Purpose: To evaluate the clinical outcomes and safety of customized corneal cross-linking (cCXL) for the treatment of keratoconus. Methods: Electronic databases were searched through May 1, 2025, yielding 394 records. Primary outcomes were 12-month changes in maximum keratometry (Kmax), corrected distance visual acuity (CDVA), and thinnest corneal thickness (TCT). Secondary outcomes included additional keratometric indices, uncorrected distance visual acuity, spherical equivalent, higher order aberrations, endothelial cell density, and demarcation line depth. Nineteen studies were included in the systematic review, with 14 studies eligible for meta-analysis. Both single-arm analyses of cCXL and comparative analyses versus non-customized CXL were performed. Subgroup analyses were conducted according to epithelial status and irradiation pattern. Results: At 12 months, cCXL significantly reduced Kmax by −1.29 diopters D (95% CI: −1.55 to −1.03 D) and achieved greater Kmax flattening than non-customized CXL (additional −1.39 D; 95% CI: −1.82 to −0.95 D; P < .001). CDVA improved by −0.10 logarithm of the minimum angle of resolution (logMAR) (95% CI: −0.12 to −0.07; P < .001), exceeding the improvement with non-customized CXL (mean difference: −0.06 logMAR; P = .031). TCT decreased modestly (−4.42 μm; 95% CI: −7.24 to −1.60 μm; P = .006). Significant reductions were observed in total higher order aberrations (−0.24 μm; 95% CI: −0.37 to −0.11 μm; P = .016) and vertical coma (−0.25 μm; 95% CI: −0.49 to −0.00 μm; P = .049). Outcomes were comparable between epithelium-on and epithelium-off cCXL, and no significant endothelial cell loss or serious adverse events were reported. Conclusions: Customized CXL is a safe and effective treatment for keratoconus and provides superior topographic and visual outcomes compared with non-customized CXL through tailored energy delivery.
Hsu et al. (2026) studied this question.