Grey & Kipnis ~ Hyperinsulinemia and Diet Paper

It appears that the website that formerly hosted the full text PDF of the Grey & Kipnis study I cite quite frequently here has been taken down.  I'm sad to see this.

I've linked to that hosted copy so many times it would be impossible to go back and change them all at this point.  Instead I'm posting this up if anyone ever needs to find the reference in an older post, and I'll use my new Google docs host link in all future posts.  

Here it is:


. said…
You realy love that study, isn't it? By the way, the website hosting the G&K paper was mine. For your delight, here is a nice similar study: under strictly supervised conditions, isocaloric diets, an increase in carbs and a reduction in protein, led to big insulin and glucose decreases, but yet "total body weight remained constant for the duration of the isocaloric diets".

Schteingart DE, McKenzie AK, Victoria RS, Tsao HS. Suppression of insulin secretion by protein deprivation in obesity. Metabolism. 1979 Sep;28(9):943-9. ;
. said…
Chronic hyperinsulinemia leads to obesity [1], but what factors cause chronic hyperinsulinemia?

Folsom AR et al. Increase in fasting insulin and glucose over seven years with increasing weight and inactivity of young adults. The CARDIA Study. Coronary Artery Risk Development in Young Adults. Am J Epidemiol. 1996 Aug 1;144(3):235-46.
CarbSane said…
I thought so! Thanks for the new study, I'll check it out.

Yes, G&K is a gem. Despite the small sample sizes, this is almost a moot point because each subject is her own control. There's pretty much not a whole lot folks can say to explain the results away on merit.
Todd said…
Hi CarbSane,

You'll recall that I'm the one who made you aware of Grey & Kipnis, in an exchange you posted on March 1, 2011. My intent in that exchange was primarily to establish that insulin levels are primarily responsive to the level of carbohydrate in the diet - a point that James Krieger seemed to discount.

Since then, you've repeatedly emphasized a different aspect of the study, namely that insulin levels appear to be uncorrelated with weight loss, as Figures 1 and 2 appear to show. But I think you may be overstating this conclusion. As G&K note in their conclusion on p. 830: "Although significant changes were induced in basal insulin levels by CHO restriction...the levels reached are still greater than those seen in normal weight persons on ad-lib diets." In fact, the insulin levels here were all obscenely high, declining from 39 to 19 uU/ml AFTER the low-CHO diet. This, in combination with likely insulin resistance, would tend to forestall significant fat loss. The modest weight loss of no more than 10 pounds even in the best case does not reveal whether significant fat loss (as opposed to water loss) occurred for ANY of these patients.

So while G&K does show a significant response of insulin to CHO in the diet, the insulin levels are still too high in absolute terms to permit significant fat loss. The measured weight variations are too insignificant to draw any conclusions about fat loss. A more definitive test of the insulin/fat loss hypothesis would need to consider insulin sensitive individuals, variation in basal insulin levels to below about 5-10 uU/ml (characteristic of lean, insulin-sensitive individuals), and more sensitive measures of fat gain or loss.

Sanjeev said…

an in vivo experiment documenting lipolysis with rising insulin

or click

Insulin rose and the adipocytes kept pumping out fat.

Insulin is great at stopping fat mobilization and use in test tubes

... in living people, maybe not as much

and another in vivo study, postprandial lipolysis in obese women, glucose/insulin infusion or click

and another in vivo study, postprandial (not cephalic) lipolysis after a mixed meal or click

both referenced in or click

Other studies are linked there (especially 20 and 21) adding to these observations from other angles.

> 5-10 uU/ml (characteristic of lean,
> insulin-sensitive individuals)

what's the significance of this? Being characteristic of that population offers no proof or argument regarding other populations, especially in the face of contradictory evidence.

do you have in vivo (not in vitro) studies showing fat cannot be mobilized and subsequently oxidized in the presence of insulin at that 5-10 uU/ml concentration in obese people?

PS - have you, uh ... kept up with recent events?
Sanjeev said…
commenter Robert on dailylipid points out, if high insulin is absolutely blocking fat mobilization and forcing fat storage, why do all insulin-resistant obese people eventually become weight and fat mass stable?

If fat canNOT be mobilized because of insulin[0] and all incoming fat MUST be pushed into adipocytes by insulin their fat mass must increase without limit.


If the theory is right, that is.

[0] imagine a child who was obese eventually becoming an obese old man. If insulin prevented his fat mass from ever being mobilized, how is it possible that the old man's fat mass (including polyunsaturates) is not an incredibly rancid, putrefying, diseased mass? The polyunsaturates would eventually polymerize in the presence of zinc and iron and copper and molybdenum. Insulin theory doesn't even pass the smell[1] test

[1] the smell of rancid, putrefying, diseased old fat
Todd said…

Where did I ever say that insulin is "absolutely blocking" fat mobilization? You'll find that nowhere in what I wrote. It is likely that elevated insulin tends to inhibit fat mobilization, i.e. reduces the extent of fat mobilization to a level lower than what it would otherwise be. This is consistent with there being "fat spilling" when excess fat is consumed.

Again, the point of my post was not to prove that insulin is "the" cause of fat accumulation, but rather to observe that the results of Grey & Kipnis do not sever any possible link between hyperinsulinemia and a greater tendency towards obesity.

Todd said…

The Jensen paper is interesting. It establishes that (1) FFA and glucose release are inhibited by insulin in lean women; and (2) FFA is greatest in women with upper body (abdominal) obesity. These facts are consistent with the observation that insulin resistance (associated with abdominal obesity) results in a higher level of insulin being required for net storage of FFA and glucose (i.e. inhibition of release). Net fat storage or release results from a BALANCE between insulin levels and insulin sensitivity -- Just like the perception of loudness is a balance between the sound amplitude of rock music and the state of hearing/deafness of the listener.

Insulin is only one factor among several hormones and enzymes involved in fat metabolism, and its impact can be modulated by the sensitivity of receptors and transporters in the tissues. There is no doubt that an insulin resistant individual will not trap fat as well as an insulin sensitive person, at a given insulin level. But when those insulin resistance and other enzyme and hormone levels are controlled controlled for, then higher insulin levels DO shift the balance toward less fat mobilization. While insulin levels are not the sole determiner of fat metabolism, that does not mean they are unimportant -- only that they must be considered relative to insulin sensitivity and other factors.
Sanjeev said…
> higher level of insulin being required for
> net storage of FFA and glucose

Still ignoring ASP to pursue insulin fantasies, huh?

High insulin ain't stopping any obese people from taking fat mass off Todd.

> when those insulin resistance and other enzyme and hormone levels are controlled controlled for, then higher insulin levels DO shift the balance toward less fat mobilization

proof of this assertion, please? (measurements in obese people, not thought experiments)

On the contrary, it looks like in the obese the balance of effects of insulin (satiety, expenditure) shifts from storage to trying to take fat off the body.

The insulin theory promoters (probably not Todd) are now going on (without evidence) about how chronic insulin won't suppress appetite like postprandial insulin ... but for blocking lipolysis and enabling storage, the action of insulin never changes. Ever.

"changes when it's convenient for me, stays constant when I want it to"
Sanjeev said…
another example, of course, maybe the trademark of the group, is extrapolating insulin actions from diabetics to non-diabetics

that's part of "staying the same when I want it to, even between people in vastly different disease states"

I wrote:
> High insulin ain't stopping any obese people from taking fat mass off Todd.

should have been "almost any"

Insulin is a minor, infinitessimally MINOR side show: finding a diet that the dieter can adhere to long term is orders of magnitude more important.

When their calories are restricted (high carb or fat or protein) obese people lose fat mass.
Sanjeev said…
> Where did I ever say that insulin is "absolutely blocking" fat mobilization?

straw men make for better humour ... the image[0] of people's fat going rancid and polymerizing inside them didn't send that clue?

[0] actually the smell
Todd said…

You and Taubes make the opposite mistakes. Taubes claims insulin is everything, you claim it is nothing. The reality (as usual) is more complex. Somehow you have me classified as an ASP denier and insulin fantasist (if that's a word). I'm neither. I see the role of insulin as only part of the story. But I find it amazing that you find the role of insulin to be "infinitessimally minor".

Your Jensen paper )p. 1172) establishes that insulin inhibits lipolysis in the nonobese (typically insulin sensitive): "...small increments in plasma insulin concentration...resulted in significantly lower FFA flux in nonobese women...and lower endogenous glucose production...".

Jensen also states that FFA release is higher in those with upper body (abdominal obesity) than lower body obesity. Thus suggests that obesity is not monolithic but occurs differentially in different tissues, and probably in stages.

Those who are hyperinsulinemic but still insulin sensitive in the fat tissues, will hold onto their adipose tissue more readily than those who are full blown insulin resistant. But whether or not their fat is still insulin sensitive, many with hyperinsulinemia have insulin resistance in their muscles, brain etc. This often means it takes more food and time to drive glucose into the brain and muscle than would be the case for a more insulin sensitive individual. It is also known that insulin resistance in the brain's appetite centers can lead to reduced satiation. This leads to overeating and reinforces the hyperinsulinemia. Whereas lean, insulin sensitive people's appetite's are more readily sated. None of this is to deny the key role of caloric balance in determining weight gain or loss. But hyperinsulinemia and insulin resistance in combination can account for drives that lead the obese to over-consume both fat and carbohydrate.
Todd said…
Take a look at Lustig:

The effect of insulin on appetite is quite interesting. In normal, insulin-sensitive individuals, it performs a dual role:

"Whereas insulin drives the accumulation of energy stores in liver, fat, and muscle, its role in the CNS tends to decrease energy intake. This is not a paradox, but rather an elegant instance of negative feedback."

However this feedback mechanism tends to break down in obese, insulin-resistant individuals:

"Although CNS insulin levels tend to reflect serum insulin levels, the relationship breaks down in obesity states. In obesity, there is proportionally less CNS insulin; the expression of the CNS insulin transporter is decreased in several obesity models. This paucity of insulin available for satiety signaling may represent a form of CNS insulin resistance."

And insulin plays a further role in leptin resistance:

"Obesity is a state in which the negative feedback pathways of insulin and leptin are ineffective: the CNS resists the regulatory effects of insulin and leptin, so that appetite remains uncurbed and weight accrues despite adequate energy stores. Despite high circulating levels of leptin and insulin in the setting of obesity, there is a paucity of CNS satiety signaling by these hormones, leading to an inappropriate perception of starvation. CNS insulin and leptin resistance are beginning to be understood, through mechanisms that are described below."

Where exactly is Lustig wrong here?

Keep in mind that he's not denying ASP. He's not saying "insulin is everything" or carbohydrates are the sole cause of obesity, or that calories don't matter. He acknowledges everything you and CarbSane say about the roles of elevated FFA and lipotoxicity in the genesis of lipotoxicity. But you can acknowledge this and still acknowledge the role of insulin in obesity.

But there is a bigger picture that acknowledges the whole truth. And part of that truth is recognizing that it's not just about insulin, but insulin relative to the state of insulin resistance in specific tissues -- including the brain -- that play out in obesity.
Todd said…
I meant to write: "...the roles of elevated FFA and lipotoxicity in the genesis of insulin resistance."
Sanjeev said…
> insulin ... you claim it is nothing


I've NEVER written that unqualifiedly. NEVER.

"adherence is orders of magnitude more important than insulin fantasies"

does NOT mean "insulin is nothing "

In one sense, CURRENTLY, insulin is nothing: anyone who wants to lose fat mass should not think about insulin at all, IMHO they should ignore insulin completely and find a diet they can adhere to.

Let aside the fact that a layperson canNOT get their insulin measured regularly - what are they supposed to do about their insulin?

What the dieter can control is diet. The diet they may be able to adhere to may be low fat or low carb or low protein ... whatever. The last thing they should care about is insulin.

The majority (maybe the vast majority) on the long term weight control registry is still low fat, and a large fraction of those, from the last time I read about them, eat a lot of starches.

> But I find it amazing that you find the role of insulin to be "infinitessimally minor".

for a layperson DIETER seeking a LONG TERM SOLUTION, available RIGHT NOW, insulin is of vanishing importance. That was my context. It may be important to researchers, should not be to dieters.

Tell me, where can they go to get their insulin measured several times a day?

pragmatically, isulin's importance is near zero. Infinitessimally small.

Or do you suggest they spend a majority of their time researching insulin, and making up stories and fantasies about it?

for the person whose ideal diet is high starch (a Kitavan immigrating to the US), you suggest they go on a high fat / low carb diet, because of insulin fantasies?

> hyperinsulinemia and insulin resistance in combination can account for drives that lead the obese to over-consume both fat and carbohydrate

Since the balance of current evidence is that insulin satiates

where's the proof for this, better proof than for satiation? (experiments with measurements, not thought experiments, not extrapolations, not fantasy)

I note you ignored the last request for measurements and respond with theories and guesses and hypothetical scenarios
Sanjeev said…
> Where exactly is Lustig wrong here?

In vitro neurons ...

New research with no real world controlled trials ...

the stuff that led to Alan Aragon's epic takedown of Lustig's fructose talk ...

In vitro neurons ...

New research with no real world controlled trials bearing on what people should pragmatically do, TODAY, what those who've been successful at this task have been doing for a long time ...

Yeah, pragmatically, based on that obese people should start using those convenient tools to measure insulin regularly ...

oh, wait ... they CANNOT
Sanjeev said…
and even after they find a place to measure their insulin regularly,

THEN they'll need to get

biopsies of their fatty tissues
biposies of their hypothalamus and muscles and liver

because it's NOT the insulin, it's the insulin PLUS the senstivities.

Which may change hourly, based on stress and exercise.

And no, one cannot use blood glucose as a proxy. One cannot use dietary carbohydrate intake as a proxy (because protein causes insulin release too).

And even if you could use those as proxies, there's the question of biopsies ...

Again, the pragmatic, real world importance of insulin: orders of magnitude smaller than finding a diet that can be adhered to.

And no, insulin considerations can't be used to FIND that adherence-prone diet, for many of the same reasons given above.
CarbSane said…
Hi Todd, I'm a bit pre-occupied at the moment, and indeed I've been thinking how best to bring the rest of your thoughtful email to my readers here. It's been a long while ... I apologize. I'm happy to see you can post here now! My discussion board topics are imbeddable into blog posts so I'm thinking that I'll share your last commentary (with links to prior exchanges) with that vehicle. Hopefully in the next day or so.

Nice to see you here. You always add food for thought although we tend to disagree.
CarbSane said…
OMG! Sanjeev ... I just listened to the whole whopping 15 minutes of Lustig's presentation at AHS. I'm left going ... huh? Question time so soon? He's never seen a fat kid on a sugar high? Really? Sigh.
Sanjeev said…

You may want to wait for confirmation on Lustig's

insulin ⇒ (hypothalamic leptin resistance)

claim. Also, look up the controlled and measured evidence for "sugar high", the first salient point he makes, around 6 minutes in.
.... that's his first major point (!!!!!!!!!!)

COME ON, Dr. Lustig; geeeez, phhhhhthhththththht
Sanjeev said…
> huh? Question time so soon? He's never seen a fat kid on a sugar

apparently they mistakenly spliced a 10 minute segment from the end into Lustig's video starting around 9 minutes in, then at around 19 minutes what should have been at 9 minutes comes in.

and the "sugar high" ... controlled studies don't find it, and that's one of the canonical examples skeptics learn about and study as part of "confirmation bias" and "availability heurystic", so unless new carefully controlled research has overturned the older debunkings ...

that was the point of my earlier OMG !!! LUSTIG !!! comment, to start his talk with a debunked myth.

click or manually:
CarbSane said…
Yeah Sanjeev ... I listened through the Q&A and it starts to sound like a lecture again. Confusing for sure!
Tonus said…
Huh! I did not know that about the "sugar rush." I just assumed that children (who are generally considered to have lots of energy) act hyper after a sugary snack because they've got additional fuel to burn. It sure seemed to make sense...

I think it's a good example, though, of how people can come to distrust science and research. Lustig, who strikes me as a respected doctor and speaker, is promoting a myth as part of his campaign against sugar. It doesn't matter if it's deliberate, does it? The impression will be that he is either dishonest or lacking in knowledge of the topic he has chosen to champion. The other reaction, of course, is to insist that the research MUST be wrong because Dr. Lustig couldn't possibly be!

(Now I will sit back and await the myriad posts and comments elsewhere decrying CarbSane's craven attack on Dr. Lustig! Sorry Eve-- the company you keep! The company you keep!!)
CarbSane said…
You crack me up Tonus!

He had me at leptin ... he lost me on ... ;-)
. said…
Sims EA, Horton ES, Salans LB. Inducible metabolic abnormalities during development of obesity. Annu Rev Med. 1971;22:235-50. Review. ; ; see also
. said…
I'm not sure if I already posted this study here (sorry if it's a repetition)... Just another shocking metabolic ward, calories rule, weight loss study!

Bortz WM. Predictability of weight loss. JAMA. 1968 Apr 8;204(2):101-5. PubMed PMID: 5694531. Full text here: