Should we patch corneal abrasions? A meta-analysis.

BACKGROUND Eye patching is commonly recommended for treating corneal abrasions. This advice seems based more on anecdotes or disease-oriented evidence theorizing that there is faster healing or less pain when the eye is patched. This meta-analysis was performed to determine if eye patching is a useful treatment for corneal abrasions. METHODS We conducted a comprehensive search of both MEDLINE (1966 to 1997) and Science Citation Index to locate relevant articles. We reviewed the bibliographies of included studies, and ophthalmology and primary care texts. Local ophthalmologists and authors were contacted to identify any unpublished data. Controlled trials that evaluated eye patching compared with no patching in patients older than 6 years with uncomplicated corneal abrasions were considered. The outcomes of interest were healing rates and degree of pain. RESULTS Seven trials were identified for inclusion, of which five could be statistically combined. Healing rates were similar in the two groups. The summary ratios (95% confidence interval) of healing rates in the patch group as compared with the no-patch group were 0.87 (0.68 to 1.13) and 0.90 (0.75 to 1.10) at days 1 and 2, respectively. Six studies evaluated pain: four found no difference and two favored not patching. No differences in complication rates were noted between the patched and nonpatched groups. CONCLUSIONS Eye patching was not found to improve healing rates or reduce pain in patients with corneal abrasions. Given the theoretical harm of loss of binocular vision and possible increased pain, we recommend the route of harmless nonintervention in treating corneal abrasions.