KEY POINTS: Brachial artery (BA) shear-mediated dilatation is a widely used assessment of vascular function with links to coronary artery health and cardiovascular risk. Cerebral vascular health is often interrogated using cerebrovascular (middle cerebral artery velocity) reactivity to carbon dioxide. We show that endothelium-dependent diameter (dilator) responses are not significantly related between the internal carotid artery (ICA) and BA; nor are endothelium-independent responses. Additionally, ICA endothelium-dependent responses are not related to middle cerebral artery velocity or ICA blood flow reactivity responses to carbon dioxide. Therefore, assessment of large extracranial cerebral artery vascular health should be quantified via methods specific to the vessel, not via peripheral endothelial function or cerebrovascular reactivity to carbon dioxide.

ABSTRACT: This study compared internal carotid artery (ICA) and brachial artery (BA) endothelium-dependent and -independent vasodilation. We hypothesized that endothelium-dependent and -independent vasodilation of the ICA and BA would be neither similar in magnitude nor correlated between vessels. In 19 healthy adults (23 ± 6 years, 24 ± 3 kg/m2 , six female), endothelium-dependent dilatation in the ICA was determined via Duplex ultrasound during transiently elevated shear stress caused by increased partial pressure of end-tidal carbon dioxide using dynamic end-tidal forcing (+9 mmHg; cerebral flow-mediated dilatation, cFMD). BA endothelium-dependent dilatation was assessed via standard flow-mediated dilatation (FMD). Endothelium-independent dilatation in the ICA and BA was assessed concurrently for 10 min following administration of 400 µg sublingual glyceryl trinitrate (GTN). Endothelium-dependent vasodilation of the ICA (3.4 ± 2.4%) was lower than (P = 0.013) and not correlated to that of the BA (7.9 ± 3.3%; r2 = 0.00, P = 0.93). Including baseline diameter and shear-rate area under the curve as covariates maintained the difference between cFMD and FMD (3.3 ± 4.2% vs. 7.8 ± 3.8%, P = 0.03), while including baseline diameter and baseline shear rate-adjusted area under the curve as covariates abolished it (5.9 ± 3.7% vs. 5.9.8 ± 3.5%, P = 0.99). GTN-mediated vasodilation of the ICA (14.3 ± 2.9%) was lower than (P = 0.002) and not correlated to that of the BA (25.5 ± 8.8%; r2 = 0.12, P = 0.19). Adjusting for baseline diameter eliminated the differences in GTN-induced vasodilation (ICA: 20.1 ± 5.8% vs. BA: 20.4 ± 5.5%; P = 0.93). Differences in endothelium-dependent responses, and the lack of correlations between arteries, indicates that endothelium-dependent function cannot be assumed to be related across cerebral and peripheral vasculatures in young, healthy humans.