Medtronic closed-loop system with bolus correction promising in first clinical trial
medwireNews: The latest iteration to the Medtronic artificial pancreas system, which includes automated correction boluses, shows promising results in its first home-based trial.
The system also works to a lower target glucose level than the commercially available system, based on feedback from testers of a previous enhanced version that incorporated improvements to reduce alarms and exits from closed-loop mode.
David O’Neal (University of Melbourne, Victoria, Australia) and colleagues tested the newest iteration in 12 adults with type 1 diabetes who were already using insulin pumps.
The trial took place during 1 week in a hotel under supervision, followed by 3 weeks free living at home. During these 3 weeks, participants spent a significantly greater proportion of their time within the blood glucose range of 70–180 mg/dL than they had done during a 1-week open-loop run-in phase, at 85.3% versus 75.0%. Average sensor glucose was also lower, at 123.0 versus 143.5 mg/dL.
However, this came at the expense of slightly but significantly more hypoglycemia, with the participants having an average 1.2 episodes of hypoglycemia per day while using the system in closed-loop mode versus 0.8 episodes per day during the run-in phase. The proportion of time spent with blood glucose below 70 mg/dL was 4.4% versus 3.0%, although there was no significant difference for time spent below 54 mg/dL (0.6 vs 0.4%).
The researchers say that these differences equate to an additional 20 minutes per day spent below 70 mg/dL and an additional 3 minutes spent below 54 mg/dL, relative to open-loop use.
“Further refinements of [closed-loop] systems are necessary to find a balance between mitigating hypoglycemia further, while maintaining excellent [time in range],” they write in Diabetes Technology & Therapeutics.
But they stress that the time in range is “among the highest of any [closed-loop] studies to our knowledge.”
To test the utility of the automated corrective insulin boluses, the supervised hotel stage included a missed meal bolus test and a late meal bolus (20 minutes after starting to eat) test. The missed bolus resulted in participants spending 46.9% of the next 4 hours with blood glucose above 180 mg/dL (but no time >250 mg/dL), compared with 8.3% and 0.0% when they delivered a premeal bolus in the open-loop run-in phase and the closed-loop home phase, respectively.
The late bolus resulted in no additional time in hyperglycemia, but 6.2% of the 4 hours spent with blood glucose below 70 mg/dL (but not <54 mg/dL) versus no time for the other conditions, and it caused significantly greater glycemic variability.
The researchers say that the hypoglycemia occurred because the automated correction boluses had already started to address the elevated blood glucose, turning a late bolus based purely on carbohydrate content into a slight overcorrection.
This “serves to emphasize the importance of premeal insulin bolus administration even when glucose control is provided by an enhanced [closed-loop] system,” they say, although they stress that despite the system being “challenged by missed and late meal boluses, the observed postmeal glucose excursions into low and high glucose ranges were modest.”
medwireNews is an independent medical news service provided by Springer Healthcare. © 2019 Springer Healthcare part of the Springer Nature group
- News story | Artificial pancreas use viewed positively by users and caregivers
- News story | Data support artificial pancreas for type 1 diabetes, but better evidence needed
- News story | Artificial pancreas benefits even well-controlled diabetes
- In-depth interview | The artificial pancreas: What’s on the horizon?