Natural history of corneal haze after collagen crosslinking for keratoconus and corneal ectasia: Scheimpflug and biomicroscopic analysis.

PURPOSE To determine the natural history of collagen crosslinking (CXL)-associated corneal haze measured by Scheimpflug imagery and slitlamp biomicroscopy in patients with keratoconus or ectasia after laser in situ keratomileusis. SETTING Cornea and refractive surgery subpecialty practice, United States. DESIGN Prospective randomized controlled clinical trial. METHODS The treatment group received ultraviolet-A/riboflavin CXL therapy. The control group received riboflavin alone without epithelial debridement. To objectively measure CXL-associated corneal haze, corneal densitometry using Scheimpflug imagery was measured and the changes in haze were analyzed over time. A similar analysis was performed using clinician-determined slitlamp haze. Correlation of CXL-associated corneal haze with postoperative outcomes was analyzed. RESULTS The mean preoperative corneal densitometry was 14.9 ± 1.93 (SD) (Pentacam Scheimpflug densitometry units). Densitometry peaked at 1 month (mean 23.4 ± 4.40; P<.001), with little change at 3 months (mean 22.4 ± 4.79; P = .06) and decreased between 3 months and 6 months (19.4 ± 4.48; P<.001) and between 6 months and 12 months. By 12 months, densitometry had not completely returned to baseline in the entire cohort (mean 17.0 ± 3.82; P<.001) and the keratoconus subgroup; however, it returned to baseline in the ectasia group (16.1 ± 2.41; P = .15). The postoperative course of slitlamp haze was similar to objective densitometry measurements. Increased haze, measured by densitometry, did not correlate with postoperative clinical outcomes. CONCLUSIONS The time course of corneal haze after CXL was objectively quantified; it was greatest at 1 month, plateaued at 3 months, and was significantly decreased between 3 months and 12 months. Changes in haze did not correlate with postoperative clinical outcomes.