AIM: To assess factors influencing glycaemic control following gastric bypass surgery in patients with Type 2 diabetes and BMI< 30 kg/m(2) . METHODS: Prospective longitudinal study of 103 patients with inadequate glycaemic control who underwent gastric bypass surgery at Soonchunhyang University, Seoul, Korea (n = 66) and Min-Sheng General Hospital, Taipei, Taiwan (n = 37). Procedures were performed August 2009 to January 2011. Key outcome measures were excellent glycaemic control of Type 2 diabetes defined as HbA1c < 42 mmol/mol (≤6%); inadequate response defined as HbA1c > 53 mmol/mol (> 7%). Analysis was conducted using binary logistic regression, and cut-points obtained from receiver operator characteristics. RESULTS: Excellent glycaemic control was achieved in 31 (30%) at 1 year. Diabetes duration of < 7 years and BMI > 27 kg/m(2) provided independent predictors and useful cut-points. Likelihood of excellent glycaemic control for an individual could be estimated using loge (Odds) = -6.7 + (0.26 × BMI) + (-1.2 × diabetes duration). Baseline BMI of < 27 kg/m(2) and baseline C-peptide of < 2.0ng/ml, best predicted a poor glycaemic response. In those with favourable baseline characteristics percentage weight loss (%WL) had a dominant influence on glycaemic outcomes. Baseline C-peptide (> 2.4 ng/ml) and subsequent percentage weight loss (> 16%) were associated with excellent glycaemic control. Higher BMI was associated with greater percentage weight loss. CONCLUSION: In patients with Type 2 diabetes and BMI < 30 kg/m(2) , glycaemic response to gastric bypass is predicted by higher baseline BMI, shorter disease duration and higher fasting C-peptide. Post-surgery weight loss has a dominant effect. Baseline BMI and weight loss have a major influence on outcomes. © 2012 The Authors. Diabetic Medicine © 2012 Diabetes UK.