Premium
F2‐02‐04: NETWORK‐SPECIFICITY OF MULTIMODAL IMAGING ABNORMALITIES IN ALZHEIMER'S DISEASE
Author(s) -
Grothe Michel,
Teipel Stefan
Publication year - 2014
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.2014.04.144
Subject(s) - default mode network , neuroimaging , neuroscience , atrophy , disease , human brain , salience (neuroscience) , psychology , alzheimer's disease , pathology , medicine , functional connectivity
Background: Neuropathological studies indicate that the distinct pathologic events in Alzheimer’s disease (AD) are not randomly distributed throughout the entire brain, but spread among discrete neuronal systems in fairly consistent temporospatial patterns. Recent neuroimaging studies have emphasized the topographical similarity between AD-related brain changes and a prominent cortical association network called the defaultmode network (DMN). However, the specificity of the diverse pathologic changes for the DMN compared to other intrinsic connectivity networks (ICNs) of the limbic and heteromodal association cortex has not yet been examined systematically. Methods: We used structural MRI, AV45and FDG-PET data from 667 participants of the Alzheimer’s Disease Neuroimaging Initiative to assess AD-related patterns of amyloid deposition, hypometabolism, and gray matter atrophy, as well as their progression across disease stages from preclinical over predementia to clinically manifest AD. Complementary region-of-interest and voxel-based analyses were used to assess the correspondence of pathologic changes with the principle ICNs of the human brain, including the visual (VIS), sensorymotor (SM), dorsal attention (DAN), salience (SAL), limbic (LIM), central-executive (CEN) and DMN, as defined by a recently published functional connectivity atlas. Results: ROI analysis revealed amyloid deposition to be equally distributed across AD stages, being most pronounced in the DMN and CEN, intermediate in DAN and SAL, and relatively low in LIM, SM and VIS. Significant hypometabolismwas first detected at the predementia stage and followed the network specificity of amyloid deposition, with the exception of intermediate hypometabolism in LIM. By contrast, the single most affected network for atrophic changes was the LIM, followed by DMN, and then DAN, CEN, and SAL. Volume of SM and VIS was spared. In AD dementia, hypometabolism and atrophy were more pronounced, but the respective ordering of affected networks was the same. Additional pattern matching analysis of the voxel-wise statistical maps (goodness-offit to ICN) revealed highest correspondence of hypometabolic and atrophic changes with the DMN and LIM, respectively; whereas the pattern of highest amyloid deposition matched equally well to the DMN and CEN across disease stages. Conclusions: AD pathology selectively and progressively affects higher order cognitive networks, but not all pathologic processes show highest affinity to the DMN.