ABSTRACT
OBJECTIVES: To develop and apply a mathematical model of the relationship between hematocrit, meter glucose and reference glucose concentrations as a tool for evaluation of whole blood glucose meters. DESIGN AND METHODS: Patient blood gas specimens were used to compare the LifeScan SureStep Flexx and the Nova StatStrip glucose meters with reference results obtained from a Radiometer 725 blood gas analyzer. Linear regression analysis was conducted to determine the extent that patient hematocrit and reference glucose concentrations predicted the performance of the glucose meters. RESULTS: Bland-Altman graphic analysis depicted that both glucose meters showed variance with the reference method. A mathematical model was derived from IFCC consensus equations that relate glucose meter results, hematocrit and plasma glucose values. Using the model, multivariate regression depicted that hematocrit affected the results of the SureStep Flexx meter in a manner dependent on the glucose concentration, whereas the StatStrip meter was not affected by hematocrit. CONCLUSIONS: Linear regression can be used to quantify the extent of hematocrit interference on the performance of glucose meters.
