Drunk Without Drinking

NAFLD, or nonacoholic fatty liver disease, is common in people with type 2 diabetes, especially when blood glucose levels aren't controlled. More than 50% may develop it.

And of course a high intake of alcohol can produce fatty liver disease.

Now researchers have discovered that more than half of patients with NAFLD have gut bacteria that produce alcohol from the sugar the people eat. There's apparently a specific bacterial strain of  Klebsiella pneumoniae that produces more alcohol than usual, and this strain was found in 61% of people with NAFLD, but only 6% of controls.

Mice fed these strains of K. pneumoniae developed signs of liver damage. When these mice got an antibiotic that killed K. pneumoniae, their condition was reversed.

One man studied had severe liver damage and a condition with the name Autobrewery Syndrome. It's normally caused by alcohol-producing yeast, but this man  had no signs of yeast infection. When tested on an alcohol-free high-carbohydrate diet, he had a very high blood alcohol concentration of 400 mg/L, or .04%. Legal intoxication is usually .08%.

Now, most people wouldn't produce as much alcohol as this man did on a high-carbohydrate diet. But the production of a smaller amount over a long period of time could damage the liver of a person who never drank alcohol.

This is another reason to avoid high carbohydrate diets. The bacteria seem to produce alcohol only when fed a lot of carbohydrates that can be broken down into glucose.

You can read the full study here.

They point out that endogenous alcohol production by particular bacteria is not the only cause of NAFLD: "It would be worth emphasizing that it has become clearer that NAFLD is a very heterogeneous disease and the findings here likely represent just one type of etiology."

But it's a fascinating finding and makes one wonder how many other unanticipated products of gut microbes contribute to disease.

YMMV

Here's a study confirming what most of us already know: YMMV, or Your Mileage May Vary. Some people prefer YDMV, or Your Diabetes May Vary.

Different people may have different reasons for getting diabetes. Some may have a lot of insulin resistance. Others may be pretty insulin sensitive but they just don't produce enough insulin. They do produce some, so unlike people with type 1, they can often get along without added insulin. And others may have a combination of these deficits.

In this study,  two groups were studied: Pima Indians from the Southwestern United States and Asian Indians from Chennai, India. They found that the Pima Indians tended to have a lot of insulin resistance (three times as much as the Asian Indians), but the Asian Indians, who were also older and thinner, tended to have defects in insulin secretion (three times less).

Unfortunately, when you're diagnosed, it's not common to have a lot of tests to find out exactly what is causing your diabetes. Most doctors tell you it doesn't matter, because they'd treat the disease the same way regardless of the cause. And a bunch of extra tests would be expensive.

However, if you come from an ethnic group that tends to have insulin resistance, that would probably be your major problem and you should focus on things like exercise and weight loss that can reduce insulin resistance. If you come from a group that tends to secrete too little insulin, it would make sense to focus on eating foods that don't require a lot of insulin, in other words, trying a low-carb diet.

Of course, many of us in the United States have a mixed heritage, so such studies would be less useful. Nevertheless, they might give hints about what kind of treatment to focus on.

Studying Rodents

Much diabetes research is done in rodents, mostly mice. But mice aren't humans, and they don't always react the same way as humans.

For example, mice have been cured of diabetes many times, but these cures don't translate into human cures. So should we abandon mouse research?

No. The mouse research makes suggestions for things that might work in humans, or in human cell cultures, and raising mice is a lot cheaper and faster than raising larger animals, so many more studies can be done.

However, more attention should be paid to how the mice are raised. Some people criticize mouse studies because they are not controlled for light intensity or electromagnetic fields, which can affect biochemical systems. Others criticize the standard mouse diets.

A recent editorial in the journal Nature discussed the problems with mouse diets. It focuses on obesity research, but obesity and type 2 diabetes are linked.

You can make mice obese pretty quickly by feeding them high-fat diets, which they love; their normal diet is relatively low in fat. Think of baiting a trap with cheese.  But the Nature editorial asks if such high-fat diets have relevance to human obesity that usually develops at much lower dietary fat levels than the mouse obesity.

The editorial also asks if the metabolism of mice raised on a very high fat diet is different from that of mice raised on more normal diets. It also points out that the very high (60%) fat diets usually used to make mice obese quickly have a lot less sucrose than a lower-fat diet would, and sugar has metabolic consequences too.

So too, the types of fatty acids in a diet can affect the metabolism, and the fatty acids in the commercial mouse diets may not be similar to those in a typical (if any diet is typical) human diet.

One thing the editorial didn't address is the fact that many of these diets use the cheapest ingredients available and often satisfy the carbohydrate goal by adding sucrose instead of some kind of healthier carbohydrate like whole grains.

This editorial obviously raises questions rather than providing answers, and at the moment it has no practical value for most people.

However, the fact that people are drawing attention to the quality of the mouse diets used in so much research is a good thing. Maybe better mouse diets will result in better research results.