Study On The Correlation Between Impaired Structural Network Properties And Cognitive Impairment In Patients With White Matter Hyperintensity

Background: White matter hyperintensity(WMH)presumed vascular origin is one of the common MRI makers of cerebral small vessel disease(CSVD).It is seen as limited or diffuse low-density shadow in CT,and as high signal lesions on T2-weighted or fluid-attenuated inversion recovery(FLAIR)sequences.The underlying pathology of WMH mostly reflects demyelination and axonal loss as consequence of chronic ischemia of cerebral small or microvascular vessels.WMH is closely related to cognitive impairment(CI)and is the main cause of vascular dementia in elderly population.Nowadays,the underlying mechanism of CI caused by WMH remains to be elucidated,and early diagnostic methods and clinical interventions are lacking.Previous studies have shown that WMH can disrupt nerve fasciculus integrity and white matter connection,resulting in CI.However,no studies have further explored the change of white matter network properties in WMH population and its correlation with cognitive function.The correlation between white matter network properties and CI has been confirmed in CSVD cohort.Therefore,exploring the change of white matter network properties in WMH patients and its correlation with CI could be helpful for clinical understanding of the potential mechanism and early evaluation of WMH related CI.Objectives: To explore the change rules of white matter structural network topological properties of WMH patients and its correlation with WMH burden and cognitive function,providing new insight and theoretical basis for elucidating the potential mechanism and early diagnosis of WMH related CI.Overall,83 subjects with WMH who met the inclusion criteria and 52 healthy elderly controls(HC)were screened in the study of CSVD cohort from January 2017 to April 2019 in inpatients and outpatients of neurological department,Nanjing Drum Tower Hospital.WMH subjects were further divided into WMH-CI group(n=41)and WMH-no-CI group(n=42)based on the Montreal Cognitive Assessment(Mo CA)and years of education.All participants underwent multimodal brain magnetic resonance imaging(MRI)scans and standardized diagnostic evaluations which included demographic data,vascular risk factors and an examination of neuropsychological status.Diffusion tensor imaging(DTI)deterministic fiber tracking and graph theoretical analysis were applied to construct the structural network.We compared network properties and clinical features among the three groups.Multiple linear regression analysis was performed to investigate the relationships among WMH volumes,impaired network properties and cognitive functions in WMH-CI group.Receiver operating characteristic(ROC)curve were used to evaluate the role of impaired network properties in the recognition of WMH related CI.Results: All subjects of the three groups showed small-world properties(γ>>1and λ ≈ 1,or δ>>1).Compared with the HC,both WMH groups showed decreased global efficiency(Eg)and increased characteristic path length(Lp)at the level of whole brain(Bonferroni-corrected,P<0.05/3).WMH-CI group displayed more profound impairments of nodal efficiency(NE)and nodal path length(NLp)within multiple regions including precentral,cingulate and medial temporal gyrus(FDR correction,q = 0.01).The disrupted network properties were associated with CI and WMH burdens in WMH-CI group.Furthermore,a mediation effect of NLp in the opercular of left inferior frontal gyrus(IFGoperc.L)was observed for the association between periventricular WMH(PWMH)and memory deficit(indirect effect:-0.486;95% CI:-1.341,-0.023).ROC curve analysis showed that the impaired network properties could effectively distinguish WMH patients with or without CI.The areas under the curve was 0.85(95% confidence interval: 0.765-0.935;P<0.001)with the sensitivity was 85.37% and the specificity was 80.95%.Conclusions Brain structural network of WMH patients is significantly disturbed and this disturbance is related to the severity of WMH burden and CI.Moreover,it can effectively identify WMH related CI.Increased NLp in IFGoperc.L was shown to be a mediation framework between PWMH and WMH-related memory deficit.All of which shed new lights on investigating underlying mechanisms and early evaluation of WMH related CI.