Abstract
Continuous glucose monitoring (CGM) systems are increasingly used, but their accuracy under flight-conditions remains insufficiently studied. We assessed sensor accuracy of four CGMs during in-flight (mid-haul) and ground conditions.
Participants (n = 20) wore four sensors (Abbott Libre 3, Dexcom G7, Medtronic Simplera, and Sinocare iCan i3) in parallel, inserted 1 day prior to experiment. Capillary reference measurements were performed every 30 min. Accuracy evaluation included: proportion of CGM values within 10% of comparator glucose levels >100 mg/dL or within 10 mg/days of comparator levels ≤100 mg/dL (%10/10), along with %20/20 agreement rates, mean absolute relative difference (MARD), Consensus error grid.
Twenty participants with type 1 diabetes (age: 37.9 ± 12.8 years, HbA1c: 52.3 ± 4.9 mmol/mol, diabetes duration 19.25 ± 10.6 years) were included in the analysis. Data from 297 sensors with 3473 matched pairs (G7/Libre3/Simplera/i3) were analysed. The overall MARDs were: 9.5% ± 3.3% (G7), 9.6% ± 3.8% (Libre3), 19.9% ± 6.2% (Simplera) and 15.7% ± 6.6% (i3). Overall %20/20 and %10/10 agreement rates were 90.6%/70.2% (G7), 90.8%/66.8% (Libre3), 63.6%/37.8% (Simplera) and 72.5%/42.2% (i3) respectively. The proportion of values in Zones A and B of the Consensus error grid was 98.9% (G7), 99.0% (Libre 3), 94.5% (Simplera) and 97.5% (i3). Prandial state and flight phase had limited influence on the relative performance ranking of the devices.
G7 and Libre3 showed consistently higher accuracy and clinical reliability across various phases of flight and ground conditions. Simplera and i3 exhibited greater variability in accuracy. These findings suggest that while modern CGMs could support diabetes management in aviation, individual system performance may differ under specific operational conditions.