Objective: To examine the relation between measures of whole-brain white matter connectivity and cognitive performance in patients with early Alzheimer disease (AD) using a network-based approach and to assess whether network parameters provide information that is complementary to conventional MRI markers of AD.
Methods: Fifty patients (mean age 78.8 ± 7.1 years) with early AD were recruited via a memory clinic. In addition, 15 age-, sex-, and education-matched control participants were used as a reference group. All participants underwent a 3-T MRI scan and cognitive assessment. Diffusion tensor imaging-based tractography was used to reconstruct the brain network of each individual, followed by graph theoretical analyses. Overall network efficiency was assessed by measures of local (clustering coefficient, local efficiency) and global (path length, global efficiency) connectivity. Age-, sex-, and education-adjusted cognitive scores were related to network measures and to conventional MRI parameters (i.e., degree of cerebral atrophy and small-vessel disease).
Results: The structural brain network of patients showed reduced local efficiency compared to controls. Within the patient group, worse performance in memory and executive functioning was related to decreased local efficiency (r = 0.434; p = 0.002), increased path length (r = -0.538; p < 0.001), and decreased global efficiency (r = 0.431; p = 0.005). Measures of network efficiency explained up to 27% of the variance in cognitive functioning on top of conventional MRI markers (p < 0.01).
Conclusion: This study shows that network-based analysis of brain white matter connections provides a novel way to reveal the structural basis of cognitive dysfunction in AD.