Fish oil and atrial fibrillation: the AF signal in omega-3 trials

For two decades, fish oil was sold to cardiovascular patients as anti-arrhythmic. Then the large outcomes trials of the late 2010s and early 2020s reported the opposite: a small but consistent excess of new-onset atrial fibrillation in the omega-3 arm. The signal is now reproducible across at least four major randomized trials and has been formally captured in pooled meta-analysis. It is dose-dependent, clinically actionable, and rarely explained on supplement labels.

What the major trials reported

REDUCE-IT randomized 8,179 high-risk patients to 4 g/day of icosapent ethyl or placebo. Atrial fibrillation requiring hospitalization occurred in 3.1% of icosapent patients versus 2.1% of placebo (HR 1.50, p=0.004) [1]. STRENGTH, which tested 4 g/day of a mixed EPA/DHA carboxylic acid against corn oil placebo, was stopped early for futility but reported new-onset AF in 2.2% of treatment versus 1.3% of placebo patients [2]. The OMEMI trial in older post-MI adults using a more modest 1.8 g/day dose of EPA+DHA still reported AF in 7.2% of omega-3 versus 4.0% of control (HR 1.84) [3].

Pooled meta-analysis

A 2021 meta-analysis published in Circulation pooled seven trials (more than 81,000 patients) and found a 25% relative increase in AF events with omega-3 fatty acids (HR 1.25, 95% CI 1.07–1.46). The absolute risk increase was modest — roughly 1 additional case per 100 patient-years — and rose with dose. Trials using doses above 1 g/day drove most of the signal [4].

Why omega-3 might be pro-arrhythmic

The proposed mechanisms are several and not mutually exclusive. EPA and DHA incorporation into cardiomyocyte membranes alters sodium and calcium channel kinetics, shortening atrial refractoriness in some preparations. High doses also lower heart rate and may increase vagal tone, which can promote re-entry in vulnerable atria. Animal models have shown both pro- and anti-arrhythmic effects depending on dose and baseline rhythm — consistent with the U-shaped clinical picture in which low-dose dietary intake appears neutral or protective while pharmacologic doses raise risk [5].

Who is affected in practice

The absolute risk increase is small and concentrated in patients already at elevated AF risk: older age, prior cardiovascular disease, left atrial enlargement, obesity, hypertension, and sleep apnea. A 70-year-old post-MI patient on 4 g/day of prescription omega-3 has measurably increased AF risk; a healthy 30-year-old taking 1 g of supplement fish oil has effectively none in the trial data. The dose threshold matters: most over-the-counter fish oil products deliver 300–600 mg of EPA+DHA per capsule, and typical daily totals stay below the level where AF risk has been demonstrated.

How clinicians are responding

The 2023 AHA/ACC chronic coronary disease guideline acknowledges the AF signal and recommends discussing it with patients before prescribing high-dose omega-3 [6]. Many cardiologists now screen for pre-existing AF before starting 4 g/day formulations, document baseline rhythm, and check for palpitations at follow-up. Patients who develop AF on therapy are typically switched off, and the rhythm usually does not recur after discontinuation — consistent with a reversible electrophysiologic effect rather than fixed structural change.

What this means for supplement users

For ordinary supplement-level dosing (1 g/day or less of combined EPA+DHA), the AF signal in the trials is not robustly demonstrated and the absolute risk is very low. For patients self-dosing at 3–4 g/day for triglyceride lowering or anti-inflammatory purposes, the signal is real and should prompt a conversation with their clinician. Patients with known paroxysmal AF, prior cardioversion, or palpitation history should be especially cautious about high-dose products.

The bottom line

Omega-3 fatty acids at pharmacologic doses (≥2 g/day) carry a small but reproducible increase in atrial fibrillation risk that should be disclosed. The signal does not negate the lipid and triglyceride benefits in appropriate patients, but it shifts the risk-benefit calculation for self-administered high-dose use. Patients should not stop prescribed omega-3 therapy without speaking to their clinician, and supplement users at typical doses can keep taking their capsules with reassurance — provided they do not stack them with prescription-strength formulations.