Quantitative plaque characteristics based on 3D high-resolution vessel wall imaging predict long-term recurrence in intracranial atherosclerotic stroke

Purpose To investigate the association between plaque characteristics observed using three-dimensional high-resolution magnetic resonance vessel wall imaging (3D HRMR-VWI) and the recurrence of ipsilateral stroke in patients diagnosed with symptomatic intracranial atherosclerotic stenotic disease. Methods The patients with intracranial atherosclerotic acute ischemic strokes between January 2018 and December 2023 were enrolled in this study. All patients underwent 3D HRMR-VWI at baseline and were followed up for? 12 months or until stroke recurrence. Baseline imaging characteristics were systematically assessed. Data analysis utilized the Cox proportional hazards model, Kaplan-Meier curves, and receiver operating characteristic (ROC) curves. Time-dependent ROC analysis quantified the dynamic predictive accuracy of plaque characteristics during follow-up. Results A total of 74 patients were followed for a mean duration of 35.6 months (range: 1-80 months), with 18 experiencing stroke recurrence. Multivariate Cox regression analysis identified age (HR = 1.055; 95% CI: 1.01-1.10, p= 0.018), plaque burden (HR = 1.066; 95% CI: 1.01-1.13, p= 0.031), and plaque enhancement ratio (HR = 3.538; 95% CI: 1.39-9.00, p= 0.008) as independent predictors of stroke recurrence. Time-dependent ROC analysis revealed that the predictive model achieved the AUC of 0.838 for stroke recurrence at 3 years, surpassing its performance at 1 year (AUC = 0.793) and 2 years (AUC = 0.822). Conclusion Plaque burden and enhancement ratio were independent predictors of recurrent symptomatic intracranial atherosclerotic stroke. Incorporating age with plaque characteristics enhanced long-term stroke recurrence risk prediction.