Intensive Glucose-Lowering Arm of Diabetes Study Halted Early

NIH's National Heart, Lung and Blood Institute recently discontinued one arm of a large, ongoing clinical trial of patients with diabetes and cardiovascular disease because of concerns raised by a number of deaths among study participants.

The Action to Control Cardiovascular Risk in Diabetes, or ACCORD, trial enrolled adults with type 2 diabetes who have an especially high risk for heart attack and stroke. The study was designed to study three different treatment approaches for these patients: intensive lowering of blood glucose levels (to below currently recommended levels for patients with diabetes) compared with more standard blood glucose-lowering therapy; intensive lowering of blood pressure; and treatment of hyperlipidemia using a statin plus a fibrate compared with statin therapy alone.

The trial's intensive blood glucose-lowering arm was halted when members of the study's Data and Safety Monitoring Board became concerned about the number of deaths in the group that underwent the intensive treatment. In this group, 257 participants have died in the past four years, compared with 203 deaths in the standard treatment group -- an increased incidence of about 20 percent.

The intensive treatment group had a target hemoglobin A1c goal of less than 6 percent, which is similar to levels in adults who do not have diabetes. The target for the standard treatment group was an A1c level of 7 percent to 7.9 percent, which is similar to what is achieved on average by U.S. patients who have diabetes. The trial was designed to investigate whether lowering blood glucose to such low levels would reduce the risk of cardiovascular events, such as heart attack, stroke or death from cardiovascular disease. Previous studies had suggested that lowering blood glucose levels in patients with diabetes to levels similar to those seen in adults without diabetes might reduce the rate of associated cardiovascular disease.

According to an editorial (10-page PDF; About PDFs) that was published in the Feb. 12 Journal of Clinical Endocrinology and Metabolism, however, the participants in these previous studies were mostly younger individuals or people who had recently been diagnosed with diabetes. That editorial, written by Irl Hirsch, M.D., a professor of medicine at the University of Washington School of Medicine, Seattle, pointed out that the ACCORD study participants were mostly older adults who either had existing cardiovascular disease or had additional risk factors for heart disease.

Study participants who were receiving intensive blood glucose-lowering treatment now will be treated to the same blood glucose levels as those already in the standard treatment group. No report on these findings has been issued, and the researchers do not yet know what may have caused higher death rates among participants in the intensive treatment group.

According to Mark Beard, M.D., an assistant professor at the University of Washington Family Medicine Residency, the best strategy for family physicians right now is to treat patients with diabetes who have known cardiovascular disease or additional risk factors less aggressively than younger patients without established disease. He points out that several other studies are in progress that may shed light on the ACCORD study results; only when the results of all the studies have been published will it be time to draw conclusions. Meanwhile, he urges family physicians to "use caution in lowering blood glucose to low or very low levels."

Beard also suggests that, once more information becomes available about this arm of the ACCORD study, the results may be better explained. "We need more analysis on when these deaths occurred," says Beard. Until information on whether study participants died after undergoing a surgical procedure, during hospitalizations related to hypoglycemia or because of other risk factors, for example, no definitive conclusions can be drawn, he notes.