MCI Diagnosis and Neighborhood Disadvantage Predict Differences in White Matter Microstructure in a Rural Aging Cohort

• Published in: Alzheimer's & dementia 2024-12, Vol.20 (S9), p.n/a
• Main Authors: Bender, Andrew R., Giriprakash, Pavithran Pattiam, Cordes, Dietmar, Zhuang, Xiaowei, Caldwell, Jessica ZK, Miller, Justin B
• Format: Article
• Language: English
• Subjects: Alzheimer's Imaging Consortium
• ISSN: 1552-5260; 1552-5279
• DOI: 10.1002/alz.094061

Background Diffusion magnetic resonance imaging (dMRI) studies show cerebral white matter alterations are associated with both mild cognitive impairment (MCI) and contextual psychosocial factors. However, these reports are almost exclusively based on older adults residing in metropolitan areas. Individuals living in rural communities face increased risk for Alzheimer’s disease and related dementias, but little is known regarding the sensitivity of dMRI markers to incident MCI or psychosocial stressors in this population. In a rural‐dwelling community cohort of older adults, this study evaluated associations between dMRI measures of white matter microstructure and both clinical MCI diagnosis and neighborhood disadvantage. Method The study sample included rural‐dwelling older adults (n = 69) from the Nevada Exploratory Alzheimer’s Disease Research Center. Participants were characterized as cognitively unimpaired (CU; n = 42) or diagnosed with MCI (n = 27) based on clinical expert consensus, and all had dMRI data and Area Deprivation Index (ADI) measures of neighborhood disadvantage. dMRI data processing leveraged the MRtrix3 fixel‐based analysis framework and free water mapping to estimate voxel‐based measures of white matter organization, transformed to a sample template. Data analysis utilized voxel‐based analysis (VBA) to model differences in clinical diagnosis and ADI scores as predictors of fiber density, crossing fiber entropy, and extracellular free water, and free water‐corrected diffusion tensor parameters, accounting for differences in age, sex, and handedness. Results The VBA models showed both MCI diagnosis and elevated neighborhood disadvantage predicted significantly greater (familywise error corrected p