Diagnostic Accuracy of Carotid Plaque Instability by Noninvasive Imaging: A Systematic Review and Meta-analysis

Background There is increasing evidence that plaque instability in the extracranial carotid artery may lead to an increased stroke risk independently of the degree of stenosis. We aimed to determine diagnostic accuracy of vulnerable and stable plaque using noninvasive imaging modalities when compared to histology in patients with symptomatic and asymptomatic carotid atherosclerosis. Methods Medline Ovid, Embase, Cochrane Library, and Web of Science were searched for diagnostic accuracy of noninvasive imaging modalities (CT, MRI, US) in the detection of 1) vulnerable/stable plaque, and 2) vulnerable/stable plaque characteristics, compared to histology. The quality of included studies was assessed by QUADAS-2 and univariate and bivariate random-effect meta-analyses were performed. Results We included 36 vulnerable and 5 stable plaque studies in the meta-analysis, and out of 211 plaque characteristics from remaining studies, we classified 169 as vulnerable and 42 as stable characteristics (28 CT, 120 MRI, 104 US characteristics). We found that MRI had high accuracy (90% [95% CI:82–95%]) in the detection of vulnerable plaque, similar to CT (86% [95% CI:76–92%]; p>0.05), whereas US showed less accuracy (80% [95% CI:75– 84%]; p=0.013). CT showed a high diagnostic accuracy to visualize characteristics of vulnerable or stable plaques (89% and 90%) similar to MRI (86% and 89%; p>0.05); however US had lower accuracy (77%, p<0.001 and 82%, p>0.05). Conclusions CT and MRI have a similar, high performance to detect vulnerable carotid plaques, whereas US showed significantly less diagnostic accuracy. Moreover, MRI visualized all vulnerable plaque characteristics allowing for a better stroke risk assessment. Registration PROSPERO ID CRD42022329690 ( https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=329690 )