Fully closed-loop insulin delivery not yet a match for precise carb counting
medwireNews: A dual-hormone fully closed-loop system did not demonstrate non-inferiority to a hybrid system in a randomized trial, despite producing a high time in range (TIR), report the researchers.
The fully closed-loop system, which was based on faster-acting insulin aspart (Fiasp, Novo Nordisk) and pramlintide, achieved an average TIR (3.9–10.0 mmol/L, 70–180 mg/dL) of 74.3% during 24 hours of use, compared with 78.1% for the hybrid system.
The study is published in The Lancet Digital Health, in which commentator Charlotte Boughton (University of Cambridge, UK) describes the TIR achieved with the fully closed-loop system as “impressive.”
She says: “In a less supervised setting, one might expect carbohydrate counting to be less reliable and thus the performance of the comparator hybrid closed-loop system to be less than the 78.1% (66.3–87.5) observed in this trial, an issue which would not affect the Fiasp plus pramlintide closed-loop system.”
The closed-loop study periods took place at a research facility, where staff counted carbohydrate content to determine participants’ meal-time insulin needs during the hybrid closed-loop phase.
The systems used consisted of a Dexcom glucose sensor (G5 or G6) with a Medtronic pump (two when pramlintide was used; Minimed 630G or Veo) and the researchers’ control algorithm. The participants undertook a 27-hour period of fully and hybrid closed-loop control, in a randomly assigned order, both preceded by an at-home run-in for up to 9 days, using the system components in open-loop mode.
A total of 24 study participants completed both study periods. They were an average of 35 years old and had an average glycated hemoglobin of 8.1% (65 mmol/mol) at baseline.
Although the participants overall had a lower TIR with the fully closed-loop than the hybrid closed-loop system, Ahmad Haidar (McGill University, Montréal, Québec, Canada) and co-researchers point out that 42% spent more time in range with the fully closed-loop system, and the non-inferiority margin of a less than 6% difference was achieved for 58% of the participants.
The poorer performance of the fully closed-loop system was due to greater postprandial hyperglycemia, despite the algorithm having a meal-detection facility and the use of pramlintide to delay glucose absorption.
There was a relatively high rate of hypoglycemic episodes, with 33% and 58% of people experiencing at least one during the fully and hybrid closed-loop phases, respectively (no significant difference between the two).
The researchers note that the study participants had relatively poor glycemic control at baseline, which “is associated with imprecise carbohydrate counting, with underestimation being reported as more common than overestimation.” They believe this may have led to inflated insulin-to-carbohydrate ratios and therefore larger than necessary prandial insulin doses when carbohydrates were counted precisely by the research staff.
Twenty-nine percent of participants reported gastrointestinal side effects while using the dual-hormone system, compared with just 8% with the insulin-only system, and they were more severe.
“Considering that the time-in-range with the fully closed-loop system was high,” the researchers advocate longer outpatient studies of the system to compare performance in the real world and ascertain whether the gastrointestinal side effects relating to pramlintide “will be transient or chronic.”
In her commentary, Boughton says: “A key question is whether the quality-of-life benefits associated with not performing carbohydrate counting and manual bolusing outweigh the side-effects of pramlintide.”
But she stresses: “The potential beneficial effects of pramlintide on body weight and reduced insulin requirements are important given the increasing prevalence of obesity in people with type 1 diabetes.”
medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2021 Springer Healthcare Ltd, part of the Springer Nature Group