Longitudinal non-linear changes in the microstructure of the hippocampal subfields in older adults

Abstract

Human brains undergo considerable morphologic variation with age, a primary risk factor for neurodegenerative disorders. While aging often causes neurocognitive decline, its governing biological mechanisms remain unclear. These age-related brain microstructural changes may be quantified by advanced diffusion MRI (dMRI) with tissue-specific compartment modeling approach. This longitudinal study investigates age-related differences in hippocampal subfields vulnerable to early stages of Alzheimer's disease (AD). Thirty-seven cognitively normal (CN) older adults (70.6 ± 6.7 years) from the Indiana Alzheimer's Disease Research Center (IADRC) underwent baseline and follow-up MRI scans, within 24 ± 11.7 months. Grey matter-specific multi-compartment diffusion model, cortical-neurite orientation dispersion, and density imaging (cortical-NODDI) was used to derive diffusion microstructural metrics, namely orientation dispersion index (ODI) and neurite density index (NDI) in hippocampal-subfields (CA1-3, CA4DG, and subiculum). We investigated rate of change in diffusion metrics and its associations with age and baseline diffusion metrics in hippocampal subfields using linear regression analysis, after adjusting for confounding factors (i.e., sex, education, Apolipoprotein E (APOE) ε4, and baseline subfield volumes). CA1-3 and subiculum volumes significantly decreased between baseline and follow-up scans. ODI rate of change was significantly higher than zero in CA4DG, while rate of change in NDI was significantly lower than zero in CA1-3 and CA4DG. ODI rate of change in CA1-3 was significantly associated with baseline age of participants and initial microstructural value of ODI in CA1-3. Results showed that Cornu Ammonis is most sensitive to age-related changes with increased microstructural dispersion and decreased neurite density with age- and initial state-dependent changes.