Ultrasound diagnostic potential of carotid plaques in symptomatic and asymptomatic patients: Insights from quantitative gray-scale analysis

Abstract

ObjectivesTo determine the ability of ultrasound gray-scale analysis to differentiate between carotid plaques in symptomatic and asymptomatic patients and to assess the reproducibility of carotid plaque gray-scale analysis. MethodsUltrasound B-mode images from patients with carotid plaques were retrospectively evaluated. Gray-scale mean and median values of carotid plaque echogenicity were determined via computer-assisted quantification using the histogram feature in Adobe Photoshop. Gray-scale values of the entire plaque obtained by the two observers were used to measure echogenicity. Receiver operating characteristic and independent t-test were used to determine potential cut-off for normalized gray-scale values, and whether there was a significant difference between the normalized gray-scale values of carotid plaques in symptomatic and asymptomatic patients, respectively. Inter-observer reproducibility of gray-scale values was assessed using intraclass correlation coefficient (ICC). ResultsEighty-seven carotid plaques (59 symptomatic and 28 asymptomatic) from 54 patients with carotid plaques were evaluated on ultrasound B-mode. Gray-scale mean and median cut-off values of 25 and 22 showed sensitivity of 75% (95% CI 56.6–87.3%) and specificity of 23.7% (95% CI 14.6–35.9%). Carotid plaques in symptomatic patients had significantly lower gray-scale mean (p = 0.03) and median (p = 0.04) values compared with asymptomatic patients. Inter-observer agreement for gray-scale mean and median was excellent, with ICC values of 0.998 and 0.997, respectively. ConclusionUltrasound gray-scale analysis is reproducible and represents a promising diagnostic additive tool to differentiate between carotid plaques in symptomatic and asymptomatic patients. Further research is needed to establish the clinical utility of ultrasound gray-scale analysis for carotid artery disease management.