Thursday, September 25, 2008

Nonalcoholic Fatty Liver Disease

Nonalcoholic fatty liver disease (NAFLD) is milder form of NASH, in which the liver becomes enlarged and accumulates fat. Ready for a shocker? The prevalence of NAFLD is thought to be between 20 and 30 percent in the Western world, and rising. It's typically associated with insulin resistance and often with the metabolic syndrome. This has lead some researchers to believe it's caused by insulin resistance. It's a chicken and egg question, but I believe it's the other way around if anything.

There are certain animal models of human disease that are so informative I keep coming back to them again and again. One of my favorites is the LIRKO mouse, or liver-specific insulin receptor knockout mouse. The LIRKO mouse is missing its insulin receptor in the liver only, so it is a model of severe insulin resistance of the liver. It accumulates a small amount of fat in its liver in old age, but nothing that resembles NAFLD. So liver insulin resistance doesn't lead to NAFLD or NASH, at least in this model.

What else happens to the LIRKO mouse? It develops severe whole-body insulin resistance, impaired glucose tolerance, high fasting blood glucose and hyperinsulinemia (chronically elevated insulin). So insulin resistance in the liver is sufficient to cause whole-body insulin resistance, hyperinsulinemia and certain other hallmarks of the metabolic syndrome, while liver and whole-body insulin resistance are not sufficient to cause NAFLD or NASH. This is consistent with the fact that nearly everyone with NAFLD is insulin resistant, while many who are insulin resistant do not have NAFLD.

In all fairness, there are reasons why NAFLD is believed to be caused by insulin resistance. For example, insulin-sensitizing drugs improve NAFLD. However, that doesn't mean the initial metabolic 'hit' wasn't in the liver. One could imagine a scenario in which liver insulin resistance leads to insulin resistance in other tissues, which creates a positive feedback that aggravates NAFLD. Or perhaps NAFLD requires two 'hits', one to peripheral insulin sensitivity and another directly to the liver.

In any case, I feel that the most plausible mechanism for NAFLD goes something like this: too much n-6 from polyunsaturated vegetable oil (along with insufficient n-3), plus too much fructose from sweeteners, combine to cause NAFLD. The liver becomes insulin resistant at this point, leading to whole-body insulin resistance, hyperinsulinemia, impaired glucose tolerance and general metabolic havoc.


Debs said...

Yikes. Throw in a dose of fat-soluble vitamin deficiency, and you've got one messed up liver.

So, does insulin resistance in the liver lead to insulin resistance throughout the body because the liver's been damaged to the extent that it can no longer play a protective/regulating role? Or is something else going on?

Food Is Love

Dennis said...

Do you know how long manufactured polyunsaturated fats have been a significant part of the western diet?

The Pima Indians became obese on a high carbohydrate reservation diet by the early 20th century; I don't know if they had fatty livers back then.

Great blog!

Scott W said...

I have seen a cause/effect argument made in low-carb books that the implementation of the food-guide-pyramid mentality led to increasing rates of obesity, metabolic syndrome, etc. I'm not totally convinced that this was a purely carbohydrate-induced phenomenon.

To expand on Dennis' question, I'm wondering if the increase of metabolic problems can be laid more squarely on the doorstep of an increased use of polyunsaturated fats and the effect their use has on carbohydrate metabolism, as opposed to the carbs themselves.

Confounding the "all carbs cause problems" hypothesis is the issue of carb source (grain vs. other).

From my understanding, the Pima indians consumed a diet that was not necessarily low in carbs as subsistence farmers pre-contact. They ate squash, beans and corn. To what percent of their diet, I'm not sure.

But post-contact, when their water was cut off, they were left to subsist on government rations, which were based on white flour and sugar...still carbs, but a radically different source from what they were used to.

I suspect that they were also exposed to vegetable oil at the same time, but not sure.

In any event, the Kitavan lesson comes back again...carb source and oil source appear to be more important than macronutrient ratios per se.


And yes...great blog that is willing to dissect any issue without bias.

Stephan Guyenet said...


The liver is important for insulin clearance and absorbing glucose after a meal. That may be why it leads to widespread insulin resistance over time.


Excellent point. I don't think we can lay all the world's woes on vegetable oil. There are a number of cultures that were destroyed without it. I looked back at USDA data, and it doesn't go past 1966 for vegetable oil. Added vegetable oils have more than tripled since then. I don't know what the trend looks like before that, except that we ate virtually no vegetable oil at the turn of the 20th century.

Stephan Guyenet said...


I agree with you completely. The type of carb, rather than carb itself, seems to be the problem. Add to that n-6 polyunsaturated oil and you have a major problem.

The Pima did indeed eat a high-carb, low-fat diet traditionally. They now eat more fat, which is what researchers like to latch on to. But the changes in the quality of carb and fat are probably the cause, in my opinion.

I'm not totally sure when vegetable oil arrived on the Pima reservation. I think the gov't rations initially included lard, possibly hydrogenated.

Pasi said...

I'd put my money on "de novo lipogenesis" and high pufa. Pufa+lipids makes unstable VLDL which is (for safety reasons) kept inside a liver cell. Fat in a hepatocyte increases its IR.

Stephan Guyenet said...


Sounds highly plausible.

Charles R. said...

Dr. Richard J. Johnson suggests it is fructose that's the culprit:

We have just published a paper in the American Journal of Physiology in which we fed animals fructose or starch for 4 months. At the end of 4 months the animals fed fructose had lost their ability to regulate their dietary intake (that is, they had become resistant to leptin, the hormone involved in satiety). When animals on fructose were switched to a high fat diet, they ate excessive amounts and became fat quickly. In contrast, the animals fed starch could control their intake, ate less food, and gained less weight gain after switching to a high fat diet. Thus one can view fructose as the fire, and high-fat diets and starches as the firewood.

Here's the link to the paper:

Clearly, what we have ended up with has different causes for different individuals, but the specific fructose hypothesis seems to explain a lot of what we're seeing, at least to my mind.

And certainly, if you see liver dysfunction as central to all of the problems, then a combination of high fructose combined with industrial oils would certainly be a good way to kill your might as well just slowly stick a knife into it.

Doesn't Peter of Hyperlipid say you want your liver to manage your blood sugar, and not your pancreas?

Sounds like a good plan.

Charles R. said...

RE: My previous comment.

Duh. Uric acid..kidney, not liver.

So he's saying it's fructose's damage to kidney function that's central to the problem?

Huh. Then there must be (in this hypothesis) a connection between the kidney impairment and the leptin resistance?

Stephan Guyenet said...

Thanks Charles,

I wasn't aware of the kidney connection. Maybe fructose and n-6 are "storage signals" from the environment. They both increase in the fall so maybe temperate creatures adapted to respond to them by storing fat for the winter. We were originally tropical but maybe we have enough of those pathways left that we can overstimulate them by consuming huge amounts of n-6 and fructose.

Regardless of the speculation, I do think fructose is part of the problem. I wasn't aware of the leptin effect; that definitely makes the story more interesting to me.

Gyan said...

Grain-eating cultures of old consumed very little fructose.

I think it is fair to blame grains-they are not responsible for West's excessive fructose consumption and a lot of grain and lentil eating cultures of today do not suffer from obseity and metabolic syndrome.

The main thing is that one must give proper attention to one's food and food needs to be prepared at home. The readymade stuff, protein bars, shakes, energy drinks are not sufficient.

Stephan Guyenet said...


I don't blame grains for everything. As a matter of fact, I think a moderate amount of non-gluten grains and legumes can be part of a healthy diet. It's critical to treat them properly however, by soaking, sprouting, and/or fermenting them. Those are the processing methods of healthy cultures.

Anonymous said...

"Do you know how long manufactured polyunsaturated fats have been a significant part of the western diet?"

Weston Price said that the children he was treating for severe cavities came from home where their primary foods were "highly sweetened strong coffee and white bread, vegetable fat, pancakes made of white flour and eaten with syrup, and doughnuts fried in vegetable fat." I think it is safe to say that vegetable fats (the high PUFA kind) have destroyed human health for a long time. They were mainly consumed by poor people until the 1950s and later, when the general public was brainwashed into thinking PUFAs were healthier than animal fats and tropical oils.