P-440: Time in range and insulin dose: searching for additional factors associated with hypoglycemia episodes in children and youth living with type 1 diabetes
C Ibeas, S Aedo, M Novoa, E Bahamonde, F GiraudoAbstract
Introduction
Hypoglycemia is a common complication of type 1 diabetes (T1D). Continuous glucose monitoring (CGM) provides metrics such as time in range (TIR) and coefficient of variation (CV), which, along with total daily insulin dose (TDD) may be associated with hypoglycemic episodes.
Objectives
To evaluate the association between TIR, CV, TDD and TBD with the occurrence of hypoglycemic events in youth living with T1D who use CGM.
Methods
Cross-sectional analytical study. Patients with T1D (aged 1.8–19.9 years, mean 11.7 ± 4.6), who used CGM and who had recorded episodes of hypoglycemia (CGM values <70 mg/dL for 15 minutes or longer) within 14 days of their last visit were included. Variables included were age, TIR, CV, TDD, and treatment type (multiple daily insulin doses [MDI] or an automated insulin delivery system [AID]). The dependent variable was the presence of a hypoglycemia event (yes/no). Binary logistic regression with backward selection was performed using the likelihood ratio (p < 0.05).
Results
A total of 91 patients were included in the final multivariable model. CV was significantly associated with a higher probability of hypoglycemia (OR 1.16; 95% CI 1.05–1.28; p=0.005). The TIR was also associated with hypoglycemia (OR 1.04 for each 1% increase; 95% CI 1.00–1.07; p=0.041). Likewise, treatment with MDI was associated with a higher odd of hypoglycemia compared with AID (OR 6.49; 95% CI 1.62–26.0; p=0.008). Age and TDD were not independently associated with hypoglycemic events.
Conclusions
Treatment with AID was associated with a lower risk of hypoglycemia than MDI in young people with T1D who use CGM. CV was independently associated with hypoglycemia, with a 16% increase in risk for each percentage point. The TIR showed a paradoxical association, possibly due to higher TDD or interaction with CV. These metrics could help to identify high-risk patients, given the multifactorial nature of hypoglycemia and the need for capillary confirmation.