Inter-rater reliability of carotid atherosclerotic plaque quantification by 3-dimensional sonography

Abstract

Objectives—Embolization from atherosclerotic carotid plaques is the most common cause of ischemic stroke; therefore, identification of high-risk plaques by sonography is important. The aim of this study was to investigate the agreement between 2 investigators in the evaluation of sonographic parameters relating to plaque stability.

Methods—The following plaque parameters were assessed: echogenicity, homogeneity, surface, maximum content, and total volume. Serial 2-dimensional (2D) image sequences were obtained. Linear motion of the probe was automatically synchronized with the electrocardiogram. The edges of the plaque in each image were manually identified by the investigators. The total plaque volume was calculated after computer transformation of 2D images into a 3-dimensinoal (3D) format. Inter-rater reliability for echogenicity, homogeneity, and the surface was assessed by the weighted κ coefficient. Parametric values were tested by a paired t test.

Results—We enrolled 30 patients (22 male; mean age ± SD, 72 ± 13 years) in the study and evaluated 28 atherosclerotic plaques. Inter-rater agreement values were as follows: homogeneity, 96% (κ = 0.84; P < .001); surface, 90% (κ = 0.77; P < .001); and echogenicity, 86% (κ = 0.60; P < .001). The significance values for plaque content and volume measurement agreement were P = .311 and .312, respectively, and the correlation coefficient was 0.808.

Conclusions—In our study, the agreement between 2 examiners in the evaluation of 2D and 3D sonographic parameters related to plaque stability was good to excellent. The sonographic measurement of plaque volume growth was the most accurate parameter; therefore, 3D sonography may be used for risk assessment of plaques in the future.