A Three-Way Accuracy Comparison of the Dexcom G5, Abbott Freestyle Libre Pro, and Senseonics Eversense CGM Devices in an Outpatient Study of Subjects with Type 1 Diabetes

There are no published studies directly comparing the accuracy of continuous glucose monitoring (CGM) devices in the outpatient setting. We tested the performance of the Dexcom G5, Abbot Freestyle Libre Pro, and Senseonics Eversense (an implantable CGM approved in Europe) during a 6-week, free-living, outpatient bionic pancreas study involving 23 subjects with type 1 diabetes who wore all 3 devices concomitantly. The primary outcome was the mean absolute relative difference (MARD) vs. plasma glucose (PG) values measured with the Nova Biomedical StatStrip Xpress meter that was also used for calibrations according to the manufacturer's instruction (except for Libre Pro that is not calibrated). We compared PG values with CGM readings when they were available from all 3 CGMS in the 5 minutes preceding the PG values (n=829 sets). Since the Libre Pro records readings every 15 minutes, we also did a two-way comparison between the G5 and the Eversense that allowed a higher number of comparisons (n=2277 sets). Statistical significance was determined using a repeated measurements model fitted with the generalized estimating equation method. All 3 CGM systems produced higher average MARDs than during in-clinic studies. However, since all three CGM systems were worn by the same individuals and used the same meter for calibration and as comparator, we were able to directly compare their performance under real-world conditions. In the 3-way comparison Eversense achieved the lowest nominal MARD (14.8%) followed by Dexcom G5 (16.3%) and Libre Pro (18.0%) (Eversense vs. Libre Pro p=0.004, other comparisons p=NS). In the 2-way comparison the MARD difference between Eversense (15.1%) and G5 (16.9%) was statistically significant (p=0.008). We found that the point accuracy of the Eversense was significantly better than two other CGM systems. The Eversense CGM system may be useful to provide glucose values to artificial pancreas devices. Disclosure R.Z. Jafri: None. C.A. Balliro: None. F. El-Khatib: Stock/Shareholder; Self; Beta Bionics. Employee; Self; Beta Bionics. M. Maheno: None. M.A. Hillard: None. A.J. O'Donovan: None. R. Selagamsetty: None. H. Zheng: None. E. Damiano: Other Relationship; Self; Beta Bionics. S.J. Russell: Other Relationship; Self; Beta Bionics, Novo Nordisk Inc.. Advisory Panel; Self; Companion Medical, Tandem Diabetes Care, Inc., Unomedical a/s. Research Support; Self; Beta Bionics, Zealand Pharma A/S, MITRE Corporation.