Small-world Brain Network And Dynamic Functional Distribution In Patients With Subcortical Vascular Cognitive Impairment

Background and Objective Recent advances in graph theoretical approaches allowus to characterize topological properties of complex networks. FMRI studies ofsubcortical vascular cognitive impairment (SVCI) have revealed abnormal functionalconnectivity. However, whether the functional network of SVCI patients is disruptedremains controversial. we aimed to investigate topological properties of the functionalbrain connectivity relationship between the brain network metrics and cognitive taskperformance in patients with SVCI using graph theoretical analysis.Materials and Methods Twenty-three patients with SVCI and20healthy individualswere recruited from the department of Neurology, First Affiliated Hospital of AnhuiMedical University in this case-control study. All of them were subject to the3.0T MRIscan and the cognitive assessment including MMSE (Mini-mental State Examination),CAMCOG-C(the Cambridge Cognitive Examination-Chinese Version) andCDR(Clinical Demential Rating) and so on. Graph theoretical analysis was applied toresting state function MRI data based on SPMZ(SPM8, http://www.fil.ion.ucl.ac.uk/spm) and DPARSFV (Data Processing Assistant for Resting-StatefMRI) software. Functional connectivities between90cortical and subcortical brain regions were estimated using bivariate correlation analysis, thresholded to construct aset of undirected graphs. The relationship between network properties and cognitivevariables were tested by computing the correlation between graph metrics and taskperformace.Results Functional brain networks of both SVCI patients and controls showedsmall-world attributes over a range of thresholds（0.15<Sthr<0.4). Global topologicalorganization of the functional brain networks in SVCI was significantly disrupted,indicated by reduced global and local efficiency, clustering coefficient, nodal degree,betweenness centrality and increased characteristic path length, relative to normalsubjects. Comparing with controls, SVCI data showed decreased activity areaspredominantly targeted in frontal-temporal lobes, such as median cingulate andparacingulate gyri bilaterally, middle temporal gyrus and superior temporal gyrus, whilesubcortical regions showed increased nodal centrality or degree which was suspected tocompensate for the inefficiency function network. In SVCI, altered brain networkproperties closely correlate with general cognition dysfunction as mainly indicated byCAMCOG-C performance.Conclusions Global and local function networks topological properties are altered inSVCI, suggesting a loss of efficiency in information exchange between both, distant andclose brain areas. This aberrant topology of the functional brain networks in SVCIappears to underlie cognitive deficits in these patients.