Cardiometabolic disorders are associated with reduced cerebral perfusion and white matter microstructure

Abstract

AbstractBackgroundCardiometabolic disorders (hypertension and diabetes) often co‐occur and are considered key modifiable risk factors for Alzheimer’s disease and related disorders; yet, the extent to which they each affect brain structure and function is unclear. We examined their associations with neuroimaging biomarkers of white matter microstructure, cerebral perfusion, and cognitive function in a diverse cohort of older adults.MethodParticipants aged 50‐85 years enrolled in Wake Forest Alzheimer’s Disease Research Center’s Clinical Core between 2016‐2020 received clinical evaluation, oral glucose tolerance testing (OGTT), neuroimaging, cognitive testing, and adjudication for mild cognitive impairment (MCI) and dementia. 3T brain MRI included T1 (for gray/white matter segmentation, regional volumes/thicknesses); FLAIR for white matter hyperintensity volume (WMHv); arterial spin labeling for cerebral blood flow (CBF); diffusion tensor imaging (DTI) for fractional anisotropy (FA); and neurite orientation dispersion and density imaging (NODDI) for Free Water. Hypertension (HTN) was staged using seated blood pressure (BP) and medications. Hyperglycemia was staged using OGTT 120 minute glucose levels (OGTT120), hemoglobin A1c, and fasting glucose as available. Cardiometabolic status was categorized (HTN only, hyperglycemia only, hyperglycemia+HTN, neither). Multivariable linear regression modeled associations with neuroimaging measures adjusted for age, gender, and race.ResultMRI was available on 478 participants (35% MCI and 10% dementia) with mean age of 70±8 years, 74% with HTN, 61% with hyperglycemia (4% with non‐insulin treated diabetes), and 15% African‐Americans. The combination of hyperglycemia+HTN was associated with significantly higher NODDI Free Water and WMHv, lower FA, and lower gray matter perfusion, compared to neither factor (Table). Continuous levels of systolic BP and OGTT120 broadly correlated with WM microstructure, WMHv, and gray matter CBF (Figure). In adjusted models, systolic BP remained significantly associated with higher NODDI Free Water and WMHv, and CBF in gray and white matter. OGTT120 and WMHv remained significant in adjusted models. Cardiometabolic factors were not associated with gray matter macrostructure (e.g. volumes, temporal lobe cortical thickness) or cognitive status in this sample.ConclusionIn this deeply phenotyped and racially diverse sample, individual cardiometabolic factors and the co‐occurrence of disorders was associated with reduced cerebral gray matter perfusion and reduced white matter microstructure.