Abstract
Aim: This trial assessed the outcomes on glucose control of the implantable Eversense real-time continuous glucose monitoring (CGM) system compared to self-monitoring of blood glucose or intermittently scanned CGM in patients with type 1 or type 2 diabetes (T1D or T2D).Patients and methods: This was a randomized (2:1), prospective, national, multicenter study. All participants, aged >18, on multiple daily insulin injections or insulin pump had a sensor inserted, activated only in the enabled group. Included patients had T1D or T2D with an HbA1c level >8% (cohort 1) or T1D with a time below range <70 mg/dL (TBR<70 ) >1.5 hour/day during the previous 28 days (cohort 2). The primary outcomes were the HbA1c change at D180 (cohort 1) or the change in TBR<54 during the period D90-D120 (cohort 2). A covariance model (ANCOVA) was used for endpoint analyses.Results: Overall, 149 patients were included in cohort 1 and 90 in cohort 2. In cohort 1, the adjusted mean [95%CI] difference (enabled-control) in HbA1c at D180 was -0.1% [-0.4; 0.1], p=0.341. No significant difference in time in range (TIR70-180 ) or time above range (TAR>180 ) was observed. In cohort 2, the mean adjusted difference in TBR<54 was -1.6% [-3.1; -0.1], p=0.039 during D90-D120 and remained at -2.6%, [-4.5; -0.6], p=0.011 during D150-D180 (pre-specified secondary outcome). There was no safety issue.Conclusion: This study shows that the Eversense CGM system can significantly decrease TBR<54 in patients with T1D prone to hypoglycemia.