Genetically Determined Omega-3 Polyunsaturated Fatty Acids and Lung Function: A Mendelian Randomization Analysis

Abstract Objectives Cross-sectional studies have found positive associations of plasma omega-3 polyunsaturated fatty acids (N-3 PUFAs) and lung function parameters, including the forced expiratory volume in the first second (FEV1) and forced vital capacity (FVC). We used Mendelian randomization (MR) to address potential limitations in previous findings, including residual confounding and reverse causality, and improve causal inference for the relationship of N-3 PUFAs on lung function. Methods We instrumented the N-3 PUFAs alpha-linolenic acid (ALA), eicosapentanoic acid (EPA), docosahexaenoic acid (DHA) and docosapentaenoic acid (DPA) with genetic variants in the fatty acid desaturase (FADS1/FADS2) and fatty acid elongase (ELOVL2) genes. We performed two sample MR, using genome-wide association data for N-3 PUFAs in the Cohorts for Heart and Aging Research in Genetic Epidemiology Consortium and for FEV1 and FVC in the UK Biobank. We also performed multivariable MR (MVMR) including linoleic acid (LA), the main dietary N-6 PUFA, to account for shared genetic predictors. We used the Wald's ratio or inverse variance weighted method in all analyses. Results In univariable MR, ALA was negatively associated with FEV1 (−0.27 ± 0.13 SD/% total FA, P = 0.02), while EPA was positively associated with FEV1 (0.05 ± 0.02 SD/% total FA, P = 0.02). The DPA—FEV1 association was similar to EPA (P = 0.05). These results align with the opposing effects of FADS1/2 variants on ALA vs EPA and DPA. DHA was not associated with FEV1 and there were no statistically significant N-3 PUFA—FVC associations. Using GWAS estimates adjusted for correlated N3-PUFAs did not alter these results. In MVMR including LA, the ALA—FEV1 associations were strengthened (P = 0.007), while the EPA—and DPA—FEV1 associations were no longer statistically significant. Conclusions Our analyses suggest that higher ALA has a direct negative effect on lung function, while the positive effects of EPA and DPA may be through the balance of N-3 and N-6 PUFA metabolism. However, interpretation of MVRM findings when modeling metabolic pathways needs further consideration. Funding Sources This work was supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health Training Program (T32) in Translational Nutrition Research at Cornell University.