I’m going to deviate from my original plan for the “Dangers of a Zero-Carb Diet” series to discuss a topic that came up in the comments to the first post.

Leonie’s Diabetes and the Rose Corn Oil Trial

What prompted this diversion is Leonie’s interesting comment from Wednesday’s post:

I developed diabetes several years after being on a low carb diet. Continuing low carb to manage the diabetes did not halt its progress. It has taken about 18 months of adding more carbs (60 – 100 gr/day) to my diet to bring my fasting glucose down by a couple of mmol and eating more carbs has also lowered my Hba1c and post meal spikes significantly. I wonder if the liver is another organ that may be affected by carbohydrate deficiency.

I had not heard of such cases before, or so I thought, but Dr. Deans in the comments reminded us that Peter at Hyperlipid had noticed two similar cases in the Rose Corn Oil trial. [1] (The Rose Corn Oil trial, of course, figures prominently in our book’s discussion of PUFA toxicity.)

In the Rose Corn Oil trial, there were three arms – a normal diet arm, a high corn oil arm, and a high olive oil arm. The normal dieters were expected to eat “fried foods, fatty meat, sausages, … ice cream, cheese, … milk, eggs, and butter” while the oil arms were supposed to restrict these foods and replace them with corn or olive oil.

Here’s what happened:

Four patients were removed from the trial for other reasons. Two developed non-cardiac thromboembolism and were given anticoagulant therapy. The other two were removed because of diabetes mellitus. One of them already had mild diabetes, but glycosuria increased considerably soon after he started oil. Oil was stopped and glycosuria disappeared. Oil was restarted, but was stopped a month later because heavy glycosuria recurred. The other patient, not a previously recognized diabetic, developed glycosuria with a diabetic glucose-tolerance test a few weeks after starting oil. [1]

The patients who developed diabetes came one from the corn oil arm and one from the olive oil arm. Likewise, the patients who developed thromboembolisms came one from the corn oil arm and one from the olive oil arm. No such disasters occurred on the “fatty meat” arm.

Since all three diets were similarly fatty, it doesn’t appear to be the quantity of fat that was the issue. Rather it was the type of lipid, or some micronutrient that was present in the animal and dairy foods but lacking in the plant oils.

For insight into what the problem might be, let’s look at how scientists poison lab animals.

Insights from Diet Animal Poisoning Research

You have to pity diet researchers. It takes 60 years for bad diets to poison humans enough to significantly raise mortality rates. Yet a diet researcher is supposed to gain a Ph.D. in 4 years (or in 5 while simultaneously obtaining an MD!), do a postdoc in 2 years, win a grant in the first years of an entry-level position with PI status, and then demonstrate productive results within the term of a 2-to-5 year grant. Deadlines are pressing: A study needs to start rats or mice on two diets, and have one diet produce much better health than the other, in considerably less than a two-year time frame.

Just comparing McDonald’s fast food with a Mediterranean diet won’t do. Two years later both sets of mice will die happily of old age, with no significant differences between groups. Peer reviewers judge you to have discovered no new results. No new results means no paper, no grant, no job.

So “diet” researchers first have to become experts at quickly inducing disease in rats and mice. Find a diet that poisons animals in a few months, compare it to another diet that doesn’t, and you have a paper. Look for variations that slow or hasten the poisoning, and you have more papers. To be a highly productive scientist, one must be a skilled animal poisoner.

Various techniques have been developed for this purpose, including: knocking out some crucial gene; breeding a mutant strain that naturally develops disease; giving the animals poison with their food; or depriving them of crucial nutrients. Almost every study of diet in mice or rats uses one of these techniques.

If a missing nutrient can cause diabetes within a few years for Leonie and 12 to 18 months for the Rose Corn Oil trial volunteers, it’s likely to be pretty good at inducing disease in animals too. There’s a good chance diet animal poisoning researchers have already stumbled upon it in rats or mice.

Choline Deficiency Diseases

One of the most popular deficiency diets among researchers is the choline-deficient diet. A useful paper by Dutch scientists [2] gives a nice look at the impact of choline deficiency on rats.

Choline deficiency (CD) by itself induces metabolic syndrome (indicated by insulin resistance and elevated serum triglycerides and cholesterol) and obesity.

A combined methionine and choline deficiency (MCD) actually causes weight loss and reduces serum triglycerides and cholesterol, but induces more severe liver damage. The MCD diet prevents the body from manufacturing choline from methionine, vitamin B12, and folate, so MCD diets severely reduce choline levels; and without choline VLDL particles are not produced. Without VLDL particles, fats and cholesterol are trapped in the liver and never reach the blood and adipose cells.

Here is a measure of insulin resistance on the two diets:

The induction of insulin resistance by the CD diet is very rapid, requiring less than a week.

Induction of insulin resistance is thought to be mediated by elevated TNF-alpha production by adipose cells and by hypertriglyceridemia. Since the MCD diet neither raised serum triglycerides nor caused obesity which induces TNF-alpha production in adipose cells, it did not cause insulin resistance.

What Does This Have to Do With Diabetes?

Insulin resistance is a key step in the development of diabetes:

• Insulin resistance in the liver causes the liver to release more glucose into the blood (since insulin inhibits glucose release by the liver). This is discussed in a nice paper [3] found by LynMarie Daye and cited in the comments by CarbSane.
• Peripheral insulin resistance means that the rest of the body is less sensitive to insulin. The pancreas has to produce more insulin to dispose of the excess glucose that the liver is releasing.

This elevation of insulin and glucose levels is a crucial step toward diabetes; it is “pre-diabetes.”

Persistently elevated glucose levels can then poison the beta cells of the pancreas, diminishing insulin secretion capability and causing diabetes. [4]

The Rose Corn Oil trial was not a low-carb diet, so postprandial glucose levels could easily have risen to toxic levels.

If a CD diet can cause insulin resistance in a week, it’s plausible that it might cause diabetes in 12 to 18 months, which is when the Rose Corn Oil trial patients developed it.

What About the Thromboembolism Cases?

MCD diets induce fibrinogenesis. In the blood, excess fibrin formation leads to clotting, and clots can block vessels to cause thromboembolisms. It may be that the thromboembolism cases in the Rose Corn Oil trial had methionine, folate, or B12 deficiencies to go with their choline deficiency.

Why Do Plant Oils Induce Diabetes But Not Animal Fats?

So why did diabetes develop in the corn and olive oil arms of the Rose Corn Oil trial but not the “fatty meat and dairy” arm?

Well, look at the choline content of these foods:

Choline content of one cup (~200 g) oil or fat or 227 g (1/2 lb) meat

Source: http://nutritiondata.com.

Take away meat and dairy and replace them with plant oils, and it’s very easy to have a choline deficiency.

What Does This Have to Do With Zero-Carb Diets?

Maybe nothing … without carb consumption, postprandial glucose levels are not as high, and beta cell poisoning is less likely … but it may be that a zero-carb diet aggravates a choline deficiency in some fashion. I will leave this as a topic for further research.

UPDATE: Leonie in a new comment gives us more information: she has PCOS, goiter with nodules, and auto-antibodies. This suggests autoimmunity as a more likely explanation for her zero-carb diabetes.

Conclusion

In the book, we recommend the use of animal fats such as beef tallow for cooking, and recommend that pregnant women and vegetarians supplement with choline. We thought seriously about recommending that everyone supplement choline, but were reluctant to recommend too many supplements.

In retrospect, we should have recommended choline supplements for everyone who is overweight, has elevated blood glucose or lipids, or has elevated liver enzymes.

We have been using beef tallow as our cooking oil for several months now. It might be good practice for everyone to favor animal fats like beef tallow over plant oils for cooking.

References

[1] Rose GA et al. Corn oil in the treatment of ischaemic heart disease.  Br Med J. 1965 Jun 12;1(5449):1531-3. http://pmid.us/14288105.

[2] Veteläinen R et al. Essential pathogenic and metabolic differences in steatosis induced by choline or methione-choline deficient diets in a rat model. J Gastroenterol Hepatol. 2007 Sep;22(9):1526-33. http://pmid.us/17716355.

[3] Sonksen P, Sonksen J. Insulin: understanding its action in health and disease. Br J Anaesth. 2000 Jul;85(1):69-79. http://pmid.us/10927996.

[4] Leibowitz G et al. Glucose regulation of ?-cell stress in type 2 diabetes. Diabetes Obes Metab. 2010 Oct;12 Suppl 2:66-75. http://pmid.us/21029302.