At least one oily fish, salmon, is in most lists of superfoods. Fish oil and cod liver oil comprise over 40% of the value of dietary supplement sales in the UK. So what nutritional characteristics of fish are responsible for its very positive health image?
What prompted this post?
Two separate sets of media headlines in the last few weeks prompted me to produce an overview of current evidence about the health benefits of eating fish or taking fish oil supplements.
The first set of headlines appeared in mid-July 2018 and these commented upon evidence from a high quality systematic review and meta-analysis of studies that had tested the effects of fish oil supplements on risk of heart disease in controlled trials. This study found that, contrary to recent popular and medical folklore, there were no measurable benefits of fish oils for heart health:
Fish oil supplements for a healthy heart ‘nonsense’ BBC news web-site 18/07/2018
The second set of headlines referred to efforts by Scottish scientists to raise the omega-3 fatty acid content of farmed salmon by using genetically modified (GM) fish food. The omega-3 content of farmed salmon, whilst still high, is much lower than that of wild salmon and has halved in the last ten years:
Trial to test if GM fed salmon are more nutritious BBC news web-site 01/08/2018
Efforts to increase the omega-3 content of farmed salmon is only practically useful if the claimed beneficial effects of these fatty acids are real, otherwise it is simply a measure to boost the marketing of farmed salmon. There are other examples of food producers trying to modify the content of foods where there is no hard evidence that such changes will benefit health:
- Increasing the antioxidant or other phytochemical content of plants or foods
- Increasing the protein content or quality of cereals like wheat
- Producing food with enhanced levels of essential nutrients for populations without a deficiency problem
- Producing gluten-free products (even shampoo!!) for the mass market rather than specifically for those with coeliac disease.
Nutritional properties of fish
From a nutritional viewpoint, fish can be divided into two broad categories:
- Oily fish like salmon, mackerel, sardines, trout and herring
- White fish like cod, haddock, hake and pollock.
The oily fish have oil in their flesh that is rich in omega-3 fatty acids, especially two long chain acids referred to as EPA and DHA which are considered to be the “active components” of fish oil. Most vegetable oils contain omega-6 fatty acids. The only difference is that omega-3 have an extra double bond between carbon atoms 3 and 4. Humans are unable to insert or remove double bonds between carbons 3 and 4 which means that omega-3 and omega-6 fatty acids cannot be inter-converted. Even in those vegetable oils that do contain good amounts of omega-3, like flaxseed oil and rapeseed oil, they are short chain versions which are not readily converted into EPA and DHA by people eating modern western diets. Fish is the only major dietary source of EPA and DHA although some algal supplements are produced as a vegetarian source of long chain omega-3s.
White fish have very lean flesh and can thus accurately be described as low-fat, low-calorie foods which are good sources of high quality protein. As noted in an earlier post, lack of protein is an unlikely problem in human nutrition and so “high protein” has limited kudos from a nutritionist’s viewpoint. It is still the cultural norm in the UK to have a “protein food” (e.g. meat, fish, eggs, cheese or a vegetarian meat substitute) as the focus of main meals. White fish is a healthy, low-fat, low-calorie option as one of these “protein foods”. White fish that is battered and deep-fried is no longer low calorie or low fat!
White fish store omega-3 rich oil and fat soluble vitamins like A and D in their livers, hence cod liver oil. Cod liver oil has been used as a medicine/supplement for over a century. It is a good source of vitamins A and D as well as containing omega-3 fatty acids. “Fish oil” supplements contain insignificant amounts of these vitamins.
What major claims have been made for fish oils and cod liver oil?
It has been claimed that taking fish oils or eating oily fish can reduce the risk of heart attack survivors dying of a further heart attack in the years following the first event. It is also suggested that regular consumption of oily fish or fish oil supplements may reduce the risk of a first heart attack in healthy middle-aged and elderly adults.
Fish oil has been said to reduce the symptoms and perhaps slow the progression of arthritis by a dampening effect upon inflammation. Those few studies that have investigated this possibility have tended to be small studies with patients suffering from rheumatoid arthritis. Nevertheless, millions of people take fish oil or cod liver oil in the hope that it might be beneficial for the relatively less severe, but much more common, osteoarthritis. Hopes that fish oil might have an anti-inflammatory effect has also led to speculative suggestions that it might be beneficial in the management of asthma, eczema and even inflammatory conditions of the bowel like ulcerative colitis and Crohn’s disease.
DHA is a normal component of brain and retinal tissue that is essential for optimal brain functioning. This has led to suggestions that a relative lack of DHA in brain tissue may lead to abnormal functioning. This has been extended to claims that oily fish or fish oil supplements might lead to improvements in conditions like depression and postnatal depression and that they might improve intellectual development and reduce behavioural problems in babies and children. Similarly there has been speculation that fish oil might slow cognitive decline in healthy elderly people or perhaps improve cognitive function in people with early dementia.
How fish oil MIGHT work
There are several “clever” theoretical mechanisms that could explain some of the proposed health benefits of fish oils. Scientists love such intellectually satisfying and elaborate theories but, in practice, they often turn out to be incorrect or fail to lead to the predicted health benefits.
EPA and its omega-6 equivalent, arachidonic acid, can both be converted into a range of substances collectively called eicosanoids that play a major part in regulating many important physiological processes including inflammation, blood platelet aggregation, blood clotting and the constriction and dilation of blood vessels. These eicosanoids include prostaglandins, thromboxanes and prostacyclins. They are called eicosanoids because they have twenty carbon atoms just like EPA and arachidonic acid; the prefix eico is from the Greek prefix for twenty icosa. This provides potential mechanisms for how fish oils might impact upon several physiological and disease processes. Eicosanoids produced from EPA (omega-3) and arachidonic acid (omega-6) have different potencies. Modern western diets are generally rich in omega-6 fatty acids and this favours the production of omega-6 eicosanoids and suppresses the formation of omega-3 varieties. Taking fish oil supplements would increase availability of EPA and thus alter the balance of eicosanoid production i.e. more omega-3 and less omega-6. There are theoretical mechanisms to explain how this might have a dampening effect upon inflammatory responses and reduce the tendency of blood platelets to aggregate and clots to form. Such changes could help explain proposed beneficial effects of fish oils upon inflammatory conditions and heart attack risk.
Increased fish oil consumption has apparently favourable effects upon blood lipid levels; it lowers blood triglyceride levels and raises the level of “good cholesterol” or HDL in the blood. Fish oil effects on “bad cholesterol” or LDL are not considered significant because of the relatively small amounts normally consumed. Such changes might be expected to have favourable effects upon the risk of a heart attack. Some high dose fish oil preparations are licensed for use in patients with pathologically high triglyceride levels and they lower these levels substantially. However, it is not clear that they also reduce the pancreatitis that such high triglyceride levels cause. Whether the smaller doses of fish oil in ordinary supplements or from eating oily fish produce useful changes in blood lipids is less clear.
Omega-3 and omega-6 fatty acids are collectively termed the essential fatty acids. Like vitamins, they are essential in the diet in small amounts and an omega-6 fatty acid (linoleic acid) was once called vitamin F. It is not easy to produce “vitamin F” deficiency even in volunteers because the body has substantial stores, the minimum requirement is small and it is difficult to eliminate all essential fatty acids even from an experimental human diet. Nutritionists generally consider that omega-3 fatty acids are specifically essential but this is very difficult to prove because the minimum requirements are so small. Remember that we cannot make omega-3 fatty acids from dietary omega-6 fatty acids. The long chain omega-3 fatty acid, DHA is a normal component of the membranes of brain and retinal cells and considered essential for normal brain functioning. Small amounts of omega-3 fatty acids are also considered essential for normal brain development in babies and young mammals. Modern diets would be expected to reduce the body’s production of DHA from short chain omega-3 fatty acids present in vegetable foods so a relative lack of DHA might alter brain cell functioning and brain development. Such arguments can be used to suggest that fish oils, rich in long chain omega-3 fatty acids might favourably affect brain functioning and development.
What is the evidence that fish oils work?
I am going to largely confine this discussion to a review of high quality controlled trials (RCT) and meta-analyses of these i.e. evidence at the very top of the evidence hierarchy or pyramid . However, it is worth looking to the origins of the current research focus upon the potential health benefits of eating oily fish or taking fish oil supplements. In the 1970s, heart disease mortality tended to be high in those populations consuming large amounts of saturated animal fats. However, during this decade Danish scientists Hans Bang and Jorn Dyerberg reported that traditional Greenland Eskimos (Inuit) had relatively low levels of heart disease despite eating large amounts of animal-derived fats; they were also said to have low rates of arthritis. Most of the fat consumed by these Inuit was from omega-3 rich marine sources like fish, seals and whales. This triggered an ongoing surge in research interest in fish oils and omega-3 fatty acids. In the early 1970s there were around 10 research papers per year about fish oil but this rose to several hundred per year in the early 1990s and a PubMed search of the term “fish oil” yielded over 1700 hits per year in each of the last few years. Elizabeth Preston has written an article reviewing evidence which claims that Dyerberg and Bang used flawed underestimates of the rates of heart disease amongst the Inuit. If, as she suggests, heart disease rates amongst the Inuit was much higher than assumed by Bang and Dyerberg then decades of work in this major research area may have been triggered by flawed data!
Two early controlled trials (1 and 2 ) reported that increased oily fish consumption or fish oil supplements reduced mortality in the period after a first heart attack. In a 1994 Department of Health report, authorities in the UK recommended doubling intakes of long chain n-3 fatty acids by increased consumption of oily fish. Similar advice was also offered by the American Heart Association. Some systematic reviews of early trials also suggested that fish oil omega-3s but not flaxseed oil might have benefits in preventing heart disease deaths, particularly in patients who had already had a heart attack (secondary prevention trials). This initial optimism about the potential benefits of fish oil supplements in reducing heart disease and cardiovascular mortality has not been sustained. A Cochrane review and meta-analysis in 2004 found no clear evidence that omega-3 fatty acid supplements from fish or plant oil reduced either total mortality or cardiovascular events even in high risk subjects. A more recent meta-analysis suggested similar lack of benefits but did suggest that long chain omega-3 fatty acids (fish oil) might have some benefit for high risk populations. A very recent and very extensive Cochrane review that gave rise to the 1st BBC headline listed above was even more negative in its conclusions about the benefits of fish oils or alpha-linoleic acid (ALA), a short chain omega-3 from plant sources. They reviewed 79 RCTs with over 112,000 participants in which intake of these omega-3 fatty acids had been increased, usually by using supplements. They found no evidence that increasing either fish oil or ALA had any beneficial effects upon total mortality, cardiovascular mortality or the number of heart attacks or strokes. The conclusion from recent careful systematic reviews and meta-analyses is thus that there is no justification for recommending fish oils for the prevention of heart disease or total mortality even in patients with existing heart disease.
Cochrane reviews are highly regarded for their scientific rigor and the latest of these found no substantial evidence to suggest that fish oils improve asthma , atopic eczema or cognitive decline (memory loss) in healthy elderly people . Available evidence for any benefits of fish oil for adult depression is low quality and any possible benefits are small and not considered clinically significant. There is no evidence of any benefit for postnatal depression. Maternal supplementation with fish oils during pregnancy and/or lactation does not have any measurable effects upon the child’s risk of allergic diseases. In an ongoing RCT, large doses of fish oil did not reduce postnatal depression and have not improved cognitive and language development in the babies when given during pregnancy. There is limited evidence that very high doses of fish oil may have some benefits for pain management in the relatively uncommon rheumatoid form of arthritis but there is very little quality evidence either way about any benefits of more moderate doses for osteoarthritis.
A discussion article published in the Journal of the American Medical Association, Internal Medicine gives an interesting perspective on fish oil research. The authors report that almost all clinical trials and meta-analyses involving fish oil supplements and published in top journals reported no beneficial effect (2007-2012) yet over this period sales of fish oils in the USA, UK and Australia had more than doubled. These negative articles were widely reported in new reports and many editorials accompanying these largely negative studies tended to be enthusiastic about omega-3 fatty acids. Does this mean that even the experts are reluctant to accept evidence that fish oils have little or no benefit for anything? The climate for heart disease risk has been transformed since the early 1990s not least by the mass use of statins. Any small effect of fish oils in those at risk of heart attacks may these days be overwhelmed by the impact of changes in risk factor management and improved treatment.
I started taking fish oils and/or cod liver oil myself several decades ago and have continued until recently, largely out of habit. I started taking them for joint pains, probably brought on by daily jogging on concrete pavements. I knew that the evidence supporting any benefits for “arthritic pain” was almost non-existent but I was persuaded to try them by the clever theoretical mechanisms suggesting how they might work.
It is difficult to prove a negative, so it is still possible that fish oil supplements may have some small beneficial effect for one or more of the conditions listed earlier. However any effect is so small that it is difficult to detect or is masked by other influences (like statins). It is particularly difficult to conclusively decide whether something has a small beneficial effect upon a symptom like depression where there is certain to be a high placebo effect.
Fish, whether oily or white, is a healthy nutritious food. However, claims that it has specific disease treatment or prevention effects are not supported by current scientific evidence. White fish is very low in fat and calories and is good non-meat source of protein, iodine and vitamin B12. A 4oz/100g salmon steak has fewer calories than 2 slices of bread has modest amounts of fat which is predominantly unsaturated fat and is a source of B12, iodine and vitamin D.