Background Previous randomized control trials showed mixed results concerning the effect of omega-3 fatty acids (n-3 FAs) on atrial fibrillation (AF). The associations of n-3 FA blood levels with heart rhythm in patients with established AF are unknown. The goal of this study was to assess the associations of total and individual n-3 FA blood levels with AF type (paroxysmal versus nonparoxysmal), heart rate (HR), and HR variability in patients with AF. Methods and Results Total n-3 FAs, eicosapentaenoic acid, docosahexaenoic acid, docosapentaenoic acid, and alpha-linolenic acid blood levels were determined in 1969 patients with known AF from the SWISS-AF (Swiss Atrial Fibrillation cohort). Individual and total n-3 FAs were correlated with type of AF, HR, and HR variability using standard logistic and linear regression, adjusted for potential confounders. Only a mild association with nonparoxysmal AF was found with total n-3 FA (odds ratio [OR], 0.97 [95% CI, 0.89-1.05]) and docosahexaenoic acid (OR, 0.93 [95% CI, 0.82-1.06]), whereas other individual n-3 FAs showed no association with nonparoxysmal AF. Higher total n-3 FAs (estimate 0.99 [95% CI, 0.98-1.00]) and higher docosahexaenoic acid (0.99 [95% CI, 0.97-1.00]) tended to be associated with slower HR in multivariate analysis. Docosapentaenoic acid was associated with a lower HR variability triangular index (0.94 [95% CI, 0.89-0.99]). Conclusions We found no strong evidence for an association of n-3 FA blood levels with AF type, but higher total n-3 FA levels and docosahexaenoic acid might correlate with lower HR, and docosapentaenoic acid with a lower HR variability triangular index.
Keywords: atrial fibrillation; heart rate; heart rate variability; omega‐3 fatty acid; rhythm type.