Diagnostic performance of MRI for detecting intraplaque hemorrhage in the carotid arteries: a meta-analysis.

Objectives: To investigate the diagnostic performance of MRI in diagnosing carotid atherosclerotic intraplaque hemorrhage (IPH) and to provide a clinical guide for MRI application.

Methods: We searched MEDLINE, Embase, and Cochrane library from the earliest available date of indexing through November 30, 2017. All investigators screened and selected studies comparing the use of MRI with histology. The accuracy to diagnose pathological IPH was expressed by sensitivity, specificity, negative likelihood ratios (LRs), positive LRs, and the area under summary receiver-operating characteristic (SROC) curve. We calculated the post-test probability to assess the clinical utility of MRI.

Results: We analyzed 696 patients from 20 articles. The sensitivity and specificity were 87% (95% CI, 81-91%) and 92% (95% CI, 87-95%), respectively. The positive and negative LRs were 10.27 (95% CI, 6.76-15.59) and 0.15 (95% CI, 0.10-0.21), respectively. The area under SROC curve was 0.95 (95% CI, 0.93-0.97). MRI was accurate in confirming or in ruling out disease over a wide range of pre-test probabilities of IPH: MRI could increase the post-test probability to > 80% in patients with a pre-test probability > 27% and could decrease the post-test probability to < 20% in patients with a pre-test probability < 64%.

Conclusion: Non-invasive MRI has excellent specificity and good sensitivity for diagnosing IPH. MRI is a tool for confirming or ruling out carotid atherosclerotic IPH.

Key Points: • Non-invasive MRI has excellent performance for diagnosing IPH, which is a component of vulnerable plaque. • The high accuracy of MRI for IPH helps clinicians analyze the prognosis of clinical events and plan personalized treatment.