The purpose of this study is to compare the long-term effects of an angiotensin II receptor blocker (ARB) and a long-acting calcium channel blocker (CCB) on left ventricular geometry, hypertensive renal injury and a circulating marker of collagen synthesis in hypertensive patients. Patients with essential hypertension (24 men and 19 women; age, 37-79 years) were treated with a long-acting CCB, amlodipine (AML; 2.5-7.5 mg once daily) for 6 months. Then, AML was switched to an ARB, candesartan (CS; 4-12 mg once daily), in 22 patients (CS group), while AML was continued in the remaining 21 patients for another 6 months (AML group). At the end of each treatment period, ambulatory blood pressure monitoring (ABPM), echocardiography and sampling of blood and urine were performed. The average office blood pressure during the latter period was comparably controlled in the AML and the CS groups (AML: 130 +/- 8/87 +/- 7 mmHg; CS: 133 +/- 11/ 88 +/- 7 mmHg), while the average systolic blood pressure of 24-h ABPM was significantly lower in the AML than in the CS group (127 +/- 9 vs. 133 +/- 14 mmHg, p<0.05). Consequently, the left ventricular mass index was significantly decreased in the AML group (102 +/- 18 to 92 +/- 12 g/m2, p<0.05), while the change was insignificant in the CS group (103 +/- 25 to 98 +/- 21 g/m2). On the other hand, plasma procollagen I C-terminal peptide (PICP), a marker of collagen synthesis, was lowered by CS (86 +/- 21 to 70 +/- 21 ng/ml, p<0.01), but was not significantly affected by AML (80 +/- 127 to 74 +/- 91 ng/ml). CS reduced urinary albumin excretion (57 +/- 123 to 26 +/- 33 mg/g creatinine, p<0.05), but AML did not bring about significant changes (85 +/- 27 to 73 +/- 19 mg/g creatinine). The results suggested that long-acting CCBs are effective in improving left ventricular hypertrophy by controlling 24-h blood pressure, while ARBs possess protective effects against cardiovascular fibrosis and renal injury beyond their antihypertensive effects.