Saturday, November 20, 2010

Glucose Tolerance in Non-industrial Cultures

Background

Glucose is the predominant blood sugar and one of the body's two main fuel sources (the other is fatty acids). Glucose, in one form or another, is also the main form of digestible dietary carbohydrate in nearly all human diets. Starch is made of long chains of glucose molecules, which are rapidly liberated and absorbed during digestion. Sucrose, or table sugar, is made of one glucose and one fructose molecule, which are separated before absorption.

Blood glucose is essential for life, but it can also be damaging if there is too much of it. Therefore, the body tries to keep it within a relatively tight range. Normal fasting glucose is roughly between 70 and 90 mg/dL*, but in the same individual it's usually within about 5 mg/dL on any given day. Sustained glucose above 160 mg/dL or so causes damage to multiple organ systems. Some people would put that number closer to 140 mg/dL.

The amount of glucose contained in a potato far exceeds the amount contained in the blood, so if all that glucose were to enter the blood at once, it would lead to a highly damaging blood glucose level. Fortunately, the body has a hormone designed to keep this from happening: insulin. Insulin tells cells to internalize glucose from the blood, and suppresses glucose release by the liver. It's released by the pancreas in response to eating carbohydrate, and protein to a lesser extent. The amount of insulin released is proportional to the amount of carbohydrate ingested, so that glucose entering the blood is cleared before it can accumulate.

Insulin doesn't clear all the glucose as it enters the bloodstream, however. Some of it does accumulate, leading to a spike in blood glucose. This usually doesn't exceed 130 mg/dL in a truly healthy person, and even if it approaches that level it's only briefly. However, diabetics have reduced insulin signaling, and eating a typical meal can cause their glucose to exceed 300 mg/dL due to reduced insulin action and/or insulin secretion. In affluent nations, this is typically due to type II diabetes, which begins as insulin resistance, a condition in which insulin is actually higher than normal but cells fail to respond to it.  The next step is the failure of insulin-secreting beta cells, which is what generally precipitates actual diabetes.

The precursor to diabetes is called glucose intolerance, or pre-diabetes. In someone with glucose intolerance, blood glucose after a typical meal will exceed that of a healthy person, but will not reach the diabetic range (a common definition of diabetes is 200 mg/dL or higher, 2 hours after ingesting 75g of glucose). Glucose tolerance refers to a person's ability to control blood glucose when challenged with dietary glucose, and can be used in some contexts as a useful predictor of diabetes risk and general metabolic health. Doctors use the oral glucose tolerance test (OGTT), which involves drinking 60-100g glucose and measuring blood glucose after one or two hours, to determine glucose tolerance.

Why do we care about glucose tolerance in non-industrial cultures?

One of the problems with modern medical research is that so many people in our culture are metabolically sick that it can be difficult to know if what we consider "normal" is really normal or healthy in the broader sense. Non-industrial cultures allow us to examine what the human metabolism is like in the absence of metabolic disease. I admit this rests on certain assumptions, particularly that these people aren't sick themselves. I don't think all non-industrial cultures are necessarily healthy, but I'm going to stick with those that research has shown have an exceptionally low prevalence of diabetes (by Western standards) and other "diseases of civilization" for the purposes of this post.

Here's the question I really want to answer in this post: do healthy non-industrial cultures with a very high carbohydrate intake have an excellent glucose tolerance, such that their blood glucose doesn't rise to a high level, or are they simply resistant to the damaging effects of high blood glucose?

The data

I'm going to start with an extreme example. In the 1960s, when it was fashionable to study non-industrial cultures, researchers investigated the diet and health of a culture in Tukisenta, in the highlands of Papua New Guinea. The eat practically nothing but sweet potatoes, and their typical daily fare is 94.6 percent carbohydrate. Whether or not you believe that exact number, their diet was clearly extraordinarily high in carbohydrate. They administered 100g OGTTs and measured blood glucose at one hour, which is a very stringent OGTT. They compared the results to those obtained in the 1965 Tecumseh study (US) obtained by the same method. Here's what they found (1):
Compared to Americans, in Tukisenta they had an extraordinary glucose tolerance at all ages. At one hour, their blood glucose was scarcely above normal fasting values, and glucose tolerance only decreased modestly with age. In contrast, in Americans over 50 years old, the average one-hour value was around 180 mg/dL!

Now let's take a look at the African Bantu in the Lobaye region of the Central African Republic. The Bantu are a large ethnic group who primarily subsist on a diverse array of starchy foods including grains, beans, plantains and root crops. One hour after a 100g OGTT, their blood glucose was 113 mg/dL, compared to 139 mg/dL in American controls (2). Those numbers are comparable to what investigators found in Tukisenta, and indicate an excellent glucose tolerance in the Bantu.

In South America, different investigators studied a group of native Americans in central Brazil that subsist primarily on cassava (a starchy root crop) and freshwater fish. Average blood glucose one hour after a 100g OGTT was 94 mg/dl, and only 2 out of 106 people tested had a reading over 160 mg/dL (both were older women) (Western Diseases: Their Emergence and Prevention, p. 149). Again, that indicates a phenomenal glucose tolerance by Western standards.

I have to conclude that high-carbohydrate non-industrial cultures probably don't experience damaging high blood glucose levels, because their glucose tolerance is up to the task of shuttling a huge amount of glucose out of the bloodstream before that happens.

Not so fast...

Now let's turn our attention to another study that may throw a wrench in the gears. A while back, I found a paper containing OGTT data for the !Kung San (also called the Bushmen), a hunter-gatherer group living in the Kalahari desert of Africa. I reported in an earlier post that they had a good glucose tolerance. When I revisited the paper recently, I realized I had misread it and in fact, their glucose tolerance was actually pretty poor.

Investigators administered a 50g OGTT, half what the other studies used. At one hour, the San had blood glucose readings of 169 mg/dL, compared to 142 mg/dL in Caucasian controls (3)! I suspect a 100g OGTT would have put them close to the diabetic range.

Wait a minute, these guys are hunter-gatherers living the ancestral lifestyle; aren't they supposed to be super healthy?? First of all, like many hunter-gatherer groups the San are very small people: the men in this study were only 46 kg (101 lbs).  The smaller you are, the more a given amount of carbohydrate will raise your blood glucose.  Also, while I was mulling this over, I recalled a discussion where non-diabetic people were discussing their 'diabetic' OGTT values while on a low-carbohydrate diet. Apparently, carbohydrate refeeding for a few days generally reverses this and allows a normal OGTT in most people. It turns out this effect has been known for the better part of a century.

So what were the San eating? The study was conducted in October of 1970. The San diet changes seasonally, however their main staple food is the mongongo nut, which is mostly fat and which is available year-round (according to The !Kung San: Men, Women and Work in a Foraging Society). Their carbohydrate intake is generally low by Western standards, and at times of the year it is very low. This varies by the availability of other foods, but they generally don't seem to relish the fibrous starchy root crops that are available in the area, as they mostly eat them when other food is scarce. Jean-Louis Tu has posted a nice analysis of the San diet on BeyondVeg (4). Here's a photo of a San man collecting mongongo nuts from The !Kung San: Men, Women and Work in a Foraging Society:

What did the authors of the OGTT study have to say about their diet? Acknowledging that prior carbohydrate intake may have played a role in the OGTT results of the San, they made the following remark:
a retrospective dietary history (M. J. Konner, personal communication, 1971) indicated that the [San], in fact, consumed fairly large amounts of carbohydrate-rich vegetable food during the week before testing.
However, the dietary history was not provided, nor has it been published, so we have no way to assess the statement's accuracy or what was meant by "fairly large amounts of carbohydrate-rich vegetable food." Given the fact that the San diet typically ranges from moderately low to very low in carbohydrate, I suspect they were not getting much carbohydrate as a percentage of calories. Looking at the nutritional value of the starchy root foods they typically eat in appendix D of The !Kung San: Men, Women and Work in a Foraging Society, they are fibrous and most contain a low concentration of starch compared to a potato for example. The investigators may have been misled by the volume of these foods eaten, not realizing that they are not as rich in carbohydrate as the starchy root crops they are more familiar with.

You can draw your own conclusions, but I think the high OGTT result of the San probably reflect a low habitual carbohydrate intake, and not pre-diabetes. I have a very hard time believing that this culture wasn't able to handle the moderate amount of carbohydrate in their diet effectively, as observers have never described diabetic complications among them.

Putting it all together

This brings me to my hypothesis. I think a healthy human body is extraordinarily flexible in its ability to adapt to a very broad range of carbohydrate intakes, and adjusts glucose tolerance accordingly to maintain carbohydrate handling in a healthy range. In the context of a healthy diet and lifestyle (from birth), I suspect that nearly anyone can adjust to a very high carbohydrate intake without getting dangerous blood glucose spikes. A low carbohydrate intake leads to impaired glucose handling and better fat handling, as one would expect. This can show up as impaired glucose tolerance or even 'diabetes' on an OGTT, but that does not necessarily reflect a pathological state in my opinion.

Every person is different based on lifestyle, diet, personal history and genetics. Not everyone in affluent nations has a good glucose tolerance, and some people will never be able to handle starch effectively under any circumstances. The best way to know how your body reacts to carbohydrate is to test your own post-meal blood glucose using a glucose meter. They are inexpensive and work well. For the most informative result, eat a relatively consistent amount of carbohydrate for a week to allow your body to adapt, then take a glucose measurement 1 and 2 hours after a meal. If you don't eat much carbohydrate, eating a potato might make you think you're diabetic, whereas after a week of adaptation you may find that a large potato does not spike your blood glucose beyond the healthy range.

Exercise is a powerful tool for combating glucose intolerance, as it increases the muscles' demand for glucose, causing them to transport it out of the blood greedily after a meal. Any exercise that depletes muscle glycogen should be effective.


* Assuming a typical carbohydrate intake. Chris Kresser recently argued, based on several studies, that true normal fasting glucose for a person eating a typical amount of carbohydrate is below 83 mg/dL. Low-carbohydrate eating may raise this number, but that doesn't necessarily indicate a pathological change. High-carbohydrate cultures such as the Kitavans, Aymara and New Guineans tend to have fasting values in the low 60s to low 70s. I suspect that a very high carbohydrate intake generally lowers fasting glucose in healthy people. That seems to be the case so far for Chris Voigt, on his diet of 20 potatoes a day. Stay tuned for an interview with Mr. Voigt in early December.

81 comments:

Anonymous said...

is the carbohydrate amount consumed and a post meal blood glucose as important as a persons glucose clearance rate. after spending a couple months priking my finger after meals varying in carb content from about 5g to 150g i found nothing consecutive or predicatble.

i did find that higher fat low carb meals left me with a fake insulin spike coupled with hypoglycemia. i found despite my ability to down 2 1lb sweet potatos ina sitting, my glucose didnt go over 105 but 2 oz of cheese will initially lower my blood sugar and later spike it high.

my point, its not necessarily always the carb content of a meal as much as the vitamin and mineral content it provides, and how well ones ability to have a high glucose clearance rate is. so maybe not 'what is you post meal blood sugar?' but what is your post meal clearance rate in relation to your cortisol pattern.

just a thought

Ryan J said...

Hi Stephan:

By chance, do you have the rough values for the amount of carbohydrate in the San's typical fibrous starches, as compared to a sweet potato/taro/potato?

Any idea how much fibre the San would consume daily, on average?

I'm constantly wondering about the higher fat eating HG's fibre intake, and how it'd affect their health.

It seems to me that even with a fairly high fat intake, the fibrous nature of these starches may still make it fairly easy to obtain the 70-100g fibre that Eaton/Cordain estimate.

PS - any gut bacteria stuff coming up soon? Fascinating stuff! Still just trying to figure out fibre's relevance(soluble in particular)

Chris Kresser said...

Excellent post, Stephan.

An interesting question I've been mulling is why some people on low-carb diets experience poor glucose tolerance once they've adapted to burning primarily fat - as might be expected - while others seem to retain their ability to process glucose efficiently.

I'll use myself as an example. I'd say I eat an average of 400 (100g) from carbohydrate each day. Yet when I challenge myself by eating a larger amount of carbohydrate than usual, I don't see any significant spike in blood sugar one or two hours post-meal.

On the other hand, I have patients with similar carbohydrate intakes and no history of diabetes or metabolic problems that experience spikes up to 175 mg/dL after a small amount of fruit juice or honey.

It's tempting to use the explanation that the body has adapted to fat metabolism in these cases, but if that's true (which is certainly plausible from a physiological perspective), I wonder why it seems to happen in some more than others?

I've been having all of my patients track post-meal blood sugars. I'll tell you one thing: I see a lot of patterns that just don't make much sense from a conventional perspective. Blood sugar regulation is incredibly complex and depends on multiple systems of the body. I suspect stress hormones - especially cortisol - play a much larger role in dysglycemia than previously acknowledged.

I've now started testing cortisol rhythms in patients with blood sugar irregularities that don't make sense. Almost without exception, they have some kind of cortisol dysregulation. Sometimes high, sometimes low, but most commonly the rhythm is off: it's low in the morning when it should be high, or high in the evening when it should be low.

Along those lines, I believe another potential cause of high fasting blood sugar is, somewhat paradoxically, low cortisol levels in the early morning hours. If blood sugar drops through the night, cortisol is supposed to kick in and raise it back up. But if there's not enough cortisol to do that job, epinephrine will take over. And epinephrine has a much more potent effect on raising blood sugar, which would explain the high morning readings. (It also explains why these folks can't sleep).

This is a fascinating subject, and I feel like the more I learn the more I have to learn. Thanks again for a great article.

Chris Kresser said...

Excellent post, Stephan.

An interesting question I've been mulling is why some people on low-carb diets experience poor glucose tolerance once they've adapted to burning primarily fat - as might be expected - while others seem to retain their ability to process glucose efficiently.

I'll use myself as an example. I'd say I eat an average of 400 (100g) from carbohydrate each day. Yet when I challenge myself by eating a larger amount of carbohydrate than usual, I don't see any significant spike in blood sugar one or two hours post-meal.

On the other hand, I have patients with similar carbohydrate intakes and no history of diabetes or metabolic problems that experience spikes up to 175 mg/dL after a small amount of fruit juice or honey.

It's tempting to use the explanation that the body has adapted to fat metabolism in these cases, but if that's true (which is certainly plausible from a physiological perspective), I wonder why it seems to happen in some more than others?

I've been having all of my patients track post-meal blood sugars. I'll tell you one thing: I see a lot of patterns that just don't make much sense from a conventional perspective. Blood sugar regulation is incredibly complex and depends on multiple systems of the body. I suspect stress hormones - especially cortisol - play a much larger role in dysglycemia than previously acknowledged.

I've now started testing cortisol rhythms in patients with blood sugar irregularities that don't make sense. Almost without exception, they have some kind of cortisol dysregulation. Sometimes high, sometimes low, but most commonly the rhythm is off: it's low in the morning when it should be high, or high in the evening when it should be low.

Along those lines, I believe another potential cause of high fasting blood sugar is, somewhat paradoxically, low cortisol levels in the early morning hours. If blood sugar drops through the night, cortisol is supposed to kick in and raise it back up. But if there's not enough cortisol to do that job, epinephrine will take over. And epinephrine has a much more potent effect on raising blood sugar, which would explain the high morning readings. (It also explains why these folks can't sleep).

This is a fascinating subject, and I feel like the more I learn the more I have to learn. Thanks again for a great article.

Eva said...

I wonder what voight's weight will do on the potato diet. I didn't see any mention of that, although apparently the goal is to eat enough to maintain weight. Then again, exclusion diets tend to make for weight loss all by themselves and that is one very exclusive diet! Also interesting to see the nutrient info laid out for his potato intake. His potato diet is only really low on a few nutrients.

Unknown said...

The !Kung San diet is interesting. Mongongo nuts are 40% PUFA. They do contain 10 - 20 times more Vitamin E than most other nuts are very rich in minerals. I'm not aware of any other hunter-gatherer groups which consume nuts as a staple. In spite of its nutritional value, there propably are some downsides to such a reliance on a linoleic acid rich food. Vitamin E supplementation has also been proven to have several negative health effects, such as increasing the risk of heart failure, casting doubts on the value of its antioxidative potential.

Justin said...

Excellent post Stephan! Chris's idea on chortisol rhythms being out of wack sounds interesting too. I wish there was an easy way to check these at home. Also, I seem to remember Matt Stone had low FBG while eating a high carb diet. I have to get a meter so I can conduct my own experiments. It would be interesting to see how those potentially stressful morning commutes effect BG levels with and with out carbs. Any suggestion on what type of BG meter to get? Thanks again for all of the hard work that you put into researching and reporting on your findings. It's one thing to research something it is another thing to spend the time to write up a post that everyone can understand. I always recommend your blog to anyone that will listen. If you are ever presenting anything again on the east side of the country I will be there.

Justin Cain

Evgeny said...

"and their typical daily fare is 94.6 percent carbohydrate"

Is it indeed possible to satisfy the needs for both the amino acids and the essential fatty acids in just 5.4% of food intake?

Amy said...

Very interesting post and comments.

Chris Kesser, I tested my blood glucose on low carb. My morning reading was high: 100 and I had trouble sleeping on low carb, especially in the beginning. The minute I went back to moderate carb, my morning reading went back to 70 and my blood sugar after eating never went over 114.

Also, on low carb I would have readings at 140. It would do this after I had a little ketchup with meat at a fast food place. I used to think the 140 reading was caused by too many carbs, but now I suspect it was caused by the fructose in the ketchup. There is at least one person in the fructose malabsorption world reporting higher blood glucose readings, that don't want to drop down after eating foods high in fructose. I think some people are sensitive to fructose and will have elevated blood sugar readings after foods high in fructose like honey and juice.

Thank you for the interesting discussion.

Chris Kresser said...

Justin: I recommend the Relion Ultima w/20 ct. test strips that can be purchased here. It's the cheapest and most reliable unit I've found yet. I don't like supporting Walmart as a rule, but I haven't been able to find it anywhere else.

Amy: I've heard anecdotal reports like your and have seen that kind of pattern in some of my patients. On the other hand, I also have patients that have FBG in the mid-70s on a low-carb diet that shoot up to 160-170 post-meal and >100 FBG if they eat any starch or fruit. If it were just a matter of fructose intolerance, they wouldn't react to things like sweet potatoes.

What's clear is that blood sugar regulation is driven by several factors that can vary widely from person to person. I'm less and less comfortable with making any broad statements about it for this reason.

As others have pointed out here previously, and in personal correspondence, micronutrient deficiencies (such as manganese, choline, etc.) may also play a significant role.

Kuntsa said...

Body's reaction to the glucose may depend on the way liver generates the blood glucose. Either liver generates glucose from its starch-like glycogen storage or it generates glucose from fats and proteins. Latter process, glyconeogenesis, also produces ketone bodies. Ĺštarting glyconeogenesis seems to relative slow and it may take several days.

Glucose is unsafe at any concentration, there is no limit where it becomes harmless. AGEs such as HbA1c are formed at reaction rate directly proportional to the glucose (and other saccharide) concentration. A single 60 minute peak at 300 mg/dL per day is not more dangerous than four 60 minutes peaks at 135 mg/dL, assuming the normal 80 mg/dL concentration.

Because upregulating glyconeogeneisis takes days, it does not make sense to downregulate it during a single glucose peak. If the carbohydrates from the food increase above the level sufficient to maintain blood glucose without glyconeogenesis I'd speculate that body first gathers sufficient glycogen storage in the liver and only then shuts down the glyconeogenesis.

I'd be more interested in HbA1c measurements from different cultures, and also what happens to them if the subjects would eat SAD food long enough. Such an unethical experiment would never get published, I bet. ;)

Chris Masterjohn said...

Awesome post Stephan. I definitely agree both insulin resistance and glucose tolerance can be adjusted homeostatically in the absence of pathology. So far what you've uncovered seems to indicate that substantial blood sugar spikes are not natural to healthy populations. It would be interesting in this respect to know how a typical !Kung meal affected their postprandial blood sugar rather than an OGTT.

I find Chris's idea of cortisol dysregulation very interesting too.

Chris

JamesSteeleII said...

I've wondered for a while whether carb refeeding every now and then might act in a hormesis type manner. If eating too much refined carbs eventually leads to insulin resistanc3e, and eating too little in some people also shows poor glucose tolerance, then maybe occaisional refeeding can give glucose homeostasis the kick in the arse it needs to do its job.

On the other hand if HIT training is on the cards that may be enough to maintain peripheral insulin sensitivity.

Scott W said...

Two comments:

As Laurent said, those nuts must give the !Kung quite a high percentage of omega 6 fats in their diet...as high as typical Western percentages that appear to cause metabolic problems over the long term? Who knows, and who knows if there may be other lifestyle factors that are compensatory.

=======

I also find it interesting when people vilify "refined carbohydrates" without quantifying what that means to them. One gets a sense that they have some vague notion that an unrefined carb is some kind of fibrous, slowly, digested thing. If they are equating "refined" with "high-glycemic-load," then those South Americans are practically mainlining glucose, because they certainly refine it to make it edible/non-poisonous...most cassava preparation methods appear to yield a very refined starch that probably breaks down nearly as fast as granulated sugar. And yet their glucose regulation has not suffered.

I think it would provide better clarity to use a term such as "industrially-refined carbohydrate" as it better communicates the inclusion of problematic grains, food additives and possibly poor fat sources. Still...there would be a range of health even within that description: a gluten-free cake donut deep-fried in lard would be "industrially-refined" but not all that different from deep-fried slices of plantain.

Scott W

stan said...

If you look at what sugars are considered worse for diabetics, they are higher in glucose. Fructose was found to be less less likely to raise blood sugar. This is why maltose, of all sugars, seems to be the worst, it is pure glucose.

Fructose is higher in fruits and vegetables. It is also sweeter than sucrose and lower in calories. Another thing about fructose is some people seem to believe it is something you only get in fruit. So they avoid fruit but the truth is you get the same fructose-glucose ratio in carrots and sweet potatoes as you do in bananas and peaches. Fructose therefore is not the demon some people want you to believe it is.

Helen said...

Great post, Stephan!

I'd like add some thoughts to this comment:

"In the context of a healthy diet and lifestyle (from birth), I suspect that nearly anyone can adjust to a very high carbohydrate intake without getting dangerous blood glucose spikes."

The process actually begins before birth, and perhaps before conception. There is evidence that fetuses exposed to high maternal glucose levels are more likely to develop insulin resistance later in life, and there may be other prenatal factors at work.

There also has been some fascinating research on epigenetic changes around food scarcity/abundance influencing predisposition to diabetes in grandchildren.

I've also noticed several articles about how novel chemicals in our environment can affect the development of insulin resistance and diabetes.

And finally, you did say "most" people. Some people have monogenic or mitochondrial forms of hyperglyemia (and sometimes hypoglycemia) that have nothing to do with diet or lifestyle, although their expression could be mitigated or worsened by various dietary, lifestyle, and environmental influences.

Since Type I diabetes is an autoimmune process, which may be triggered by a leaky gut caused by gluten and/or other factors, it does fall somewhat into the "right lifestyle" category, but as with Type II diabetes, you need the predisposing genes, too.

Fasting Guy said...

QUOTE:
"diabetics have reduced insulin signaling, and eating a typical meal can cause their glucose to exceed 300 mg/dL"

Oh, how well I know that. Sometimes my glucose meter seems to be very friendly, with a message saying "HI." Unfortunately, that's not a greeting, but a warning that my blood glucose is over 500.

3D Face Analysis said...

So they avoid fruit but the truth is you get the same fructose-glucose ratio in carrots and sweet potatoes as you do in bananas and peaches.

You're wrong. If you look at the USDA database, sweet potatoes actually have a low fructose-glucose ratio. Sweet potatoes that are cooked have an even lower ratio.

In South America, different investigators studied a group of native Americans in central Brazil that subsist primarily on cassava (a starchy root crop) and freshwater fish. Average blood glucose one hour after a 100g OGTT was 94 mg/dl, and only 2 out of 106 people tested had a reading over 160 mg/dL

I don't think they have good insulin sensitivity. Two people out of 106 is a lot for a reading over 160 mg/dL. They should test how many people have a blood sugar reading over 120 mg/dL. I bet it's a lot.

Also, I don't think cassava is a good choice for a staple food. Cassava is high in goitrogens, which can inhibit thyroid function.

Check out Stephan's other article on sweet potatoes and insulin resistance. It said that 3.8 percent of the population had a blood glucose over 160 from a 100 gram glucose tolerance test. That's a lot.

So I don't think sweet potatoes are also a good choice for a staple good. Sweet potatoes are relatively low in fructose, but they are lightly goitrogenic, so I'd be careful to consume them as a staple.

Peter said...

Sweet potatoes are the only starch that doesn't raise my blood sugar.

Colldén said...

Stephan, have you ever written some kind of summary on the macronutrient ratios of the healthy non-industrial cultures of the world? I remember you wrote in a comment some time ago that there were more examples of healthy cultures eating a relatively high-carb diet than cultures eating a high-fat diet, but have you elaborated on that anywhere?

Geoff said...

Hey Stephen,

Thanks for the post. Very interesting. I wanted to add something that might help you make sense of the Kung section.

Have you seen Robb Wolf's post on Gestational Diabetes? Here's the link if you have not seen it: http://robbwolf.com/2010/09/06/gestational-diabetes-what-constitutes-low-blood-sugar/

The reason I bring this up is that when the body us running primarily on ketone bodies, it can become insulin resistant since the organs are using ketones for their primary source of fuel rather than glucose. There's an adaptation period required for the organs to switch back over, prior to which extremely healthy people may test insulin resistant/glucose intolerant.

On his podcast, Robb has mentioned the Kung before, and I believe that they are a tribe that for a short seasonal period each year switch over to a 90% carbohydrate (high fructose) diet. The rest of the year they seem to be on a very high fat diet. So even within this maladapted population, they seem to be able to switch their metabolism over with relative ease and maintain their health.

Chris Kresser said...

More along these lines:

This study shows that glucose utilization drops by 30% after only 4-weeks on a ketogenic diet. It's not hard to imagine that number going up to 50% over a longer period of time.

This study suggests some individuals can function at BG as low as 40 mg/dL when sufficient ketone bodies are present.

It seems pretty clear that the body can and does adapt to different energy inputs. When fat is the primary energy source, less glucose is needed in the blood and a natural state of insulin resistance occurs.

Still, I wonder what accounts for the individual differences in this adaptation, as it seems more pronounced in some than others (in spite of similar carbohydrate intakes).

Ned Kock said...
This comment has been removed by the author.
Ned Kock said...

I agree with Chris on the complexity of the phenomenon, and the possible role of cortisol and other hormones. Indeed, it is not only insulin that regulates blood sugar levels.

In addition to cortisol, glucagon is also particularly important. Abnormally high glucagon levels lead to abnormally high blood glucose levels. Even when insulin levels are normal.

Also, as we talk about glucose, we cannot forget fructose. The reason is that many traditional cultures with good health consume both together in fruits.

Fructose by itself does not lead to any insulin response. On top of that, fructose is prioritized by the liver (and possibly muscle) for glycogen replenishment. Let me provide, as a reference, Parniak and Kalant's (1988) study (see link below), which is based on rat liver tissue, but I also have a reference (not with me at the moment) that deals with humans. The results are very consistent.

http://healthcorrelator.blogspot.com/2010/06/fructose-in-fruits-is-good-for-you.html

If your glycogen stores are depleted, the body will not make any fat based on fructose. Glycogen stores will be replenished, and insulin will not even come into the picture. If you do sprints or strength training, fructose in moderation is your "friend".

If you are sedentary, and on top of that eat constantly, the liver will still prioritize fructose processing, but it will make fat with it (releasing it in VLDL particles). When glycogen stores are full this is what happens.

One interesting factoid. In birds, diabetes type 2 is generally caused by abnormally high glucagon secretion. Insulin levels are usually normal in diabetic birds.

Stephan Guyenet said...

Hi Ryan,

Off the top of my head, I think their root vegetables were roughly 30-90% as starch-dense as a potato by wet weight. I don't remember how much fiber they contained, but it's in the book I referenced. They also get a fair amount of carb from fruit.

Hi Chris,

Thanks for the comment. I don't know what causes people to adapt differently to different carbohydrate intakes. I agree it's complicated and there are important differences between individuals. Perhaps cortisol plays a role. In the sleep post, I suggested a connection between lack of sleep and both disturbed cortisol rhythm and glucose intolerance.

Hi Eva,

Chris Voigt has already lost weight, but he's not focusing on that because he doesn't want this to turn into a weight loss craze.

Hi Laurent,

I also think it's interesting that mongongo nuts have such a high n-6 PUFA content. They are supposedly very rich in micronutrients, including vitamin E as you mentioned.

Hi Evgeny,

The investigators reported a protein intake of 3% and fat 2.4% over the course of 3 days' observation in the village. These people eat pork from time to time as well, but the investigators didn't catch them during a feast day. There may also have been insects that were eaten outside of the village, as is common throughout PNG. I also have a hard time taking those numbers at face value, considering they observed no evidence of protein deficiency and the men were fit and decently muscled.

Stephan Guyenet said...

Hi Chris,

Yeah, I also really wanted to see post-meal blood glucose in these people, but I found no data on that. Unfortunately we're left extrapolating from OGTTs.

Hi Scott,

I agree that the glycemic index probably isn't the critical harmful factor in refined carbohydrate. I also agree that the word "refined" is quite slippery and needs to be defined.

Hi Helen,

I agree that prenatal factors can probably influence glucose tolerance as well, and perhaps even how your parents and grandparents lived (via epigenetics).

Hi organism as a whole,

Those numbers are extremely low by Western standards, so to the contrary I think they suggest sweet potatoes are a good staple food.

Hi Collden,

I haven't, but Loren Cordain and his colleagues have estimated macronutrient ratios in hunter-gatherers in several papers.

Andreas said...

My perspective on this: Put very simple, insulin lovers blood glucose by 2 mechanism: 1 Helps Glut 4 (and perhaps similar proteins in non muscle tissue) to clear glucose from the blood. 2 down regulate/stops liver production and subsequent blood release of glucose. I have come to believe that it,s a matter of supply and demand. If the cells need glucose they will express glut 4 in a linear way, and I suppose that organ tissue, including liver have a similar mechanism. If one were to eat no carb (or very little) the body stops producing Glut 4 in muscle tissue and maybe something similar happens in other tissue as well. This explains the impaired glucose clearance in people following low carb diets. The reason could be that the tissue start expressing other proteins than Glut 4, that help too remove fat from the blood since the cells needs some form of energy. (it's been showed that Insulin resistant people have elevated level of intra muscular triglycerides, to compensate for the low level of glycogen perhaps).
I suppose that in the normal individual, glucose clearance ability will be present at sufficient level to handle the amount of carbs that are eaten.
If one were to overfeed oneself for prolonged period of time with more calories than the tissue would need, then maybe the tissue would down regulate Glut 4 and similar transport proteins since enough energy will enter the cell with less of these, due to higher blood levels. This situation is toxic to the body due the the level of blood glucose, and several alternative actions can be taken to lover blood glucose levels. Tissue expansion in form of more adipose (fat) tissue can help with clearance of glucose. And of course as a last resort it can be flushed out in the urine. This situation is what you see in the obese type 2 diabetic/insulin resistant individual.
This would explain why type 2 diabetes vanishes in a matter of days in people who have had gastric bypass. Since the stomach is reduced in size, the individual will have a very hard time ingesting the normal high amount of energy. This would lead to lover levels of available energy in the blood and subsequently will the cells again need more energy, than would passively leek from the blood and this would lead to Glut 4 and similar again being produced.
I admit that this model is somewhat simplistic, but it fits with my current knowledge.

Dune said...

Although it seems to be the case that people living an active= hunter gatherer/natural lifestyle can handle varying intakes of Carbs, this doesn't exactly translate to us in our modern lives. Sweet potatoes and Mongongo nuts are one thing. Coca Cola and French Fries with Ketchup is quite another, especially if taken with a side of no exercise. Also, as a new reader, I read some post you did on acidity and how you disagree with the theory and believe that the body can handle it. I would disagree, because while our ancestors were certainly adapted to handling the acidity of meat, they did not have to deal with things like Soda drinks with PHs of 2.5 I know people who drink 4 or more cups of soda a day. Our body is not able to handle that kind of acidity over the long term. People with more acidic saliva have more illness including cancer.

wjones3044 said...

Stephan,

I've been reading your blog for the last two months or so and have read many archived posts as well. Fascinating stuff. Yesterday's post intrigued me sufficiently to get me to purchase a glucose meter and I've been testing myself every hour or two since.

Basically my levels run from 70 to 100, fasting to post-meals. The one outlier is what intrigues me. After playing a soccer game last night (as goalie, not field player) I had two glasses of wine. Checked my levels and I was at 167 mg/dL! I wish I had tested myself before the wine. Now I wonder, was it the exercise or the wine? Cortisol effect or booze effect?

Curious to know your thoughts. Unfortunately, I won't be able to check myself for two weeks as I'll be going out of town and will miss my next games.

Note that I've been on a very low carb diet for almost a year.

When I got home I had another glass of wine, some beef and some cheese. An hour later my reading was in the mid 70s...

Peter said...

What are some examples of non-industrial societies where type 2 diabetes is common?

David Pier said...

If you are going to buy a glucose monitor, know that that information can enter your "permanent file" and mark you as a diabetic. Several years ago I bought a monitor for the type of self-investigation that Stephan is advocating. I bought it at Wal-Mart with my credit card, and a few weeks later I started getting mailings from pharmaceutical companies for diabetes medications, etc.
If I did it again, I would pay cash.

John said...

Weird, the comment I posted yesterday is gone. Blogger bug? I'll just try again.

According to an extensive insulin review, the idea that insulin works mainly by clearing/offloading excess blood glucose into cells seems seriously flawed. Here are the main points I took from that review: [1]

a) Insulin decreases blood glucose mainly by inhibiting the liver's glucose production. (Andreas made a similar point.)

b) Glucose uptake by cells is in a large part determined by the difference in glucose concentration between the blood and the cell (glucose concentration gradient): a high concentration difference leads to more cellular uptake.

c) Insulin actually reduces tissue glucose uptake in hyperglycaemic diabetics when given insulin (how's that for cognitive dissonance?). Reason: insulin causes reduced glucose production by the liver, which lowers blood glucose, leading to a smaller glucose concentration difference between blood and cells. This results in less cellular glucose uptake.
Note: these findings were already published in 1978. [2]

d) Insulin increases glucose transporter (Glut 4) availability at the cell's surface in muscle and adipose tissues. However, even without insulin there are more than enough available transporters for glucose transport into these tissues.

(Continued in next comment below)

John said...

e) High ketone concentrations cause tissue insulin resistance. Reason: cells capable of 'burning' the ketones to do so exclusively. Simultaneously, the glucose inside the cell is converted (glycolysis) for eventual burning.
However, ketones occupy the 'stove', so the conversion products (glucose 6-phosphate) accumulate, as they cannot be transported out of the cell. This continues until glucose conversion is impossible due to the high concentration of these products. At that point, glucose can still enter the cell, but it will leave it again unused, i.e., inflow = outflow.
Result: no more glucose usage by the cell regardless of insulin levels, i.e., insulin resistance.

(Continued in next comment below)

John said...

The latter point helps to understand the oral glucose tolerance test on a low-carb diet. And also why carb refeeding works: the carb clears the ketones, so cells start burning the previously accumulated glucose conversion products. When these run out, the cell will convert glucose extracted from the blood again for fuel, i.e., the 'normal' non-ketogenic situation returns.

Note: I found this review via James Krieger's insulin series, part 4 (at http://weightology.net).


John

References:
[1] Sonksen P. Sonksen J. Insulin: understanding its action in health and disease. Br. J. Anaesth. 2000 Jul;85(1):69-79. http://www.ncbi.nlm.nih.gov/pubmed/10927996
[2] Brown P.M. et al. Mechanism of action of insulin in diabetic patients: a dose-related effect on glucose production and utilisation. Br. Med. J. 1978 May 13;1(6122):1239-42. http://www.ncbi.nlm.nih.gov/pubmed/647213

John said...

Stephan,

My apologies for the split up comment mess. It seems that Blogger removed the comments when they were either too long, or when they contained user friendly clickable links (I probably used too many HTML tags in total, for italics etc.).

Stephan Guyenet said...

Hi Dune,

I agree that not all sources of carbohydrate are the same. My point in this post is that high-carbohydrate diets do not inherently cause damaging blood glucose spikes.

As far as the acid/base theory, it has nothing to do with the pH of the substance you're ingesting. For example, lemon juice is alkalinizing. It has to do with the mineral content of the food and how it's metabolized in the body. That being said, I think I was too hasty in my post on the acid/base balance. I'm still not convinced that the actual balance is important, rather than just getting enough of the alkaline minerals (which are important in and of themselves: magnesium, potassium, calcium), but it's a fine distinction. In any case, getting enough Mg, K and Ca is important and that means eating plant foods.

Hi Wjones3044,

The first thought that comes to mind is that your meter gave you a bogus reading. Did you double-check it? If the reading was accurate, it probably indicates that your liver decided it was a good time to dump glucose into the bloodstream. I don't know why that would be.

Hi John,

I saw your original comment in my Yahoo account but it never posted on the blog. I don't know why. Google has tightened its spam filters and it may have thought it was spam.

I know James thinks insulin's function to promote glucose transport from the blood is overrated, but I think he exaggerates a bit. The fact is that in a normal person, insulin promotes tissue transport of glucose and suppresses hepatic production of glucose, both resulting in decreased blood glucose. Both sides of the equation are important for blood glucose control.

Roberto said...

Hi Stephan

Have you ever encountered a study examining the glucose tolerance in non-industrial cultures with a high carbohydrate intake primarily from grains? It would be especially interesting to see in a culture eating non-refined gluten grains like rye and wheat.

Stephan Guyenet said...

Hi John,

I found your comments in my spam filter. I can post them if you want, but it looks like you got your message across in subsequent comments. Let me know if you want me to post them (minus duplicates).

Hi Roberto,

The Bantu culture I included eats grains, but I'm not sure of the exact proportion in that particular Bantu subgroup. Generally I think African grain-eating cultures probably have good glucose tolerance, judging by their low level of diabetes and associated diseases. I don't know about wheat and rye, as I don't have good data from any traditionally-living culture that eats them.

shawn57187 said...

Hi Stephen,
If glycation occurs when sugars react with proteins, wouldn't a consistently higher fasting blood sugar result in increased AGE formation? These high carbohydrate diets may be fine in the short term, but what are the long term implications regarding glycation?

Also, having recently watched a documentary on the Korowai & Kombai tribes ("Living with the Kombai"), it seems like many of these traditionally living high-carbohydrate cultures may also be undergoing long periods of fasting. The Korowai live predominantly off of starch harvested from sago palms growing wild in the jungle. But there are also days where food intake is absent or extremely reduced, plus they are extremely active. Would you not agree that activity level and intermittent fasting are confounding factors for these high carbohydrate diets?

Robert Martini said...

Stephan said: "In any case, getting enough Mg, K and Ca is important and that means eating plant foods."

Isn't it possible to get enough of these types of nutrients through animal products such as liver, bone marrow, eggs and other meats which most Americans don not find palatable?

Brooklyn said...
This comment has been removed by the author.
Brooklyn said...

Awesome blog!!!

Your blog is very interesting. Your postings are really exceptional and I love your blog.

I have added your link in my my blog ttp://denbarblog.blogspot.com/ under "Health and Food Blogs" Section, so that my blog visitors could also enjoy reading through your blog.

I would be glad if you could also place my link in your blog.

Thanks for your time.

Hope u too would like my blog...

Nick said...

It seems that the amount of carbohydrates eaten has nothing to do with whether someone becomes diabetic or not. Perhaps this is what Jenny Ruhl means when she says, 'you did not eat your way to diabetes'.

It begs the question, again - what is the cause of the higher rates of diabetes in industrialized nations (let's say during the '60s, when the research was done)?

Stephan, I think you speculated that refined sugar may be one of the main causes of 'diseases of civilization' in a prior post. Certainly, if food is a cause, we would need to understand what people were eating (and drinking) in the '50s.

You (and Taubes) also showed that non-industrial people who move to industrial societies (not necessarily just in the USA) begin to suffer from the diseases of civilization. This seems to indicate that genes are not the primary cause of the problem, but that something may very quickly change how they express themselves.

What is it? Carbohydrates load seems to be a red herring as the cause. Refined sugar intake, trans fats, alcohol consumption, refined grains and exposure to pollutants likely all increased dramatically when individuals transitioned to industrialized cultures.

New York said...
This comment has been removed by the author.
New York said...

It's an instalanche.

Congratulations Stephan.

Dave said...

Interesting, but you're missing the elephant in the room: EXERCISE.

The reason the NG peoples can eat all those carbs is they have to spend all day pounding those roots to make them edible.

Humans are poorly evolved for the amount of leisure highly productive societies like ours allow.

Stephan Guyenet said...

Hi Shawn and Dave,

Yes, exercise is certainly a factor and variable energy intake may be as well in some cases (although the Kitavans for example rarely experience food shortage). Exercise will be protective against metabolic disease no matter what the diet composition is, and was probably a factor in the health of all traditionally-living cultures including the Inuit and other low carbohydrate hunter-gatherers.

Hi Nick,

I think a variety of factors are converging and it's too simplistic to pin it on one thing. Changes in diet quality, including industrial foods like sugar, seed oils and white flour are a large part of the equation. Inactivity, lack of sleep, lack of sunlight and a poor psychological milieu are also probably contributing. There may also be a role of man-made toxins but that's difficult to assess.

MB said...

Your article makes me wonder if there is a muscle glycogen link. That is does elevated levels of glycogen reduce skeletal muscle and hepatic uptake of serum glucose, reducing the rate of clearance for the pathway of glycogen storage, resulting in clearance becoming dependent on hepatic conversion to fatty acids? I can think of two means to observe this, first by increasing the rate of glucose skeletal muscle by increasing skeletal muscle mass by making a larger reservoir, and the second is depleting the existing volume through exercise. This may potentially enhance the activity of both insulin dependent and independent glucose transport proteins.

This could be tested in an animal model by reducing skeletal muscle glycogen by using a thermogenic agent such as 2,4-DNP and sampling the serum glucose clearance rate vs glycogen content

John said...

Stephan,

For the record: my 5 points were based on my own analysis of the review paper, not James Krieger's. I did read his post first though, so I may have been biased.

About insulin's transport function being overrated or not. The review paper says this:
"It is now well established that what Schafer called insulin's 'chalonic' (or inhibitory) actions are the physiologically more important. Indeed, its autacoid (excitatory) action has recently been shown to be, on the whole, physiologically unimportant."

Sounds plausible to me: inhibiting a glucose producer seems more direct than waiting for the glucose to passively diffuse into cells, enabled by transport proteins. Even if insulin brings more transporters to the surface, the transport mechanism remains passive.
If a bathtub is about to overflow, I would turn off the faucet. Opening the drain would be another option, but that seems less efficient. Even if there were several drains.

However, I agree with you that both sides of the equation are important. I just didn't find that information in the sentence: "Insulin tells cells to internalize glucose from the blood". ;-)
But, to be fair, on August 28 2010 you wrote: "Insulin is a hormone that drives glucose and other nutrients from the bloodstream into cells, _among other things_".

About the comments in the spam box: feel free to delete them.


John

Stephan Guyenet said...

Hi John,

True, I didn't mention suppressing hepatic glucose production in the post, which was an important oversight.

Anand Srivastava said...

Could it be that the Mongongo nuts and the high Omega6 is to be blamed for the low glucose tolerance of the !Kung. They might not have discarded the Mongongo nuts because they do not have much carbs anyway. So the glucose intolerance is not of much consequence to them. Also the nuts provide them with a lot of energy easily.

If this is true then the nuts would not be a great thing for people eating more carbs.

DocPenn said...

In case no one mentioned it earlier, James V. Neel's Thrifty Gene hypothesis plausibly explains your observations.

karl said...

A couple of notes:

Insulin meters are not as accurate as you may think - and need a correction factor to compare with blood-lab levels.

Insulin is also the hormone that tells the body to store fat and induces hunger.

Dr. Davis has put out a BG goal of 110

David Pier said...

Which would you pick as the less harmful sweetener in small amounts, sucrose or glucose? It sounds to me that if one were to insist on sweetening things (such as ice cream) despite your sage warnings, than glucose would be less harmful.

Richard Nikoley said...

Hey Stephan.

We seem to cross subjects on post now and then. I had not read this before posting on one of my pregnant Paleo reader's OGTT failure (1st & followup) a few days back.

http://freetheanimal.com/2010/11/gestational-diabetes-and-paleo-you-horny-women-you.html

Very interesting, all of this.

Unknown said...

I can say that Interesting read and conclusions. Thanks for posting.Exercise is the absolute apparatus for combating glucose intolerance, as it increases the muscles’ direct for glucose.

Fort Worth real estate

Anonymous said...

Stephan said:

"I think a healthy human body is extraordinarily flexible in its ability to adapt to a very broad range of carbohydrate intakes, and adjusts glucose tolerance accordingly to maintain carbohydrate handling in a healthy range."

I agree with this. Excellent post.

@All using glucose meters.. they are WILDLY variable in readings. You must use three strips for each prick and then throw away the outlier - even that is still a guess. You can easily get +- 20 mg/dl between readings.

@Chris Kesser
Based on self-experimentation and experience of my readers, I think there is a big difference between 100g and say, 50 g a day carb intake. I feel and have BG results more like a carb eater on 100g/day, but on 50 g/day I have OGTT response typical of VLC eaters - 100g of potatoes gives about 155 at one hour..

@Anand
I too wonder if excess N-6 could affect IR in the Kung San. If we think excess N-6 is inflammatory and causes liver IR in our western diets, why can't the same effect be had with large intake of nuts?

I think the category of "refined carbohydrate" is meaningless. I do not think "refined carbohydrates" cause disease. I think specific carbohydrate sources like WHEAT and excess fructose from sugar or hfcs cause disease - in the case of wheat, it is of course the protein fraction in the form of gluten that causes much of the trouble, even if the lectins are chemically carbohydrates...

I think the only sense in which "refining" is contributing to disease is by enhancing the quantities eaten - more gluten with introduction of white flour and more fructose with refined sources of sucrose.

Does anyone fear refined fats like butter* or refined proteins? Is refined saturated fat dangerous and natural linoleic acid safe in all quantities?

*other than Dr. Davis!

Chris Kresser said...

Kurt:

I agree, re: variability, though I haven't seen quite as much in my patients using the Relion Ultima meters. They seem to be more accurate than most.

In any case, I don't focus too much on any one reading. I use the post-prandial testing to establish patterns. If the overall pattern is good, and there are one or two aberrant readings, I don't worry about it. However, if BG is consistently spiking after meals, or FBG is consistently elevated beyond what might be expected considering their diet, I start to pay attention.

Peter at Hyperlipid linked to this paper by Colagiuri and Brand Miller a while back. They hypothesize that insulin resistance is an adaptive mechanism that protected our ancestors from hypoglycemia during the Ice Ages when glucose was scarce and protein and fat comprised the bulk of caloric intake. While some populations later exposed to higher carb intake (i.e. Kitavans, Okinawans, etc.) adapted again, the "low-carb insulin resistant" phenotype is still predominant in many parts of the world, and when mixed with the Standard American Diet, is a blood sugar train wreck.

EL 66K said...

@Kurt

Butter may be somewhat refined, but not stripped off its nutrients. And there's no proof that protein isolates are any good. It's what rats are given.

Anonymous said...

@Chris

I've used four different meters - they all display substantial variability with the same stick - perhaps yours is some new technological advance but I would still encourage every one to measure at least three times if they expect less than 10 -15 mg/dl error.

Thanks for the link-I've read every one of Peter's posts and most of the papers he's referenced.

As far as the insulin resistant low carb phenootype, that is just a hypothesis. The IRLC phenotype certainly exists as a PRODUCT of the SAD - those who must keep low carb and can't tolerate higher carb intakes - but what evidence is there that there is a phenotoype that absent damage from the SAD is born that way? It is certainly possible, but remains unproven.

I agree with Stephan that most humans have a multifuel capable metabolism that over the intermediate term, and absent metabolic damage, is equally capable of using starch or fats as the predominant fuel source. For one thing, we know the adaptation to starch came after our arboreal existence but many hundreds of thousands of years before homo sapiens was under the selective pressure of ice ages. (R Klein The Human Career) Switching to ketone metabolism is likely phylogenetically a very old capability. The ability to produce pancreatic and salivary amylase is all that is needed to use starch. And we still retain the ability to eat fruit just like the common ancestor and our cousins P troglodytes.

So I think the idea that we went from fruit (common ancestor) to meat and starchy tubers (H ergaster) to meat (late paleolithic ice ages).... and the ability of the kitavans to eat lots of starch is a recrudescence seems highly implausible. It makes much more sense that we have genetically retained the multi-fuel capability all along, and those we find are completely intolerant of carbohydrate are actually just metabolically damaged, perhaps even epigenetically in utero.....


* I don't discount the idea that there may be genetic differences is tolerance of macronutrient extremes - say, northern european vs kitavan. I am just doubtful that genetic differences in "preferred" macronutrient ratios explains the DOC. I eat a very high fat diet myself - because I prefer it- but people I share genes with eat vey high carb and we have very little DOC in my family.

@66 ELK

Butter is indeed very highly refined, consisting of pure fat compared to the whole milk it is refined from. It has had all the lactose and all but a trivial amount of the protein "stripped" from it. Are these not nutrients?

Are you claiming that stripping of micronutrients is what causes fructose and pure linoleic acid to be bad for us - or could it be the fructose and the linoleic acid themselves that are bad?

Now do you get my point? The fat in butter is not harming you and would not even if there were no vitamins in the butter. Drinking a soup of pure amino acids equivalent to the ones your small bowel drinks up after digestion likewise is not harmful per se.

The idea that REFINING per se is some kind of evil dietary magic is nonsense. Look up "refine" in the dictionary.

Coconut oil, butter, a grilled steak, etc.. everything we eat has some degree of "refining" unless you kill it and eat it raw - even cooking is refinement.

Refined foods are bad for us when they increase the QUANTITY of something found in a BAD food, not just because micronutrients are stripped out.

Anonymous said...

apologies for the multiple posts - google is telling me the posts are too big, then I split them up and they are posted anyway!

feel free to delete the redundant posts, Stephan

Chris Kresser said...

Kurt:

I agree that there are a lot of unanswered questions here. As fascinated as I am by the theories, I'm much more interested in what it all means from a practical, clinical perspective.

For example, should we be concerned about a patient following a paleo, low-ish carb diet (50-100g/d) with a FBG of 95-100 mg/dL and normal A1c?

What about a pregnant woman following such a diet with a FBG of 95 mg/dL (technically past the cutoff for gestational diabetes) and normal A1c?

If Peter is correct, and we test those people with post-meal or OGTT, it's likely they'll fail. But is that a problem? I suppose it isn't provided their A1c and post-meal numbers are normal when eating what they normally eat. In that case, the only thing the OGTT tells us is that said person can't tolerate a 75g bolus of pure glucose.

I suspect that a FBG of 95 mg/dL does not carry the same level of risk in someone eating a paleo/LC diet as it does in someone eating a SAD. That's where A1c and OGT are supposed to fill in the gaps, but OGT may not be accurate for an LC dieter unless they've had 3 days of higher carb consumption beforehand.

That's why I prefer post-meal testing with the kinds of foods the patient typically eats. Aside from the problems with accuracy, this provides a much better idea of what's happening with the patient's blood sugar when they're eating the foods they normally eat - and should be eating.

If a patient has high OGTT with 75g of glucose, but normal post-meal readings with low-carb meals, then I'd say "don't drink 75g of glucose!" From a practical perspective, it's really neither here nor there why they can't tolerate the glucose (i.e. whether they're "fat adapted" or have the "IR LC phenotype") if the recommendation is going to be the same in either case: restrict carb intake.

Where I do want to do more investigation, though, is when I see elevated FBG, A1c and post-meal BG in a person following a paleo/LC diet. And I'm seeing this fairly regularly. I suspect cortisol dysregulation secondary to stress, inflammation and chronic infection. There's support for this in the literature, and I do see changes in BG when these mechanisms are addressed. But there are still many questions.

Anonymous said...

"should we be concerned about a patient following a paleo, low-ish carb diet (50-100g/d) with a FBG of 95-100 mg/dL and normal A1c?"

No, we should not.

"What about a pregnant woman following such a diet with a FBG of 95 mg/dL (technically past the cutoff for gestational diabetes) and normal A1c?"

No, we should not.

"If Peter is correct, and we test those people with post-meal or OGTT, it's likely they'll fail. But is that a problem?"

No, it is normal physiology and is well described.

"In that case, the only thing the OGTT tells us is that said person can't tolerate a 75g bolus of pure glucose."

I agree

"I suspect that a FBG of 95 mg/dL does not carry the same level of risk in someone eating a paleo/LC diet as it does in someone eating a SAD."

Definitely. Although I don't have to time to expand on it - you can be seriously misled by A1c in HFLC eaters as well!

"That's why I prefer post-meal testing with the kinds of foods the patient typically eats."

Of course


"Where I do want to do more investigation, though, is when I see elevated FBG, A1c and post-meal BG in a person following a paleo/LC diet."

Example: 40 g/day carbs FBG = 92, A1c of 5.6 and post meal bg of 120 this is all all NORMAL IMO.

same values but 1hr pp BG is 160 - that could be pre-diabetes, BUT IT MAY IMPROVE ON HIGHER CARB INTAKE!

Be very careful of using lab values from the world of the sick to "treat" those who are perfectly well.

Not saying this applies to you - but as with the obsession over lipoproteins, a lot of disease is being created by too much testing and over-interpretation of same. If your post meal BG is relatively normal after you've eaten healthy for many months, then you're done. Quit measuring it unless you suspect actual diabetes and are willing to start taking pills.

*good diet assumes no wheat, limited n-6 and fructose and adequate magnesium.

Anonymous said...

Kurt,

I am very curious about why A1c can mislead for an LC person. Please expand when you have time!

Anonymous said...

@toddhargrove

My impression from reading many posts by LC eaters with zero history of DM or Metabolic syndrome that have tested their A1c is that the A1c may run higher than some of the actual diabetics who eat a Bernstein like diet. I suspect differences in red cell turnover between DM and non-DM may account for the difference. Fructosamine (a glycation end product that is caused by glycation - not due to fructose per se..._) is not affected by red cell survival and can be used as a check on A1c.

My own A1c runs 5.6, but a 24 hr serum BG correlates with a level closer to 5.0 and my fructosamine is lower than predicted by A1c.

This is one example of using a lab value that is highly predictive in population studies but may be quite imprecise at the individual level.

For that reason I like a 24 hr BG on your usual diet more that relying on A1c levels. Then when it's normal, you can keep eating healthy and forget about the measuring and testing....

Stephan Guyenet said...

Hi Richard,

Thanks, I did read that post.

Hi Kurt,

I deleted your redundant posts as requested. Interesting discussion.

Chris Kresser said...

I agree on the inaccuracy of A1c as an individual marker. In my practice, I see a lot of people with borderline anemia. This can artificially lower A1c results because there's a shortage of hemoglobin to get glycated. On the other hand, even mild dehydration, which is also relatively common, can increase HGB levels and cause an artificially high reading.

I like the idea of using fructosamine as a reality check.

David Pier said...

Do any of you have a clear understanding of how the spleen decides when to take a red blood cell out of circulation? It apparently isn't A1c, and I doubt there is "time stamp" on each cell. My quick look at the research suggests maybe it is the degree of oxidative damage, affecting cell shape and flexibility? If so, an excellent antioxidant status (and lower pufa content in membranes) would result in longer circulating hemoglobin and hence the elevated A1c. I've been wondering this since my recent and first A1c test came in a little high-normal at 5.1%, while my fructosamine was right in the middle of the "normal" range.

John said...

Chris, Kurt, Stephan,

What about the following theory: at high ketone concentrations, OGTT results tell you something about the insulin sensitivity of the liver.

My reasoning is as follows: high ketone concentrations cause ketone oxidation to displace glucose oxidation. Therefore glucose 6-phosphate accumulates in tissue, until further phosphorylation of glucose becomes impossible. At that point the tissue doesn't utilize glucose anymore, regardless of insulin levels (insulin resistance). High insulin due to an OGTT will then only affect hepatic glucose production.

(Continued in next comment below)

John said...

In other words: an OGTT with ketones doesn't fail, it just provides slightly different information.
If you would also do an OGTT without ketones and subtract the results, this could provide some approximation of the tissue glucose uptake due to insulin.

Given that I'm no expert, I'm open to the idea that my theory is seriously flawed, or not new.
I also admit that this wouldn't be a very practical measurement method.
Feedback anyone?

John

JBG said...

"24 hr BG"

I didn't know what this meant and went to look it up.

There are gizmos available now that make it simple and easy to take a blood glucose reading every five minutes for three days running. (You can even buy one for $1000 or so plus $30-$100 for a sensor for each set of readings.)

Apparently the readings wander so that it's necessary to do a regular BG reading several times a day for calibration. And apparently the accuracy of regular BG readings isn't all that great, either.

But the 24-hour device provides a line graph whose representation of direction of increase/decrease apparently is reliable, so one can get a good picture of how the level behaves during the course of the day.

Docs can set you up to do such readings. It involves a subcutaneous placement of the small sensor plus wearing a cell-phone-size device that gets the data by telemetry.

Here are a few links with further info:

http://diabetes.webmd.com/continuous-glucose-monitoring

http://www.diabetesnet.com/diabetes_technology/continuous_monitoring.php#axzz16j1wz6Ql

http://www.dlife.com/diabetes/information//daily_living/Viewpoints/amy_august06.html

My description above is from a pretty quick session of book-larnin' so some correction/addition may be needed.

Unknown said...

I read many useful health tips about the glucose.I known about the content of glucose in various level of our body and also in food particles.

Health News

Anonymous said...

@David

You said: "If so, an excellent antioxidant status (and lower pufa content in membranes) would result in longer circulating hemoglobin and hence the elevated A1c."

This is my own rather self-serving speculation for lower red cell turnover with (low oxidative damage (esp High Fat relatively low carb) diets.
No proof AFAIK but makes some sense intutively.

It's good to remember that the population studies that use A1c to correlate to coronary events, etc (EPIC as blogged by Peter at Hyperlipid) are mostly populated by SAD eaters, not low carbers, HGs or Paleo or weston price enthusiasts.

It may be that A1c may be diet-context dependent in the same way that lipoprotein measurements like HDL and LDL are, and hence lack predictive value outside that context. I think this is very likely, in fact.

@Chris

Even in the anemic folks, I believe it is actually red cell lifespan that is affecting the A1c level, as A1c as calculated is dimensionless - it is calculated as the percentage of Hb that is glycated, which should be independent of the hemoglobin level....

Anonymous said...

@JBG

I am not advocating a fancy device - just hourly fingersticks while you are awake. 15 fingersticks and 50 or so test strips.

EL 66K said...
This comment has been removed by the author.
EL 66K said...
This comment has been removed by the author.
EL 66K said...
This comment has been removed by the author.
EL 66K said...

@Kurt

Severely refined foods (industrial) are indeed a problem when the diet barely meets the requirements for nutrients, or it plain doesn't. I doubt your health would ever be optimal in the long term with a diet of rice, RBD coconut oil and lean chicken. There may be some inherent problems of refining, or maybe there aren't, but it definitively can be problematic in situations like the SAD.

If mongongo nuts, which are so high in O6, can be consumed, apparently safely, by a unmodernized population, then more may be at play. Refined oils not only have loads of PUFAs, but by property of this, they deplete vitamin E (and have none to give), etc. The same may apply to other foods, don't you think?

Sorry for the erased posts.

David Pier said...

Stephan had a similarly high HbA1c of 5.8% in this post from a ways back (and we're pretty sure his average glucose is very good):
http://wholehealthsource.blogspot.com/2008/09/few-numbers.html

robrob said...

I can't explain the contradictions either maybe it is a combo of genetics, nutritional status as well as other factors not yet figured out? I was thinking based on my understanding one's calcium sat fat and vita d magnesium and chromium factors have a big bearing on glucose tolerance.

so maybe the disparities are due to low calcium or vita d status? those with good glucose tolerance despite starchy diets are that way because they are better able to get and hold on to minerals and vitamines? it is possible that some are genetically unable to absorb their calcium and vita d they need without sufficient saturated fat? (of course if they get alot of sun but don't darker skin people have harder time makeing vita d in their skin?)

we have to remember the tests results could of been skewed due to bad test materials or bad handling or just plain dishonesty in some cases.

glucose and fructose only cause harm if your already insulin resistant, I really don't like insulin resistance as a term to use here I don't believe the cells are resistant to insulin at all but resistant to glucose itself.

they don't have the nutrition they need to handle the glucose and thus it would posion the cells so they refuse it. calcium, vita d are needed as well as proper amounts of cholesterol, sat fat (for the lipid raft to transport glucose to the mitrochonria without oxidizing in cytoplasm damaging the cells.) so maybe the whole thing is dependant on your nutritional status and your toxic load.

just some random thoughts here.

Unknown said...

Some alternative practitioners believe that
leaky gut syndrome
is a possible starting point or connection with many disorders such as asthma, diabetes, autoimmune diseases like lupus, diseases like scleroderma, internal colitis, long term disorders like rheumatoid arthritis, severe illnesses like multiple sclerosis and chronic fatigue syndrome and Crohn's disea