Abstract
PURPOSE: Corneal crosslinking (CXL) is the standard treatment of progressive keratoconus (KC). We evaluated the safety and 10-year outcomes of conventional "epithelial-off" CXL for progressive KC for the first time in a cohort in France. METHODS: We conducted a retrospective review of patients undergoing conventional CXL (Dresden protocol) in our tertiary ophthalmology department from 2006 to 2011 with 10-year follow-up. The primary outcome was change in preoperative versus postoperative keratometry measured by maximum keratometry (Kmax), steep keratometry (K2), flat keratometry (K1), mean keratometry (Km), and topographic cylinder. Secondary outcomes were changes in visual and refractive outcomes. We report postoperative complications and adverse events. RESULTS: Eighty-nine eyes from 76 patients (67% male patients, mean age 22.7 ± 7.6 years) were included. Mean Kmax (-2.31 ± 2.98 diopters (D); P < 0.00001), K2 (-2.07 ± 3.15 D; P < 0.00001), K1 (-1.00 ± 2.29 D; P = 0.00008), Km (-1.53 ± 2.47 D; P < 0.00001), and topographic cylinder (-1.15 ± 2.53 D; P = 0.00004) significantly decreased 10 years after CXL compared with preoperative baseline. Significant decreases were still observed between 5 and 10 years after for mean Kmax, mean K2, mean K1, and mean Km. Mean distance best spectacle-corrected visual acuity and mean manifest refraction spherical equivalent were significantly improved after 10 years versus before CXL. The 10-year rate of repeat CXL was n = 3/76 patients (4%) (all younger than 18 years at first CXL) and of loss of >3 lines in best spectacle-corrected visual acuity was n = 1/76 patients (1%). CONCLUSIONS: Progressive KC was effectively stabilized with a prolonged flattening and maintenance of functional vision improvements after 10 years. Repeat CXL was rare and only required among younger patients.