Presenter of 1 Presentation
SENSOR-DETECTED HYPOGLYCAEMIA IN SHORT-DURATION CGM DATA FROM MULTIPLE CLINICAL TRIALS IN DIABETES. PRELIMINARY RESULTS FROM THE HYPO-RESOLVE DATABASE.
Abstract
Background and Aims
Compared to time below range less is known about the frequency and temporal distribution of hypoglycaemic episodes in CGM data. Pooled analysis of multiple clinical trials from the Hypo-RESOLVE (Hypoglycaemia - REdefining SOLutions for better liVEs) database allows for large-scale exploration of sensor-detected hypoglycaemia (SDH).
Methods
We identified 15 studies (7 in T1D) with short-duration (<40 days) CGM (2165 participants, 45% with T1D). We explored frequency of SDH lasting ≥15 minutes on different levels (<70, <54, <40 mg/dl) by time of day and sensor runtime and glycaemic control by clock time.
Results
While nighttime accounts for 25% of hours in a day (0-6 am), we found 32% of SDH<40, 30% of SDH<54 and 28% of SDH<70 there (figure 1). This pattern was more pronounced in T2D (42, 40, 37%, respectively) than in T1D (30, 27, 24%, respectively). The number of daily SDH episodes <70 mg/dl and <54 mg/dl dropped after the first days and stabilized over sensor runtime while readings <40 mg/dl had less variability (figure 2). The drop was more pronounced in T1D while in T2D the curve over time was flatter for all SHD levels. Analysis by clock time showed tighter glycaemic control for T2D and a surplus of very low readings <40mg/dl for T1D (figure 3).
Conclusions
Further research will aim at explaining the detected patterns in SDH. This will provide people using CGM systems (including people living with diabetes, healthcare professionals as well as researchers and health authorities) with a benchmark of expectable patterns of hypoglycaemia in CGM data.