Are diets high in omega-6 polyunsaturated fatty acids unhealthy?
Yes, so don’t eat ‘em. Thanks for reading.
Whaddya mean you want more of an explanation? They just are, okay!?
Look, there are plenty of health professionals who tell you to eat or avoid certain things with little in the way of an explanation, so why can’t I?
Tough crowd.
For the vast majority of you, I am preaching to the converted when it comes to warning of the perils of consuming large amounts of vegetable and seed oils, and the many processed foods that contain them, due to their relatively high omega-6 polyunsaturated fatty acid content (herein abbreviated as n6 PUFA). Specifically, we are talking about the n6 PUFA – linoleic acid (LA). Others might require a bit more convincing. Here goes…
The main reason we are consuming large amounts of this fat is due to the diet-heart hypothesis (Omega 6 fats, by the way, are the major fat present in the typical Western diet – so when health authorities/researchers point out that we consume too much fat, or are eating high-fat diets, this n6 PUFA is the fat they are referring to… even if they don’t realise it). This hypothesis states, as we all know, that high cholesterol levels, specifically high LDL (“bad”) cholesterol, is a strong risk factor for cardiovascular disease (*cough*bullshit*cough*). And by reducing our dietary intakes of saturated fat and replacing some of this energy with polyunsaturated fats such as corn oil, soybean oil, safflower oil, and so on, we reduce our total and LDL cholesterol levels and achieve immortality a reduced cardiovascular disease risk.
My last series of posts tackled some of the issues surrounding this hypothesis and the deficiencies that lie within it. Following on from those, but sticking with a similar theme, I’ve been reading through some papers that look specifically at healthy halo that surrounds n6 PUFA LA-rich vegetable and seed oils and asks whether the biochemistry matches the marketing hype. We’ll focus, in this post, on a review paper from 2001 published in the European Heart Journal, possibly drawing on a handful of other papers and maybe spilling over into another post – we’ll see how it all goes.
Before we crack into the gems contained within the 2001 paper, let’s set the scene with the abstract from a more recent meta-analysis on this very topic, just so we can firmly establish my confirmation bias from the get go…
Randomised controlled trials (RCT) of mixed n-6 and n-3 PUFA diets, and meta-analyses of their CHD outcomes, have been considered decisive evidence in specifically advising consumption of ‘at least 5-10 % of energy as n-6 PUFA’.
Here we (1) performed an extensive literature search and extracted detailed dietary and outcome data enabling a critical examination of all RCT that increased PUFA and reported relevant CHD outcomes; (2) determined if dietary interventions increased n-6 PUFA with specificity, or increased both n-3 and n-6 PUFA (i.e. mixed n-3/n-6 PUFA diets); (3) compared mixed n-3/n-6 PUFA to n-6 specific PUFA diets on relevant CHD outcomes in meta-analyses; (4) evaluated the potential confounding role of trans-fatty acids (TFA).
n-3 PUFA intakes were increased substantially in four of eight datasets, and the n-6 PUFA linoleic acid was raised with specificity in four datasets. n-3 and n-6 PUFA replaced a combination of TFA and SFA in all eight datasets. For non-fatal myocardial infarction (MI)+CHD death, the pooled risk reduction for mixed n-3/n-6 PUFA diets was 22 % (risk ratio (RR) 0.78; 95 % CI 0.65, 0.93) compared to an increased risk of 13 % for n-6 specific PUFA diets (RR 1.13; 95 % CI 0.84, 1.53).
Risk of non-fatal MI+CHD death was significantly higher in n-6 specific PUFA diets compared to mixed n-3/n-6 PUFA diets (P = 0.02).
RCT that substituted n-6 PUFA for TFA and SFA without simultaneously increasing n-3 PUFA produced an increase in risk of death that approached statistical significance (RR 1.16; 95 % CI 0.95, 1.42).
Advice to specifically increase n-6 PUFA intake, based on mixed n-3/n-6 RCT data, is unlikely to provide the intended benefits, and may actually increase the risks of CHD and death.
We have a lot going on here. Firstly, when we review research around any potential benefits of n6 PUFA’s, we also need to be certain that alongside an increased downing of soybean oil, subjects haven’t been barbecuing large lumps of n3-PUFA (omega-3)-rich salmon, and the like, on the side. And if we are going to be making claims regarding the benefits of n6 PUFA replacing saturated fat (SFA), we need to be sure that they are indeed replacing saturated fat alone and not saturated fat PLUS the infinitely more troublesome trans fatty acids (TFA’s).
What we can’t have is omega-3 fatty acids replacing trans fatty acids and then claim that substituting out your butter for margarine is doing your cardiovascular system any favours. Because clearly it’s not. And in fact, based on the above meta-analysis, replacing SFA and TFA with n6 PUFA actually leads to an increased CVD risk. But selling the opposite message has obvious financial benefits for stakeholder companies if the public and their guiding health authorities are sucked in by it all.
Almost a decade prior to this meta-analysis, we have questions being asked in a review by Elliot M. Berry from the Department of Human Nutrition at the Hebrew University Hadassah Medical School, Israel. He writes;
However, in spite of the widespread enthusiasm for increasing PUFAs in the diet some 20 years ago, more recent studies from a variety of disciplines suggest possible deleterious effects of a high consumption of LA.
Israel has one of the highest dietary polyunsaturated/saturated fat (P/S) ratios in the world; the population consumes some 8% more omega-6 PUFAs than in the U.S.A., and some 10–12% more than in most European countries. Israeli Jews may be regarded as a population-based dietary experiment of the effect of a high omega-6 PUFA diet on disease, one which, until recently, was widely recommended. Despite such national habits, paradoxically there is a high prevalence of cardiovascular diseases, hypertension, non-insulin dependent diabetes mellitus (NIDDM) and obesity—all diseases which are associated with hyperinsulinaemia and insulin resistance.
Berry touches on what is commonly known as the Israeli Paradox, with paradox being code for “these bastards aren’t doing what they are supposed to when eating foods that we recommend they do or don’t eat and they are making us look quite silly”. A prime example is the French – they eat a fair amount of saturated fat and yet the buggers just won’t die of heart disease like they are supposed to. Because we all know that saturated fat is artery clogging, yet the French don’t generally have clogged arteries to the same degree as other countries eating less arterycloggingsaturatedfat, we call that a paradox (and we make a big deal of it, suck up some research money over it, and generally look for all sorts of explanations for it other than the patently obvious one that saturated fat doesn’t actually clog arteries).
Conversely, the Israeli’s eat seemingly large amounts of hearthealthyvegetableoils – way more than the recommendations, yet they are having cardiac events left, right, and centre. Now of course the argument could potentially be made that, despite the high number of events, their mortality rates are quite low. Israeli men, at least, are ranked very high in terms of lifespan. They tend to live a good length if they survive the high rates of heart disease, cancer and diabetes and this likely is more reflective of Israeli doctors have good success in treating these diseases than any protective effect of vegetable oils.
Here is a study on exactly this paradoxical phenomenon…
Israel has one of the highest dietary polyunsaturated/saturated fat ratios in the world; the consumption of omega-6 polyunsaturated fatty acids (PUFA) is about 8% higher than in the USA, and 10-12% higher than in most European countries. In fact, Israeli Jews may be regarded as a population-based dietary experiment of the effect of a high omega-6 PUFA diet, a diet that until recently was widely recommended. Despite such national habits, there is paradoxically a high prevalence of cardiovascular diseases, hypertension, non-insulin-dependent diabetes mellitus and obesity-all diseases that are associated with hyperinsulinemia (HI) and insulin resistance (IR), and grouped together as the insulin resistance syndrome or syndrome X. There is also an increased cancer incidence and mortality rate, especially in women, compared with western countries. Studies suggest that high omega-6 linoleic acid consumption might aggravate HI and IR, in addition to being a substrate for lipid peroxidation and free radical formation. Thus, rather than being beneficial, high omega-6 PUFA diets may have some long-term side effects, within the cluster of hyperinsulinemia, atherosclerosis and tumorigenesis.
There are a number of ways in which excess linoleic acid may adversely affect metabolic pathways. Omega-6 and omega-3 fatty acids compete for the same enzymes involved in their conversion to the main fatty acids used within the body (AA on the omega-6 side of things, and EPA and DHA on the omega-3 side of the ledger). Each of these fatty acids are involved in the production of eicosanoids – signalling compounds involved primarily in inflammatory and immune function pathways. The omega-3-derived eicosanoids generally tend to drive anti-inflammatory pathways, whilst the omega-6-derived eicosanoids are inflammatory (keeping in mind that a degree of inflammation is needed for healthy functioning).
In general, eicosanoids formed from the omega-3 fatty acids are much less potent in causing biological responses than those formed from the omega-6 fatty acids, including such adverse reactions as the stimulation of cytokine production and inflammatory responses.
So to give omega-3′s a fighting chance, they have to be within cooee of omega-6 concentrations. Within the modern diet, however, this ratio is heavily weighted in favour of omega-6. And most health authorities want to make this worse.
This image is a nice summary of the specific essential fatty acid pathways and their influencers.
Lets get into specific health endpoints with regard to omega-6 fats. From Berry, again;
Atherosclerosis
Omega-6 PUFAs increase the susceptibility of lowdensity lipoprotein (LDL) to oxidative modifications and, perhaps because of this, the risk for acute myocardial infarction and coronary thrombosis. LA consumption may reduce the level of high density lipoprotein (HDL) cholesterol, increasing the risk for coronary heart disease (CHD) mortality. Lipid peroxidation mediated by free radicals and/or hydroxy radicals is considered associated with the activation of radical scavengers, initiation and development of atherosclerosis…
…LA and LA hydroperoxides reduce the activity of prostacyclins in the vascular wall, alter the production of thromboxane B2 and the tendency for platelet aggregation.
Translation: Linoleic acid and its oxidised metabolites make your blood vessels more constricted and drives your blood toward clotting. If you want a heart attack, those are a good couple of ingredients to start with.
Arrhythmogenicity
The metabolites of arachidonic acid (20:4n-6) derived from LA via the cyclooxygenase reactions may be arrhythmogenic.
Arachidonic acid (AA) can be considered an animal fat and as such can get itself into trouble (think “animal fats cause inflammation”). However, as it is also derived from linoleic acid, lots of linoleic acid can increase arachidonic acid levels markedly. Those people who suffer arrhythmia may be advised to cut down on animal fats in order to reduce AA, with little to no attention paid to the intake of LA chemically derived from animal fats.
Cancer
Some reports suggest that LA potentiates tumourigenesis by providing structurally and functionally essential fatty acids for the growth of dividing cells, and by serving as precursor for eicosanoid metabolites of arachidonic acid (AA). Leukotrienes C4 and D4 might act as tumour-enhancing agents, and PGE2 in turn may be involved in tumour development, promotion and immunosuppression. This reduces macrophage tumouricidal activity and inhibits interleukin-2 production, which activates the natural killer cell and cytotoxic T cell activity.
…higher linoleate to arachidonate concentrations were found in gynaecological tumours, perhaps indicating decreased insulin sensitivity.
Insulin resistance
Omega-6 fatty acids may increase the secretion of insulin, and/or reduce insulin catabolism, causing impaired insulin action and leading progressively to insulin resistance, which determines accelerated atherosclerosis. Insulin activates the enzyme phospholipase A2, which hydrolyses membrane phospholipids to generate free PUFA. These are substrates for eicosanoid formation via the cyclo-oxygenase and lipoxygenase pathways, during which processes free radical are generated which may enhance lipid peroxidation.
The same relationships hold for eicosanoids formed from dietary omega-6 lipids affecting thrombosis, vasospasm, arrhythmia and chronic inflammatory processes. In the absence of adequate insulin activity there is a relative decrease in the conversion of LA to AA.
Studies have shown that AA is inversely related to both HbA1c concentrations, as well as to the degree of diabetic complications. There is also a correlation between AA and insulin sensitivity in muscle biopsies.
In summary, there appears to be a reciprocal relationship between insulin and the omega-6 fatty acid pathway; these fatty acids impair insulin activity which in turn regulates their metabolic conversions.
The inverse relationship between AA and HbA1c (consider this a rolling average of your blood sugar levels and an indicator of how well your blood sugar is being controlled), is particularly interesting given that AA (preformed) is derived from animal foods, suggesting that such foods may improve insulin sensitivity.
Berry includes a section within his review which can be applied to all areas of human nutrition and how it relates to disease, but particularly with regard to studying the effect of dietary fat on health and disease…
Translating these animal and experimental studies into the clinical situation requires knowledge of the long-term habitual diet. The evidence required to link diet and disease may be assessed at a number of different levels. Work in cell culture (in vitro ‘feeding’) and animals may suggest possible biological mechanisms for findings from epidemiological/geographical studies. However, these relationships suggest association rather than causality, which is also the limitation of case control retrospective studies. Cohort studies are prospective and more reliable. However, dietary studies are notoriously difficult to evaluate because of the extended time-scale of the exposure and the lack of suitable instruments for assessing the long-term habitual dietary intake in the absence of objective biomarkers.
An example of such confusion is provided in the field of carcinoma of the breast, where most animal and case control studies suggest a relationship between fat and cancer, while cohort studies (meta-analysis) fail to do so. An answer, therefore, has to await a long-term dietary intervention randomized control trials with all their difficulties of compliance. However, work on animals suggests an alternative explanation, in that it is total energy intake which is of more relevance than the type or quantity of fat ingested.
Much is made of the ratio between omega-6 to omega-3 fatty acids in the diet. And for a while there, in the paleosphere, this lead to the recommendation of taking enormous doses of fish oil capsules in order to rebalance this ratio. However, in recent years there has been more focus on the absolute amounts rather than the ratio between the two (though it pays to keep the two in reasonable balance even if consuming them in relatively small amounts).
Berry makes a good point when considering the plant-derived precursors to AA (omega-6) and EPA (omega-3), and how we interpret any research that uses these plant derived fatty acids…
The competition between the two pathways of essential fatty acids is very difficult to untangle biologically. Thus it is not known with any confidence in what proportions C20 [AA] and C22 [EPA] long-chain fatty acids will be produced from a given mixture of linoleic and alpha-linolenic acids fed to any animal.
I’ll cap this post off with one last comment from Berry. This time in reference to the perception that omega-6 vegetable oils are a suitable and heart-healthy replacement for saturated fats in the diet…
The above considerations suggest that a reductionist (single nutrient) approach to nutritional recommendations may not be appropriate when considering fatty acids. They have to be considered in terms of their biological function in relation to other nutrients (integrationist approach)…
…attention to lifestyle and the quantities eaten are of far more importance than the type of fatty acid consumed.
To be continued…**
**If some of my sentences seem to slur a bit more than usual, it is because of the two glasses of Pinot I had whilst completing this post. And yes, I’m a cheap date when it comes to alcohol.
