P-146: Long-term glycemic trajectories and behavioral determinants during three years of advanced hybrid closed-loop therapy in youth with type 1 diabetes
S Passanisi, B Bombaci, G Salzano, M Bassi, N Minuto, G Tornese, E Catamo, R Bonfanti, A Rigamonti, P Reinstadler, C Mameli, C Maffeis, C Piona, M Delvecchio, C Arnaldi, B Felappi, E Mozzillo, R Roppolo, G Maltoni, D Lo Presti, G Pezzino, A Zanfardino, F Lombardo, , ISPED Diabetes Group Study CollaboratorsAbstract
Introduction
Data on the persistence of glycemic benefits beyond the first year of automated insulin delivery are scarce, especially in large pediatric multicenter settings.
Objectives
This study assessed long-term effectiveness and determinants of sustained optimal control over three years of MiniMed™ 780G use in young individuals with type 1 diabetes.
Methods
In this real-world, longitudinal multicenter study, 359 participants (median age 12.2 years; 50.9% female) from 20 Italian pediatric centers were followed for 3 years after MiniMed™ 780G initiation. Glucose outcomes, insulin delivery patterns, and system settings were evaluated at 1, 2, and 3 years. Trends over time were analyzed using linear mixed-effects models, and predictors of achieving time in tight range (TITR ≥50%) at 3 years were identified through multivariable logistic regression.
Results
Median time in range (TIR) remained within recommended targets, with a slight decline from 75% at year 1 to 74% at year 3 (p=0.002). Automated correction boluses progressively increased, while user-initiated boluses and carbohydrate entries decreased (p<0.001). Participants with higher 1-year TIR showed a gradual decline over time alongside a greater rise in automated corrections (interaction p<0.001) (Figure 1). At 3 years, greater SmartGuard use (OR 1.12, p=0.008), optimized system settings (OR 2.99, p=0.013), and fewer automated corrections (OR 0.85, p<0.001) independently predicted TITR ≥50%.
Conclusions
MiniMed™ 780G maintains effective glycemic control over three years. However, increasing dependence on automation and reduced user engagement may limit optimal outcomes, underscoring the importance of continuous education and device optimization.