SAT-638 An Academic Center Experience with the Eversense Continuous Glucose Monitoring System and Assessment of Intrapatient Variability with Repeated Same-Pocket Insertion

BACKGROUND: The Eversense continuous glucose monitoring (ECGM) system is a 90-day implantable device approved for patients with diabetes. The manufacturer recommends alternating arms each subsequent insertion to ensure adequate healing. To date, the intra-patient sensor variability using the same subcutaneous pocket has not been evaluated. We aim to assess change in glucometrics upon continuous use and the use of the same insertion pocket. Methods: Retrospective study (October 2018-October 2019), that included all patients using the ECGM. Demographics, and glucometrics data: transmitter wear time (TWT), average glucose (AG [mg/dL]), coefficient of variation (CV), time in range (TIR), time below range (TBR) and time in serious hypoglycemia (SH) data were obtained. Paired T-test was used to compare means (±SD) of the initial 90 days of sensor data (P1) with data of subsequent device insertions, grouped by periods of data time: period 2, 3 and 4 (P2, P3, P4). Analysis was done as a group, by insertion period, and by cohorts: cohort 1: same subcutaneous pocket, cohort 2: alternating insertion site. Results: Nine female, 7 male patients, age 37±11 years old, weight 85±10 Kg. Fourteen with T1DM, one with T2DM and one from pancreatectomy. There were 7 patients in each cohort, 2 patients had only one device inserted resulting in only 1 period of 3-month data; 7 patients had 2 consecutive devices inserted i.e. 2 consecutive 3-month periods of data (P1 and P2), 4 had 3 consecutive devices inserted, three 3-month periods of data (P1, P2 and P3) and 3 had 4 consecutive devices inserted, i.e. four 3-month consecutive periods of data (P1, P2, P3 and P4). Baseline, first insertion, initial 3-month period of data (P1) showed TWT 69.7±23 for the entire group, AG 175±30 mg/dL, CV 36.7±7.6, TIR 52.6±17, TBR 2.69±2.5, SH 1.5±2.3. We found no statistical significance for TWT, AG, CV, and TIR, as a whole and when analyzed by cohorts and period of data: P1 vs P2, P1 vs P3, P1 vs P4. TBR and SH decreased with continuous ECGM use. TBR mean difference comparing P1 vs P2, P3 and P4 was -0.15±0.9 (p=0.9), 1.16±1.8 (p=0.02), and 1.8±1.6 (p=0.001) and for SH -0.05±0.8 (p=0.8), 0.6±1.3 (p=0.09), and 0.9 ±1 (p=0.004). There were 3 device failures; 2 post first insertion, one post repeat insertion in the same pocket. No infections, or serious bleeding were observed. Conclusions: Continuous ECGM use decreased time below range and time in serious hypoglycemia, without change in TWT, AG, TIR and CV. There was no significant intrapatient variability comparing consecutive device insertions regardless of whether patients elected to use the same pocket or alternate arms, suggesting that using the same insertion pocket may have the same efficacy as placement into a fresh insertion site, while decreasing the number of punctures, pain, procedure time, risk of bleeding and infection.