目的：缺血性脑卒中后交叉性小脑神经机能联系不能（crossed cerebellar diaschisis，CCD）的发生对评估患者的神经功能和临床结局具有重要作用。本研究探讨磁共振（magnetic resonance，MR）体素内不相干运动（intravoxel incoherent motion，IVIM）定量参数诊断亚急性缺血性脑卒中后CCD的价值。

方法：回顾性收集2015年1月—2017年12月就诊的56例亚急性缺血性脑卒中患者的病历资料，其中39 例符合病例选择标准，均接受头颅常规MR成像（MR imaging，MRI）、IVIM 成像和三维动脉自旋标记（arterial spin labeling，ASL）成像检查。测量ASL成像序列病灶侧及健侧小脑半球的脑血流量（cerebral blood flow，CBF）及相应的不对称指数（asymmetry index，AI）_CBF。以AI_CBF＞10%判定为CCD阳性。测量双侧小脑半球IVIM成像定量参数，包括真扩散系数（D）、假扩散系数（D﹡）和灌注分数（f）以及单指数模型参数ADC，并且计算各参数的AI_D、AI_D﹡、AI_f和AI_ADC。采用配对t检验比较双侧小脑半球各参数值的差异，Pearson相关分析比较IVIM成像各参数与CBF之间的相关性，采用受试者操作特征（receiver operating characteristic，ROC）曲线评估IVIM成像相关参数检出CCD阳性的诊断效能。

结果：39 例患者中，CCD阳性者15例（38.5%），CCD阴性者24例（61.5%）。CCD阳性组入院和出院时的美国国立卫生研究院卒中量表（National Institute of Health Stroke Scale，NIHSS）评分均高于CCD阴性组（P＜0.05）。CCD阳性组双侧小脑半球D﹡值和f值的差异均有统计学意义（P＜0.001），健侧D﹡值低于患侧，而f值高于患侧。AI_D﹡与AI_CBF呈正相关（r＝0.502，P＝0.001），AI_f与AI_CBF呈负相关（r＝－0.347，P＝0.003）。AI_D﹡和AI_f在检出CCD阳性方面显示出较好的诊断效能，曲线下面积分别为0.744和0.717。

结论：IVIM 成像定量参数有助于检出幕上缺血性脑卒中后CCD。

Objective: Crossed cerebellar diaschisis (CCD) after ischemic stroke plays an important role in the evaluation of neurological function and clinical outcome. The aim of this study was to investigate the value of magnetic resonance (MR) intravoxel incoherent motion (IVIM) quantitative parameters in detecting CCD after subacute ischemic stroke.

Methods: The medical records of 56 patients with subacute ischemic stroke were collected and analyzed retrospectively from January 2015 to December 2017. Of 56 patients, 39 patients met the criteria of case selection. All patients underwent routine MR imaging (MRI), IVIM sequence on 3.0T MR and three-dimensional arterial spin labeling (ASL). The cerebral blood flow (CBF) and asymmetric index (AI) of CBF (AI_CBF) of bilateral cerebellar hemisphere were detected on ASL perfusion MRI. AI_CBF > 10% was considered as CCD positive. The IVIM derived parameters (D , D﹡, f and ADC ) and the corresponding asymmetric indexes (AI_D , AI_D﹡, AI_f and AI_ADC ) were measured. Paired t test was used to compare the differences in parameters of cerebellar hemispheres. Pearson correlation analysis was used to evaluate the correlation between IVIM parameters values and CBF. Receiver operating characteristics (ROC) curve was used to evaluate the diagnostic accuracy of the statistical significant IVIM parameters in detecting CCD.

Results: Among 39 patients, 15 (38.5%) were CCD positive and 24 (61.5%) were CCD negative. The National Institute of Health Stroke Scale (NIHSS) scores at admission and discharge of the CCD positive group were higher than those of the CCD negative group (P > 0.05). The values of D﹡ and f of the CCD positive group showed statistically significant differences between the bilateral cerebellar hemispheres (P < 0.001), and the D﹡ values of the healthy side decreased compared with the affected side, while the f values increased. There was a positive correlation between AI_D﹡ and AI_CBF (r = 0.502，P = 0.001), and a negative correlation between AI_f and AI_CBF (r = －0.347，P = 0.003). The parameters AI_D﹡ and AI_f showed high accuracy for detecting CCD with the areas under the curve of 0.744 and 0.717, respectively.

Conclusion: IVIM-derived quantitative parameters are useful to detect CCD after supratentorial ischemic stroke.