In routine practice, warfarin is widely used in dialysis patients with atrial fibrillation (AF) for stroke prevention though the ratio of risks to benefits remains unclear. Recent cohort studies investigating the association between warfarin use and the risks of stroke and bleeding in dialysis patients with AF present conflicting results. The objective of this study was to assess the effectiveness and safety of warfarin use in patients with AF undergoing dialysis. Three databases PubMed, EMBASE, and OVID were searched from their inception to August 2015. Observational studies which assessed the ischemic stroke or bleeding risk of warfarin use in dialysis patients with AF were included. Two reviewers independently extracted data and assessed methodological quality based on the Newcastle-Ottawa Scale score. Combined hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using the random-effects model and heterogeneity was assessed based on the Cochrane Q-statistic test and the I statistic. Metaregression analyses were performed to explore the source of heterogeneity. A total of 11 eligible studies with 25,407 patients were included in the analysis. Warfarin use, in comparison with no-warfarin use, was not associated with a lower risk for ischemic stroke (HR 0.95, 95% CI 0.66-1.35). Sensitivity analyses found results to be robust. Metaregression analysis showed that demographic feature, clinical characteristics, or study-level variable had no impact of warfarin use on stroke risk. In addition, warfarin use was associated with a 27% higher risk for bleeding (95% CI 1.04-1.54). Overall, warfarin use did not have a significant association with reduced mortality (95% CI 0.96-1.11). It appears that warfarin use is not beneficial in reducing stroke risk, but with a high risk for bleeding in dialysis patients with AF. Randomized trials are needed to determine the risk-benefit ratio of warfarin in dialysis patients with AF.