Methods: A total of 436 participants from the Radboud University Nijmegen Diffusion Tensor and Magnetic Resonance Cohort (RUN DMC), a prospective hospital-based cohort of elderly without dementia with cerebral SVD, were included in 2006. During follow-up (2011–2012), dementia was diagnosed. The structural network was constructed from baseline diffusion tensor imaging followed by deterministic tractography and measures of efficiency using graph theory were calculated. Cox proportional regression analyses were conducted.
Results: During 5 years of follow-up, 32 patients developed dementia. MRI markers for SVD were strongly associated with network measures. Patients with dementia showed lower total network strength and global and local efficiency at baseline as compared with the group without dementia. Lower global network efficiency was independently associated with increased risk of incident all-cause dementia (hazard ratio 0.63, 95% confidence interval 0.42–0.96, p = 0.032); in contrast, individual SVD markers including lacunes, white matter hyperintensities volume, and atrophy were not independently associated.
Conclusions: These results support a role of network disruption playing a pivotal role in the genesis of dementia in SVD, and suggest network analysis of the connectivity of white matter has potential as a predictive marker in the disease.