0585 Use of a Hybrid Closed Loop Insulin Delivery System Improves Sleep and Glycemic Control in Adults with Long-Standing Type 1 Diabetes and Hypoglycemia Unawareness

Abstract Introduction Insulin delivery and continuous glucose monitoring systems (CGMs) have been reported to disrupt sleep in individuals with type 1 diabetes (T1D), potentially thwarting the adoption and continued use of diabetes therapeutic technologies. This study assessed changes in actigraphic sleep and glycemic outcomes in individuals at high risk for life threatening nocturnal hypoglycemia after initiating a hybrid closed loop (HCL) insulin delivery system with integrated CGM. Methods 10 adults (median age=51y) with long-standing T1D (median duration=34y) and hypoglycemia unawareness participated in an 18-month ongoing clinical trial assessing the effectiveness of a HCL system. Wrist actigraphs and CGMs measured sleep and glycemic control, respectively, at baseline (1 week) and at months 3, 6, 9, 12, 15, and 18 (3 weeks) following HCL initiation. Body mass index and hemoglobin A1c (HbA1c) were also collected at these timepoints. Hypoglycemia awareness was assessed using the Clarke hypoglycemia questionnaire, HYPO score, and glycemic lability index. Paired sample t-tests and Cohen’s d effect sizes modeled changes in sleep, glycemic control, and hypoglycemic awareness and the magnitude of these changes from baseline to 18 months. Results Sleep improved from baseline to 18 months [shorter sleep latency (p<0.01), later sleep offset (p<0.05), and less wake after sleep onset (WASO) (<0.01)]. Medium effect sizes were found for later sleep onset (d=0.74) and later sleep midpoints (d=0.77). HCL also improved hypoglycemia awareness from baseline to 18 months [Clarke score (p<0.01), HYPO score (p<0.01), lability index (p<0.05)]. Medium to large effect sizes were found for reduced nocturnal hypoglycemia (percent time glucose was <54mg/dL,<60mg/dL,<70mg/dL; d=0.66 – 0.81), daytime and nocturnal hypoglycemia (percent time glucose was<54mg/dL,<60mg/dL,<70mg/dL; d=0.61 – 0.69), and glucose variability (coefficient of variability; d=0.62). Conclusion HCL insulin delivery with CGM improved sleep over time as indicated by shorter sleep onset latency, later sleep offset, and less WASO. HCL insulin delivery also improved hypoglycemia awareness and led to clinically significant reductions in hypoglycemia and glucose variability. Support (If Any) NIH (R01DK117488 (NG), R01DK091331 (MRR), K99NR017416 (SKM), and UL1TR001878 (University of Pennsylvania Center for Human Phenomic Science)). National Aeronautics and Space Administration (NASA) (NNX14AN49G and 80NSSC20K0243 (NG)) and from the Pennsylvania Department of Health (SAP 4100079750 (IL)).