Advanced Closed-Loop Control System Improves Postprandial Glycemic Control Compared With a Hybrid Closed-Loop System Following Unannounced Meal

Abstract

OBJECTIVE Meals are a major hurdle to glycemic control in type 1 diabetes (T1D). Our objective was to test a fully automated closed-loop control (CLC) system in the absence of announcement of carbohydrate ingestion among adolescents with T1D, who are known to commonly omit meal announcement. RESEARCH DESIGN AND METHODS Eighteen adolescents with T1D (age 15.6 ± 1.7 years; HbA1c 7.4 ± 1.5%; 9 females/9 males) participated in a randomized crossover clinical trial comparing our legacy hybrid CLC system (Unified Safety System Virginia [USS]-Virginia) with a novel fully automated CLC system (RocketAP) during two 46-h supervised admissions (each with one announced and one unannounced dinner), following 2 weeks of data collection. Primary outcome was the percentage time-in-range 70–180 mg/dL (TIR) following the unannounced meal, with secondary outcomes related to additional continuous glucose monitoring-based metrics. RESULTS Both TIR and time-in-tight-range 70–140 mg/dL (TTR) were significantly higher using RocketAP than using USS-Virginia during the 6 h following the unannounced meal (83% [interquartile range 64–93] vs. 53% [40–71]; P = 0.004 and 49% [41–59] vs. 27% [22–36]; P = 0.002, respectively), primarily driven by reduced time-above-range (TAR >180 mg/dL: 17% [1.3–34] vs. 47% [28–60]), with no increase in time-below-range (TBR <70 mg/dL: 0% median for both). RocketAP also improved control following the announced meal (mean difference TBR: −0.7%, TIR: +7%, TTR: +6%), overall (TIR: +5%, TAR: −5%, TTR: +8%), and overnight (TIR: +7%, TTR: +19%, TAR: −5%). RocketAP delivered less insulin overall (78 ± 23 units vs. 85 ± 20 units, P = 0.01). CONCLUSIONS A new fully automated CLC system with automatic prandial dosing was proven to be safe and feasible and outperformed our legacy USS-Virginia in an adolescent population with and without meal announcement.

Performance Effect of Adjusting Insulin Sensitivity for Model-Based Automated Insulin Delivery Systems

“…The IS-informed BC would then modify the amount obtained through 7 by a factor of S I r a t i o , in a similar manner as the method used in Fabris et al 1 Note that this would imply a proportional modification of both CR and CF, instead of solely modifying the correction element in the meal bolus as proposed for bolus calculators in Schmidt and Norgaard. 36…”

A Deep Learning Framework for Automatic Meal Detection and Estimation in Artificial Pancreas Systems

“…Zheng et al [23] employ the minimal model and run multiple simulations when the difference between the model trajectory and CGM trajectory is larger than a set value, to explain the likely meal size to reduce this divergence. Finally, Garcia-Torado et al [24] rely on a classification model (logistic regression) that continuously estimates the probability of an unannounced meal and provides a scaled insulin bolus based on the total daily insulin instead of estimating the carbohydrate content.…”

Diabetes Technology Meeting 2021

“…Four days were repeated for each individual, with an unannounced meal on the second day. Garcia-Tirado et al 98 present results comparing the new controller (“RocketAP”) with the previous controller (“USS-Virginia”). Particularly during the six hours following the unannounced meal, the RocketAP outperformed the USS-Virginia, with a reduced time-above-range.…”