Save sight keratoconus registry study: Transepithelial versus epithelium-off corneal crosslinking

• What is already known on this topic? Corneal collagen crosslinking (CXL) is the primary treatment for progressive keratoconus and can be performed with the epithelium on or off. • What does this study add: Both transepithelial and epithelium-off corneal collagen crosslinking were similarly effective, measured by visual acuity and corneal curvature, to stabilise keratoconus and safe based on adverse events at 12 months follow up. • Implications of this study: Transepithelial and epithelium-off corneal collagen crosslinking to treat keratoconus were safe and effective at 12 months follow up. Corneal collagen crosslinking (CXL) is the primary treatment for progressive keratoconus and can be performed with the epithelium on or off. Evidence is needed to guide clinicians and patients on whether they should have transepithelial or epithelium-off CXL. The aim of this study was to determine the safety and efficacy of transepithelial and epithelium-off CXL. An observational prospective study was conducted using real-world data from patients in the Save Sight Keratoconus Registry. Patients with no previous intervention before CXL were included. The primary outcomes were adverse events, change in Kmax, habitual visual acuity and minimum corneal thickness from baseline to 12 months follow up. Mixed effects regression models evaluated changes in outcomes adjusted for age, sex, eye laterality, practices, and baseline outcomes. There were 46 eyes (37 patients) and 1203 eyes (976 patients) included treated with transepithelial and epithelium-off CXL, respectively. At 12 months follow up, the habitual visual acuity, pinhole visual acuity, Kmax, K2 and minimum corneal thickness were not significantly different between epithelium-off and transepithelial CXL. There were fewer adverse events recorded in transepithelial compared to epithelium-off CXL. Epithelium-off and transepithelial CXL were similarly effective, measured by visual acuity and corneal curvature, to stabilise and prevent progression between baseline and 12 months follow-up in keratoconus. Transepithelial CXL had fewer adverse events compared to epithelium-off CXL.