Elevated pulse pressure predicts longitudinal accumulation of tau PET in older adults without dementia: Neuroimaging / Optimal neuroimaging measures for tracking disease progression

Background Cardiovascular risk is associated with Alzheimer’s disease (AD) pathology, with possible mechanisms including increased blood‐brain barrier permeability, white matter changes, inflammatory cytokine activation, and neuronal vulnerability due to cerebral hypoperfusion. Prior research has linked pulse pressure (PP), an index of arterial stiffening, to abnormal cerebrospinal fluid measures of tau. However, to our knowledge, no research has examined PP in relation to tau PET imaging. Method Data were analyzed for 139 older adults without dementia from the Alzheimer’s Disease Neuroimaging Initiative who had at least two tau PET scans. Baseline PP was measured as systolic ‐ diastolic blood pressure (with higher values denoting increased arterial stiffening). Tau PET was measured with the AV‐1451 tracer, intensity normalized, partial‐volume corrected, and parcellated into cortical regions using FreeSurfer. A priori regions of interest (ROIs) included AD‐vulnerable cortical areas across Braak stages I‐V including the entorhinal cortex (EC), the inferior temporal gyrus (ITG), and the inferior parietal lobule (IPL). Iterated re‐weighted least squares regression, robust to the influence of outliers, modeled the association between baseline PP and tau PET accumulation across 12 months within each ROI adjusting for age, sex, cognitive diagnosis (i.e., normal or mild cognitive impairment), APOE status, and baseline tau PET signal. Result Higher PP predicted higher baseline tau PET levels only in the ITG ( t = 2.09, p = .04), although it was strongly predictive of increased tau PET accumulation over 12 months for all three ROIs (EC t = 3.15, p = .002; ITG t = 2.98, p = .003; IPL t =2.62, p = .01) over and above the effects of covariates. Conclusion Results indicate that higher PP is strongly predictive of tau PET accumulation across multiple AD vulnerable regions, whereas it is less predictive of baseline tau PET levels. Further, associations were stronger within earlier Braak regions. Findings confirm prior CSF studies and further suggest that arterial stiffening may be especially associated with a trajectory of increasing tau burden. Although the nature of the relationship between vascular risk and tau PET may be bidirectional, our results highlight the importance of cardiovascular intervention in the context of encroaching AD pathology.