Hermanns, NorbertNorbertHermanns0000-0002-2903-2677Freckmann, GuidoGuidoFreckmannEhrmann, DominicDominicEhrmann0000-0002-5794-5596Wintergerst, PeterPeterWintergerstFaber-Heinemann, GabrieleGabrieleFaber-HeinemannHeinemann, LutzLutzHeinemann2019-09-192018-07-3020170012-1797https://fis.uni-bamberg.de/handle/uniba/44229It is well known that people with type 1 diabetes (pw1d) and hypoglycemia unawareness have an increased risk for severe hypoglycemia (SH); however, it is unknown if continuous glucose monitoring (CGM)helps to prospectively evaluate the risk of SH in such patients. We analyzed CGM recordings of 126 participants on MDI in the HypoDE study with hypoglycemia unawareness wearing blinded CGM (DexCom Gen 4) for 28 days (age 46.5±11.5 yrs. 36.5% female, HbA1c 7.5±1.0%, unawareness score 5.0±1.1). From these subjects, 65 (51.6%) reported occurrence of SH in the last year prior to study start. In a high-risk group of pw1d and hypoglycemia unawareness, CGM data on the exposure regarding hypoglycemic glucose values or number of hypoglycemic events ≤55 mg/dl predict the occurrence of SH. ROC analyses suggested meaningful cut-off values for the duration of hypoglycemic episodes, respectively the number of hypoglycemic events. These cut-off values can facilitate the identification of people with an elevated risk for SH even in a hypoglycemic high-risk group with reasonable sensitivity and specificity.engarticle