Point-of-Care Glucose Monitoring in a Critical Care Setting

Maintenance of appropriate glycemic control (AGC) is known to reduce morbidity and mortality in critical-care patients [N Engl J Med. 2001;345(19):1359-1367; Mayo Clin Proc. 2004;79(8):992-1000; N Engl J Med. 2006;354(5):449-461]. Since point-of-care (POC) blood glucose test systems provide rapid results at the patient's bedside, they offer potential value in the maintenance of AGC protocols, but it is important to demonstrate that results are comparable to reference methodology and that they are not adversely affected by clinical scenarios, such as maltose therapy, hypernatremia and hyponatremia, and abnormal hematocrit. The aim of this evaluation was to determine whether the new Accu-Chek Inform II test strip for use with the Inform II POC blood glucose test system (Roche Diagnostics), Nova StatStrip (Nova Biomedical), and Medisense PXP (Abbott Diagnostics) demonstrated the required accuracy and precision to allow implementation of a AGC protocol in critically ill patients compared with the reference blood gas methodology currently in routine use within our critical care and intensive care areas (cobas b221; Roche Diagnostics). In addition, this study sought to determine whether these POC blood glucose test systems are suitable for use in patients receiving treatments that contain, or are metabolized to, maltose, and to determine the clinical implications of bias in patients with hypernatremia/hyponatremia and in patients with hematocrit values outside the manufacturer-specified limits. All POC blood glucose test systems within this evaluation demonstrated statistically significant correlation with reference methodology across the working range for critically ill patients. The performances of all POC systems tested are not adversely affected by the presence of maltose containing solutions. All devices within this study showed no demonstrable bias in both hypernatremic and hyponatremic patient groups. A minimal positive bias was evident when hematocrit values were reduced, and a minimal negative bias when hematocrit levels were elevated. These clinical scenarios, however, would not preclude the POC system implementation in routine clinical practice.