Omega-3's and heart disease: where's the evidence?

Nov 22 2010 Published by under Medicine

Omega-3 fatty acids (more properly called "n-3 fatty acids") are a group of naturally occurring fat molecules.  They are found mainly in fish and other marine-derived oils, but some can also be extracted from plants.  Omega-3's are currently very popular, but the evidence for their usefulness isn't so clear.   A recent study failed to show any benefit in preventing dementia.   A new study out of the Netherlands looked at omega-3's in heart disease.  In order to understand the study, we need to back up for a moment.

We already know that some dietary fats have a strong effect on heart disease: certain types of cholesterol are elevated in heart disease, and altering the levels of these fats reduces the risk of heart disease.  The treatment of cholesterol problems is a mainstay of heart disease treatment and prevention.   The idea that omega-3-fatty acids may help prevent heart disease is a few decades old.  Like many interesting hypotheses, it was based on observations.  In this case, investigators noted that Greenland Inuit had a much lower rate of heart attack than "Westerners".  Researches posited that it was the high levels of dietary fish oils that protected Inuit from heart disease.

We've already discussed some of the ways scientists evaluate risk and benefit; in this case, much of the evidence for the benefit of fish oil in heart disease is based on observational studies rather than randomized controlled trials.   Many of these studies looked at the dietary choices of subjects to assess their omega-3 consumption.  There is evidence for a role for omega-3's in the primary prevention of heart disease, that is, preventing someone who doesn't have heart disease from getting it, but the evidence isn't terribly strong.  The evidence for secondary prevention is a bit meatier.  One large Italian study gave omega-3 supplements (rather than guessing at dietary consumption) to people who had had a recent heart attack.  The results were very encouraging, with a large reduction in the risk of death from another heart attack.  But---there's always a 'but'--the study was done at a time when many of the effective medications (such as statins) weren't yet in wide use, so it's unclear what effect there would be in a more modern context.

A newer study out of the Netherlands took an interesting approach.  Rather than simply guessing at diet or giving a pill, the researchers developed margarines supplemented with omega-3's.  They gave the margarines to people who already had heart disease and observed them for a period of time.  The doses of omega-3's were relatively low, but when they compared the regular margarine group to the omega-3 margarine group, there were no significant differences in the rates of major heart attacks.

This isn't the final word on omega-3's for heart disease-prevention.  Part of the problem in studying the question is that the current treatment of heart disease is impressively effective.  Any intervention that doesn't have a dramatic effect is going to be hard to evaluate.  Given the epidemiologic evidence, and the questions still out there, this study isn't going to end the question of omega-3's in heart disease.


Kromhout, D., Giltay, E., & Geleijnse, J. (2010). n–3 Fatty Acids and Cardiovascular Events after Myocardial Infarction New England Journal of Medicine, 363 (21), 2015-2026 DOI: 10.1056/NEJMoa1003603

17 responses so far

  • James Sweet says:

    My wife is really big on the fish oil. Her midwife is really bullish on vitamin D these days (in fact I very much appreciated your post on Vitamin D the other month for exactly that reason).

    Both aren't necessarily wrong, but they are way too optimistic about the benefits on what amounts to very scant evidence. I find this sort of thing mildly frustrating, but probably not worth making a big deal over, in person at least (by contrast, if her midwife start pushing homeopathy or decrying vaccinations, I'd demand my wife find a new practitioner posthaste).

    That said, I am so very grateful to see a balanced-yet-skeptical eye turned to these sorts of trendy notions. Any fool with half a brain can see that homeopathy is ridiculous -- but what's an intelligent layperson like myself to make of the health claims surrounding omega-3s? It's prima facie plausible, so I can't reject it on those grounds. The literature is conflicting, and I don't have the time to do an exhaustive literature search on every single controversial topic I might encounter in my daily life.

    I guess what I'm saying is that I'd love to see you take on more of this borderline stuff like vitamin D and omega-3s and such, where it's not outright quackery, but it's possibly being oversold at the same time. Fighting quackery is important too, of course -- but on a personal note, I can usual spot the blatant quacks. It's the that-could-be-but-maybe-it's-too-soon-to-tell-but-I'm-not-really-sure zone that gives me troubles.

    In a roundabout way -- thanks, and keep it up!

  • Isabel says:

    "certain types of cholesterol are elevated in heart disease, and altering the levels of these fats reduces the risk of heart disease."

    I heard that this was debunked, and it was okay to eat eggs/egg yolks again. Do you have a citation? I eat eggs every morning. I'm an egg fiend!

    I also remember studies that strongly indicated that eating nuts had positive effects on long term health, which seems to point to the healthier fats thing. I consider nuts a major food group.

    Also, do you allow your patients to eat coconut oil? I eat it and bathe in it, so I hope it gets your approval.

    Sorry I just finished a grant and am too burnt out to look for citations myself. Maybe in a few days.

    • PalMD says:

      Um, no...cholesterol as a risk for heart disease has not been "debunked". It's always been "OK" to eat eggs---dietary requirements depend on the patient. People with high LDL-C levels and high TGs probably benefit form reducing the intake of animal fats.

      Many nuts have n-3 fatty acids, but as we've seen, these observational studies aren't all that reliable.

    • Vicki says:

      As I understand it, what has been "debunked" was the assumption of an association between dietary cholesterol, and blood serum cholesterol.

      That is, eating more or fewer eggs won't (or usually won't?) affect how much cholesterol is in your bloodstream, which is what matters. Fats, or some kinds of fats, do affect that.

      And some of it is genetic. Both my aunts inherited high cholesterol from their father, and are careful about fat in their diet. My mother and I didn't inherit that, so we don't need to pay as much attention there. I still use whole milk for my tea and hot chocolate, and sometimes drink 2%; my aunts use only skim. (I'm not basing this just on "Mom didn't get it, so I didn't"--I did have a father, after all--I'm basing it on what my doctors tell me my blood tests say.)

      • PalMD says:

        Cholesterol levels are affected by diet, and by genetics.

        • James Sweet says:

          Indeed, my diet is pretty lousy, and I am overweight as a result, and if I don't whip that into shape I can probably expect all kinds of health problems twenty years down the road... my blood pressure is not great... but my cholesterol is just fine. Go figure.

          Then there are people I know whose diet is as bad as mine but they are barely overweight at all -- and some of them have high cholesterol.

          Sigh. We deal with the cards we've been dealt, eh?

      • Isabel says:

        Oh that's right Vicki, that's what I was thinking of. So what are the 'bad' fats that raise cholesterol? Long chain saturated fats and trans fats? And isn't the whole idea not to just add omega-3's, but to increase their relative proportion? I hear a lot about the balance between the omega-6's and omega 3's being out of whack in modern diets for example.

    • D. C. Sessions says:

      One way to moderate your egg consumption is to use an egg/tofu mix. My farm-boy uncle's doc got him to switch from eggs alone to about one part eggs to two parts tofu for breakfast, and with a bit of onion you could hardly tell the difference.

  • Dianne says:

    I've heard it claimed that the reason the Inuit population has a lower rate of heart disease is because they have a high rate of alcoholism and tend to die of cirrhosis before heart disease can manifest. I don't know whether it is true or not, but I mention it to point out the difficulty of comparing two different populations, which may have many differences in behavior and genetics, and stating definitively that their risk of disease X is due to behavior Y.

    Also, assuming the observation is correct at face value (i.e. the Inuit population has a lower heart attack rate due to their diet), is the effect due to eating omega-3 fatty acids or due to not eating high amounts of other fatty acids? If the latter then all pills aren't going to be effective unless one takes enough of them to replace food.

    In short, interesting but I'm sticking with statins and aspirin for now.

  • Kevan Gelling says:

    The suspect "fish = good health" is due to vitamin D rather than omega 3. Many of the supposed health benefits of omega 3 are also benefits of vitamin D. Solid RCT evidence is lacking in both camps, but vitamin D has plausibility whereas omega 3 does not.

    • PalMD says:

      There's actually a good deal of plausibility to the omega-3 hypothesis, but plausibility is necessary but not sufficient.

  • daedalus2u says:

    Some of the Inuit effect might be from being in ketosis all the time. But I think that ketosis is incompatible with pregnancy and lactation.

    Alcohol probably is really bad if you are in ketosis all the time. Even in ketosis you still need gluconeogenesis. Some cells only run on glucose (immune cells and red blood cells mostly).

  • Jeff says:

    Does it make sense to combine Omega-3s with margerine? Margerines contain trans fats, which are thought to promote inflammation and heart disease.

    There is one study which actually compared Omega-3s and Statins:

    Omega-3 Fatty Acids, but Not Statin Therapy, Cuts Mortality and Hospitalizations in Heart Failure

    • Isabel says:

      I thought of that also, but it just seemed too obvious for the researchers to miss. Also, there are lots of products out nowadays that are vegetable oil derived that are solid at room temp and advertise as not trans fat and even 'more omega 3's' though not sure how they manage the trick. Also not sue if they are technically margarine, usually they are just called spreads I think.

  • Bill in NC says:

    Do you have any thoughts on omega-6 fatty acids?

    Are we still supposed to avoid those like the plague?

  • daedalus2u says:

    There is some need for some amount of omega-6 fatty acids. A diet with zero omega-6 FA would not be a good one, and in any case would be very hard to get and would violate much of what is well known about what a good diet is.

    The essential fatty acids are used for structural and signaling compounds. EFA in excess of those needs are oxidized for energy just like non-essential FA. Once you have met the needs for EFA, excess levels likely don't do much.

    Individual needs may be idiosyncratic, but unless you have a genetic metabolic disorder an extreme diet is likely not helpful.

  • Kevan Gelling says:

    Vitamin D is fat-soluble, so does omega-3 obtain from fish contained dissolved vitamin D?