Association Between Serum Lipid Traits and Cognitive Function in Middle-aged and Elderly Adults: A Longitudinal Study.

Background: It is debatable whether demographic factors alter the relationship between serum lipid traits and cognitive function. Few data have examined the effects of non-traditional lipid metrics on the lipid-cognition relationship. We aim to test the generality of relationships between lipid traits and cognitive function in Chinese adults.

Methods: Data from 5,959 participants were obtained from the China Health and Retirement Longitudinal Study (2011-2020). The cognitive function was assessed via the Mini-Mental State Examination. Effects of traditional lipid metrics (Total Cholesterol, TC, Triglycerides, TG, Low-Density Lipoprotein, LDL, High-Density Lipoprotein, HDL) and non-traditional lipid metrics (TC/HDL, LDL/HDL) were analyzed. We employed mixed-effect models, Group-Based Trajectory Models (GBTM), and logistic regression to examine the associations between baseline serum lipid traits and cognitive function.

Results: As continuous variables, higher TG levels were correlated with higher cognitive scores (P=0.036), and similar patterns were found in TC/HDL (P < 0.01) and LDL/HDL (P < 0.01). In contrast, higher HDL levels were associated with lower cognitive scores. Similar trends were observed when lipid traits were analyzed as categorical quartiles, and grouped by gender and age. Non-traditional lipid metrics (LDL/HDL, TC/HDL) had higher contributions to the variation of cognitive scores than traditional lipid metrics (TC, TG, LDL, HDL).

Conclusion: Our study provided evidence for the generality of a significant association between traditional/non-traditional lipid metrics and cognitive function in middle-aged and elderly adults. The factors that vary with genders and age groups do not appear to significantly alter the lipid-cognition relationship.