Cross-sectional associations of cumulative triglyceride-glucose index with cognitive function and mild cognitive impairment: Evidence from two community-based cohorts

Abstract

Background The triglyceride-glucose (TyG) index is a surrogate marker of insulin resistance, implicated in cognitive decline and Alzheimer's disease (AD), but whether long-term TyG exposure is associated with cognitive health remains unclear. Objective To examine whether cumulative TyG index is associated with cognitive performance, mild cognitive impairment (MCI), and follow-up cognitive performance in two community-based cohorts. Methods We analyzed adults aged ≥60 years in the Beijing Disability Risk and Ageing Monitoring Study (BEAM) and ≥45 years in the China Health and Retirement Longitudinal Study (CHARLS). Cumulative TyG was estimated from four annual measurements (2020–2023) in BEAM and two measurements (2012 and 2015) in CHARLS. Cognitive outcomes were assessed in 2023 and 2015, respectively. Multivariable linear and logistic regression models were applied. Results A total of 3857 participants were included, comprising 585 participants from BEAM (mean age 73.0 ± 5.1 years; 64.4% women) and 3272 participants from CHARLS (mean age 58.9 ± 8.8 years; 51.4% women). Higher cumulative TyG was associated with higher global cognitive scores in BEAM (β=0.11, 95% CI 0.02–0.20, p = 0.013) and CHARLS (β=0.13, 95% CI 0.05–0.23, p = 0.003). Higher cumulative TyG was also associated with lower odds of MCI (BEAM: OR = 0.85, 95% CI 0.73–0.98, p = 0.028; CHARLS: OR = 0.92, 95% CI 0.86–0.99, p = 0.041). However, cumulative TyG was not associated with follow-up cognitive performance after adjustment for baseline cognition. Conclusions Cumulative TyG was modestly associated with higher cognitive performance and lower odds of MCI in cross-sectional analyses, but not with follow-up cognitive performance.