Prevalence of Carotid Plaques with High-risk Features in Embolic Stroke of Undetermined Source: Systematic Review and Meta-Analysis.

Introduction: Recent evidence suggests a possible role of non-stenotic carotid atherosclerotic plaques in the aetiology of embolic stroke of undetermined source (ESUS). Methods: We conducted a systematic review and meta-analysis of prevalence and characteristics of non-stenotic carotid plaques (NSP) with high-risk features (complicated NSP) in internal carotid artery in unilateral ESUS in the anterior circulation. We searched Medline and Ovid-Embase databases. High-risk features were intra-plaque haemorrhage, thickness ≥ 3 mm, ulceration and hypodensity. We assessed the risk-of-bias (RoB), extracted the data, calculated the pooled-prevalence and 95% confidence intervals using Inverse Variance Weighting method and Random Effect models. Results: We included 16 studies and 1406 patients with different imaging for NSP assessment (1 Ultrasound, 11 CT-angiography, 4 MR-angiography). The RoB was moderate to low in most studies. Definition of complicated NSP differed across studies. The combined prevalence of any complicated NSP was 31% (95% CI=27-36%) ipsilateral and 14% (95%CI=9-19%) contralateral to the index stroke, the finding of any high-risk NSP was four-fold higher ipsilateral to the index stroke (OR=3.63; 95%CI=2.09-6.33). Prevalence of single high-risk features ipsilateral to ESUS was as follows: 35% (95%CI= 30-41%) for thickness ≥ 3 mm; 24% (95%CI= 8-39%) for ulceration; 45% (95%CI= -2; 93%) for hypodensity, 16% (95%CI=5-26%) for intraplaque haemorrhage. Conclusion: Complicated NSP are present in around a third of all ESUS, and are four times more frequent ipsilaterally to the index stroke. Our results confirm the possible causal role in ESUS and highlight the need for greater diagnostic uniformity of plaque at risk.