Reproducibility of noninvasive ultrasonic measurement of carotid atherosclerosis. The Asymptomatic Carotid Artery Plaque Study.

Abstract

To determine the effect of a lipid-lowering agent and/or a low-dose antithrombotic agent on the progression of early-stage carotid atherosclerosis, noninvasive B-mode ultrasound was used to measure intimal-medial thickness in asymptomatic individuals with moderately elevated lipids as part of the ongoing multicenter Asymptomatic Carotid Artery Plaque Study. Uniform ultrasonic scanning and reading protocols were implemented to obtain maximum intimal-medial thickness measurements in 12 standard segments in patients having a small to moderate wall thickness (1.5-3.5 mm) in at least one of the carotid arteries. Paired B-mode image recordings on 858 patients, performed 1 month apart and read at a core laboratory (each pair by the same reader), determined both within-sonographer (W, n = 405) and between-sonographer (B, n = 453) reproducibility. The primary end point (mean +/- SD), defined in each individual as the mean value of the 12 maximum intimal-medial thickness measurements, was 1.31 +/- 0.21 mm (W) and 1.32 +/- 0.22 (B) at the time of the second examination. The mean difference in the primary end point (exam 2-exam 1) was -0.01 +/- 0.13 mm (W) and 0.00 +/- 0.15 mm (B). The Pearson correlation coefficients were 0.79 (W) and 0.75 (B). In 90% of the patients, the absolute difference in the primary end point was less than 0.22 mm (W) and less than 0.24 mm (B). Variability of the secondary end point, defined as the single largest intimal-medial thickness measurement in a patient, was between three and four times larger than the variability for the primary end point. Differences in sonographer performance between clinical centers were very small. The results demonstrate that standardized noninvasive ultrasonic techniques yield highly reproducible measures of carotid intimal-medial thickness, which can serve as a measure of carotid atherosclerosis in clinical trials that monitor small rates of lesion progression.