Fourteen non-insulin-dependent diabetic (NIDDM) patients continued their previous medication (7 on glyburide, 7 on glipizide) for 6 mo, after which they switched to the alternate treatment for another 6 mo. The treatment periods were followed by 1 mo of placebo. The sulfonylurea dose was increased to achieve fasting plasma glucose levels less than 9 mM or to a total maximum daily dose of 25 mg. The mean final doses of glyburide (14.7 +/- 2.4 mg/day) and glipizide (15.2 +/- 2.2 mg/day) were similar. Postprandial (postdose) glipizide levels were higher than those of glyburide, whereas fasting (predose) glyburide concentrations were higher than those of glipizide. Both treatments improved glucose control by 25% compared with placebo. Glipizide therapy evoked higher postprandial insulin concentrations than did glyburide, whereas basal insulin concentrations were higher during glyburide. Insulin sensitivity, assessed by an insulin tolerance test, was more improved with glyburide than with glipizide. In conclusion, overall glucose control is similarly improved by glyburide and glipizide. However, glipizide amplifies the plasma insulin response to meals more than glyburide, whereas glyburide enhances basal insulin secretion more than glipizide. Both pharmacokinetic and pharmacodynamic factors may contribute to these differences.