颅内动脉粥样硬化斑块、软脑膜侧支循环与脑血流低灌注的关系:多模态MRI分析

Relationship between intracranial atherosclerotic plaque,pial collaterals and cerebral hypoperfusion:a multi-modal MRI study

目的:利用三维快速自旋回波和超长回波链采集序列(3D-CUBE)、TOF-MRA及三维动脉自旋标记(3D-ASL)灌注成像等多模态磁共振技术探讨颅内动脉粥样硬化斑块特征、软脑膜侧支与脑灌注之间的关系。方法:将72例症状性大脑中动脉狭窄患者纳入本研究。利用三维动脉自旋标记(3D-ASL)成像技术获得所有患者的脑血流量(CBF)值,根据CBF值将患者分为正常组(23例)和低灌注组(49例)。在3D-CUBE序列上测量大脑中动脉(MCA)狭窄段的管腔狭窄程度,分析和测量责任斑块的特征(包括分布、长度和斑块强化程度),在TOF-MRA原始图像上评估软脑膜侧支循环情况(MCA狭窄侧侧支血管较对侧减少<50%定义为侧支循环差,≥50%则定义为侧支循环良好)。先使用单因素分析两组患者在管腔狭窄程度、斑块特征及软膜侧支之间的差异,再将有统计学差异的变量纳入二元logistic回归模型,最后绘制ROC来评估单变量和组合变量预测供血区脑组织血流低灌注的效能。采用CBF结合阿尔伯塔省卒中项目早期CT评分方法(CBF-ASPECTS)评估脑灌注减低的范围,应用Mann-Whitney U检验评估斑块强化程度、软脑膜侧支情况与CBF-ASPECTS评分之间的关系。结果:二元logistic回归分析结果显示,斑块呈偏心分布(OR=5.099,P=0.030)、斑块长度(OR=1.713,P=0.023)及强化程度(OR=5.460,P=0.027)及软脑膜侧支循环差(OR=6.967,P=0.009)是供血区脑组织低灌注的危险因素。ROC曲线分析显示,斑块偏心分布、长度、强化程度及软脑膜侧支情况预测低灌注的AUC分别为0.724、0.792、0.750和0.736(P均<0.001),低于回归方程(4个变量组合)的AUC(0.907,P<0.001)。Mann-Whitney U检验结果显示,斑块明显强化组的CBF-ASPECTS评分低于无和轻度强化组,软脑膜侧支循环不良组的CBF-ASPECTS评分低于侧支循环良好组,且差异均有统计学意义(P<0.05)。结论:偏心分布、明显强化和较长的斑块及软脑膜侧支循环较差是脑血流灌注减低的危险因素,多因素组合可明显提高对低血流灌注的诊断效能。

Objective:To investigate the relationship between the characteristics of intracranial atherosclerotic plaque,pial collaterals and cerebral perfusion using multi-modal magnetic resonance technology including three-dimensional fast spin echo(3D-FSE)T_(1)-weighted imaging with variable flip angles(named as 3D-CUBE),time of flight(TOF)magnetic resonance angiography and 3D arte-rial spin labeling(ASL)perfusion imaging.Methods:72 patients with symptomatic middle cerebral artery(MCA)stenosis were collected.The cerebral blood flow(CBF)values of all patients were obtained by 3D-ASL technology.According to CBF values,they were divided into normal group(23 cases)and hypo-perfusion group(49 cases).The plaque characteristics and luminal stenosis degree of MCA responsible stenosis segment were measured on 3D-CUBE sequence.The plaque characteristics included plaque distribution,plaque length and plaque enhancement degree.The pial collaterals were evaluated on the original TOF-MRA images(>50%reduction in pial collaterals on the side of MCA stenosis compared to the contralateral side was defined as poor collateral circulation,and≤50%was defined as good collateral circulation).Univariate analysis was used to analyze the differences in the degree of lumen stenosis,plaque characteristics and pial collaterals between the two groups of patients,then the variables with statistical differences were included in the binary logistic regression model,and the receiver operating characteristic curve(ROC)was drawn to evaluate the efficacy of univariate and combined variables in predicting cerebral hypo-perfusion in blood-supply area.CBF combined with the early CT Score of Alberta Stroke Project(CBF-ASPECTS)was used to evaluate the range of cerebral hypo-perfusion.Mann Whitney U rank sum test was used to evaluate the relationship between plaque enhancement degree,pial collaterals and CBF-ASPECTS scores.Results:Binary logistic regression results showed that the eccentric distribution of plaque(OR=5.099,P=0.030),the length of plaque(OR=1.713,P=0.023),the degree of plaque enhancement(OR=5.460,P=0.027)and poor pial collaterals(OR=6.967,P=0.009)were the risk factors of hypoperfusion in blood-supply area.ROC curve analysis showed that the AUC of plaque eccentric distribution,length,enhancement degree and pial collaterals in predicting hypoperfusion were 0.724,0.792,0.750 and 0.736,respectively(all P<0.001),which were lower than the AUC(0.907,P<0.001)of regression equation(combination of 4 variables).Mann Whitney U-test results showed that the CBF-ASPECTS scores of the significant enhancement plaque group were lower than that of the non and mild enhancement group,and the CBF-ASPECTS scores of the poor pial collateral compensation group were lower than that of the good pial collateral compensation group,and the differences were statistically significant(P<0.05).Conclusion:Eccentric plaques,significantly enhanced plaques,longer plaques and poor pial collaterals are the risk factors of cerebral hypo-perfusion.The combination of multiple factors can significantly improve the diagnostic efficiency of hypo-perfusion.