Characteristics of Automated Insulin Suspension and Glucose Responses with the Predictive Low-Glucose Management System

Abstract

Background: The Predictive Low-Glucose Management (PLGM) system suspends basal insulin when hypoglycemia is predicted and reduces hypoglycemia. The aim of this analysis was to explore the characteristics of automated insulin suspension and sensor glucose (SG) responses following PLGM-initiated pump suspension.

Research design and methods: Children and adolescents with type 1 diabetes used the Medtronic MiniMed™ 640G pump as part of a randomized controlled trial. Data collected on a subgroup of participants on PLGM (suspend before low enabled) from CareLink^® Therapy Management Software were analyzed to explore the time and duration of PLGM-initiated pump suspension. Day and nighttime were defined as 06:00 am to 10:00 pm and 10:00 pm to 6:00 am, respectively.

Results: There were 20,183 suspend before low events in 8523 days (2.37 events/day). The mean suspend duration was 55.0 ± 32.7 min (day 50.0 ± 30.1, night 71.7 ± 35.1; P < 0.001). Although a 2-h pump suspension was more frequent at night (day 5%, night 18%), a patient-initiated resumption occurred more during day (day 34%, night 12%). SG values did not reach <3.5 and <3 mmol/L in 79% and 91% of the events, respectively. The 2-h SG following pump resumption was higher following autoresumption during the day (day vs. night 9.3 mmol/L vs. 8.4 mmol/L; P < 0.001).

Conclusions: Longer suspends and fewer glycemic excursions occur at night compared with day. The higher glycemic daytime excursions could be due to carbohydrate consumption to increase glucose levels and highlights the need for health care professionals to educate patients about carbohydrate intake around pump suspension.

Keywords: Predictive Low-Glucose Management; Sensor glucose.; Suspend before low; Type 1 diabetes.