The investigation of hippocampal free‐water as a biomarker for cognitive impairment and decline in older adults

Abstract

AbstractBackgroundWhile diffusion MRI derived free‐water (FW) metrics have shown promise in predicting cognitive impairment and decline in aging and Alzheimer’s disease (AD), it is unclear whether hippocampal FW metrics would be superior to traditional structural neuroimaging biomarkers (i.e., hippocampal volume, AD Schwarz Signature, AD McEvoy Signature). We examined the relationships between hippocampal FW metrics and traditional neuroimaging biomarkers with longitudinal memory and executive function.MethodVanderbilt Memory & Aging Project participants [n = 298, age at baseline: 73 ± 7 years, 40% mild cognitive impairment (MCI)] underwent longitudinal neuropsychological assessment and brain MRI over a 7 year follow‐up (mean: 3 ± 2 years) period. Structural MRI was used to quantify hippocampal volume in addition to Schwarz and McEvoy signatures. Diffusion MRI in conjunction with the FW elimination technique was used to quantify hippocampal FW and FW‐corrected fractional anisotropy (FAFWcorr). Linear regression was used to determine the association of each biomarker with baseline memory and executive function. A subsequent competitive model analysis was conducted to determine how much additional variance was provided by each biomarker compared to the covariate only model, which included age, sex, race/ethnicity, apolipoprotein‐e4 status, cognitive status, and modified Framingham Stroke Risk Profile scores. Additionally, linear mixed‐effects regression was used to determine the association between baseline biomarker values with longitudinal memory and executive function scores. All analyses were corrected for multiple comparisons using the FDR procedure.ResultAt baseline, all biomarkers (excluding hippocampal FAFWcorr) were significantly associated with cognitive performance (Table 1), with hippocampal FW exhibiting the most significant association for memory (pFDR = 3.92×10−6) and executive function (pFDR = 5.35×10−3). Competitive models revealed that hippocampal FW contributed the most unique variance beyond covariates for memory (ΔRadj 2 = 3.87%) and executive function (ΔRadj 2 = 1.59%). Longitudinal models (Table 2, Figure 1) largely mirrored cross‐sectional models by demonstrating that hippocampal FW was most significantly associated with longitudinal decline in memory (pFDR = 5.04×10−13) and executive function (pFDR = 6.15×10−5).ConclusionThis study shows that hippocampal FW is a sensitive biomarker for cognitive impairment and decline, and provides strong evidence for further exploration of this measure in aging and AD.