Tuesday, April 13, 2010

Good Calories Bad Calories Is Not Necessarily Infallible


When Good Calories Bad Calories (abbreviated here as GCBC) was published in 2007, the low-carb community was ecstatic. Dr. Robert Atkins and the Doctors Eades had discussed the scientific basis for the low-carb lifestyle, but their writings were usually presented in the context of clinical observations. With GCBC, Gary Taubes gave low-carbers 460 pages of tightly reasoned discussion and another 113 pages listing many specific citations from the scientific literature.

For a layperson, the book was not easy to read, but with effort it was comprehensible. At last low-carbers had access to information that cast doubt on the hypothesis that excessive consumption of fat raises cholesterol levels, which in turn causes heart disease and early death. Taubes presented plausible evidence for an alternative hypothesis--that excessive carbohydrate consumption, not fat consumption, is the cause of diabetes, heart disease, hypertension and even cancer.

Since the publication of GCBC, two interesting things have happened. (A) GCBC has moved into the position of holy writ in the eyes of many low-carbers and (B) several low-carb blogs and forums have arisen to discuss the scientific and practical aspects of low-carbing.

A rereading of GCBC in 2010 shows that many of its ideas have been supported by the subsequent publication of prospective dietary studies, including Weight Loss with a Low-Carbohydrate, Mediterranean, or Low-Fat Diet, published in the New England Journal of Medicine. However, recent discussions in the blogosphere show that some statements in GCBC may need to be reconsidered.

Specifically, on page 394 of the hardbound edition of GCBC, Taubes states, "By the mid-1960s, four facts had been established beyond reasonable doubt: (1) carbohydrates are singularly responsible for prompting insulin secretion; (2) insulin is singularly responsible for inducing fat accumulation; (3) dietary carbohydrates are required for excess fat accumulation; and (4) both Type 2 diabetics and the obese have abnormally elevated levels of circulating insulin and a 'greatly exaggerated' insulin response to carbohydrates in the diet..."

Let's address these statements in order.

1. Although consumption of carbohydrates does prompt insulin secretion, it is a well-known physiological fact that consumption of proteins also prompts insulin secretion. The amount of insulin released in response to protein is about a third of that released in response to carbohydrate on a gram-for-gram basis, but the increase is still measurable. Dr. Mike Eades has an illustration of this on page 37 of the paperback edition of Protein Power. Scientific articles measuring the insulin release in response to protein can be found here and here. Insulin response to various foods in terms of 120 minute area under the curve can be found in Table 4 here.

2. Insulin release does promote the storage of fat in adipocytes, but it is not the only signaling protein that produces fat storage. Acylation Stimulating Protein (ASP) is secreted by fat cells and allows fat to be removed from chylomicrons and stored in fat cells. Acylation Stimulating Protein permits the body to store fat in the absence of insulin. The process is discussed here by Dave Dixon and here by Petro Dobromylskyj (Hyperlipid).

3. While it is difficult to accumulate excess fat in the absence of dietary carbohydrates, it is not impossible. On various discussion boards, a few zero-carbers have related anecdotal evidence that they gained weight while eating large amounts of protein and fat. From a theoretical perspecive, on pages 388-392 of GCBC Taubes goes into great detail about the necessity of glycerol phosphate for the storage of fat in adipose tissue. (Glycerol phosphate is the precursor to the molecule used as the backbone of a triglyceride, the storage form of fat.) On page 392 Taubes says, "Dietary glucose is the primary source of glycerol phosphate. The more carbohydrates consumed, the more glycerol phosphate available, and so the more fat can accumulate. For this reason alone, it may be impossible to store excess body fat without at least some carbohydrates in the diet and without the ongoing metabolism of these dietary carbohydrates to produce glucose and the necessary glycerol phosphate." This sounds logical. However, biochemists know that glycerol phosphate can readily be produced from protein via glyceroneogenesis. The absence of dietary carbohydrate in no way prevents the synthesis of triglycerides from a high-protein or even a high-fat diet.

(4) It is true that high insulin is often associated with type 2 diabetes, but it is important to remember that type 2 diabetics do not always have an excess of circulating insulin. Instead they have insulin resistance. If their body tries to control high blood glucose levels with excess insulin production by the pancreas, this can result in beta cell burnout and a patient who actually has less endogenous insulin production than a person without diabetes.

As described here the scientific method is an ongoing process. Good Calories Bad Calories is an excellent book and provides many good arguments for the low-carb lifestyle. But the scientific method requires that we keep testing and evaluating our hypotheses, and it is important to realize that not everything we read in GCBC will necessarily stand the test of time.

9 comments:

Charles R. said...

I'm trying to find the passage in GCBC where Taubes claims infallibility, and I can't seem to locate it ;=)

I can find passages where that claim seems to be being made by other researchers in the field, however.

What I always tell people when I get into a discussion about this or that specific point that Gary brings up is the only thing he claimed to be doing was bringing up an alternate hypothesis that needed to be tested.

From my reading, he didn't claim to have the ultimate truth, he only argued that there was little or no support in the research for the prevailing hypothesis ("gluttony or sloth" or saturated fat), and a fair amount for his hypothesis, and why not test it and find out if it's helpful or not.

It is already three years old, and has spurred a great deal of discussion, further research, and more scrutiny of the existing literature.

And there is now this incredible network of people out there talking about the research, drawing new conclusions and coming up with new hypotheses, like Peter you mentioned and Stephen over at wholehealthsource.com.

I have the feeling Gary thinks that's all great, as is someone coming up with questions like yours...That's what science is all about, seems like.

Stargazey said...

Well said, Charles R. I agree completely!

Pooja said...

Diabetes is a very common disease. The deficiency of insulin leads to type 1 diabetes and the inability of the body to absorb insulin leads to type 2 diabetes. It is commonly found in children and teens. There are many signs and symptoms like fatigue, weakness, excessive urination, weight gain or loss, itching in genital organs etc. Cholesterol, hypertension, obesity are few causes of type 2 diabetes. Regular exercise and right diet helps you to prevent
from diabetes. For more details refer Type 2 Diabetes

Stargazey said...

Thanks for the link, Pooja. Although the American Diabetes Association believes that cholesterol and obesity cause type 2 diabetes, evidence is mounting that a high carbohydrate intake may be the culprit. For the prevention and treatment of diabetes please consider the low-carb approach of Dr. Richard Bernstein.

Just as a technical point, insulin resistance means that a person is able to absorb insulin, but that his or her tissues don't respond to it as efficiently as they should. Welcome to the blog, though, and I hope you enjoy reading some of the previous posts on the science behind low-carbing!

Mike said...

Always awesome stuff, Stargazy. I do agree that GCBC is *not* infallible; Taubes, while having an eye for detail, has not covered ALL the details. A task like that would be near insurmountable.

I'd like to counter a few of your points with questions/comments of my own; in no way am I challenging, I am merely trying to grasp the concepts fully.

1)Protein does cause insulin release, but it also stimulates glucagon release from the alpha cells, especially in complete proteins from animal sources. This attenuates the insulin response. Both studies make mention on this, but it was not measured.

2) I got nothing on that, lol, but I do agree there's more to fat storage than just insulin. Master hormone? Sure. Single player? No way. Looks like I need to do more reading, thank you. ;)

3) While I cannot argue with this point either, would, in your experience, say that this point could be considered "splitting hairs" for the vast majority of the general population? I would suspect that only a genetic defect would exacerbate glycerolneogenesis (hyperglycerolneogenesis?) I have yet to see someone actually NOT lose weight from a lower carb, high pro/fat dietary change. I do get your point, though that Taubes is making it a black and white issue, which it is clearly not.

4) I agree, and a Type II with low insulin would be transitioning to type I as beta cells are completely spent. I don't remember Taubes clarifying this point; I was (perhaps wrongly) under the assumption when he talked about Type II, he was referring to beginning stages, during hyperinsulinemia.

I always appreciate the phenomenal content you post up, please keep up the great work!

Stargazey said...

Hi, Mike! Your'e right about the hairsplitting. Normally I wouldn't have addressed these points, but when I read them again, I realized that certain low-carbers have taken the first three of them as holy writ and are leading others astray as a result.

For instance, some people have used Taubes's point #1 to claim that if they avoid carbs altogether, they are no longer secreting insulin because they don't need it. This simply isn't true. If it were, type 1 diabetics could have used a zero carb diet to survive in the days before exogenous insuin was available. They couldn't and they didn't. To address your specific question, the secretion of glucagon is necessary to counteract the blood glucose lowering effect of the insulin, but it doesn't stop the insulin from being released.

On point #3, there are lots of people who do low-carb but who can't lose weight effectively on it. If their "satisfactometer" is broken, they will overcome the metabolic advantage and the satiety effect of low-carbing and eat enough calories to maintain an overweight condition. They don't succeed at weight loss until they find a way to decrease their caloric intake along with their carbs. Dr. Richard Bernstein uses Byetta and Symlin in patients like this and has some success.

On point #4, Gary Taubes was probably talking about the early stages of type 2 diabetes. It's important to realize that he is self-taught in this area, and that he is covering material that is normally taught in two years of medical school basic science. He got many things right in his book, but has a few holes in his knowledge base. On the other hand, he doesn't try to come across as a know-it-all and will probably be willing to change his views as more knowledge becomes available to him.

Thanks for reading here and taking the time to comment!

LynMarie Daye said...

Hi Stargazey!

The way I see it, Taubes doesn't claim to be infallible but some of his followers (for lack of a better word) seem to think he is. Some call GCBC their bible.

Now if I could ask you a completely unrelated question. According to this post by Dr. James Carlson, cholesterol synthesis only comes from glucose. I thought it could come from either glucose or fatty acids because they both yield acetyl CoA. Am I missing something?

Thanks in advance!

Stargazey said...

I agree with your assessment of Taubes. My quarrel is with those of his followers who don't seem to understand that his alternative hypothesis is not Scripture, but is subject to testing and revision just like any other scientific hypothesis.

As for Dr. Carlson, what shall we say? He took biochemistry as an undergrad and again in medical school and seems to think that that makes him an expert. Long story short, yes, fatty acids yield acetyl co-A and can be used for cholesterol synthesis. Sigh.

LynMarie Daye said...

Thanks for your reply! I was on a mission to figure this out yesterday LOL! In fact, last night my husband said to me "what are you obsessing about now?" :~)