Background To investigate prognostic factors for complete remission in type 2 diabetes mellitus (T2DM) patients who underwent gastric bypass (GBP) and to establish a prognostic model for risk stratification. Methods We evaluated the baseline clinical features of patients with T2DM who received at Beijing Tian Tan Hospital from April 2012 to December 2015. Complete remission of T2DM was defined as meeting the following criteria: HbA1c <6.5%, fasting plasma glucose (FPG) <100 mg/dL, and absence of hypoglycemic drugs for 1 year following GBP. ResultsA total of 101 patients were enrolled in our study, and the complete remission rate of T2DM was 70.3% (71/101). Compared with patients with incomplete remission, patients with complete remission of T2DM had higher C-peptide levels, lower HbA1c, shorter disease duration, better cell function, and an absence of insulin therapy. HbA1c level, fasting C-peptide, duration of T2DM, and history of medical therapy were important prognostic factors for complete remission of T2DM (P=0.001, 0.002, 0.01, 0.028, respectively). Patients with HbA1c lower than 7.5%, a history of T2DM shorter than 9.5 years, fasting C-peptide higher than 1.2 ng/mL, and absence of insulin therapy before GBP achieved a higher complete remission rate of T2DM after GBP (AUC of the model was 0.825, 95% CI, 0.741-0.910; P=0.001). Conclusions The duration of T2DM, history of medical therapy, and levels of HbA1c and fasting C-peptide are independent predictors for the prognosis of T2DM patients undergoing GBP. Patients with HbA1c lower than 7.5%, a history of T2DM shorter than 9.5 years, a fasting C-peptide higher than 1.2 ng/mL, and an absence of insulin therapy may have a higher complete remission rate of T2DM after GBP.