In a previous retrospective study, long-term treatment of essential hypertensive patients with a slow-release calcium channel blocker resulted in normal resistance artery structure and endothelial function, which did not occur with a beta-blocker. In the present prospective study, 19 previously untreated essential hypertensive patients (aged 47 +/- 2 years, 75% male) were treated for 1 year in a double-blind randomized study with the long-acting calcium channel blocker amlodipine or the beta-blocker atenolol. Resistance arteries (lumen diameter, 150 to 350 microm) dissected from gluteal subcutaneous biopsies were studied on a pressurized myograph. Blood pressure (BP) control (129 +/- 2/85 +/- 2 mm Hg) was identical in both groups for the last 6 months of the study. After 1 year of treatment with amlodipine, the media-to-lumen ratio (M/L) of resistance arteries decreased from 7.89% +/- 0.40% to 6.81% +/- 0.41% (P < .05). Acetylcholine-induced endothelium-dependent relaxation tended to improve from 84.3% +/- 5.5% to 90.5% +/- 4.8% (P = .06), whereas sodium nitroprusside-induced relaxation was unchanged in the patients treated with amlodipine. In the beta-blocker-treated group there was no significant change in M/L or acetylcholine-induced relaxation. In conclusion, treatment with the calcium channel blocker amlodipine corrected altered resistance artery structure and tended to improve endothelial function in essential hypertensive patients, whereas similar good control of BP with the beta-blocker atenolol did not. Whether the vascular protective effect of amlodipine will result in improved outcomes in hypertension remains to be demonstrated.