Abstract
Background and Aims
Background: The long-term implantable Eversense CGM System was first commericalized in 2016 in Europe and South Africa for adults with diabetes. Sensor performance over sequential 90- or 180-day cycles has not been published.
Methods
Methods: Sensor glucose (SG) and SMBG measurements obtained from the Eversense data managment system from 6/2016 to 8/2019 were used to evaluate accuracy in Eversense patients with at least four sensor cycles from European and South African clinics. Twenty-four hour transmitter wear time, mean SG with variability metrics, glucose management indicator (GMI), and percent and time in various glycemic ranges were calculated over each cycle.
Results
Results: Among the 945 users included in the analysis (Table), the mean absolute relative difference (MARD) using 152,206 to 206,024 calibration-matched pairs against SMBG ranged from 11.5 to 11.9% over the four sensor cycles. Mean values of the CGM metrics ranged from 156.5 mg/dL to 158.2 mg/dL for SG, 54.7 mg/dL to 55.8 mg/dL for SD, 0.35 to 0.36 for CV, and 7.04% to 7.08% for GMI. The range of %SG across the 4 cycles in the various glycemic ranges was as follows: 1.1-1.3% (<54 mg/dL), 4.6-5.0% (<70 mg/dL), 63.2-64.5% ( ≥70 to 180 mg/dL), 22.4-23.2% (>180-250 mg/dL) and 8.1-8.8% (>250 mg/dL). Median transmitter wear time ranged from 83.2-85.8%.
Conclusions
Conclusion: This real-world longitudinal evaluation of the implantable Eversense CGM System demonstrated that the accuracy and glucometrics are stable over multiple, consecutive sensor cycles with no degradation of performance.
