Monday, August 1, 2011

There is no doubt that abnormally elevated insulin is associated with body fat accumulation

For as long as diets existed there have been influential proponents, or believers, who at some point had what they thought were epiphanies. From that point forward, they disavowed the diets that they formally endorsed. Low carbohydrate dieting seems to be in this situation now. Among other things, it has been recently “discovered” that the idea that insulin drives fat into body fat cells is “wrong”.

Based on some of the comments I have been receiving lately, apparently a few readers think that I am one of those “enlightened”. If you are interested in what I have been eating, for quite some time now, just click on the link at the top of this blog that refers to my transformation. It is essentially high in all macronutrients on days that I exercise, and low in carbohydrates and calories on days that I don’t. It is a cyclic approach that works for me; calorie surpluses on some days and calorie deficits on other days.

But let me set the record straight regarding what I think: there is no doubt that insulin is associated with body fat accumulation. I was told that an influential health blogger (whom I respect a lot) denied this recently, going to the extreme of saying that no professional metabolism or endocrinology researcher believes in it, but I couldn’t find any evidence of that statement. It is not hard at all to find professional metabolism and endocrinology researchers who have asserted that insulin is associated with body fat accumulation, based on very reliable evidence. Actually, this is Biochemistry 101.

What I think is truly unclear is whether insulin spikes associated with carbohydrate-rich foods in general are the cause of obesity. This idea is, indeed, probably wrong given the evidence we have from various human populations whose members consume plenty of non-industrialized carbohydrate-rich foods. On a related note, I particularly disagree with the notion that the pancreas gets tired over time due to having to secrete insulin in bursts, which seems to also be one of the foundations on which many low carbohydrate diet varieties rest.

As with almost everything related to health, the role of insulin in body fat gain is complex, and part of that complexity is due to the nonlinear relationship between body fat gain and postprandial insulin release. Industrial carbohydrate-rich foods have a much higher glycemic load than natural carbohydrate-rich foods, even though their glycemic index may be the same in some cases. In other words, the quantity of easily digestible carbohydrates per gram is much higher in industrial carbohydrate-rich foods.

In normoglycemic folks, this leads to an abnormally elevated insulin response, among other hormonal responses. For example, circulating growth hormone, which promotes body fat loss, is inversely correlated with circulating insulin. Insulin drives fat, typically from dietary sources of fat, into adipocytes. That fat may also come from excess carbohydrates, packaged into VLDL particles.

Under normal circumstances, that would be fine, since our body is designed to store fat and release it as needed. But the abnormal insulin response elicited by industrial carbohydrate-rich foods, together with other hormonal responses, leads to a little more body fat accumulation, and for longer, than it should. And I’m talking here about people without any metabolic damage. Saturated and monounsaturated fats are healthy when eaten, but when they are stored as excess body fat, they become pro-inflammatory.

Body fat is like an organ, secreting many hormones into the bloodstream, several of which are pro-inflammatory. One of those pro-inflammatory hormones, which I believe is closely linked with many diseases of civilization, is tumor necrosis factor. (The acronym is now TNF. Apparently the “-alpha” after its name and acronym has been dropped recently.) Dietary fat, particularly saturated fat, seems to be anti-inflammatory. In other words, body fat accumulation is the problem. You only need 30 g/d of excess body fat accumulation to gain around 24 lbs of fat per year. Over three years, that will add up to over 70 lbs of body fat.

In my view, ultimately it is excess inflammation (which is, in essence, a vascular response) that is at the source of most of the diseases of civilization.

That is where the nonlinearity comes in. Insulin is healthy up to a point. Beyond that, it starts causing health problems, over time. And one of the main mechanisms by which it does so is via excessive body fat accumulation, with different damage threshold levels for different people. Insulin may decrease appetite as it goes up, but it increases it if goes down too much. If it goes up abnormally, typically it will go down too much. As it reaches a trough it induces hypoglycemia, even if mildly.

Take a look at the graph below, from this post showing the glucose variations in normoglycemic individuals. There is a lot of variation among different individuals, but it is clear that the magnitude of the hypoglycemic dips is inversely correlated with the magnitude of the glucose spikes. That inverse correlation is due primarily to the effect of insulin. Under normal circumstances, a decrease in circulating insulin would promote an increase in free fatty acids in circulation, which would normally have a suppressing effect on hunger in the hours after a meal. But industrial carbohydrate-rich foods lead to increases and decreases in glucose and insulin that are too steep, causing the opposite effect.

You may ask: why do you keep talking about industrial carbohydrate-rich foods? Why not talk about industrial protein- or fat-rich foods as well? The reason is that the food industry has not been very successful at producing industrial protein- or fat-rich foods that are palatable without adding a lot of carbohydrate to them.

More often than not they need enough carbohydrate added in the form of sugar to become truly addictive.


Jenny said...

I think you are way oversimplifying the role of insulin here, which is, unfortunately, the tendency among all people who write about obesity.

There is a huge variation in insulin effectiveness among individuals which is influenced by genes, exposure to environmental toxins like herbicides and pharmaceuticals, and the amount of intracellular fat stores in the liver.

A normal person who has normal insulin resistance might secrete the equivalent of 5 units of insulin in response to a huge piece of iced cake. A normal person who is insulin resistant might secrete 15, 25, or even 50 units of insulin in response to the same food intake.

This is why there are people who can eat extremely high carb diets full of junk food and stay thin.

It is also why there are obese people who have completely normal blood sugar and can live to be 100 years old without developing diabetes. (See Dr. Nir Barzilai's centenarian study for a surprising report on how many of his centenarians were obese at midlife.)

Health bloggers need to stop looking for One Size Fits All explanations for metabolic functions, because the variation among humans is huge.

They also have to stop blaming foods for the damage that is more likely due to the growing burden of pesticides, herbicides, insulin resistance-causing prescription drugs, plasticizers, BPA etc.

People were mostly thin in the 1950s when I grew up and they ate all kinds of dreadful food all the time. The obesity crisis maps very closely to things like the growing use of statins, SSRIs, BPA, atrazine in water supplies, etc.

These raise insulin resistance and some poison the beta cell.

The impact of high fructose intakes from HFCS on the liver is another major cause of obesity.

High carb diets make the damage worse, but to blame them for the epidemic of metabolic dysfunction is to give a pass to the real culprits--the industries that have wantonly poisoned our environment.

Go look at the recent studies of the toxic industrial chemical burden found in the blood of pregnant women if you doubt that these chemicals are a huge and undiscussed problem.

Beth@WeightMaven said...

Re Stephan's comment, you can find it here:

All this postprandial insulin spikes = fat gain stuff is nonsense as far as I can tell. I still haven't seen a shred of convincing evidence to support it, and in fact, the evidence I've seen mostly supports the opposite hypothesis, that insulin spikes oppose fat gain. The idea that postprandial insulin causes fat gain is contradicted by the most basic empirical facts in both human and animal studies. As I said before, I don't know a single person who studies metabolism/endocrinology professionally who takes that idea seriously-- it is confined to the popular press and internet blogs/forums.

I think the key word though is "postprandrial" ... thus his position seems in line with yours re insulin spikes after meals.

Jeff Consiglio said...

I like your thoughts about "lipotoxcicity" being the result of the fat stored on one's body...Becoming in essence, an active endocrine-gland spitting out nasty inflammatory opposed to the fat one eats being the cause. At least as long as it's not trans-fats or excess omega-6 of course.

By what mechanism do you think saturated fats are actually anti-inflammatory? I would also wonder if the medium chain fatty acids in coconut oil are more anti-inflammatory than long-chain sat-fats from animal sources. Just a theory, for which I have absolutely no proof.

Anonymous said...

Ned, what are your thoughts about James Kriegers writings on Insulin? His multipart series pretty much refutes a lot of what you have said, and provides huge amounts of research backing it...

Ned Kock said...

Hi Jenny. Environmental toxins are definitely a factor, but I don’t think they fully explain all cases of diabetes. There are several examples of formerly isolated populations where T2D became rampant, and where the main causative factor seems to have been massive consumption of wheat flour and sugar.

Ned Kock said...

Hi Beth. Stephan clearly knows that a very low insulin-to-glucagon ratio is associated with diabetic ketoacidosis, where the very high blood concentration of ketones is a direct result of out-of-control body fat release.

An increase in circulating insulin solves the problem, because insulin is a fat storage hormone, among other things.

So I don’t know what he meant to say with that quote.

Ned Kock said...

Hi Jeff. I have a post here somewhere showing a decrease in pro-inflammatory hormones in circulation after a high saturated fat meal. The exact mechanism I don’t know.

Ned Kock said...

Hi Anon. I read those articles, some of which only briefly, and I liked them. I don’t see many blatant contradictions between those articles and this post though. Can you point a few out?

Ned Kock said...

Btw, Jenny, on your site you say that blood sugars > 140 mg/dl "cause" diabetes.

Do you really believe in that?

If yes, how does that fit with what you said above?

Jeff Consiglio said...

Regarding possible anti-inflammatory mechanism for sat-fats...I would wonder if sat-fat's ability to help "stabilize" poly-fats plays a part in that.

Or at least it is my understanding that chemically unstable polyunsaturated fatty acids are made less reactive to heat, light and oxygen in the presence of sat-fats.

I think of Kitavans who presumably get a fair amount of poly-fats from fish...but within the context of also eating coconut which I've heard may "stabilize" poly-fats even better than animal based sat-fats.

That's why I'm VERY wary of trying to mimicking "parts" of traditional diets, without knowing the full context that diet occurs in.

So many intellectual rabbit-trails to explore, yet so little time in a day. (<:

Anonymous said...

Regardless of the role of insulin, I feel that for a lot of people that eating too many carbohydrates is an issue regarding digestion. (From mild all the way to ceoliac to IBS or Crohns). So you can argue all you want (to the commentors that stated) that insulin is not too much of a problem as many people have been saying it is.

I guess it goes back to eat what makes you feel good.

I do agree thought that the topic of nutrition can be a bit sticky. Unless we understand 100% bio-chemistry within the body, we'll never truly understand nutrition. For example, some people argue that dairy raises insulin, but that it doesn't raise blood sugars. (This would be for the argument that it's not insulin that causes fat accumulation). But what about type I's who gain weight (weight they need to gain) once they start taking insulin? What about the (stupid) teenage girl type I's who dick around with their insulin in order to 'lose weight'. (That's a scary thought).

I would love to know what developments happen within the next 50 years regarding nutrition. So far I know that Peter (Hyperlipid) and Stan (Heretic) have been LC'ing for approx. 12 years. They seem to be in good fact 'corrected' their health. (Hence their reasoning for switching to LC).

*I say 'corrected' because no type of diet will make you super human. Rather it is a normalization (or approaching normalization) that makes people feel superhuman when comparing to a sick population.


Jeff Consiglio said...

BTW - Came across your blog via a comment you made about adiponectin on another blog.

Came across a few abstracts today doing further research on adiponectin you might find interesting, if you're not already familiar with them.

Adiponectin and cancer: a systematic review -

High-salt diet increases plasma adiponectin levels independent of blood pressure in hypertensive rats: -

David Isaak said...

Well, I can't say that this delves deep into biochemistry, but in Taubes recent book "Why We Get Fat," he has an absoutely horrific picture of a woman with two melon-sized lumps of fat on her upper thighs...caused by injecting insulin in the same spot over many years. So apparently locally high concentrations of insulin have a dramatic fat storage effect on at least some people.

I'm thinking of patenting this as an alternative to silicone breast implants.

js290 said...

FWIW: In contrast to the hormonal activation of adenylate cyclase and (subsequently) hormone-sensitive lipase in adipocytes, the mobilization of fat from adipose tissue is inhibited by numerous stimuli. The most significant inhibition is that exerted upon adenylate cyclase by insulin. When an individual is in the well fed state, insulin released from the pancreas prevents the inappropriate mobilization of stored fat. Instead, any excess fat and carbohydrate are incorporated into the triacylglycerol pool within adipose tissue.

Jenny's site is really useful, particularly the tip on keeping blood sugars low. This is probably good advice for everybody. But, I agree with Ned in using Ockham's Razor in explaining the causes of diabetes. Let's not get sidetracked by the edge cases.

Aaron Blaisdell said...

Ned, your description of the see-saw act between circulating glucose and insulin is the model of what in psychology we call opponent processes. The a process is the initiator of the b process which attempts to bring the system back towards homeostasis. The larger the a process, the larger the b process. Unfortunately, when the a process is excessively large, it can bring the system so far out of homeostasis that the reactionary b processes is so large that it overshoots optimal values and pushes the system out of homeostasis in the opposite direction. It's like a bicycle that begins to wobble after hitting a bump, once the deviation is large enough, the wobbles become so large that they destabilize the bicycle. This may be a nice analogy for the role that industrial carbs play in modern obesity.

js290 said...


In engineering, that's what's known as an undamped system. One can figure out ones own system response as suggested at Jenny's page on how to lower blood sugar. Altering the carbohydrate intake is essentially changing the input function to the system. Healthy metabolisms are probably critically damped for a variety of input functions. Damaged ones probably not so much.

malpaz said...

I am always left confused when i read your posts, haha.

Is the take home message here that industrialized carbohydrates are the cause of excess-long runs of high insulin? If so, I highly doubt 99%of the readers of your blog even eat industrialized carbohydrates to begin with.

is there anything to eating non-industrialized carbohydrates and extended bouts of high BS? is this even possible? i would think, even an attempt at keeping your BS spiked long times would be physically hard to do eating real food.

for one, real food carbs are packed with fiber and water which makes eating them in excess pretty limiting. most people are apt to feel like a whale if they eat a pint of strawberries, same with even trying to throw back 3 sweet potatoes.

junk food is obviously obnoxiously unnatural and going to have unnatural consequences, however is there any proof out there that overeating fruit, or overeating starch is even possible and harmful or even having the ability to produce long bouts of spiked BS??

Beth@WeightMaven said...

Ned, I took Stephan's comment to mean that in general, normal postprandrial insulin response to dietary carbs is unlikely to be the primary (or even a major) causative factor wrt obesity.

I.e., a normal postprandrial spike of insulin as a result of a high carb meal is a very different thing from "abnormally elevated insulin" (which is presumably related to other factors e.g., peripheral or hepatic IR).

So I don't think he's saying insulin isn't associated with body fat storage, just that he doesn't ascribe to Taubes et al's theory that obesity is a result of dietary carbs driving insulin driving fat storage.

I've always liked Peter@Hyperlipid's hepatic injury theory myself. Aaron's homeostasis/wobble theory could tie in nicely, as could the role of both environmental and food toxins.

js290 said...

Some of you are starting to sound like Milhouse van Houten:

In the treehouse, the neighborhood kids try to figure out what's up with
the adults.

Bart: So finally, we're all in agreement about what's going on with
the adults. Milhouse?
Milhouse: [steps up to blackboard] Ahem. OK, here's what we've got: the
Rand Corporation, in conjunction with the saucer people --
Bart: Thank you.
Milhouse: -- under the supervision of the reverse vampires --
Lisa: [sighs]
Milhouse: -- are forcing our parents to go to bed early in a fiendish
plot to eliminate the meal of dinner. [sotto voce] We're
through the looking glass, here, people...
-- A conspiracy theory Oliver Stone would be proud of, "Grampa vs.
Sexual Inadequacy"

Your connecting all the wrong dots and confusing yourselves.

There are a few facts that should not be controversial:

1. Insulin regulates fatty acid metabolism.
2. Excessive glucose metabolism is not good for long term health.

There's easy and relatively cheap ways to track this for each of us individually.

1. Insulin probably tracks blood sugar levels. Even if it doesn't exactly, this is the cheapest way to measure right now.
2. Use Ned's HCE to make some correlations on what works for you.

80/20 rule: 20% effort to solve 80% of the problem. Don't get distracted by the higher order effects.

David Isaak said...


That medical biochemistry page you linked to is actually quite a nice compilation of the basics!

Ned Kock said...

Hi Jeff. Cool articles on adiponectin, particularly the second! Thanks.

Ned Kock said...

Hi David. Some argue that lipomas can develop around the insulin injection sites. Here is an anecdotal account:

So the patent idea will not work, as no product based on it will be approved by the FDA.

What is it with breast augmentation anyway? So many women doing it; I don’t see the point.

Ned Kock said...

Nice analogy Aaron.

Ned Kock said...

Hi Mal. There is no evidence that I know of implicating natural carb-rich foods in the diseases of civilization. Refined carbs tip a fine balance in favor of disease, and it is a fine balance indeed, as Aaron and js290 noted.

Btw, most readers of this blog arrive via Google searchers and are probably on the disease-promoting SAD (standard American diet). The blog has currently about 40,000 visits per month; 43 percent are new visitors.

Ned Kock said...

Hi Beth. I see your point. Many people misinterpreted that quote by Stephan though.

Anonymous said...

Jenny 'gets' it.

High intakes of PUFAs in conjunction with carbs (especially fructose) clearly play a role in developing that liver fat, too.

In regards to Ned's remark about flour and sugar: Isn't this just restating Jenny's point? These populations are clearly genetically sensitive.


iolite said...

After the blood stream has been flooded with an abnormally high level of insulin for an hour or so, the blood sugar level finally begins to tumble

CarbSane said...

@Jenny: How do you explain remissions from diabetes (in those 80% for whom obesity triggers the disease) with GBP surgery, intense early insulin treatment, or that crash diet? Presumably these people are still living in their same environments, and perhaps no longer take meds to control diabetes, but one can hardly finger those meds from causing the disease in the first place.

@js290: You cite Jenny's site in stating that insulin levels likely correlate with BG levels. How does that square with her own statements here: "A normal person who has normal insulin resistance might secrete the equivalent of 5 units of insulin in response to a huge piece of iced cake. A normal person who is insulin resistant might secrete 15, 25, or even 50 units of insulin in response to the same food intake. "
BG levels are good markers of glycemic control and tell us nothing about our insulin levels other than that if they remain within normal ranges, our bodies are secreting the appropriate amount of insulin to maintain that control. That amount may be much higher in an IR person who is able to compensate.

CarbSane said...

@Ned: Your post got me to drag out my copy of Guyton Textbook of Medical Physio yesterday. Of COURSE it outlines all the ways insulin works to put and keep fatty acids in the fat cells. We all have that triglyceride/fatty acid cycle functioning at all times or we're dead. But I believe Stephan is referring to the carb -> postprandial insulin -> NET fat accumulation hypothesis when he says he knows of no researcher in that arena who believes it. I've yet to come across a study demonstrating that carbohydrate consumption causes disruption in the fatty acid metabolic cycle in adipocytes.

I've blogged a ton on how Taubes seems to have the progression of pathological insulin backwards. Small adipocytes are sensitive to insulin but they become resistant as they get large and bloated. This results in lessening of insulin's action on HSL and excessive free fatty acid release. Elevated FFA delivery to the cells forces the cells to use them over glucose -> decreased glucose disposal (IGT) and hepatic IR.

There do appear to be segments of the population who are hyperinsulinemic w/o being IR and vice versa indicating that perhaps we're all talking past one another whether HI causes IR or IR causes HI ... or perhaps they occur simultaneously but some are less predisposed to one or the other and a fatty acid induced model is consistent with these observations.

If ppInsulin is responsible for net fat accumulation, and a T2 no longer mounts a ppInsulin response to a carb load, why don't they drop fat like gangbusters? Because the basal insulin levels are not ultimately related to carb consumption.

In the infamous Grey & Kipnis study that supposedly demonstrates that fasting insulin levels are altered by rather drastic changes in diet (iso or hypocaloric) we see that the fasting insulin levels do jump considerably with the marked change in diet, but they also appear to be "normalizing" in the third week after the change.

The more convincing take home message to me is that there was absolutely no correlation between the change in fasting insulin levels and weight change.

Ned Kock said...

Hi CS. To me the effect of refined carbs consumption on insulin, which I see as abnormal, results in a tip of the balance toward body fat gain. This is relatively easily counteracted in some people (like me), by sticking with natural carbs, and in others by simply consuming fewer calories or increasing caloric expenditure.

In my case refined carbs lead to the mistimed hunger that we hear about so often, which I think is reasonably well explained by abnormal insulin and glucose fluctuations. Maybe not fully explained, as hunger is a complex phenomenon that has (among other things) an important psychological component.

Even my view on refined carbs is not dogmatic, and I see gradations of risk – e.g., regular soda right at the top. Also, I see the calorie control part of the equation as critical. In fact, a key advantage of dropping refined carbs from one’s diet is in my view that it frequently leads to a reduction in caloric intake, with a concomitant increase in nutrient variety.

David Isaak said...

"What is it with breast augmentation anyway? So many women doing it; I don’t see the point."

Neither do I. And neither do most males I know. So there seems to be some sort of disconnect here...or the guys I know aren't a very representative group.

Jack C said...

Jeff, in response to your question about the mechanism by which saturated fat decreases inflammation:

"Ingestion of dietary fat stimulates cholecystokinin (CCK)receptors, and leads to attenuation of the inflammatory response by way of the efferent vagus nerve and nicotinic receptors"

(Pubmed 16216887)

CCK is a hormone best known as an appetite suppressant that slows the digestive system in response to fat. Interestingly, CCK responds more strongly to dairy fat than to non dairy fats that have an equal mix of fats by degree of saturation. (Pubmed 14652359)

David Isaak said...

@Anon: "In regards to Ned's remark about flour and sugar: Isn't this just restating Jenny's point? These populations are clearly genetically sensitive."

That's circular reasoning.

An equivalent statement: "Not everyone who smokes gets lung cancer. The subpopulation who get lung cancer are clearly genetically sensitive."

And how do we know they are genetically sensitive, rather than it being a basic problem with smoking? Well, becasue they got lung cancer. And why did they get lung cancer? Because they are genetically sensitive...and so on.

It might in fact be true. Or not. What it isn't, at our present state of knowledge, is helpful.

Anonymous said...

May I know if you have reviewed any research on meal frequency and its impact on blood glucose and insulin.

4rx said...

It makes sense because insulin is in charge of sugar regulation on blood and the best way to take it out from there. It is to gather in glycogen deposits on the cell.