Optimizing UV-A Irradiation Profiles in Crosslinking for Keratoconus: Comparison of Outcomes After Standard Accelerated and Topography-Guided Protocols

Abstract

INTRODUCTION: Our purpose was to compare visual and tomographic outcomes of crosslinking treatment for progressive keratoconus, utilizing excimer-laser assisted epithelium removal and either central uniform irradiation (C-CXL) or customized, topography-guided irradiation (TG-CXL).
METHODS: Retrospective study. We included patients with progressive keratoconus who underwent TG-CXL or C-CXL. In both procedures the epithelium was removed using phototherapeutic keratectomy (PTK) with a 50 µm ablation within a 7.0 mm optic zone, followed by riboflavin application every 2 minutes for 10 minutes. In TG-CXL this was followed by topography-guided ultraviolet-A (UVA) irradiation, with treatment energies ranging from 5.4 to 10 J/cm2 and fluence of 10 mw/cm2, while in C-CXL the cornea was uniformly irradiated with UVA with treatment energy of 6.0 J/cm2 and fluence of 10 mw/cm2. Patient data was collected at baseline, 6 and 12 months postoperatively including maximum keratometry (Kmax), anterior keratometry values within 5 points of a 3 mm diameter circle centered at the pupil, in the superior and inferior halves of the cornea, central corneal thickness, thinnest point pachymetry, subjective refraction and best corrected visual acuity (BCVA).
RESULTS: Fifty-four eyes from 48 patients were included (27 eyes for each group). Baseline characteristics were not significantly different between groups. Kmax was significantly lower 1 year (-0.83 ± 1.64 D; p=0.016) postoperatively for TG-CXL, but not for C-CXL (-0.46 ± 2.04 D; p=0.256). Inferior-superior (I-S) asymmetry index decreased significantly at 1 year for TG-CXL (-8.17 ± 9.56 D; p<0.001), but not for C-CXL (-3.69 ± 11.69 D; p=0.113). There were no significant differences in the evaluated structural parameters between 6 and 12 months postoperatively in both groups. The BCVA improved significantly at 1 year (difference to baseline: TG-CXL -0.13 ± 0.14 logMAR; p<0.001 and C-CXL -0.24 ± 0.38 logMAR; p=0.018; No difference between groups; p=0.244), while there was a significant myopic increase in spherical refractive error in both groups (difference to baseline: TG-CXL -1.05 ± 2.08 D; p=0.017 and C-CXL -0.90 ± 1.49 D; p=0.025).
CONCLUSION: One year after surgery, TG-CXL leads to greater Kmax reduction and topographic regularization than C-CXL. Both procedures lead to improved visual acuity and a myopic shift, with no superiority among the two. These results support the use of topography-guided crosslinking as a new valuable solution in the treatment of progressive keratoconus.