Tuesday, July 20, 2010

A Common Sense Defense of Animal Fat

When one looks in the books and on the internet about a "paleolithic-style" or "evolutionary-based" diet, one will likely be confused at the end of it. Some will focus on what Kurt Harris calls "paleo re-enactment" - meaning we should be hunting wild boar, digging up roots, and honey is okay. (When I first told a friend of mine, who happens to be a gastroenterologist at Johns Hopkins, about my interest in paleolithic diets, she remarked, "The only guy I ever saw on a paleo diet was pooping leaves and blood." Take that as an object lesson, paleo re-enactors, don't forage unless you know what you are doing!). Some paleo folks will be (once again using Kurt Harris' term - he has a knack for them), "pc-paleo" - meaning low fat paleo. Loren Cordain and Boyd Eaton are scientists and paleolithic nutrition pioneers, and their first look at the research focused on the fact that animals in the wild tend to be leaner than our domesticated animals. Cordain's The Paleo Diet: Lose Weight and Get Healthy by Eating the Food You Were Designed to Eat and The Paleo Diet for Athletes: A Nutritional Formula for Peak Athletic Performance will both advocate a lower-fat approach. Notice both were published prior to 2007. And as this is where the research was headed at the time, the "paleo" diets studied in diabetics by Staffan Lindeberg and others were relatively low-fat and sometimes a little weird - not sure how many of us foraged for canola oil and mayonnaise. But there you go. That's where an IRB gets you, I guess.

What's so important about 2007? Well, that's when Good Calories, Bad Calories: Fats, Carbs, and the Controversial Science of Diet and Health (Vintage) came out. I think Taubes' work is incredible and amazing, and required reading for any health professional or nutritionist. However, I do believe he focuses a bit too much on the carbohydrate hypothesis as the cause of Western Disease. He does, however, do a heck of a lot to exonerate fat.

Since 2007 and "Good Calories, Bad Calories", there has been the advent of what many now call "primal" style paleo, exemplified best by Mark Sisson's excellent book, The Primal Blueprint: Reprogram your genes for effortless weight loss, vibrant health, and boundless energy. Like Stephan Guyenet, I do think Mark is a little too hard on the carbohydrates (starch has a lot in common with, but ultimately is very different from sugar. This could be its own post, but if you have the time, I highly recommend this youtube video of a lecture by Robert Lustig, a pediatrician and obesity expert. He has a rather strange idea of what a paleo diet is, but hey, doesn't everyone?).

Anyway, the "primal" folks love fat, real food, and allow for some "sensible indulgences" such as high fat dairy, dark chocolate and red wine. And it is rumored that even Loren Cordain is more pro-fat than he used to be. Anthropological reports from all over the world will show us that our ancestors favored the fatty cut of meat, anyway, leaving carcasses and the lean meat to rot, and savoring the glands, liver, brains, and marrow, all high in saturated fat (1)(2)(3). In fact, when our ancestors were forced to survive on lean meat alone, they endured a life-threatening condition called "rabbit starvation." These symptoms of fatigue, weakness, diarrhea, hunger, headache, and low blood pressure and heart rate come from a diet of too much protein, and can be ameliorated by substituting some of the protein with carbohydrates or fat.

I've been told most of my life that animal fat is bad for me. The scientific story why this is not true is explained in exhaustive detail in a number of different, excellent books (including GCBC, but also Eat Fat, Lose Fat: The Healthy Alternative to Trans Fats) I especially like Mary Enig, as it seems to me she is the researcher primarily responsible for finally getting the FDA to come down against the nutritional horror that is industrial trans fats.

There are two, final scientific arguments against saturated ("animal") fats, in this case palmitic acid, that are not addressed in full by any of the previous sources I've named above. One is that a high amount of free fatty acids (palmitic acid for the most part) in the blood causes insulin resistance. Any erudite fat-lover will say to this, yes, of course it does. Palmitic acid is released into the bloodstream by the metabolic action of our own livers every time we fast (overnight, for example) or lose weight. It has to be. Otherwise we would never burn our fat stores, and we can't store much glycogen. The release of palmitic acid signals our body that we might not have a lot of food around, and we better preserve our precious glucose stores (the glycogen) for our brain to use. Therefore our muscles become more insulin resistant, and we shift to burning fat as fuel rather than glucose (burning fat sounds good, right?) This mechanism is physiologic, and I would say has very little to do with the full body insulin resistance of diabetes. Stephan and Peter have their own, brilliant takes on the matter (more required reading in my mind - no one said learning about nutrition didn't take a lot of time).

But let's get down to brass tacks on the insulin resistance/palmitic acid debate. Remember, and this is key, that when we burn fat and lose weight, we release a flood of palmitic acid into the blood stream. Deadly, hard core saturated animal fat. Yes, our own livers are trying to do us in. So why is fat loss, really, the best treatment for diabetes? Why does bariatric surgery, with the resultant forced semi-starvation, fasting, and FAT LOSS often result in the immediate turnaround of diabetes (4)? (Let's see - maximum of 2 pounds of fat loss a week = 7200 calories of fat burned, more or less, which would equal 1029 calories of fat a day = 114 grams of saturated fat (more or less) a day supplied by our own rumps! (or, hopefully, our visceral belly fat)).

This physiologic insulin resistance in fasting can, in the short term, increase fasting glucose levels in diabetics, it's true, and Gary Taubes' explanation of muscle insulin resistance vs overall body insulin resistance is a good one. This finding led the American Diabetic Association to recommend high carbohydrate diets to diabetics for years, and, in the long term, that is a huge mistake.

Did you know that feasting on your own stores of animal fat can help heart disease too (5)?

All right, I think I've fairly addressed the insulin resistance. Now let's go to the last, best argument against fat. Lipotoxicity. Perhaps I shouldn't have waited until the end for this one, as it is rather biochemistry-heavy. We'll give it a whirl. I have the full text for this paper (6), but let's summarize with the abstract:

"Insulin resistance is one of the pathophysiological features of obesity and type 2 diabetes. Recent findings have linked insulin resistance to chronic low-grade inflammation in white adipose tissue. Excess storage of saturated fat in white adipose tissue due to a modern life style causes hypertrophy and hyperplasia of adipocytes, which exhibit attenuated insulin signaling due to their production and release of saturated fatty acids. These adipocytes recruit macrophages to white adipose tissue and, together with them, initiate a proinflammatory response. Proinflammatory factors and saturated fatty acids secreted into the bloodstream from white adipose tissue impair insulin signaling in non-adipose tissues, which causes whole-body insulin resistance."

Let's leave off the insulin resistance for a bit and go to the proinflammatory section. Basically, what this paper says is that if we stuff our fat cells full of fat and then release saturated fat, we get inflammation. We hate inflammation here at Evolutionary Psychiatry, and this paper seems to implicate saturated fat as a major cause in a huge part of our bodies, the white adipose tissue (much of our stored body fat). This inflammation is called "lipotoxicity."

Paleo-diet aficionados tend to agree that doing the following are anti-inflammatory in the diet: omega-3 fatty acids (and avoiding omega 6), avoiding grains (especially wheat) and casein, eating vegetables and fruits, and avoiding sugar (fructose) and processed food.

Using common sense and biochemistry, saturated fat is anti-inflammatory compared to either cooked monounsaturated or polyunsaturated fat, given that heat and air can make the vulnerable unsaturated bonds oxidize and become rancid, which everyone can agree is highly inflammatory and bad for you. (Cold, fresh virgin-pressed olive oil would escape the oxidation, I hope!)

But here we have some science telling us saturated fat is inflammatory. And I don't think we have a full handle on that. I think what makes the most sense is to speculate that lots of saturated fat is inflammatory when combined with lots of sugar (big thanks to Dr. BG for helping me clarify this in my own mind! *edit* But I've further refined my thinking on this matter since publishing this post - please see the comments! Thank you *end edit*) Burning fat is a signal to our bodies that there might not be a lot of food around. Eating carbohydrates signals summertime! Lots of tubers and fruit and foliage! We stumbled upon a beehive! Yippie! Store it up! There was never a situation in our evolutionary past when we combined the grotesque amount of sugar we consume year round on the western diet, combined with the all the fats. Dr. BG called this, metabolically speaking, putting our foot on the brakes and the gas at the same time. Not great for the transmission.

I think, overall, the most important anthropological lessons are these - hunter-gatherers are healthy and, excepting infant mortality and accidents, long-lived eating a wide variety of macronutrient ratios. The Kitavans are high carb, the Inuit and Tokelau and Masai high fat. Most other folks were likely somewhere in between. But they did not consume tons of sugar. Or industrial seed oils, or wheat.

Again and again I come to that point. Epidemiology and observation won't show us the truth, but it will show us what could possibly be true, and what cannot. Saturated fat alone does not cause heart disease or diabetes. Neither do starchy carbohydrates alone. Those facts are true unless the Kitavans and the Inuit are blessed with magical pancreases or livers or hearts. My guess is they have the same old hearts you and I have.

Hopefully, science, a holistic approach, and sense will tell us the rest.


  1. Hi Emily

    I checked out your last reference.

    My reading is this happens only when an over-large number of "full to bursting" adipocytes can finally no longer manage the process of assembling/disassembling TAGs from FFAs, which is what is required both for storage and release of fat to/from the fat cells. At his point some fat cells begin to die and macrophages gather in WAT. Thus leading to low grade inflammation.

    I understand the paper to imply this happens in the extremity of severe obesity, not at a BMI of 27.5 for example.

  2. Thanks LeonRover. So we're looking at a morbidly obese situation - but I also wonder about the case of type II diabetes - here's a quote from a 2002 Endocrine Review "Disordered Fat Storage and Mobilization in the Pathogenesis of Insulin Resistance and Type 2 Diabetes" (pg 112)

    "It is becoming more clear that lipotoxic events are tightly regulated to excess glucose levels... in fact, hyperglycemia is suggested to be a prerequisite for lipotoxicity in some models." It goes on to discuss results with experimentation with Beta cells, suggesting "elevated glucose and the resulting increase in malonyl CoA impairs fatty acid oxidation... and contributes to the accumulation of lipid metabolites... excess glucose levels will stimulate the expression of genes involved in lipogenesis. Taken together, the effects of glucose ultimately elevate intracellular fatty acids, which adversely affect beta cell function and insulin secretion. However, the exact mechanisms involved in lipid-induced beta cell death remain to be determined... free fatty acid induced lipoapoptosis has been demonstrated in... pancreatic beta cells."

    But of course, there has to be free fatty acids floating around in type II diabetes if one is losing fat. And due to the peripheral insulin resistance, there can be hyperglycemia with fasting, for example, as I wrote above.

    I suppose the last meta analysis of the epidemiology supports the fact that saturated fat, on a population level, is not associated with heart disease. But I still struggle with what is going on in the metabolism with a bad type II diabetic. There is no paleolithic model for that!

  3. I see the circumstances you describe as being pathological hyperglycemia and likely hyperinsulemia as well. And thus very little to do with palmitic acid and physiological insulin resistance, which as you described is glucose sparing and entirely normal!

  4. The increase in muscle IR is a necessary physiologic response. The confusion comes from conflating glucose tolerance with pathologic liver insulin resistance. Glucose tolerance measured with a CHO meal measures whole body IR, and the CW medical researchers think IR is a disease because they assume everyone eats the SAD and they have no understanding of high fat. It has been known for about 80 years that muscle and whole body IR adapt to the CHO fraction in your diet. If you eat VLC and fail an OGTT you may be normal, conversely a borderline diabetic may have a more normal OGTT only because they eat 300g carbs a day...

    It's the liver IR that tells if you are damaged and that is not being measured in isolation with OGTT.

    There is no evidence that eating saturated fat causes inflammation.

    The term "lipotoxicity" is a made up term that has no real meaning. I see it used to mean a variety of things, usually that saturated fat is toxic or that this food or that raises your TC or LDL, in addition to the sense used in the quotes you provide. The term is a tautology - it assumes that fat in the diet or lipids in the blood (even FFAs on LC diets) are intrinsically toxic in a Keyesian sense.

    For a good example of belief in "lipotoxicity", see William Davis's posts where he assumes that if fasting TG being high is bad, so must all postprandial elevations in TG. He obviously is a believer in the lipotoxicity concept. The reason elevated fasting TGs indicate risk is because they area sign of metabolic syndrome, not because the TGs themselves are "clogging your arteries" or something - that is nonsense and is usually the sense in which I see "lipotoxicity" used in the literature.

    The fact is that liver cells or adipocytes losing their ability to metabolize fatty acids and either abnormally accumulating (steatohepatosis) or releasing (hypertriglyceridemia) fatty acids is an effect of inflammation. The pathologic fat metabolism is EFFECT, not CAUSE. The cause is leaky gut/ PUFA consumption/fructose and it has zero to do with sat fat consumption being "lipotoxic".

    I do not believe the canard about consuming glucose and fat at the same meal, either. Not biologically plausible that you have to pick either fat or carbs to predominate at each meal. A sliding scale will do nicely, I think., as long as you avoid the neolithic agents of disease.

    Taubes' Carbohydrate hypothesis is correct if modified to include wheat and sugar, the two agents invariably introduced with civilization.

    I agree with Stephan and others that starch is not a neolithic agent and in those with healthy livers and no Metsyn should cause no problem.

    Good summary and nice blog..

  5. Hi Kurt - thanks! That makes sense. A belief in lipotoxicity necessitates a belief in all sorts of weird things. You ultimately have to believe that the best diet for diabetics is olive oil, wood chips, and lean protein, and secretly have to believe (or not think about) the fact that fat loss releases saturated fat, so fat loss must be bad for diabetics (okay, you could remark on the difference between fasting and eating saturated fat, but really...). And you have figure out where that point is between "bad" diabetes and no metabolic syndrome where saturated fat suddenly becomes harmful. And you have to come up with strange metabolism metaphors involving transmissions. I'm not sophisticated enough a molecular biologist, though, to poke at some of those lipotoxicity references and elucidate about what might be wrong with the beta cell culture media... but the cognitive dissonance resolves nicely if one can put aside lipotoxicity altogether. I like it.

  6. Until two years ago, I always ate a low fat diet. I especially avoided saturated fat. I absolutely could not get my Vitamin D or testosterone up to optimum even with large doses of each.

    I switched to the paleo diet with generous amounts of saturated fats mainly from coconut oil products. Result-both my Vitamin D and testosterone levels soared and I had to cut back on my doses. It is very probable that a contributing factor in our current epidemic of low Vitamin D levels and infertility is do to our under consumption of saturated fat.

    Also my HDL went from 51 to 79 and my LDL went from 150 to 143.

    Stephen has said that our body fat stores contain the same ratio of omega 6/omega 3 that was in our bodies at the time the body fat was created. For everybody obese in the US, that means their body fat contains high amounts of omega 6. When they lose weight, those omega 6 fats pour into the blood stream and cause high amounts of inflammation. It is the omega 6 causing the problem when we lose weight not the saturated fat.

    He also said that it takes two years to have a complete turnover in body fat. So I have a year to go before my 20-1 body fat is replaced with my present ratio of omega 6 to omega 3 of 3-1.

  7. Excellent post (and comments)! I also don't subscribe to hypothesis that consuming starch+lipids in same meal is bad (presuming no metabolic syndrome). Don Matesz at his blog Primal Wisdom made a good case that when compared to the great apes, humans are especially well adapted to handle starches like from tubers.

  8. Hi Jake - yes, I definitely oversimplified in suggesting that when we burn fat, we release only saturated fat. However, since we break down fats into fatty acids and glycerol before reforming them within adipocytes, and we make excess carbohydrates into palmitic acid, a large percentage of our stored fat will be palmitic acid. It is my understanding that omega 3 and omega 6 fats are preferentially stored in cell membranes all over our bodies rather than within the adipocytes, and that they compete for space on our cell membranes, (which is why a small amount of omega 3 can undo some of the damage of an oversupply of omega 6 - though Stephan's 4% rule from Dr. Lands' research makes the most sense here), so their metabolism is somewhat different, but I really have to look into the specifics of that in more detail.

    Hi Aaron - I consider this blog my unofficial (and unpaid!) "fellowship training" in evolutionary psychiatry, so believe me I am very grateful for all the comments. I don't believe in "food combining" or anything like that either. The concept of lipotoxicity has been troubling me for some time now, because it simply doesn't make sense. And I do worry about what to recommend for diabetics. I eat "primal" (or Pa-Nu style) myself - I don't have insulin resistance, and I strongly feel that is the healthiest way to eat (though in my mind - and for the kids - one can go plus/minus on the non-grain non-fructose non-processed food carbs and still be fine, with no fat to lose). The only way I could comprehend the concept of lipotoxicity was in some combination of vast oversupply of carbs and fat and inflammatory neolithic agents, which is what I was (likely unsuccessfully) trying to convey in my post, but I agree it is a stretch. I would much prefer to believe that lipotoxicity does not exist. Maybe I'll sleep better now.

  9. Late to the party but this is worth considering: ketosis attenuates lipotoxicity:

    Carbohydrate-restricted ketogenic diets were used as sole treatment for diabetes before the insulin era beginning in 1922 and are reported to outweigh the performance of isocaloric high-carbohydrate fat-restricting diets in alleviating glycemic control, dyslipidemia, and insulin resistance in type 2 diabetes (13–15). The efficacy of ketogenic diets is surprising in view of their inherent lipotoxic potential (16). This apparent paradox may be resolved by proposing that the lipotoxicity of high-fat diets may be because of downstream fatty acyl metabolites (e.g., diglycerides, triglycerides, ceramide) derived under conditions of carbohydrate and insulin excess, whereas the free long-chain fatty acid (LCFA) precursors (or their respective LCFA-CoA thioesters) may account for the surprising efficacy of carbohydrate-restricted ketogenic diets, if not allowed to be further metabolized into downstream lipotoxic products.



Tired of receiving spam comments! Sorry, no new comments on the blog

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