242-OR: The Impact of Glucose Threshold on the Number and Duration of Glucose Episodes

Introduction and Objectives: The number of sensor-detected hypoglycemia (SDH) episodes depends on the hypoglycemia threshold used. We investigated the relationship between glucose threshold and the number and duration of events in the Hypo-METRICS dataset. Methods: Individuals with insulin-treated diabetes wore a blinded continuous glucose monitor for 10 weeks. SDH was defined as 15 min below each threshold between 5 mmol/L (90 mg/dL) and 2.2 mmol/L (40 mg/dL) at 0.1 mmol/L increments. The number and average duration of episodes was calculated at each glucose threshold. Results: Data from 599 (T1D=276 and T2D=323) people were analyzed; median age 56 years, HbA1c 7.4%. Event rate was higher for events between 3-5 mmol/L than < 3mmol/L (10.5 vs 1.1 episodes/week, p < 0.01) and average duration longer (63.1 vs 35.5 minutes/episode, p < 0.01). This is shown in Figure 1a and b. Each 0.1 mmol/L increment in hypoglycemia threshold resulted in 0.5 more episodes/week between 3-5 mmol/L vs 0.05 additional episodes/week at a threshold < 3 mmol/l (p < 0.01). Each 0.1 mmol/l increment in hypoglycemia threshold increased average duration by 3.1 minutes/episode between 3-5 mmol/l vs 0.4 minutes/episodes for those < 3 mmol/l (p < 0.01). Conclusion: These results may be useful in understanding the alarm burden at different thresholds allowing personalization in proportion to individual risks.