Abstract
Purpose
To evaluate long-term outcomes of corneal collagen crosslinking (CXL) using riboflavin and UV-A irradiation and to determine when to repeat CXL.
Methods
In this retrospective consecutive interventional case series 131 eyes of 131 patients (95 male, 36 female, mean age 29.7 ± 11.4 years) between 2006 and 2016 received standard CXL (Dresden protocol, epithelium-off) for progressive keratoconus. Corrected distance visual acuity (CDVA) and corneal tomography (K1, K2, Kmax) were repeatedly recorded 1 year (n = 103 eyes) to 10 years (n = 44) postoperatively. Only one eye per patient was included. Paired t-test or Wilcoxon matched-pairs signed rank test was used for parametric and nonparametric data, respectively.
Results
1–3 years preoperatively, median K2 significantly increased by 1.1 D (p < 0.001). Postoperatively, median K2 increased by 0.1 D after 1 year, then decreased over the remaining postoperative period by 0.85 D (p = 0.021). Kmax fluctuated without significant change. Median apical corneal thickness decreased by 16 µm (p = 0.012) after 5 years and then returned to preoperative values. Mean CDVA showed a significant improvement (decrease in logMAR 0.08 after 10 years, p = 0.010). CXL non-responders, defined by a postoperative increase in Kmax>2 D, increased from 16% after 5 to 33% after 10 years. Risk factors for non-response were young age, high astigmatism (>4.3 D), thin cornea (<480 µm), poor initial visual acuity (CDVA ≥0.3 D), and atopic dermatitis. 4 eyes were re-treated 3–4 years after first CXL without complications and keratoconus stabilized thereafter.
Conclusions
CXL can slow or stop keratoconus progression. However, as the number of responders declines after 5 years, especially patients with risk factors may require re-treatment.
Acknowledgments
The authors are grateful for the invaluable advice of the Institute for Clinical Epidemiology and Biometry (IKE-B) at the University of Würzburg in study planning and statistical analysis.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
The data that support the findings of this study are available from the corresponding author upon request.