The Independent Effect of Various Cross-Linking Treatment Modalities on Treatment Effectiveness in Keratoconus

To investigate the 1-year outcomes of using various corneal cross-linking (CXL) techniques for treating keratoconus.

Setting: This is a prospective longitudinal cohort study performed at a tertiary academic medical center.

Six hundred seventy eyes of 461 patients with progressive keratoconus who underwent CXL were followed up for 1 year.

Eight combinations of CXL modalities were assessed, including 2 different CXL techniques (transepithelial or epithelium-off), 7 riboflavin formulations, and 2 ultraviolet-A protocols (conventional 3 mW/cm or accelerated 9 mW/cm). Patients treated using the Dresden protocol were used as the reference group.

Primary outcomes were maximum keratometry and mean keratometry 1 year after treatment. Multivariable linear regression was used which provides β coefficients (β). Secondary outcomes were uncorrected and corrected distance visual acuity, manifest refractive spherical equivalent, and corneal thickness 1 year after treatment.

Four treatment modalities differed significantly from the reference group regarding both maximum keratometry and mean keratometry: the transepithelial CXL (TE-CXL) group (β = 1.422; P = 0.001), 1 riboflavin formulation (Meran β = 1.210; P = 0.02), and both the 9 mW/cm protocols (Vibex Rapid β = 1.751; P < 0.001 and Collagex β = 1.170, P < 0.001). Overall, the visual outcome, manifest refractive spherical equivalent, and corneal thickness were similar among the treatment modalities. Infections were rare (1.6% of cases); however, re-treatment was required for 33.3% of cases that underwent TE-CXL.

TE-CXL, the use of Meran riboflavin, and applying the accelerated irradiation protocol appeared to be associated with reduced efficacy regarding controlling keratoconus progression. One-third of cases treated using TE-CXL required re-treatment.