Dementia is strongly associated with medial temporal atrophy even after accounting for neuropathologies

2021 
Objective: Brain atrophy is associated with degenerative neuropathologies as well as clinical status of dementia. Whether dementia influences atrophy independent of neuropathologies is not known. In this study, we examined the pattern of atrophy associated with dementia while accounting for the most common dementia-related neuropathologies. Methods: We used data from National Alzheimer9s Coordinating Center (NACC, N=129) and Alzheimer9s Disease Neuroimaging Initiative (ADNI, N=47) participants with suitable in-vivo 3D-T1w MRI and autopsy data. We determined dementia status at visit closest to MRI. We examined the following dichotomized neuropathological variables: Alzheimer9s disease neuropathology, hippocampal sclerosis, Lewy Bodies, cerebral amyloid angiopathy, atherosclerosis. Voxel-based morphometry (VBM) identified areas associated with dementia after accounting for neuropathologies. Identified regions of interest were further analyzed. We used multiple linear regression models adjusted for neuropathologies and demographic variables. Results: We found strong associations for dementia with volumes of the hippocampus, amygdala, and parahippocampus (semi-partial correlations≥0.28, P<0.0001 for all regions in NACC; semi-partial correlations≥0.35, P≤0.01 for hippocampus and parahippocampus in ADNI). Dementia status accounted for more unique variance in atrophy in these structures (~8%) compared with neuropathological variables; the only exception was hippocampal sclerosis which accounted for more variance in hippocampal atrophy (10%). Conclusion: Even after accounting for the most common neuropathologies, dementia still had among the strongest correlations with atrophy of medial temporal lobe structures. This suggests that atrophy of the medial temporal lobe is most related to clinical status of dementia as opposed to Alzheimer9s or other neuropathologies.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    34
    References
    0
    Citations
    NaN
    KQI
    []