Magnetic resonance angiography-detected vulnerable plaque features at 3.0 T field strength are associated with symptomatic presentation in carotid artery disease: A systematic review and meta-analysis.

Abstract

To evaluate the association between 3.0 T Magnetic resonance angiography with vessel wall imaging (MRA-VWI) detected vulnerable plaque features and the risk of symptomatic presentation and first-ever neurovascular events. This study followed PRISMA guidelines and was registered in PROSPERO. We systematically searched MEDLINE, Embase, CENTRAL, Web of Science, Epistemonikos, and the gray literature up to July 2024. Eligible studies included five or more adult patients undergoing 3.0 T MRA-VWI for carotid plaque characterization with reporting outcomes of stroke or transient ischemic attack. Exclusion criteria were recurrent or post-revascularization events, animal studies, and reviews. Data were extracted from cross-sectional, case-control, and prospective cohort studies. Pooled odds ratios (OR) were recorded for symptomatic presentation, and hazard ratios (HR) for incident first-ever events. Random-effects models were applied, and heterogeneity was assessed using I². A parallel analysis was performed for cohort studies to assess incident risk of asymptomatic patients including 1.5 T and 3.0 T strength fields. Thirteen studies met the inclusion criteria, of which 11 contributed to the quantitative analysis. In unadjusted meta-analyses of 3.0 T case-control/cross-sectional studies, type VI plaques (including intraplaque hemorrhage [IPH], ruptured cap, or surface thrombus) showed a strong association with symptoms (pooled OR, 5.03; 95% confidence interval [CI], 3.22-7.85; I² = 0%), as did IPH with or without thrombus (OR, 5.31; 95% CI, 3.10-9.11; I² = 8%) and IPH alone (OR, 4.88; 95% CI, 1.90-12.50; I² = 43.6%). Ruptured fibrous cap showed a large but nonsignificant association unadjusted (OR, 4.59; 95% CI, 0.77-27.32; I² = 15.2%), and type IV/V plaques showed no association (OR, 0.50; 95% CI, 0.12-2.01; I² = 6.6%). After adjustment, the associations remained and were significant for type VI (pooled adjusted OR, 3.56; 95% CI, 2.04-6.24; I² = 0%), IPH with or without thrombus (OR, 4.92; 95% CI, 1.56-15.53; I² = 19.8%), and ruptured cap (OR, 2.85; 95% CI, 2.28-3.56; I² = 0%); IPH alone remained strong but imprecise (OR, 5.36; 95% CI, 0.56-51.46; I² = 43.5%). In the parallel analysis, results remained consistent after including three additional cohorts of 1.5 T with one of 3.0 T, as type VI (pooled HR, 3.61; 95% CI, 2.32-5.62; I² = 0.4%) and IPH (HR, 3.51; 95% CI, 2.42-5.10; I² = 0%) predicted first-ever events in both 1.5 T and 3.0 T strength fields. Lipid-rich necrotic core was inconsistent and overall nonsignificant with high heterogeneity (HR, 2.38; 95% CI, 0.15-38.76; I² = 65.8%). We found the3.0 T MRA-VWI detection of vulnerable plaque was consistently associated with symptomatic status. Furthermore, despite field strength, vulnerability was associated with first-ever ipsilateral events. In contrast, type IV/V (lipid-rich necrotic core) alone was not discriminatory. Incorporation of vulnerable-feature imaging, particularly identification of type VI plaques into routine evaluation of carotid stenosis refines risk stratification beyond luminal narrowing alone.