Background:Angiotensin II (A II) accelerates the development of atherosclerosis in hypertension. The plausible mechanisms are that A II promotes superoxide anion generation and activates PAI-1 expression and nuclear transcription factor. These effects are mediated by angiotensin II type 1 (AT1) receptor, however, blocking activity of maximal A II binding to the AT1 receptor are different among ARBs. Therefore, ARBs may have different vascular effects. Methods: We administered placebo, losartan 100 mg, irbesartan 300 mg, and candesartan 16 mg daily during 2 months to 122 patients with mild to moderate hypertension. This study was randomized, double-blind, placebo-controlled in design. Results: Compared with placebo, ARBs significantly improved the percent flow-mediated dilator response to hyperemia (p=0.019 by ANOVA) with no differences among each. Compared with placebo, ARBs significantly reduced plasam levels of malondialdehyde (MDA) (p=0.005 by ANOVA) and candesartan therapy reduced to the greatest extent. Compared with placebo or losartan, irbesartan and candesartan therapies significantly lowered plasma levels of PAI-1 antigen (p<0.001 by ANOVA) with no differences between both. Only candesartan therapy significantly lowered plasma levels of MCP-1 (p=0.004 by ANOVA) compared with others.There were no significant correlations between these changes and reduction of systolic blood pressure (-0.151���r���0.106) and between these changes and reduction of diastolic blood pressure (-0.146���r���0.118). There were significant correlations between the degree of changes in MDA and PAI-1 antigen levels following ARBs (r=0.265, p=0.011). Conclusions: We observed that ARBs improved endothelial function in hypertensive patients independent of lowering blood pressure. However, the clinical significance of different vascular effects among ARBs should be investigated.