Purpose: To compare efficiency and safety of epithelium-off corneal cross-linking (CXL) and transepithelial cross-linking (TE-CXL) in pediatric patients with progressive keratoconus.
Methods: Uncorrected and corrected visual acuity, corneal topography and pachymetry (Pentacam; Oculus Pentacam), and in vivo confocal microscopy (HRT II, Rostock Cornea Module, Heidelberg Engineering, Heidelberg, Germany) were evaluated at baseline and at 3, 6, and 12 months.
Results: In the epithelium-off CXL group (19 patients, 23 eyes; mean age, 14.75 ± 2.1 years), a significant improvement at month 12 was present for Kmax [-1.11 diopters (D), P = 0.01], Kmin (-3.2 D, P = 0.001), mean K (-1.47 D, P = 0.01), surface asymmetry index (-0.64 D, P = 0.001), inferior-superior symmetry index (-0.54 D, P = 0.01), index of height asymmetry (-2.97, P = 0.03), and anterior elevation at the thinnest location (-2.82 D, P = 0.01) and at the apex (-2.27 D, P = 0.01). Postoperative corneal edema lasted 3 months in 16 eyes (69.5%) and more than 6 months in 2 eyes (8.7%). In the TE-CXL group (10 patients, 14 eyes; mean age, 15 ± 4.2 years), a significant improvement at month 12 was present for Kmax (-1.14 D, P = 0.02), Kmin (-2.04 D, P = 0.01), mean K (-1.63 D, P = 0.01), surface asymmetry index (-0.86 D, P = 0.001), inferior-superior symmetry index (-0.55 D, P = 0.001), index of height asymmetry (-2.95, P = 0.01), and anterior elevation at the thinnest location (-2.96 D, P = 0.01) and at the apex (-2.19 D, P = 0.01). No postoperative corneal edema after TE-CXL was observed. Changes at month 12 from baseline were not significantly different between the 2 groups (P > 0.05). TE-CXE was significantly less painful than epithelium-off CXL.
Conclusions: In pediatric patients with progressive keratoconus, TE-CXL was less painful, provided similar effectiveness and fewer complications than epithelium-off CXL at 12-month follow-up.