Retreatment for significant regression after excimer laser photorefractive keratectomy. A prospective, randomized,...

Regression, a gradual partial or complete return to the myopic state, remains a common complication of excimer laser photorefractive keratectomy (PRK) and limits the predictability of refractive outcome, especially in high myopia. An estimated 10% to 20% of patients, therefore, request a repeat PRK procedure. This study was designed to provide patient selection criteria and guidelines for successful retreatment.

One hundred six patients who had regressed were randomized to 1 of 4 retreatment groups comprising (1) those with minimal haze after their first PRK who received an exact retreatment; (2) those with minimal haze treated with a 50% deliberate overcorrection; (3) those with significant haze (> or = 2+ haze) given an exact retreatment; and (4) those with significant haze who received a 50% overcorrection. A Visx 20/20 laser was used in each case. Mean follow-up after retreatment was 12 months (range, 6-18 months).

Deliberate overcorrection (groups 2 and 4) resulted in a statistically significantly better refractive outcome (P = 0.026 at 6 months). Analysis of variance showed that significant haze after the first PRK was the most important predictor of a poor outcome after retreatment, other factors being high original myopia, marked regression, and loss of best-corrected visual acuity.

A retreatment PRK procedure for significant regression will reduce residual myopia significantly in the majority of patients, and a deliberate overcorrection (50%) reduces the chance of further regression. However, patients with high myopia who have regressed beyond approximately -3.50 diopters originally and who show significant anterior stromal haze (> 2+) should be retreated only with great caution, because of the risk of further regression, haze, and loss of visual acuity.