Repeat Corneal Collagen Cross-Linking Following Primary Cross-Linking Failure In Keratoconus: A Retrospective Study With Long-Term Follow-Up

To evaluate the long-term corneal and visual outcomes of repeat corneal cross-linking (CXL) in patients with progressive keratoconus following a failed primary CXL.

Tertiary referral centre - Department of Ophthalmology, Royal Victoria Infirmary, Newcastle Upon Tyne, UK.

This retrospective observational study analysed medical records of patients who underwent repeat CXL for keratoconus progression between January 2014 and December 2024. Disease progression was determined by changes in tomographic and visual parameters. All procedures involved epithelium-off CXL. Depending on corneal thickness, either isotonic riboflavin 0.1% with HPMC or hypotonic riboflavin 0.1% was administered. All CXL procedures followed an accelerated protocol (A-CXL), with the corneas exposed to 9mW/cm² UV-A irradiation for 10 minutes, continuous mode (Avedro KXL System, Avedro). Only one repeat CXL was performed using the Sub400 protocol. Patients were followed for a mean period of 37.6 (SD 21.1, range: 19-75) months postoperatively.

Out of 201 consecutively treated eyes, five (2.48%) required repeat CXL due to evidence of progression, none bilateral. Four patients (80%) were female. The median age at primary CXL was 18 years, and 22 years at repeat CXL. Mean Kmax increased from 70.5 ± 11 D after primary CXL to 73.1 ± 11.1 D just before repeat CXL and subsequently decreased to 69.14 ± 9.7 D at the final follow-up. Best-corrected LogMAR visual acuity (BCVA) worsened from 0.38 ± 0.3 after the first CXL to LogMAR 0.56 ± 0.12 before repeat CXL but showed partial improvement (LogMAR 0.52 ± 0.2) at the final visit. One patient developed mild anterior stromal haze, but no other complications were observed.

Repeat CXL was effective in stabilizing keratoconus progression, with a trend toward corneal flattening and visual stability. Advanced disease is a risk factor for primary CXL failure, which aligns with our findings in this study. The mean interval of nearly five years between primary and repeat CXL highlights the slow and prolonged course of disease progression. These findings emphasize the importance of long-term follow-up after the first CXL procedure, as disease progression may occur several years after initial treatment. Close monitoring, particularly in younger patients with advanced keratoconus or risk factors such as atopy and eye rubbing, is essential for timely intervention.