Background: Despite the increasing prevalence of diabetes and its complications in the world, no comprehensive studies have investigated the prevalence of diabetic retinopathy (DR) as the most important ocular complication. In this study, we conducted a systematic search to estimate the global and regional prevalence of DR and its prevalence in different age and sex groups, and to evaluate the determinants of heterogeneity in its prevalence. Methods: A systematic and comprehensive search was done in international databases, including Scopus, PubMed, Web of Science, Embase, and other data sources to find studies related to the prevalence of DR. Prevalence and 95% CI were calculated using binomial distribution. The Cochrane's Q-test and I2 index were applied to investigate heterogeneity, a random-effects model was used to estimate the pooled prevalence, and a meta-regression method was administered to investigate the determinants of heterogeneity among studies. Findings: Of 5648 studies, 87 articles with a sample size of 541256 individuals and 212497 diabetic patients were analyzed. The majority of the studies were from the WHO-Regional Office for Europe (n=24). The estimated pooled prevalence of DR in the diabetic population was 28.41% (95% CI: 25.99 to 30.84). The pooled prevalence of DR in the diabetic population was 26.10 (95% CI: 3.34 to 28.86) in female and 28.44 (95% CI: 25.56 to 31.32) in male patients. The heterogeneity of the pooled prevalence according to I2 was 99% (p<0.001). According to the meta-regression results, the variables of WHO region (b: 3.681; p<0.001), age (b: 8.01; p: 0.003), type of DR (b: 18.51; p<0.003), and publication year (b:-0.58; p: 0.001) had a significant correlation with heterogeneity. Interpretation: One third of diabetic patients suffer from DR, mostly NPDR. Age, WHO region, type of DR, and publication year affected the heterogeneity in the prevalence of DR. Our knowledge of the global prevalence of DR can be helpful for effective screening of DR, planning eye care services, management of DR risk factors, and also for prioritization of resource allocation. Funding Statement: Noor Research Center for Ophthalmic Epidemiology. Declaration of Interests: The authors declare that: No conflicting relationship exists for any author. Ethics Approval Statement: All parts of this meta-analysis were conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA guidelines).