Abstract
Introduction: This study evaluated 6-month effectiveness and safety of automated insulin delivery (AID) in comparison with multiple daily injections (MDI) in pediatric and adult type 1 diabetes (T1D).
Materials and methods: Individuals with T1D, aged 2–80 years, were enrolled across 32 international centers (in the United States, Europe, Canada, and New Zealand) and randomized 1:1 to AID intervention (MiniMed™ 670G or 770G system) or MDI with or without continuous glucose monitoring. Primary endpoints were change in mean HbA1c for participants with a baseline HbA1c >8.0% (Group 1) and percentage of time spent below 70 mg/dL (%TBR <70 mg/dL [<3.9 mmol/L]) for participants with baseline HbA1c ≤8.0% (Group 2), to show superiority of AID intervention versus MDI. Safety endpoints including rates of severe hypoglycemia and diabetic ketoacidosis (DKA), and difference in diabetes treatment satisfaction score were assessed.
Results: For Group 1, N = 56 participants (aged 29.4 ± 17.0 years) were randomized to AID intervention and N = 54 participants (aged 36.8 ± 19.6 years) were randomized to MDI. For Group 2, N = 73 (aged 37.4 ± 21.0 years) and N = 69 (aged 39.2 ± 19.3 years), respectively, were randomized to AID and MDI. Change in HbA1c (mean [95% CI] difference of −0.7% [−1.1, −0.3], P = 0.0002) and difference in %TBR <70 mg/dL (4.8% [−6.4, −3.1], P<0.001) favored AID intervention versus MDI. Rates of severe hypoglycemia (AID: 1.82/100 patient-years) and DKA (MDI: 3.52/100 patient-years) were low and met preestablished success criteria for safety.
Discussion: This large, international, multicenter randomized controlled trial demonstrates safety of the MiniMed™ 670G/770G systems. AID significantly improved HbA1c and time spent in hypoglycemia when compared with MDI, in both youth and adults living with T1D.
Clinical trial registration: https://clinicaltrials.gov/, identifier NCT02748018.