The Relationship Between Clinical Profiles, Glycemic Parameters, and Hypoglycemia in Pediatric Patients with Type 1 Diabetes

Background/Objectives: Our objective was to assess the role of clinical and continuous glucose monitoring (CGM) parameters in predicting the risk of hypoglycemia in pediatric patients with type 1 diabetes. Methods: Pediatric patients with type 1 diabetes (n = 71) at the Oradea County Clinical Emergency Hospital, Romania, who underwent CGM during their initial visit and were followed for at least 6 months with in-clinic visits every 3 months were enrolled in this study. Age, body mass index, time in range, the mean daily glucose (MDG) concentration, and the coefficient of variation (%CV) were considered as potential predictors of the risk of hypoglycemia, which was defined as the percentage of time spent below two glycemic thresholds of 3.9 and 3.0 mmol/L, corresponding to mild and clinically significant hypoglycemia, respectively. Results: Among a total of 142 glycemic profiles, the MDG concentration was significantly lower in those with hypoglycemia compared to those without, whereas %CV was significantly higher (p < 0.0001). Regression tree models identified %CV as the dominant variable for both thresholds, whereas classification tree models identified %CV as the dominant variable for clinically significant hypoglycemia and MDG for mild hypoglycemia. In profiles with a %CV of less than 36.15% and an MDG concentration greater than 7.16 mmol/L, the mean percentage of time spent below the 3.9 mmol/L threshold was 4.8%, which is close to that recommended by the American Diabetes Association guidelines. Patients younger than 7 years presented the highest frequency for both mild and clinically significant hypoglycemic episodes. Conclusions: Our study supports %CV and the MDG concentration as key factors in predicting hypoglycemia risk. Minimizing the risk of hypoglycemia in pediatric patients requires a %CV of less than 36%.