Medtronic MiniMed 780G's PID algorithm improves postprandial glycemia for unannounced meals in T1D adolescents

Managing Type 1 Diabetes (T1D) presents significant challenges, especially concerning postprandial glucose excursions. Unannounced meals, where carbohydrate intake is not pre-bolused, are a major hurdle, often leading to hyperglycemia. Current standard-of-care often involves manual insulin bolusing, which can be imprecise and burdensome. Automated Insulin Delivery (AID) systems aim to mitigate this by autonomously adjusting insulin delivery, but their performance varies. This study addresses the critical gap in understanding how different AID algorithms handle the real-world scenario of unannounced meals, impacting glycemic control and patient quality of life.

This randomized controlled crossover pilot-study enrolled 20 adolescents aged 11 to 18 years with Type 1 Diabetes, who were using either the Medtronic MiniMed 780G (utilizing a proportional-integral-derivative, PID, algorithm) or the Tandem t:slim X2 (employing a model predictive control, MPC, algorithm) AID system. Participants consumed two types of meals: a carbohydrate-only meal and a mixed meal, both containing 30 g of carbohydrates, with the mixed meal including an additional 15 g of protein. Each meal type was tested in both announced (AM) and unannounced (UM) settings. Primary endpoints included 2-hour and 4-hour blood glucose difference (ΔBG), glucose peak, time to peak, and time spent above (TAR), below (TBS), in range (TIR), and in tight range (TITR).

For unannounced meals (UM), the PID algorithm demonstrated significantly better glycemic control compared to the MPC algorithm. Specifically, PID achieved a lower 2-hourΔBG than MPC, alongside a reduced glucose peak, lower TAR (time above range), and improved TIR (time in range) and TITR (time in tight range). Differences in 4-hourΔBG and time-at-peak were not statistically significant between the two algorithms. Both AID algorithms effectively mitigated postprandial glycemia and returned glucose levels to baseline within 4 hours without any reported safety concerns.

In mixed announced meals, the PID algorithm also showed lower glucose peaks and TAR, with higher TIR and TITR compared to MPC. However, when meals were properly announced (AM), the performance of both algorithms was comparable for carbohydrate-only meals. This suggests a distinct advantage for PID in scenarios where meal boluses are not pre-administered.