The aim of this prospective, randomized, double-blind, placebo controlled study was to investigate the effect of nifedipine on carbohydrate metabolism in diabetic patients after a 3-day and a 3-month course of treatment. Sixteen non obese, well controlled non-insulin dependent diabetics, (HbA1 less than 10%), with moderate untreated hypertension were divided in two groups: nifedipine (group N, 8 patients) and placebo (group P, 8 patients). An oral glucose tolerance test (OGTT, 75 g glucose) and an arginine infusion were performed before, after a 3-day, and a 3-month course, either of nifedipine 30 mg/D or placebo. Blood samples obtained during OGTT were assayed for glucose and insulin, and during arginine infusion for insulin, glucagon and growth hormone. The differences between basal and peak values during tests were compared between both groups before and after treatment using Wilcoxon's rank sum test. Neither acute nor chronic administration of nifedipine or placebo modified the glucose tolerance. However, basal insulin levels were reduced by 3 month-administration of nifedipine (from 19 +/- 2 micromicrons/ml to 10 +/- 1 micromicrons/ml, p = 0,01). Otherwise the basal and peak hormonal values during tests were not significantly affected by nifedipine either at the start of after 3 months of treatment. These results suggest that nifedipine, when given in standard dosage for 3 months, has minor effects on carbohydrate metabolism in non-insulin dependent diabetic patients.