By David Kiefer, MD, Editor
Synopsis: In more than 12,000 patients at high risk of cardiovascular events, daily supplementation with 1 g of fish oil (85% eicosapentaenoic acid plus docosahexaenoic acid) did not reduce cardiovascular morbidity nor mortality over approximately 5 years of follow up.
Source: Risk and Prevention Study Collaborative Group. n-3 fatty acids in patients with multiple cardiovascular risk factors. N Engl J Med 2013;368:1800-1808.
The authors of this study recruited patients at high risk of cardiovascular (CV) disease from 860 general practitioners in clinical practice throughout Italy. The inclusion criteria for research participants are listed in Table 1. Patients were excluded from this study if they previously had a myocardial infarction, were hypersensitive to omega-3 fatty acids, had poor short-term prognosis, or were unable to provide informed consent.
The 12,513 enrollees in the study were randomly assigned, in a double-blind fashion, to either a 1 g capsule of omega-3 fatty acids (omega-3; n = 6244) or an olive oil placebo (n = 6269) once daily for a median follow-up of 5 years. The omega-3 fatty acid capsule contained the ethyl ester polyunsaturated fatty acids, at least 85% of which were eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in a ratio ranging from 0.9:1 to 1.5:1.
Enrollees had a baseline visit and yearly follow-up visits. These visits were with their general practitioner who collected information about interim health history, CV risk factors, lifestyle, new diagnoses, and changes in medications, in addition to physical examination parameters and laboratory testing. The primary endpoint of this study included time to CV death or first hospitalization for CV causes, both listed as event rates or percentages. Table 2 lists these endpoints and the numerous secondary endpoints of these studies.
The study had an intention-to-treat methodology, in the sense that even patients whose treatment was stopped were followed until the end of the study. This resulted in 12,505 patients being analyzed — 6239 in the omega-3 group and 6266 in the placebo group. Baseline characteristics were similar between the two groups, except for slightly more men and people on calcium channel blockers in the omega-3 group. The most common criteria being met for inclusion in the study was the presence of diabetes mellitus plus one or more CV risks (47.9% of patients).
|Table 1: Criteria for Inclusion|
|"Multiple cardiovascular risk factors" (at least four of the following)|
|"Clinical evidence of atherosclerotic vascular disease"|
|"Any other condition putting the patient at high cardiovascular risk in the opinion of the patient's general practitioner"|
After a median follow up of 5 years, only a few differences were noted in the omega-3 group vs the placebo group (See Table 3). No differences were noted for blood pressure, LDL cholesterol, total cholesterol, blood glucose, hemoglobin A1C, nor CV drug prescriptions. The primary event rate occurred in 11.8% of patients: 11.7% in the omega-3 group and 11.9% in the placebo group; this was statistically similar (hazard ratio, 0.97; confidence interval, 0.88-1.08; P = 0.64). Similarly, there was no difference between the two groups for any of the secondary endpoints as listed in Table 2.
|Table 2:Primary and Secondary Endpoints Measured|
|Primary Endpoints||Secondary Endpoints|
|Time to cardiovascular disease death||Time to death|
|Time to first hospitalization for cardiovascular causes||Time to nonfatal myocardial infarction|
|Time to nonfatal stroke|
|Death from coronary heart disease|
|Sudden death from cardiac causes|
|Table 3: Two Differences Between Omega-3 Group and Placebo Group After 5 Years of Follow Up|
|Omega-3 Group||Placebo Group||Statistics|
|Triglycerides lower by 28.2 mg/dL||Triglycerides lower by 20.1 mg/dL||
P < 0.001
|"Slight increase in HDL"||Not provided|
|96 admissions for heart failure||142 admissions for heart failure||P = 0.002|
|Lower primary event rate for women than for men||HR = 0.82, CI = 0.67-0.99, P = 0.04|
The most common adverse effects were gastrointestinal, although the incidence was similar between the omega-3 and placebo groups. Two severe cases of epistaxis occurred in the omega-3 groups; both of these people were taking blood-thinning medication. Interestingly, "bleeding" as an adverse event was similar between the two groups
(n = 16 in the omega-3 group and n = 12 in the placebo group, P = 0.44).
As an aside, the funding for this trial came from Pfizer and two societies, Sigma Tau and an antibiotic production organization, but the authors explicitly stated that these groups had no role whatsoever in the planning nor implementation of the research nor in the analysis of the results.
This study advances our knowledge about the use of fish oil supplementation. Common omega-3 recommendations involve the consumption of two servings of fish twice weekly for primary CV prevention, and 1 g of EPA plus DHA daily for secondary CV prevention.1 This article is one of the first, and largest, to examine fish oil supplementation for primary CV prevention in a high CV risk population. If we believe the veracity and applicability of these results, then it appears that fish oil supplementation does not function in this capacity for this demographic.
The results of this study agree with a recent meta-analysis of any CV outcomes across demographics.2 This congruence improves its believability. What may detract from its relevance to clinical medicine in the United States is the location of the trial. The Italian patients studied, likely closely approximating the country as a whole given the large number of clinics and geographic locations covered, had a lower than expected rate of "hard endpoints," or CV morbidity and mortality, especially as the trial progressed over the years. The authors comment that there could have been a significant effect of preventive CV measures throughout Italy or dietary habits (the Mediterranean diet). It could be argued the typical diet in the United States diverges from that seen in Italy, and that a "nudge" from omega-3 supplementation might make more of a CV difference. A U.S. study of the size and methodological quality as seen in the Italian analysis would be welcome to shed light on this.
Possibly a minor detail, but there are different formulations of omega-3 fatty acids that may affect physiology and, therefore, clinical outcomes. In this study, the product used was an ethyl ester omega-3, whereas other studies have used a triglyceride form. Manufacturers of each product type try to tout the benefits of their particular form of omega-3, but the medical literature is relatively mute on the topic. A not-so-recent review doubted that different omega-3 sources had CV relevance,3 but the possibility that omega-3 form is important, or not, needs to be clarified at some point. In addition, most experts use much higher doses (2-4 g of DHA plus EPA daily) of omega-3 to have meaningful physiological effects, such as the lowering of serum triglycerides.1 Therefore, this study may not have documented improvements in the primary outcomes simply due to underdosing.
Omega-3 research does not seem to be waning, but rather more, and more specific, studies are likely to continue to be published in the years to come. All of this should help to clarify the dose, form (supplementation, diet), and demographic for which this intervention is most useful or minimally effective. That said, the current state of the literature still supports the normalization of the omega-6 to omega-3 ratio (through the lowering of omega-6 intake and the increase of omega-3 intake through diet or supplementation) for inflammatory disorders (i.e., arthritis) and fish oil supplementation for hypertriglyceridemia. Intriguing results also exist for many psychological diagnoses — such as mood disorders and attention deficit disorder, and during pregnancy and lactation — that warrant our clinical attention, especially given omega-3's favorable side effect profile.