基于体素的形态学测量方法诊断轻中度血管性痴呆的可行性研究

The feasibility of voxel based morphometry in the diagnosis of mild to moderate vascular dementia

目的:探讨基于体素的形态学测量方法(VBM)分析MRI图像诊断轻中度血管性痴呆(VaD)的可行性。方法:搜集轻中度VaD患者(VaD组)及同年龄段、受教育年限相匹配的无认知功能障碍者(NCI组),获得受试者T1WI结构像,采用SPM8软件对受试者图像进行处理与统计学分析。将存在统计学差异的脑区容积值与认知功能量表行相关性分析,具有相关性的脑区分别绘制ROC曲线,获得具有最佳诊断效能的脑区及分界值。结果:2组受试者临床基础资料差异无统计学意义(均P>0.05),MMSE、MoCA量表评分差异均有统计学意义(均P<0.05)。与NCI组比较,VaD组双侧丘脑、双侧中央后回及右侧海马等脑区灰质容积减小。双侧中央后回、双侧丘脑、右侧海马、左侧颞中回及左侧顶下缘角回的容积与MMSE、MoCA量表评分均具有正相关性,其中左侧中央后回、左侧顶下缘角回的AUC最大,均为0.83。结论:应用VBM方法分析颅脑MRI图像可为轻中度VaD的诊断提供新的影像学生物标志物。

Objective:To explore the feasibility of voxel based morphometry(VBM)in the diagnosis of mild to moderate vascular dementia(VaD).Methods:The patients with mild to moderate VaD(VaD group)and non-cognitive impairment group(NCI group)with matched age and education years were recruited.Statistical Parametric Mapping(SPM8)was used to process and statistically analyze 3D T1WI.The brain regions with different volume between VaD group and NCI group were shown.Correlation analysis was conducted to investigate the correlation between the volume of these brain regions and clinical cognitive scales scores.The ROC curves of the brain regions exhibiting significant correlations were drawn and obtain brain regions with the maximum diagnostic efficiency and cut-off value.Results:There were no statistical differences in gender,age and education years between the two groups.There were statistical differences in MMSE and MoCA scores(both P<0.05).Compared with NCI group,the volumes of gray matter in bilateral thalamus,bilateral posterior central gyrus and right hippocampus decreased in VaD group.The volumes of left posterior central gyrus,right posterior central gyrus,left thalamus,right thalamus,right hippocampus,left middle temporal gyrus and left parietal inferior marginal angular gyrus were positively correlated with MMSE and MoCA scores,and the AUC of the left posterior central gyrus and left inferior parietal angular gyrus was the largest(0.83).Conclusion:VBM can provide a new imaging biomarker for the diagnosis of mild to moderate VaD.