Omega-3 Fish Oil Supplements – Taking Stock and Awaiting Vital New Studies

Controversy about the role of n-3 polyunsaturated fatty acids (PUFAs) in cardiovascular disease has a long and decorated history. Beginning in 1971, Dyerberg and Bang published a series of highly-cited articles on Greenland Eskimos, suggesting that they have lower-than-expected rates of cardiovascular disease that the Danish investigators attributed to anti-atherosclerotic and antithrombotic effects of high n-3 PUFA intake.1, 2 However, the extraordinary Ancel Keys, Director of the Laboratory of Physiological Hygiene at the University of Minnesota School of Public Health, had already concluded in 1957 that “on no grounds is it possible to suggest that the case of the Eskimo contributes anything.”3 Unfortunately, Keys’ caution itself appears to be a misreading of an even earlier survey of clinical manifestations of atherosclerosis, which also suggested a lower rate among Greenland Eskimos than among poorly-matched Finns who had a much lower overall mortality rate.4

Despite this challenging history, the importance of dietary n-3 PUFA intake has gained a substantial measure of consensus. The American Heart Association (AHA) itself includes fatty fish intake as part of a healthy diet, one of Life’s Simple Seven, and specifically recommends that adults consuming a 2000-kilocalorie diet aim for at least 8 ounces of fatty fish per week,⁵ at least in part to ensure adequate intake of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). This recommendation owes much to the pioneering work of David Siscovick, whose case-control study in Seattle and suburban King County, Washington demonstrated that, compared with no n-3 PUFA intake, consumption of one fatty fish meal per week was associated with 50% lower risk of primary cardiac arrest.⁶

Siscovick and colleagues, in their new Science Advisory, now tackle the challenge of n-3 PUFA supplementation, a topic previously addressed by many of the same authors in an AHA Scientific Statement in 2002.⁷ At that time, the evidence supporting a benefit of n-3 PUFA supplementation seemed clear, at least among individuals who had survived a myocardial infarction. The GISSI-Prevenzione trial confirmed 73 fewer deaths among participants randomized to 1 gram of n-3 PUFA supplementation daily than to control, 42 of which were sudden deaths (relative risk for sudden death 0.74; 95% confidence interval 0.58–0.93).⁸ The DART trial, in a similar population, found a significant reduction in risk of coronary death (despite a numerical increase in cases of myocardial infarction) with advice to consume at least two fatty fish meals per week.⁹ Thus, these two large and complementary trials both suggested that n-3 PUFA intake has a specific benefit on reducing risk of cardiac death (but not necessarily incident events) among survivors of previous myocardial infarction. Unfortunately, evidence in the ensuing years has not necessarily been entirely consistent, and the new Advisory therefore serves a highly welcome function in collecting and synthesizing this evidence.

The Advisory specifically targets n-3 PUFA supplementation and thus addresses the clinically relevant question of whether physicians should recommend supplements (which can be costly and have unpleasant sensory side effects) to their patients. The authors correctly caution that their results do not bear directly on n-3 PUFA intake by diet. Of course, several reasons exist to justify fish consumption well beyond its n-3 PUFA content, including its rich protein and low saturated fat contents, and hence this Advisory does not alter the AHA recommendation to consume fatty fish regularly. Nonetheless, even answering the more limited clinical question of n-3 PUFA supplementation poses substantial hurdles, such as the host of cardiovascular conditions for which it has been tested in randomized trials, the varied and overlapping trial populations, and the sometimes contradictory results that trials have yielded.

The authors focus their work on long-term trials of clinically meaningful endpoints, which are obviously of most relevance for patients and physicians. Necessarily, this restricts most of their analyses to EPA+DHA doses generally in the range of 1 gram per day, well within the range attainable with over-the-counter or prescription n-3 PUFA supplements but substantially below the 4 gm per day dosing approved for hypertriglyceridemia; the one exception is trials of incident atrial fibrillation, where doses have occasionally been higher. Because trials on endpoints have not yet attempted to determine dose-response relationships, the Advisory provides expert guidance on commonly-used doses of these supplements (as might be used in clinical care) but not on the long-term health effects of these fatty acids at pharmacological doses, nor on whether an ideal dose of supplementation might exist.

What did the Advisory find, and what did the authors conclude? As the authors did, it may be easiest to summarize their results by clinical indication:

1. Primary prevention
   Obviously this is potentially the largest pool of patients for whom n-3 PUFA supplements could be recommended. However, the authors make no recommendation about the use of supplements for primary prevention of coronary heart disease, heart failure, or atrial fibrillation, because the only trial to test any of these indications – the The VITamin D and OmegA-3 TriaL (VITAL) – is still underway.¹⁰ VITAL began in 2010 and will enroll over 25,000 adults in factorial-design, randomized trial of 2000 IU vitamin D and 1 gram of fish oil daily for an average of 5 years of follow-up. Clearly, based both on its size and the more broadly representative nature of its study population, VITAL has the potential to alter recommendations about n-3 PUFA supplementation substantially; even for higher-risk subgroups, such as those with diabetes, who have been the subject of previous n-3 PUFA trials, VITAL may well enroll enough individuals in those strata to modify the conclusions of this Advisory. VITAL investigators have also identified a broad range of cardiovascular outcomes of interest with which to evaluate primary prevention, although power to detect benefits of supplementation will inevitably differ between them. Until the results of VITAL are available, the Advisory concludes, reasonably, that no recommendation for primary prevention can be made in either direction. Although n-3 PUFA supplementation may have other effects in a primary prevention context for which its use might be justified even by randomized trials, these benefits extend well beyond the scope of the Advisory and none are of sufficient in themselves to justify a recommendation for routine supplementation.
2. High cardiovascular risk
   Across a range of conditions marking cardiovascular risk, including diabetes, hypercholesterolemia, and macular degeneration, the authors found no consistent benefit of n-3 PUFA supplementation, although one large trial from Japan found a significant benefit on non-fatal coronary events.¹¹ A sizable fraction of participants in this subgroup had existing prevalent cardiovascular disease, but the trials typically excluded individuals with a recent myocardial infarction. The Advisory recommends against supplementation in this group of individuals.
3. Prevalent atherosclerotic cardiovascular disease
   As noted above, n-3 PUFA supplementation had specifically been found to improve prognosis among survivors of acute myocardial infarction when the AHA last reviewed this topic.⁷ In the updated Advisory, three new trials have been included, none of which showed statistical evidence of benefit from n-3 PUFA supplementation, although, as the authors note, pooling of results across studies still suggests a benefit specifically on cardiac death. Consistent with the long history of controversy in the field, this Advisory wisely did not include one positive trial cited in the prior Scientific Statement¹² that has since come under serious scrutiny.¹³ The authors advise that supplementation among individuals with prevalent coronary heart disease be considered reasonable but make no recommendation about secondary prevention of stroke.
4. Prevalent congestive heart failure and atrial fibrillation
   Relatively recent trials have addressed the possible benefit of n-3 PUFA supplements among patients with heart failure or atrial fibrillation, which was not addressed in the 2002 Scientific Statement. The GISSI-HF trial, which randomized nearly 7000 patients with largely systolic heart failure, found significant benefits on total mortality and hospitalization in adjusted but not in unadjusted analyses, leading the Advisory authors to consider supplementation reasonable among these patients. In contrast, trials to prevent recurrent atrial fibrillation were negative, and supplementation was deemed not indicated.

What questions and controversies remain after this thorough review? At least three areas of uncertainty still require clarification. First, the weakening over time in the observed effect of n-3 PUFA supplementation even among patients with previous myocardial infarction remains perplexing. Attribution to increasing background rates of n-3 intake is difficult, given that the only trial to observe any possible benefit among high-risk individuals was conducted in Japan,¹¹ where dietary n-3 intake is consistently high. However, if it reflects a smaller pool of susceptible patients in the context of modern and aggressive coronary therapies, with smaller and better-managed regions of myocardial scarring, then there may truly be no role for n-3 supplementation among myocardial infarction survivors in the decades to come.

Second, the role of n-3 supplementation in heart failure calls for more research. Given that approximately half of heart failure cases occur with a preserved ejection fraction,¹⁴ that population needs formal study. Moreover, the statistically borderline effect observed in GISSI-HF¹⁵ suggests that even the systolic heart failure population might require reexamination as resynchronization and other novel therapies grow in importance.

Finally, the role of n-3 supplementation in primary prevention of any cardiovascular disorder requires urgent clarity. Fortunately, this question has a pending answer on the horizon, as the VITAL study progresses toward its conclusion. Its results will be eagerly awaited.

In the interim, Siscovick and colleagues have provided an important review of the state of the science to date and provide a well-argued and comprehensive assessment of the role that n-3 PUFA supplementation should play in the prevention and management of cardiovascular disorders.