From reading reports in the news media, one might assume it is a given that fish consumption promotes health. In reality, the results from observational studies are conflicting and the available evidence does not support such a claim. The whole area remains controversial.
Before discussing the individual issues, we should mention some caveats that should be remembered when one is interpreting the results of new studies.
First, one must be aware of the "substitution effect." For example, if fish is substituted for beef in the diet and health benefits are observed, is the result due to the consumption of fish or to the elimination of beef? A number of the early claims ascribed to fish oil, such as the lowering of cholesterol concentrations in the blood or the lowering of blood pressure, were later disproven when better studies came along that showed that fish consumption per se was not responsible for the observed findings.
In 1996, results of a World Health Organization-supported study of Tanzanian Bantu villagers appeared in the respected British journal The Lancet. One group of villagers was said to consume a diet consisting almost exclusively of plant foods. The other group, which lived by a lake, consumed fish as well. The cholesterol concentrations and the blood pressures of the fish-eating group were lower. The authors theorized that fish oils were responsible for these beneficial effects and that a fish-containing diet is, therefore, superior to a vegetarian diet. These results were picked up by a variety of journals and nutrition newsletters, which trumpeted recommendations to increase fish consumption.
When I first read this report, it seemed familiar. I went through my files and found a previous report from this study that had been published in a journal with much more limited circulation, Advances in Prostaglandin, Thromboxane, and Leukotriene Research. I found that, contrary to the impression given in The Lancet article, the plant food portions of the diets of these two groups of villagers were actually quite different. The fish-eating group was said to have derived almost half its calories from a cassava root preparation, similar to what we know as tapioca. It turns out that there is experimental evidence that cassava root, by virtue of its fiber content, can lower cholesterol in the blood and, by virtue of a neurotoxin it contains called scopoletin, can lower blood pressure! My response, which I felt refuted the conclusions of The Lancet article, was published in a later issue.1
It is instructive with regard to the politics of the nutrition establishment that, after my letter appeared in The Lancet, the groups that reported initially on the results of this study did not offer any modifications. The desire to extol the consumption of fish (and other animal products) and to deprecate vegetarian diets appears to be fairly well ingrained.
A second important caveat is that fish consumption may be a marker for a more healthful lifestyle. In other words, people who consume more fish may tend to follow a lifestyle that in other ways is more conducive to health. Studies have shown, for example, that people who consume above average amounts of fish also tend to consume above average amounts of fruits and vegetables. Therefore, if we are analyzing a study that evaluated the effect of fish consumption on a particular aspect of health, we must be sure that the study investigators controlled for fruit and vegetable consumption when they presented the results and drew their conclusions. Clearly, fish consumption may also be associated with other lifestyle practices that investigators have not taken into account.
With that in mind, let us now examine some of the issues. Does fish consumption help protect against heart disease? Certainly there are theoretical reasons why such may be the case. Fish oils are rich in the long-chain omega-3 fatty acids eiscosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). These fatty acids inhibit the action of platelets, the tiny cellular fragments in the blood that aggregrate to initiate the process of blood clot formation. Platelet inhibition can reduce the risk of heart attack or stroke in people at risk; that is why aspirin, an inhibitor of platelet aggregation, is frequently taken by people who have one or more major risk factors for cardiovascular disease. On the other hand, following a healthful, high-fiber, low-fat diet reduces the tendency of the blood to coagulate and represents one of the means by which such diets exert their protective effects.
Fish oil consumption also reduces the viscosity of blood, which roughly corresponds to its "thickness." A higher blood viscosity has been associated with the risk of heart disease. But an appropriate lifestyle incorporating both exercise and a healthful, plant-based diet also reduces blood viscosity.
Fish oils reduce endothelial dysfunction, that is, the inability of blood vessels to dilate when more circulation is needed to a given area. But endothelial dysfunction is also reduced by reducing the fat content of the diet, including trans fatty acids, and by increasing antioxidant intake.
The omega-3 fatty acids found in fish oils reduce concentrations of triglycerides in the blood. These are fats that increase the risk of cardiovascular disease. However, a healthful diet and exercise help keep levels down where they should be.
Still another important property of fish oils is their antiarrhythmic effect, that is, their ability to prevent abnormal heart rhythms that can lead to sudden death.
Fish oils have anti-inflammatory properties. They influence the balance of eicosanoids in the body. Eicosanoids, such as prostaglandins, are synthesized from fatty acids and can have either proinflammatory or anti-inflammatory effects, depending on their chemical structure. This is a very important property, given the growing body of evidence that an inflammatory tendency in the body increases the risk of heart and other chronic diseases. In many people, this inflammatory tendency is due to the intake of too many calories. Intake of arachidonic acid, an omega-6 fatty acid that comes mainly from animal products, may also be a contributing factor. Arachidonic acid leads to the formation of a proinflammatory series of eisosanoid, but that result can be nullified by increasing the intake of omega-3 fatty acids such as those found in fish oils.
Despite all of these theoretical advantages of fish consumption, a balanced review of the medical literature indicates that there is no good evidence that fish consumption reduces the risk of heart disease. In fact, some clinical trials have even shown an increase in the mortality rate when more fish and/or fish oil is consumed! The evidence is well summarized by Dr. Peter Markmann in an editorial he wrote for The American Journal of Clinical Nutrition in 2003.2 It appears that some of the properties of fish oils may be beneficial whereas others may be harmful. For example, not only does fish oil not lower cholesterol levels, it appears to increase the susceptibility of LDL-cholesterol (the "bad" variety) to oxidation. The ability of LDL-cholesterol to cause atherosclerosis (hardening of the arteries) is enhanced when it becomes oxidized. Therefore, fish oil may actually accelerate the process of atherosclerosis.
Prevention of arrhythmias (heartbeat irregularities) that can lead to sudden death is one area in which fish oils may truly help people who already have heart disease. The Physicians' Health Study,3 a major prospective study of over 20,000 male physicians, showed that, whereas fish consumption did not reduce the risk of myocardial infarction (heart attack), it did reduce the risk of sudden death. Sudden death is generally caused by a conduction abnormality in the heart, that is, an arrhythmia.
Recently, some studies have cast doubt on the ability of fish oils to prevent arrhythmias. A prospective study in Denmark did not show any protective effect from fish oil intake with regard to the incidence of atrial flutter or fibrillation, two common arrythmias involving the atria (upper chambers) of the heart.4 A Dutch clinical trial showed that the number of premature ventricular contractions (PVCs) in patients with frequent PVCs was not reduced in those who received fish oils.5 Nevertheless, the heart rate was slightly slowed in those patients, an effect on the conduction system of the heart that might predict a lower risk of sudden death.
From my review of all of the studies published to date, my feeling is that the omega-3 fatty acids found in fish oils do indeed reduce the risk of the arrhythmias leading to sudden death in people with heart disease. But there are other dietary components that also have antiarrhythmic properties. The very important mineral magnesium appears to have such an effect with regard to both the atria (upper chambers) and the ventricles (lower chambers) of the heart. The richest sources of magnesium are whole grains, legumes, dark green vegetables, and nuts and seeds. Whether increased intake of the mineral magnesium or of the shorter omega-3 fatty acids derived from plant foods would also decrease the risk of sudden death in such patients remains to be determined.
Some studies have shown that fish consumption may have a protective effect with respect to a number of other diseases, including prostate cancer, Alzheimer's disease, stroke, and age-related macular degeneration. However, the results have been mixed and the caveats discussed at the beginning of this page need to be considered. The effects of the omega-3 fatty acids in fish oil on the production of prostaglandins and other eiscosanoids, the fat-derived compounds mentioned earlier than play a role in inflammation and possibly even the growth of tumors, may indeed provide some limited benefits to those who fail to follow a healthful lifestyle.
The negatives associated with fish consumption must also be considered. First and foremost, fish consumed today are frequently contaminated with the pollutants that have invaded our seas, rivers, and lakes. PCBs, mercury, and other harmful substances are frequently found. Even farmed fish are contaminated with potentially hazardous chemicals. Recommendations to pregnant women to reduce their fish consumption should perhaps be considered by everyone.
Fish, of course, is a high-protein food, but, as discussed on the page about meat, protein is really not an issue at all, because more than enough protein can easily be obtained from plant foods. A diet very high in protein, in fact, can be harmful by increasing the risk of kidney stones6 and probably osteoporosis as well.7 Fish, like all flesh foods, is high in purines, high intake of which increases the risk of gout.
Fish contains approximately the same amount cholesterol as beef and chicken, and about 20% of the cholesterol in our blood is derived from the cholesterol in our foods. That is why fish consumption does not lower total cholesterol concentrations as some erroneously believe. However, the low saturated fat content of fish as compared with beef and fowl means that fish will not raise cholesterol levels as much as the other forms of animal flesh.
In summary, the results from studies published to date regarding the potential benefits of fish consumption are frequently conflicting, and evidence of benefit to the general population is insufficient to warrant inclusion of fish as an obligatory component of an optimal diet.
1Lavine JB. Blood pressure and cholesterol in fish-eaters vs. vegetarians [letter]. Lancet 1996;348:1460.
2Marckmann P. Fishing for heart protection. Am J Clin Nutr 2003;78:1-2.
3Albert CM, Hennekens CH, O'Donnell CJ, et al. Fish consumption and risk of sudden cardiac death. JAMA 1998;279:23-28.
4 Frost L, Vestergaard P. n-3 Fatty acids consumed from fish and risk of atrial fibrillation or flutter: the Danish Diet, Cancer, and Health Study. Am J Clin Nutr 2005;81:50-54.
5Geelen A, Brouwer IA, Schouten EG, Maan AC, Katan MB, Zock PL. Effects of n-3 fatty acids from fish on premature ventricular complexes and heart rate in humans. Am J Clin Nutr 2005;81:416-420.
6Curhan GC, Willett WC, Rimm EB, Stampfer MJ. A prospective study of dietary calcium and other nutrients and the risk of symptomatic kidney stones. N Engl J Med 1993;328:833-838.
7Meyer HE, Pedersen JI, Loken EB, Tverdal A. Dietary factors and the incidence of hip fracture in middle-aged Norwegians. Am J Epidemiol 1997;145:117-123.