A meta-analysis published last week in the Journal of the American Medical Association reported no effect of omega-3 fatty acid supplements—often referred to as "fish oil"—on cardiovascular event rates. Predictably, the study made headlines. It was welcomed by those inclined to disparage the touchy-feely world of alternative medical treatments and was lamented and rejected by advocates for the kinder, gentler remedies of nature. Which view is correct? Let's see if we can fillet open the details.
The study conclusion, and the parade of related pop culture headlines, indicated no significant effects of omega-3 supplementation on any of the five outcomes of interest. But, actually, something about that punch line is more than a little fishy.
The study showed, in fact, that omega-3 supplementation was associated with a significant reduction in cardiac death, and nearly significant reductions in dying of any cause, sudden cardiac death, and heart attack.
In some ways, even a suggestion of clinically relevant effects of omega-3 fatty acids in such a context is rather extraordinary. Consider that the fundamental requirement for a randomized, controlled trial testing a hypothesis—in this case, that omega-3 fatty acid supplements can reduce cardiac event rates—is that the two treatment groups be alike in all ways but the intervention. In other words, the study participants should have everything in common, on average, except that one group gets fish oil, and the other group gets placebo.
Since these were studies of people with significant cardiac risk factors, or prior cardiac events, they were of course receiving state-of-the-art treatments. It would be unethical to take cardiac patients off of their beta-blockers, anti-arrhythmic drugs, platelet inhibitors, or statins to test omega-3s. It would be unethical to extract pacemakers or defibrillators to assess the effects of omega-3 fats on heart rhythm.
In other words, to see an effect of omega-3 on cardiac outcomes in this meta-analysis, it would have to be in addition to the best effects achieved with modern drug therapy, and, presumably, the lifestyle intervention of cardiac rehabilitation programs as well. There is nothing wrong with this, per se. As noted, testing this hypothesis any other way would be unethical. But note that virtually no drug would show any effect on the outcomes studied here if it were added to state-of-the-art treatment. The deck is stacked mightily against observing any effect.
Of course, some drugs do clear such hurdles—and are then added to the conventional cocktail. But the more advanced standard treatment becomes, the higher the bar any treatment needs to clear to show additionalbenefit.
And, of course, a study of fish oil supplementation does not preclude participants from eating fish, or walnuts, or flaxseeds. So the baseline here is not "no" omega-3; all study participants were getting some. Again, it would not be ethical to eliminate all omega-3 intake from the control group, because omega-3 fatty acids are considered essential nutrients. So whereas drug trials truly compare a therapeutic dose of a drug to none, the omega-3 trials compare "more" to "some." Many drugs that pass the test of a "some to none" comparison would fail a "more to some" comparison.
The fact that omega-3 fatty acids are essential nutrients is far from trivial. What this means is that these fatty acids—like all other essential nutrients, including vitamins, minerals, and amino acids—are used routinely by the body in critical metabolic functions. In the case of omega-3 fats, they are incorporated into cell membranes and are a major construction material for a large family of hormones known as prostaglandins.
Prostaglandins are, among other things, regulators of the immune system and the body's inflammatory responses. Some classes of fat—including most omega-6 polyunsaturates, and saturated fatty acids—are used to construct prostaglandins that accentuate inflammatory responses. The prostaglandins manufactured from omega-3s tend to help weaken such responses, and this is why fish oil is often called "anti-inflammatory"—because it leads to the manufacture of hormones that blunt inflammatory responses.
Pro- and anti-inflammatory hormones are not all good, or all bad. The target here is balance. An immune system that is too depressed will do a poor job of fighting infection and cancer. An immune system that is too active will fan the flames of chronic inflammation, which injures and ages our cells and tissues. The two effects are the Yin and Yang of healthy immune system function.
Modern living tends to give us an excess of inflammatory factors, including those that come from our diets. Paleoanthropologists estimate that in our native diet, the ratio of anti-inflammatory omega-3s to pro-inflammatory omega-6s was somewhere between 1 to 1 and 1 to 4. The ratio in the typical American diet today is between 1 to 11 and 1 to 20. This represents a serious potential excess of pro-inflammatory input, and/or a deficiency of anti-inflammatory omega-3s that help control that fire.
A large and expansive body of literature across many different fields highlights the potentially important effects of omega-3 fats—from fish, seafoods, nuts, seeds, and/or their oils—on inflammatory conditions. I know this in part because I have written two editions of a nutrition textbook, and am now writing the third. Doing so, and reviewing literally thousands of research papers, provides a bird's eye view of the literature that's otherwise quite elusive. From altitude, omega-3s clearly matter.
I also see this from the more intimate perspective of patient care. We are all cautious about anecdotes, but it is not unimportant when a patient with arthritis adds an omega-3 supplement (from fish, krill, flaxseed, or algae) and gets better. My colleagues and I have seen this many times.
What if the true benefit of omega-3 fats to cardiovascular risk is not about treatment, but prevention? What if those anti-inflammatory effects are most important over a span of years to decades by preventing the inflammation that propagates vascular injury and plaque formation in the first place? The meta-analysis in JAMA last week would have been blind to any such effects.
That doesn't mean we can or should dismiss the study. It does indeed indicate that the effects of fish oil supplements, when added to state-of-the-art conventional care for high-risk cardiac patients with established vascular disease, are apt to be modest at best. But going from that to headlines announcing "no cardiac benefit" from omega-3 fats is rather a fish story.
I remain convinced of the health benefits of omega-3 fats across an array of conditions, likely including cardiovascular health. It is established fact that these nutrients are essential, and that the typical American diet provides a relative deficiency of them. It is all but established fact that an imbalance in our dietary fats contributes significantly to inflammation. And it is established fact that inflammation is one of the key processes propagating all chronic diseases—cancer, diabetes, and heart disease alike—as well as many others. I make a concerted effort to include sources of omega-3 in my diet regularly; take a supplement myself; and routinely recommend omega-3 supplements to my patients.
So while the JAMA study is useful, the headlines it has spawned are a bit fishy overall. Digest the information accordingly.
Hungry for more? Write to firstname.lastname@example.org with your questions, concerns, and feedback.
David L. Katz, MD, MPH, FACPM, FACP, is a specialist in internal medicine and preventive medicine, with particular expertise in nutrition, weight management, and chronic-disease prevention. He is the founding director of Yale University's Prevention Research Center, and principal inventor of the NuVal nutrition guidance system. Katz was named editor-in-chief of Childhood Obesity in 2011, and is president-elect of the American College of Lifestyle Medicine.