Bergenstal and colleagues compared the efficacy and safety of sensor-augmented insulin pumps and insulin injections in reducing levels of glycated hemoglobin (HbA1c) in 485 patients with type 1 diabetes. Patients were randomized to receive pump (n = 244) or injection (n = 241) therapy. The groups were also stratified by age (7 to 18 years and 19 to 70 years). All patients received diabetes management training, and those in the pump group received online pump training as well as additional training at the sites.
The mean baseline HbA1c level in both groups was 8.3%. One year after study initiation, the mean HbA1c level in the pump group was 7.5% compared with 8.1% in the injection group. In general, individual HbA1c levels were lower in both children and adults in the pump group compared with those in the injection group. In fact, post hoc analysis of a subset of patients showed that HbA1c levels in the pump group fell quickly between baseline and three months and remained below those in the injection group for the remainder of the trial.
After one year, more patients in the pump group than in the injection group reached the target HbA1c level of 7% or less (27% versus 10%, respectively). In addition, a greater percentage of both adults (34%) and children (13%) in the pump group achieved the target HbA1c level than did adults (12%) and children (5%) in the injection group. Incidence of severe hypoglycemia and diabetic ketoacidosis was comparable between groups.