Comparison of laser epithelial keratomileusis and photorefractive keratectomy for the correction of myopia: a meta-analysis.

BACKGROUND It is unclear whether a laser epithelial keratomileusis (LASEK) has any significant advantage over a photorefractive keratectomy (PRK) for correcting myopia. We undertook this meta-analysis of randomized controlled trials to examine possible differences in efficacy, accuracy, safety and side-effects between two methods, LASEK and PRK, for correcting myopia. METHODS A systematic literature retrieval was conducted in the PubMed, EMBASE, Chinese Bio-medicine Database, and Cochrane Controlled Trials Register to identify potentially relevant randomized controlled trials. The statistical analysis was performed using a RevMan 4.2 software. The results included efficacy outcomes (proportion of eyes with uncorrected visual acuity (UCVA) > or = 20/20 at 1 month and 12 months post-treatment), accuracy outcomes (proportion of eyes within +/-0.50 diopters (D) of target refraction at 1 month and 12 months post-treatment), safety outcomes (loss of > or = 2 lines of best spectacle-corrected visual acuity (BSCVA) at > or = 6 months post-treatment), mean pain scores on day 1 post-treatment, and mean corneal haze scores at 6 and 12 months post-treatment. RESULTS Seven articles describing a total of 604 eyes with myopia from 0 to -9.0 D were identified in this meta-analysis. The combined results showed that the efficacy and accuracy outcomes between the two groups at 1 month and 12 months post-treatment were comparable. No patient lost > or = 2 lines of BSCVA at > or = 6 months post-treatment in four relevant trials. Compared with PRK, LASEK did not relieve discomfort on day 1 post-treatment or reduce corneal haze intensity at 6 and 12 months post-treatment. CONCLUSIONS According to the available data, LASEK does not appear to have any advantage over PRK for correcting myopia from 0 to -9.0 D. This meta-analysis focuses mainly on the comparison of the early, mid-term and mid-long term results of the two methods. Additional studies to compare the long-term (> one year) results should be considered.