Coronary atherosclerosis is characterized by an early loss of endothelium-dependent vasodilation. However, the methods of assessing coronary endothelial function are invasive and difficult to repeat over time. Recently, a noninvasive ultrasound method has been widely used to measure flow-mediated dilation in the brachial artery as a surrogate test for endothelial function. We seek to further validate this method of measuring vascular function. The brachial artery diameters and blood flow of 20 normal volunteers (10 males and 10 females) were measured using high resolution (7.5 MHz) ultrasound and strain gauge plethysmography. Flow-mediated endothelium-dependent vasodilation was measured in the brachial artery during reactive hyperemia after 5 minutes of cuff occlusion in the upper arm. The brachial artery diameter increased maximally by 9.7 +/- 4.3% from baseline at 1 min after cuff release and blood flow increased by 1002 +/- 376%. Five min of cuff occlusion was sufficient to achieve 97 +/- 6% of maximal brachial artery dilation and degree of dilation was not different whether the cuff was inflated proximally or distally to the image site. The intraobserver variability in measuring brachial diameters was 2.9% and the variability of the hyperemic response was 1.4%. In young, healthy men and women, the baseline brachial artery diameter was the only factor that was predictive of the flow-mediated vasodilation response. The brachial noninvasive technique has been further validated by the determination of flow-mediated dilation. This method of assessing endothelial function may help to determine the importance of vasodilator dysfunction as a risk factor in the development of atherosclerosis.