Introduction

Omega-3 fatty acids come in several forms but eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) have been most widely investigated with regard to their cardiovascular benefits. Both forms are abundantly available in oily fish (Table 1) and in supplements provided by numerous companies. Many epidemiological studies and randomized control trials have established the beneficial effects of omega-3 fatty acids on cardiovascular health.1,2,3 In response to this abundance of evidence, the AHA has, for the first time, endorsed a nutritional supplement for the secondary prevention of cardiovascular events in patients with documented coronary heart disease (CHD);2 they recommend the consumption of omega-3 in the form of oily fish, but cite supplements as another option. However, some studies have indicated a less robust relationship between omega-3 fatty acids and reduction in cardiovascular risk.4 Furthermore, some researchers argue that wild fish populations are collapsing and blanket recommendations to increase omega-3 fatty acids from marine sources will only exacerbate this problem.5 In this Review, we will examine the currently available evidence on the importance of omega-3 fatty acids for cardioprotection, address whether the benefits are best conferred by consuming fish or supplementing with fish oils, and discuss possible future sources of omega-3 fatty acids.

Table 1 Content of docosahexaenoic acid and eicosapentaenoic acid in fish1,58
Full size table

Omega-3 and cardiovascular health

Thousands of epidemiological, observational, and experimental studies and randomized controlled trials conducted over the last three decades have established the positive cardiovascular effects of the long-chain omega-3 fatty acids DHA and EPA.1,3 These benefits seem to result primarily from DHA and EPA enrichment of membrane phospholipids,6 which leads to improved arterial and endothelial function,7 reduced platelet aggregation,8 improved autonomic tone,9,10 increased arrhythmic thresholds11 and reduced blood pressure (Box 1, Figure 1).10,12,13,14 Omega-3 fatty acids may also reduce insulin resistance15 and suppress production of proinflammatory cytokines (interleukin-6, interleukin-1β, and tumor necrosis factor).16 In patients with heart failure, doses of omega-3 fatty acids of 8 g daily or more improve body composition and have anti-inflammatory effects.17,18 Omega-3 fatty acids also have favorable effects on lipid profiles, and the AHA recommends 2–4 g daily of DHA and EPA for patients with high triglyceride levels.2

Figure 1: Schematic of potential dose–response curves that indicate the beneficial physiological effects for intake of long-chain omega-3 fatty acids.
figure 1

Antiarrhythmic and antithrombotic effects can be clinically beneficial within weeks. Beneficial clinical effects of heart-rate lowering can be achieved over a period of months, and those for triglyceride and blood-pressure lowering over a period of months to years. Abbreviations: DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid. Reproduced with permission from Mozaffarian, D. & Rimm, E. B. JAMA 296, 1885–1899 (2006).42 © 2009 American Medical Association. All rights reserved.

Full size image

In a meta-analysis by Harris and colleagues, in which data from 25 trials that evaluated the correlation between in vivo omega-3 fatty acid levels and risk of CHD were analyzed, the number of CHD events was inversely related to levels of DHA in phospholipids.19 These markers are closely related to the amount of DHA in the myocardium.19,20 This study did not, however, demonstrate a statistically significant relationship between plasma EPA level and risk of CHD, but interpretation of this finding is difficult since DHA and EPA are almost always consumed together. A randomized controlled trial conducted in the late 1980s (the Diet and Reinfarction Trial [DART]) demonstrated that 2-year all-cause mortality was reduced by 29% in patients who receive omega-3 fatty acids after a myocardial infarction. The benefit was equal for oily fish (about 300 g per week, providing approximately 1 g of EPA/DHA per day) and fish-oil capsules (about 900 mg of DHA + EPA per day).21 More recently, the effects of omega-3 fatty acid supplementation on CHD risk have been examined in two large trials. In the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardico (GISSI)-Prevenzione study,22 11,323 patients who had experienced a myocardial infarction in the preceding 3-month period were randomly assigned to receive 850 mg daily of DHA and EPA or not. All patients also received optimal care with pharmacological therapy and lifestyle interventions. By the end of the 3.5-year trial, the relative risk of death from any cause had fallen by 21% and the relative risk of sudden cardiac death by 45% among patients who received omega-3 treatment. In the largest omega-3 supplementation trial to date—the Japan EPA Lipid Intervention Study (JELIS)23—18,645 patients with hypercholesterolemia were randomly assigned to receive a statin alone or a statin with a highly concentrated form of EPA at a dose of 600 mg three times per day. During the 5-year follow up, the relative risk of major adverse cardiovascular events in the EPA group was reduced by 19%. The third study to document benefits of omega-3 was the GISSI Heart Failure trial, in which 6,975 patients with Class II–IV heart failure were randomly assigned to receive 1 g of prescription omega-3 acid ethyl esters or placebo and followed for all-cause mortality for 3.9 years. The investigators observed a 9% reduction (P = 0.04) in the death rate in the omega-3 group.24 However, the results of the randomized, placebo-controlled OMEGA-Trial, which were presented at the ACC Scientific Sessions in March 2009, indicated that treatment with 460 mg EPA plus 380 mg DHA conferred no benefit in the first year after myocardial infarction among 3,851 patients receiving modern medical therapy.4 This lack of effect could be related to the very low rates of sudden cardiac death (1.5%), the short duration of follow-up (1 year), as well as to the fact that all patients were already being treated with advanced medical therapies (including acute revascularization in 78% of participants) that together could have reduced adverse cardiovascular outcomes leaving very little room for improvement. In our opinion, this relatively small and underpowered study does not negate the large evidence base that supports the benefits of omega-3 fatty acids in CHD prevention. Two other studies that did not have positive outcomes have been published.25,26 In the randomized controlled trial by Burr et al., patients with angina who were given fish-oil capsules had higher rates of sudden cardiac death than untreated controls.25 However, this study was described by von Schacky et al. as suboptimally “conducted and reported”, and thus these results are questionable.27 Unlike the Burr study, the study by Nilsen et al. did not show any harmful effects of omega-3 supplementation.26 However, it did not demonstrate a significant benefit for omega-3 fatty acid supplementation in patients after myocardial infarction either,26 perhaps because the Norwegian participants had a high background intake of fish oils, which could have masked the treatment effects and, with only 300 patients, this study was seriously underpowered.

The effects of omega-3 fatty acids in the primary prevention of cardiovascular mortality have yet to be established in a randomized control trial. Although the primary prevention cohort of the JELIS (n = 14,981 hypercholesterolemic, statin-treated patients) had the same relative risk reduction (18% versus 19%) as the secondary prevention cohort (n = 3,664), the reduction was not statistically significant for the former group (P = 0.13) whereas it was for the latter (P = 0.048). The similarity in effect sizes indicates that this lack of statistically significant effect in the primary prevention cohort was related to sample size and not a difference in efficacy. A subanalysis of data for the primary prevention cohort of JELIS found that EPA treatment reduced major cardiac events by 53% (P = 0.043) in patients with HDL-cholesterol levels less than 40 mg/dl and triglyceride levels greater than 150 mg/dl.28 In a primary prevention trial that is currently underway (NCT00135226) and scheduled for completion in 2011, investigators are enrolling 10,000 patients with type 2 diabetes mellitus and randomly assign them to 100 mg aspirin, 1 g omega-3 fatty acids, both or neither; patients will be followed up to evaluate rate of major cardiac events.

An inverse relationship between CHD risk and fish intake has been consistently observed in prospective observational cohort studies.29 Typically, these primary prevention studies indicate that one fish meal per week is associated with a 15% reduction in CHD risk, and that five or more fish meals per week are associated with a 40% decrease in risk.29 Importantly, two separate observational studies have demonstrated that blood or erythrocyte EPA and DHA levels in the highest quartile are associated with a 70–80% reduction in risk for sudden cardiac death compared with the lowest quartile among apparently healthy men.30,31

Many international bodies including the AHA, ACC, and the European Society of Cardiology have found the overall evidence for benefit sufficiently strong to make public recommendations for increased omega-3 fatty acid intake for primary as well as secondary prevention of cardiovascular disease.1,2,32,33

Fish or fish oils?

The AHA recommends that patients with CHD consume a total of 1 g per day of DHA and EPA, preferably from oily fish, but considers fish oil (in capsules or liquid form) taken in consultation with a physician to be an acceptable option.2 Moreover, the FDA has approved an omega-3 fatty acid formulation at a dose of 4 g daily for the treatment of patients with very high triglyceride levels. This formulation is available with a prescription in the US and contains the highest concentration of omega-3 acid ethyl esters available, with both DHA and EPA represented.

Some experts advocate dietary sources of fatty fish as the main source of omega-3 fatty acids because several important nutrients such as vitamin D, selenium, and naturally occurring antioxidants that are found in oily fish are missing from fish-oil supplements. Indeed, although omega-3 fatty acids have not been consistently shown to lower C-reactive protein levels,34,35 overweight, insulin-resistant patients had significant reductions in C-reactive protein levels after 4 weeks on a diet supplemented with cod protein, when compared with those on diets containing similar quantities of other types of protein.36 Furthermore, selenium, which is present in fish in varying concentrations, is thought to have antioxidant properties as well as the ability to attenuate the adverse affects of methyl mercury and provide cardiovascular benefits.37,38,39,40 Selenium has antithrombotic properties, reduces lipid peroxidation, myocardial infarct size, and ischemia-induced ventricular arrhythmias, improves recovery from ischemia or reperfusion injury, and protects against free radicals.38

However, convincing evidence—particularly from the landmark omega-3 trials GISSI and JELIS22,23,24 clearly indicates that omega-3 fatty acids alone also confer cardiovascular benefits. Furthermore, a study by Harris et al. showed that equivalent amounts of EPA and DHA from either a fish diet or a fish-oil supplement enriched the omega-3 content of erythrocyte membranes equally well.6,41

Certain fish can contain environmental toxins such as mercury (Table 2), polychlorinated biphenyls, chlordane, dioxins, and dichlorodiphenyltrichloroethane (better known as DDT). Many of these toxins have been greatly reduced in the environment through changes in regulatory and industrial practices, but mercury is still of concern. Mercury rarely exists as a free element in nature; it is generally released into the air as a by-product of volcanic activity and industry,38 and is then carried by rain into the water supply where microorganisms convert it into the organic form methyl mercury, which is readily taken up by aquatic organisms. The concentration of methyl mercury increases with progression up the food chain and the highest levels are found in top predators, such as swordfish and shark (Table 2).42 Eating these fish on a frequent basis is not advisable. Aside from occupational exposure or industrial accident, mercury toxicity in humans largely results from consumption of contaminated fish.38,43,44,45 The adverse effects of overexposure to mercury include paresthesia, ataxia, and sensory abnormalities, but are easily reversed with decreased exposure.43 With higher exposure levels and inhalation—not likely from fish ingestion—permanent damage to the kidneys, lungs and even death is possible but rare.46 Although much concern has been expressed about the health risks associated with mercury exposure through fish consumption, a meta-analysis of five published studies showed no significant relationship between increased mercury levels and the risk of CHD (relative risk 1.12, 95% CI 0.71–1.75, P = 0.62).42 Many of the most popular fish varieties, such as salmon, shrimp, canned light tuna, pollock, and catfish, are relatively low in mercury (Table 2) and other chemical contaminants and are safe to eat up to several times per week.44 Unfortunately, with the exception of salmon, these species are relatively poor sources of EPA and DHA (Table 1). The levels of other contaminants, such as polychlorinated biphenyls and dioxin, in fish have been declining steadily as a result of changes in governmental regulatory policies in the US and other developed countries. The risk of cancer from these compounds is minimal compared to the cardiovascular benefits derived from fish consumption.47

Table 2 Levels of methyl mercury* in fish44
Full size table

Given that fish are such a diverse type of food and provide a wide range of omega-3 fatty acids plus many other potentially protective (and harmful) components, at this stage a firm conclusion cannot be reached on whether consumption of fish is better than use of fish-oil supplements. Further research is, therefore, needed. New preparations of fish oil are under investigation, in studies examining various ratios of EPA and DHA, different chemical forms (ethyl esters versus triglycerides versus free fatty acids), and different delivery systems (emulsions, liquid oils, and capsules).

Sustainability of wild fish stocks

Global fish stocks have declined rapidly over the past 50 years. Populations of some commercially popular fish species have collapsed to only 10% of their historic maximum, with over 100 confirmed extinctions of marine species.48,49 Some experts have suggested that worldwide fish stocks could be depleted within 40 years if the harvest continues at the current rate.5,48 This prediction has been severely criticized,50 however, and could be overly pessimistic. Nevertheless, pressure on fish stocks is likely to increase with the number of recommendations for increased consumption of fish and fish oil.

Whether increasing omega-3 intakes can be accomplished in a sustainable manner remains to be seen. More than 90% of US citizens have a suboptimal intake of omega-3,51 and correction of deficiencies in this essential nutrient should be an important objective both for care of individual patients and for public health. Major changes in the way we source omega-3 fatty acids, such as aquaculture (fish farming) and bioengineering, seem to be logical and practical solutions to the conundrum. Nontraditional sources of marine-based omega-3 fatty acids, such as algae (for DHA) and yeast (for EPA) are also being developed. Additionally, the plant-derived omega-3 fatty acids, α-linolenic acid, and stearidonic acid—the δ-6 desaturase product of α-linolenic acid, which can be produced in soybeans by biotechnology—could provide some of the benefits of DHA and EPA.52 One study found that dietary supplementation with walnuts (which contain α-linolenic acid) significantly lowered cholesterol levels while fatty fish (DHA and EPA) significantly lowered triglycerides in healthy or mildly hyperlipidemic individuals.53

Worldwide production of Atlantic salmon from aquaculture has grown from 55,000 tons in 1985 to 1.2 million tons in 2006.54,55 Although this industry might seem to resolve the problem of fish shortages, every kilogram of fish produced is estimated to require an equal amount of fish-oil-based feed. The amount of omega-3 required can vary depending on whether the farmed fish is carnivorous, in which case a higher omega-3 ratio is necessary, as opposed to omnivorous or herbivorous fish, which require a lower ratio on omega-3 in their feed.56 In turn, an estimated 3.2 kg of wild fish stock (commonly from small, pelagic fish like anchoveta, Chilean jack makerel, and Atlantic herring) is required to produce this 1 kg of feed.54,56 Aquaculture might seem inefficient but, to put the issue into perspective, a wild salmon must consume 10–15 kg of fish to gain 1 kg in weight.54 In this respect, although not perfect, aquaculture does have an advantage over the 'by-catch' method of supplying fish.

One adage that holds true for fish is that 'you are what you eat'. The omega-3 content of fish is entirely dependent on their diet; the higher the amount of omega-3 fatty acids they are fed, the greater their omega-3 content. For reasons of cost and supply, fish farmers have attempted to modify the diet of their stock by reducing the proportion of fish oils and increasing the amount of animal and plant oils. Fish fed with a diet of 75% vegetable oil develop without any compromise in their growth and function.54,57 Such fish, however, have a significantly lower omega-3 fatty acid content than fish fed on a 100% fish-oil diet.54 The omega-3 fatty acid content (and, therefore, the cardiovascular benefit of eating the fish produced by this method) depends on the proportion of vegetable oils used in the feed.

One possible solution to the aquaculture problem is to develop an alternative sources of omega-3 for use in fish feed. Miller and colleagues have outlined several possibilities that are currently being investigated, including microbes, marine organisms such as krill and copepods, biosynthetic precursors that contain vegetable oils capable of being metabolized by Atlantic salmon to produce omega-3 fatty acids, and genetically modified plants or microbes that produce omega-3 fatty acids.54 Although they could be developed and implemented in the near future, the cost of using these other sources of omega-3 fatty acids for fish feed could be prohibitive.

Conclusions

Currently available data indicate that patients with known CHD should consume at least 1 g daily of long-chain omega-3 fatty acids from either oily fish or fish-oil supplements, and that individuals without cardiovascular disease should consume at least 250–500 mg daily. Whether oily fish or fish oils are the best source of omega-3 fatty acids is yet to be determined, so either should be used until more data are available. Emerging techniques to produce increased amounts of omega-3 fatty acids without endangering our world fish stocks are being investigated. The provision of omega-3 fatty acids from either traditional or novel sources will need to be accomplished in a sustainable fashion if long-term cardiovascular risk reduction is to be achieved at the population level.

Review criteria

The initial literature review was conducted using the PubMed database and the search terms: “omega-3”, “EPA”, “eicosapentaenoic acid”, “DHA”, “docosahexaenoic acid”, “ALA”, “alpha-linolenic acid”, and “cardiovascular disease”. Full-text articles published in English from 1970 to 2009 were included. Reference lists of comprehensive review articles were further examined for additional reference sources.