Whether AT(1) (angiotenin II type 1) receptor blockade can prevent the decrease in conduit artery FMD (flow-mediated dilatation) during NOS (nitric oxide synthase) inhibition by alternative endothelial pathways has not been explored previously in humans. In 12 healthy subjects, we measured radial artery diameter (echotracking) and flow (Doppler) during FMD induced by sustained reactive hyperaemia during a control period and following NOS inhibition [1.5 mg.min(-1).l(-1) L-NMMA (N(G)-monomethyl-L-arginine)], after a single oral administration of telmisartan (80 mg) or placebo, using a randomized double-blind cross-over design. In six volunteers, we also assessed the roles of prostacyclin and EDHF (endothelium-derived hyperpolarizing factor) during radial FMD after AT(1) receptor blockade by oral administration of aspirin (500 mg) alone, aspirin+L-NMMA or aspirin+L-NMMA+fluconazole (a cytochrome epoxygenases inhibitor; 0.37 mg.min(-1).l(-1)). Telmisartan did not affect radial artery FMD in the control period (10.9+/-0.6% with placebo compared with 9.9+/-0.7% with telmisartan), but prevented its decrease after L-NMMA (9.3+/-0.8% with placebo compared with 12.6+/-1.2% with telmisartan; P<0.05) with no modification in baseline parameters, hyperaemia and radial artery endothelium-independent dilatation to sodium nitroprusside. Moreover, in telmisartan-treated subjects, radial artery FMD, compared with control (9.0+/-1.0%), was not modified by aspirin alone (9.4+/-0.7%) or associated with L-NMMA (9.5+/-0.5%), but was reduced by the combination of aspirin, L-NMMA and fluconazole (7.5+/-0.6%; P<0.05). These results demonstrate that AT(1) receptor blockade prevents the decrease in conduit artery FMD during NOS inhibition in humans, suggesting the development of a compensatory endothelial mechanism. This mechanism appears to be independent of prostacyclin and could possibly be related to an EDHF release.