OUP user menu

Margarines supplemented with low dose n-3 fatty acids are not effective in secondary prevention

Ulrich Laufs, Stephan H. Schirmer
DOI: http://dx.doi.org/10.1093/eurheartj/ehs063 1555-1557 First published online: 27 March 2012

This editorial refers to ‘Effects of n-3 fatty acids on major cardiovascular events in statin users and non-users with a history of myocardial infarction’, by S.R.B.M. Eussen et al., on page 1582

The concept of ‘functional foods’ is to add biologically active compounds to natural or processed foods to achieve a health benefit for the prevention or treatment of chronic diseases. The functional food industry continues to be an area of fast growth, in part because regulations and marketing strategies differ substantially from those of the pharmaceutical industry. Prospective randomized studies testing the impact of a functional food on cardiovascular events are rare. Therefore, Kromhout et al. are to be congratulated on their important Alpha Omega trial that randomized 4837 patients post-myocardial infarction to margarines supplemented with eicosapentaenoic acid (EPA) plus docosahexaenoic acid (DHA) (with a targeted additional daily intake of 400 mg) and/or plant-derived alpha-linolenic acid (ALA) (with a targeted additional daily intake of 2 g).1 In agreement with recent studies such as OMEGA2 and SU.FOL.OM3,3 Alpha Omega showed that margarines with low-dose n-3 polyunsaturated fatty acids (n-3 FAs) do not prevent cardiovascular events in patients post-myocardial infarction. The result of the study is important, clear, and consistent: all three n-3 FA-supplemented margarines tested did not reduce the number of cardiovascular endpoints.

Of the 4837 patients randomized in the Alpha Omega trial, only 2531 underwent a follow-up measurement of baseline characteristics due to budgetary constraints; the remaining individuals were followed by questionnaires. A total of 413 patients were classified as statin non-users. Eussen et al. have now performed a retrospective analysis on this small subgroup of ‘statin non-users’.4 The reasons for non-statin use were not assessed and non-statin use was not confirmed; the study did not record serial serum cholesterol concentrations. The two subgroups (statin users and non-users) were very different in size (3740 vs. 413). The non-users received four margarine regimens. As expected in these small groups, significant differences in baseline characteristics were observed. The study collected a combined endpoint consisting of different cardiovascular events including percutaneous coronary interventions (PCIs). In the control margarine group, only 20 combined events were recorded compared with 16 in the EPA–DHA group (P = 0.60), 17 in the ALA group (P = 0.85), and 9 in the EPA–DHA + ALA group (P = 0.07). The authors provide no information on the reasons for PCI nor on the relative contribution of PCIs to the combined endpoint. Functional food has the disadvantage that the amount consumed may vary. The molecular mechanism(s) of n-3 FAs is not fully understood. Plasma levels of the different fatty acids or a surrogate parameter for effects of an n-3 FA were not measured. One can only speculate on whether intake of higher amounts of n-3 FAs would have shown effects. In summary, functional foods supplemented with low dose EPA–DHA showed no effect even in the absence of statin treatment. A trend was observed for the EPA–DHA + ALA margarine; however, one could debate whether this may represent a chance finding in the light of the very small number of combined events and the multiple limitations of the analysis as acknowledged by the authors.

Physicians and patients have to choose between multiple drugs that are advertised for individuals post-myocardial infarction. Patients and their healthcare providers need to implement several lifestyle changes such as smoking cessation and physical activity.5,6 The Alpha Omega trial shows that margarines with low-dose n-3 FAs are not effective in secondary cardiovascular prevention and provides important evidence not to use this food supplementation. The data provide guidance to focus resources on preventive strategies with documented effects such as medication adherence with regard to statin therapy.7 Medication adherence inversely correlates with the number of tablets. Any ineffective therapy can therefore be harmful by impairing the intake drugs whose action is of evidence based.

Based on extensive prospective evidence,8 patients post-myocardial infarction benefit from statin treatment. There is broad consensus that novel preventive treatments have to prove their efficacy on top of statin treatment. The disappointment of investigators testing a preventive treatment that does not reach significance in the presence of statins is understandable. The requirement for novel strategies may appear ‘unfair’ compared with old drugs such as aspirin, where the clinical effect in the presence of a current therapy including statins and angiotensin-converting enzyme (ACE) inhibitors is unknown. However, from the patient's perspective, a new drug or functional food preparation that does not add to the risk reduction by statins is not needed.

Regulatory agencies such as the German Bundesinstitut für Risikobewertung (BfR) have issued advice against the supplementation of foods with DHA and EPA.9 The Alpha Omega data support this critical point of view and add to the list of dashed hopes in the use of functional foods for cardiovascular prevention. While epidemiological studies show improved health in individuals consuming fruit, vegetables, or fish, prospective randomized studies of supplementation with vitamins, plant sterols, and—now—low-dose n-3 FAs do not demonstrate a reduction of cardiovascular events.10,11 Studies such as Alpha Omega remind us that an epidemiological association does not prove the causality of an intervention. Consumption of foods such as fish or red wine may be indicators of a healthy lifestyle, income, and education unrelated to direct health effects. There is no convincing positive study to date to recommend any type of supplemented margarine for the purpose of cardiovascular prevention.

Raffaele De Caterina has recently summarized the current knowledge on n-3 FAs in cardiovascular disease in a seminal review article.12 Multiple ‘orthodox pathways’ and a wide variety of ‘novel pathways’ of polyunsaturated fatty acid metabolism have been elucidated. However, after decades of research, the precise mechanisms of action remain incompletely understood. The majority of the effects have been attributed to a reduction of arachidonic acid in membranes of different cells, but this process has not been fully elucidated (Figure 1). In the Alpha Omega subgroup analysis,4 the effects of fatty acids are reported to be more than three times higher in non-statin users who take the combination of EPA–DHA plus ALA as compared with the ‘classical’ n-3 FAs (EPA–DHA) only. A molecular explanation for the reported role of ALA is not known. The mechanistic uncertainty related to n-3 FAs is in contrast to the molecular mechanisms of statins, ACE inhibitors, or beta-blockers that are well defined and that can be quantitated and monitored (Figure 1). The clinical data on the prevention of myocardial infarction and cardiac arrhythmias by n-3 FAs are also heterogeneous, and many findings remain unexplained or have not been reproducible. In the conclusion of his paper, De Caterina therefore points out that ‘… important gaps in knowledge remain. Data are lacking from clinical and mechanistic studies of the putative benefits of n-3 FA for both primary and secondary prevention’.12

Figure 1

The molecular targets of established drugs in cardiovascular secondary prevention. n-3 fatty acids (FAs) exert multiple molecular mechanism(s) that are still incompletely understood, and the list most probably is not complete. Thromboxanes, leukotrienes, and prostaglandins are thought to be modulated by replacement of arachidonic acid by n-3 FAs.12,14 eNOS, endothelial nitric oxiude synthase; HMG-CoA, 3-hydroxy-3-methyl-glutaryl-CoA; LV, left ventricular; NADPH, nicotinamide adenine dinucleotide phosphate; LDL, low density lipoprotein; NFkB, nuclear factor kappa-light-chain-enhancer of activated B cells.

The analysis by Eussen et al. adds to the concern that low-dose n-3 FAs are not effective in the prevention of cardiovascular diseases for a majority of individuals.4 At present, we do not need to implement functional foods supplemented with low dose n-3 FAs, plant sterols, or vitamins. However, in higher doses, n-3 FAs may play a significant role for cardiovascular drug therapy because of their potential to lower serum triglycerides. In contrast to low-dose n-3 FAs, the effect of high-dose (2–4 g EPA–DHA per day) n-3 FAs to lower serum triglyceride concentrations by 20–40% is well documented. This clinical effect is reproducible and quantifiable. Therefore, high-dose EPA–DHA represent an important option for the treatment of hypertriglyceridaemia. High-dose EPA–DHA do not change serum cholesterol and could therefore test the importance of triglyceride lowering independent of cholesterol concentrations. Broadly implemented statin therapy makes the study of on-top medication cumbersome. However, in an individually tailored approach, selected hypertriglyceridaemic patients appear to be likely to benefit from high-dose therapy with n-3 FAs provided as a pill, and this concept is worth testing prospectively. In summary, while the effect of n-3 FAs in this small subanalysis in non-statin patients promotes an interesting hypothesis for future research,4 the focus of patient care for now is to implement statin treatment in secondary prevention. Recent large statin trials confirm that statins—even in very high doses—are well tolerated by the vast majority of patients.13


The Deutsche Forschungsgemeinschaft [KFO 196]; Deutsche Stiftung für Herzforschung.

Conflict of interest: none declared.


  • The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.


View Abstract