Body mass index and allostatic load are directly associated with longitudinal increase in plasma neurofilament light among urban middle-aged adults

Date

2021-10-29

Department

Program

Citation of Original Publication

May A Beydoun, Nicole Noren Hooten, Ana I Maldonado, Hind A Beydoun, Jordan Weiss, Michele K Evans, Alan B Zonderman, Body Mass Index and Allostatic Load are Directly Associated with Longitudinal Increase in Plasma Neurofilament Light Among Urban Middle-aged Adults, The Journal of Nutrition, 2021;, nxab381, https://doi.org/10.1093/jn/nxab381

Rights

This is a pre-copyedited, author-produced version of an article accepted for publication in The Journal of Nutrition following peer review. The version of record May A Beydoun, Nicole Noren Hooten, Ana I Maldonado, Hind A Beydoun, Jordan Weiss, Michele K Evans, Alan B Zonderman, Body Mass Index and Allostatic Load are Directly Associated with Longitudinal Increase in Plasma Neurofilament Light Among Urban Middle-aged Adults, The Journal of Nutrition, 2021;, nxab381, https://doi.org/10.1093/jn/nxab381 is available online at: https://academic.oup.com/jn/advance-article/doi/10.1093/jn/nxab381/6413996?login=true#311625329
Access to this item will begin on 10/29/22.

Subjects

Abstract

Background Plasma neurofilament light, a novel biomarker for age-related neurodegenerative disease, may be linked to cardiometabolic risk factors, including body mass index (BMI), the allostatic load total score (ALtotal) and related continuous components (ALcomp). These relationships may differ by sex and race. Methods We used data from the Healthy Aging in Neighborhoods of Diversity across the Life Span (HANDLS) study [n = 608, Age at visit 1 (v1:2004-2009): 30-66y, 42% male, 58% African American] to investigate associations of initial cardiometabolic risk factors and time-dependent plasma NfL levels over 3 visits (2004-2017; mean follow-up time±SD:7.72±1.28y), with outcomes being NfLv1 and annualized change in NfL (δNfL). We used mixed-effects linear regression and structural equations modeling (SM). Results BMI was associated with lower initial (γ01 = –0.014±0.002, P<0.001) but faster increase in plasma NfL over time (γ11 = +0.0012±0.0003, P<0.001), a pattern replicated for ALtotal. High sensitivity C-reactive protein (hsCRP), serum total cholesterol, and resting heart rate at v1 were linked with faster plasma NfL over-time increase overall, while being uncorrelated with NfLv1 (e.g. hsCRP×TIME, full model: γ11 = +0.004±0.002, P = 0.015). In SM analyses, BMI's association with δNfL was significantly mediated through ALtotal among women (Total effect = +0.0014±0.00038, P<0.001; Indirect effect = +0.00042±0.00019, P = 0.025; mediation proportion = 30%), with only a direct effect detected among African Americans (Total (effect = +0.0011±0.0004, P = 0.015; Direct effect = +0.0010±0.00048, P = 0.034). The positive associations between ALtotal/BMI and δNfL were mediated through increased HbA1c levels, overall. Conclusions Cardiometabolic risk factors, particularly elevated HbA1c, should be screened and targeted for neurodegenerative disease, pending comparable longitudinal studies. Other studies examining the clinical utility of plasma NfL as a neurodegeneration marker should account for confounding effects of BMI and AL.