Calcium antagonists are known to decrease insulin release in vitro. In vivo, their effects on insulin secretion and glucose tolerance are more controversial. We carried out a double-blind, double-dummy, controlled trial to study the effects of nifedipine (40 mg/day; n = 9), nitrendipine (20 mg/day; n = 9) and placebo (n = 10) on insulin release in 3 parallel groups of obese patients with mild or transient hypertension and a normal oral glucose tolerance test. The treatment lasted one week. Patients were asked not to modify their diet throughout the trial, and their body weight did not vary significantly. A 2-hour i.v. glucose tolerance test was performed twice in every patient, just before the first drug intake and one hour after the last one. The following parameters were measured or calculated during each test: basal, peak and early phase levels of blood glucose, insulin and C-peptide; glucose disappearance rate; and glucose, insulin and C-peptide incremental areas under the time-curves. On day 1, the 3 groups of patients were comparable for age, sex, body weight and every biological parameter. One-way analysis of variance did not show any significantly different evolution of those parameters between the 3 treatment groups, but calcium antagonists tended to slightly reduce the early phase of insulin release compared with placebo.