Omega-3 has previously been touted for its ability to protect against dangerous heart rhythm due to beneficial membrane effects in cardiac muscle cells. In more recent times, studies such as the STRENGTH study and Reduce-IT studies [1] have turned this to a suspicion that supplementation could do the opposite, and promote atrial fibrillation(AF).

During the last month (August 2026), 4 papers have been published looking at Omega-3 and incidence of AF all supporting an overall recommendation of fish oil for hearthealth.  

A systematic review [2] of 16 studies was performed with studies from the UK Biobank with sample sizes from 20,338 to 273,835. The review demonstrated an inverse association of plasma total omega-3 fatty acids with coronary heart disease, CVD mortality, heart failure, haemorrhagic stroke and severe fatty liver disease.

Although the study did not specifically assess AF, the presence of AF is a contributing factor for heart failure and stroke and thereby an implicit indication of reduced AF.

A more focussed approach was seen with a study by Versnjak et al who assessed 483,372 individuals, again from the UK Biobank showing that those with AF had lower circulating levels of PUFAs [3].

A study performed with 9,178  Japanese[4] in- and out-patients also showed that increasing plasma levels of EPA, DHA or both was significantly inversely associated with AF events.

The above 3 studies are observational studies which provide strong in-sights into the potential effects of omega-3 but are limited by co-variants that may also impact on health outcomes. As an example, it can be considered that individuals that eat fish or take supplements also choose other healthy lifestyle options compared to those who do not eat fish and the additional factors are the drivers of cardiac health improvements.

With this criticism in mind, the observational studies are nicely complemented by a study by Abuknesha et al [5] which retrospectively assessed 114,592 individuals above the age of 50, from interventional studies selected based on the use of more than 500mg/day EPA+DHA with an intervention period of at least 12 months. The study assessed new AF (individuals with known AF were excluded).

The study was funded by GOED and included 35 randomized controlled trials with 114,592 participants. Approximately half of the assessed trials had previously unpublished AF data.

The results strongly support the safe use of EPA and DHA as supplements. Healthy subjects showed no increase in AF at any dose, whereas high risk subjects had increased AF only with doses exceeding 1500mg EPA+DHA daily.  

The study provides a powerful message that normal supplemental use of EPA and DHA (i.e.,<1500mg/day) is safe and beneficial for cardiac health, or as the leadauthor Abuknesha concludes:

 “Given that elevated blood omega-3 levels … are associated with reduced atherosclerotic cardiovascular disease risk and other favourable healthoutcomes, the small AF signal in High-risk high-dose populations should beweighed against the substantial cardiovascular benefits when making treatment decisions”.

1.           Olshansky, B., et al.,Cardiovascular benefits outweigh risks inpatients with atrial fibrillation in REDUCE-IT (Reduction of CardiovascularEvents with Icosapent Ethyl-Intervention Trial). European Heart Journal,2021. 42(Supplement_1).

2.           Firdous, W.T. and P.C. Calder, Circulating Omega-3 Fatty Acids andCardiometabolic Outcomes: A Systematic Review of Publications From UK Biobank.Lipids, 2026.

3.           Versnjak, J., et al., The Role of Circulating PolyunsaturatedFatty Acids in Atrial Fibrillation and Atrial Cardiomyopathy: A MachineLearning-Informed Structural Modelling Approach. Eur J Heart Fail, 2026.

4.           Yanagisawa, N., et al., Association Between Plasma Omega-3 andOmega-6 Fatty Acid Levels and Atrial Fibrillation: A Large-Scale, Real-WorldRetrospective Study in Japan. J Atheroscler Thromb, 2026.

5.           Abuknesha, N.R., et al., Effects of Omega-3 Fatty Acid Treatment onRisk for Atrial Fibrillation: An Updated Meta-Analysis of 35 Trials including114 592 Individuals. Circ Arrhythm Electrophysiol, 2026: p. e014785.

‍

‍