Background: We have previously demonstrated that tight glycemic control (80-120 mg/dL) improves outcome in critically injured patients. However, many centers have gotten away from aggressive glucose control due to the workload and risk of hypoglycemia. The objective of this pivotal trial is to evaluate the first in human continuous inline glucose monitor (OptiScanner) in critically ill patients.
Methods: A multicenter pivotal trial was conducted over a 1-year period (2014-2015) at four major academic centers in 200 critically ill patients. Three thousand seven hundred thirty-five glucose measurements were obtained and measured. A paired blood sample was then collected to coincide with the OptiScan measurement. The OptiScanner withdraws 0.13 mL of blood every 15 minutes from a central venous line, centrifuges the sample, and uses midinfrared spectroscopy to directly measure glucose levels in blood plasma. We plotted a Clarke Error Grid, calculated mean absolute relative difference (MARD) to analyze trend accuracy, and population coefficient of variation (PCV) to measure deviations. OptiScanner and Yellow Springs Instrument values were "blinded" from clinicians. Treatment was guided by the standard point of care meters.
Results: 95.4% of the data points were in zone A of the Clarke Error Grid and 4.5% in zone B. The MARD was 7.6%, the PCV 9.6%. The majority of data points achieved the benchmark for accuracy. The MARD was below 10%, which is the first inline continuous glucose monitor to achieve this result in a clinical trial. The PCV was less than 10%. We confirmed that the OptiScanner outperformed every 1- to 3-hour glucose measurements using point of care meters which prevents glucose excursions and variability and achieves a higher amount of time the patient's glucose values remain in range.
Conclusion: This pivotal multicenter trial demonstrates that the first inline CGM monitor is safe and accurate for use in critically ill surgical and trauma patients.
Level of evidence: Diagnostic study, level I.