Abstract:

Objective  To investigate the relationship between computational fluid dynamics (CFD)-based noninvasive hemodynamic features and the risk of ischemic stroke recurrence in acute ischemic stroke (AIS) patients with intracranial atherosclerosis stenosis (ICAS).

Methods  The AIS patients with symptomatic ICAS diagnosed by MRI and MRA at Department of Neurology, Beijing Tiantan Hospital, Capital Medical University from June 2018 to June 2019 were included in this cohort study. All patients completed 12-month follow-up. The patients were divided into two groups: stroke recurrence group and no recurrence group, according to with or without recurrence of stroke/TIA within 12 months. The clinical and imaging data were collected. The hemodynamic parameters (including blood flow fraction, blood flow velocity ratio and vascular wall shear force ratio) were calculated using MRA-based CFD model. The association between hemodynamic parameters and ischemic stroke recurrence within 12 months was analyzed.

Results  A total of 48 eligible patients were included in this study, with the median age of 58 (52-68) years old and 36 males (75.0%). The median rate of intracranial artery stenosis was 60.3% (53.2%-82.0%), and the median stenosis length was 3.3 (2.6-4.9) mm. Ischemic stroke recurrence within 12 months occurred in 10 patients (20.8%). The blood flow fraction at the culprit vessels in stroke recurrence group was lower than that in no recurrence group [0.74 (0.69-0.81) vs. 0.90 (0.82-0.97), P=0.01]. The area under the ROC curve of blood flow fraction for predicting the recurrence of ischemic stroke was 0.796 (95%CI 0.669-0.923), the cutoff value was 0.815, the sensitivity was 76.3% and the specificity was 80.0%.

Conclusions  Blood flow fraction calculated by CFD model may be a potential predictive factor of ischemic stroke recurrence in patients with symptomatic ICAS.

Key words: Symptomatic intracranial atherosclerosis stenosis; Blood flow fraction;Computational fluid dynamics; Ischemic stroke; Recurrence

摘要：

目的 评估基于计算机流体力学（computational fluid dynamics，CFD）技术的无创血流动力学参数与缺血性卒中复发之间的关系。 

方法 连续入组2018年6月-2019年6月在首都医科大学附属北京天坛医院脑血管病中心住院治疗的急性缺血性卒中且经头颅MRI和MRA证实为症状性颅内动脉狭窄的患者。收集患者的临床及影像学检查信息。对患者的MRA原始图像进行后处理分析，建立CFD模型，并通过计算获得责任血管的血流分数、血流速度比和血管壁剪切力比。所有入组患者完成12个月随访，根据随访期内是否有缺血性卒中或TIA复发分为卒中复发组和无卒中复发组，分析各血流动力学参数与12个月内缺血性卒中复发的关性。 

结果 共入组48例患者，中位年龄58（52～68）岁，其中男性患者36例（75.0%）。责任血管中位狭窄率为60.3%（53.2% 82.0%），狭窄长度为3.3（2.6～4.9）mm。12个月随访缺血性卒中复发10例（20.8%）。卒中复发组责任血管的血流分数低于无卒中复发组[0.74（0.69～0.81）vs . 0.90（0.82～0.97），P =0.01]。ROC曲线显示，血流分数预测缺血性卒中复发AUC为0.796（0.669～0.923），截点值为0.815，敏感度为76.3%，特异度为80.0%。 

结论 通过CFD技术模拟计算的血流分数可能是症状性颅内大动脉狭窄患者缺血性卒中复发的预测因子之一。

关键词: 症状性颅内动脉狭窄; 血流分数; 计算机流体力学; 缺血性卒中; 复发