Effects of Omega-3 Fatty Acid Supplements on Arrhythmias

Sarah Parish, DPhil; Marion Mafham, MRCP, MD; Alison Offer, PhD; Jill Barton; Karl Wallendszus, MSc; William Stevens, PhD; Georgina Buck, MSc; Richard Haynes, DM; Rory Collins, MBBS, MSc; Louise Bowman, MD, MRCP; Jane Armitage, MRCP, MFPH


Circulation. 2020;141(4) 

For several decades there has been interest in the possible role of omega-3 fatty acids (FAs) in the prevention of cardiovascular disease and the possibility that they have antiarrhythmic effects. Higher consumption of fish (which contains the long-chain omega-3 FAs, eicosapentaenoic acid, and docosahexaenoic acid) is associated with lower risks of coronary heart disease events, and particularly of cardiac deaths in observational studies, but meta-analyses in 2018 of randomized trials of ½ to 2 g omega-3 FA supplementation for a mean of 4 to 5 years failed to show convincing benefits on cardiovascular outcomes.[1] However, recently, the REDUCE-IT trial (Reduction of Cardiovascular Events with Icosapent Ethyl–Intervention Trial) of a daily dose of 4 g icosapent ethyl (a purified form of eicosapentaenoic acid) in participants with high triglycerides reported a highly statistically significant 25% proportional reduction in various cardiovascular outcomes.[2] The results in both low and high dose large trials are consistent with a pro rata to dose reduction in coronary heart disease events of 7% to 8% per 1 g eicosapentaenoic acid + docosahexaenoic acid supplementation.[1,2] However, in REDUCE-IT,[2] supplementation was also associated with a nominally statistically significant (without adjustment for multiple comparisons) ≈50% increase in hospitalization for atrial fibrillation (AF) or flutter. It is not clear whether this excess is reflected in the low dose trials.[2] In ASCEND (A Study of Cardiovascular Events in Diabetes), we previously reported the effects of omega-3 FA on participant-reported arrhythmia adverse events, but here we report more comprehensively on AF and other arrhythmias, using additional data extracted from linked electronic health records.[3]

ASCEND was a large mail-based randomized trial that assessed the effect of daily 1 g omega-3 FA supplementation (containing 0.84 g eicosapentaenoic acid + docosahexaenoic acid) on the risk of serious vascular events in people with diabetes mellitus but without known atherosclerotic cardiovascular disease (and, in a 2×2 factorial design, also the effects of low-dose aspirin; current Controlled Trials number, ISRCTN60635500; ClinicalTrials.gov number, NCT00135226.).[3] The protocol was approved by the North West Multi-centre Research Ethics Committee and all of the participants provided written informed consent. Anticoagulant use at randomization was an exclusion criterion (because of the aspirin comparison), and this would have excluded most people with known AF (as anticoagulant use is recommended in AF). The main results on omega-3 FA supplementation, reported previously, showed no significant effect on the primary outcome.[3] This investigation includes added data extracted from electronic record linkage to hospital episodes, available for 97% of participants both during the trial and for 14 years before randomization.

Arrhythmia outcomes were defined based on either hospitalizations or serious episodes reported by the participants during follow-up or from International Classification of Diseases, Tenth Revision diagnoses or OPCS Classification of Interventions and Procedures, Version 4 procedure codes in the electronic data. AF diagnoses in hospital episodes before randomization were used to define previously known AF. Arrhythmia outcomes considered are AF (among participants without any previously known AF) and nonfatal ventricular arrhythmia. We only considered nonfatal ventricular arrhythmias, with survival to the next day, because we could not reliably ascertain all arrhythmias contributing to death, and it was not possible to determine which cardiac arrests without resuscitation had involved an arrhythmia. Therefore, we also consider the outcome cardiac deaths overall, which would include sudden deaths attributable to fatal cardiac arrhythmias (see Arrhythmia Outcome Definitions on the ASCEND study web page: https://ascend.medsci.ox.ac.uk/professionals). Statistical methods are as used in the main study report.[3] P values <0.01 were considered statistically significant in this secondary analysis.

Among 15 480 participants randomized in ASCEND, less than 1% had a previous hospital admission with an AF diagnosis. Among the remaining 99%, AF was recorded from either electronic health records or participant reports in 1177 participants, compared with 287 by self report alone. AF occurred in 7.7% of participants in the omega-3 FA group and in 7.6% in the placebo group, with a nonsignificant rate ratio (RR) of 1.02 (95% CI, 0.91 to 1.15). A non-fatal ventricular arrhythmia was recorded in nonsignificantly more participants in the omega-3 FA group than in the placebo group (81 vs 54; RR, 1.49; 95% CI, 1.06 to 2.09; P=0.02). Most (106/135) of the nonfatal ventricular arrhythmias required intervention (resuscitation, ablation, or implantable cardioverter defibrillator); sensitivity analyses restricting to these ventricular arrhythmias yielded a similar RR of 1.90 (95% CI, 1.30 to 2.77). Cardiac death occurred in fewer participants in the omega-3 FA group than in the placebo group (1.7% vs 2.2%; P=0.04).

The 47% proportional excess of hospitalization for AF observed in REDUCE-IT with 4-g daily supplementation corresponds to a 10% (95% CI, 3% to 18%) proportional excess per 1 g, and 2 other of the 10 large low-dose trials have reported on AF and found nonsignificant positive associations with hospitalization for AF (Figure).[2,4,5] The association in the low-dose trials, including ASCEND, in combination is not statistically significant, but does not exclude the level of effect per gram seen in REDUCE-IT.


Effects of omega-3 fatty acid supplementation on atrial fibrillation in large trials.
The atrial fibrillation (AF) outcomes were: GISSI-HF (Gruppo Italiano per lo Studio della Sopravvivenza nell'Insufficienza Cardiaca - Heart Failure),4 among participants without ECG evidence of AF at randomization, AF in the ECGs taken at each visit during the trial or an event between visits causing or worsening heart failure/hospitalization; Risk and Prevention Study,5 hospitalization for AF; ASCEND (A Study of Cardiovascular Events in Diabetes), new onset of AF; REDUCE-IT (Reduction of Cardiovascular Events with Icosapent Ethyl–Intervention Trial),2 hospitalization for AF or atrial flutter. The risk ratio and CI in REDUCE-IT were estimated from the reported percentages getting the outcome in each group and the P value (3.1% vs 2.1%, supplementation vs placebo, P=0.004). DHA indicates docosahexaenoic acid; and EPA, eicosapentaenoic acid.

In conclusion, evidence from large trials now suggests that omega-3 FA supplementation may have a dose-related protective effect on coronary events. However, systematic reporting of arrhythmia outcomes in existing and future trials is required to clarify whether supplementation also has any adverse effects on nonfatal arrhythmias.