In the Journals Plus

Implantable glucose sensor safe, accurate in type 1, type 2 diabetes

Show Citation

February 14, 2018

An implantable, subcutaneous continuous glucose sensor for patients with type 1 and type 2 diabetes was shown to be safe and accurate over 90 days, with more than 90% of continuous glucose monitoring system readings within 20% of reference values and few mild to moderate adverse events, according to findings from the PRECISE II study.

“The clinical benefits of CGM technology are maximized in patients who regularly wear their CGM devices for at least 6 days per week; however, many patients struggle to achieve consistent adherence,” Mark P. Christiansen, MD, of Diablo Clinical Research in Walnut Creek, California, and colleagues wrote in the study background. “Moreover, many individuals who initiate CGM frequently discontinue its use.”

In a nonrandomized, prospective, masked, single-arm study, researchers evaluated the safety and accuracy of an implantable CGM system (Eversense, Senseonics) in 90 adults with type 1 and type 2 diabetes between January and July 2016 at eight sites in the United States (mean age, 45 years; 54 men; 77 white). Within the cohort, 61 participants had type 1 diabetes and 29 had type 2 diabetes; 47.8% were using insulin pump therapy. The CGM system consisted of an implantable, fluorescence-based, cylindrical glucose sensor (3.5 mm x 18.3 mm) designed to operate for 90 days and a “smart” transmitter worn over the sensor, which wirelessly powered it to initiate glucose measurement and transfer the data to a mobile medical app. Alerts for hyperglycemia and hypoglycemia were provided on the mobile device as well as with on-body vibratory alerts from the transmitter.

Patients attended seven visits, including an initial screening visit, sensor insertion visit, four accuracy assessment visits (days 1, 30, 60 and 90), and a follow-up visit after sensor removal. At accuracy assessment visits, researchers compared the CGM values with a bedside glucose analyzer, and qualified participants underwent hyperglycemia and hypoglycemia challenges on days 30, 60 and 90. Primary endpoint was the mean absolute relative difference for the paired sensor and the reference glucose measurements collected over 90 days across a glucose range between 40 mg/dL and 400 mg/dL. Additional endpoints included Clarke error grid analysis and

sensor longevity, and the primary safety endpoint was the incidence of device-related serious adverse events.

Among the cohort, 75 participants received a single sensor, and 15 received bilateral sensors; 82 participants completed the 90-day study with 16,653 matched glucose measurements.

Researchers found that the overall mean absolute relative difference against the reference glucose analyzer was 8.8% (95% CI, 8.1-9.3), and analysis showed that 93.3% of the CGM values were within 20 mg/dL, or 20%, of the reference values over the total glucose range.

Researchers observed similar results in the post hoc analysis of all 90 participants, with a mean absolute relative difference of 8.9% (95% CI, 8.3-9.4) and 93% of CGM values within 20% of the reference range. Additionally, the system correctly identified 93% of hypoglycemic events and 96% of hyperglycemic events identified by the reference glucose reader. Researchers observed 12 mild adverse events and two moderate adverse events, most related to bruising or pain/discomfort at the insertion site.

“Additional clinical studies will be required to evaluate the accuracy and usability of the Eversense CGM system among pediatrics, with reduced calibration frequency and for extended durations through 180 days,” the researchers wrote. – by Regina Schaffer

Disclosures : Christiansen reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.