medwireNews: The Dexcom G7 continuous glucose monitor (CGM) is accurate and well tolerated for up to 10.5 days when worn on the arm or abdomen, a study of adults with diabetes suggests.
“G7 is significantly smaller than G6 (by 60%) and introduces new features such as sensor/transmitter integration with a simplified insertion process”, write Satish Garg, from the University of Colorado Denver in Aurora, USA, and co-workers.
“The enhanced features of G7 may increase clinical adoption, encourage sustained use, and reduce the burden of diabetes management”, they suggest.
As reported in Diabetes Technology & Therapeutics, a total of 619 sensors were tested by 316 participants who had type 1 diabetes, or type 2 diabetes with or without intensive insulin therapy, and were given training on the wearable transmitter.
The participants wore the sensors concurrently on the arm and abdomen for 10 days plus a 12-hour grace period, attending three 12-hour clinic sessions on days 1 or 2, and again on days 4 or 7, with a final visit on the second half of day 10 or the first half of day 11.
CGM values during these sessions were compared against arterialized venous blood glucose concentrations using the YSI 2300 Stat Plus glucose analyzer, giving a total of 77,774 matched pairs of CGM–YSI values within the reportable range of 40–400 mg/dL.
The researchers report that the G7 CGM was “accurate and safe”, with mean absolute relative difference (MARDs) compared with blood glucose of 8.2% for sensors placed on the arm and 9.1% for those on the abdomen.
Accuracy was evaluated from the proportion of CGM values that were within either 15% of blood glucose levels above 100 mg/dL, or within 15 mg/dL of blood glucose levels of 100 mg/dL or less (%15/15). The agreement rates for the arm and abdomen were 89.6% and 85.5%, respectively.
The corresponding arm and abdomen agreement rates when instead using %20/20 evaluation were 95.3% and 93.2%, while the %30/30 rates were 98.8% and 98.1%, the researchers say.
Garg and co-workers note that “accuracy remained high” for both arm and abdomen sensors throughout the 10-day wear period, the 12-hour grace period, and across glucose ranges. Nor did accuracy significantly differ according to diabetes type or insulin regimen, or by gender or BMI category.
Further analysis indicated that the G7 sensor was most accurate when the CGM readings were increasing or decreasing by up to 1 mg/dL per min. The MARDs for arm sensors when CGM readings were increasing and decreasing by over 2 mg/dL per min were 9.3% and 9.7% with %20/20 agreements of 90.1% and 90.3%. The corresponding MARDs and %20/20 values for abdomen sensors were 10.4% and 10.3%, and 87.8% and 89.4%.
The investigators also assessed the true alert rates for detection of induced hypoglycemia and hyperglycemia using thresholds of 55 mg/dL and 300 mg/dL, respectively. The arm and abdomen sensors sent alerts when levels fell below 70 mg/dL in 91.3% and 85.2% of cases, respectively, and when levels were above 250 mg/dL in 99.9% and 99.8% of cases, respectively.
Garg et al report that four participants had mild-to-moderate adverse events during the study, namely moderate erythema, mild skin tearing, and mild or moderate discomfort during sensor removal. There were no serious adverse events.
Of the 241 participants who had prior experience with a CGM, 68.1% said the G7 was “easier to insert than their current personal device,” and of the participants who stated a site preference, 51.7% would choose to wear the device on their arm.
“As is the case with G6, calibration of G7 sensors remains optional”, observe the study authors.
“Unlike G6, G7 allows for temporary silencing of all audible alerts, including Urgent Low. Taken together, these attributes are anticipated to provide for a better end-user experience with G7 and help reduce diabetes burden.”
medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2022 Springer Healthcare Ltd, part of the Springer Nature Group
Diabetes Technol Ther 2022; doi:10.1089/dia.2022.0011