Over a dozen years ago, when I was a medical student, treatment of type 2 diabetes was very simple. There was only one family of oral diabetes medicine — sulfonylureas. In patients for whom the sulfonylureas failed, the only option was insulin injections. Sulfonylureas suffer from two serious side effects. They cause weight gain, an especially frustrating problem since weight loss is so important in diabetes. They can also cause blood sugar levels to become too low (hypoglycemia) which can have uncomfortable and dangerous consequences.

In 1994 metformin (Glucophage) was approved in the US, providing the first non-sulfonylurea oral treatment for diabetes. Glucophage does not cause weight gain, and, by itself, does not cause hypoglycemia. Its most common side effect is annoying but not dangerous — diarrhea and stomach upset. The most serious side effect, lactic acidosis, is rare and usually preventable. Glucophage was a boon for diabetics.

More recently, a third group of oral diabetes medications became available — the thiazolidinediones (TZDs) which include Avandia (rosiglitazone) and Actos (pioglitazone). The TZDs also do not cause hypoglycemia, but they cause fluid retention (edema) which can be dangerous in patients with heart failure or kidney disease. They can also cause weight gain, but I’ve always believed (perhaps incorrectly) that this was fluid weight, not a gain in fat, like with sulfonylureas.

Diabetes is a progressive disease, meaning that even with optimal treatment sugar levels slowly increase and more medication must be used to achieve adequate control. Many diabetics therefore eventually need more than one medication, and some eventually need all three of the above classes.

A large trial in this week’s New England Journal of Medicine helps doctors and patients decide which of these three classes of medications are best used first in newly diagnosed diabetics. Over four thousand recently diagnosed patients were randomly assigned to receive Avandia, Glucophage, or glyburide (a sulfonylurea). They were followed for an average of four years to see which medication, when used alone, would control their diabetes longest, and to follow side effects.

Avandia controlled sugars longest, keeping the glycated hemoglobin (a long-term measure of sugar control) normal for an average of 60 months, compared to 45 months with Glucophage, and 33 months with glyburide. This benefit, however, came with increased side effects. Avandia caused the most weight gain, 10.6 lbs on average. The patients on glyburide surprisingly gained only an average of 3.5 lbs in the first year, and then did not gain more. The group on Glucophage actually lost weight. As expected, patients on glyburide also had occasional hypoglycemia.

The authors conclude

Our findings confirm the value of metformin (Glucophage) as an initial treatment for type 2 diabetes and the greater efficacy of metformin than of glyburide.

Tangential Miscellany:

Some ideas are so preposterous, they never go away. The purported link between cell phone use and cancer always struck me as such an idea. This Forbes article cites a recent large Danish study showing no connection between cell phone use and the risk of malignancy. I hope this will be the last word on the topic. So go ahead and use your cell phone. It won’t give you cancer, though you might drive into a tree.