[KempyKolibri]

This is just mechanistic speculation based on animal studies. There is no good evidence that I’m aware of showing harms from high n6 consumption in humans, outside of their contribution to caloric excess.

The n3:n6 ratio was a hypothesis that never panned out. If you look at the studies that “support” it, the “bad ratio” is brought about by reducing n3 levels below sufficiency, not by keeping n3 levels at the RDA or higher and then boosting n6 further.

So it’s no more evidence that n6 is harmful than taking a cohort of people, reducing half of the group’s iron intake to a minuscule amount and claiming that because that group’s “iron:magnesium” ratio is wrong, then consuming more magnesium is clearly harmful. It just doesn’t add up.

[davebrown10]

Actually, there is human research. Below is a message Olaf Adam sent to myself and several others on September 5, 2021.

    I refer to the very readable review by Philip Calder “A systematic review of the effects of increasing arachidonic acid intake on PUFA status, metabolism and health-related outcomes in humans.” His final statement is that an increase in arachidonic acid intake up to 1.5 grams per day does not significantly change the parameters associated with inflammation, blood clotting or atherogenesis. In this very interesting observation, I was astonished by the fact that the background diet was not taken into account. Although the intake of arachidonic acid with the background diet is reported several times, the resulting metabolic consequences are not discussed.
    From the data provided, it can be concluded that the background diet in all studies included in the review was a Western diet, the proportion of arachidonic acid being estimated at 200 to 400 mg per day. Our studies on healthy volunteers were carried out with formula diets and allowed a precisely defined supply of arachidonic acid over a period of 6 weeks. These studies have shown that the exclusion of arachidonic acid from the diet (vegan diet) causes a progressive decrease of this fatty acid from 11 + 3% of the total fatty acids in the cholesterol esters of the plasma to 8 + 2% after 6 weeks. The later studies on patients with rheumatoid arthritis have shown that an intake of arachidonic acid amounting to not more than 80 mg/day does not increase the concentration of arachidonic acid in the phospholipids of the plasma and in the erythrocyte lipids. From these findings I have concluded that the body's own production of arachidonic acid is around 80 mg per day. This means that the Western Diet provides approximately 2.5 to 5 times the estimated need for arachidonic acid.
    This intake that is higher than the requirement primarily has no negative consequences. We know from many studies that the "silent inflammation" characteristic for the prevalent diseases of western societies has a latency period of more than 10 years before the consequences such as arteriosclerosis and myocardial infarction become apparent. The body is evidently able to avert the consequences of an unfavorable diet for a long time. To do this, there are numerous regulatory options, such as substrate or product inhibition in the case of enzymes or the inhibition of transport to or incorporation into cells. Arachidonic acid has a very special metabolic pathway that offers possibilities for regulating absorption from the intestine, transport in the chylomicrons, metabolism via the enzymes involved and also for incorporation into the cells. For example, we have found a completely different efficiency for the uptake of arachidonic acid into the cell membrane for platelets compared to erythrocytes or granulocytes. It is therefore very likely that regulation options on the metabolic pathway of arachidonic acid can, to a certain extent, compensate for changes in intake.
    Only when too much arachidonic acid is present in the food for a prolonged time do these protective mechanisms apparently fail and inflammation and the manifestation of lifestyle diseases is seen. This explains the long latency period with which the diseases of civilization occur. It is documented in the literature that unhealthy supplementation, such as megadoses of vitamin E, has no deleterious effects on health-related outcomes for humans in the short term. Only long-term observations and meta-analyzes have been able to prove the increased overall mortality. These protective functions with which our body is endowed are very important, because our body is often confronted with megadoses of vitamins or other food stuff or unreasonable diets. Otherwise humans would already be extinct.
    In summary, I would like to note that the human metabolism has many opportunities to compensate for unreasonable interventions for a limited time. The studies available so far relate to arachidonic acid intake with the Western Diet. This condition may already have provoked defense mechanisms that delay the occurrence of Western Diet diseases. We all agree that the Western Diet is too meat-based. The studies included in the review gave me the impression that an attempt had been made to further aggravate an already bad situation, present in Western civilizations. The negative result of these studies is reassuring for me, but does not mean that this supplementation has to be harmless and without long-term effects.
    From my point of view, it would have been more productive from the experimental approach if vegans had been given the doses of arachidonic acid employed in the studies that are included in the review. This would come closer to the desired goal of the effects of arachidonic acid on PUFA status, metabolism and health-related outcomes in humans. Then one could also come to a result for the desirable intake of arachidonic acid, which I estimate for patients with inflammatory rheumatic diseases at 80 mg per day, corresponding to 560 mg per week. This corresponds to a diet with 5 vegan days and 2 days with consumption of animal products per week.

[KempyKolibri]

I don’t mind discussing arachidonic acid after this - I’m less woke on the literature on it and am always up for learning something new. When I was referring to no evidence in humans, I meant with regards to the idea of n3:n6 ratios.

If n3 is at sufficient levels, I see no reason to believe the n3:n6 ratio should be of any concern in terms of health risks. All the evidence I’ve seen cited to support the claim that a given n3:n6 ratio raises risks of negative health outcomes has induced “unhealthy” n3:n6 ratios by dropping n3 levels to insufficiency.

It’s on that point that I believe there is no supporting evidence in humans, and the fears of an issue are rooted in speculation. Happy to be proven wrong, though!

[davebrown10]

It's an omega-6 toxicity problem. Bruce Hammock writes, "Fatty acid composition in the Western diet has shifted from saturated to polyunsaturated fatty acids (PUFAs), and specifically to linoleic acid (LA, 18:2), which has gradually increased in the diet over the past 50 y to become the most abundant dietary fatty acid in human adipose tissue." https://pmc.ncbi.nlm.nih.gov/articles/PMC9060469/

Vijay p. Singh says, "Separately, on analyzing global COVID-19 mortality data and comparing it with 12 risk factors for mortality, they found unsaturated fat intake to be associated with increased mortality. This was based on the dietary fat patterns of 61 countries in the United Nations' Food and Agricultural Organization database. Surprisingly, they found saturated fats to be protective."https://www.medpagetoday.com/reading-room/aga/lower-gi/86940

It's interesting that fasting and exercise furnish some protection from excessive polyunsaturated fatty acid intake. For example, "The increased proportional intake of dietary fat, decrease in feeding frequency and increased physical activity in free-ranging compared to captive cheetahs are all predicted to result in enhanced mitochondrial FA oxidation through the lowering of circulating glucose concentrations and insulin:glucagon ratios. During fasting/refeeding cycles and increased levels of exercise, tissue PUFA concentrations have been shown to deplete rapidly in both humans and rats. These studies show that most PUFAs, including α-linolenic acid (ALA) and linoleic acid (LA), are preferentially oxidized in periods of exercise or fasting. During refeeding, SFAs and monounsaturated fatty acids (MUFAs), such as palmitic acid and oleic acid, are also more rapidly replaced than any of the PUFAs. Similarly, the concentrations of most plasma PUFAs and MUFAs have been shown to be significantly lower in rats fed a high fat ketogenic diet than in controls. The predicted increase in FA oxidation in free-ranging cheetahs is therefore likely to also skew their serum FA profiles toward lower proportional serum concentrations of PUFAs and MUFAs relative to SFA." https://pmc.ncbi.nlm.nih.gov/articles/PMC5167222/

In the final analysis, dietary saturated fats are benign, if not outright beneficial over a wide range of intakes as long as they are consumed in the context of healthy nutrient configuration as in whole foods. https://pmc.ncbi.nlm.nih.gov/articles/PMC7846167/

[KempyKolibri]

I read those papers and I don’t see any compelling evidence anywhere in there that n6s are toxic. The first one is just discussing burn patients and yes, we consume more n6s than we used to. But that’s not evidence of toxicity.

The COVID one is disappointing. I was expecting a longitudinal study where they perhaps hadn’t adjusted for confounding variables like obesity, etc. Unfortunately, the reality is far worse.

The only human outcome data they seem to have is a cross-sectional analysis where (as far as I can tell, correct me if I’m wrong) they just looked at gross consumption at a national level of different FA types, then saw if there was an association with COVID mortality. This is an insane way to test the hypothesis “consumption of n6 increases Covid mortality.” It just isn’t evidence at all.

I would take evidence to be something that is expected on a given hypothesis and not the negation of that hypothesis. For example, “we see the sun rise in the sky” is not evidence supporting the hypothesis “the sun orbits the earth”, because it would also be expected on the negation of that hypothesis, for example under another hypothesis: “the earth orbits the sun”.

In this case, we can think of any number of hypotheses that negate the hypothesis “n6 consumption increases covid mortality”. For example, “n6 consumption proxies for junk food consumption and population obesity, which increases COVID mortality.” The outcomes of the study are equally expected on both hypotheses. The outcomes they’re seeing could be explained by this alternate hypothesis, for all we know.

Fundamentally, though, there’s actually a much bigger issue here - the data are just cross-sectional. We have no idea if the increased COVID mortality is even taking place in the people that consumed more UFA - the data just don’t tell us!

This leads to an awkward bullet-bite one has to make in order to make a causal inference from data like these: you would also have to affirm that smoking increases lifespan. Unadjusted cross-sectional data shows that countries with greater cigarette consumption have longer lifespan (http://web.archive.org/web/20220325085356/http://www.thefunc...). Now, clearly this is not because of the beneficial effects of smoking. Perhaps it’s because more cigarette consumption occurs in those countries that are wealthier. The point is, cross sectional data is unsuitable for making causal inferences like this.

The rest is just mechanistic speculation and animal studies, not something that can be extrapolated to human outcomes (more about why later). So the study doesn’t actually show any negative health outcomes in humans from n6 consumption.

The next study is literally a study of cheetahs. Might be of interest if you’re a cheetah, or deciding on your pet cheetah’s diet. But we’re talking about human health here.

And then the final paper suffers from the issue of earlier one - trying to make a causal inference based on cross-sectional data. We have zero idea if the individuals suffering from SAP are even the ones consuming more n6s. Again, if we are to find this convincing evidence of n6 toxicity then we also have to grant that cigarette consumption increases lifespan.

A commonality in these studies is that they try to back up the cross sectional data with animal modelling and in vitro studies, but we have to bear in mind that the best data we have on translation of animal studies to human outcomes suggests the success rate is atrocious. The confidence interval for toxicity studies in animal studies translating to human outcomes includes .5, so you may well be better off tossing a coin than you are relying on animal studies to make inferences about human health outcomes: https://translational-medicine.biomedcentral.com/articles/10...

To sum up, this seems to be a collection of data from the lower end of the evidence hierarchy, none of which is even in the category of data suitable to provide information about what is healthful or harmful to humans. I don’t see why we would find this convincing evidence of n6 toxicity on its own, even before we get into all the much higher quality evidence pointing to benefits from n6 consumption.

[davebrown10]

    (1996) "Excessive signaling of arachidonic acid (AA) metabolites has been associated with various chronic degenerative or autoimmune diseases, and intervention with the metabolism of AA is widely employed therapeutically in these afflictions. In essence, AA is the most biologically active unsaturated fatty acid in higher animals. Its concentration in membranes and its magnitude of effects depend on its amount, or that of its precursors and analogues, in the diet. The tendency of the field of nutrition to ignore the role of dietary AA will optimistically be reversed in the future." The article also said, "The underlying rationale for this symposium is that dietary AA is perhaps the single most important nutritional determinant in regulating AA levels in Americans. This may ultimately account in part for the striking differences in chronic diseases between strict vegetarians and the bulk of the omnivorous population." https://pubmed.ncbi.nlm.nih.gov/8642436/

Both animals and humans have endocannabinoid systems that operate identically in terms of physiological outcomes. Consequently, excessive arachidonic acid in cell membranes has pathologically consequences. https://johnditragliamd.substack.com/p/the-other-fat-science

[KempyKolibri]

As I said, happy to move onto AA once we’ve closed off the n3:n6 ratio thing. So do you have any evidence that the ratio is an issue in humans, or do you concede there’s no such evidence currently?