Accuracy of the intermittently scanned continuous glucose monitoring system in critically ill patients: a prospective, multicenter, observational study

Objective Continuous glucose monitoring (CGM) has the potential to improve glucose control in the intensive care unit (ICU) setting. We sought to evaluate the accuracy of the intermittently scanned CGM (isCGM) system in critically ill patients. Research design and methods Adult patients were consecutively enrolled from three ICUs from August 2020 to January 2021. The performance of FreeStyle Libre Pro was evaluated against the venous blood glucose samples as a reference. Numerical accuracy was examined by the mean absolute relative difference (MARD), the Bland–Altman analysis, and the International Organization for Standardization criteria. Clinical accuracy was assessed by performing the Clarke and consensus error grid analysis. Results A total of 122 patients were included and 3416 matched glucose pairs were analyzed. The overall MARD was 18.0%, and the highest MARD (33.1%) was observed in the hypoglycemic range (<70 mg/dL). The Bland–Altman analysis revealed a mean bias of −11.7 mg/dL, with the 95% limits of agreement of −73.0 to 49.5 mg/dL. The percentages of isCGM glucose values within ±15%/15, ±20%/20, and ±30%/30 mg/dL were 49.8%, 64.7%, and 84.5%, respectively. The Clarke and consensus error grid analysis showed acceptable clinical accuracy with 98.5% and 98.8% of glucose values falling into zones A and B. Conclusions Our study demonstrated suboptimal overall accuracy of isCGM for critically ill patients. Whether the adjunctive use of isCGM could improve glucose management and health outcomes in the critically ill needs further investigation. Clinical trial registration ChiCTR2100042036, Chinese Clinical Trial Registry.