August 23, 2018

A Cochrane Review of Fish oil: a Whopper of a disappointment


The earliest reference I have found to fish oil having a protective effect against cardiovascular disease is Bang and Dyerberg’s 1971 paper. [1]  This perturbs me as I clearly recall my 10th biology grade teacher mentioning this Eskimo and fish business a few years earlier.  Suffice to say the idea that fish oil offers protection against heart disease has been around for about half a century.  Recall in the late 1960s, heart disease was supposedly linked to high animal fat diets.  Yet Eskimos didn’t seem to have the predicted risk.  The explanation was that the high fish oil in their diet was somehow protective.


We’ve swallowed this idea, hook, line and sinker, if you can tolerate the fishing expression.  It’s only recently that this common assumption is being questioned.  Most of us have strenuously made excuses and rationalized why these contrary studies didn’t find cardiovascular benefit to supplementing with fish oil or consuming more fish.


The strongest argument against our fish oil belief was published in mid-July on the Cochrane Database and at this point we’re left with a pretty kettle of fish. [2] Abdelhamid and colleagues performed the most comprehensive meta-analysis we’ve seen to date.


Being a Cochrane Review they did a meticulous job.  The authors searched CENTRAL, MEDLINE and Embase up until April 2017.  Studies were compiled for meta-analysis.  The authors included randomized controlled trials (RCTs) that lasted at least 12 months and compared supplementation and/or advice to increase omega-3 polyunsaturated fatty acids from oily fish [long-chain omega-3 (LCn3) or alpha-linolenic acid (ALA)] intake versus usual or to lower intake.



Their literature search yielded 79 randomized controlled trials (RCTs) with a total of 112,059 participants.


Two review authors independently assessed studies for inclusion, extracted data and assessed validity. The authors performed separate random-effects meta-analysis for ALA and LCn3 interventions and assessed dose-response relationships through meta-regression.


Trials were of 12 to 72 months’ duration and included adults at varying cardiovascular risk, mainly in high-income countries. Most studies assessed LCn3 supplementation with capsules, but some used LCn3- or ALA-rich or enriched foods or dietary advice compared to placebo or usual diet.


The authors did everything right.  We will have a hard time finding fault with their methodology.  They didn’t bother looking at blood markers associated with CVD, instead they looked at bottom line measures that included; all-cause mortality, cardiovascular mortality, cardiovascular events, arrhythmia, stroke, bleeding, and coronary heart disease.  These are what we really care about when it comes right down to it.


The conclusion of this large meta-analysis was that increasing fish oil consumption either as oil supplements or by eating more fish had no effect all-cause mortality, cardiovascular mortality, cardiovascular events, coronary heart disease (CHD) mortality, stroke or arrhythmia.

Increasing ALA didn’t have much more benefit. It was not associated with any significant change in all-cause mortality or congestive heart disease events.  Increasing ALA did however, was associated with a non significant reduction in cardiovascular events from 4.8% to 4.7%[RR 0.95, 95% CI 0.83 to 1.07, 19,327 participants; 884 CVD events, 5 RCTs, low-quality evidence), and probably reduces risk of CHD mortality (1.1% to 1.0%, RR 0.95, 95% CI 0.72 to 1.26, 18,353 participants].  Mind you these benefits were not statistically significant.  Not even close, but I’m trying to lessen the shock of these outcomes.

Screen Shot 2018-08-26 at 9.10.35 AM

The only statistically significant benefit to show up in the data was for congestive heart disease.  Fish oil was associated with a 7% drop in risk.  Well it was until the researchers performed a sensitivity analysis and the benefit disappeared.  Still it’s better than nothing.


These data suggest some potential of risks of morbidity from things we had assumed fish oil would lower risk or.  One of those conditions is stroke, that there was a slight uptick in stroke in the fish oil groups. This uptick in strokes was not statistically significant and it was super small.  Still that should give us pause as we were considering multiple numbers that were not statistically significant. It kind of works both ways. This does go contrary to the common belief that fish oil is an anticoagulant and so decreases the risk of ischemic strokes. Bleeding also is a potential risk of fish oil and we saw that there was no difference in either group with increased risk of bleeding We aren’t very certain of this, but this again argues that fish oil probably isn’t an adequate anticoagulant to prevent strokes and this is further evidenced by the fact that it doesn’t increase bleeds. If something is going to prevent strokes because it’s an anticoagulant, it should also increase bleeds because that’s one of the downsides of anticoagulant therapy.

There are many of us who will be unhappy with these results.  We have been advising patients to take fish oil for so long it will be hard to change our thinking.

Screen Shot 2018-08-26 at 9.13.13 AM

This isn’t the first large study that has called the fish ‘myth’ into question.


In September 2012 JAMA published as systematic review by Rizos et al that also assessed the effect of fish oil supplementation on cardiovascular events. [3]  Data from twenty studies that included 68,680 patients were analyzed. Their conclusions were similar to what we see in the current Cochrane Review: “Overall, omega-3 PUFA supplementation was not associated with a lower risk of all-cause mortality, cardiac death, sudden death, myocardial infarction, or stroke based on relative and absolute measures of association.”


A Korean meta-analysis also published in 2012 failed to find any benefit of fish oil supplementation in people with a history of heart disease.  This study analyzed data from 14 randomized, double-blind, placebo-controlled trials involving 20,485 patients with a CVD history. [4]



In 2013 NEJM published the results of a clinical trial giving fish oil to patients with multiple CVD risk factors.  Patients had multiple CVD risk factors but had not had a myocardial infarct yet. 12,513 patients enrolled, 6244 were assigned to n-3 fatty acids and 6269 to placebo, which we should mention was olive oil. After a median time of 5 years 11.7% of those taking fish oil and 11.9% of those receiving placebo had died or had a non-fatal MI or stroke.  There is no significant difference between these two rates. [5]


Screen Shot 2018-08-26 at 9.20.22 AM

A 2014 paper published in the Canadian Journal of Cardiology tokk a second look at that original Bang and Dyerberg paper that started this whole fish thing off. They reviewed the literature and reported that risk of cardiac arterial diseases among both the Greenland Eskimo and Alaskan Inuit never did differ from other populations and that the original basis for the “Eskimo diet” never existed. Bang and Dyerberg never studied native populations directly; they used death reports and hospital admission figures to make their calculations.  It turns out that CVD is common in the Inuit and possibly even higher than expected; curiously rates may decrease as people are ‘Westernized.” [6] [7] [8]

There is a new trial scheduled to be published at the end of this year. The trial ended last year. This trial is a large trial with 25,000 participants and it has examined the effect of vitamin D and Omega-3s on the primary prevention of cardiovascular disease over a five-year period. This is the largest study to ever be done on this topic. It looks at long term supplementation of fish oil and is robustly designed.[9]  It may overturn other things we thought we know about fish oil and cardiovascular disease. It will have the statistical power to question these current Cochrane Review findings, that is if they disagree.



How could we be so far off with this?  Science isn’t perfect and the idea that dietary interventions might have a large impact on disease prevention is attractive to practitioners, manufacturers and consumers alike. The idea took on a life of its own.   Let’s return to Greenland for a moment.  A paper published in 1992 offered an alternative explanation for why Eskimos have such good lipid profiles.  Eskimos have significantly lower lipoprotein(a) [Lp(a)] and this is apparently do to their genetics rather than diet. [10]  We were too caught up with the fish business to notice this and other early reports.



Note parts of this article are from an online lecture given by Drs. Andrew Day and Joshua Goldenberg as part of their Doctors Journal Club.


white kayak at cherry creek


[1] Bang HO, Dyerberg J, Nielsen AB. Plasma lipid and lipoprotein pattern in Greenlandic West-coast Eskimos. Lancet. 1971 Jun 5;1(7710):1143-5.

[2] Abdelhamid AS, Brown TJ, Brainard JS, et al. Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease.  Cochrane Database Syst Rev. 2018 Jul 18;7:CD003177.


Rizos EC, MD,  Ntzani EE,  Bika E et al.  Association Between Omega-3 Fatty Acid Supplementation and Risk of Major Cardiovascular Disease Events A Systematic Review and Meta-analysis.  JAMA. 2012;308(10):1024-1033. doi:10.1001/2012.jama.11374

[4] Kwak SM1, Myung SK, Lee YJ, et al. Efficacy of omega-3 fatty acid supplements (eicosapentaenoic acid and docosahexaenoic acid) in the secondary prevention of cardiovascular disease: a meta-analysis of randomized, double-blind, placebo-controlled trials. Arch Intern Med. 2012 May 14;172(9):686-94. doi: 10.1001/archinternmed.2012.262.

[5] Risk and Prevention Study Collaborative Group, Roncaglioni MC, Tombesi M, Avanzini F, et al. n-3 fatty acids in patients with multiple cardiovascular risk factors.  N Engl J Med. 2013 May 9;368(19):1800-8.

[6] Howard BV, Comuzzie A, Devereux RB. Cardiovascular disease prevalence and its relation to risk factors in Alaska Eskimos.  Nutr Metab Cardiovasc Dis. 2010 Jun;20(5):350-8.

[7] Bjerregaard P1, Young TK, Hegele RA. Low incidence of cardiovascular disease among the Inuit–what is the evidence?  Atherosclerosis. 2003 Feb;166(2):351-7.

[8] Fodor JG, Helis E, Yazdekhasti N, Vohnout B. “Fishing” for the origins of the “Eskimos and heart disease” story: facts or wishful thinking? Can J Cardiol. 2014 Aug;30(8):864-8.

[9]  Vitamin D and Omega-3 Trial (VITAL) Identifier: NCT01169259

[10] Klausen IC, Gerdes LU, Schmidt EB, Dyerberg J, Faergeman O. Differences in apolipoprotein (a) polymorphism in west Greenland Eskimos and Caucasian Danes. Hum Genet. 1992 Jun;89(4):384-8.