medwireNews: The DREAM5 randomized trial has confirmed the efficacy of closed-loop insulin delivery with the MD-logic algorithm in adolescents and young adults without remote monitoring.
The algorithm is one that uses fuzzy logic to control glucose with a basal–bolus insulin approach, so as well as controlling the basal insulin infusion rate it can give one-off larger amounts in response to a predicted high glucose level. Users still have to give pre-meal insulin boluses based on carbohydrate content, however.
Previous trials established the system in controlled settings, for overnight use at home, and during unannounced exercise. In the current study, 34 adolescents and young adults with type 1 diabetes used the closed-loop system or a sensor-augmented pump for two 60-hour periods over two consecutive weekends spent at home, without being remotely monitored.
“We specifically chose this time period, since the weekend poses additional challenges to diabetes management such as large meals, social events with possible alcohol consumption, outdoor activities and more,” write Revital Nimri (Schneider Children’s Medical Center of Israel, Petah Tikvah) and study co-authors in Diabetes, Obesity and Metabolism.
They note that although other closed-loop systems have already been successfully tested in unsupervised adolescents, these systems could only alter basal insulin and could not administer boluses.
During the closed-loop period, the participants spent 66.6% of their time within the glucose range of 70–180 mg/dL, which was significantly greater than the 59.6 % achieved during the sensor-augmented pump period.
This was largely due to a significant reduction in the time spent above 180 mg/dL, at a corresponding 28.3% versus 36.4%. There were no significant differences for higher glucose levels or for time spent in hypoglycemia, but there was a significant reduction in glycemic variability.
The team notes that the effect of closed-loop delivery on time in range was even more pronounced in the per protocol population, at 73.8% versus 58.5% during sensor-augmented pump delivery. The per protocol group comprised 23 patients, with the main protocol issue being interruption of closed-loop delivery because of miscommunication between the algorithm, which was run on a Windows tablet, and the pump.
“By integrating the MD-Logic algorithm in the pump itself this could be alleviated,” say the researchers. “Therefore, the per-protocol results maybe more indicative for the closed-loop performance in this study.”
There were no serious adverse events, and the common complaints of headache, abdominal pain, dizziness, and feeling ill “were equally distributed between both groups.”
medwireNews is an independent medical news service provided by Springer Healthcare. © 2018 Springer Healthcare part of the Springer Nature group