High Fat Diets: The Sweet Deception

Robert Lustig was recently interviewed by KQED and had this to say:

How do you get diet induced obesity in a rat? People say put them on a high fat diet. Garbage.

You can't get a rat to eat a high fat diet...unless you add 20% sucrose to the diet. Basically you're giving them cookie dough.

But if you give them lard they won't eat it. They actually lose weight on that because that's the Atkins diet for them. It's not palatable. They don't like it and they actually lose weight and their metabolic parameters improve.

The only way to get an animal; a rat, monkey, or any other animal, for that matter, to eat a high fat diet is to lace the fat with sucrose.

So the question is: which is doing the damage, the fat or the sucrose in that case? And the answer is both.

In a recent post entitled Diabetic nephropathy and the lost Swede at Hyperlipid, Peter links to a recent paper in PLoS which demonstrated a partial reversal of diabetic nephropathy in mouse models.

“These studies demonstrate that diabetic nephropathy can be reversed by a relatively simple dietary intervention,” conclude the authors of the paper.

Peter goes on to cite a Swedish paper - that the recent paper above did not cite - which demonstrated a reversal of renal failure by a low-carbohydrate diet.

While the PLoS study neglected to mention the Swedish paper, they referenced a 2007 paper entitled A high-fat, ketogenic diet induces a unique metabolic state in mice.

Bottom line for these researchers? “These data indicate that KD [Ketogenic Diet] induces a unique metabolic state congruous with weight loss.”



The diets were:

1) Control, or, Chow (16.7% fat, 26.8% protein, 56.4% carbohydrate (6.5% sucrose))
2) High Fat (45% fat, 24% protein, 35% carbohydrate (17% sucrose))
3) Ketogenic Diet (95% fat, 5% protein, 0% carbohydrate)

Wait, so which one is the “high-fat” diet, again? “High-fat” is 45% fat, while KD is 95% fat. It has been mentioned before at this blog, and warrants mentioning again: in science, you need well-defined variables.

When researchers compare a low-carbohydrate diet to a low-fat diet, what constitutes “low” depends on which team of researchers you ask. There is no well-defined number or percentage. Also, invariably, when the diets are compared, the low-carbohydrate diet is ad libitum, which is a fancy way of saying “all-you-can-eat.” The low-fat diet is restricted in calories, but more often than not, the researchers do not make this distinction throughout their articles. These are just a couple of examples to the limitations of the current definitions, and there are more, to be sure.

With a “high-fat” diet which happens to contain 162% more sucrose as a percentage of carbohydrates compared to the chow, perhaps the researchers and the people reporting the results should mention something about this? Considering when the researchers remove the sucrose and add fat (i.e., the ketogenic diet), the problems magically disappear, it just might warrant mentioning, and might bring into question the quite liberal use of the term "high-fat" in this context.

While Dr. Lustig and I can agree that we think it’s garbage to claim that we induce obesity by feeding mice, or humans for that matter, a “high-fat diet,” it seems we part ways regarding why the mice eat less on a ketogenic diet (i.e., a  high-fat, little-to-no-carbohydrate diet) and also regarding what is doing the damage. Lustig thinks its sucrose and fat? A little bit odd coming from the guy who said “It ain't the fat, people.” Perhaps I'm picking nits, but doesn’t this imply that you need to feed a human both sucrose and fat to induce damage? (I would also like to address the idea that the KD is not “palatable” and the absurd implication that dieters eat less out of boredom that many researchers, academicians, journalists, physicians, and authors alike invoke, but will save it for another post. To keep you “satiated” in the interim, here is a great post by Dr. Eades on the subject.)

In other words, if we had two arms of a randomized controlled trial (RCT):

1. 2500 kcal: 90% Fat; 10% Protein; 0-1% CHO
2. 2500 kcal 0-1% Fat; 10% Protein; 90% Sucrose;

Would both of these groups lose weight and improve their metabolic parameters? Let’s say prior to baseline they have an average BMI of 32, and they have been relatively weight stable in the past 6 months eating:

2500 kcal: 30% Fat; 10% Protein; 60% CHO (about 20% sucrose)

As Lustig has said, a calorie is not a calorie, and calories can be isocaloric, but not isometabolic.

In other words, the second arm of the RCT is taking in 90%, or about 2250 calories of sugar, and 1125 calories of fructose. This amounts to about 281 grams. Lustig has said to shoot for less than 50 gm.

Why would the fructose be particularly damaging? After all we’re talking about “fruit” sugar. How can something in fruit be bad for us? Sugar is natural. Have we just refuted the idea that sugar is toxic, poison, and evil?

As Gary Taubes pointed out in a recent interview on NPR, and Paracelsus (not sure if it was on NPR as well) in the early 16th century, the effective dose makes the poison.

One of the points biochemists will make is you can’t just say that something is natural.

A metaphor is I could lean on you and apply a gentle pressure for 10, 20 minutes, an hour, it’s not going to do you any harm. But if I take that same amount of pressure and condense it into a split-second, then I’m hitting you in the arm pretty hard, and it hurts, and I’m going to cause swelling and pain, etc.

And that’s what we’re doing when we consume these sugars in this quantity and especially in liquid-form, it’s like we’re hitting the liver with them, and the liver responds by going “ouch,” and turns it into fat and becomes a little insulin resistant, and then the rest of the body becomes insulin resistant, and over the course of months or years, you end up quite likely with this problem called metabolic syndrome, and all of these chronic diseases stem from there.

As Dr. Lustig pointed out in an interview with Jimmy Moore:

Basically, if you look at the stereoscopic model of fructose, you can’t even imagine how it can stay together as a ring. It’s going to fly apart almost on contact and that means that that reactive ketone group that’s in the fructose is going to react with just about any protein around and it’s going to contribute reactive oxygen species, which ultimately cause vascular damage. That’s basically you down the road to cardiovascular disease. If you’ve let that fructose get past your liver, you’re screwed.

We shouldn’t let researchers, or journalists, or professors, or anyone for that matter, off the hook when they implicate a “high-fat” diet in causing, or even curing, an ailment, without them making the distinction that a “high-fat” diet is a “high-fat-high-sugar” diet in this context.

It could be likely that the increase in sugar is the driver of disease, while the “high-fat” content is along for the ride. The results of the 90% fat ketogenic diets certainly seem to make a case for it.