The predictive value of HR-MRI for stroke recurrence in patients with non-stenotic intracranial atherosclerosis

ZHU Xiufang; CHEN Lei; DUAN Xinxiu; LI Lei

doi:10.20258/j.cnki.1006-9011.2025.10.005

Abstract：Objective To explore the predictive value of high resolution magnetic resonance imaging(HR-MRI)in stroke recurrence caused by non-stenotic intracranial atherosclerosis.Methods The clinical and HR-MRI examination data of 153 patients with stroke caused by non-stenotic intracranial atherosclerosis were collected at admission.During a 3-month follow-up,according to whether they experienced another stroke during the follow-up period,the patients were divided into two groups:the recurrent group(n=42)and the stable group(n=111).The HR-MRI imaging features and clinical parameters were compared be-tween two groups of patients.Then,using logistic regression analysis,we screened the influencing factors of stroke recurrence,and evaluated the predictive power of relevant influencing factors for stroke recurrence by means of ROC curve analysis.Results Patients in the recurrence group had a higher proportion of diabetes(P<0.05);in the recurrent group,more plaques were located in the intracranial segment of the internal carotid artery,and the plaque burden,eccentric index,plaque volume,and intra-plaque hemorrhage were higher(all P<0.05).Logistic regression analysis showed that plaque location,plaque burden and plaque volume were risk factors for stroke recurrence(all P<0.05).ROC curve analysis showed that the area under the curve for predicting stroke recurrence based on plaque position,plaque burden,and plaque volume was 0.637,0.660 and 0.860,re-spectively.The AUC of the three indicators combined was 0.890,which was higher than the prediction of three indicators alone(P<0.05).Conclusion HR-MRI has a high predictive value for stroke recurrence caused by non-stenotic intracranial athero-sclerosis.