SUMMARYWhat is known and objectives: Aspirin is one of the most widely used medications in the world. The evidence on its effect on the risk of age-related macular degeneration (AMD) appears inconsistent across different types of studies. The aim of this metaanalysis was to evaluate the association between aspirin use and the risk of AMD. Methods: Relevant studies were searched using databases including PubMed, EMBASE, Cochrane Library and MEDLINE up to March 2014. Summary relative risks (RRs) and 95% confidence intervals (CIs) were calculated by random-effects or fixed-effect models. The heterogeneity was assessed by the inconsistency index (I 2 ). The publication bias was evaluated by Begg's funnel plot and Egger's weighted regression. Sensitivity analysis was also performed in different ways. Results: Ten eligible studies including 180 834 individuals based on the inclusion criteria were analysed in this metaanalysis. The pooled RR for the aspirin use on the risk of AMD was 1Á137 (95% CI, 1Á003-1Á289; I 2 , 68Á4%). The pooled RR for the aspirin use on the risk of early and late AMD was 1Á19 (95% CI, 0Á92-1Á53; I 2 , 82Á6%) and 1Á22 (95% CI, 0Á87-1Á72; I 2 , 55Á7%), respectively. In different types of late AMD, the pooled RR was 1Á95 (95% CI, 1Á40-2Á72; I 2 , 27%) for neovascularization and 0Á84 (95% CI, 0Á62-1Á15; I 2 , 0%) for geographic atrophy. The pooled RR in studies with standardized AMD classification was 1Á307 (95% CI, 1Á006-1Á698; I 2 , 79Á2%). What is new and conclusion: This meta-analysis updates similar reviews that included studies with various types of biases. A rigorous analysis shows a weak but statistically significant association between aspirin use and the risk of AMD; a result which is different to that previously reported.
WHAT IS KNOWN AND OBJECTIVESAge-related macular degeneration (AMD) is a common disease affecting the posterior segment of the eye, and it represents the leading cause of irreversible blindness and low vision in developed and many developing countries with the longest life expectancies.