2型糖尿病合并轻度认知障碍患者脑葡萄糖代谢连接网络拓扑结构分析

Topological Analysis of Brain Glucose Metabolic Connectivity Networks of Patients with Type 2 Diabetes Mellitus Combined with Mild Cognitive Impairment

目的:分析2型糖尿病(T2DM)合并轻度认知障碍(MCI)(T2DM&MCI)患者脑葡萄糖代谢连接网络拓扑结构的改变情况。方法:选择阿尔茨海默病神经影像学倡议(ADNI)数据库收录的T2DM&MCI受试者(观察组)和健康受试者(对照组),每组35例。从数据库获取2组简易智能精神状态检查量表(MMSE)评分、患者报告日常认知量表的记忆评分(EcogPtMem)、患者亲属报告日常认知量表的记忆评分(EcogSPMem)以及氟代脱氧葡萄糖-电子发射断层扫描(FDG-PET)图像。利用SPM12将FDG-PET图像标准化到蒙特利尔神经病学研究所(MNI)空间脑模板上。以小脑作为参考区,提取每个受试者解剖自动标记法(AAL)模块脑区的标准摄取值比率(SUVR),构建葡萄糖代谢连接脑网络。采用GRETNA软件分析全局效率、度中心性、节点局部效率等脑网络拓扑属性;采用Spearman相关性分析差异脑区度中心性与EcogPtMem评分以及EcogSPMem评分的相关性、节点局部效率与EcogPtMem评分以及EcogSPMem评分的相关性。结果:与对照组比较,观察组MMSE评分明显更低,EcogPtMem和EcogSPMem评分均明显更高,差异具有统计学意义(P<0.05)。与对照组比较,观察组的全局效率组间差异无统计学意义(P>0.05)。与对照组比较,观察组左侧海马旁回(P=0.005)、左侧海马(P=0.013)、左侧顶下缘角回(P=0.031)、左侧角回(P=0.034)、右侧角回(P=0.013)、右侧顶下缘角回(P=0.044)、右侧海马旁回(P=0.024)度中心性明显降低;左侧直回(P=0.028)、左侧海马旁回(P=0.044)、左侧海马(P=0.026)、左侧顶下缘角回(P=0.031)、左侧角回(P=0.024)、右侧角回(P=0.004)、右侧顶下缘角回(P=0.039)节点局部效率明显降低。相关性分析结果显示,观察组左侧海马度中心性(r=-0.273,P=0.022)和左侧海马旁回度中心性(r=-0.341,P=0.004)与EcogPtMem评分呈负相关关系;左侧海马度中心性(r=-0.391,P=0.001)、左侧海马旁回度中心性(r=-0.410,P<0.001)、右侧海马旁回度中心性(r=-0.240,P=0.045)与EcogSPMem评分呈负相关关系。观察组左侧海马节点局部效率(r=-0.257,P=0.032)、左侧海马旁回节点局部效率(r=-0.251,P=0.036)、右侧角回节点局部效率(r=-0.265,P=0.027)与EcogPtMem评分呈负相关关系;左侧海马节点局部效率(r=-0.363,P=0.002)和左侧海马旁回节点局部效率(r=-0.362,P=0.002)与EcogSPMem评分呈负相关关系。结论:T2DM&MCI患者脑葡萄糖代谢连接网络拓扑结构异常可能是引起其记忆功能下降的原因之一,这可为探究T2DM&MCI神经机制提供参考。

Objective:To analyze the changes in the topology of the brain glucose metabolic connectivity networks of patients with type 2 diabetes mellitus(T2DM)combined with mild cognitive impairment(MCI).Methods:A total of 35 T2DM&MCI subjects(observation group)and 35 healthy subjects(control group)were included from the Alzheimer's Disease Neuroimaging Initiative(ADNI)database.Mini-mental state examination(MMSE)score,everyday cognition:the patient-reported memory(EcogPt‐Mem)score,everyday cognition:the study partner-reported memory(EcogSPMem)score,and fluoro-2-deoxyglucose(FDG)-positron emission tomography(PET)images of both groups were included from the ANDI database.FDG-PET images were normalised to the Montreal neurological institute(MNI)spatial brain template by SPM 12 software.Using the cerebellum as reference area,standardized uptake value ratio(SUVR)of the anatomical automatic labeling(AAL)module brain regions of each subject was extracted to construct the glucose metabolism connectivity brain network.GRETNA software was used to analyze the topological properties of the brain network,including global efficiency,degree centrality,and node local efficiency.Spearman's correlation analysis was used to analyze the correlation of differential brain area degree centrality with EcogPtMem score as well as EcogSPMem score,and the correlation of node local efficiency with EcogPtMem score as well as EcogSPMem score.Results:Compared with the control group,MMSE score of the observation group was significantly lower,EcogPtMem and EcogSPMem scores were significantly higher,and the differences were statistically significant(P<0.05).There was no statistically significant difference of the global efficiency between the control group and the observation group(P>0.05).Compared with the control group,the degree centralities of the left parahippocampal gyrus(P=0.005),left hippocampus(P=0.013),left subparietal marginal angular gyrus(P=0.031),left angular gyrus(P=0.034),right angular gyrus(P=0.013),right subparietal marginal angular gyrus(P=0.044),and right parahippocampal gyrus(P=0.024)of the observation group were significantly lower.The nodal local efficiencies of the left straight gyrus(P=0.028),left parahippocampal gyrus(P=0.044),left hippocampus(P=0.026),left subparietal angular gyrus(P=0.031),left angular gyrus(P=0.024),right angular gyrus(P=0.004),right subparietal angular gyrus(P=0.039)decreased significantly.Correlation analyses showed that,degree centrality of left hippocampal(r=-0.273,P=0.022)and left hippocampal parahippocampal gyrus(r=-0.341,P=0.004)were negatively correlated with EcogPtMem score.The degree centrality of left hippocampal(r=-0.391,P=0.001),left hippocampal parahippocampal gyrus(r=-0.410,P<0.001),and right hippocampal parahippocampal gyrus(r=-0.240,P=0.045)were negatively correlated with EcogSPMem score.The nodal local efficiencies of the left hippocampal(r=-0.257,P=0.032),left parahippocampal gyrus(r=-0.251,P=0.036),and right angular gyrus(r=-0.265,P=0.027)were negatively correlated with the EcogPtMem score.The nodal local efficiencies of left hippocampal(r=-0.363,P=0.002)and left parahippocampal gyrus(r=-0.362,P=0.002)were negatively correlated with EcogSPMem score.Conclusion:The abnormal topological structure of brain glucose metabolic connectivity network in patients with T2DM&MCI might be one of the reasons for the decline in memory,which could provide a reference for the exploring of the neural mechanism of patients with T2DM&MCI.