction with a potentiometer to allow regulation of voltage. Before treatment, an intended 3.0 mW/cm2 of surface irradiance (5.4 J/cm2 surface dose) was calibrated using a UV light meter at a working distance of 10 cm. Irradiance was performed for 30 min. During treatment, the riboflavin solution was applied every 2 min to ensure saturation while balanced salt solution was used to maintain corneal stromal hydration. At the end of the procedure, a bandage soft contact lens was kept in place until full corneal re-epithelialization occurred.
The postoperative treatment was as follows: topical combined steroid and antibiotic drops that was administrated in all patients five times daily for 4 weeks with monitoring of IOP, then tapered to four times/day for the next week, then two times/day for the next 2 weeks, and lacrimal substitutes (preservative-free artificial tears) was administered four times daily for 4–6 weeks.
Group II: epi-on
This group included 15 eyes of 12 patients; 10 eyes of male patients and five eyes of female patients. The preoperative Kmax ranged between 43 and 52.7 D and astigmatism ranged between −2.7 and −8.3 D.
This group underwent riboflavin-UV-A CXL with the epithelium intact (epi-on).
The same optical system machine as in group I was used in all cases to avoid bias in the results.
- Anesthesia: Anesthesia was done the same as mentioned before.
- Operative details: After applying the eyelid speculum, instillation of Paracel (riboflavin 0.25%, hydroxypropyl methylcellulose, benzalkonium chloride, and ethylene diamine tetraacetic acid; Avedro Inc.) was done every one and a half minute for 6 min followed by instillation of VibeX (riboflavin 0.1%, dextran 500 disodium hydrogen phosphate, sodium phosphate, and sodium chloride; Avedro Inc.) every 2 min for 30 min The UVA irradiation was performed using an optical system (UV-X illumination system version 1000; IROC AG) with a light source consisting of an array of UV diodes (365 nm) in conjunction with a potentiometer to allow regulation of voltage.
- Postoperative medication: The same as in group I.
Postoperative follow-up for the two groups
The patients were followed up postoperatively at 1 week, 1 month, 3 months, and 6 months as follows: At 1 week, 2 weeks, and 1 month postoperatively, the patients were examined for their uncorrected visual acuity (UCVA) and best spectacle-corrected visual acuity (BSCVA), refraction, IOP, corneal examination, and patients’ complaints were all documented.
At the third and sixth month postoperatively, the patients were examined for their UCVA, best corrected visual acuity, refraction, IOP, corneal examination, and for any complaints. The patients were then investigated as follows: pentacam was used to assess the patient’s topography, keratometry, corneal astigmatism, and pachymetry. ORA was used to measure the CH and CRF and the specular microscope was used to count the endothelial cells.
| Results|| |
The study included 30 eyes of 18 patients, 10 of whom were men and eight were women with a mean age of 27.08 years (range: 17–30 years) ([Figure 1]). The preoperative mean Kmax was 47.9 D (range=41.1–62.7 D). The preoperative mean astigmatism was −3.75 D and the mean pachymetry of the corneal center was 467.1 µm.
Group I (epi-off) included 15 eyes of 12 patients, seven eyes of six male patients and eight eyes of six female patients. The preoperative Kmax ranged between 41.3 and 62.7 D with a mean of 50.9±5.21 D and corneal astigmatism ranged between −1.7 and −8.3 D with a mean of −3.99±1.78 D.
Group II (epi-on) included 15 eyes of 12 patients, 10 eyes of eight male patients and five eyes of four female patients. The preoperative Kmax ranged between 41.1 and 51.6 D with a mean of 46.15±3.44 D and corneal astigmatism ranged between −1.2 and −8.0 D with a mean of −4.06±1.97 D.
The mean preoperative UCVA in group I was 0.118±0.05 while the mean postoperative UCVA at 6 months was 0.26±0.18. The mean preoperative UCVA for group II was 0.278±0.14, while the mean postoperative UCVA at 6 months was 0.408±0.3 ([Figure 2] and [Table 2]). Comparing the difference in the mean preoperative UCVA and mean postoperative UCVA at 6 months between the two groups, using the analysis of variance: single factor test revealed the data as shown in [Table 3]. There was a statistically significant difference between the two groups regarding UCVA improvement.
|Figure 2 Comparing both preoperative and postoperative means uncorrected visual acuity at 3 and 6 months between the two groups.|
Click here to view
|Table 2 Comparing the preoperative mean uncorrected visual acuity, best spectacle corrected visual acuity, Kmax, spherical equivalent, and corneal astigmatism between the two groups|
Click here to view
|Table 3 Comparing preoperative and postoperative mean uncorrected visual acuity and best spectacle-corrected visual acuity readings at 6 months among the two groups|
Click here to view
The mean preoperative BSCVA for group I was 0.218±0.204 while the mean postoperative BSCVA at 6 months was 0.48±0.18. The mean preoperative BSCVA in group II was 0.6±0.305, while the mean postoperative BSCVA at 6 months was 0.745±0.21 ([Figure 3]). Comparing the difference in the mean preoperative BSCVA and mean postoperative BSCVA at 6 months between the two groups, using the analysis of variance: the single factor test revealed the data as shown in [Table 3]. There was a statistically nonsignificant difference between the two groups regarding BSCVA improvement.
|Figure 3 Comparing both preoperative and postoperative means best spectacle-corrected visual acuity at 3 and 6 months between the two groups.|
Click here to view
Comparing groups I and II revealed a statistically nonsignificant difference in BCSVA improvement.
The mean preoperative spherical equivalent (SE) was −5.61±2.01 D in group I and −8.48±2.7 D in group II. After 6 months, the mean SE was −3.66±1.01 D in group I and −4.32±2.69 in group II.
The mean preoperative corneal astigmatism was −4.06±1.97 D in group I and −3.9±1.7 D in group II. After 6 months, the mean SE was −3.64±2.3 D in group I and −3.19±1.17 in group II. Comparing the change in the mean SE between the two groups using the t-test: two-sample assuming unequal variances were found to be statistically significant (P=0.04). On the other hand, comparing that of corneal astigmatism was found to be statistically nonsignificant.
The mean preoperative flattest meridian keratometry (K1), steepest meridian keratometry (K2), and average keratometry (Kmax) were 44.24±2.65, 48.25±4.44, and 46.15±3.44 D, respectively in group II and were 49.45±5.03, 52.15±5.08, and 50.9±5.21 D, respectively in group I. At 6 months, these readings were 43.2±3.4, 46.64±5.17, and 44.84±4.19 D, respectively in group II and were 45.27±5.7, 49.39±5.63, and 47.2±5.58 D, respectively in group I. Using the t-test: two-sample assuming unequal variances, the difference in the Kmax improvement between the two groups was found to be statistically nonsignificant.
The mean preoperative pachymetry at its thinnest location was 444.71±38.75 µm and the central corneal pachymetry was 467.17±31.68 µm. After 3 months, the mean pachymetry (thinnest location) was 433.78±46.71 µm and the mean central corneal pachymetry was 459.78±39.63 µm. Later on, in follow-up the corneal thickness was almost back to its original preoperative value (a statistically nonsignificant difference between the two groups).
The mean preoperative CH was 7.65±1.46 in the epi-off subgroup and 7.98±0.89 in the epi-on subgroup while the mean postoperative CH after 6 months was 8.2±1.76 and 7.97±1.2, respectively ([Figure 4]). Moreover, the mean preoperative CRF was 7.19±1.66 in the epi-off subgroup and 6.51±0.89 in the epi-on subgroup. The mean postoperative CRF at 6 months was 7.38±1.31 and 6.5±1.25, respectively ([Figure 4]). There was a statistically nonsignificant difference between the result of CH and CRF in the two groups (P=0.44).
|Figure 4 Comparing both preoperative and 6 months postoperative means corneal hysteresis and corneal resistance factor between groups 1 and 2.|
Click here to view
After 1, 3, and 6 months there were no significant changes in the endothelial cell count during the procedure, and yet, there was no endothelial damage ([Figure 5] and [Figure 6] and [Table 4]). There was a statistically nonsignificant difference between the result of endothelial cell density by specular microscopy in the two groups (P<0.05).
|Figure 6 Specular microscope 1 month after the procedure for the left eye.|
Click here to view
|Table 4 Comparing the average endothelial count in both groups at 1, 3, and 6 months|
Click here to view
Minimal ocular adverse events were observed, and no significant IOP changes were seen. There was no delayed re-epithelialization in any of the patients. In addition, no endothelial damage was detected during the follow-up. night glare and haloes were reported by 80% of the patients, but only in the first 3 months. Stromal edema was clinically detectable by slit-lamp examination in 70% of the patients and it occurred in the first 30 postoperative days. Temporary haze occurred in 20% of the cases. The haze regressed after 1 month with a topical steroid regimen.
| Discussion|| |
In the present study, the results were comparable with the results of the study conducted by Caporossi et al.  in 2010 (a study that included 44 eyes with a mean follow-up of 48 months). Caporossi et al.  in their study had a hyperopic shift in the mean SE with a value of +1.87 D at 12 months and +2.15 D by the time of the last visit. Statistical analysis with the paired t-test revealed this to be statistically significant. These results are relatively comparable to the current study results .
Topographic analysis done for our patients preoperatively revealed a mean Kmax of 48.06±5.05 D. The mean postoperative Kmax at 6 months was 45.725±4.49, thus showing a mean reduction of −2.33 D. There was a statistically significant difference between the result of mean preoperative and mean postoperative Kmax at both 3 and 6 months. Kymionis et al.  in 2012 (a study on 14 eyes, maximum follow-up of 12 months) had a flattening of the mean keratometry readings from 51.99±5.57 to 49.33±4.82 D at the last follow-up visit with a mean reduction of −2.66 D which is similar to the present study results, while Vinciguerra et al.  on the other hand showed better results with a decrease in average keratometry from 48.08 D preoperatively to 42.01 D at 12 months with a mean reduction of −6.07 D.
One of the important criteria to evaluate the effects of CXL, in the present study, was to evaluate its effect on corneal thickness. The mean preoperative pachymetry at its thinnest location was 444.71±38.75 µm and the central corneal pachymetry was 467.17±31.68 µm. After 3 months, the mean pachymetry (thinnest location) was 433.78±46.71 µm and the mean central corneal pachymetry was 459.78±39.63 µm. Later on, in follow-up the corneal thickness was almost back to its original preoperative value. There was a statistically nonsignificant corneal thinning recorded in the first, three postoperative months and no statistically significant difference in central corneal thickness or corneal thinnest location was observed thereafter. These initial insignificant changes we think may be attributed to the corneal de-epithelialization that was performed during the operative procedure. These results are similar to the study conducted by Caporossi et al.  as they also recorded an initial nonsignificant decrease in the pachymetry values of all patients and then no statistically significant difference in central corneal thickness was observed after 3 months of follow-up. On the contrary, Vinciguerra et al.  found that there was a significant decrease in corneal thickness after 12 months of follow-up. The corneal pachymetry at the thinnest point changed from 451.14±25.97 to 436.23±29.38 µm, a change which was found to be statistically significant .In the present study, there were no significant changes in the biomechanical properties of the cornea after CXL as measured in vivo by ORA. The baseline preoperative values of CH and CRF were 7.785±1.14 and 6.734±1.25 mmHg, respectively. The mean CH and CRF values were transiently increased at 3 months after CXL (mean CH was 8.2±1.42 mmHg and mean CRF was 7.06±1.22 mmHg), with the differences from baseline being not statistically significant. At 6 months follow-up visit, a drop of the mean CH and CRF than the third month visit was found where the mean CH at 6 months was 8.085±1.48 mmHg and that of CRF was 6.94±1.33 mmHg. Goldich et al.  in 2009 conducted a study (on 10 eyes with a maximum follow-up of 6 months) comparing the ORA parameters pre-CXL and for up to 6 months afterwards. Their results were comparable to the current study and they concluded that there was a statistically significant change in CH and CRF after CXL. Their baseline values of CH and CRF were 8.44 and 7.15 mmHg, respectively. At 1 month, the values were 8.22 and 7.91 mmHg and at 6 months were 8.14 and 7.16 mmHg, respectively . Our results were also similar to the results published by Gkika et al.  in 2011 (on 50 eyes with a follow-up period of 12 months). Nonsignificant differences were detected between preoperative and postoperative CH and CRF measurements in keratoconic eyes (P=0.518 and 0.479, respectively). Significant correlations were found between ORA parameters and BSCVA, Kmax and astigmatism .
We compared the epi-off and epi-on groups aiming to document if removing the epithelium had any additive effect to CXL that would affect the visual, topographic, and biomechanical results.
When assessing the visual results, we found that in group I improvement in the UCVA after 6 months of the procedure was on average 0.14 Snellen’s lines while it was 0.13 Snellen’s lines in group II. In addition, the mean increase in the BSCVA in group I after 6 months was 0.262 Snellen’s lines and in group II 0.145 Snellen’s lines. On comparing the above results we found that the change in the UCVA between the two groups was statistically significant, while that of the BSCVA was nonsignificant.
Comparing the two groups, both groups showed a significant change in the SE, yet group I revealed better results. The SE showed a hyperopic shift of 3.56 D in group I and 1.29 D in group II after 6 months of follow-up. We think this may be attributed to the change in corneal symmetry and the centralization of the cone that occurs in patients of the first group.
Topographic analysis of both groups using the Oculus Pentacam showed a significant reduction in keratometric values in each group. However, the change in Kmax was found to be more in the epi-off group but still this difference was found to be nonsignificant. We think this imposes the idea that removing the epithelium during CXL potentiates the effect of CXL mainly through centralization of the cone.
In the present study, though there was a change in ORA parameters during follow-up, this change was statistically nonsignificant. There was no statistically significant difference between the two groups in terms of CH and CRF indicating that there is still a lot we need to learn to understand from the corneal biomechanics, rigidity and yet distensibility.
To our knowledge no case series was done to compare the corneal biomechanical changes in these two groups, thus further studies and work are needed to validate these results.
| Conclusion|| |
Corneal CXL treatment is a relatively new, revolutionary, minimally invasive procedure aiming at the stabilization of corneal ectatic disorders. From the present study results, CXL appears to be effective in improving uncorrected and best spectacle-corrected visual acuities in eyes with progressive ectasia by causing significant refractive, topographic, and clinical modifications. Though the corneal biomechanical changes induced by CXL were insignificant, they were positively correlated with topographic and visual acuity improvements. CXL with removing the epithelium helps to provide even better visual results for patients than transepithelial CXL. Longer duration of follow-up and a larger sample size may further provide more reliable results.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Caporossi A, Mazzotta C, Baiocchi S, Caporossi T. Long-term results of riboflavin ultraviolet a corneal collagen cross-linking for keratoconus in Italy. Am J Ophthalmol 2010; 149:585–593.
Kymionis G, Diakonis V, Portaliou D, Siganos C. One-year follow-up of corneal confocal microscopy after corneal cross-linking in patients with post-LASIK ectasia and keratoconus. Am J Ophthalmol 2009; 147:774–778.
Mazzotta C, Balestrazzi A, Baiocchi S, Tommasi C. Treatment of progressive keratoconus by riboflavin-UVA-induced cross-linking of corneal collagen. Cornea 2007; 26:390–397.
Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-A-induced collagen cross-linking for the treatment of keratoconus. Am J Ophthalmol 2003; 135:620–627.
Khan Y, Behrens A. Corneal collagen cross-linking for the treatment of diseases of the cornea. Contemp Ophthalmol 2009; 8:1–8.
Gipson IK, Joyce NC, Zieske JD. The anatomy and cell biology of the human cornea, limbus, conjunctiva and adnexa. Cornea 2005; 4:1–35.
Kymionis G, Poertaliou D, Diakonis V, Kounis G, Panagopoulou S, Grentzelos M. Corneal collagen cross-linking with riboflavin and ultraviolet-A irradiation in patients with thin corneas. Am J Ophthalmol 2012; 153:24–28.
Vinciguerra P, Albe E, Trazza S, Rosetta P, Vinciguerra R, Seiler T, Epstein D. Refractive, topographic, tomographic, and aberrometric analysis of keratoconic eyes undergoing corneal cross-linking. Ophthalmology 2009; 116:369–378.
Goldich Y, Barkana Y, Morad Y, Hartstein M, AvniIandZadok D. Can we measure corneal biomechanical changes after collagen cross-linking in eyes with keratoconus? A pilot study. Cornea 2009; 28:498–502.
Gkika M, Labiris G, Giarmoukakis A, Koutsogianni A, Kozobolis V. Evaluation of corneal hysteresis and corneal resistance factor after corneal cross-linking for keratoconus. Graefes Arch Clin Exp Ophthalmol 2012; 250:565–573.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]
[Table 1], [Table 2], [Table 3], [Table 4]