BACKGROUND: Numerous studies have investigated the association between adiponectin concentrations and diabetic retinopathy (DR) caused by type 2 diabetic mellitus. However, the results remain conflicting. We performed a meta-analysis to explore the relationship between adiponectin concentrations and risk of DR caused by type 2 diabetic mellitus from published articles. METHODS: A published literature search was performed through the PubMed, Cochrane Library, EMBASE, Science Citation Index Expanded database, Chinese CNKI, and Chinese Wan Fang databases for articles published in English and Chinese. Pooled standardized mean differences (SMDs) and 95% confidence intervals (95% CIs) were calculated using random or fixed effects model. Heterogeneity between studies was assessed using the Cochrane Q test and I(2) statistics. RESULTS: Nineteen studies with a total of 1 545 cases and 1 502 controls were retrieved. The original meta-analysis found a significant difference in the adiponectin concentrations between the DR and non-DR (NDR) groups. After excluding the high heterogeneity studies, the second meta-analysis also demonstrated the significant association (SMD (95% CI) = -0.62 (-0.80 to -0.44), P = 0.0001). According to the available data, there was statistical significance in the adiponectin concentrations considering non-proliferative DR (NPDR) versus NDR, PDR versus NPDR in Chinese populations with high heterogeneity. CONCLUSION: Adiponectin concentrations are correlated with DR; however, the relationship between adiponectin concentrations and DR needs more in-depth investigations with larger sample sizes.