Plasma-equivalent glucose at the point-of-care:: evaluation of Roche accu-chek Inform® and Abbott precision PCx® glucose meters

Background: Glucose testing at the bedside has become an integral part of the management strategy in diabetes and of the careful maintenance of normoglycemia, in all patients in intensive care units. We evaluated two point-of-care glucometers for the determination of plasma-equivalent blood glucose. Methods: The Precision PCx (R) and the Accu-Chek Inform (R) glucometers were evaluated. Imprecision and bias relative to the Vitros (R) 950 system were determined using protocols of the Clinical Laboratory Standards Institute (CLSI). The effects of low, normal, and high hematocrit levels were investigated. Interference by maltose was also studied. Results: Within-run precision for both instruments ranged from 2-5%. Total imprecision was less than 5% except for the Accu-Chek Inform (R) at the low level (2.9 mmol/L). Both instruments correlated well with the comparison instrument and showed excellent recovery and linearity. Both systems reported at least 95% of their values within zone A of the Clarke Error Grid, and both fulfilled the CLSI quality criteria. The more stringent goals of the American Diabetes Association, however, were not reached. Both systems showed negative bias at high hematocrit levels. Maltose interfered with the glucose measurements on the Accu-Chek Inform (R) but not on the Precision PCx (R). Conclusions: Both systems showed satisfactory imprecision and were reliable in reporting plasma-equivalent glucose concentrations. The most stringent performance goals were however not met. (C) 2007 Elsevier B.V. All rights reserved.