Lipid levels and the risk of dementia: A dose-response meta-analysis of prospective cohort studies

Objectives: We performed a dose-response meta-analysis to estimate the association between lipid profiles with the risk of dementia and the potential differences according to the subtype of dementia based on prospective studies. Methods: We searched PubMed, Embase and Web of Science for relevant articles and performed a meta-analysis. We applied fixed or random-effects models to calculate pooled relative risk (RR) with their 95% confidence intervals (CI). The dose-response relationship was assessed by restricted cubic spline. Results: Twenty-five prospective studies comprising 362,443 participants and 20,121 cases were included in the final analysis. We found that increased risk of all-cause dementia could be predicted by elevated total cholesterol (TC) (RR = 1.13, 95% CI 1.04-1.22). When looking at dementia subtypes, we also observed high TC and triglycerides (TG) may increase the future risk of Alzheimer's disease (AD), with a pooled RR of 1.13 (95% CI: 1.06-1.21) and 1.10 (95% CI: 1.04-1.15) respectively. Moreover, a dose-response analysis revealed a linear association between TC or TG and the risk of AD, with a pooled RR of 1.09 (95% CI: 1.02-1.16) and 1.12 (95% CI: 1.05-1.21) for per 3-mmol/L increment in TC and TG, respectively. Conclusions: Current evidence suggest that every 3-mmol/L increase in blood TC or TG is linearly associated with a 9% or 12% increase in RR of AD, supporting the notion that high TC and TG levels appear to play a causal role in the development of AD.