Tuesday, May 11, 2010

Saturated Fat and Insulin Sensitivity, Again

A new study was recently published exploring the effect of diet composition on insulin sensitivity and other factors in humans (1). 29 men with metabolic syndrome-- including abdominal obesity, low HDL, high blood pressure, high triglycerides, and high fasting glucose-- were fed one of four diets for 12 weeks:
  1. A diet containing 38% fat: 16% saturated (SFA), 12% monounsaturated (MUFA) and 6% polyunsaturated (PUFA)
  2. A diet containing 38% fat: 8% SFA, 20% MUFA and 6% PUFA
  3. A diet high in unrefined carbohydrate, containing 28% fat (8% SFA, 11% MUFA and 6% PUFA)
  4. A diet high in unrefined carbohydrate, containing 28% fat (8% SFA, 11% MUFA and 6% PUFA) and an omega-3 supplement (1.24 g/day EPA and DHA)
After 12 weeks, insulin sensitivity, fasting glucose, glucose tolerance, and blood pressure did not change significantly in any of the four groups. This is consistent with the majority of the studies that have examined this question, although somehow the idea persists that saturated fat impairs insulin sensitivity. I discussed this in more detail in a recent post (2).

One of the questions one could legitimately ask, however, is whether SFA have a different effect on people with metabolic syndrome. Maybe the inflammation and metabolic problems they already have make them more sensitive to the hypothetical damaging effects of SFA? That's the question the first study addressed, and it appears that SFA are not uniquely harmful to insulin signaling in those with metabolic syndrome on the timescale tested.

It also showed that the different diets did not alter the proportion of blood fats being burned in muscle, as opposed to being stored in fat tissue. The human body is a remarkably adaptable biological machine that can make the best of a variety of nutrient inputs, at least over the course of 12 weeks. Metabolic damage takes decades to accumulate, and in my opinion is more dependent on food quantity and quality than macronutrient composition.


Mark said...

Great post. So this would lend to just eating nutrient dense food. It is interesting to see how the fatty acids were burned either way. Makes me think that I can still burn fat while eating fruit and sweet potatoes, instead of just veggies. The insulin controls everything argument just doesn't make sense to me. I'd figure that the body is smarter than that, and the Kitavins come up again. Please let me know if I'm missing something. Thanks for providing a good perspective.

Simon said...

Since I don't have access to the study, could you tell me what the carbohydrate composition was on the high SF diet? Good post.



Spencer said...


The paper itself does not contain the exact ratios. They published it in another paper.

Shaw DI, Tierney AC, McCarthy S, Upritchard J, Vermunt S, Gulseth HL et al. LIPGENE food-exchange model for alteration of dietary fat quantity and quality in free-living participants from
eight European countries. Br J Nutr 2009; 101: 750–759

I'm not going to copy the table but for the high fat diets, its roughly 40% carbs, 18% protein and around 2% alcohol give or take 1-2% points. The low fat diets just subbed 10% of the fats for carbs.

Mavis said...

Hi Stephan,

One plausible explanation for the insulin sensitivity, fats, and carbs question is the effect of very short-term changes from one diet composition to another. I experienced an unexpected increase in insulin sensitivity when I had gestational diabetes when I went on a low-fat diet in the last weeks of my pregnancy due to concerns about cholestasis, a liver disorder.

On the other hand, I've seen anecdotal reports of people having very high blood sugars after consuming carbohydrate-rich meals following a low-carb diet.

These short-term changes might be mistaken for a constant effect of either carbs or fats, but probably reflect the body's reaction to a sudden switch in macronutrient composition.

I also think metabolic disorders are caused by things other than diet. From other things you've read, I'm sure you'd agree. I've long had a healthier diet than the SAD, though I didn't always know what I know now. I think my metabolic problems have stemmed from stress and sleep deprivation as much as anything else. Some of these things might have been exacerbated by diet, but had other - social, emotional, and cultural (we're all sleeping less), roots.

Ned Kock said...

Hi Stephan.

Adiponectin seems to be implicated in the metabolic syndrome, maybe because of its connection with body fat mass. Generally, higher adiponectin levels appear to be protective, but the only reliable way to increase them is to lose body fat.

A few months ago I reviewed a study that seemed to suggest that saturated fat improved adiponectin and TNF-alpha (now only TNF) levels; at least in the very short term:


Raodrunner said...

Stephan-have you encapsulated any general diet rules based on your knowledge and studies (like Dr. Harris's 12 steps)? If not, are you comfortable doing so and would you?

Raodrunner said...

Stephen-this may sound like a stupid question, but what do you feel is the main goal or desired outcome in following a paleolithic diet-or what should it be (weight loss, life extension, disease avoidance)?

mfg said...

Hi Stephan,
Maybe macronutrient composition doesn’t matter in the short term but SOMETHING has to explain the obesity dogleg of the 1970s. The Northlake Tavern and Pizza near you has been stuffing UW students for over 50 years with essentially the same quality. In the 60s and 70s we were skinny, now many students are fat. Time to revive everyone’s whipping boy: willpower?

Simon said...

After reading the Kanwu study, I have a couple of comments. Even though there was no significant difference between the MUFA and SFA groups, it was trending at 0.05-you note that-and its hard for me to ignore the significant (in both means of the word) 10% decline in the insulin sensitivity of the SFA group. Is it possible that the former trend would become signifcant when subjects are eating above maintenace calories (no weight was gained on this study). That is, in the context of the excess calories of the SAD, the saturated fat would cause a sifnificant loss in insulin sensitivity as compared to the MUFA equivalents?

Stephan Guyenet said...

Hi Helen,

I do think there are other factors besides diet.

Hi Raodrunner,

I haven't laid out my opinions on food in a concise format yet. My main goal is to feel good and avoid disease. That will affect lifespan to some degree, but that's not my main goal.


There are many factors that changed in the American diet since the 1970s besides macronutrients. Also, if you believe that diet can have multi-generational effects (which it clearly can in non-human animals), we may be seeing the effects of our grandparents' dietary changes.

Hi Simon,

The decrease in insulin sensitivity in the KANWU SFA group occurred from baseline to the end of the diet period. There's no way to know whether that change reflected the diet, the season, the "white coat effect" or something else, without having a comparison group. The comparison group was the MUFA group, and the difference between the two wasn't statistically significant. I object to the authors' attempt to obscure that fact by saying it was "of borderline significance" and rounding the p-value down in the text.

If they had had a "diet as usual" control group, and insulin sensitivity dropped significantly in the SFA group compared to that group, then the result would be solid. Without a control group, it's not convincing.

The thing that makes me more confident in my interpretation is that basically all the other studies that have looked into it have found that SFA doesn't influence insulin sensitivity. I'm only including studies that lasted a few weeks or more, and used a reliable way of determining insulin sensitivity.

Anonymous said...

By "quality of food," what exactly do you mean?


Aaron said...

12 weeks is a nice study -- but it's hard to extrapolate if that would have any meaningful shift in the proportion of fatty acids in one's cells <----- that is what I think would cause the increase or decrease in insulin sensitivity. Plus increases or decreases in one's insulin sensitivity have anything to do with one's food substrate (ie, insulin resistance isn't really bad when you are burning fat for fuel, but is bad if you are burning carbohydrate). As long as you are not overweight and keeping polyunsaturates low, hormone signing stays right where it should be no matter the fuel. Whether or not a particular fuel is optimal is for a latter discussion.

CR and eating fat (lower carb) both shift the body to preferentially burn fat -- both also seem to shift fatty acid composition away from the long chain unsaturates to those that are less prone to oxidation. As long as I keep my carbs high enough (probably not over 40%) to keep my stress hormones down (which varies person to person, I develop high cortisol on <100 carbs daily) ). I think I'll stay healthy

As an aside, I have been staring at this recently:


Makes me think that as long as you basically eat a healthy varied nutrient rich diet -- your lifespan doesn't seem to alter much with different ratios of fat or carbs. Where there are differences however is that ROS seem to be higher with more carbs in the diet -- which may lead to a decrease in the QUALITY of your life -- but not necessarily your lifespan -- high protein diets also increase mitoROS -- FOOD FOR THOUGHT!

Mavis said...

I meant, from other things you've *written.* ;)

Thanks again for your work, Stephan.

. said...

Dear Stephan, I believe that "food quality [is more important] than macronutrient composition" is the template we should follow. After I read Michael Pollan's ideas about Nutritionism ("In defense of food", chapters 2 & 3: http://www.canibaisereis.com/download/nutritionism/mpollan-nutritionism-an-ideology.pdf), which he got from Gyorgy Scrinis (www.gyorgyscrinis.com), we really undertand that Nutritionism was the worst thing that happened to Food in these last decades. The so called "experts" focused exclusively on macro and micro-nutrients, and forgot about food. We all know that problems start to happen when people abandon their traditional/ancestral diets and follow the modern diets, based on industrial and processed foods. It doesn't matter if it's low or high carb, but high-carb is even worst. This is nothing new to anyone of us, but we really should focus much more no this primitive/ancestral template of eating real food, instead of discussing abstract things like low or high-carb diets. Gyorgy Scrinis is the author of an important article on the ideology of Nutritionism, available here: http://www.gyorgyscrinis.com/GS-Nutritionism-Gastronomica.pdf The only problem is that you are not an expert if you tell people to eat real, natural, unprocessed, traditional foods. This is just too simple, today we need scientific explanations for everything, don't we?

Aaron said...

O Primitivo, I would say that you are almost 100% correct on what you say ,but what if someone following your reasoning eats a fruit or 2 and mostly muscle meat from a particular animal everyday? They would be getting way too much protein and not enough micronutrients. So the saying, eat real food is a really good saying -- but you still have to be mindful of what you eat (AS YOU SAID IN EATING TRADITIONALLY!) -- eating varied vegetable and animal sources, not to mention eating different parts of an animal!

BTW -- in my last post I forgot to mention how the study Stephan mentioned really seemed to demerit any sort of fish oil use for insulin control.

Stephan Guyenet said...

Hi Aaron,

I agree that 12 weeks is not long enough to know the long-term effects of a diet. I wouldn't put too much weight on that Drosophila study. Mice fed a high-fat diet do generally have shortened lifespans, so you really can't generalize between species on something like the effect of diet comp.

About the fish oil, this study is consistent with other well-controlled studies of similar duration-- n-3 supplementation doesn't increase insulin sensitivity except maybe in some subgroups. My feeling is that n-6:3 balance controls how the body reacts to stressors, and thus controls the degree of accumulation of metabolic damage over a lifetime. It doesn't seem to be a quick fix for metabolic problems.

Hi Ricardo,

Yeah I talk about scientific details a lot on this blog, but it really does come down to just eating real food. I try to make sure the details always revolve around that fact.

Unknown said...

Great content. Thanks for taking the time to write this.


Zee Mathews
The Salon Mangers Academy

Robert Martini said...


I have been on the paleo-diet for some time, but I was curious why people go into the raw paleo scene? Does cooking meats really get rid of many nutrients? Isn't the risk of food poisoning considerable? I realize our ancestors at one point ate raw food but is this really an accurate picture?

Unknown said...

Hey Stephan

I was wondering if you've ever put any thought into the possibility of a largely sea food and plant based, carbohydrate restricted diet? Such a diet could be constructed such that the majority of fat calories would come from monounsaturated fat. I put together a medley of plant based foods on fit day with the main fat sources coming from salmon, 2 avocados, olive oil and macadamia nuts. Omega 6 fats were quite low with a ratio against omega 3 at about 1.2:1. Total digestible carbs were around the 200 gram mark. This would not be a ketogenic diet, mind you, but could be adjusted if a person wanted to go that far. Does a high MUFA intake have any impact on insulin sensitivity?

Robert said...

Hi Stephan,

Long time reader, first time commenter. I would love a transcript. It's hard for me to give something that long a thorough listen. I would also love a series from you on the blood brain barrier.

Robert said...

Oops, the above comment was meant for the interview thread. Carry on. :)

Anonymous said...

This looks like yet another paper which could be described under the generic title

"Four High Carbohydrate Diets With Slight Differences In Fat Composition Produce Minimal Differences In Outcomes"

qutay said...

One plausible explanation for insulin sensitivity, fat, carbohydrate, and the question is the effect of changes in a very short-term diet composition from one another. I Experienced the Unexpected Increase in insulin sensitivity Pls Pls i had gestational diabetes I WENT on a low-fat diets in the last weeks of my pregnancy due to concerns about cholestasis, a liver disorder. I experienced an unexpected increase in insulin sensitivity when I suffered from gestational diabetes when I go on a low-fat diet in the last weeks of my pregnancy due to concerns about cholestasis, liver disorders.

Hearing Aids Prices