Menstrual cycle-related variations in insulin requirements and glycemic control during automated insulin delivery in adolescents with type 1 diabetes.

OBJECTIVES: Data on menstrual cycle-related metabolic variability in adolescents with type 1 diabetes (T1D) using automated insulin delivery (AID) systems are scarce. We aimed to quantify cycle-dependent differences in insulin requirements and glycemic outcomes in this population. METHODS: In this retrospective observational cohort study, adolescents with T1D aged 12.0-17.9 years who were at least 1 year postmenarche and had used AID for ≥6 months were included. Menstrual cycles were tracked using smartphone applications, and nine complete cycles per participant were analyzed. Insulin delivery and continuous glucose monitoring data were extracted from Glooko. Outcomes included total daily insulin dose (TDD), time in range (TIR), time above range (TAR), time below range (TBR), mean glucose, glucose coefficient of variation (CV), and recorded carbohydrate intake. Linear mixed-effects models were used. RESULTS: Fifty-one adolescents contributed 459 menstrual cycles. Mean age was 14.7 ± 1.8 years, and median HbA1c was 7.2 % (interquartile range 6.8-7.5 %). TDD was higher during the luteal than the follicular phase (1.05 vs. 0.95 U/kg/day; mean difference +0.10 U/kg/day, 95 % CI 0.05-0.15; p<0.001). The luteal phase was associated with reduced TIR (-6.3 %, p=0.001), increased TAR (+5.5 %, p=0.01), higher mean glucose (+12.5 mg/dL, p=0.03), and higher CV (39.4 vs. 35.0 %, p=0.03). TBR and carbohydrate intake did not differ significantly. CONCLUSIONS: In adolescents with T1D who menstruate and use AID systems, insulin requirements and glycemic control vary across the menstrual cycle, with increased insulin demand and hyperglycemia during the luteal phase. These findings highlight clinically relevant cycle-related metabolic variability despite the use of AID.