Meta‐Analysis of the Association of Omega‐3 Fatty Acids Biomarkers with Pulmonary Function

Abstract

OBJECTIVEOmega‐3 fatty acids (ω‐3 FAs) are hypothesized to have protective effects on pulmonary function, but few population‐based studies have investigated this association. We conducted a large‐scale meta‐analysis across multiple cohorts to study the association of ω‐3 FA biomarkers with pulmonary function.METHODSSeven cohorts in the Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium (CHARGE) had measurements of ω‐3 FAs biomarkers (phospholipid FAs in plasma in 6 cohorts and red blood cells [RBCs] in 1 cohort) and pulmonary function tests (PFTs) in 13,629 European ancestry (EA) and 2,505 African ancestry (AA) participants. The individual ω‐3 FAs, included in separate models, were α‐linolenic acid (ALA), eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA), and docosahexaenoic acid (DHA). The PFT outcomes were forced expiratory volume in the first second (FEV1), forced vital capacity (FVC), and the ratio (FEV1/FVC). All linear regression models were adjusted for age, age2, sex, smoking status, pack‐years, height, height2, study site, and weight (FVC analysis only). Effect modification by smoking status (current/former/never) was tested in models extended to include interaction terms. Cohort‐specific results were combined by fixed‐effect meta‐analysis.RESULTSDPA and DHA had statistically significant associations with FEV1 and FVC. For DPA in the EA group, a 1 standard deviation (SD; 0.08% of total FAs) higher plasma DPA was associated with 10.6mL higher FEV1 and 11.7mL higher FVC (both p<0.0001), on average. For DHA in the EA group, a 1 SD (0.43% of total FAs) higher plasma DHA was associated with 6.7mL higher FEV1 (p=0.001) and 4.4mL higher FVC (p=0.071), on average. The DHA and DPA associations were mostly positive regardless of the biomarker compartment (plasma or RBC) or ancestry group (AA or EA). No significant associations of ALA or EPA with FEV1 or FVC were found. However, several EPA (or ALA)—FEV1/FVC associations were statistically significant in both ancestry groups with small effect estimates (between −0.15% and 0.01% per 1 SD higher ω‐3 FA). No significant associations of DHA or DPA with FEV1/FVC were found. Meta‐analysis of smoking status × ω‐3 FAs revealed an interaction of plasma DHA with current smoking status in the EA group: DHA—FEV1 (p=0.042) and DHA—FEV1/FVC (P<0.0001) associations were stronger in current smokers compared to never smokers.CONCLUSIONSMeta‐analysis showed significant associations of higher plasma DHA and DPA with higher pulmonary function, particularly FEV1 and FVC, in the EA population. To put the finding in context, a year of cigarette smoking is associated with a 15mL lower FEV1, whereas a 1 SD higher plasma DHA or DPA is associated with an approximately 9mL higher FEV1, on average. Longitudinal studies are needed to understand whether increasing plasma DHA or DPA would improve lung function or slow its decline.Support or Funding InformationFunded by NIH HL125574 (PAC and DBH), HL077612 (RGB)