n–6 fatty acid biomarkers and incident atrial fibrillation: an individual participant-level pooled analysis of 11 international prospective studies

The presence of atrial fibrillation (AF) is associated with an over 2-fold increased risk of stroke, heart failure, and cardiovascular mortality. Long chain n–6 PUFAs have been suggested to have a variety of beneficial biologic effects that may reduce AF development; however, prior studies evaluating this relationship are limited. We prospectively evaluated the association between circulating levels of linoleic acid (LA) and arachidonic acid (AA) with incident AF. We used participant-level data from a global consortium of 11 prospective cohort studies with measurements of LA and AA in adults (aged ≥18 y). Participating studies conducted de novo analyses using a prespecified analytical plan with harmonized definitions for exposures, outcomes, covariates, and subgroups. Associations were pooled using inverse-variance weighted meta-analysis. Among 41,335 participants, 6173 incident cases of AF were ascertained, with median follow-up time of 14 y. In multivariable analysis, per interquintile range (difference between the 10th and 90th percentiles for each fatty acid), circulating n–6 levels were not associated with incident AF. For LA, the hazard ratio per interquintile range was 0.96 (95% confidence interval [CI]: 0.89, 1.04), and for AA, 1.02 (95% CI: 0.94, 1.10), with little evidence of heterogeneity between cohorts. Associations were similarly nonsignificant across subgroups of age, race, and biomarker fraction. Biomarkers of n–6 fatty acids including LA and AA are not associated with incident AF. These findings suggest that overall effects of n–6 PUFAs on influencing AF development are neutral.