Risk prediction model for ischemic stroke in symptomatic intracranial atherosclerosis patients based on high-resolution MRI and arterial spin labeling
10.3969/j.issn.1002-1671.2025.05.002
- VernacularTitle:基于高分辨率MRI及动脉自旋标记的症状性颅内动脉粥样硬化患者发生缺血性脑卒中风险预测模型
- Author:
Ling LI
1
;
Qianqian WANG
;
Min TANG
;
Na ZHANG
;
Yu WEN
;
Xiaoling ZHANG
;
Xiaoyan LEI
;
Xuejiao YAN
Author Information
1. 陕西省人民医院磁共振室,陕西 西安 710068
- Publication Type:Journal Article
- Keywords:
transient ischemic attack;
acute ischemic stroke;
arterial spin labeling imaging;
high-resolution magnetic resonance imaging
- From:
Journal of Practical Radiology
2025;41(5):726-731
- CountryChina
- Language:Chinese
-
Abstract:
Objective To develop a risk prediction model for ischemic stroke in symptomatic intracranial atherosclerotic stenosis(ICAS)patients based on high-resolution magnetic resonance imaging(HR-MRI)and arterial spin labeling(ASL)imaging.Methods A total of 142 patients were included and divided into acute ischemic stroke(AIS)and transient ischemic attack(TIA)groups based on stroke occurrence.Clinical risk factors,plaque characteristics,and arterial transit artifact(ATA)presence on ASL images were compared between the two groups.Multivariate logistic regression analysis was performed,incorporating clinical risk factors,plaque characteristics,and double post labeling delay(PLD)ATA presence.The predictive value of different models was compared using receiver operating characteristic(ROC)curve and DeLong tests.Results Hypertension,positive lumen remodeling,plaque enhance-ment rate,1.5 s-ATA presence,and 2.5 s-ATA presence were independent risk factors for AIS(P＜0.05).The combination of HR-MRI and ASL imaging predicted AIS most effectively[area under the curve(AUC)=0.908;95％ confidence interval(CI)0.862-0.954].No significant difference was found between the prediction performances of HR-MRI and ASL(95％CI-0.041-0.082,Z=0.659,P=0.509).Conclusion ASL is more convenient than HR-MRI for predicting ischemic stroke in ICAS patients.A model combining plaque characteristics and ATA presence effectively predicts AIS occurrence.
