September 30, 2009
Dutch Study Finds No Fish Heart Benefit

A prospective study does not find a heart benefit from fish.

With heart failure treatments often limited to palliative care, much rests on prevention; this latest report from the Rotterdam Study was to investigate whether intake of the long-chain n-3 polyunsaturated fatty acids (PUFAs) found in fish conferred protection against heart failure as they seem to do against coronary heart disease.(3)

The analysis comprised 5299 subjects (41% men, mean age 67.5 years) who were free from heart failure and for whom dietary data were available. During 11.4 years of follow-up, 669 subjects developed heart failure. Their habitual diet had been assessed at baseline (in a self-reported checklist and by expert interview), with subjects specifically asked to indicate the frequency, amount, and kind of fish they had eaten, either as a hot meal, on a sandwich, or between meals.

Results showed that the dietary intake of fish was not significantly related to heart failure incidence. This relative risk was measured according to five levels of fish consumption as reflected in intake of two long chain n-3 PUFAs (eicosapentaenoic acid [EPA] and docosahexaenoic acid [DHA]), both of which have been shown to exert some cardiovascular benefit via anti-inflammatory mechanisms, anti-arrhythmic effects and/or a reduction in serum triglycerides, blood pressure, and heart rate.

I am pretty confident that we should eat lots of fruits and vegetables and when eating grains they should be whole grain and high in fiber. But just how much omega 3 fatty acids to consume?

One of the problems is that there's genetic variability in determining what's the ideal diet. We need a level of understanding of nutritional genomics that does not exist yet. Though with the big drops in DNA sequencing costs I'm hopeful genetic alleles for dietary guidance will become known in a few years.

Update: Note that the problem with the Dutch study might be a dosing problem. Not enough people in the sample might have eaten enough fish for enough time to deliver a benefit. This study does contradict a lot of other studies. You can't conclude from just this study that eating fish doesn't help.

Also, omega 3 fatty acid concentrations vary greatly between species of fish. For example, you are going to get a lot more DHA and EPA from salmon than from tuna. Take a lot at the table at the bottom of this page for omega 3 in various fish. Salmon, anchovy, and sardines have the most omega 3 fats.

Share |      Randall Parker, 2009 September 30 11:08 PM  Aging Diet Heart Studies

David R. French MD said at October 1, 2009 7:27 AM:

Interesting study, however, I think the important thing to keep in mind is that heart failure cause by coronary artery disease only represents a little under half of all heart failure cases. Isolated hypertension, valvular heart disease, cardiomyopathy (alcoholic, viral, idiopathic, etc), and COPD are other potential causes. I would not expect omega 3 supplementation to alter the clinical course of these other diseases.

It would be interesting if they would, retrospectively, try to look at the patients diagnosed with heart failure who also carry the diagnosis of CAD and if, in these patients, the occurence of heart failure was altered.

In the end, this study should not alter anyone's views toward the well documented health benefits of omega 3's.

David Govett said at October 1, 2009 8:18 AM:

Study 1 proves A. Study 2 proves not-A.
This means that our understanding of the fantastic complexity of the human organism is primitive, at best.
It is a system of systems, a hierarchy of hierarchies, each level of which is currently--but only currently--beyond our understanding.
In about 20 years, a lot of hands will slap foreheads, as life scientists exclaim, "How could we have been so stupid!"
Not stupid--ignorant. Human, in fact.

James Bowery said at October 1, 2009 10:32 AM:

Its racist to study consequential genetic variation in humans and its all so complicated. A little knowledge is the first step to another Holocaust. And, besides, no one in my family ever had an dietary problems so you barbarian pastoralists can all just go off and die the deaths you deserve for being so ill adapted to civilization. Your time is past.

notthefatone said at October 1, 2009 2:28 PM:

I'g study. It seems from a quick read however that the studies absence of findings does not indicate a non correlation. how much fish was consumed, what types, etc. once those variables are accounted for, how valid is the sample size. further complicating this is the connection between coronary heart disease and failure.

there is heart failure unconnected to coronary buildup (I believe, Randall, please let me know if this is incorrect, if you know, or can find anything on it - it has huge consequences either way.)That is, much heart failure is unexplained. if the study attempted to isolate out just heart failure, as opposed to documented coronary buildup and or disease,this would be a different question than whether fish was overall "good for heart health."

that is, fish would still be good for health (or could be) even if this studies finding were duplicated many times.

IF the study included coronary buildup and disease (which I'm not sure is the case) AND all heart failure is a result of this (which I don' think is the case, just look at thoseo who die from bad valves alone) then perhaps if the study was duplicated many many tims it would raise more questions. but otherwise, even if these two factors laid out are both answered "yes" the study still does not seem to necessarily (yet) undermine what we think we are coming to learn

SteveSC said at October 1, 2009 3:02 PM:

A key comment from the researchers in the article cited is:

"Fish intake in the Netherlands is extremely low - on average less than one portion per week - so maybe higher intakes are needed for any protection against heart failure."

The high quintile of fish consumption was only 20 grams per day, which is less than one ounce. If it was all a relatively good source of omega-3s, such as salmon, this would provide about 300 mg of DHA and EPA per day, about the equivalent of one 1000 mg fish oil supplement capsule. If, as is more likely, it was a typically preferred 'white fish' such as flounder, this would amount to only 40 mg of DHA and EPA per day. No wonder there wasn't a big effect.

And yet there WAS an effect, although it was not statistically significant. The article states that the relative risk dropped 11% overall, 25% for women, and an astounding 42% for diabetics. Now the uncertainty was large, which is why there was no statistical significance, but this study hardly finds 'no benefit from fish'. A more scientific headline would be: Small amounts of fish may prevent heart failure: bigger studies needed.

Carl Pham said at October 1, 2009 3:06 PM:

As they say, absence of evidence is not evidence of absence. I'd keep eating the fish.

Michael G.R. said at October 1, 2009 5:00 PM:

SteveSC makes a good point.

health nut said at October 1, 2009 6:46 PM:

Chris Masterjohn has a pay report about omega 3 dosing. from the abstract:
"Current reviews and textbooks call the omega-6 linoleic acid and the omega-3 alpha-linolenic acid “essential fatty acids” (EFA) and cite the EFA requirement as one to four percent of calories. Research suggests, however, that the omega-6 arachidonic acid (AA) and the omega-3 docosahexaenoic acid (DHA) are the only fatty acids that are truly essential. Eicosapentaenoic acid (EPA) occurs in fish products but is probably not a normal constituent
of the mammalian body, and in excess it interferes with essential AA metabolism. The EFA requirement is inflated in the scientific literature by several factors: the use of diets composed mostly of sucrose, glucose, or corn syrup; the use of diets deficient in vitamin B6; the use of purified fatty acids instead of whole foods; the use of questionable biochemical markers rather than verifiable symptoms as an index for EFA deficiency; and the generalization from studies using young, growing animals to adults. The true requirement for EFA during growth and development is less than 0.5 percent of calories when supplied by most animal fats and less than 0.12 percent of calories when supplied by liver."

Ray Peat also has some articles warning against excessive fish oil dosing.

bbm said at October 1, 2009 7:30 PM:

Heart "failure" seem like a poor endpoint to measure as well. MI and/or stroke would probably be a better endpoint.

Bob said at October 1, 2009 7:38 PM:

When I was in holland I saw many street booths selling pickled herring, which is a fatty fish.

Maybe it is a novelty and not popular with the average dutchman.

Bob said at October 1, 2009 7:39 PM:

When I was in holland I saw many street booths selling pickled herring, which is a fatty fish.

Maybe it is a novelty and not popular with the average dutchman.

coldequation said at October 2, 2009 2:30 AM:

The real lesson here might be that you're better off getting your omega 3s from fish oil supplements than actual fish, since it's hard to eat enough fish (those fatty fish like salmon aren't cheap). Steve in a comment above bounds the amount of omega 3s that the top quintile of fish eaters in the study consumed between 40 and 300 mg. Most research has been done with more like 4000 mg, which is the daily dose of prescription fish oil (Lovaza).

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©