Thursday, November 9, 2017

Omega-6 Fats: The Alternative Hypothesis for Chronic Disease

People have been asking me to write a post on what I've been learning about diet for a long time, specifically focusing on linoleic acid and omega-6 fats. Well, here it is.  Credit goes to Raphi Sirt of Break Nutrition, who hit me up at just the right time.

[P.S. Sadly Break Nutrition has gone offline. Here's a PDF version of the article as it was published there, I'll try to update this post with the references at some point—links in the PDF don't work.]

Here's the published version over at BN ("Omega-6 Fats: the Alternative Hypothesis for Diseases of Civilization"), and here's the tweet announcing its publication:

Reception seems to have been quite good.  Mark Sisson mentioned it in his weekly 'net round-up, and then re-tweeted it separately:
Prof. Tim Noakes also retweeted the original tweet. Probably my favorite reaction was this one:
So I was pretty happy with the reception, and with the piece. Raphi was a good editor, and added a lot to it with some citations to studies backing up the argument that I'd never seen before. So if you're interested in the topic, go read that version.

Raphi also asked me to be on his podcast. Due to my verbosity, and prediliction to avoid brevity and drill down, it turned into two episodes. They're below, in order.

Episode 23 – Tucker Goodrich dishes on bad fats

Episode 24 – Tucker and Gabor on Seed Oils vs Refined Carbs – Part 2

I thought I would post the first draft here, mainly to get it off my hard drive, but also to have something to reference in posts on this blog. Links to citations, if you're curious, are in the final.

So here's the rough draft as delivered to Raphi:

Omega-6 Fats: The Alternative Hypothesis for Chronic Disease

The world is facing a health crisis of unprecedented proportions. What have become known as chronic diseases, Western Diseases, or Diseases of Civilization (DC, a better term), have become pandemic as populations around the world adopt the lifestyle that first became prevalent in the country that perfected industrialization, the United States.

The DCs revolve around the Metabolic Syndrome, a set of signs of disease that include central obesity, excess fats in the blood, high blood pressure, and excess blood sugar. The diseases that associate with the MetSyn include the biggest killers in the industrial nations: heart disease, cancer, diabetes, neurological illnesses such as Alzheimer’s, and host of seemingly unrelated auto-immune conditions.

Many causes have been blamed for the spread of the DCs, including lack of exercise; wheat consumption; excess consumption of calories, carbohydrates, sugar, animal fats, or meat; environmental toxins such as pollutants or pesticides; genetics (most implausibly); and the dread “multi-factorial”.

Each of these proposed causal factors fails, in my view, to explain the pandemic. What I’d like to introduce is an alternative hypothesis, one that better fits the observed epidemiology, and that has clearly-described mechanisms that explain much of the pathology for the DCs.

Since it started in America, we’ll start with America. In the late 1800s and early 1900s three events occurred almost simultaneously. First, a method of detoxifying cottonseed oil, which had been an industrial waste product of the cotton industry, was discovered. Next came a method to “hydrogenate” cotton seed oil, or make it into a solid-at-room-temperature edible fat. Thus cottonseed oil and the hydrogenated derivative known as Crisco—introduced in 1911—entered the food supply in large quantities for the first time.

Simultaneously, America began to experience heart disease in large quantities. Previously a very rare condition, heart disease quickly became an epidemic, and the relatively new modern medical profession began to track it, and attempt to devise treatments. Similarly, cancer “control” became a concern, with the forerunner of the American Cancer Society being founded in 1913.

The single biggest change in the American diet over the 20th century was the increase in seed oils, which increased 1000-fold.

America, a massive agricultural exporting nation, exported the fruits of the industrial revolution, which included new foodstuffs, including white wheat flour, sugar, and vegetable fats— so called to distinguish them from the traditional animal fats that had been used as food sources from the beginning of human evolution. I will refer to vegetable fats as seed oils, to distinguish fats made from seeds such as cotton, corn, soy, rape, and others; from fats made from fruits, like olives or palm. This will become important later on.

Wherever American industrially-produced foods were introduced, DCs soon followed. Although not generally well known now, a number of doctors and scientists recognized the impact of industrial foods on populations abandoning their traditional diets for what has become known as the Standard American Diet (SAD). I prefer the acronym MAD, Modern American Diet, as the American diet prior to industrialization was a largely meat-based one, and did not produce the same diseases, and will use that throughout.

While there are many cases we could examine, perhaps the most telling was the MAD being introduced to post-WWII Japan. After the Japanese surrender, America took over the southern Japanese island of Okinawa, and used it for their base of operations. Unlike most other traditional cultures into which the MAD was introduced, Japan was a highly industrialized country. Yet Okinawans still ate a traditional diet that revolved around pork, yams, and fresh vegetables, and made them famous as one of the “Blue Zones”, a population with exceptionally long lifespans—in fact, the longest in the world.

What is less known is what happened after America took control. The first American fast-food restaurant opened in Okinawa, 9 years before Tokyo, to meet the American soldiers’ appetite. Okinawans also enjoyed American fast food, and rapidly adopted it as their own.

In that one generation, DCs exploded into Okinawa. Fathers attending the deaths of their sons became a common occurrence, with obesity, heart disease, and cancer becoming common.

Harumi Okuyama, a Japanese scientist exploring the dramatic change in longevity in Okinawa, in 1996 published a paper pointing the finger clearly: Dietary fatty acids – the N-6/N-3 balance and chronic elderly diseases. Excess linoleic acid and relative n-3 deficiency syndrome seen in Japan. (Omega-6 = n-6, omega-3 = n-3.)
“The age distribution of the survival rates of male Okinawans in 1990 is also interesting. The mortalities from all causes for the generations over 70 years of age were the lowest, whereas for those males less than 50 years old they were the highest among the 47 prefectures...”
In a single generation, Okinawa went from the lowest mortality in Japan to the highest.

This one incident clearly disproves a number of the leading hypothesis for the emergence of DC. It’s clearly not genetic (although that’s a factor, especially in Japan), as genes don’t change that quickly. It’s not caused by carbohydrates, as the Okinawans had a high carb diet prior to Americans arriving. It’s not caused by meat, as the rest of Japan had a huge increase in meat consumption after WWII, and longevity increased, unlike in Okinawa. It’s not caused by animal fats, as they were rapidly replaced by cheap seed oils imported from America—which introduced a program to “Americanize” the Japanese diet. It’s unlikely to be caused by saturated fats, as consumption of saturated fat was and remains lower in Japan than it was or is in America, and there are no mechanisms to explain how natural saturated fat causes disease. Environmental toxins are an unlikely explanation too, as, while Japan was industrialized, Okinawa didn’t experience a major change in that regard. There’s also no possibility of a local environmental factor in Okinawa, as Japanese who moved to America saw a similar increase in DCs. And just to throw it out there, this also disproves the cholesterol-heart disease hypothesis, as cholesterol is not associated with heart disease in Japan, but with increased longevity.

Background of Excess Linoleic Acid Syndrome—ELAS

In my reading, the diseases that surround the MetSyn share common traits. Inflammation and insulin resistance are oft-cited, but perhaps more significant traits include mitochondrial dysfunction, apoptosis, necrosis, and genetic damage. These point to the common mechanism, named by Okuyama the ELAS, that I think is the best candidate for cause, and best explanation of, the MetSyn and related, chronic DCs.

Linoleic acid (LA) is an omega-6 fat which is considered essential to human and animal function.

(To head folks off at the pass, n-6 fats are found in all natural foods, so this isn’t something you need to avoid entirely, it’s not possible, nor necessary.)

LA is a polyunsaturated fat (PUFA, and I’ll spare you the chemistry), joined by the monounsaturated fats (MUFA) such as oleic acid (named for olive oil) and the saturated fats (SFA) such as stearic (named for steers, beef), or palmitic (named for palm oil). Fish oil is also a PUFA, but of the omega-3 variety.

N-6 PUFAs are primarily made by plants, as are the similar n-3 PUFAs, and are concentrated up the food chain by animals eating those plants. The major sources of n-6 PUFAs in the MAD are oils refined from seeds, and animals fed a high proportion of seeds. PUFAs have traits which make them of interest: they’re highly susceptible to oxidation (rancidity), unlike MUFA or SFA, and they’re used throughout the body as building blocks for tissues and for various signaling functions, after being converted into other chemicals.

The rancidity of PUFAs is the root of the problem. Both n-6 and n-3 PUFAs are highly likely to go rancid. Humans don’t detect the rancidity of n-6 PUFAs particularly well, they smell slightly stale, and people actually prefer the taste. Contrast this to rancid n-3 PUFAs, and imagine eating a rotten fish. No thanks! This is likely because concentrated n-6 foods were rare until the modern era.

Both n-3 and n-6 PUFAs going rancid produce toxins, but the n-6 fats produce worse toxins. Most notable of these—and best studied—are acrolein, HNE, and MDA; although there are many others. Collectively, they’re called oxidized linoleic acid metabolites, OxLAMs. Acrolein is the acute toxin found in cigarette smoke. HNE is the best marker of effects of ELAS, as it is only produced from n-6 fats. All three are both produced in cooking or heating n-6 fats, but are also produced in the body. How toxic are these products? Cooking with seed oils is the leading cause of lung cancer in non-smoking women in China.

The list of toxicities of these three chemicals is most impressive. Acrolein is a biocide, meaning toxic to all life. HNE and MDA are less bad than that but are cytotoxic (kill living cells), mutagens (induce mutations in DNA), and genotoxic (destroy DNA). OxLAMs are also highly reactive, which means they can combine with other molecules in the body, inducing and stimulating malfunction.

A Primary Mechanism of ELAS

An increase in n-6 consumption rapidly remodels the tissues in the body, as the fats are replaced throughout. In some tissues it happens within weeks, in others, like the human brain, it appears to take much longer. Increased n-6 consumption rapidly remodels cartilage, for instance, in all species studied, driving out the more stable omega-9 fatty acids (Oleic is a n-9 MUFA). The same happens in mitochondria.

As mentioned, mitochondrial dysfunction and DNA damage is a signature of the MetSyn and all related diseases. It’s seen in: obesity in the fat cells, diabetes in the pancreas, atherosclerosis in the linings of the vessels, heart failure, fatty liver disease, neurological diseases such as Alzheimer’s and Parkinson’s, and, most notably, cancer.

The mechanism for this is well-described, although not well-recognized. Excess n-6 linoleic acid (LA) consumption causes a remodeling of a molecule called cardiolipin in the mitochondria, the key energy-producing part of cells in all higher life forms. Cardiolipin comprised of LA is uniquely susceptible to oxidation compared to n-3 PUFAs, MUFAs or SFAs, and this can happen spontaneously, as LA oxidation can be catalyzed by iron, and cardiolipin is in constant contact with iron atoms in the mitochondria. When cardiolipin oxidizes, a chain reaction can start. In vitro, this reaction continues until all cardiolipin is consumed, but luckily our body has counter-measures. In this process OxLAMs are produced. HNE , for instance, causes other cardiolipin molecules to oxidize, thus potentially causing a self-sustaining chain reaction. Reactive Oxygen Species (ROS) are produced in the reaction, which can also cause adjacent cardiolipin to oxidize. However, OxLAMs are several orders of magnitude more damaging to the bodies than simple ROS. HNE itself can induce the production of ROS. Oxidized cardiolipin causes mitochondrial dysfunction, as mitochondria are impaired with oxidized cardiolipin. What follows is mitochondria either collapsing, inducing apoptosis (controlled cell death–cell suicide) or necrosis (uncontrolled cell death). Apoptosis is seen though out the DCs—in cancer the cells are essentially ignoring the apoptotic signal and going rogue. The Warburg Effect noted in cancers is a cellular reaction to mitochondrial dysfunction in which the cells adopt an alternative energy pathway. Thomas Seyfried, who has studied this, notes that dysfunctional cardiolipin are always seen in cancer cells. Necrosis is seen in late-stage DCs, such as atherosclerosis, cirrhosis of the liver, heart failure, and Alzheimer’s.

Once loose in the cells, the OxLAMs rapidly propagate, in a process known as Oxidative Stress. HNE and MDA are the primary markers used to measure OxStr. ROS cannot leave the mitochondria, which is well prepared for them, but OxLAMs, being water-soluble, rapidly distribute throughout the cells, and beyond. OxLAMs are also a regular part of mitochondrial function: HNE induces the mitochondria to down-regulate, as a basic feedback mechanism, presumably to limit its creation, and the important antioxidant glutathione (GSH) as well as the aldehyde dehydrogenase enzyme (ALDH) are both important in protecting the body against evolutionarily-expected levels of HNE. Unfortunately for us, excess HNE can impair the function of both GSH and ALDH, a basic defense mechanism, and allowing propagation of a runaway chain-reaction. Decreased levels of GSH are a regular sign of excess production of HNE, and a dietarily induced deficiency in GSH production predisposes to the DCs. A genetic deficiency in ALDH, which is highly prevalent in Japan and China, predisposes to all the DCs.

Confusingly, assays of n-6 status in pathological tissues often show a lower level of n-6 than other fats, and in these cases, addition of n-6 can actually improve function. This appears to be due to the chain reaction depleting LA or arachidonic acid (AA, a long-chain n-6 fat produced in the body from LA). N-6 levels are lower, but HNE levels have risen, as N-6 is converted into OxLAMs.

OxLAMs can bind to and alter the function of DNA, both in the mitochondria and in the cell nucleus. In fact, they appear to be the leading cause of genetic damage, as the markers used for genetic damage are those generated by OxLAMs. Widespread generation of mutagenic, genotoxic chemicals in vivo would go a long way toward explaining the genetic damage common in the DCs.

OxLAMs such as HNE directly induce inflammation, increasing inflammatory markers. Excess levels of LA-derived AA also induces inflammation, as it is used to build chemicals that send an pro-inflammatory message to the body. The mechanism of anti-inflammatory drugs such as aspirin, NSAIDs, and Cox-2 inhibitors is to partially impair this pathway.

It appears that a fundamental job of macrophages, an immune-system cell that attacks foreign cells, is to remove toxic OxLAMs from the tissues. Macrophage infiltration into tissue is seen in various DCs other than atherosclerosis, including obesity. One explanation is that the modifications made by OxLAMs to molecules cause those molecules to resemble Pathogen-Associated Molecular Patterns (PAMPs)—the molecules appear the same to macrophages as those on bacteria. Antibodies for OxLDL exist, and development of these antibodies for therapy against atherosclerosis has revealed that the antibodies are equally sensitive to bacteria-derived lipopolysaccharides and OxLDL. Anti-cardiolipin antibodies are seen in several severe auto-immune diseases, and are only sensitive to oxidized cardiolipin. Thus excess n-6 is a known cause of autoimmunity, and may be the fundamental cause of the increase in allergic diseases seen in Okinawa and around the world.

Specific Disease Pathologies

Cardiovascular Disease (CVD)

Goldstein and Brown received a Nobel Prize for discovering the LDL receptor. The next thing they tried to do was to induce the first stage of atherosclerosis, the conversion of macrophages into foam cells, which form the core of the atherosclerotic plaques that are thought to cause heart disease. They failed. Steinberg and Witztum then discovered that LDL must be modified, through oxidation, to cause macrophages to become foam cells.

“The nature of the substrate for lipid peroxidation, mainly the [PUFA]s in lipid esters and cholesterol, is a dominant influence in determining susceptibility. As noted by Esterbauer et al. (52), there is a vast excess of [PUFA]s in LDL, in relationship to the content of natural, endogenous antioxidants. The importance of the fatty acid composition was impressively demonstrated by our recent studies of rabbits fed a diet high in linoleic acid (18:2) or in oleic acid (18:1) for a period of 10 wk. LDL isolated from the animals on oleic acid-rich diet were greatly enriched in oleate and low in linoleate. This LDL was remarkably resistant to oxidative modification, measured either by direct parameters of lipid peroxidation”

(Emphasis mine.)

Esterbauer discovered HNE. HNE is always present in atherosclerotic lesions, in all species, and oxidized n-6 PUFAs comprise a large proportion of the fats found in those plaques.

OxLDL is the second-best known predictor of myocardial infarction, exceeded only by the OxLDL/HDL ratio. Other aspects of circulatory disease, such as varicose veins and erectile dysfunction, also display increased rates of OxStr, as determined by the presence of OxLAMs. HNE and oxLDL are active throughout the pathological process, being required for creation of foam cells, inducing macrophages to enter the lining of the vessels, and being able to induce the apoptosis, necrosis, and DNA damage seen in atherosclerosis.

Steinberg and Witztum followed up their rabbit study with a human study, which confirmed the importance of LA in creating OxLDL. Other studies have confirmed the effect. The most successful CVD prevention trial ever, the Lyon Diet Heart Study, which produced a 70% reduction in CVD rates, specifically reduced the consumption of LA and increased the consumption of n-3 and n-9 fats, thus validating the mechanism in humans (although the metabolites were not measured).

Similarly, oxidative activities of OxLAMs are seen in varicose veins, with these markers always being present at affected areas. Erectile dysfunction also appears to be a consequence of this process. Erectile dysfunction drugs, aside from the obvious effects, also have an antioxidant effect, and appear to prolong life in those with vascular diseases.

Cancer is not, of course, a single disease, but a wide array of diseases with, so far as we know, different causes. Viruses are a well-known cause of certain cancers, so it is safe to say that there is no single cause of cancer, however appealing that prospect would be. Much of the pathological behavior of cancer cells can be explained by the effects of OxStr: mitochondrial dysfunction, genetic damage, and a shift to glycolysis. HNE damages and impairs the function of pyruvate dehydrogenase, the enzyme that allows substrates produced by glycolysis to enter the mitochondria. This loss, combined with malfunctioning mitochondria emitting high rates of ROS and OxLAMs, may explain the metabolic dysregulation and anti-oxidant upregulation often seen in cancer cells. Epidemiological work has shown low rates of cancer in populations eating traditional diets; in Asians, migration to industrialized countries increases breast cancer rates many-fold, as they reach Western levels. LA in the diet is, in fact, required to induce cancer in animals experimentally, and the cancer-promoting effects of LA increase as it increases in the diet. This effect plateaus at 4% of energy, well below industrial levels of consumption. One of the hallmark traits of cancer cells are apparently random mutations, often tens of thousands of them, and sometime none at all. For many cancers, the theory that it was the result of a single mutation, this is apparently invalid for most cancers. HNE and MDA both damage DNA, and HNE preferentially damages the p53 gene, which is part of the body’s natural cancer-control mechanism, as seen in colorectal cancer.

N-6 PUFA consumption appears to be involved in obesity through several pathways. HNE, which is elevated in obesity, directly induces fat-storage across species at an intra-cellular molecular level and induces pathological behavior of adipose tissues, including a failure of those tissues to differentiate, leading to the engorged adipose cells typical of pathological obesity. As LA converts to AA, AA upregulation leads to increased production of endocannabinoids, including 2-AG, the endogenous mimetic of THC, found in marijuana. Injection of either 2-AG or THC in animal models induces overeating, regardless of satiety. Elevated 2-AG is a typical feature in human obesity. Several studies by a group at the NIH has shown that dietary LA modulates production and levels of 2-AG, and directly induces obesity. Blocking the endocannabinoid receptor using a drug prevents obesity and metabolic syndrome, in animals and humans; however the side effect is an increased rate of suicides, so the drug has been withdrawn.

One long-observed medical observation is that insulin resistance (IR) accompanies sepsis, a condition caused by infection. OxStr precedes IR in humans, and OxLDL antibodies acutely reduce IR in a primate model of atherosclerosis. So the PAMPs found in OxLDL appear to be interpreted by the immune system, and IR may be a reaction to a perceived infection. A reduced LA diet has been shown to reduce IR in two human studies; one measured OxLAMs, which were reduced. HNE injected into skeletal muscle cells directly induces IR in those cells. Mitochondrial dysfunction is a common feature of diabetes, and the resulting shift in energy production may play a role in IR. LA directly induces hyperinsulinemia in vivo in animal models of beta cells (which produce insulin in the pancreas), and OxLDL causes beta cell death directly. Hyperglycemia, high levels of blood glucose, is a central feature of Type 2 diabetes, as glucose production is dysregulated in the liver. High levels of HNE accompany dysregulation of gluconeogenesis, although I’ve not seen a mechanism to explain it.

Neurological Diseases
OxStr has become recognized as a major pathological factor in several severe neurological conditions, which incidence has been increasing along with increasing n-6 consumption such as Alzheimer’s, Parkinson’s, and Lou Gehrig’s disease (ALS). The human brain (as opposed to rat brains) appears to have a rate-limiter for uptake of LA. However, AA, which is only produced in small quantities from LA but concentrates in tissues, does pass the blood-brain barrier. It’s been observed that AA increases prior to the onset of Alzheimer’s, but decreases after the onset of Alzheimer’s. AA is more subject to oxidation to HNE than LA is. HNE and other OxLAMs increase as AA decreases, perhaps indicating the self-perpetuating reaction is underway. HNE is always found in the pathological areas of the brain, and in an animal model, injecting HNE induces the formation of beta-amyloid plaques, the signature of the disease.

Liver Disease
Non-Alcoholic Fatty Liver Disease (NAFLD) is a new disease to humanity which only appeared as n-6 consumption levels reached the current high levels. It mimics the effects of alcohol-induced Fatty Liver Disease (FLD). Like cancer, LA is required to induce FLD in animals, and very low levels of LA in the diet allow animals to consume up to 30% of energy as alcohol without pathology. Total Parenteral Nutrition (TNP) is a feeding strategy in humans with damaged or malformed guts. Fatty liver is a common consequence in TNP, and replacing LA-rich oils with fish oils cures the condition in human infants. A small pilot study from Poland in humans examining NAFLD reduced the dietary levels of N-6, while providing the bulk of calories as carbohydrates. OxLAMs were reduced, as was insulin, HOMA-IR, weight, and NAFLD resolved in 100% of subjects. HNE directly induces the fibrosis seen in advanced liver disease and in other DCs.

The leading cause of the blindness in the United States is Age-related Macular Degeneration (AMD). This is probably the only condition where the causal role of n-6 has not only been established, but is becoming widely recognized. It’s illustrative of principles that should likely be informative for treating other DCs. The retina of the eye is rich in PUFAs, and like other tissues are affected by diet. N-3 supplementation does not affect the progression of AMD, unless it’s accompanied by low-n-6 intake. That combination is preventative—an evolutionarily-appropriate balance of the fats appears to be crucial. N-6 fats are very susceptible to oxidation by radiation, even visible light; blue light will induce retinal n-6 PUFAs to oxidize to highly toxic HNE. This may also be the causal pathway behind sunburn and skin cancer.

And More!
Osteoporosis, osteoarthritis, diabetic side-effects such as kidney failure, chronic pain, and my personal favorite, sunburn, all have pathological roots in ELAS. Even high rates of violence have been compellingly linked to ELAS.

Pathological Co-Factors

Several co-factors induce worsened OxStr is humans. In vitro, mixing LA, glucose, and water at physiological concentrations induces peroxidation of LA into OxLAMs; the same has been shown in an in vivo animal model, where increased n-6 feeding induced cardiolipin breakdown, and subsequent induced hyperglycemia caused mitochondrial collapse and cardiac necrosis, a condition seen in humans in heart failure, which is now epidemic. Alcohol, fructose, smoking, radiation, and infection all induce increased levels of OxStr, and, as in FLD, the effects of OxLAMs may play a role in the pathology associated with those factors. If two factors contribute to a disease, but the disease only appears when one is present, it’s logical to conclude that the required factor is causal.


The epidemiology around n-6 consumption and DCs reminds me of an old joke.
A policeman sees a drunk crawling around under a street light.

“What are you doing?”

“I’m looking for my keys!”

“You lost them here?”

“No, I lost them over there.”

“Why are you looking here, then?”

“Because this is where the light is!”
Epidemiology on nutrition is a very daunting task, and conducting it in a modern, industrial society is much easier than going to a traditional society with no government statistics or wealthy research institutions.

So most of the epidemiology looking at food consumption is done in the industrial nations, which have mostly had high incidences of seed oils for a very long time, before nutritional epidemiology became a science. Seed oils appear to cease increasing their promotion of cancer after they comprise 4% of energy, and saturate tissues at 5%. Most industrial populations get more than 5% of energy from seed oils, so comparing one to another is to compare high to high, and what one wants to see is high vs. low. Nevertheless, there are a number of studies looking at the DCs in populations with differing food patterns, and they strongly support the hypothesis, with DC incidence in populations consuming fewer seed oils being either fractional or non-existent.

This likely explains the rapid increase of DC in those countries with the introduction of the MAD.

What’s Missing?

According to several scientists I’ve read or listened to lately, including Dr. Ron Krauss, who would know, the National Institutes of Health have largely stopped funding clinical nutrition research. That, combined with the pro-n-6 PUFA bias in the American nutrition establishment, means that it’s unlikely much research will be done in the U.S. Much of the research seems to be done in either second-tier US institutions, or in Europe or Asia. There’s a lot of lab work looking at mechanisms, but not a lot of human interventional studies. The few that do exist, like the Lyon Diet Heart Study, Christopher Ramsden’s work, and the pilot LA Metabolite intervention from Poland, are very compelling, however. Research I would like to see are examinations of PUFA status and OxLAM load in those few people still eating a traditional diet without DCs, and interventional studies lowering n-6 PUFA. Most such studies make no attempt to lower n-6, but just add n-3 on top of it. Due to competition between the two types of fat, and as seen in AMD, this is unlikely to be a successful strategy. It’s essentially the modern Japanese diet: excess n-6, but good n-3. This has not prevented DC in Japan, but they do have lower rates of some diseases, like cardiovascular diseases.


HNE was discovered in the early 1980s, many decades after seed oils were introduced. The relationship between n-6 and endogenous production of signaling molecules was discovered later. Omega-3 fats became recognized as important after that. Research into this topic was in the 10s of papers in the 1970s, and has increased by 135-fold today. It’s a burgeoning field, but one that appears to be very much under the radar. In reading through the literature, I have come to the conclusion that the case for ELAS as the root cause of Diseases of Civilization is overwhelming.

Personally, I came to this topic through Stephan Guyenet and his excellent Whole Health Source blog. After months of reading his posts, and reading the papers he linked to, I decided to cut seed oils from my diet on the spur of the moment, standing in front of the salad dressings in a cafeteria. My irritable bowel syndrome of 16 years disappeared in two days. My carb cravings disappeared as quickly, allowing me to discover an underlying wheat sensitivity. Sunburn became a thing of the past, and as a pale blonde who had always assumed I was ill-adapted to life under the sun, this was revolutionary. My excess weight dropped off, along with my now too-large pants one morning, never to return. And after six broken bones in two years, I haven’t broken one since.

I started reading on the topic because I wanted to understand what had caused my personal health recovery, and I guess because I like puzzles.

What to do?

Avoid eating seed oils, foods containing seed oils—junk food, and animals fed high levels of grains and seed oils, like pork and chicken. Don’t go crazy with the nuts. Eat some fish. But no, you can’t fix ELAS with extra n-3 fats like fish oil, as the Japanese demonstrate.

This is the simplest health intervention ever, as it’s impossible when eating whole foods to become deficient in n-6 fats, and if even a fraction of the diseases with pathological signs pointing to n-6 are proven, you’ll still be far better off.

William E.M. “Bill” Lands spent his scientific career studying the role of omega-6 and -3 fats in the body. His conclusion, the title of an article written during his retirement, deserves repeating here:

“Prevent the cause, not just the symptoms”


  1. You write "An increase in n-6 consumption rapidly remodels the tissues in the body, as the fats are replaced throughout. In some tissues it happens within weeks, in others, like the human brain, it appears to take much longer. Increased n-6 consumption rapidly remodels cartilage, for instance, in all species studied, driving out the more stable omega-9 fatty acids (Oleic is a n-9 MUFA). The same happens in mitochondria."

    We'd all like to know how quickly a reduction in LA to <3% reverses this! Is there any data on that?

  2. When you say "Avoid eating ...animals fed high levels of grains and seed oils, like pork and chicken" are animals fed seed oils in addition to the grains they eat, and do we have any studies that measure the amount of LA in grain fed vs. grain + seed oil vs. grassfed/pastured animals?


Please don't comment here. Go to

Note: Only a member of this blog may post a comment.