To compensate for this digression, let me show you a graph, which pretty much summarizes the "punch line" of this post. The graph below shows the omega-6 fat contents of 1 lb (454 g) of grain-fed beef and 1 tablespoon (roughly 14 g) of a typical industrial vegetable oil (safflower oil, of the high linoleic acid variety). As you can see, there is a lot more omega-6 in the much smaller amount of industrial vegetable oil. A gram-for-gram comparison would practically make the beef content bar disappear.
It has been estimated that our Paleolithic ancestors consumed a diet with an omega-6 to omega-3 ratio of about 1. While other estimates exist, the general consensus seems to be that that ratio was not much greater than 5. Western diets, in contrast, typically have omega-6 to omega-3 ratios of between 15 and 40. In some cases, the ratio is even higher.
Omega-6 fats are essential fats, meaning that they must be part of one’s diet. Fats make up about 60 percent of our brain. About 20 percent is made up of omega-6 and omega-3 fats. The primary omega-6 fat found in our brain is arachidonic acid, which is either synthesized by our body based on linoleic acid from plant foods or obtained directly from animal foods such as meat and eggs. The predominant omega-3 fat found in our brain is docosahexaenoic acid (DHA), of which certain types of fish and algae are rich sources.
Inflammation is an important process in the human body, without which wounds would never heal. Incidentally, muscle gain would not occur without inflammation either. Strength training causes muscle damage and inflammation, after which recovery leads to muscle gain. Omega-6 fats play an important role in inflammation. Generally, they are pro-inflammatory.
Too much inflammation, particularly in a chronic fashion, is believed to be very detrimental to our health. A very high omega-6 to omega-3 ratio seems to cause excessive and chronic inflammation. The reason is that omega-3 fats are generally anti-inflammatory, counteracting the pro-inflammatory action of omega-6 fats. Over time, a very high omega-6 to omega-3 ratio is believed to cause a number of Western diseases. Among them are cardiovascular complications, cancer, and various autoimmune diseases.
So, should you worry about too much omega-6 from grain-fed meats?
If you think that the answer is “yes”, consider this. Apparently the (arguably) longest-living group in the world, the non-Westernized Okinawans, consume plenty of pork. Pork is a staple of their traditional diet. It is true that the average cut will have an omega-6 to omega-3 ratio of more than 7, which is not very favorable. Pork in general, whether grain-fed or not, is relatively high in omega-6 fats. As a side note, pork is not a good source of linoleic acid (found in plants), even though it is a rich source of arachidonic acid, the omega-6 fat synthesized from linoleic acid by various animals.
It is difficult to estimate the exact amounts of omega-6 and omega-3 fats from grain-fed cuts of meat; different sources provide different estimates. Here are some reasonable estimates based on various sources, including Nutritiondata.com. A typical 100 g portion of grain-fed pork should contain about 690 mg of omega-6 fats, and 120 mg of omega-3 fats. A typical 100 g portion of grain-fed beef should have about 234 mg of omega-6 fats, and 12 mg of omega-3 fats. It does not take that much omega-3 to counterbalance the omega-6 obtained from grain-fed pork or beef, even if one eats a lot of them. Two softgels of fish oil will normally contain about 720 mg of omega-3 fats (they will also come with 280 mg of omega-6 fats). Three sardines will have over 2 g of omega-3 fats, and less than 200 mg of omega-6 fats.
Industrial vegetable oils (made from, e.g., safflower seeds, soybean, and sunflower seeds) are very, very rich sources of omega-6 fats, in the form of linoleic acid. There is a lot more omega-6 in them than in grain-fed meats. One tablespoon of safflower oil contains over 10 g of omega-6 fats, in the form of linoleic acid, and virtually zero omega-3 fats. About 2 kg (4.4 lbs) of grain-fed pork, and 5 kg (11 lbs) of grain-fed beef will give you that much omega-6; but they will also come with omega-3.
How much fish oil does one need to neutralize 10 g of pure omega-6 fats? A lot! And there is a problem. Excessive fish oil consumption may be toxic to the liver.
If you cook with industrial vegetable oils rich in linoleic acid (this excludes olive and coconut oils), or eat out a lot in restaurants that use them (the vast majority), you will probably be consuming significantly more than 10 g of omega-6 fats per day. The likely negative health effects of eating grain-fed meats pales in comparison with the likely negative health effects of this much omega-6 fats from industrial vegetable oils.
You should reduce as much as possible your consumption of industrial vegetable oils rich in linoleic acid, as well as other products that use them (e.g., margarine). Keep in mind that industrial vegetable oils are in many, many industrialized foods; even canned sardines, if they are canned with soybean oil.
It is also advisable to couple this with moderate consumption of fish rich in omega-3, such as sardines and salmon. (See this post for a sardine recipe.) Taking large doses of fish oil every day may not be such a good idea.
Should you also consume only grass-fed meat? Do it if you can. But, if you cannot, maybe you shouldn’t worry too much about it. This also applies to eggs, dairy, and other animal products.
References:
Elliott, W.H., & Elliott, D.C. (2009). Biochemistry and molecular biology. New York: NY: Oxford University Press.
Ramsden, C.E., Faurot, K.R., Carrera-Bastos, P., Cordain, L., De Lorgeril, M., & Sperling (2009). Dietary fat quality and coronary heart disease prevention: A unified theory based on evolutionary, historical, global, and modern perspectives. Current Treatment Options in Cardiovascular Medicine, 11(4), 289-301.
Schmidt, M.A. (1997). Smart fats: How dietary fats and oils affect mental, physical and emotional intelligence. Berkeley, CA: North Atlantic Books.
50 comments:
Pork in general, whether grain-fed or not, is relatively high in omega-6 fats.
Could you give me the source of this? I have heard Matt Stone say that the Kitavan pork is low in omega-6 because they are fed coconuts.
Great article, BTW.
I live with my grandparents, but my grandparents cook with corn oil. Your article explains why it's important to switch their cooking oils to beef tallow or coconut oil. But coconut oil isn't available here in China, so I must pay for expensive shipping costs...
Sorry for triple posting, but Ray Peat has a strong case against fish oil.
Luming Zhou: Have you considered Tea Seed Oil? Per Wikipedia it is heavily used in parts of China and apparently very low in Omega 6 oils. Good for cooking with almost no taste and a high flash point.
Scott W
Ned, great post. I have been diligently talking a few capsules of fish oil for years every day but recently rethinking this approach. I actively avoid all Omega 6 oils and have for a long time and I hardly ever eat out. I'm starting to think that a person who avoids N-6 oils should consume fish oil like a prescription medicine...if you eat something high in N-6 (such as a restaurant meal), take a fish oil capsule or two. Otherwise, it is not needed.
Now that you point out the important role of inflammation in building strength while lifting weights, which I am actively trying to do, I'm even more convinced that too much fish oil is working against my goals.
Scott W
When you said, "Western diets, in contrast, typically have omega-6 to omega-3 ratios of between 15 and 40.", shouldn't the numbers be the other way around? I thought the point is that western diets have too much omega-6.
Speaking of which, what would be a good ratio to shoot for, and/or how much omega-3's do you consume a day?
I also agree that a chart of grain fed beef vs vegetable oil would be be more appropriate since it's more common.. although I wonder how gamier meats, like kangaroo, match up!
I use coconut oil for frying, and macadamia nut oil for everything else. It has a high smoke point, very little polyunsaturated oil, and the little it does have contains a good omega-3 to omega-6 ratio.
Finally, I've read that the paradigm of omega-3 ("anti-inflammation")is good, and omega-6 ("pro-inflammation") is bad, is clearly an oversimplification. Data shows a diet rich in linoleic acid reduce the risk of CVD.
What are your thoughts on this study:
http://www.ncbi.nlm.nih.gov/pubmed/18990555
Ned
I like this, simple and straight forward. People are out there trying to weigh what is good or bad for them. This chart tells them what they should worry about.
It would have been even more informative, if butter could have been shown on the same chart.
Hi Luming.
The ratio for pork follows from the numbers provided in the post. It is quite possible that the numbers would be different in animals with a diet primarily of coconuts.
Butter is also a good choice for cooking. Certainly much, much less LA than industrial vegetable oils. Oliver oil is another good option, as it is mostly oleic acid (omega-9).
Yes, Ray Peat is someone that I like to read, even though I may not agree with everything he says.
Hi Scott.
Indeed, effects tend to be dose-dependent, and follow U-curve distributions. Too much or too little of anything, even a ratio, can have negative consequences.
Many bodybuilders take omega-6 supplements to amplify the inflammatory response to exercise, and thus gain more muscle. Not very wise, in my opinion, in the same way that loading on omega-3 is not wise.
On top of that, the supplements are expensive, and have essentially the same effect that industrial vegetable oils do.
Hi snoop911.
I suspect that if someone's diet places emphasis on natural foods, and includes organ meats and seafood, the ratio will be just fine.
In the article linked below, Simopoulos seems to think that around 4 or less is close to ideal:
http://www.ncbi.nlm.nih.gov/pubmed/12442909
The Fritsche article seems to be an opinion/review piece. I would have to look at the actual articles or numbers he is referring to.
I've just got the full text for the Fritsche article. It is a short opinion piece; not convincing at all. Let's see if Blogger doesn't cut off the link (it is big):
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6WPH-4TVJ01N-2&_user=952828&_coverDate=11/30/2008&_rdoc=1&_fmt=high&_orig=search&_origin=search&_sort=d&_docanchor=&view=c&_acct=C000049193&_version=1&_urlVersion=0&_userid=952828&md5=6a7306c439c3a5eb883c3e4343f85804&searchtype=a
Hi Michael.
Here is the makeup of what I think is 1 tbsp (14 g) of grain-fed butter:
Total Omega-3 fatty acids: 44.1 mg
Total Omega-6 fatty acids: 382 mg
Not a great ratio, but the omega-6 content graph bar for butter would look like nothing compared with the whopping 10 g of omega-6 provided by the same amount (14 g) of safflower oil.
The bottom line is that the amount provided by butter it tiny. Easy to counter with a bit of seafood; maybe half a sardine.
Another reason why the Fritsche article does not inspire a lot of confidence:
"... diets high in saturated fat and cholesterol are known to increase the risk of developing cardiovascular disease ..."
A bad start for any opinion piece, IMHO. Especially one published as recently as 2008.
Replacing refined carbs and sugars with saturated fat and cholesterol in one's diet seems like a very healthy choice:
http://healthcorrelator.blogspot.com/2010/02/want-to-improve-your-cholesterol.html
With reference to the Okinawan's and pork consumption, I think it is very important to note that they NEVER consume pork meat without first boiling it (anywhere from 30 minutes to 2 hours depending on the recipe) after which they further cook it in soups, stir fry's etc. The reason isn't so much to rid the meat of fat (although this process does), but to destroy the impurities that they believe exist in pork.
I own the cookbook entitled: "Okinawan Mixed Plate - Generous Servings of Culture, Customs and Cuisine" by Hui O Laulima, which contains MANY recipes using pork and ALL of them feature this beginning process of boiling. A check on the Internet will also reveal this interesting aspect.
I question whether the Okinawan people consume much omega 6 in the form of pork.
Hi Jenny.
Thanks much for that info. I think there is a lot in the traditional preparation of foods by certain cultures that is very healthy, and that we don't know about.
And I agree with you on that the Okinawans who adhere to their traditional diet would have very little to be concerned about regarding omega-6. In addition to preparation methods, like you mentioned, they also eat plenty of seafood.
Still, I think someone living the in US and eating bacon every day would also probably be generally fine, as long as s/he had an otherwise whole foods diet and avoided industrial vegetable oils.
Ned,
I have been analyzing data on fat consumption which has led to some interesting conclusions:
(1) About 60% of fat consumed in the U.S. now comes from vegetable sources, 40% from animal sources. About 80% of fat from vegetable sources comes from "added fats" or "visible fats", the rest from nuts, grains etc.
(2) 83% of all vegetable oils comes from soybean oil which is 62% linoleic acid.
(3) Based on 33% of caloric intake from fat and a 2400 cal/day diet, the average daily intake of linoleic acid comes to around 30 grams per day, much more than the 10 grams per day you mentioned.
On a different subject, per person intake of hydrogenated oils (margarine and shortening) peaked in 2000 at about 25% of total fat consumption,then fell by 50% between year 2000 and 2008, but total vegetable oil consumption remained about the same.
Hi Ned. Chicken is mostly Omega 6 as well, isn't it?
Hi Jack C., thanks for that info.
Hi Ellen.
Good point, the average cut of chicken meat has more omega-6 than pork or beef. Here is the content of 100 g:
Total Omega-3 fatty acids: 120 mg
Total Omega-6 fatty acids: 1240 mg
Still, not much compared with omega-6-rich vegetable oils, or even with nuts, like peanuts (which Jack C. reminded me of). One ounce (28 g) of peanuts has the following:
Total Omega-3 fatty acids: 0.8 mg
Total Omega-6 fatty acids: 4393 mg
One ounce of walnuts will have a much better ratio than peanuts, but still plenty of omega-6:
Total Omega-3 fatty acids: 2542 mg
Total Omega-6 fatty acids: 10666 mg
good post! got me to thinking... ieat a lot of sardines. one i like them and two there cheap. organ meats are also pretty cheap.
what do you think about conventional organ meat? does it have an ok ratio? mostly liver/heart is what i eat
after this post i als think i may stop taking fish oil. i wuld rather keep at sardines(way ceap then fish oil)
Hi Ned, You have a nice way of putting things. And I totally agree with you that the consumption of all these omega 6 containing seed oils should be restricted, they are not as healthy as both the 6- and 3-industry wants us to believe: http://bit.ly/ar6JLe
VBR Hans
Ned,
It seems to me that your statement that fish oil may be toxic to the liver is misleading. The study stated that fish oil, like corn oil, may support "alcohol induced liver injury". There is no indication in the abstract that fish oil injures the liver in absence of alcohol.
I am not aware that the Inuit have any liver problems from all the omega-3 they eat.
been reading more past posts of yours on the blog...and i have a question.
what is your opinion on using a primal ketogenic diet to gain weight vs a primal nonketogenic diet to gain weight?
i ask because it seems from the posts that overeating and gaining on a keto-diet will induce a further and worse insulin resistance and set you up for future failure unless you plan to stay keto for life.
Hi Mal.
Definitely the sardines provide omega-3 in a much more natural way, and in a relatively large quantity. A lot of other macro- and micro-nutrients come together, perhaps some very important ones.
Hi Jack.
Good point. My understanding is that the alcohol was added to provide a baseline that would make the effects occur within a short timeline (assuming that the effects would occur anyway, but within a much longer timeline without the alcohol). This is not uncommon in experimental studies. Here are two that are a bit more to the point, comparing different fats, still with alcohol intake as a moderator:
http://www.ncbi.nlm.nih.gov/pubmed/9397995
http://jpet.aspetjournals.org/content/299/2/638.long
If I recall it properly, the two studies above have been discussed by Peter in the same context. I know, rats are not humans, but the case for fish oil supplementation, at least in large amounts, doesn’t look very good.
And note that I refer to fish oil, which has never been part of the traditional Inuit diet. Consumption of omega-3 through natural foods is another story. A lot of other macro- and micro-nutrients come together, perhaps some very important ones. In this context you can also expect some hunger regulation of intake, which doesn’t happen with supplements.
Hi Hans, thanks.
Re. the primal ketogenic versus non-ketogenic. My take on this is that if you are not insulin resistant, both are pretty effective at getting your weight up to healthy levels.
From a naturalness perspective, Sapiens evolved to be an omnivore, so fruits and tubers are in. In fact, a diet with plenty of fruits and tubers will probably have an edge in terms of hunger up-regulation.
The insulin spikes created by fruits and tubers are not all that higher than those elicited by natural protein-rich foods. They are much lower than those elicited by bagels and doughnuts though.
Having said that, depending on one’s level of insulin resistance, a primal ketogenic diet may be much better. Carbs, even from fruits and tubers, are toxic for some people; that is an undeniable reality.
Via Wikipedia:
Linoleic acid is 18:2(n-6).
Arachidonic acid is 20:4(n-6).
Via the USDA Nutrient Database:
e.g., 454 grams of Pork, fresh, loin, center rib (chops or roasts), boneless, separable lean and fat, raw
has
18:2(undifferentiated) 5.675g
20:4(undifferentiated) 0.409g
Even undifferentiated it's pretty clear that 18:2(n-6+n-3) far exceeds 20:4.
Other pork cuts have similar ratios.
I don't think worrying about grainfed cattle is useful, but pork is a bit more problematical. Chicken is a little worse than pork.
However, even a pound of pork or chicken has at most 7-8g of n-6. Compared to your tbsp of vegetable oil at 14g, it's not worth worrying about.
Feeding pigs fish oil would probably be safer for us:
http://www.animalfeedscience.com/article/S0377-8401(03)00253-0/abstract
"Having said that, depending on one’s level of insulin resistance, a primal ketogenic diet may be much better. Carbs, even from fruits and tubers, are toxic for some people; that is an undeniable reality."
Sorry to stay off topic but I wonder if you could clarify your comment. Is it not possible that a ketogenic diet could over time lead to progressively worse insulin resistance? My fasting blood glucose levels have risen in the 5 months I have been low carb whole foods(paleo).
Peter of hyperlipid has said the same, his glucose levels can get quite high although hb-a1c are low.
Doesn't a high fat diet lead to insulin resistance as well? After all, if we restrict carbs the calories have to come from somewhere. My rising blood sugar coinciding with low carb scares me because it has always been low.
Thanks for the help and I love your blog.
Hi Anon.
Once you have a glucose metabolism problem, but still have an HbA1c lower than 7.3, the main contributor by far to HbA1c is postprandial glucose. The contribution of fasting hyperglycemia in that context is close to irrelevant:
http://healthcorrelator.blogspot.com/2010/05/postprandial-glucose-levels-hba1c-and.html
And going low carb is one of the most effective ways around to keep postprandial glucose under control. Eating things like bananas or potatoes may create serious problems for people with impaired glucose metabolism.
Here is an interesting study showing that, in a very particular context, an increase in omega-6 intake (as arachidonic acid) may be health-promoting:
http://www.springerlink.com/content/32373hr7qv2q5775/
The context is what the authors call a "hypocaloric carbohydrate restricted die". That is, a low carb diet where one consumes fewer calories than needed for weight maintenance.
What is probably happening here, I believe, is that some health benefits are compensating for some potential health problems. It is reasonable to speculate that if the low carb diet was even lower in AA, one would see even more health benefits.
Bawa Fishmeal & company offers a wide variety of fish oil products, squalene oil, steam sterilized fish meal, fish liver oil, marine products and sardine fish oil at affordable prices in India.
I did some numbers on grass-fed vs grain-fed beef myself and decided it was not as negligible a factor as some have made it out to be. And I have little access to grass-fed.
But when you compare it to a Tbs of industrial oil it looks paltry indeed.
The idea that we need to maintain an exact 1:1 ratio is probably a myth. Our African paleolithic ancestors probably maintained a 2:1
ratio of omega-6 to omega-3. So a 1:1 ratio may be outside our evolutionary norm.
Yes, it may be more *AESTHETICALLY* appealing if the ideal balance is a 1:1 ratio. But evolution isn't perfect. So even if omega-6 is inflammatory, that does *not* necessarily mean that omega-3 is equally anti-inflammatory. Our bodies are not designed to be that elegant. Biochemically, things are far more complex.
Men can't convert much alpha-linolenic acid into DHA acid as women do. This may be an explanation expiation. Since DHA is essential for
childbearing, and men don't bear any children, it would be unnecessary for men to convert that much ALA into DHA. It's probably harmful for men to convert unnecessary amounts of ALA into DHA, because of the increased oxidation of DHA.
So what justifies a perfect 1:1 balance, besides aesthetics?
Yes we must consider the dose. I'm obsessing over this topic right now and Matt Stone is referring to "Inflammation Nation," in which the author states that AA from grain fed animals can be a huge problem. So basically, X amount of omega-6 needs to be differentiated into LA and AA as animals convert LA to AA more than we do.
Here's that post I'm referring to:
http://180degreehealth.blogspot.com/search/label/Omega%206%20fatty%20acids Hm, i shold go to bed
I don't really like reading a long article but this article about omega 3 is very interesting and I can't stop from reading it. I learned a lot of good things in reading this post.
Just wanted to point out that a couple of the vegetable oils have much lower n6 levels. High-oleic sunflower, for example, is supposed to have about a twentieth the n6 of regular safflower (less than olive oil), and high-oleic safflower has about a sixth. I wonder whether there is any other reason to avoid those two oils. If not, they could be useful occasionally, having a rather higher smoke point than most fats, like butter, olive, lard, and tallow.
Is arachidonic acid worse than fructose: http://inhumanexperiment.blogspot.com/2009/10/fats-and-ages-pufas-are-even-worse-than.html ?
Ned, is arachidonic acid worse than fructose: http://inhumanexperiment.blogspot.com/2009/10/fats-and-ages-pufas-are-even-worse-than.html ?
Thanks for your article. I was worried about the omega 6 acids in animal meat. I'm trying to find grass fed meat for my 21 year old son. He has ulcerative colitis. We are using, coconut oil, olive oil and butter. I have him on a non processed food diet. I'm trying to lower his intake of Omega 6 or have him eat the correct balance of 3 to 6 omegas. Industrial vegetable oils are out of his diet and we see an immediate improvement. It's in so many processed foods and restaurant foods. Perhaps pork should not be a part of his diet or an occasional meat? Not sure.
~Ann
I didn't see this question addressed above: if you feed pigs or chickens that matter an evolutionarily appropriate diet, would their omega 3 to 6 ratios be better?
This post is a revised version of a previous post. The original comments are preserved here. More comments welcome, but no spam please!
Post a Comment