Rebound hyperglycemia in children with type 1 diabetes using automated insulin delivery

Background and aims: To investigate how rebound hyperglycemia (RH) affects CGM-assessed time-in-range and how different automated insulin delivery systems (AID) manage RH.

Methods: This cross-sectional study included children and adolescents with type 1 diabetes (T1D) using either Tandem Control IQ (CIQ) or Medtronic Minimed 780G (780G) at Steno Diabetes Center Copenhagen, Denmark. RH was defined as ≥1 sensor glucose value (SG) >10.0 mmol/l within two hours of a preceding SG <3.9 mmol/l, with RH-severity measured by area under the curve (AUC).

Results: Among 190 children and adolescents, 94 used CIQ and 96 used 780G. The median age (range) for CIQ and 780G was 11.9 (1.8-18.0) versus 11.3 (1.2-17.4) years (P=0.49), diabetes duration was 7.0 (1-16) versus 4.0 (0-13) years (P<0.001), and HbA1c was 53 (37-102) versus 53 (37-147) mmol/mol (P=0.29). RH-severity (AUC) was inversely related to time-in-range (TIR: 3.9-10.0 mmol/l) (P<0.001), regardless of the AID systems (P=0.57). The frequency of hypoglycemia, frequency of RH, and the percentage of hypoglycemia leading to RH were similar for 780G and CIQ (P>0.05). However, CIQ-users had a significant increased RH-duration of 30% (95%CI:16-46%) and RH-severity of 34% (95%CI: 18-53%) compared with 780G-users, despite having a shorter duration of hypoglycemia leading to RH of 23% (95%CI:14-32%).

Conclusions: Even though the frequency of RH was similar between AID systems, the severity and duration of RH were less and shorter for children and adolescents with T1D using 780G than CIQ. The inverse association between RH-severity and TIR suggests that addressing hypoglycemia management in AID-users could improve glycemic outcomes.