To compare efficacy and tolerability of combination treatment with metformin and sulfonylurea with each of these drugs alone in the treatment of type 2 diabetes, 88 type 2 diabetic subjects (hemoglobin A1c [HbA1c] levels, 8.0%+/-1.0%; age, 57.3+/-7.1 years; body mass index [BMI]. 27.0+/-2.6 kg/m2; diabetes duration, 9.8+/-8.2 years; means +/- SD) were randomly assigned to double-blind treatment with metformin (500 to 3,000 mg/d), glibenclamide (5 to 15 mg/d), or their combination (metformin 400 to 2,400 mg/d + glibenclamide 2.5 to 15 mg/d) for 6 months. Thereafter, groups were crossed over for the following 6 months. Thus, each patient received metformin or glibenclamide alone, and the combination treatment. Doses were titrated to obtain HbA1c levels < or = 6.0% and fasting plasma glucose levels less than 140 mg/dL. Eighty patients concluded both treatment periods and were included in the analysis of treatment efficacy. In patients receiving metformin or glibenclamide alone, the maximal dose was reached in 21 and 25 patients, respectively; 8 and 15 of these subjects, respectively, required the maximal dose when they were on the combination treatment. During the study, 4 (10.0%) subjects receiving metformin, 7 (17.1%) receiving glibenclamide, and 31 (39.2%) receiving the combination treatment reached HbA1c levels < or = 6.0%. Moreover, when efficacy of the combination treatment on glycemic control was compared with that of single-drug therapies in each individual patient, the combination was significantly more effective than either metformin or glibenclamide (HbA1c after treatment, 6.1%+/-1.1% v 7.3%+/-1.4%, and 6.5%+/-0.7% v 7.6%+/-1.5%, respectively, both P<.0001). In conclusion, combination treatment with metformin and sulfonylurea is more effective than these drugs alone in improving glycemic control in type 2 diabetes, while also allowing a reduction of the dosage of each drug.