Continuous intra-arterial blood glucose monitoring using quenched fluorescence sensing in intensive care patients after cardiac surgery: Phase II of a product development study

Purpose: Variations in blood glucose (BG), hyperglycaemia and hypoglycaemia are associated with adverse clinical outcomes in intensive care unit patients. Continuous glucose monitoring (CGM) offers the potential to improve BG control, leading to improved patient outcomes. In our product development study, we determined the safety and performance of the GluCath Intravascular CGM System for up to 48 hours in 20 patients admitted to the ICU after cardiac surgery. Methods: The GluCath system uses a quenched chemical fluorescence mechanism to optically measure glucose in blood. After undergoing elective cardiac surgery, 20 patients had a GluCath sensor inserted through a preexisting radial artery (RA) catheter, and BG was monitored for up to 48 hours. Qualitative measures included effects on patient care, blood pressure monitoring, and ease of blood sampling through the arterial catheter. Safety assessment of the sensor involved ultrasound (US) monitoring for intra-arterial thrombi. Quantitative measures were the accuracy of the sensor in comparison with the reference analyser, and the proportion of paired BG measurements that were compliant with the ISO15197:2003 and CLSI POCT 12-A3 accuracy reference standards. BG was managed according to usual protocols. Results: Twenty sensors were successfully deployed through pre-existing RA catheters and stayed in the RA of the 20 patients for between 6 and 48 hours, with a median time of 45.0 hours (interquartile range, 42.0-47.3 hours). Sixteen of the inserted sensors (80%) remained in situ for more than 40 hours. Three catheters were removed due to clinically significant sampling difficulty or waveform dampening. Two patients had US evidence of a thrombus; in neither patient was the sensor removed early, and there were no significant sequelae or adverse effects detected. For the 758 paired measurements available for performance analysis, reference BG values ranged between 5.3mmol/L and 12.8mmol/L. Of the 758 paired sensor measurements, 735 (97.0%) met the ISO15197:2003 criteria (within 20% of a reference measurement when BG is >=4.2mmol/L [75mg/dL]), and 648 (85.5%) met the CLSI POCT 12-A3 criteria (within 12.5% of a reference measurement when the BG level is >=5.6mmol/L [100mg/dL] and within 0.67mmol/L [12mg/dL] when the BG level is < 5.6mmol/L [100 mg/dL]). The aggregate mean absolute relative difference (MARD) between the sensor and the reference BG was 6.4%, with individual sensor MARDs ranging from 3.6% to 12.4%.