Automated insulin delivery significantly reduced glycated hemoglobin levels in adults with insulin-treated type 2 diabetes compared to continuous glucose monitoring alone, according to findings from a 13-week, multicenter, randomized controlled trial.

Researchers found a mean adjusted reduced glycated hemoglobin (HbA1c) reduction of –0.6 percentage points in the automated insulin (AID) group compared to the control group (95% confidence interval [CI] = –0.8 to –0.4; P < .001). Their findings were published in The New England Journal of Medicine.

Three hundred and nineteen patients across 21 centers in the United States and Canada were enrolled in the trial. Participants were randomized 2:1 to receive either AID or their pretrial insulin regimen with continuous glucose monitoring (CGM). At baseline, mean HbA1c levels were 8.2% (±1.4) in the AID group and 8.1% (±1.2) in the control group. By week 13, HbA1c levels dropped to 7.3% in the AID group and 7.7% in the control group.

Time in the target glucose range (70–180 mg/dL) increased from 48% to 64% in the AID group, compared to 51% to 52% in the control group (mean difference = 14 percentage points; 95% CI = 11%–17%; P < .001). Mean glucose levels decreased from 194 mg/dL to 170 mg/dL with AID vs 190 mg/dL to 188 mg/dL in the control group (P < .001). 

"This difference represents a mean time in the target glucose range that was 3.4 hours per day longer in the AID group than in the control group," wrote lead study author Yogish C. Kudva, MD, of the Mayo Clinic, and colleagues.

AID remained in automated mode for a median of 93% of the time (interquartile range = 87%–95%). Of the 215 patients receiving AID, 98% completed the trial. Adverse events occurred in 30% of AID users and 18% of controls, with device-related hyperglycemia due to infusion-set failures reported in 6% of AID recipients. Severe hypoglycemia was rare, and no cases of diabetic ketoacidosis or hyperosmolar hyperglycemic syndrome were observed.

Investigators found that AID was effective across diverse patient demographics, including those new to using insulin pumps. Greater HbA1c reductions were observed in patients with higher baseline levels.

"The results of this trial suggest that previous experience with an insulin pump or in-depth training in carbohydrate-counting methods are not prerequisites to the successful and safe use of AID," the authors wrote. 

The results support AID as an effective intervention for insulin-treated type 2 diabetes, providing greater HbA1c reductions and improved time-in-range compared to CGM alone. Further studies are needed to evaluate long-term outcomes.

Full disclosures are available in the study.