Performance of three blood glucose meters.

Abstract

To evaluate the performance of three blood glucose meters. The One Touch II (LifeScan, Milpitas, CA), Glucometer Elite (Bayer, Elkhart, IN), and Accu-Chek Advantage (Boehringer Mannheim, Indianapolis, IN) were compared with a reference laboratory method (Technicon Chem System, Tarrytown, NY). Blood glucose meters used in this study were validated by a clinically oriented approach known as the error grid analysis (EGA), for which the performance of the meters was compared to a laboratory standard, and by the criteria of the American Diabetes Association (ADA). Limits of agreement were evaluated using differences from the reference laboratory method and 95% Cls. Capillary blood was obtained from study participants in fasting state with the morning blood draw and tested on the three meters simultaneously. A total of 120 blood glucose meter readings were analyzed; values ranged from 62 to 396 mg/dL. For all three meters, at least 75% of the capillary blood glucose values fell into zone A (acceptable) of the EGA. The number of values falling into zone B (unacceptable) were 10, 8, and 6 for the Accu-Chek Advantage, the One Touch II, and the Glucometer Elite, respectively. Only 15-25% of the meter glucose readings met the ADA criteria of being within 5% of the laboratory standard. The mean difference from the reference values was least with the Glucometer Elite. The majority of blood glucose determinations obtained on the meters used in this study were within the acceptable limits using the EGA. The Glucometer Elite meter had the fewest values in the unacceptable range and had the least mean difference from reference laboratory values.