The effects of two antihypertensive regimens (isradipine and hydrochlorothiazide-amiloride) on the ratio between media thickness and lumen diameter of subcutaneous arteries and on left ventricular mass in essential hypertension were compared. Fifty patients, aged 46.3+/-8 (mean+/-SD) years, with newly diagnosed or poorly controlled essential hypertension were randomized to treatment with either isradipine or hydrochlorothiazide-amiloride. Atenolol and hydralazine were added in both groups as secondary and tertiary drugs, respectively, when needed for normalization of diastolic blood pressure. A subcutaneous gluteal biopsy was taken surgically before medication and again after 9 months of successful antihypertensive treatment. Two small resistance arteries were isolated from each biopsy and mounted in a Mulvany-Halpem isometric small vessel myograph. The media thickness-to-lumen diameter ratio (percentage) of the vessels was measured under standardized conditions and meaned. Left ventricular mass (LVM) index was determined by echocardiography according to the Penn convention. Ten patients were treated with isradipine as monotherapy, whereas only one patient was well controlled on diuretics as monotherapy. Mean blood pressure was reduced equally with the two regimens, from 131+/-9 mm Hg to 101+/-10 mm Hg with the isradipine and from 128+/-9 mm Hg to 99+/-7 mm Hg with the thiazide/atenolol regimen. LVM decreased significantly in both groups by 130+/-75 g with the isradipine-based regimen and by 70+/-53 g with the hydrochlorothiazide/atenolol-based regimen. The reduction of LVM was significantly greater on the isradipine-based regimen than on the thiazide-based regimen (P < .01). There was a significant reduction of media thickness-to-lumen diameter ratio during treatment with the isradipine-based regimen from 10.9% to 8.8% (P < .01). The reduction in the thiazide regimen was from 9.7% to 8.5%, which was not significant (P = .07). The study demonstrated significant reduction of hypertensive changes in peripheral resistance artery structure during antihypertensive treatment with an isradipine-based regimen. The thiazide/betablocker-based regimen did not have a significant effect on the vessels. Significant reduction of LVM was achieved with both isradipine-based and thiazide/atenolol-based regimens. The reduction of LVM obtained with the isradipine-based regimen was significantly greater than that of the thiazide/atenolol-based regimen.