Comparative reproducibility of the noninvasive ultrasound methods for the assessment of vascular function

The aim of this work was to assess the reproducibility of ultrasound parameters of vascular function, since these measurements are currently recommended by the guidelines for the evaluation of the cardiovascular risk. Twenty subjects (51 +/- A 17 years, 11 men) had vascular ultrasound (Aloka Prosound alpha 10) performed by two observers, at the level of the right common carotid artery for assessment of intima-media thickness (IMT), wall tracking, and wave-intensity analysis, and at the level of the right brachial artery for the assessment of flow-mediated dilation (FMD). Wave intensity is a hemodynamic index, evaluating ventriculo-arterial interaction and can be measured in real time by a double-beam ultrasound technique through simultaneous recording of carotid arterial blood flow velocity and diameter. Carotido-femoral pulse wave velocity (PWV) was determined using the Complior method. Intra- and inter-observer reproducibility was assessed during a first session, when three consecutive acquisitions were made (first observer -> second observer -> first observer); repeatability was evaluated 1 week later (second observer). The most reproducible and repeatable parameters were PWV (intraobserver +/- 3.3%, interobserver +/- 2.6%, repeatability +/- 5.6%) and IMT (+/- 3.7, +/- 4.3, +/- 4.9%, respectively). Intraobserver reproducibility for arterial stiffness and ventriculo-arterial coupling parameters was the highest for the beta index (+/- 3.8%), and the lowest for the second systolic peak (+/- 22.4%). Interobserver reproducibility and repeatability varied between very good for the wave speed (+/- 5.5 and +/- 4.3%), and unsatisfactory for the negative area (+/- 31.8 and +/- 38.6%). FMD had good reproducibility (intraobserver +/- 11.6%, interobserver +/- 8%, repeatability +/- 7%), whereas augmentation index had only satisfactory results (+/- 17.8, +/- 8.4, +/- 23.8%, respectively). Only some parameters of vascular function have good reproducibility and repeatability, better or similar to other ultrasound methods and, therefore, these are ready to be used in routine clinical practice.