Cardiovascular magnetic resonance imaging for non-invasive assessment of vascular function: validation against ultrasound.

Ultrasound is an established modality for quantification of vascular function in clinical studies of cardiovascular disease. We determined whether cardiovascular magnetic resonance imaging (CMR) can provide a comparable assessment of vascular function. In seventeen control subjects, we used CMR to quantify endothelium-dependent (flow mediated dilatation, FMD) and endothelium-independent dilatation of the brachial artery, brachial and carotid distensibility, aortic compliance, and pulse wave velocity. These were compared to brachial and carotid measurements obtained by established ultrasound protocols. Twelve of the volunteers then underwent repeated measurements with both modalities. There was good agreement between imaging modalities for measures of endothelial function and arterial structure in the same subjects (difference between CMR and ultrasound for FMD = 0.14 +/- 6.8%, and brachial artery area = - 0.7 +/- 2.2 mm2, correlation between modalities for FMD = 0.62, p = 0.01 and for area = 0.87, p = < 0.0001). Inter-study reproducibility was also similar (coefficient of variation (CV) for FMD: CMR = 0.3, ultrasound = 0.3, CV for brachial artery area: CMR = 0.1, ultrasound = 0.1). Comparability and reproducibility were not as strong for functional measures if repeated studies were several days apart (CV for FMD by ultrasound on the same day = 0.1 and several days apart = 0.4). CMR and ultrasound show good agreement for quantitative measures of vascular structure and function with good reproducibility for both modalities. The major advantage of CMR is that it allows one-stop integrated assessment of both peripheral and central measures of vascular function.