PURPOSE: To identify preoperative parameters that may predict flattening of the keratoconic cornea after collagen crosslinking (CXL).

SETTING: Institut f€ur Refraktive und Ophthalmo-Chirurgie (IROC), Zurich, Switzerland.

DESIGN: Cohort study.

METHODS: Patients with verified progressive primary keratectasia received standard corneal CXL. Factors such as corrected distance visual acuity (CDVA) and Scheimpflug tomography (Pentacam) were used to follow the evolution from preoperatively to 12 months after CXL. Statistical analysis included U tests and Spearman rank correlation tests to detect risk factors for flattening of the keratoconus.

RESULTS: The study enrolled 151 eyes of 151 patients; more than 80% completed the 12-month follow-up. The flattening rate (flattening of the maximum curvature >1.00 diopter [D]) was 37.7%. A preoperative maximum keratometry (K) reading of more than 54.00 D was identified as the only significant risk factor for this effect (odds ratio, 1.88; 95% confidence interval, 1.01-3.51). A restriction to corneas with a maximum K value greater than 54.00 D would have resulted in a significant flattening in 51% of the cases.

CONCLUSIONS: Statistically significant flattening occurred during 1 year after CXL in more than 50% of cases when the preoperative maximum K reading was more than 54.00 D. None of the other preoperative parameters evaluated (eg, age, sex, diagnosis, CDVA, corneal shape factors) had a statistically significant impact on corneal flattening after CXL.