AIMS: Weight loss is expected to improve glycaemic control in patients with diabetes or at high risk hereof. Sibutramine causes weight loss and is associated with an increased risk of myocardial infarction and stroke in high-risk patients. We examined the impact of sibutramine-induced weight loss on glycaemic control. METHODS AND RESULTS: In total, 8192 obese patients with diabetes were randomized to sibutramine or placebo plus diet and exercise after a preliminary 6 weeks in which all patients received sibutramine. Patients were classified into four groups of weight change. A total of 1582 patients had a weight loss >5.7 kg; 2047 patients lost 3.7-5.7 kg; 2432 patients lost <3.7 kg, and 1875 patients gained weight. Patients on sibutramine lost slightly more weight than those on placebo (-0.2 kg on average, P < 0.0001). Mean blood glucose changes in the placebo group were -0.6 mmol/L (±3.1, P = 0.0002), -0.2  mmol/L (±2.7, P = 0.04), and -0.1  mmol/L (±3.0, P = 0.01) in the moderate, modest, and mild weight loss groups, respectively; in the weight gain group blood glucose levels increased by +0.2 mmol/L (±3.1, P = 0.003). Corresponding mean blood glucose changes in the sibutramine-treated patients were -0.4 mmol/L (±3.2, P = 0.0002), +0.1 mmol/L (±3.0, P = 0.04), +0.4 mmol/L (±2.8, P = 0.01), and +0.2 mmol/L (±3.4, P = 0.003). Mean values of HbA1c followed the same pattern though the HbA1c changes were smaller with weight loss and greater with weight gain in the sibutramine group. All results were statistically significant (P < 0.0001). CONCLUSION: Weight loss induced by sibutramine, diet, and exercise attenuates falls in blood glucose levels and HbA1c compared with similar weight loss with placebo, diet, and exercise.