Safety of Overestimation of a Meal Insulin Bolus in the Context of Closed-Loop Operation Combined with a Simplified Semi-Quantitative Carbohydrate Intake Estimation

Abstract

A closed-loop system (CLS) combined with a simplified meal bolus has been shown to achieve comparable glucose control as CLS with carbohydrate (CHO) counting for patients with type 1 diabetes. It is, however, important to establish the safety of this meal bolus approach. We compared in a randomized inpatient trial: 1) Dual-hormone CLS with an adequately estimated CHO intake, 2) Dual-hormone CLS with an overestimated CHO intake. The simplified strategy tested involved a meal bolus based on a semi-quantitative meal CHO-content size assessment (regular, large or very large). A large size standardized meal of 75 g of CHO was provided to participants and was either adequately categorized (large meal 60 to 90 g of CHO; bolus for 65 g) or overestimated (very large meal >90 g of CHO; bolus for 95 g). The primary outcome was time below 4.0 mmol/L over a 4-hour postprandial period. We aim to recruit 10 participants who will be tested for the 75 g CHO meal and 10 participants for a 45 g CHO meal. Six participants completed the study for the 75 g meal so far (40.4±17.2 years old; HbA1c 7.4±0.5%). Similar mean sensor glucose level was achieved with the adequate meal category and the overestimation (median [IQR]: 10.8 [8.4 to 12.5] vs. 10.2 [8.5 to 10.5] mmol/L; p=0.46). No participant spent any time <4.0 mmol/L in both interventions and no difference was observed in the amount of glucagon infused (p=0.91). These preliminary analyses show that the dual-hormone CLS with a simplified meal bolus calculation is able to avoid hypoglycaemia in the event of overestimation of the meal insulin bolus.