Subanalysis Sheds Light on ACCORD Mystery

Baseline characteristics linked with increased risk for severe hypoglycemia included African American race, male gender, increased age, and longer diabetes duration. Higher BMI actually protected against severe hypoglycemia, presumably owing to greater insulin resistance. Insulin treatment at baseline nearly doubled the risk of severe hypoglycemia in the intensive group, but quadrupled it in the standard group (BMJ 2010;340:b5444).

Among just the patients with no severe hypoglycemic events, the mortality risk was increased by 25% in the intensive treatment group compared with standard treatment, similar to the 22% increase for the entire intensive treatment group, suggesting that severe hypoglycemia was not the reason for the increased deaths in the intensive group, he said.

Importantly, the higher the average HbA_1c achieved and maintained during the trial, the more the incidence of severe hypoglycemia increased. The risk of hypoglycemia was greatest among those whose HbA_1c barely declined in the first 4 months of treatment, and was least among those whose HbA_1c fell rapidly.

Finger-stick data implied that the occurrence of severe hypoglycemia identifies patients with type 2 diabetes at increased risk for death, particularly in those whose HbA1_c does not respond to intensification of treatment, Dr. Genuth concluded (BMJ 2010;340:b4909).

More ACCORD data are due to be published during 2010, including results on microvascular outcomes.

Disclosures: ACCORD was funded by the National Institutes of Health and the Centers for Disease Control and Prevention, with supplies and medications contributed by 13 companies. Dr. Ismail-Beigi is a consultant to Eli Lilly. Dr. Seaquist and Dr. Genuth reported no relevant disclosures.