扩散峰度成像评估多发性硬化患者脑结构网络拓扑属性改变及其微结构损伤  

Diffusion kurtosis imaging in assessment of structural brain network topology alteration and microstructural damage in patients with multiple sclerosis

作　　者：严子淳 丁爽 史卓伟 祝琦源 殷菲悦 王小花 谭泽昀 李咏梅[1] Yan Zichun;Ding Shuang;Shi Zhuowei;Zhu Qiyuan;Yin Feiyue;Wang Xiaohua;Tan Zeyun;Li Yongmei(Department of Radiology,the First Affiliated Hospital of Chongqing Medical University,Chongqing 400016,China;Department of Radiology,the Childrens′Hospital of Chongqing Medical University,Chongqing 400015,China;College of Medical Informatics,Chongqing Medical University,Chongqing 400016,China)

机构地区：[1]重庆医科大学附属第一医院放射科,重庆400016 [2]重庆医科大学附属儿童医院放射科,重庆400015 [3]重庆医科大学信息学院,重庆400016

出　　处：《中华放射学杂志》2023年第11期1222-1230,共9页Chinese Journal of Radiology

基　　金：重庆市科学技术局技术创新与应用发展(重点项目)(CSTC2021 jscx-gksb-N0008);重庆市科卫联合医学科研重点项目(2023ADXM006);重庆医科大学附属第一医院重庆医科大学博士研究生创新项目(CYYY-BSYJSCXXM-202311)。

摘　　要：目的探讨多发性硬化(MS)患者脑结构网络拓扑属性改变及其微结构损伤,并分析其与认知功能的相关性。方法回顾性分析2021年5月至2022年9月于重庆医科大学附属第一医院确诊为MS患者114例(MS组)的临床和影像资料。另招募同期71名志愿者作为健康对照组(HC组)。对所有受检者进行认知功能评估及3D-T1磁化准备快速梯度回波成像序列、3D液体衰减翻转恢复序列和扩散峰度成像(DKI)扫描。采用GRETNA软件获取网络拓扑属性,全局属性包括全局效率、局部效率和小世界属性[聚类系数(Cp)、最短路径长度(Lp)、标准化Cp(γ)、标准化Lp和小世界指数(σ)]。局部属性包括节点介数(BC)、节点度(DC)、节点集聚系数(NCp)、节点效率、节点局部效率(NLe)和节点最短路径长度。采用后处理软件生成的DKI参数图,提取脑结构网络局部属性异常脑区的DKI参数值。采用独立样本t检验或Mann-WhitneyU检验分析2组全局属性、局部属性差异及局部属性异常脑区DKI参数值[各向异性分数(KFA)、平均峰度(MK)、径向峰度(RK)和轴向峰度(AK)值]的差异,并经错误发现率(FDR)校正。采用Spearman或Pearson相关分析评估MS组异常脑结构网络拓扑属性与认知量表评分的相关性。结果MS组和HC组结构网络均表现出小世界属性,MS组的γ和σ值均明显低于HC组(FDR校正,P<0.05)。与HC组相比,MS组BC、DC、NCp、NLe值广泛降低,主要位于双侧额叶、颞叶、楔前叶、杏仁核及丘脑(FDR校正,P<0.05)。经过FDR校正,与HC组比较,MS组23个网络局部属性异常脑区中KFA、MK、RK、AK值在多个脑区改变差异有统计学意义(FDR校正,P<0.05)。相关分析显示,经过FDR校正,MS患者右侧壳核的DC值与数字广度试验(DST)量表评分呈正相关(r=0.318,P=0.001)。结论MS患者脑结构网络存在广泛改变且伴有不同程度的微结构损伤,基底神经节壳核区域节点度降低与认知功能障碍存在相关Objective To investigate the changes in structural brain network topology and microstructural damage in patients with multiple sclerosis(MS),and to analyze its correlation with cognitive function.Methods Clinical and imaging data of 114 patients with MS(MS group)diagnosed in the First Affiliated Hospital of Chongqing Medical University from May 2021 to September 2022 were analyzed retrospectively.In addition,71 volunteers were recruited as a healthy control group(HC group)during the same period.All subjects were performed on cognitive assessment and 3D-T1 magnetization-prepared rapid gradient echo,3D-fluid-attenuated inversion recovery,and diffusion kurtosis imaging(DKI)scans.GRETNA software was used to obtain network topology attributes,and global attributes included global efficiency,local efficiency,and small-world attributes[clustering coefficient(Cp),shortest path length(Lp),normalized Cp(γ),normalized Lp,and small-world index(σ)].Local attributes included betweenness centrality(BC),degree centrality(DC),nodal clustering coefficient(NCp),nodal efficiency,nodal local efficiency(NLe)and nodal shortest path length.The DKI parameter map generated by the post-processing software was used to extract the DKI parameter values of the brain region with abnormal local topology of the brain structure network.The differences of global attributes,local attributes and DKI parameter values[kurtosis fractional anisotropy(KFA),mean kurtosis(MK),radial kurtosis(RK)and axial kurtosis(AK)values]were analyzed by independent sample t-test or Mann-Whitney U test,and corrected by false discovery rate(FDR).Spearman or Pearson correlation analysis was used to evaluate the correlation between abnormal brain structure network topology attributes and cognitive scale scores in the MS group.Results Both the MS group and the HC group structure network showed small-world attributes,and theγandσvalues of the MS group were significantly lower than those in the HC group(FDR correction,P<0.05).Compared with the HC group,BC,DC,NCp and NLe bro

关 键 词：多发性硬化 脑结构网络 扩散峰度成像 微结构损伤 认知功能 

分 类 号：R744.51[医药卫生—神经病学与精神病学]