Friday, 4 July 2014

How low & very low-carbohydrate diets result in more weight loss than high-carbohydrate diets for people with Insulin Resistance or Type 2 Diabetes.

See The Battle of the Diets: Is Anyone Winning (At Losing?) for trials where insulin resistant people lose more weight on low-carbohydrate diets than on high-carbohydrate diets and insulin sensitive people lose more weight on high-carbohydrate diets than on low-carbohydrate diets.

If Gary Taubes's carbohydrate/insulin hypothesis of obesity is correct, everyone would lose more weight on low-carbohydrate diets than on high-carbohydrate diets. This isn't the case, therefore Gary Taubes's hypothesis is not correct.

Although insulin is involved, it has nothing to do with "Hormonal clogs" or "Insulin fairies".
The Aragon Insulin Fairy

The Energy Balance Equation

Change in Body Stores = Energy in - Energy out, where... 

Energy in = Energy entering mouth - Energy exiting anus, and... 

Energy out = BMR/RMR + TEF + TEA + SPA/NEAT

See The Energy Balance Equation to find out what the above terms mean.

People with Insulin Resistance (IR), Impaired Glucose Tolerance (IGT) & Type 2 Diabetes (T2DM) have excessive insulin secretion in response to meals (postprandial hyperinsulinaemia). See Hyperinsulinaemia and Insulin Resistance - An Engineer's Perspective.

People with Insulin Resistance (IR), Impaired Glucose Tolerance (IGT) & Type 2 Diabetes (T2DM) also have impaired/no 1st phase insulin response to a sudden rise in blood glucose level. This introduces a time-lag into the negative feed-back (NFB) loop that regulates blood glucose level. If the input rise-time is less than the time-lag in a NFB loop, the output of the NFB loop overshoots. This is standard NFB loop behaviour. See Control of overshoot for more information.

1. On a high-refined-carbohydrate or high-GL diet, blood glucose level rises rapidly, with a rise-time that's less than the time-lag in the blood glucose regulation NFB loop. Insulin secretion from the pancreas overshoots in a positive direction. The resulting postprandial hyperinsulinaemia results in down-regulation of insulin receptors in the brain, which reduces insulin action in the brain. When the insulin level eventually falls to normal a few hours later, the brain interprets a normal insulin level as hypoinsulinaemia. Hypoinsulinaemia results in ravenous hunger, as insulin is a short-term satiety/satiation hormone in the brain (leptin is a long-term satiety/satiation hormone in the brain). Ravenous hunger results in over-eating. Energy in increases. Postprandial hyperinsulinaemia also results in postprandial sleepiness. Energy out decreases. Body stores increase. There are also accusations of gluttony & sloth!

2. On a low-carbohydrate or low-GL diet, there are small fluctuations in blood glucose & insulin levels. There is no ravenous hunger. There is much less/no over-eating. Energy in decreases. There is no massive postprandial hyperinsulinaemia. There is much less/no postprandial sleepiness. Energy out increases. Body stores decrease.

In addition, there is a loss of water weight due to a loss of liver & muscle glycogen. This can be ~2kg in one day (it varies from person to person). Kidneys can increase their output of urine for hormonal reasons. This can increase water weight loss to ~5kg. See Why counting Calories and weighing yourself regularly can be a waste of time.

There are also other hormones involved. For a Facebook discussion with James Krieger that led to the updating of this post, see

In Metabolic Ward studies, food intake is tightly controlled, so postprandial hunger doesn't result in over-eating. Energy expenditure is also controlled, so postprandial sleepiness doesn't significantly affect energy expenditure. This is why varying Fat:Carb ratios (with Protein held constant) makes no significant difference to weight in a Metabolic Ward. See Energy intake required to maintain body weight is not affected by wide variation in diet composition.

Inter-personal variations in postprandial hyperinsulinaemia, postprandial sleepiness & energy out explain the inter-personal variations in weight gain seen under hypercaloric conditions. See Bray et al shows that a calorie *is* a calorie (where weight change is concerned).

Insulin Resistance can be fixed. See Insulin Resistance: Solutions to problems.

Type 2 Diabetes can also be fixed. See Reversing type 2 diabetes, the lecture explaining T2D progression, and how to treat it.

Aim to fix the problem. If it's impossible to fix the problem, a low-carbohydrate diet as an adjunct to medication is fine.


billy the k said...

"People with Insulin Resistance & Type 2 Diabetes have no 1st phase insulin response to a sudden rise in blood glucose level."

This was my understanding of the point of our 1st phase insulin response: upon detection of new incoming dietary glucose, to release that burst of stored insulin in order to rapidly suppress hepatic glucose output--otherwise the body will have to endure the double-whammy of 2 sources of glucose simultaneously entering the blood--exogenous + endogenous.

Folks who lack a decent 1st phase response probably can't fail to experience abnormal post-carb BG elevations.  Peter D uses the term "physiological insulin resistance", and we would expect that his particular VLC diet would result in a rather pronounced inability to tolerate dietary carbs.  Apparently, however, one can suffer the loss of 1st phase response on diets having considerably more carbs than ketogenic ones:

"In the past, it was recommended that the patient should consume
a diet containing 250g of carbohydrate or more, and calories for
maintenance of body weight, for at least 3 days before the [GTT]
test to obtain the best reproducibility of the test and to minimize 
falsely abnormal results.  Ingestion of smaller amounts of 
carbohydrate (150 to 200g) is probably sufficient to prevent plateau
or diabetic-type curves in healthy subjects.  Presently, no formal
dietary preparation is recommended UNLESS IT IS ESTIMATED 
CONSUMED.  A low carbohydrate intake for several days preceding
the glucose tolerance test OBLITERATES the first-phase immediate 
insulin response to administered glucose that is most important in
determining normal glucose tolerance.  In subjects who have been 
on a...decreased carbohydrate intake..., a diet containing 200g of 
carbohydrate or more should be taken for a period of one week to
ensure maximal glucose use during the glucose tolerance test."
[Stefan S. Fajans, MD. Classification and Diagnosis of Diabetes 
(Chap. 20) in: Ellenberg &Rifkin's Diabetes Mellitus. 5th ed. p.367]

Does this seem correct to you?

billy the k said...

Unfortunately, I can't provide a link to Fajans' piece--mine is a photocopy I made of various chapters in this reference book (@ Univ. of Arizona Medical Library).

I was very eager to find out about Phase 1 insulin response years ago when I read Bernstein's second book [Diabetes Type II (1990)], wherein he said that impairment of Phase 1 was the first sign of things going wrong, and this impairment made it impossible to attain normal BG levels after high carb meals, and that carbs were in fact responsible for this impairment.  

But this claim was the direct opposite of the evidence that H.P. Himsworth presented back in 1935:  [The Dietetic Factor Determining
the Glucose Tolerance and Sensitivity to Insulin in Healthy Men. Clinical Science (1935) vol.2; p.67-94]

"It will be seen from the insulin sensitivity curve that as
dietary carbohydrate is increased from 50g to 200g marked
increase in insulin sensitivity, and consequently of the ability 
of the body to dispose of carbohydrate occurs;...[the results
of his experiment] suggest that the maximum disposal of 
dietary carbohydrate by each unit of insulin will be obtained 
when a diet containing 200g of carbohydrate is given. For amounts
both above and below this figure more insulin will be required for 
each gram of dietary carbohydrate."

Seeing my own BG tests worsen following Bernstein's dietary advice, I was concerned that perhaps I was in the process of losing my pancreatic Phase 1 ability.  But I was relieved to read the following information about this from DeFronzo: (p.321):

"The loss of the first phase of insulin secretion has important
pathogenic consequences, because this early burst of insulin 
release plays an important role in priming those insulin target
tissues, especially the liver, that are responsible for the 
maintenance of normal glucose homeostasis.  However, loss of 
first-phase insulin secretion does not appear to be the primary 
defect responsible for NIDDM.  Recent studies have shown that 
progression from normal to IGT or to NIDDM is associated with a 
reduction in insulin sensitivity and an increase in insulin-secretion with and intact first-phase response.  Although the first-phase 
insulin-secretory response is characteristically lost in patients 
with well-established NIDDM, this defect does not occur UNTIL
RESPONSE, indicating that the defect is acquired and not 

I think you don't even need to be on a long-term ketogenic level of carb restriction to have kittens--upon losing [temporarily!] your Phase 1 insulin response.  A few days @ ~50g/day and if you so much as sniff a piece of whole-kernel rye toast your glucometer will register triple digits!

Nigel Kinbrum said...

It looks as though there is a "sweet spot" of carbohydrate intake.

Too little carbohydrate kills the 1st phase insulin response (mechanism unknown).

Too much carbohydrate for one's activity level overfills the liver, causing chronic hyperglycaemia & hyperinsulinaemia, depleting the insulin store used for 1st phase insulin response (I think).