Evaluation of HemoCue Glucose Meter (201+)

The most recent HemoCue glucose meter measures B-Glucose and converts the result to a plasma-equivalent glucose concentration (P-Glucose) by multiplying with 1.11. The authors studied the accuracy and precision of such a conversion with 120 samples from Herlev Hospital. The imprecision (CV) of the HemoCue glucose meter was 1.5%. HemoCue's conversion of B-Glucose to P-Glucose yielded a less than 10% difference from their routine method for P-Glucose (Eppendorf EBIO+) in more than 90% of samples. The mean bias was not statistically significantly different from zero. This performance satisfies the requirements for self-monitoring, point-of-care testing, and diagnostic determination of P-Glucose. The bias between HemoCue P-Glucose and Eppendorf P-Glucose in individual samples was related to hematocrit. The magnitude of the bias could only be partly explained by the difference in water concentration between blood and plasma. The glucose concentration in blood, but not in plasma, decreases with increasing hematocrit, because glucose in blood distributes like water, and the water concentration is lower in erythrocytes than in plasma. An additional hematocrit effect of the same magnitude and direction, perhaps caused by a slightly incorrect algorithm for background absorbance, was observed for HemoCue's B-Glucose. The two effects together could explain the observed bias in P-Glucose that was related to hematocrit.