Efficacy of automated insulin delivery after recent pump therapy initiation in type 1 diabetes.

Abstract

Introduction Automated insulin delivery (AID) has been shown to be effective in routine care in type 1 diabetes (T1D). However, its effectiveness without prior insulin pump therapy lasting more than 6 months remains poorly investigated. Material and methods This prospective, multicenter study assessed the 6-month AID effectiveness in patients with T1D for more than one year, not previously treated with an insulin pump, with HbA1c level <10% and aged over 8 years (y). The primary study endpoint was the change in time in range (TIR, 70-180 mg/dl) from AID initiation to 6 months. Results A total of 43 patients with T1D (38 adults, 5 aged under 18y) were enrolled in 9 centers. Mean age was 37±16y, 56% were male, T1D duration was 14±12 years and baseline HbA1c level was 7.6±0.8%. Patients initiated AID (Medtronic® MiniMedTM 780G), after 2.9±1.0 months of non-automated pump therapy. After 6 months of AID (n=39), TIR increased by 17.6±11.2% (p<0.0001), % time >180 mg/dl decreased by 15.3±12.1% (p<0.0001), % time >250 mg/dl by 5.9±11.2% (p=0.0001), % time <70mg/dl by 1.5±3.4% (p<0.002), % time <54mg/dl by 0.6±2.3% (p=0.05). Mean HbA1c level decreased by 0.6±0.6% (p<0.0001) and mean sensor glucose by 20.3±37.5mg/dl (p= 0.0004). The INSPIRE questionnaire score was unchanged. Adverse events included 2 non-severe ketosis episodes and 2 local skin reactions. Conclusion AID initiation after a short period of pump therapy in T1D patients significantly improved glycemic control over 6 months, with no safety concerns but no measurable improvement in quality of life.