Variation in the reporting of glucose values during simultaneous glucose sensor wear

Abstract

Background: We evaluated variations in glucose values recorded by simultaneously worn glucose sensors. Method: Children (n=44) randomised to the hybrid closed-loop arm of the CLOuD study (Closed-Loop from Onset of Diabetes) had multiple two-week periods of simultaneous sensor wear (sensors A and B or sensors C and D worn together) during unrestricted living. Glucose metrics were calculated from each sensor and paired sample T-tests were used to evaluate differences. Results: Sensor A recorded lower mean glucose by 0.4mmol/L (95% CI -0.8, -0.1) (mean±SD 7.6±0.5mmol/L vs 8.0±0.9mmol/L), 2.1 percentage points (95% CI -3.4, -0.9) less time <3.9mmol/L (median [IQR] 3.2% [2.3, 4.4] vs 4.3% [2.5, 7.4]), and 7.9 percentage points (95% CI 5.0,10.7) greater time 3.9–10.0mmol/L (78.7±5.1% vs 70.8±8.0%) than sensor B. Sensor C recorded higher mean glucose by 1.2mmol/L (95% CI 1.0, 1.5) (8.3±0.7mmol/L vs 7.1±2.7mmol/L), 9.9 percentage points (95% CI -11.8, -8.0) less time <3.9mmol/L (median [IQR] 2.2% [1.6, 3.4] vs 12.8% [8.0, 17.2]), and 3.9 percentage points (95% CI 2.2, 5.7) greater time 3.9–10.0mmol/L (73.1±6.9% vs 69.2±8.3%) than sensor D. Conclusions: There is significant variability in reporting of glucose values by different sensors worn during unrestricted living. Copyright © 2023 John Wiley & Sons.