Multimodal ultrasound and computed tomography imaging may improve the discrimination between symptomatic and asymptomatic patients with carotid stenosis

Abstract

Study objective To evaluate whether ultrasound (US) plaque features are associated with plaques from symptomatic patients and whether the combination with computed tomography (CT) parameters improves diagnostic accuracy. Design Single-center prospective cohort study. Setting Tertiary academic center for cardiovascular medicine in Germany. Participants 188 patients who underwent carotid endarterectomy (CEA) with 156 receiving US and 152 receiving CT. Interventions US assessment of plaque echogenicity, heterogeneity, calcification, discrete white areas, juxtaluminal echolucency. CT assessment of calcified plaque volume and heterogeneity. Multimodal combination of US and CT parameters. Main outcome measures Comparison of plaque features between patients with symptomatic and asymptomatic carotid stenosis. Diagnostic performance using logistic regression, receiver operating characteristic (ROC) curve analysis and area under the curve (AUC). Results Patients with symptomatic carotid stenosis had lower echogenicity (plaque with gray-scale (GS) <25: 61.6 ± 35.9% vs. 30.9 ± 27.4%, P < 0.001), greater US heterogeneity (coefficient of variation (CV): 1.20 ± 0.54 vs. 0.82 ± 0.48, P < 0.001), and more juxtaluminal echolucency (23.2% vs. 3.0%, P < 0.001). Plaque with GS <25 achieved an AUC of 0.76 (95% confidence interval, 0.68–0.84). Combining this US measure with CT-derived heterogeneity yielded a higher AUC of 0.83 (0.74–0.92), P = 0.062, with 80.0% sensitivity and 82.0% specificity. Conclusions US identifies patients with symptomatic carotid stenosis via echogenicity, heterogeneity, and qualitative features. Combining US with CT may improve diagnostic accuracy and warrants further validation. Multimodal plaque imaging should be evaluated concerning prospective stroke risk to assess its role in risk stratification before CEA.