Endocrine Abstracts

It was reported that outpatients with type 1 diabetes using real-time continuous glucose monitoring (rtCGM) reduce HbA1c more than those using self-monitoring of blood glucose (SMBG). However, it is not well known whether inpatients with type 2 diabetes using rtCGM reduce glycemic variability more than those using SMBG. This is a prospective parallel-group comparative study. Sixteen patients with type 2 diabetes hospitalized for diabetes treatment were randomly allocated to 2 groups. In group 1, patients used personal continuous glucose monitor (CGM) [GUARDIAN CONNECT], and drug treatment intervention (DTI) is performed based on the sensor glucose levels (SG) in the rtCGM (CGM group). In group 2, patients used professional CGM (iPro2) and DTI is performed based on the capillary blood glucose levels (BG) measured using “glucometers for SMBG, that were compliant with ISO15197:2013 diagnostic test systems” (“GISO”) [ACCU-CHEK Guide] (BGM group). In both groups, CGM was attached on the day of hospitalization (day 1) and used for 6 days. In group 1, all 288 SG were referenced for the intervention in real-time. In group 2, BG referenced for the intervention in real-time were measured at pre- and post- prandial time and bedtime (7 points) on days 2 and 5 and pre-prandial time and bedtime (4 points) on days 3 and 4. We preliminarily analyzed the 24-h SG measured using professional CGM (iPro2) for 150 patients with type 2 diabetes at our hospital to determine the target range for intervention using rtCGM to achieve both time-in-range (70–180 mg/dl) [TIR] > 70% and coefficient of variation [CV] < 36%. From this analysis, we determined the target range to achieve TIR>70% as “75–160 mg/dl” and that to achieve CV<36% as “within±30% of mean glucose levels”. In both groups, DTI was performed at evening from day 2 to day 5 and morning from day 3 to day 5 based on the DTI algorithm unified for each CGM group and BGM group, where that determened target range was used. Two patients in the BGM group were excluded from this study because they could not carry out the research protocol (CGM group: n=8, BGM group: n=6). Patients in the CGM group achieved both TIR>70% and CV<36% earlier than those in the BGM group (1.4 days vs 2.7 days; P=0.03: Log-rank test). During hospitalization, the DTI with reference to rtCGM may achieve both TIR>70% and CV<36% slightly earlier than that with reference to BGM.