MINNEAPOLIS, February 7, 2008—A Minnesota-based project to lower the blood-sugar levels of diabetics returned startling results, causing the government to shut down part of the study, and causing physicians to wonder if they have been pursuing the wrong diabetes strategies.

The findings throw into question the central thrust of diabetes treatment, that lower blood sugar controls the disease and keeps patients safer.

The study shows that aggressive glucose lowering to a glycosylated hemoglobin target of less than 6 percent for type 2 diabetes was associated with excess mortality. Lowering theuir blood sugar using aggressive means actually increased their chances of dying from a heart attack or stroke.

The study was designed in part by University of Minnesota cardiologist Richard Grimm Jr., M.D., and included mainly diabetic subjects in the Twin Cities.

The trials showed there were 257 deaths in the intensive-treatment group and 203 in the standard-treatment group—an excess of 54 deaths, or three per 1,000 participants each year, since the 10,251-patient trial was launched six years ago.

The discovery prompted federal health officials to step in and stop one part of the three-part trial in the interests of the health of the subjects.

Grimm, who also practices and conducts research at Hennepin County Medical Center, described the results in a Star Tribune article as "very surprising, shocking." The finding reverses conventional thinking about type 2 diabetes treatment, which is that lowering blood sugar reduces the risk of death.

Irl Hirsch, M.D., a diabetes researcher at the University of Washington, told the Star Tribune that the results will be hard to explain to patients who have made tremendous efforts to get their blood sugar under control.

The trial goes by the full name Action to Control Cardiovascular Risk in Diabetes, or ACCORD. It was designed to test whether aggressive treatment strategies lessen the likelihood of heart disease. It is sponsored by the National Heart, Lung and Blood Institute.

The study involves 10,251 type 2 diabetes patients ages 40 to 82 at 77 U.S. and Canadian sites at high risk for heart disease for any reason, such as having high blood pressure, high cholesterol or being obese.

Study patients were assigned to one of three treatments: one comparing intensity of blood sugar control; another comparing intensity of cholesterol control; and the third comparing intensity of blood pressure control.

It is the first part of the trial, about blood sugar control, that has been shut down. The cholesterol and blood pressure components of the trial will go on.

About half the patients were placed on a regimen combining diet and exercise with drugs designed to lower blood sugar levels to that of the average person with diabetes, while the other half was put on a regimen designed to drive it closer to that of someone without diabetes.

The process was aggressive. Many subjects took four or five shots of insulin per day. Some used insulin pumps. Some measured their blood sugar seven or eight times a day. Some took medication to lower blood sugar, on top of medicines for other medical conditions. They came to a medical clinic every two months and had frequent talks with clinic staff by phone.

After about four years, about half of the participants in the intensive treatment group achieved blood sugar levels close to normal. About 50 percent in the standard group achieved levels close to the average diabetic.

New York Times story