Dr. William Lands, one of the pioneers of the eicosanoid field, compiled this graph. It may be the single most important clue we have about the relationship between diet and ischemic heart disease (heart attacks).To explain it fully, we have to take a few steps back. Dietary polyunsaturated fatty acids (PUFA) are primarily omega-6 and omega-3. This is a chemical designation that refers to the position of a double bond along the fatty acid's carbon chain. Omega-6 fats are found abundantly in industrial vegetable oils (corn, soybean, sunflower, cottonseed, etc.) and certain nuts, and in lesser amounts in meats, dairy and grains. Omega-3 fats are found abundantly in seafood and a few seeds such as flax and walnuts, and in smaller amounts in meats, green vegetables and dairy.
The body uses a multi-step process to convert omega-3 and omega-6 fats into eicosanoids, which are a diverse and potent class of signaling molecules. The first step is to convert PUFA into highly unsaturated fatty acids, or HUFA. These include arachidonic acid (AA), an omega-6 HUFA, eicosapentaenoic acid (EPA), an omega-3 HUFA, and several others in the 20- to 22-carbon length range.
HUFA are stored in cell membranes and they are the direct precursors of eicosanoids. When the cell needs eicosanoids, it liberates HUFA from the membrane and converts it. The proportion of omega-6 to omega-3 HUFA in the membrane is proportional to the long-term proportion of omega-6 and omega-3 in the diet. Enzymes do not discriminate between omega-6 and omega-3 HUFA when they create eicosanoids. Therefore, the proportion of omega-6- to omega-3-derived eicosanoids is proportional to dietary intake.
Omega-6 eicosanoids are potently inflammatory and thrombotic (promote blood clotting, such as thromboxane A2), while omega-3 eicosanoids are less inflammatory, less thrombotic and participate in long-term repair processes.
Many of the studies that have looked at the relationship between HUFA and heart attacks used blood plasma (serum lipids). Dr. Lands has pointed out that plasma HUFA do not accurately reflect dietary omega-6/3 balance, and they don't correlate well with heart attack risk. What does correlate strikingly well with both dietary intake and heart attack risk is the proportion of omega-6 HUFA in tissue, which reflects the amount contained in cell membranes. That's what we're looking at in the graph above: the proportion of omega-6 HUFA in the total tissue HUFA pool, vs. coronary heart disease death rate.
You can see that the correlation is striking, both between populations and within them. Greenland Inuit have the lowest proportion of omega-6 HUFA, due to a low intake of omega-6 and an exceptionally high intake of seafood. They also have an extraordinarily low risk of heart attack death. The red dots are from the Multiple Risk Factor Intervention Trial (MRFIT), which I'll be covering in a bit more detail in a later post. They're important because they confirm that the trend holds true within a population, and not just between populations.
In the next post, I'll be delving into this concept in more detail, and explaining why it's not just the ratio that matters, but also the total intake of omega-6. I'll also be providing more evidence to support the theory.
35 comments:
How long does it take before the O-6 content of cells is balanced with O-3 content after switching from a diet with substantial O-6's to more Sat Fat, Monounsaturateds, and O-3's? I know it can't be exact but are we talking weeks or years?
Well done Stephan. Before long you'll have enough good health info in your blog to publish a book! :)
I've almost never seen a regression line of a scatter plot (if that's what the solid line running through the data points is) with so little variance around it! Amazing!
I second the book nomination :-P
Would be interested to know how unsaturated we are. I guess there's no simple test to determine this.
Darren's question is a good one. I completely eliminated vegetable oils from my diet over a year ago, and I have my cholesterol results to prove it.
Darren...the turnover of fats in the cell membrane takes about two years - heard that in an radio interview with Ray Peat.
It takes about 600 days to change 50% of adipose tissue, and change will still be going on at year 5.
The relationship is inversely exponential.
The rate of change varies between fat depots.
Breast fat will see significant change in months.
Author Omega 6 The Devils Fat
www.Omegasixthedevilsfat.com
This is a video lecture by Professor Lands who is truly one of the Heroes in the field of Omega 3 and 6, and has been carrying the Omega 6 torch for 50 years often in the face of ridicule, and is still referred to as a rebel on occasions by the media.
http://videocast.nih.gov/ram/crii01c103202000.ram
BBC food program While not the in-depth lecture Robert supplied, you can hear 5 minutes of Bill Lands discussing his work 8 minutes into this recent BBC food program. Here is a link to the
KIM software to monitor omega 3<>6 intake mentioned by Bill Lands.
Stephan,
I know this is off-topic, but have you read “Riddled with Life: Friendly Worms, Ladybug Sex, and the Parasites That Make Us Who We Are” by evolutionary biologist Marlene Zuk? I haven’t read it but stumbled on a little review http://www.sciencebasedmedicine.org/?p=498 of the book. Supposedly she claims that diseases have good effects on us. For example, the cystic fibrosis gene may protect us from cholera. And, she debunks the benefits of the Paleolithic diet claiming that there was too much variation in diets back then to know which to follow.
I’m all for reading opposing views in an attempt to prevent confirmation bias. However, I’m not a biologist and can use only my limited science knowledge to evaluate various claims. That is why I try to find experts like you that have a valid epistimological method. I’m curious as to your opinion on the book.
Ron
Darren,
It looks like Braesikalla and Robert B. have answered your question.
rkelley23,
I haven't read that book. There may be something to the idea that certain parasites help us function optimally. One of my readers has pointed me to a few studies on using intestinal worms to treat digestive illnesses and allergies. It seems to be effective. I don't know if that's because we're adapted to having parasites, or if the parasite is able to reduce the excessive inflammatory signaling caused by the modern diet.
The author's argument against "paleolithic"-type diets is a straw man attack. Who says a paleo diet has to mean one specific thing? The diet gives you a palate of foods to work with and you can paint whatever meals you want with it. That could mean high-fat, low-fat, highly saturated, mostly unsaturated (but low in omega-6), lots of fruit, no fruit, or whatever makes you feel good and satisfies your tastes. But it will always exclude neolithic foods such as grains, legumes, dairy, and all processed food.
"she debunks the benefits of the Paleolithic diet claiming that there was too much variation in diets back then to know which to follow. "
I haven't read her book. But this position strikes me as illogical and dismissive, and without much thought (or perhaps some opposing bias). There isn't ONE paleo diet for humans, 30,000 years ago or today. Perhaps 2.5 million years ago there was one basic diet, but humans have spread too far and wide since then to eat one diet.
Just wanted to point out that walnuts, while having a lot of omega-3 fat, have the most omega-6 PUFA per serving of these popular nuts. Macadamia nuts are extremely low in omega-6. I'd skip the walnuts, take some fish oil, and go for the Macadamias myself.
(n-6 is linoleic, n-3 is linolenic, data from the USDA database)
g. per 100 g. of nuts
n-6 n-3
Almonds 13.52 0.00
Peanuts 15.10 0.00
Pecans 20.63 0.99
Walnuts 38.09 9.08
Pistach 13.64 0.26
Cashew 8.48 0.07
Macadam 1.30 0.20
"And, she debunks the benefits of the Paleolithic diet claiming that there was too much variation in diets back then to know which to follow."
I would guess she knows little if anything about paleolithic diets or climate, or human migrations out of Africa.
Genetic data has confirmed that humans left Africa about 50,000 years ago, at the height of an ice age, and migrated north into colder climates. They sure weren't seeking plant foods. The first migration followed a coastal route and ended in Australia. Later groups made it into central Asia, Europe, China, Mongolia, Siberia, and then into the Americas. People could not have followed this pattern of migration unless they were preadapted to a carnivorous diet and they were hunting big game.
Read about the Journey of Man: http://news.nationalgeographic.com/news/2002/12/1212_021213_journeyofman.html
Watch the video: http://www.youtube.com/watch?v=OV6A8oGtPc4
In fact, archaeological evidence tends to support the thesis that paleolithic people, at least in the northern hemisphere, ate a diet similar to the arctic wolf (namely, carnivorous). See A brief review of the archaeological evidence for Palaeolithic and Neolithic subsistence at http://www.nature.com/ejcn/journal/v56/n12/full/1601646a.html
Quote from this paper: "There have only been two studies of Palaeolithic modern
humans, Homo sapiens sapiens. A study of the isotope values of humans from the late Upper Palaeolithic (ca 13 000 years old) site of Gough’s and Sun Hole Cave in Southern England
(Richards et al, 2000a) indicated, again by the d15N values, that the main source of dietary protein was animal-based, and most likely herbivore flesh. The second study (Richards
et al, 2001) was a survey of isotope values of humans from Gravettian and later (approximately 30 000 – 20 000 years old) Eurasian sites. The d13C and d15N values here indicated
high animal protein diets, but the type of animal protein was more varied than the Neanderthals, incorporating aquatic foods in their diets."
I would submit that in fact paleolithic diets had very little variation, the data we have indicating a diet composed largely of flesh from large herbivores, but this is a politically incorrect view, so people refuse to believe it.
Interestingly, this paper takes seriously Milton's hypothesis: "Milton (1993) writes that mandible size decreased due to the increased consumption of energy-rich plant foods such as fruits, and not necessarily meat. The complex skills required to harvest these energy-rich plants would also result in a selection for more intelligent hominids, with resulting increased brain size through time."
I don't see how anyone can take Milton's idea seriously. Modern chimps eat a diet that averages 75% of energy from fruits, yet they have neither a reduced mandible nor a big brain. Then, she needs to explain what "complex skills" an animal needs to outwit and capture those oh so witty and mobile fruits. By her hypothesis also the first human remains should be found in forests, but they are not, they are found on savannahs where fruits are relatively scarce. Also, we should find that human cultures place a higher value on fruit than meat. Then, she has to show us some fruits capable of supplying all the neural fatty acids needed for brain expansion (none exist). If she was correct, if hunting down and eating loads of fruit makes for big brains, chimps should have large brains and be writing her papers, not the reverse. In short her hypothesis is dead on arrival, yet this paper gives it undeserved respect.
Don
Thanks Don, Anna, and Stephan. I have a little reading to do. Don, I'll check out the links you provided.
Ron
Interesting series of posts. Apart from olive oil, vegetable oil is only edible after having gone through a refine/bleach/deodorize process.
The deodorization process creates a significant amount of trans-fats, as well as other molecules that don't exist in unprocessed foods. The longer the oil is deodorized, the more stable it is, and the more trans-fats are created.
In addition, as Stephan mentioned, vegetable oil is lightly hydrogenated to extend it's shelf life, which also creates significant trans fats.
And of course hydrogenation creates trans-fats and likely other 'unnatural' molecules.
On the other side, vegetable oil tends to be high in omega 6.
What's your take on these questions:
Assuming a high correlation between vegetable oil intake and CHD, how can we effectively distinguish the impact of trans-fats + other 'unnatural' molecules - vs. the impact of a higher level of omega 6?
A lot of these studies that control diet use a source of vegetable oil/margarine that is specially formulated. Is it possible that this 'special formulation' involves a very tightly controlled deodorization process that minimizes trans-fats? Could that also explain some of the positive results?
I don't mean to imply that I disagree with the concern re omega 6 - omega 3 ratios. It also seems that the processing component could have an even larger impact, and potentially make the walnut vs. macadamia question, for example, somewhat academic in terms of overall CHD impact
Marlene Zuk is one my favorite Evo-Biologist. You can either read her book "Riddle with Life" or search for her recent essay in the NY Times from which the following quotes were taken:
"As an evolutionary biologist, I was filled with enthusiasm at first over the idea of a modern mismatch between everyday life and our evolutionary past. But a closer look reveals that not all evolutionary ideas are created equal; even for Darwinians, the devil is in the details. The notion that there was a time of perfect adaptation, from which we’ve now deviated, is a caricature of the way evolution works."
"First, when exactly was this age of harmony, and what was it like? Scavenging, or eating the carcasses of dead animals left by (or stolen from) predators like lions, was probably replaced by active hunting and accumulation of wild plants about 55,000 years ago, and agriculture seems to have begun a mere 10,000 years ago. We did a lot of different things during each of these times."
And it's difficult to argue with her conclusion:
"This isn’t to say that we wouldn’t be better off eating fewer processed foods. And certainly we have health concerns that never struck our ancestors. But we shouldn’t flagellate ourselves for having modern bodies, and we shouldn’t assume that tweaking our diets or our posture will rescue us from all our current ills. That’s just a paleofantasy about the future."
There had never been perfect adaptation during the long and variable Pleistocene and we have never stopped evolving (*).
*: Cochran & Harpending "The 10,000 Year Explosion."
"But we shouldn’t flagellate ourselves for having modern bodies, and we shouldn’t assume that tweaking our diets or our posture will rescue us from all our current ills. That’s just a paleofantasy about the future."
Well, I'm not sure what she means by "modern bodies", but genetically we are not significantly different than Paleo bodies. And who's "assuming" about tweaking diets and posture? There is a fair amount of evidence that "tweaking" these things does cure some of our current ills.
Don,
Nice comment. I noticed you started blogging recently. I've added it to my reader.
I think it's interesting that we have more isotope data from neanderthals than paleo humans, why the heck is that? I wonder if it's because scientists know what they're going to find but don't want to face the implications.
Dr. Milton is a troublemaker. I have a book titled "Human Diet: its Origins and Evolution", that contains a chapter by her. It's mostly about how we can learn about the optimal human diet by looking at what our closest surviving primate relatives eat-- mostly leaves and fruit. Her whole taxonomy argument disintegrates when faced with the simple fact that neanderthals were carnivores. And our direct ancestors also ate meat as a significant part of the diet. Oops.
The other thing that's ironic is that all those leaves gorillas eat get turned into fatty acids by gut flora. From what I understand, the cellulose gets digested and the bacteria strip all the oxygen for oxidative phosphorylation, secreting fatty acids that the gorilla absorbs. So the diet that appears to be high-carb is actually deceptively high in fat. Oops.
usfoodtrends,
I agree it's very difficult to tease apart the effects of omega-6 itself and the results of harsh processing. My opinion is that there's enough mechanism to indict omega-6 itself, but I think there's room for both to be a problem.
John,
See, I think that's another straw man attack. Who is claiming that humans ate the exact same thing for our entire evolution and completely stopped evolving when we hit agriculture? The argument is that we haven't had sufficient time to fully adapt to agricultural foods, not that we haven't adapted at all. And there's good evidence to support that from clinical trials.
The scavenger thing is a theory. There's evidence for scavenging early in our history, but there's also clear evidence of hunting live prey hundreds of thousands of years ago and perhaps before that. Paleolithic people couldn't have subsisted on a mostly meat diet if they were scavengers. And modern hunter-gatherers are generally not scavengers.
But even if we were scavengers, it doesn't change the argument. We still weren't eating grains, legumes, dairy, sugar and vegetable oils. We were still eating, on a qualitative level, the same types of foods for roughly 2.6 million years: meats, tubers, vegetables, nuts, fruit, eggs, etc. in varying proportions. There has probably been some genetic adaptation to an agricultural diet, but 10,000 years (and much less for many people) really is a blink of an eye next to 2.6 million.
usafoodtrends
"Apart from olive oil, vegetable oil is only edible after having gone through a refine/bleach/deodorize process."
There are sources of cold pressed minimally processed oils like coconut and palm oil.
SMALL scale oil extraction is an ancient art that has been going on I suspect almost as long as there were sufficient food to press to extract the oils.
Also small specialist health manufacturers like UDOs claim to cold press a number of seed products.
They may be more palatable after treatment but were certainly edible before industrial treatment.
Yes I agree it is difficult to separate out the multiple effects effects of significant processing of oils at many levels. eg trans fats, effect on mineral and vitamin content, antioxidants, other organic compounds, vitamin K etc and then we add back a whole range of additives,
Many of the above will also apply to processed lard.
There are the mechanisms to explain why Omega 3 and 6 are so fundamental to health. In biological importance they were in the game at kick off. Some suggest they do not rank far behind oxygen in importance to life.
Author Omega Six The Devils Fat
www.omegasixthedevilsfat.com
I have enjoyed reading this blog since it was pointed out to me. Just out of curiosity, what is the leading cause of death for the Inuit people? Also, what research is there that compares the stress levels of these different groups? I have a personal feeling that the amount of "real" stress (i.e. physical exertion, trying to hunt for your food in the elements, etc) compared with the amount of "perceived" stress (trying to meet deadlines at work, worrying about what others think about you, etc.) may contribute as much to the well-being of people as their diet. I am not sure how one could quantify these two "stresses", or dissect the effects of the diet and the effects of the stress. i am a research technician that studies immunology, and am interested in how stress and diet can affect our immune systems...
There's a good comment on Marlene Zuk's article by Keith Norris on Theory to Practice here:
http://theorytopractice.wordpress.com/2009/01/22/nyt-article-the-evolutionary-search-for-our-perfect-past/
I like the closing bit: "That’s all fine and well, and far be it from me to stand in the way of those who choose to “take a needle for Team Human” in the advance of that evolutionary inevitability."
In other words, it's clear there has been some evolution towards coping with agricultural diets, but it's clear that that evolution is by no means finished. Many individuals discover that paleo/primal/real food or some variation thereof works for them and fixes their health problems - that's what counts.
Just for the record, olives are a fruit, so olive oil is not a vegetable oil but a fruit oil.
@ Stephan,
I have Food and Evolution which contains a similar essay by Milton. She doesn't think very carefully.
"The other thing that's ironic is that all those leaves gorillas eat get turned into fatty acids by gut flora. From what I understand, the cellulose gets digested and the bacteria strip all the oxygen for oxidative phosphorylation, secreting fatty acids that the gorilla absorbs. So the diet that appears to be high-carb is actually deceptively high in fat. Oops."
Exactly. In fact, after accounting for hindgut fermentation of fiber, which produces the short chain fatty acids (like butyrate, the signature of butter) that provide the major energy source for the gorilla, you find that these critters actually get 60% of energy from fat, 24% from protein, and only 16% from digestible carbohydrate (sugars and starches). So these guys are on an Atkins diet!
All this is discusse in this paper:
http://jn.nutrition.org/cgi/content/full/127/10/2000
But you have to read the data rather than the text, because Popovich and Jenkins interpret it as indicating that humans need a high fiber, high carbohdyrate, low fat diet. They appear incapable of accepting the implications of the data, which actually indicate that gorillas, 98% identical to humans, are adapted to low carbohydrate, high fat, high butyrate, high protein diets.
These people also seem to ignore or forget that the primate line comes from insectivores. Insects are animal food, and their macronutrient profile is almost identical to wild game.
Don
"Just for the record, olives are a fruit, so olive oil is not a vegetable oil but a fruit oil."
Anybody have any thoughts on pressed avocado oil? Isn't avocado oil pretty much like EV olive oil nutritionally?
I just bought some really nice locally grown avocado oil, www.BellaVado.com, that tastes very nice. They also make a version with lemon oil, super drizzled on freshly grilled fish or a salad.
The only retail store source I know of is their online store and SD area Whole Foods (not sure of WF nationally). But they had a booth at the fledgling farmer's market at my son's school grounds on the weekend, which was great to see.
While watching them wrap my bottles in green tissue and a nice earthy handled brown kraft paper "wine bag", I realized what great gifts these would make - at 250 ml each they are reasonably priced (a bit less at the FM than at WF); relatively easy and small to mail or pack for carry-on when traveling; they have "local flavor"; are healthy (I'm pretty sure, but I'll look further into this); and not so common, like olive oil. They also make bars of avocado oil soap, but I haven't tried the soap yet.
Anna, avocado oil has a few percent more omega-6 than olive oil, but is otherwise very similar in the major components. I put together a table of fats and oils from USDA data that compares omega-6, omega-3, pufa, mufa,and saturated percentages and also an Excel dietary nutrition calculator that has the detailed fat analyses from USDA.
Seeing the Mediterraneans smack in the middle of the graph (and understanding that more and more of them these days, do not necessarily follow a traditional diet of their region, but rather a version of the SAD, diet makes me wonder.
Are we looking at this wrong? Couldn't it more likely be a cluster of elements in the diet specific maybe even to the fat soluble vitamins A,D,and K? The reason I say this is:
1. We know that O-3s tend to act as anti-inflammatory almost like a cortisone-mimetic by not only reducing inflammation through the prostaglandin pathways but also because they reduce the repress the immune system. So you have a native diet rich in O-3 from fish, seafood. You also have a diet rich in O-9 (from olive oil) which according to some scientists is considered a "neutral" oil.
2. Because of the relationship to the sun, most people of the region have a higher production of Vitamin D (theoretically). And the association between health and higher Vitamin D levels is being shown in study after study.
3. Also, their Vitamin A intake has also got to be higher (again, because of the fatty fish along with a high level of greens in their diet).
4. And a high level of Vitamin K2 from greens via K1 and eating a higher level of fermented greens and food, so K2 through fermentation. As a region they tend to eat some sort of deep leafy green at almost every meal (kale, escarole, mustard greens, etc).
Just a thought. But it occurs to me that the collection of fat soluble vitamins may have more in relationship to each other?
What do you think about natural sources of omega-6 such as nuts and seeds?
According to this chart, it looks like the less omega-6 the better.
So should nuts and seeds be avoided?
I read the link on gorilla diets (Popovich et al 1997). They cite evidence that in humans a gram of fiber produces about 1.5 calories worth of SCFA via fermentation in the colon. Since the feces contain SCFA, does this mean we don't absorb it? I'm no expert on digestive organs, but I thought the colon was just for dehydrating and shaping turds. Can somebody explain whether (and why/not) fiber contributes energy in humans? Isn't the mainstream idea that fiber is zero calories?
Pooti,
I do think it's multi-factorial.
Jon W,
Humans do derive calories from cellulose, a type of fiber. It gets turned into short-chain saturated fatty acids in the colon and we absorb it. We just aren't as good at it as gorillas.
The list of nuts and their o3 and o6 fats does NOT suggest avoiding walnuts at all. if what we are concerend about is the RATIO or proportion of 3's to 6's, and not the total amt of both in the food, then walnuts have by far the more favorable ratio (let's call it 4 to 1), about twice as favorable as macadamias, and infinitely better than the nuts that are listed at zero grams of 3's, no matter how much O6 they contain.
if what you want to do is simply avoid all these vegetable fats, then certainly drop walnuts from your diet, but i didn't understand that to be the point here, at all.
i am looking to improve (increase) the ration of omega 3 fats to both O6 and O9 in my diet, and therefore find the listed data very supportive of adding walnuts to my snack menus, or other recipes.
numbers, or graphs or tables, are notoriously hard to read correctly (I think the reasons are emotional and go back to being humiliated by junior high school math teachers), but this one is pretty clear.
Stephan,
To what extent do you think this HUFA/mortality curve would change if Lands had included more data? For example, inclusion of the JELIS data (using mortality endpoints) would not fit the curve. Thanks, Dan
Hi Dan,
I don't know how it would change if he included more data. Perhaps the variability would increase. I suspect the trend would remain consistent though.
Keep in mind JELIS was an omega-3 intervention trial, whereas all of the points on Lands' graph are from observational studies reflecting lifelong consumption patterns. JELIS lasted 5 years.
It's always more difficult to cure a problem than to prevent it.
"Humans do derive calories from cellulose, a type of fiber. It gets turned into short-chain saturated fatty acids in the colon and we absorb it. We just aren't as good at it as gorillas. "
My ex would agree
"Darling, could you please retrieve the duvet, I seem to have blown it across the room again"
Ironic that we are supposed to be eating soy rather than cows to reduce methane output, yet *my* methane output is several orders of magnitude less than it was since I stopped eating Healthy Whole Grains and started eating animals that aren't fed on grains either.
Post a Comment