Premium
[IC‐P‐026]: FUNCTIONAL CONNECTIVITY WITHIN NEUROANATOMICAL SUBSTRATES OF ALZHEIMER's DISEASE
Author(s) -
Bishnoi Ram J.,
Zhang Wei,
Fox Peter T.
Publication year - 2017
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2017.06.2298
Subject(s) - voxel , resting state fmri , functional connectivity , grey matter , alzheimer's disease , node (physics) , pattern recognition (psychology) , computer science , neuroscience , disease , artificial intelligence , psychology , medicine , pathology , magnetic resonance imaging , physics , quantum mechanics , white matter , radiology
impairment (MCI). Methods: 19 patients with MCI and 18 agematched controls underwent magnetic resonance imaging and cognitive assessment. Diffusion tensor imaging-based tractography was used to construct whole-brain tractograms, which were represented as network graphs weighted by fractional anisotropy of the edge between any two nodes. Whole-brain graph theoretical measures and local measures for the hippocampi and thalami were compared between groups and correlated with cognitive scores. Linear regression models were constructed to investigate how local and whole-brain measures predict episodic memory. Results:MCI patients showed reduced global efficiency (t1⁄42.6, p1⁄4.019) and mean clustering coefficient (t1⁄43.1, p1⁄4.005). In MCI, global efficiency was associated with episodic memory performance (r1⁄4.50, p1⁄4.042). Local efficiency of the left hippocampus was decreased in MCI (t1⁄42.1, p1⁄4.045), while there were no group differences in clustering and efficiency for the right hippocampus and the thalami. Episodic memory correlated with local measures of the left hippocampus (local efficiency: r1⁄4.68, p1⁄4.003; clustering coefficient: r1⁄4.59, p1⁄4.012) and the left thalamus (local efficiency: r1⁄4.51, p1⁄4.038; clustering coefficient: r1⁄4.48, p1⁄4.049). Measures of network topology did not correlate with cognition in controls. When global efficiency was added to the regression models for episodic memory performance in MCI, the relationships with local efficiency and the clustering coefficient for the left thalamus were no longer significant. In contrast, measures of the left hippocampus retained independent relationship in combined regression models. Conclusions:Both local and whole-brain measures of network topology correlate with episodic memory in MCI. The influence of hippocampal connections is independent of global network structure. In contrast, alterations in global efficiency mediate the effect of thalamic connectivity on memory performance. Our results challenge the hippocampal-centred view by suggesting that episodic memory impairment might also be dependent on a distributed mnemonic system.