Affiliations: [a]
Department of General Medicine, No. 970 Hospitalof the People’s Liberation Army Joint Logistics Support Force, Yantai, Shandong, China
| [b] Unit 71217 of People’s Liberation Army, Yantai, Shandong, China
Correspondence:
[*]
Correspondence to: Fangkai Xing, Unit 71217 of People’s Liberation Army, Yantai, Shandong, China. Tel.: +86 18153572482; E-mail: [email protected].
Note: [1] These authors contributed equally to this work.
Abstract: Background:The relationship between serum fatty acids and cognitive function has been the subject of extensive study. Objective:To analyze the relationship between serum fatty acids composition and cognitive function by NHANES database and multivariate Mendelian randomization (MR) analysis. Methods:A sub-cohort of 1,339 individuals with serum fatty acids and Digit Symbol Substitution Test (DSST) examinations from the 2011–2014 wave of the NHANES were analyzed using fully adjusted multiple linear regression models for associations between serum hydrolyzed fatty acid levels and cognitive function. Univariable and multivariable MR was used to analyze the correlation between 98 exposures related to serum fatty acids and cognitive function. Results from different database sources were combined using meta-analysis. Results:The fully adjusted regression analysis showed that linoleic acid (LA), Omega 6, fatty acids (FAs), and LA/FAs were positively correlated with DSST. 27 exposures were included for univariate MR analysis. Ultimately, only 2 traits had IVW test p-values ranging between 0.0019 and 0.05, both of which were LA/FAs. The meta-analysis of univariate MR revealed that LA/FAs was positively associated with cognitive function (β: 0.040, 95% CI = 0.013–0.067, p = 0.0041). In multivariate MR analysis, after adjusting for education, ischemic stroke, and age, LA/FAs was positively independently associated with cognitive function (IVW β: 0.049, 95% CI = 0.021–0.077, p = 0.0006). The results of MVMR are well in line with the univariate results. Conclusions:Both the Cross-sectional observational analyses and MR-based studies supported a suggestive causal relationship between the serum ratio of Linoleic acid in fatty acids and cognitive function.
