Objective To assess the therapeutic effect of gastric bypass on type 2 diabetes mellitus (T2DM) after a one-year treatment in Mainland China. Methods Databases including The Cochrane Central Register of Controlled Trials, MEDLINE, EMbase, CBM and CNKI were searched from inception to February 2012, and the relevant journals and references of articles were also searched to collect randomized controlled trials (RCTs) or before-after self-controlled trials on gastric bypass in treating T2DM in Mainland China. Two reviewers independently screened articles according to the predefined inclusion and exclusion criteria, extracted data, and evaluated quality of the included studies. Then meta-analyses were performed using RevMan 5.1.0. Results A total of 6 before-after self-controlled trials involving 131 patients were finally included. All these trials were graded as low quality. The results of meta-analysis showed that the therapeutic effect of gastric bypass on T2DM after a one-year treatment was good. There were significant reductions in both fasting plasma glucose (1 year: SMD=–2.55, 95%CI –3.40 to –1.69, Plt;0.000 01) and glycosylated hemoglobin (1 year: SMD=–1.98, 95%CI –2.33 to –1.62, Plt;0.000 01); there was no marked change in fasting insulin (SMD=–2.03, 95%CI –4.41 to 0.35, P=0.10). Sensitivity analysis indicated that these results were stable, but funnel-plots indicated possible publication bias existed. Conclusion One year after gastric bypass, T2DM patients in Mainland China get reduced in both fasting plasma glucose and glycosylated hemoglobin, but get no improvement in fasting insulin. However, this conclusion still needs to be further proved by more high-quality and large-scale clinical trials with long-term follow-up because of the limitation of quantity, scale and quality of the included studies.