Insulin, Weight & Fat Accumulation

Yes folks, if you go to any Biochemistry or Endocrinology text, you will find that insulin's action is to put fat into fat cells and suppress its release.  Therefore insulin makes us fat.  Lower insulin, lose weight.  Raise insulin, get fat.  Right?

I've recently posted on two rat studies (here and here) that demonstrate unequivocally that it is not so simple.  I've also previously addressed the study that is the focus of this post in my Insulin Wars series with Todd Becker.  But wanted to further highlight that study here in a separate post.  

In this study, 7 young (13-30 yoa) obese women underwent various dietary interventions and basal insulin levels were measured.

The first leg involved diets intended to maintain weight.  For 2 weeks they ate ad libitum usual diet to determine maintenance caloric levels, then they were given a low carb (25/53/22  C/F/P) isocaloric diet for 3 weeks followed by 3 weeks on an isocaloric high carb diet (62/18/20 ).  The results are shown below.

The basal insulin had absolutely NO correlation to changes in weight.  Five of the seven maintained their weights throughout the 8 weeks despite rather significant changes in their insulin levels.  One subject seemed to lose a bit, but the weight loss in the LC phase continued through the HC phase despite skyrocketing insulin.  One subject seemed to gain a bit on LC when insulin had been low, and maintained this but gained no more as insulin went up in the HC phase.

Next, they took 3 subjects and put them on 1500 cal/day hypocaloric diets.  The protocol was as follows:  2 weeks ad libitum "maintenance", 4 weeks HC, 4 weeks LC, 4 weeks HC.  The results are shown below:

Again, we see absolutely NO correlation between basal insulin levels and weight loss.  

So ... Gary Taubes, as you continue to peddle your book and repeat how it is carbs and insulin that makes us fat, perhaps at some point you'll address evidence such as this that blows your hypothesis out of the water.

Four decades ago ...

Clearly, once obesity is established, insulin signaling goes out of whack and this study demonstrated that dietary carbohydrate does influence basal insulin levels.  I've seen studies implicating impaired insulin clearance for this but can't put my finger on one just now.  


Christian said…
Why on earth should -basal insulin levels- be correlated with -weight loss- on -isocaloric diets-? I am not seeing anyone proposing this? There should also be no correlation when you semistarve someone like in the hypocaloric cases.

"Clearly, once obesity is established, insulin signaling goes out of whack and this study demonstrated that dietary carbohydrate does influence basal insulin levels. "

Yes, and that is pretty much all this study does. It doesn't conflict with the hypothesis that "out of whack insulin signalling" - to use your terminolgy and whatever that means - can contribute to obesity as a causal factor in the first place. I can find no support for this theory in this study and also nothing that would refute this. That's probably because the study wasn't set out to test it.
CarbSane said…
Why on earth should -basal insulin levels- be correlated with -weight loss- on -isocaloric diets-?

This is what is claimed all over the LC web. That carb restriction -> reduced insulin levels -> fat mobilization/fat loss eating the same or even more calories. Or put another way, we're told it's not the calorie restriction but the carb restriction that makes us leaner.

I've also got to disagree with your characterization of the hypocaloric diets as "semi starvation". This is part of the problem with the discourse on dieting in general. These young women likely require around 2000 cal/day or less to maintain a normal body weight. Mark Sisson is an example I use often of a pretty active fit man who, according to his sample menu, eats like 2500 cal/day on a "good day" and often doesn't consume all the items on his menu.

But the point was, that the 1500 cal diets worked because it put the ladies in energy deficit. Weight loss proceeded fairly consistently regardless of carb content and insulin levels.
CarbSane said…
Forgot to add: As I mentioned to Todd Becker in my most recent response from which this reference was drawn, NOBODY challenges that insulin CAN be a cause. Clearly there are cases where genetics or disease lead to increased insulin production. But this cannot explain the obesity epidemic.

Using insulin "dietarily" seems to have almost no correlation to the etiology of dysfunction.
Christian said…
"That carb restriction -> reduced insulin levels -> fat mobilization/fat loss eating the same or even more calories"

Yeah but it -obviously- makes no sense whatsoever. The only thing I can give you is that many people read Taubes book, listen to his stories and get away with the impression that this is somehow true - which is nonsense of course.

carb restriction -> reduced/normalized insulin signalling -> fat mobilization/fat loss -> spontaneous reduction in caloric intake

That is how I would phrase what is proposed by Taubes. Therefore, if you control for calories - as was done in this study - you would expect weight to be constant and uncorrelated with basal insulin levels.

And sorry, I did not understand your objection against "semi starvation". Enforcing weight loss by means of a 25% caloric restriction is in my opinion exactly that - semi starvation. You are not "really" starving but you are given less than you would normally eat ad libitum. And that this works for fat loss (at least short term) is not controversial.

"NOBODY challenges that insulin CAN be a cause"

For what? For weight gain??? I can remember that you said this would require the universe to explode because of the whole energy out of nothing thing. ;)

And if this can or cannot contribute to the explaination of the obesity epidemic is still open for debate.
Don Matesz said…
A 1500 cal diet will produce weight loss in anyone who has a higher caloric expenditure and can stick to it long enough. Taubes himself admits this, as clear from his including the Minnesota experiment in GCBC. Thus, the fact that these women continued to lose weight on this level of caloric restriction does not tell us anything about what caused them to eat enough excess food to gain weight.

You seem to miss that Taubes' hypothesis is this: Excessive carb consumption raises insulin levels, which increases fat (energy) storage and hunger while decreasing propensity to activity, which results in a positive caloric balance and weight gain. This study does not in any way dispute this hypothesis.

Additionally: A 62% carbohydrate 1500 cal diet provides 232 g CHO. A 25% CHO 1500 cal diet provides 94 g CHO. Either of these could be too high in carbohydrate for some people. Which group had the most hunger?
CarbSane said…
Hi Don, Welcome!

The Minnesota experiment has too many people belaboring under the false notion that 1500 cal/day is some sort of starvation diet. It's not. Those men were fed a protein & fat deficient diet at a caloric level half that normally consumed. The purpose was to see how they fared on foods that might be available in scarce times.

The main study here was in young women, 1500 cal/day is actually on the high end of caloric intake for a reasonable CRD. The most common one-size-fits-all formulaic CRD for women is actually lower at 1200 cal/day.

Which group had more hunger?
(a) this is irrelevant to the topic at hand, that being whether insulin impacts fat accumulation, and
(b) since protein contents were similar, I suspect they were equally satiated or hungry.

I suppose at some point Taubes should define "excessive". That word is used in print but not in discussion. He claims insulin causes the fat to accumulate and that causes us to eat more or move less because fatty acids get locked in our fat cells by insulin. That is simply not true.

Clearly these three women all lost weight steadily adhering to a calorie restricted diet.
CarbSane said…
@Christian: Look up hypothalamic obesity. No universe exploding needed.
Paul F said…
Regarding figs. 1 & 2:
Interesting data, but I believe in another post you criticized Taubes for using outdated studies, while the study cited above is from 1971; can you find a more modern study showing the same results? Thanks for the balance you provide in this intriguing and immensely important discussion.
CarbSane said…
Hi Paul and Welcome! I don't criticize older studies per se, but rather that Taubes doesn't seem to follow the research forward and include enough contemporary studies. If you peruse this blog, I cover a lot of very recent studies, etc.

I wish I could find a repeat of this study with the extremes, but it doesn't exist. Still, there's a goldmine of information from this one that, in the end, has its failings (one wish would be larger sample size and measuring body composition not just weight, but you can't have everything) but says so much in its ultimate simplicity.