The Atkins Decline – Hormonal Obesity Part XX

/, Health and Nutrition, Hormonal Obesity/The Atkins Decline – Hormonal Obesity Part XX

The Atkins revolution was well and truly underway by the mid 2000s.  Millions of people around the world were following the Low-Carbohydrate lifestyle.  Burger King introduced lettuce wrapped burgers to reduce the carb content.  Everybody seemed to know somebody who was on the diet and had lost a lot of weight.  The possibility of permanently reducing weight and all its associated health complications seemed within grasp.LC vs LF long term

But alas, it was not to be.  Longer term studies of the Atkins diet failed to confirm the hoped-for benefits.  Dr. Foster’s randomized controlled study of 307 patients failed to find improved weight loss in the low carbohydrate group.  The experience of this group was documented in the excellent book “Rethinking Thin” by New York Times writer Gina Kolata.  Both the low-fat and the Atkins groups lost weight, and both groups regained the weight.

A systematic review of all the dietary trials showed that much of the benefits of a low carbohydrate approach faded by 1 year.In the follow up to the DIRECT study, it was found that by 12 months, weight had started to regain in all patients, including the low carbohydrate group. Lc vs Lf longterm2

Some of this was due to the low compliance in both the low calorie as well as the low carbohydrate group.  Because calories are not restricted, many assumed that the Atkins diet was going to be easier to follow and therefore have better compliance.  It turned out that compliance with the diet was equally low in the low calorie as well as the low carbohydrate group.  Upwards of 40% of each group would abandon the diet by 1 year with no better compliance in the Atkins group.

In a sense, this was completely predictable.  Many of the foods that were restricted were indulgent in nature.  Cakes, cookies, ice cream, and other desserts.  Everybody knew that these foods are not healthy for us and no help to weight loss efforts.  But we continue to eat them, simply because they are indulgent.  Food is a celebration.  Every celebration in human history is accompanied by a feast.  This is as true in year 2014 as it was in year 2014 BC.  Birthday, wedding, graduation celebrations – what do we eat?  Cake.   Ice Cream.  Not pork chops.  Why?  Because we want to indulge.  The Atkins diet does not allow for this simple fact, which doomed it to failure.

This confirmed the clinical observation of most people that the Atkins diet was not a lasting one.  Millions of people abandoned the Atkins approach and the New Diet Revolution faded into just another dietary fad.  The company Atkins Nutritionals, founded in 1989 by Dr. Atkins filed for bankruptcy amongst heavy losses as the user fled.

What happened?

One of the founding principles of the low carbohydrate approach is that insulin is the key driver of obesity.  This fact seems solidly grounded in science as we have discussed before.  Carbs are the major macronutrient (out of fat, protein and carbs) that cause blood sugars to go up.  As blood sugar goes up, insulin is assumed to also follow.  Therefore, carbohydrates are assumed to play the major role in stimulating insulin release.  This is true.  However, we have not considered the possibility that food may increase the insulin levels without raising blood sugars.

The breakthrough was discovered in 1997, with the publication of the insulin index.  Measuring the insulin response of 1000kJ portions of food, it was no surprise that the foods that stimulated insulin the most were the carbohydrate rich foods, bakery products, and snacks and confectionary.  What was astounding was the fact that protein rich foods also significantly raised insulin levels.Insulin Index

Where most people assumed that there was a tight correlation between glucose levels and insulin levels, it turned out that only 23% of the variability in the insulin response was due to the glucose.  In other words, only 23% of the insulin response is determined by how high the blood sugars rise.  The vast majority of the insulin response (77%) is not related at all.  Since it is insulin, and not glucose that drives weight gain, this changes everything.  This is precisely where the glycemic index diets failed.  They were targeting the glucose response with the assumption that the insulin response mirrored the glucose.  But this is not the case.  You could lower the glucose response by diet, but you didn’t necessarily lower the insulin response.  In the end, it is only the insulin response that matters.

The recent discovery of the incretin effect partially explains things.  What has been recently appreciated is that the gastrointestinal tract, particularly the stomach and small intestine actively secrete hormones.  There are so many neurone in the GI system that it has sometimes been called the ‘second brain’.  The incretin hormones, Glucagon Like Peptide (GLP-1) and Glucose-dependent insulinotropic polypeptide (GIP) are secreted by the fundus of the stomach in response to food.  GLP-1 is known to increase insulin release by the pancreas.  The main role of GIP, found in the duodenum and jejunum (small intestine) is to stimulate insulin release.  Fats, amino acids and glucose all stimulate GLP-1 release and thus, increased insulin.  The amino acid glutamine may be a particularly potent stimulus.

Even non nutritive sweeteners, which have no calories at all, can stimulate the insulin response.  Sucralose, when studied in obese human volunteers, raises the insulin level 22% higher than water.  The GI tract and the pancreas can detect sugars through taste receptors similar to those in the mouth.  This may activate secretion of GLP-1 and insulin.

Furthermore, there is a well-known effect called the ‘cephalic phase’ of insulin secretion.  This refers to the fact that the body starts to anticipate food as soon as you put it in your mouth and long before any nutrients hit the stomach.  Studies have shown insulin release in response to sucrose and saccharine even if you swish it around your mouth and spit it out.

Insulin does not respond only to high sugars, but the incretin effect shows that fatty acids and amino acids also play a role in the stimulation of insulin.  Hold on.  This changes everything.  The simple implication is this.  If carbohydrates are not the only, or even the major stimulus to insulin secretion, then restricting carbohydrates may not have as large an effect on insulin levels as we may have believed.

Continue here to Hormonal Obesity Part XXI – Replace, Don’t Add Fruit

Start here with Calories Part I – How do We Gain Weight?

See the entire lecture – The Aetiology of Obesity 3/6 – Trial by Diet

2017-09-02T11:54:16+00:00 12 Comments

About the Author:

Dr. Fung is a Toronto based kidney specialist, having graduated from the University of Toronto and finishing his medical specialty at the University of California, Los Angeles in 2001. He is the author of the bestsellers ‘The Obesity Code’ and ‘The Complete Guide to Fasting’. He has pioneered the use of therapeutic fasting for weight loss and type 2 diabetes reversal in his IDM clinic.

Leave a Reply

12 Comments on "The Atkins Decline – Hormonal Obesity Part XX"

newest oldest most voted
Notify of

[…] Continue here for Hormonal Obesity XX – The Atkins Decline […]

Marty Kendall

This is really interesting stuff. I came across a more current food insulin index list in the appendix to a recent thesis at The thesis is excellent too and reveals some great benefits of looking at the protein content of food as well as just the carbohydrates.


This *is* intriguing – ‘Even non nutritive sweeteners, which have no calories at all, can stimulate the insulin response.’. That screws the notion of all diet products – would this apply to stevia?


Clearly the notion of diet products is screwed. It is based on the caloric reduction as primary (CRAP) theory. However, this does explain why diet products don’t work at all. It seems obvious that if diet products worked, we would already have been thin, considering the thousands of diet products available. Stevia is very new, so data are sparse.

With regard to your last paragraph: Have you looked at Brand-Miller et al. (2009): Food insulin index: physiologic basis for predicting insulin demand evoked by composite meals. Am J Clin Nutr 2009;90:986–92? “Interestingly, in the present study, the fat content of the mixed meal showed a significant inverse relation (r = -0.60) with observed insulin responses and was a more reliable predictor of insulin demand than the amount of carbohydrate.” “(…) protein (r = -0.04, P = 0.88; Figure 2E) and fiber (r = -0.46, P = 0.116; Figure 2F) showed no relation” “The calculated GL of the mixed meals… Read more »
This is a very subtle point. Fat acts as a protective factor – it reduces insulin in a mixed meal. So does fibre, as a side note. So, eating Low Carb but High Fat helps to reduce insulin. However, eating LCHF still raises insulin at least a little bit no matter how low carb and how high fat you go, unless you eat pure olive oil, for example. That means that it will still contribute to weight gain, although much less than a high carb meal. It changes everything, because the only way to get zero insulin rise is to… Read more »

I am not sure if there are specific studies out there focusing on LCHF dieting people’s insuline response, but the last graph and the commentary in below post suggest that even an OGTT might provoke much lower insulin response in these people.


Dr. Fung, is there any benefit to taking a tablespoon of coconut oil in my morning coffee on my 24 hour and 36 hour fast days? I feel like it really helps keep hunger away and makes the fasting days a breeze. I literally do not get hungry and feel alert and energetic and yet can sleep very well. Thanks!


What about your fasting regime for patients, I found that in your instruction Milk, cream, coconut oil and cinnamon are permitted
 ? I would understand coconut oil, pure fat not so much insulin reaction but milk and cream ? Doesn’t it cause insulin spikes? Thanks very much for answer.


Yes – we recommend cream over milk (more fat) in coffee, but there is a small spike due to the protein in milk/ cream. However, it is usually small enough to be acceptable.


[…] פוסט זה מוגש כשירות לציבור. הזכויות על התוכן שייכות לכותב של הפוסט המקורי. את הפוסט המקורי ניתן למצוא בכתובת הזאת. […]


So if proteins and whey can increase insulin but you are not consuming glucose, what are the effects of the higher insulin?