Laboratory assessment of three new monitors of blood glucose: Accu-Chek II, Glucometer II, and Glucoscan 2000.

We describe a laboratory assessment of three new monitors of blood glucose concentrations: the Boehringer "Accu-Chek II" (B), the Ames "Glucometer II" (A), and the Lifescan "Glucoscan 2000" (L). Inherent imprecision (CV) of each monitor was less than 2%. Maximum difference between individual monitors of the same type was less than or equal to 0.5 mmol/L. The volume of blood applied to the test strips is not critical, but duration of blood incubation or color development should be precise. Two types of test strips retained sufficient color 48 h after development to allow checking of the original measurement, and would be suitable as quality-control "spot" checks. Correlation coefficients for results for whole-blood glucose vs those for serum glucose (measured with the Beckman ASTRA-8) were: 0.992 (B), 0.967 (A), and 0.988 (L). Bias plots of these data showed positive bias for A (0.45 mmol/L) and L (0.17 mmol/L) in relation to serum-glucose measurements, but a negative bias of 0.32 mmol/L for B. Calibration solutions are not interchangeable. Although these versions of the monitors are probably not analytically superior to earlier models, they are easier to use.