Abstract: OBJECTIVE To evaluate the efficacy and safety of sulfonylureas for treating type 2 diabetes(T2DM). METHODS Bayesian network meta-analysis was used to compare the effects of sulfonylureas on treatment for T2DM. The effectiveness of sulfonylureas were assessed by changes of heamoglobinA1c(HbA1c) and fasting blood glucose(FBG) before and after treatment, and safety were assessed by hypoglycemia events and adverse events. RESULTS A total of 50 RCTs involving 201 339 patients were included. According to the results of surface under the cumulative ranking(SUCRA), network meta-analysis showed that metformin, glipizide and glibenclamide were better in reducing HbA1c compared with other treatments, and the SUCRA were 94.3%, 83.9% and 49.6%, respectively. Compared with other treatments, glipizide, gliclazide and glibenclamide had better effects in reducing FBG, and the SUCRA were 70.7%, 59.6% and 59.2%, respectively. Glibenclamide and gliclazide did not increase the risk of hypoglycemia compared with the untreated group, and the OR(95%CI) were 1.51(0.89-2.16, P>0.05) and 1.36(0.68-2.05, P>0.05), respectively. CONCLUSION Glipizide and glibenclamide are more effective in the treatment of T2DM, but glipizide is more prone to adverse events, and its conclusions needs further study.