Evaluation of a reagent-strip method for glucose in whole blood, as compared with a hexokinase method

The hexokinase method (I) for glucose in biological fluids, proposed by the FDA as a potential national standard method, is compared with the Eyetone/Dextrostix/Control method (II) for measurement of glucose in whole blood. I used a new, two-step calibration procedure based on a 1.30 g/liter standard for the II system. The correlation coefficient for results by the two methods was 0.99 for 142 assays in duplicate. Recovery values were near 100% (I, range 100.0 to 106.2%; II, range 100.8 to 105.1%), and the coefficients of variation were respectively +/- 4.1% and +/- 4.5%. Considering the close correlation between the two methods, together with the simplicity, speed, and convenience of II, I conclude that it has great clinical utility, particularly in emergency situations. The precision and accuracy of II depends greatly on proper calibration and control procedures and on adequate training of the operator.