Dramatic Reduction in Corneal Transplants for Keratoconus 15 Years After the Introduction of Corneal Collagen Crosslinking.

PURPOSE: The aim of this study was to investigate the effect of the implementation of corneal collagen crosslinking (CXL) on the frequency of corneal transplants among patients with keratoconus (KC) in the same region. METHODS: Before the introduction of CXL in 2007, 55 primary corneal transplants had been conducted in patients with KC (2005 and 2006) at the Department of Ophthalmology, Oslo University Hospital, Norway. We collected data from our corneal transplant registry for 2021 and 2022. The primary outcome was the number of corneal transplants performed in patients with KC. Age, sex, visual acuity (logarithm of the minimal angle of resolution), KC stage according to the Amsler-Krumeich classification system, and steepest keratometry reading (maximum keratometry, Pentacam, HR) were recorded. Furthermore, we registered the annual number of CXL treatments conducted from 2007 to 2022. RESULTS: A total of 352 corneal transplants were performed in 2021 and 2022. Among them, 11 (3.1%) were transplants for patients with KC. All included patients were male; further, 90.1% and 9.1% of the patients were graded stages 4 and 3, respectively. The mean maximum keratometry was 79.0 diopter (range 61.0-109). The mean best-corrected visual acuity (logarithm of the minimal angle of resolution) was 1.3 (range 0.2-3.0). In 2021 to 2022, 431 CXL treatments were performed. CONCLUSIONS: There was a significant decrease in the number of corneal transplants performed in patients with KC 15 years after the introduction of CXL. This indicates that the availability of CXL treatment over many years may considerably reduce the need for keratoplasties in this group of patients.

Superior outcome of corneal collagen cross-linking using riboflavin with methylcellulose than riboflavin with dextran as the main supplement.

PURPOSE: To compare the effect of corneal collagen cross-linking (CXL) on progressive keratoconus using 0.1% riboflavin with either dextran or methylcellulose as the main supplement. METHODS: In a comparative case series, CXL was performed in 40 patients (40 eyes) using a riboflavin solution containing either dextran (dextran-riboflavin; n = 20) or methylcellulose (methylcellulose-riboflavin; n = 20). Changes in central corneal thickness (CCT), Scheimpflug tomography, maximal keratometry reading (Kmax ), visual acuity (VA) and endothelial cell density (ECD) were recorded. Stromal changes one month after surgery were analysed using optical coherence tomography (OCT) and in vivo confocal microscopy (IVCM). RESULTS: The CCT was significantly higher in the methylcellulose-riboflavin group during the CXL procedure. The IVCM demarcation line depth was 274 ± 80 (SD) µm in the dextran-riboflavin group and 442 ± 80 µm in the methylcellulose-riboflavin group (p < 0.001). Complete absence of keratocytes in the pre-endothelial stroma was found in none of the corneas treated with dextran-riboflavin and in 42% of the corneas treated with methylcellulose-riboflavin. Visibility of the OCT demarcation line was significantly lower in the methylcellulose-riboflavin group. Kmax and corrected distance visual acuity were improved in the methylcellulose-riboflavin group and stable in the dextran-riboflavin group after 2 years. Endothelial cell density (ECD) was stable in both groups. CONCLUSION: We found deeper structural changes in the methylcellulose-riboflavin group than in the dextran-riboflavin group. This may be explained by different riboflavin solution properties and raises safety concerns. The study also indicates improved effect using methylcellulose-riboflavin than dextran-riboflavin, possibly explained by deeper stromal CXL effect.

Randomized Study of Collagen Cross-Linking With Conventional Versus Accelerated UVA Irradiation Using Riboflavin With Hydroxypropyl Methylcellulose: Two-Year Results.

PURPOSE: To compare the clinical outcome 2 years after corneal collagen cross-linking (CXL) with conventional and accelerated ultraviolet A (UVA) irradiation using riboflavin with hydroxypropyl methylcellulose. METHODS: Prospective randomized controlled study. Forty patients with keratoconus (40 eyes) were randomized to either CXL using conventional 3 mW/cm UVA irradiation for 30 minutes (CXL30 group) or accelerated 9 mW/cm UVA irradiation for 10 minutes (CXL10 group). In both groups, a solution of 0.1% riboflavin with 1.1% hydroxypropyl methylcellulose (methylcellulose-riboflavin) was used. Uncorrected distance visual acuity, corrected distance visual acuity (CDVA), and Scheimpflug tomography were performed at baseline and after 24 months. RESULTS: Both groups had statistically significant improvement in CDVA and maximum keratometric reading compared with baseline; however, with no statistically significant difference in the change between the 2 groups. No significant changes in flattest, steepest and mean keratometry (K1, K2 and K mean) were found in either of the groups. There were no statistically significant changes in ECD in either group after 2 years or in the difference in the change between the 2 groups. A literature review showed comparative clinical outcome after accelerated CXL compared with conventional CXL; however, in several studies, there was a tendency for less pronounced corneal flattening after accelerated CXL. CONCLUSIONS: Improvement in visual acuity and maximum keratometric reading 2 years after CXL was found after both conventional and accelerated UVA irradiation using methylcellulose-riboflavin. This suggests that when using riboflavin with methylcellulose, the less time-consuming accelerated protocol is a valuable and effective option in CXL treatment.

Collagen crosslinking with conventional and accelerated ultraviolet-A irradiation using riboflavin with hydroxypropyl methylcellulose.

PURPOSE: To evaluate corneal collagen crosslinking (CXL) with conventional and accelerated ultraviolet-A (UVA) irradiation using riboflavin with methylcellulose. SETTING: Department of Ophthalmology, Oslo University Hospital, Oslo, Norway. DESIGN: Prospective randomized case series. METHODS: Patients with keratoconus were randomized to have CXL using conventional 3 mW/cm2 UVA irradiation for 30 minutes (CXL30) or accelerated 9 mW/cm2 UVA irradiation for 10 minutes (CXL10). In both groups, a solution of riboflavin 0.1% with hydroxypropyl methylcellulose 1.1% (methylcellulose-riboflavin) was used. The endothelial cell density (ECD), visual acuity, and tomography were measured at baseline and after 12 months. Anterior segment optical coherence tomography and in vivo confocal microscopy (IVCM) were performed after 1 month. RESULTS: The study comprised 40 patients (40 eyes). A complete absence of keratocytes in all eyes at 100 µm depths was found on IVCM. At 300 µm, 400 µm, and preendothelial levels, the differences were 83.3% versus 31.3% (P = .02), 64.7% versus 20.0% (P = .01), and 42.1% versus 5.9% (P = .02) in the CXL30 and CXL10 groups. No statistically significant differences were found in the change in visual acuity or maximum keratometry between the groups after 12 months. There was no relationship between the depth of keratocyte absence and the ECD change after 12 months. CONCLUSIONS: Marked deep structural changes with an absence of keratocytes occurred when CXL was used with conventional or accelerated UVA irradiation; however, the changes were more pronounced with the use of conventional UVA irradiation. The use of methylcellulose-riboflavin might explain the deep alterations and raises a long-term safety concern.

Measuring the depth of crosslinking demarcation line in vivo: Comparison of methods and devices.

PURPOSE: To assess the interrelationship of different methods of measuring the demarcation line depth after corneal collagen crosslinking (CXL). SETTING: University eye clinic, Oslo, Norway. DESIGN: Prospective case series. METHODS: Eyes having CXL for progressive keratoconus were evaluated 1 month after CXL by in vitro confocal microscopy (IVCM), optical coherence tomography (OCT), and Scheimpflug imaging. When applying IVCM, the depth of the CXL demarcation line was measured with 2 methods; that is, IVCM keratocyte disappearance and IVCM intensity increase. With OCT, the evaluations were made by measuring the depth of the corneal stromal demarcation line. Scheimpflug imaging was used with 2 depth-measuring methods; that is manual Scheimpflug and objective Scheimpflug intensity change. The demarcation line depths in the central cornea were compared by the intraclass correlation coefficient (ICC) and pairwise comparison of the measured treated depth. If acceptable correlations (ICC > 0.7) were found, Bland-Altman analysis was performed. RESULTS: Twenty eyes of 20 patients were evaluated. Acceptable correlations were found between depth measurements using OCT-IVCM keratocyte disappearance (ICC = 0.80), OCT-IVCM intensity increase (ICC = 0.75), and IVCM intensity increase-IVCM keratocyte disappearance (ICC = 0.91). The Bland-Altman plots of these 3 pairs showed sufficient levels of agreement. Using pairwise comparison of these pairs, the measured depths were in the same level by the OCT-IVCM intensity increase only (P = .529). CONCLUSIONS: Scheimpflug images were inaccurate for measuring the CXL demarcation line depth. The 2 confocal microscopy methods and OCT images showed good correlation. Of these 3 pairs, only measurements with OCT and IVCM intensity increase depths were in the same level.

Does Corneal Collagen Cross-linking Reduce the Need for Keratoplasties in Patients With Keratoconus?

PURPOSE: To investigate whether the introduction of corneal collagen cross-linking (CXL) influences the frequency of keratoplasties in patients with keratoconus. METHODS: Data were obtained from a cohort of patients from our corneal transplant registry. Two different periods were compared, 2005 to 2006 (period 1) and 2013 to 2014 (period 2). Patients during period 1 had surgery before the introduction of CXL treatment, and patients in period 2 had surgery after this treatment was well established in our department. Age and gender were registered, and the Amsler-Krumeich classification system was applied to grade the degree of keratoconus. RESULTS: The total number of keratoplasties performed during period 1 was 137, and keratoconus was the cause of surgery in 55 eyes (55 patients). The corresponding numbers in period 2 were 231 and 26 eyes (26 patients), respectively. The difference in the number of keratoplasties for keratoconus in both periods was statistically significant (P = 0.003). There were no significant differences in the distributions of age and gender between both periods. In period 1, 63.6% of the eyes were graded as stage 4 in the Amsler-Krumeich classification, compared with 96.2% in period 2 (P = 0.001). CONCLUSIONS: The frequency of keratoplasty for keratoconus has been more than halved in our department over the last decade. There is reason to believe that this reduction is for a great part caused by the introduction of CXL treatment.

Corneal collagen crosslinking in vitro: inhibited regeneration of human limbal epithelial cells after riboflavin-ultraviolet-A exposure.

PURPOSE: To assess the hypothesis that during corneal crosslinking (CXL) treatment, riboflavin and ultraviolet-A (UVA) may have a toxic effect on human limbal epithelial cells. SETTING: Center for Eye Research, Department of Ophthalmology, Oslo University Hospital Ullevål, Oslo, Norway. DESIGN: Experimental study. METHODS: In this vitro study, limbal biopsies from corneoscleral rims collected after corneal transplantation were treated with the following combinations: riboflavin-UVA, riboflavin only, or UVA only; a control group received no treatment. After 3 weeks of cell culture, outgrowth of epithelium from the biopsies was evaluated by measuring the area of cell expansion and the number of cell layers. The explanted biopsies were analyzed for proliferation using immunohistochemistry marker Ki-67 and for apoptosis using the terminal deoxynucleotidyl transferase deoxy-UTP-nick end labeling (TUNEL) assay. RESULTS: The mean outgrowth from the biopsies was 2.25 mm(2) ± 6.90 (SD) in the riboflavin-UVA group, 181.4 ± 94.8 mm(2) in the riboflavin-only group, 128.5 ± 129.5 mm(2) in the UVA-only group, and 176.2 ± 114.0 mm(2) in the control group. There were no statistically significant between-group differences in the number of cell layers except in the riboflavin-UVA group, in which no cells were found. Detection of apoptosis with the TUNEL-assay was found in the riboflavin-UVA group only (4/5 sections). The proliferation marker Ki-67 was positive in some sections in all groups. CONCLUSION: Cytotoxicity and reduced cell expansion of human limbal epithelial cells occurred after riboflavin-UVA treatment in vitro, emphasizing the importance of avoiding riboflavin-UVA on the limbus during CXL.