OBJECTIVES: Bariatric surgery has been recommended as an effective treatment for overweight or obese patients with type 2 diabetes mellitus. The long-term follow up of such patients, however, has been limited. We performed a systematic review and meta-analyses to evaluate the long-term effect of bariatric surgery on patients with type 2 diabetes METHODS: We searched PubMed, EMbase, and Cochrane Central Register of Controlled Trials up to September 2013. Trained reviewers, working in pairs, independently screened for eligible studies and extracted data. We pooled the data by using random-effects meta-analysis; qualitative description was applied for the data not available for pooling. We conducted pre-specified subgroup analyses. RESULTS: We included 25 studies (n=23,628), consisting of 8 cohort studies with 17407 patients (moderate or high risk of bias) and 17 non-controlled follow up studies with 6221 patients (high risk of bias). At baseline, the mean age of 48.5 years (69% were female), mean body mass index (BMI) was 45 kg/m, 67% of patients received oral antidiabetic drug prior to surgery. 2.6% of patients (130/4950) died in the surgical group. Improvement on cardiovascular complications, nephropathy and quality of life was observed. BMI (-13.13 kg/m[-17.66,-9.07]), Fasting Plasma Glucose (-59.74 mg/dl [-74.59,-44.89]), and hemoglobin A1c (-1.83% [-2.4,-1.25]) declined significantly postoperatively. 67.1% of patients were in remission; diabetes was improved or resolved in 89.2% of patients. Subgroup analysis for alterative surgical procedures, length of follow up and the methods of data collection suggested no significant different. CONCLUSIONS: Bariatric surgery should be considered as a long-lasting and effective treatment for patients with type 2 diabetes. However, our results are based on a small number studies with low quality, additional carefully designed studies on mortality, cardiovascular events and adverse events may provide further clarification.