Simple meal announcements and pramlintide delivery versus carbohydrate counting in type 1 diabetes with automated fast-acting insulin aspart delivery: a randomised crossover trial in Montreal, Canada
Cohen E; Tsoukas MA; Legault L; Vallis M; Von Oettingen JE; Palisaitis E
The Lancet. Digital health
- PubMed
- 38906614
- Evidence level
- moderate
- Verification
- verified
Summary
Cohen et al. 2024 (Lancet Digital Health) randomised crossover trial (NCT04163874, n=32 enrolled/30 analysed) showing that an insulin-and-pramlintide closed-loop system with simple meal announcements is non-inferior to an insulin-and-placebo closed-loop with carbohydrate counting for time-in-range (difference −5%, 95% CI −9·0 to −0·7, p<0.0001 for non-inferiority) but does not improve the Emotional-Burden score; mild-to-moderate gastrointestinal symptoms were more common on the pramlintide arm; no serious adverse events.
Limitations
Small sample (n=30 analysed); short 14-day arms; single-centre (McGill); GI symptoms substantially more common on pramlintide (47% mild + 7% moderate) than on placebo (7% mild); several author industry ties disclosed (Eli Lilly, Novo Nordisk, Dexcom, Tandem, etc.).