Impaired White Matter Network Topology Mediates the Association Between White Matter Hyperintensities of Presumed Vascular Origin and Cognitive in the Elderly

We aimed to investigate the alterations in white matter (WM) networks topology caused by varying white matter hyperintensities (WMH) burden, and assess the relationship between topological changes of WM networks and cognitive function in the moderate-to-severe WMH (m/sWMH) patients. The study enrolled 54 cognitively unimpaired individuals, stratified into two groups: 26 with m/sWMH and 28 with mild WMH (mWMH). We used graph theory analysis to investigate the global and nodal topological disruptions between two groups and relate WM networks topological alterations to cognitive function. The results provided that, compared to mWMH group, m/sWMH group exhibit disruptions of global properties and nodal properties, including decreased clustering coefficient (Cp) and shortest path length (Lp), decreased nodal clustering coefficient (Ncp), increased nodal efficiency (Ne), both increased and decreased nodal degree centrality (Dc) and betweenness centrality (Bc) in regions such as body of corpus callosum, left sagittal stratum (SS) (including the inferior longitudinal fasciculus and inferior fronto-occipital fasciculus, ILF and IFOF), the anterior limb of the internal capsule, cingulate gyrus, and inferior longitudinal fasciculus. The pearson’s correlation analysis revealed that the nodal properties (Ncp and Dc) in SS (including ILF and IFOF) were significantly correlation with scores of cognitions in m/sWMH group. Increased WMH burden leads to the disintegration of WM network topology, and this disruption is closely associated with cognitive dysfunction (global cognitive function, attention and processing speed). The SS may be a specific target for cognitive impairment in patients with m/sWMH.