Death risk halts diabetes study

Sharp blood sugar cuts were perilous

A major national diabetes study was halted Wednesday after researchers found that cutting blood sugar to near normal levels in certain patients actually increased the risk of death.

The findings raise questions about the best way to manage patients with the disease.

"The thought was maybe what you need to do is get even [lower] with blood sugar and get these patients as close to normal as you can get -- conventional wisdom isn't always right," said Dr. Mark Feinglos, chief of endocrinology at Duke University Medical Center and an investigator in the study. Duke is among the largest recruitment sites in the state with 174 patients enrolled in the trial.

About 18 million Americans -- about 500,000 North Carolinians -- have diabetes. Type 2 accounts for about 90 percent of cases, and cardiovascular disease is one of the most common complications of the disease. The Action to Control Cardiovascular Risk in Diabetes, or ACCORD trial, sought to reduce heart disease risk in patients at the greatest risk of it.

The federal government said Wednesday that it would stop one part of the large trial, which involves 10,251 patients with advanced diabetes. Data showed that Type 2 diabetes patients who had had intensive therapy of pills and insulin to reduce blood sugar levels were more likely to die than patients on standard therapy. Several smaller studies had suggested that reducing blood sugar to near normal levels cut the risk of death.

Study leaders, including Dr. John Buse at UNC-Chapel Hill, said Wednesday that 257 patients receiving aggressive treatment to cut blood sugar have died, compared to 203 deaths among patients in a control group. There were about 14 deaths per 1,000 patients in the intensive group, compared with 11 deaths per 1,000 patients in the standard therapy group.

Deaths do not appear linked to any of the study drugs, including Avandia, a GlaxoSmithKline diabetes drug that has been linked to cardiovascular problems. Last fall, the U.S. Food and Drug Administration issued sterner warnings for the drug's label.

Study leaders emphasized that the findings apply only to patients similar to those in the ACCORD trial and urged people with diabetes to keep taking medication to control their blood sugar unless otherwise advised by their physicians. To qualify for the study, patients must have had Type 2 diabetes for at least 10 years, have higher blood sugars than most other people with the disease and known heart disease or multiple risks for it.

"This isn't directly relevant to 99 percent of people being treated for diabetes," said Buse, who is medical director of the Diabetes Care Center at UNC Hospitals and the American Diabetes Association's president for medicine and science.

Patients in the intensive therapy group often took as many as three different pills to manage blood sugar and four shots of insulin a day, Buse said. A more typical treatment regimen for someone with Type 2 diabetes might include one or two tablets and a single shot of insulin.

"Not very many doctors in the U.S. would even attempt this," Buse said of the regimen. "Very few people are being treated like this."

ACCORD seeks to reduce the risk of heart disease among people with diabetes at the greatest risk by better controlling blood sugar, blood pressure and blood cholesterol levels. The news reported Wednesday affects only the blood sugar arm of the study. The blood pressure and blood fats portions will continue until June 2009, when the study concludes.

Patients on intensive therapy will transition to standard treatment, which allows for higher blood sugar levels, and will continue to be monitored as part of the clinical trial. If patients' blood glucose levels are below standard recommendations for patients with advanced diabetes, they will be managed until they are up at appropriate levels, Buse said.