Omega-3 Fatty Acids in Cardiovascular Disease

Re-assessing the Evidence

Alan Begg; Susan Connolly; Julian Halcox; Agnes Kaba; Linda Main; Kausik Ray; Henry Purcell; Helen Williams; Derek Yellon


Br J Cardiol. 2012;19(2):79-84. 

In This Article

What Have the Clinical Trials of Fish Oils Shown Us?

The panel reviewed clinical studies with omega-3 PUFAs. Early promise shown in clinical studies, most notably in GISSI-P,[16] has not been borne out in more recent studies, which – with the exception of the JELIS and GISSI-HF studies – have shown equivocal results.

The majority of intervention studies in those at high cardiovascular risk show a beneficial impact of omega-3 supplementation on major cardiovascular events, although heterogeneity between studies has been noted. Large-scale randomised trial data are most compelling for omega-3 supplementation in the post-MI setting.


The largest of these, the GISSI-P (GISSI-Prevention) study (n=11,324 randomised to omega-3 PUFAs within three months of MI),[16] demonstrated relative risk reductions in overall mortality, cardiac mortality and SCD of 20%, 30% and 45%, respectively, with 1 g/day of highly purified omega-3 acid ethyl esters (Omacor®) over a 3.5-year period. Absolute risk reductions over the same period were 2.1%, 2% and 1.6% for overall mortality, cardiac mortality and SCD, respectively. Significant benefits of supplementation emerged within three to four months and were most marked in those with more extensive left ventricular dysfunction. Considered together, these data suggest a reduction in ventricular arrhythmia as the likely mechanism of benefit.


Results from JELIS (Japanese Eicosapentaenoic Acid Lipid Intervention Study),[17] an open-label, blinded analysis conducted in Japan, showed that 1.8 g/day of EPA reduced combined cardiovascular events. About 36% of the subjects recruited were hypersensitive, 20% had coronary artery disease, and 15% had diabetes. Review of the individual component end points in JELIS demonstrated that the main driver of benefit was reduction in coronary instability, suggesting benefits on clinical lesion progression, destabilisation and thrombosis. No effect on the incidence of arrhythmia was noted, although such events were extremely rare in this Japanese population and the study was not powered to evaluate this end point.


Recommendations supporting intake of 1 g/day omega-3 PUFAs in those with established coronary artery disease, including the post-MI setting, are currently under review by NICE following publication of several recent studies. The German OMEGA (Effect of Omega 3-Fatty Acids on the Reduction of Sudden Cardiac Death After Myocardial Infarction) study randomised over 3,000 patients recruited early post-MI to 1 g Omacor® or placebo to examine the impact on SCD within the first year.[18]

Although the study outcome was neutral, it was substantially underpowered and the results should, therefore, be interpreted with caution. It was designed with the expectation of a 45% reduction in the relative risk of SCD after one year with omega-3 treatment as seen in GISSI-P (which had a duration of three years). But due to far lower than planned event rates (1.5% actual vs. 3.5%, which was expected based on the results from GISSI-P) and without changing the number of subjects recruited or the duration of follow-up, the study was estimated to have a power of only 45% or less to address the primary hypothesis.

Alpha Omega

Similarly, Alpha Omega (Study of Omega-3 Fatty Acids and Coronary Mortality),[19] which studied omega-3 PUFA supplemented margarine in post-MI patients, was also neutral. This study was also underpowered. The study looked at relatively low-dose supplementation with either EPA+DHA or ALA (add doses), EPA+DHA+ALA or a control margarine in just under 5,000 patients who had suffered an MI an average of 2.5 years previously. The study was powered to show a 25% reduction in death due to CHD and it was reported in the online appendix that the final study power to address this end point was approximately 35%. The primary comparisons reported in the manuscript were also unusual, comparing outcomes in EPA+DHA versus ALA + placebo and ALA versus EPA+DHA + placebo rather than comparisons of the omega-3 supplement groups (alone or in combination) versus placebo alone.


GISSI-HF (GISSI-Heart Failure),[20] a large and adequately powered trial looked at omega-3 PUFAs in heart failure. Results demonstrated a statistically significant and clinically important reduction of all-cause mortality, despite the modest relative risk reduction, with 1 g Omacor® in a contemporary heart failure population.

Omega-3 PUFA supplementation showed an absolute risk reduction of 1.8% in all-cause mortality seen over two years of follow-up. It was felt that half of this benefit was due to a reduction in arrhythmic deaths with the other half due to reduction of admission for ventricular arrhythmia. There was little benefit in atherothrombotic events, such as MI and stroke. A subsequent pharmacoeconomic evaluation of the GISSI-HF data suggests that this dose of Omacor® is a cost-effective intervention in patients with heart failure.[21]

Cochrane Meta-analysis

A Cochrane review, published in 2009,[22] concluded that it was still not clear whether dietary or supplemental omega-3 PUFAs, either from fish or vegetable sources, did reduce total deaths or cardiovascular events in both patients with cardiovascular disease or those at risk of developing the disease. This conflicting evidence is reflected in the various meta-analyses on the possible outcomes of fish or omega-3 PUFA consumption and cardiovascular disease (CVD) outcomes.[13] The wide range of study patient characteristics, together with the varying nature and dose of supplement, as well as trial size and quality, all contribute to the lack of clarity in this context.

The panel felt that recent post-MI studies have been inadequate in their design to elucidate the role of omega-3 PUFA supplements in a contemporary post-MI cohort. They noted recent studies in this area were often underpowered, were likely to be using suboptimal doses and may not have been conducted for long enough to show any benefit. It was reassuring, the panel felt, that the more recent JELIS and GISSI-HF studies, which were adequately powered and allowed several years of follow-up, demonstrated positive results.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: