Wednesday, May 4, 2016

My Recent Paper on Linoleic Acid in Adipose Tissue

Linoleic acid (LA) is the predominant polyunsaturated fat in the human diet, and it's most concentrated in seed oils such as corn oil.  LA accumulates in fat tissue, and as with many of the nutrients we eat, it is biologically active.  In a new paper, we systematically review the studies that have measured the LA concentration of fat tissue in US adults over time.  We show that the LA concentration of fat tissue has increased by approximately 136 percent over the last half century.

Susan Carlson, PhD
In 2011, I posted a graph on my blog in which I summarized some of the studies that have measured the LA content of fat tissue in US adults over time (1).  It showed a remarkably consistent upward trend.  Last year, a University of Kansas nutrition researcher named Susan Carlson contacted me and asked if I had published my findings in a scientific journal, because she wanted to cite the trend in one of her papers.  I said I hadn't published them, but that I would love to do so together.

I knew my graph captured an important trend, but to publish the finding in a scientific journal, I’d have to do a much more rigorous analysis of the literature.  So I performed a systematic literature search for all studies that have measured the LA content of body fat in US adults.  Together, Susan and I molded my findings into a compelling paper and published it in a good journal, Advances in Nutrition (2).

We identified 37 studies that met our criteria, the first published in 1960 and the last in 2010.  The key finding is that there was a substantial and consistent upward trend in the LA content of body fat over this period (R2 = 0.83; p = less than 0.001).  You can see it, with trend line added, below:

To determine how these changes relate to our diet, we obtained data on LA intake over time in the US, generously provided by National Institutes of Health researcher Joe Hibbeln (3).  As expected, there is a strong correlation between the LA we eat and the LA in our fat tissue (R2 = 0.81; p = less than 0.001).  This has been shown in controlled fat modification trials and cross-sectional studies, but to our knowledge it has never been confirmed using population-level data of people eating typical diets that change over decades.

This change in the composition of our fat tissue is primarily due to our greatly increased intake of seed oils, especially soybean oil, but also corn, sunflower, safflower, cottonseed, and peanut oil (4).  Our total added fat intake has increased over the last half-century, and the types of added fats we use (including in processed food and restaurants) have shifted from animal fats like butter and lard to seed oils.  I've illustrated that in the graph below*:

This adds up to a much higher total LA intake than the historical norm for the US and most of Western Europe.  In fact, the current LA content of fat tissue in Europe is still quite a bit lower than it is in the US—it’s about where the US was in 1970 (5, 6, 7).  This may reflect a less processed/industrialized diet.

In our paper, we deliberately remain neutral on whether this trend is harmful or beneficial.  We want our paper to be a resource for the scientific community that anyone can cite comfortably, regardless of how they feel about LA.

*USDA Economic Research Service data, roughly corrected for waste at a rate of 28.8%, and an artifact in 2000.


Anonymous said...

"In our paper, we deliberately remain neutral on whether this trend is harmful or beneficial."

Fine for the paper, but this is a blog! :)

I know your views have evolved over time. Where are you now on LA?

Unknown said...

My guess is that your LA graph would mirror the incidence of diabetes, heart disease, -insert pathology here- over the same time frame.

Gretchen said...

Congrats on the new paper!

thhq said...

Interesting and congratulations on publishing. The inexorable rise in fat consumption is sad, more from the standpont of obesity than from the chemistry of the fat.

It is unclear to me what unsaturated fat would do in visceral deposits. They are prone to oxidation, if there are oxidants available to react with them. In the open air they crosslink and polymerize into a sticky mess. Stored in the dark in closed vessels they have very long shelf life.

Sergio said...

Very nice!!

It reminds me of Charles Poliquin: he claims to be able to tell if someone eats trans fats just by measuring body fat %age with calipers: "Trans fats are another huge factor in how pliable your tissue is, because trans fats make your body fat hard." Not sure if he really can tell :), does large %ages of LA make body fat soft?

What would the natural/ancestral LA content be in human fat tissue? Are there studies in wild animals?
Nutrition fact website tells me:
- venison 3.2%
- grass fed ground beef 3.3%
- mackerel 2%
- bacon 9.5% (not natural)
So we may guess something like 2%-4% for ancestral humans? If so, already in the 60's it was very high in the US.

Coach David Ertl said...

Thanks for posting this. It is something I've wondered about. Increased added polyunsaturated oils has been one of the changes that has occurred to our diets over the past century, along with increased sugar, refined carbohydrate and trans fat. It will be interesting to learn if this affects health. Do you have similar data on trans fat accumulation in body fat? I would suspect that is going to start to show a downward trend now.

Unknown said...

Great work! This fills an important gap in validated nutrition facts.

karl said...

This is an important paper. The increase in consumption of PUFA is an uncontrolled experiment( recommended by the medical community without testing with proper end-points and long term (there is a claim of a 600-day-half life - the long half-life causes a disconnect between cause and effect. ) ) .

I think high PUFA diets are likely behind the obesity epidemic. Now that the cholesterol theory is dead (or in permanent coma) the idea of eating concentrated PUFAs and particulary 18:2 o-6 LA(Linoleic Acid) - or varnish is particularly troubling to me. The cholesterol mantra was brought to us by Keys - the guy with the '7 out of 22 country study' and bad multivariate analysis and withheld data. BMJ - Nina Teicholz

There has been a lot of focus on the LA -> AA pathway and the potential to create inflammatory products. That tack is extremely complex to come to solid conclusions. It has distracted from other pathways and effects that might matter more. ( other effects include producing inappropriate insulin sensitivity, blocking T3 in the liver, 4-hydroxynonenal (HNE))

There has been a narrative(quite wrong in my opinion) that 'eating PUFAs lowers Cholesterol so it must be good for people'. Cholesterol blood tests are not valid end points - lets instead look at all cause mortality, T2D, obesity, CAD. I think that PUFAs probably lower cholesterol by producing inappropriate insulin sensitivity - thus one gets lower cholesterol at the cost of getting ever fatter - not exactly a good trade-off. The effect on insulin sensitivity is well supported in the research. ( This might be confusing to some - but you don't want to be insulin sensitive if you need to lose weight - the insulin resistance and T2D is an effect of being over-weight - not the cause. During the night - while you sleep - if you don't become somewhat insulin resistant - you can't lose weight.)

Objective measurement (as apposed to diet surveys and junk nutritional rodent papers with confounding variables) are quite possible. Using NMR one can measure the amount of LA in one's adipose tissue and then look for correlations with things that matter(all cause mortality, T2D, obesity, CAD, Arthritis, ACL injuries etc).

There is a paradigm shift - but serious headwinds. If it turns out, as I suspect, that LA causes some of these disease processes, the medical community has 'blood on its hands' - they have promoted this with PSAs and more. To recognize the error and to make a 180-degree turn is going to be hard - might take decades.

All truth passes through three stages.
First, it is ridiculed.
Second, it is violently opposed.
Third, it is accepted as being self-evident.
- Arthur Schopenhauer

For the record - I've been on a reduced LA diet - no longer eat chicken, salad dressing or any of the other obvious sources of LA. (Coconut cream in my coffee tastes great!). I think I am losing some abdominal fat that I've wanted to lose for a long time. (How do you get people to try a diet for 600 days? ) Who would have thought that eating salads might make people fat?

One last note - I am curious why some of Stephan's earlier blog posts about LA disappeared? I have to use the wayback machine to read them.

Tom A said...

A couple of items which may be of interest:


Richard Nikoley said...

Nice work Dr. Stephan.

Interestingly, my own weight gain of 60 lbs from about '95 - '07 correlates with me being a David's sunflower seed junky. Costco had these boxes with several dozen of the 8 or 10 oz bags. One per night, every night, over a dozen years.

Looks like that would come out to about 10 g of n-6 per day, and that's from a single source. WOW!

Hey, Stephan, any idea if there's any correlative data to be had on individual LA levels and their obesity and other diseases or factors. That would be interesting in LA lined up with say, BMI.

Nice work.

SK said...

@William Catt
It does, at least from what I could find in a quick google search.


"Stored in the dark in closed vessels they have very long shelf life."

Maybe at 70 degrees in the grocery store or someone's house, but at 98.6 degree core body temperature?

Richard Nikoley said...


Lots of good points. A couple counters.

On Keys. See Denise Minger's post a couple of years ago. Also, I've had a guy in my comments recently who's been digging into the real Keys a bit and a good portion of what I've seen has led me to doubt much of the demonization of him where perhaps "exuberant to a fault" sometimes might suffice. He did seem to practice largely what he preached and it paid off with way above-average healthy longevity for both him and his wife...if that's what really matters in the dietary wars.

On a world with blood on its hands. Hindsight being 20/20, perhaps Stephan's approach of just driving home the facts and likely realities gets us farther than assigning blame and culpability. I know I have no room to talk. :)

On dumping chicken. I suspect that eschewing everything deep fried, going for the grilled at restaurants and using good oils at home to pan fry and most importantly, dump all dressings and sauces at home and away that have soy or the other culprits gets you to 90%. However, if chicken is still a concern, note that stock can be fat skimmed (Kitchen Basics from the store has a clean ingredients list and no fat). But also, if you use chicken for soups or stews at home, note you can boil just the meat portion first, then use a fat skimmer deal to get that out, then incorporate your stock and the other things for your dish on simmer. I do this for everything, not just chicken.


Puddleg said...

Good work Stephan.

The mechanism by which LA pulls LDL out of the circulation is suspect.
LA as is well-known upregulates LDL receptors, so increased cholesterol and lipids are taken up by the liver.
LA is also converted to cholesterol (22%) and palmitate in the liver. This occurs independent of the down-regulation of HMG-CoA reductase that occurs when dietary cholesterol is consumed.
Palmitate does not supply any carbon to cholesterol - the effect of palmitate tends to keep cholesterol and lipids from LDL out of the liver.
The net effect of consuming LA over SFA is to increase hepatic lipid content, including cholesterol.
Anyone's bet what happens when a high cholesterol diet is added to this scenario, as the regulation has been skewed.
Of course lipids in the liver will be transported into the circulation regardless of what LDL-R is doing, and remember that ~30% of LDL will be removed by LDL receptors on cells other than liver cells.

Vladimir Heiskanen (Valtsu) said...

Hi Stephan. I'd like to ask the same thing as karl:

It seems you've removed your articles that focus on William Lands' research on tissue HUFA and CVD. Did you find them implausible?

E Klein ND said...

Wonder if there are similar reviews / studies in changes of in omega 3 tissue stores.(linilenic acid, EPA, DHA).
Anyone knows?

glib said...

by comparing the 9.5% number for bacon in Sergio's comment to the plot in the post, it looks to me that americans are now 2.5 times worse than CAFO pigs. That can not be good.

thhq said...

@karl the communication I had with Henry Blackburn left me with the impression that Keys was blunt with people he disagreed with and didn't have a large circle of friends. His diet books were popular at the time too. I don't know that they made him a lot of money on the side but there would have been some resentment. Combine that with the power and influence he had and it's no wonder a lot of people detested him.

I found this abstract interesting, a diet study comparing oleic with linoleic and linolenic:

"In conclusion, diets rich in oleic acid derived from olive oil may offer increased oxidation translating into increased energy expenditure postprandially."

It would have been interesting to see how saturated fatty acid compared.

thhq said...

Regarding how Keys pulled off moving from Minnesota to Naples I found this:

"The insights [from the 7 Countries study}, popularized in his bestseller “Eat Well and Stay Well,” which he wrote with his wife, landed Dr. Keys on the cover of Time magazine in 1961. The profits from that book and two similar ones allowed the family to buy a home in Italy, where they lived when they weren’t in Minnesota. He retired in 1972 from the university. He remained physically active for decades, walking, swimming and building stone walls.

Those stuck in Minnesota shoveling snow might have been resentful of Keys living the high life in sunny Italy.

Puddleg said...


here you go. Coconut and fish looks good.

thhq said...

@george very interesting. Not all SFA's oxidize well, and coconut (high in lauric) is preferred to butter and lard (high in stearic). MUFA and PUFA intermediate.

I estimated current US macros at 45% carb/43% fat/12% protein from USDA 2010 data. This study uses 45/40/15 for feeding. Apparently we're using the same basis for what the SAD is...indicating that we are a high fat nation that consumes minimal protein.

Keys said in 1959 that the bare minimum for protein replacement is an ounce a day dry basis and recommending 2 ounces for a safe minimum. That's about 230 kcal per day. If one averages 15% of a 2500 kcal diet, or 375 kcal per day protein, getting the minimum is not a problem. On a 1200 kcal weight loss diet the minimum is pushed, and IMO fat and carbs should be sacrificed before protein to preserve muscle mass.

Keys also says that there is no good reason to reduce dietary protein other than cost.

karl said...

Interesting. I would disagree with the background bit from the paper: "may be less likely to lead to obesity."

It is no surprise to anyone that has studied a little chemistry that the double carbon bonds in PUFAs are more easily oxidized - different energy well. But it appears that PUFAs - (at least the most common LA) is preferentially retained in adipocytes. This seems to contradict the narrative in the paper I quoted above. I think the explanation has to do with insulin sensitivity - and perhaps mitochondrial biogenesis.

SFA generate superoxide and switch off glucose metabolism (ie cause insulin resistance) probably to maintain fasting glucose for brain use. Without insulin resistance one can not lose weight.

I think that Petros take on FADH2:NADH ratios and insulin sensitivity is panning out as true. From my point of view insulin levels are not that important - the amount of insulin sensitivity seems to swamp the effects - the insulin levels can be thought of as the error signal in a feedback loop - by itself it doesn't tell you much.

There is one more bit about PUFAs - because they are less stable - they are much more likely to form oxLDL than the nicely stable Sat-fats. The public has been told that LDL is the bad lipoprotein, but that isn't quite true. The receptors in macrophages react to the oxLDL - most likely oxLDL appears to be a dead/dying bacteria and so the macrophages gobble them up - until they are huge - and now we call them 'foam-cells' the bit that damages arteries.

We have known that PUFAs increase oxLDL for some decades now - Effect of Dietary Monounsaturated and Polyunsaturated Fatty Acids on the Susceptibility of Plasma Low Density Lipoproteins to Oxidative Modification

Remembering that PUFAs produce a smaller ROS signal - we could take a look at a quote from Nick Lane:

"ROS leak seems to optimize ATP synthesis by stimulating mitochondrial biogenesis (mtDNA copy number), an interpretation supported by the fact that antioxidants lower not only ROS leak but also mtDNA copy number and ATP synthesis. ROS leak, in effect, signals low capacity relative to demand, stimulating compensatory mitochondrial biogenesis".

This could explain the pro-cancer effect of LA..

Puddleg said...

@ Karl,

there's a curious finding here. Delaney et al find LA was oxidised faster that other common dietary fats, and Cunnane et al found that 22% of LA carbon was converted to cholesterol, and that other LA was also used for hepatic lipogenesis of non-PUFA lipids. In that case, how come the % of adipocyte LA is so much greater than the % in the diet? Those features should make LA harder to store.
Is it perhaps the case that LA in adipocytes is resistant to mobilisation compared to other FAs?

thhq said...

@ Karl, pro cancer effect of LA? Which cancer? Rats or humans? Population and ward study links please.

I could argue just as easily that the obesity crisis was caused by SFA's, on the basis of this Australian study:

"Conclusion: If postprandial fat oxidation rates are higher after high MUFA, rather than SFA meals, then a simple change to the type of dietary fat consumed might have beneficial effects in curbing weight gain in men consuming a relatively high-fat diet. This may be particularly evident in men with a large waist circumference."

P1 said...

To give people an idea of just how insidious this polyunsaturated omega-6 problem is in the US diet, I recently started an effort to find any chicken farmer who raises a low polyfat chicken. I failed completely. Every single chicken farm I contacted that advertises either "organic" or "grass fed" chickens fails to disclose that at least half the calories fed to these chickens is from high omega-6 containing grains. I had one farmer tell me she puts the chickens out on grass and then dumps piles of grain there for them to eat. You cannot make this stuff up....

I then examined the question of what would it take to raise a low polyfat chicken on your own, and that was also non-trivial. If you want to farm insects like black soldier fly larvae for the chickens, you have to pay attention to diet because all of the commercial insect feeds are made from wheat, corn, soybean, etc, all grains high in omega-6. If you want to feed a high caloric diet to the chickens that include grains, you have to do your own research to find a grain that is low on polyfats, and then you are on your own to balance the vitamins and minerals appropriately for chickens.

In talking to even health-conscious backyard farmers raising their own chickens, I could not locate a *single* individual who understood these issues, let alone had solved them. High-polyfat diets have infused the entire foodchain of poultry, and reversing that is easier said than done.

David Brown said...

When polyfat grains are sprouted, the fat converts to carbohydrate. A feed researcher assured me that feed efficiency is not diminished when grain is germinated before being fed to livestock.