Laboratory and clinical evaluation of two glucose meters for the neonatal intensive care unit

Objective: To evaluate the analytical and clinical performance of the One Touch II and Advantage glucose meters for use in neonatal specimens. Design and Methods: For the laboratory evaluation, a total of 96 umbilical cord whole blood specimens were analyzed on the One Touch II and/or Advantage meters. Samples were centrifuged after analysis on the meters and plasma glucose was determined on the Hitachi 917. For the clinical evaluation, a total of 64 infants had specimens analyzed on each of the meters as well as on the laboratory analyzer. Results: In the laboratory and clinical evaluations, both meters correlated well (r > 0.9, p < 0.001) with the plasma values from the Hitachi 917. However, the mean difference between the Advantage and Hitachi 917 was lower than that of the One Touch II in both the laboratory (-0.23 vs -0.64 mmol/L) and the clinical evaluations (-0.08 vs -0.60 mmol/L). 53.1% of One Touch and 26.6% of Advantage results from the clinical study had a discrepancy of >0.5 mmol/L from the laboratory value. Conclusions: For neonatal specimens, glucose meters must have good low end precision, sensitivity and accuracy. In this study, the Advantage meter had fewer discordant results and better correlation with the Hitachi 917. Overall, nursing and laboratory staff preferred the performance and characteristics of the Advantage meter. Copyright (C) 1998 The Canadian Society of Clinical Chemists.