Marine-Derived Omega-3 Fatty Acids and Cardiovascular Disease
Thomas G. Guilliams, PhD
Jørn Dyerberg, MD, DMSc, DHC
Omega-3 Fatty Acids and Cardiovascular Disease
One of the most complex (and controversial) associations between food intake and cardiovascular disease (CVD) risk involves the dietary intake of fatty acids. Epidemiological and animal model data from the early 20th century linked the dietary intake of fat (primarily saturated) and cholesterol with increased risk for cardiovascular events in Western countries, which was followed by decades of recommendations for a low-fat/low-cholesterol diet to prevent CVD.1,2,3 Today, there is a much more nuanced understanding of the complex relationship between fats and CVD risk, including their types (eg, saturated, unsaturated, omega-3, omega-6, omega-9), molecular structures (eg, trans versus cis bonds, triglycerides, phospholipids [PLs], free fatty acids, ethyl esters [EEs]), sources (eg, plant, animal, marine), and the doses and ratios of each. Among other discoveries, a significant shift in our understanding of the association between fatty acid intake and CVD occurred 50 years ago when Hans Olaf Bang and Jørn Dyerberg discovered that the Inuit populations of Greenland, eating a traditional diet with an extremely high intake of dietary fat, had a lower risk for cardiovascular events than Western populations consuming less fat. This discovery, and the subsequent publication of these findings by Bang and Dyerberg, initiated a revolution in both nutritional research and cardiovascular intervention.4,5,6 Since then, the long-chain marine omega-3 fatty acids discovered in the Inuit diet and blood samples have been investigated for nearly every disease outcome and biomarker, including hundreds of clinical trials in healthy and at-risk populations. In that time, omega-3 fatty acid supplementation (primarily eicosapentaenoic acid [EPA] and/or docosahexaenoic acid [DHA] from fish) has become increasingly popular as a means for reducing CVD risk, as illustrated by the inclusion of omega-3 fatty acids in the American Heart Association dietary recommendations for the prevention of heart disease and the US Food and Drug Administration (FDA) approval of several omega-3 pharmaceutical products for a CVD-related biomarker.
However, the increased popularity of omega-3 fatty acid supplementation is not without controversy or confusion. Recent reevaluation of clinical trials performed years ago, through meta-analytical methodology, has led to several publications suggesting there is limited cardiovascular risk reduction associated with EPA and DHA supplementation, whereas epidemiological studies on the effect of a high fish intake continue to show a positive association with reduced CVD risk. These controversies are further mixed with numerous other debates surrounding the design of clinical trials testing omega-3 fatty acids for CVD outcomes, including how CVD risk should be measured (ie, biomarker versus end points), whether primary or secondary prevention studies are more appropriate to measure risk, bioavailability issues of omega-3 fatty acid supplements, the baseline omega-3 status of trial participants, and the concomitant use of other risk-reducing agents (ie, statin drugs) in these trials. At nearly the same time, the results of large and well-designed intervention trials have been recently published demonstrating significant CVD risk reduction after consuming n-3 fatty acid supplements. It is not surprising, then, that there is still much confusion as to the utility of supplementing omega-3 fatty acids for cardiovascular event risk reduction.
In this chapter, we will review these recent studies, as well as several seminal clinical trials from the recent past. In addition, we will discuss the mechanisms underlying the association between omega-3 fatty acid intake and the pathophysiology and biomarkers known to influence the risk for CVD. Included in this chapter is a discussion of various products (eg, dietary supplements, pharmaceuticals) designed to deliver omega-3 fatty acids for therapeutic use and important distinctions between these products.
Omega-3 Fatty Acids
Omega-3 (n-3) fatty acids are a class of polyunsaturated fatty acids derived from the 18-carbon essential fatty acid alpha-linolenic acid (ALA, 18:3 (n-3)). Although ALA is only synthesized in plants, some plants (eg, algae) and animals consuming ALA can elongate and desaturate this molecule to produce several different long-chain polyunsaturated fatty acids (LCPUFAs) such as stearidonic acid (18:4 (n-3)), eicosatetraenoic acid (ETA, 20:4 (n-3)), EPA (20:5 (n-3)), docosapentaenoic acid (DPA, 22:5 (n-3)), and DHA (22:6 (n-3)). Although each of these fatty acids have been associated with health benefits in humans and garnered research attention, the two fatty acids that have gained the most attention and research focus are EPA and DHA. In fact, the vast majority of epidemiological studies connecting n-3 fatty acids and cardiovascular risk often only measure dietary intake of, or biomarkers for, EPA and DHA; furthermore, intervention trials using supplemental n-3 fatty acids are often described solely by their EPA and/or DHA content (even though they may contain other n-3 intermediates). Therefore, before diving into the epidemiological, interventional, and mechanistic studies of EPA and DHA, it is important to briefly describe the sources and molecular structures of these compounds as they are found in foods, supplements, and pharmaceuticals—because these factors can greatly influence their bioavailability, efficacy, and cost.
Sources of Marine Omega-3 Ingredients
For the most part, the marine n-3 fatty acid category is dominated by products best described as “fish oil”—that is, although there are products available that deliver n-3 fatty acids from other marine sources, nearly all the available research has been done with fish oil-derived fatty acids. These data using fish oil have become the benchmark for efficacy and safety and are the standard to which we compare other sources. Although pharmaceutical products often avoid the use of the term “fish oil,” these products are currently all made from fish-derived fatty acids.
The following are the main sources of marine omega-3 fatty acids.
Fish Body Oil: The largest biomass used to create marine-derived n-3 fatty acids are small oily fish caught in the cold waters off the coast of Chile and Peru. The fish species most commonly used are anchovies and sardines, with some mackerel. Concentrations of these purified oils are the most common therapeutic ingredient used in dietary supplements and pharmaceutical products throughout the world. Other species used to produce fish oil may include salmon, tuna, menhaden, herring, and other minor species. The EPA and DHA (and other fatty acids) content, which is predominantly in the triglyceride form, is dependent on the species of fish, the water temperature, and other variables on a seasonal basis.
Cod Liver: As a by-product of the cod meat market, cod livers are used to provide a blend of fatty acids similar to unconcentrated fish body oil.
Krill: These small crustaceans feed on plankton and are subsequently eaten by many marine mammals, especially penguins and whales. Factory ships process krill immediately upon capture off the coast of Antarctica. Krill oil, in which the n-3 LCPUFAs are predominantly in a free fatty acid and PL form, is relatively low in EPA and DHA but contains small amount of the carotenoid astaxanthin.
Calamari: A more recent, but small, player in the n-3 fatty acid industry is calamari or squid oil. This oil, which is predominantly in the triglyceride form, has a higher ratio of DHA over EPA than is typical of fish oil. This material is a by-product of the calamari food industry.
Mussels: Shellfish are only a minor source of commercially available n-3 fatty acids. Nonetheless, several products are currently available from the fatty acids derived from green-lipped mussels (Perna canaliculus). These ingredients are not typically marketed for cardiovascular benefits.
Algae: Various species of algae are commercial sources for n-3 fatty acids. Algae can be grown in large inland production sites where access to sunlight is plentiful. These products are very high in DHA, with only small amounts of EPA, predominantly in the triglyceride form. Most of the pure DHA raw materials, especially pure DHA used for the fortification of infant formula, is sourced from algae. In addition, algae are currently the only vegan source of DHA available.
EPA and DHA from genetically modified plants: Various algae, plants, and fungi have been genetically modified to produce various fatty acids, including both EPA and DHA. These ingredients are designed to help increase the global supply of these fatty acids, while limiting the harvesting burden on marine animals. As of 2018, these ingredients were only being produced for the supplementation of farm-raised fish (not directly used in dietary supplement ingredients).7,8 It is possible these plant-derived EPA and DHA fatty acids may be approved for direct human consumption in the future.
Delivery Forms for Supplementation
When fatty acids are harvested from their source, they are typically in the form of triglycerides (TGs), PLs, or free fatty acids (FFAs) and are relatively low in total EPA and DHA (<30%). When consuming fish or unconcentrated fish oil (ie, fish body oil or cod liver oil), these fatty acids are in the TG form, as they are in most plant and animal sources of fat. However, because the recommended doses of EPA and DHA are often difficult to consume using unconcentrated oils, several steps can be used to increase the EPA and DHA concentration of the product while increasing the purity of the fatty acids delivered. The EPA and DHA fatty acids can be removed from their glycerol backbone and separated from other fatty acids (via hydrolysis and distillation). These fatty acids are then concentrated as EEs of EPA and DHA. These
concentrated fatty acids can be reattached to a glycerol backbone to form re-esterified TG (rTG) molecules that contain a much higher concentration of EPA and DHA compared with the original TG molecule. These two forms of concentrated fish oil (EE and rTG) are the most common sources used in clinical trials and often recommended by physicians (as dietary supplements or pharmaceuticals). It is important to note the distinctions between the various delivery forms of these fatty acids, as this often impacts their bioavailability and efficacy. The use of FFAs and PLs (primarily from krill), as well as other factors affecting the quality, safety, and bioavailability of n-3 products (eg, heavy metals, pesticides, oxidation), will be discussed further.
concentrated fatty acids can be reattached to a glycerol backbone to form re-esterified TG (rTG) molecules that contain a much higher concentration of EPA and DHA compared with the original TG molecule. These two forms of concentrated fish oil (EE and rTG) are the most common sources used in clinical trials and often recommended by physicians (as dietary supplements or pharmaceuticals). It is important to note the distinctions between the various delivery forms of these fatty acids, as this often impacts their bioavailability and efficacy. The use of FFAs and PLs (primarily from krill), as well as other factors affecting the quality, safety, and bioavailability of n-3 products (eg, heavy metals, pesticides, oxidation), will be discussed further.
The Epidemiology of Omega-3 FA and CVD
Epidemiological and cohort studies have repeatedly shown that higher dietary intake of fatty fish and/or a person’s n-3 status (as measured by EPA and DHA in serum, plasma, or red blood cell [RBC] membranes) is inversely associated with cardiovascular events and/or cardiovascular mortality.9,10,11,12 For instance, in a cohort of 20,551 men from the Physician’s Health Study, the multivariate relative risk (RR) for sudden cardiac death in those consuming one fish meal per week was 0.48 (P = .04), compared with men who consumed fish less than once per month.13 The adjusted RR in the highest quartile of RBC n-3 levels (compared with the lowest quartile) in this population was just 0.19 (P = .007).9 The Honolulu Heart Program followed Japanese-Americans living in Hawaii and found the RR for coronary heart disease (CHD) mortality was cut in half for heavy smokers (>30 cigarettes/d) if they consumed more than two fish meals per week.14 Several recent meta-analyses of prospective cohort studies have confirmed these overall results. Chowdhury et al. (2014) analyzed 16 studies exploring the relationship between long-chain n-3 dietary intake (lowest versus highest tertile) and coronary outcomes and reported an RR of 0.87 (95% confidence interval [CI], 0.78-0.97).15 By comparison, ALA intake had no statistical relation to coronary outcomes (RR = 0.99, N = 7 studies), whereas total trans fatty acid intake contributed to a significant increase in coronary outcomes (RR = 1.16, N = 5). Furthermore, their analysis of studies comparing coronary outcomes based on circulating fatty acids (top versus lowest tertile, N = 13) revealed statistically significant risk reduction for EPA (RR = 0.78), DHA (RR = 0.79), and EPA + DHA (RR = 0.75). In a more recent meta-analysis, Alexander et al. (2017) analyzed 17 prospective cohort studies and found a significant reduction in CHD events (RR = 0.82), coronary deaths (RR = 0.82), and sudden cardiac deaths (RR = 0.53) comparing subjects consuming the highest versus lowest intake of n-3 fatty acids.16 Notably, an analysis of data generated through the National Health and Nutrition Examination Survey (NHANES-2012) estimated that insufficient intake of marine n-3 fatty acids was the “cause” of over 54,000 CVD-related deaths annually.17
Omega-3 Index and CVD Risk
Because the dietary intake of omega-3 fatty acids from foods or supplements may not always correlate with biomarkers of omega-3 status, several investigators have focused on the incorporation of long-chain n-3 fatty acids (primarily EPA and DHA) within RBC membrane fatty acids as a way to measure the long-term absorption and tissue deposition of n-3 fatty acids.18 In fact, the percentage of EPA and DHA within RBC membranes, known generally as the “Omega-3 Index” or O-3I, is inversely related to cardiovascular events and mortality, whereby the highest risk is associated with subjects with an omega-3 index less than 4% and the lowest risk is in subjects with an omega-3 index greater than 8%.19 A recent meta-analysis of 10 cohort studies measuring risk in subjects based on their estimated omega-3 index predicted that subjects with an omega-3 index of 8% had a 30% lower risk for fatal CHD compared with those with an omega-3 index of 4%.20 Therefore, because the risk reduction potential of supplemental EPA and/or DHA is likely dependent on dose, absorption, and tissue incorporation, determining a subject’s baseline and postsupplemental omega-3 index may be necessary to optimize risk reduction-related n-3 supplementation.
Testing a patient’s omega-3 index may be especially important to ensure that they are being recommended the correct dose to optimize their CVD risk reduction. Flock et al. have convincingly shown that the treatment dose of EPA and DHA (TG from fish) has a predictable effect on the change in omega-3 index, but they also discovered interindividual differences (especially based on a person’s body weight) account for a high degree of variability in the omega-3 index changes seen after consuming EPA and DHA (Figure 7.1).21 Therefore, based on this study, it is essential for the clinician to understand that, without testing a person’s omega-3 index, it is not possible to predict the patient’s current (or change in) omega-3 index, based simply on the n-3 fatty acid dose given (an important factor in interpreting clinical trials where all subjects are given the same dose of n-3). The omega-3 index is readily available through numerous laboratories and can easily be incorporated into a clinician’s CVD risk assessment.
CVD Intervention Studies Using Omega-3 Fatty Acid
Although epidemiological and cohort studies have consistently shown substantial risk reduction in subjects consuming higher amounts of long-chain n-3 fatty acids, primary and secondary prevention trials have resulted in much more heterogeneity. One of the first studies assessing the secondary prevention potential of n-3 fatty acids from fish was the Diet And Reinfarction Trial (DART).22 Men (N = 2033) recovering from a myocardial infarction (MI) were randomized to receive one of three different dietary recommendations: to increase fatty fish consumption, to increase fiber, or to reduce fat intake. Those advised to increase fatty fish
consumption had a 29% reduction in 2-year all-cause mortality, whereas neither the low-fat or fiber recommendation groups had a meaningful risk reduction. Unfortunately, like many lifestyle changes, this advice was difficult to maintain over many years and both compliance and benefits diminished after a decade.23
consumption had a 29% reduction in 2-year all-cause mortality, whereas neither the low-fat or fiber recommendation groups had a meaningful risk reduction. Unfortunately, like many lifestyle changes, this advice was difficult to maintain over many years and both compliance and benefits diminished after a decade.23
 Figure 7.1 Change in omega-3 index after 5 months of different doses of EPA + DHA in healthy subjects. A, Treatment dose significantly predicted changes in omega-3 index. Baseline omega-3 index for groups was 4.3%. Note that, although the average increases as the dose increases, there are many individuals with much higher (or lower) than average changes with each dose. B, Body weight effects on omega-3 index changes. The amount of EPA + DHA in grams consumed per kilogram of body weight significantly predicted changes in the omega-3 index. (From Flock MR, Skulas-Ray AC, Harris WS, et al. Determinants of erythrocyte omega-3 fatty acid content in response to fish oil supplementation: a dose-response randomized controlled trial. J Am Heart Assoc. 2013;2:e000513.) |
Until recently, the largest and most cited secondary prevention trial investigating the supplementation of EPA and DHA was the GISSI-prevention trial.24 In this study, over 11,000 patients (surviving a recent MI) were randomized to one of three supplement groups: those given one gelatin capsule containing 850 to 882 mg of EPA and DHA (as EEs in the average ratio of EPA:DHA, 1:2), those given 300 mg of vitamin E (as acetyl alpha tocopherol, synthetic all racemic), or those give both n-3 fatty acids and vitamin E. Most of these patients were concomitantly taking nonstatin cardiovascular pharmaceuticals of various kinds, as well as advised about diet and lifestyle changes. Total (RR = 0.59) and cardiovascular mortality (RR = 0.66) were significantly reduced in the fish oil groups as early as 3 and 4 months into the study, respectively. The most dramatic reduction was in sudden deaths, for which RRs of 0.37 (after 9 months) and 0.55 (42 months) were reported.25 Among the lipids measured, only triglyceride levels showed significant improvements. The results of the GISSI-prevention trial initiated the wide-spread use of concentrated EPA and DHA products around the world and justified numerous official recommendations for the use of EPA and DHA for CVD prevention and management (eg, American Heart Association’s recommendation to consume 1 g/d of EPA and DHA from fatty fish or supplements).26
In the years between the publication of the DART and GISSI trials (1988-1999) there were over 20 randomized clinical trials evaluating the role of n-3 supplementation or increased fatty fish consumption on CVD, 10 of which met the criteria for a meta-analytical study.27 Although many of these trials were of suboptimal quality, these data showed that the daily intake of long-chain n-3 fatty acids from fish for an average of 3 years resulted in a 16% decrease in all-cause mortality and a 24% decrease in the incidence of death from MI. We should note that none of the subjects in these 10 trials were taking statin drugs and, with the exception of the GISSI-prevention trial, the n-3 fatty acid doses given were all > 1.5 g of EPA and DHA per day. Both are important factors in comparing these data with more recent clinical trials, which are almost universally performed in statin-treated individuals (for secondary prevention) and often use n-3 fatty acid doses below 1 g/d.
In the past two decades, hundreds of clinical trials have been performed using a variety of doses, combinations, and types of EPA and DHA with respect to nearly every CVD-related end point. Although many of these were trials exploring the effects of n-3 FA on a variety of biomarkers (eg, TG, C-reactive protein [CRP], low-density lipoprotein-cholesterol [LDL-C], lipoprotein number and size, blood pressure), several of these studies were designed as primary and secondary prevention trials, measuring MACE (major adverse cardiac events; nonfatal MI/strokes, CVD deaths) and/or all-cause mortality. In 2017 and 2018, several well-publicized meta-analyses examined and compared these latter trials with some of the earlier trials mentioned before, determining there was no significant reduction in fatal or nonfatal CVD events in subjects randomized to n-3 FA therapies in these trials.16,28,29 Of these, Aung et al., published in JAMA Cardiology, has had a significant negative impact on cardiologists’ view of n-3 supplementation.28
However, as with many large intervention trials evaluating nutrients for drug-like outcomes, this meta-analysis, and the studies upon which it was based, has significant limitations. First, the analysis included only studies with greater
than 500 participants treated for more than 1 year, which restricted their analysis to only 10 trials. Furthermore, 83.4% of the nearly 78,0000 subjects included in the analysis were concurrently using statin therapy, which reflects the changes in cardiovascular therapy from the previously mentioned cohort of n-3 studies. Perhaps more problematic was the fact that the omega-3 status was not used as an inclusion/exclusion criterion for any of these trials nor reported as a biomarker in these studies. This is important because, unlike drugs, participants would have started each trial with varying levels of EPA and DHA, greatly influencing their ability to achieve risk reduction through n-3 supplementation. In addition, as pointed out by von Schacky, these trials used a fixed dose of EPA and DHA, most often below 1 g (usually during a low-fat breakfast), which was likely to result in poor bioavailability (overall) and a large interindividual dose-response on improved omega-3 index.30 Although some of the same limitations apply to the meta-analysis performed by Alexander et al., published in Mayo Clinic Proceedings, subgroup analysis of their 17 included trials revealed statistically significant benefits in subjects with elevated baseline TG > 150 mg/dL (RR = 0.84; 95% CI, 0.72-0.98) or baseline LDL-C >130 mg/dL (RR = 0.86; 95% CI, 0.76-0.98). Importantly, in this study, the strongest benefit was seen in subjects with elevated baseline TG given an n-3 fatty acid dose greater than 1 g/d of EPA and DHA (RR = 0.75; 95% CI, 0.64-0.89). This magnitude of risk reduction aligns with the prospective cohort data mentioned previously, suggesting that higher doses of n-3 fatty acids over a short-term period (ie, less than 5 years) may be needed to realize benefits that are typically achieved by extended periods of low to moderate n-3 intake. Data from recent large clinical trials may confirm this notion.
than 500 participants treated for more than 1 year, which restricted their analysis to only 10 trials. Furthermore, 83.4% of the nearly 78,0000 subjects included in the analysis were concurrently using statin therapy, which reflects the changes in cardiovascular therapy from the previously mentioned cohort of n-3 studies. Perhaps more problematic was the fact that the omega-3 status was not used as an inclusion/exclusion criterion for any of these trials nor reported as a biomarker in these studies. This is important because, unlike drugs, participants would have started each trial with varying levels of EPA and DHA, greatly influencing their ability to achieve risk reduction through n-3 supplementation. In addition, as pointed out by von Schacky, these trials used a fixed dose of EPA and DHA, most often below 1 g (usually during a low-fat breakfast), which was likely to result in poor bioavailability (overall) and a large interindividual dose-response on improved omega-3 index.30 Although some of the same limitations apply to the meta-analysis performed by Alexander et al., published in Mayo Clinic Proceedings, subgroup analysis of their 17 included trials revealed statistically significant benefits in subjects with elevated baseline TG > 150 mg/dL (RR = 0.84; 95% CI, 0.72-0.98) or baseline LDL-C >130 mg/dL (RR = 0.86; 95% CI, 0.76-0.98). Importantly, in this study, the strongest benefit was seen in subjects with elevated baseline TG given an n-3 fatty acid dose greater than 1 g/d of EPA and DHA (RR = 0.75; 95% CI, 0.64-0.89). This magnitude of risk reduction aligns with the prospective cohort data mentioned previously, suggesting that higher doses of n-3 fatty acids over a short-term period (ie, less than 5 years) may be needed to realize benefits that are typically achieved by extended periods of low to moderate n-3 intake. Data from recent large clinical trials may confirm this notion.
REDUCE-IT (2018)
One of these large clinical trials, the Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial (REDUCE-IT), has garnered much attention.31 Like the successful Japan EPA Lipid Intervention Study (JELIS), the REDUCE-IT trial employed the use of the approved drug Vascepa, which is an EPA-only, EE product.32 In this secondary prevention study, 8179 subjects with elevated fasting TG (between 150 and 499 mg/dL) and LDL-C of 41 to 100 mg/dL (all subjects were concurrently medicated with a stable dose of statin therapy) were randomized to receive placebo (mineral oil) or 4 g daily (2 g twice daily, with food) of EPA-EEs and followed for nearly 5 years. The major clinical end point was the cumulative incidence of cardiovascular events (CVD deaths, nonfatal MI, nonfatal stroke, coronary revascularization or unstable angina), which was 25% lower in the fish oil therapy group compared with placebo (hazard ratio 0.75, P < .001). As anticipated, TG levels in these subjects, which averaged 216.5 mg/dL at baseline, were significantly reduced (18.3%-21.7%) during the trial. Although the company producing Vascepa (Amarin) often remarks that EPA therapy (as opposed to DHA therapy) does not raise LDL-C, these subjects realized a small but statistically significant increase in LDL-C from baseline (3.1%, P < .001). Because the LDL-C increase was coincident with an increase in HDL-C and a decrease in ApoB of similar magnitudes, these changes likely reflect a beneficial shift in LDL particle size and number (not reported). The lowest statistically significant hazard ratio (HR) following n-3 therapy was for subjects having baseline TG > 200 mg/dL and HDL <35 mg/dL (HR = 0.62). There is significant debate (both scientifically and commercially) as to the applicability of these data to other (similar) available products (ie, dietary supplements) that provide concentrated EPA or EPA/DHA in either the EE or rTG form. For a discussion of the potential differences between EPA and DHA and the bioavailability differences between EE and rTG see subsequent text.
VITAL (2018)
Published in the same November 2018 issue of the New England Journal of Medicine as the REDUCE-IT trial, the Vitamin D and Omega-3 Trial (VITAL) evaluated a more traditional n-3 dose and product in a large primary prevention study.33 Participants (N = 25,871) were randomized to one of four groups: (1) n-3 fatty acids (1 g fish oil capsule/day [Omacor/Lovaza] providing EE forms of EPA [460 mg] and DHA [380 mg]), (2) vitamin D3 (2000 IU/d), (3) both n-3 and vitamin D3, or (4) both placebos (olive oil used for fish oil placebo). This dose and n-3 form was based on the recommendation of the American Heart Association (for cardioprotection) and the late-1990s GISSI-prevention trial. As a primary prevention trial in subjects over the age of 50 years (mean 67.1 years), only 4.2% of the subjects in VITAL experienced a cardiovascular event (defined in the composite end points: MI, stroke, cardiovascular deaths, or coronary revascularization) during the 5 years of the trial compared with greater than 19% of subjects in the REDUCE-IT trial. However, when subjects taking the n-3 supplements were compared with those taking olive oil, there was only a nonstatistical trend in the reduction of major cardiovascular events (HR = 0.93; 95% CI, 0.80-1.04). Secondary end-point analysis suggested some benefits with n-3 supplementation, such as total MI (HR = 0.72; 95% CI, 0.59-0.90), deaths from MI (HR = 0.50; 95% CI, 0.26-0.97), and events in subjects with fish consumption <1.5 servings per week (HR = 0.81; 95% CI, 0.67-0.98).
Overall, these data are not unexpected, based on the shortcomings of the trial design: primarily the low number of events in this populations, the comparatively low n-3 dose to achieve meaningful event reduction (evidenced by achieving an omega-3 index of only 4.1% [from 2.7%] after 1 year of n-3 supplementation), and the use of olive oil (known to reduce CVD events at higher doses) as a placebo.34 These shortcomings should be considered when evaluating the (mostly) negative results of VITAL, one of the largest clinical trials performed using EPA and DHA for the primary prevention of CVD. Unfortunately, some of these same shortcomings exist in a similar trial published earlier in 2018, ASCEND.
ASCEND (2018)
A Study of Cardiovascular Events in Diabetes (ASCEND) was also a primary prevention trial including only diabetic subjects (N = 15,480) with no evidence of CVD.35 Subjects were randomized to receive a 1-g fish oil capsule per day (Omacor/Lovaza, providing EE forms of EPA [460 mg] and DHA [380 mg]) or placebo (olive oil) and followed for an average of 7.4 years. The primary outcome was a composite of nonfatal MI, stroke, or vascular deaths, whereas secondary end points included other serious vascular events or any arterial revascularization. Over the length of the trial, the group randomized to fish oil had 689 events (8.9%), whereas the olive oil group had 712 events (9.2%); this difference was not statistically significant (RR = 0.97; 95% CI, 0.87-1.08). In fact, with the exception of vascular deaths (RR = 0.81; 95% CI, 0.67-0.99), there were no beneficial differences in events reported between the n-3 and olive oil groups.
Again, these data are not surprising for some of the same reasons discussed previously in the VITAL study (eg, low n-3 dose, olive oil placebo), with two important differences. Because this group included only diabetic subjects (all types), these subjects had a higher CVD risk, which was evidenced by the higher number of events recorded in ASCEND (9.0%) compared with VITAL (4.2%), although still much fewer than the secondary prevention REDUCE-IT trial (19%). However, unlike the American subjects recruited for VITAL who had a mean baseline omega-3 index of just 2.7% (high risk), the UK subjects recruited for ASCEND had a mean baseline omega-3 index of 7.1% (which increased to 9.1% after supplementation). This suggests that the ASCEND subjects were already at low risk based on their baseline n-3 levels, limiting the ability of n-3 supplementation to alter that risk. On the other hand, because the VITAL group was only able to achieve an omega-3 index of 4.1% after supplementation, these subjects never achieved an omega-3 status associated with lower risk. Because neither trial reported TG levels of the participants, it is difficult to know whether either group included subjects with TG-associated risk and whether n-3 supplementation altered subjects’ TG levels. Overall, these data suggest that routine n-3 supplementation of less than 1 g/d is unlikely to reduce measures of CVD events in populations with an average baseline omega-3 index of less than 4% or greater than 7%, at least when compared with subjects given olive oil placebos, over a span of less than 7 years.36 However, several large clinical trials are currently underway that are soon likely to influence our understanding of the role of n-3 fatty acid supplementation in CVD primary and secondary prevention.37,38,39
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
