Prospective Study on Self-Calibrating Continuous Glucose Monitoring Practicality and Accuracy in Noncritically Ill...

Aims: This study is aimed at evaluating the accuracy and feasibility of real-time continuous glucose monitoring (rt-CGM) in non-ICU hospitalized adult COVID-19 patients who had hyperglycemia requiring insulin therapy during admission. Materials and Methods: Medtronic Guardian Sensor 3 and transmitter were placed on the patient's abdomen. The patient performed a self-calibration of CGM via the application installed in the smartphone. Paired CBG and sensor glucose values were analyzed for accuracy of CGM using mean absolute relative difference (MARD) and Clarke error grid analysis (CEGA). Results: Fifteen patients were enrolled. Mean age was 48.6 ± 17.9 years; 53.3% were female. Thirteen patients (86.7%) had pre-existing diabetes. Mean HbA1c was 10.6 ± 3.6%. Mean duration of CGM use was 6 ± 1.2 days, and mean calibration was 2.6 ± 0.7 times/day. There were 253 paired CBG and CGM measurements. The overall MARD was 9.9 ± 9.3%. The lowest MARD was observed in the CBG range of 70-180 mg/dL (9.6 ± 9.0%) and on Day 3 of sensor wear (8.0 ± 4.6%). The percentages of glucose readings within 15/15%, 20/20%, and 30/30% were 81.0%, 89.7%, and 95.3%, respectively. A total of 99.2% of the data points were in Zones A and B of CEGA, and none were in Zone E. CGM reduced POC testing by approximately 30%. Conclusions: Rt-CGM use in hospitalized patients with COVID-19 infection demonstrates high accuracy, reduces the frequency of CBG testing, and preserves medical resources. The patient's self-calibration of rt-CGM in this setting is feasible. Although the COVID-19 pandemic tends to improve, our research could be applied to new emerging infectious diseases. Trial Registration: Thai Clinical Trials Registry: TCTR20230426007.