Analytical evaluation of the Optium Xido blood glucose meter.

Background: The main prevention of chronic diabetic complications is maintaining near normoglycemia. In this study, evaluation of analytical performance of the Optium Xido blood glucose meter designed for patients' self-monitoring of glycemia was carried out.

Methods: Glucose concentrations were measured in 118 EDTA blood samples using the Xido glucose meter and the GOD/PAP method on a Modular P clinical analyzer. The results obtained were used for the assessment of accuracy, precision and linearity. Performance of the Xido glucose meter was also assessed using four different reagent strip lots, and the between-lot variation was calculated.

Results: The within-run imprecision coefficient of variation (CV) amounted to 4.2%. Good response linearity was found in glucose concentrations of 31-444 mg/dL (1.7-24.7 mmol/L). In the concentration range studied, the glucose meter error was 5.14% and the linear regression equation was y=0.91x+6.2, (r=0.984). The Passing-Bablok agreement test indicated good concordance of results. However, for glucose concentrations<100 mg/dL (5.6 mmol/L) (n=69) the error was 6.82% with regression equation y=0.86x+5.9 (r=0.757). Between-lot differences amounted to 0.7%-18.2%.

Conclusions: The Xido glucose meter has good precision and accuracy when compared to the laboratory method and meets the quality recommendations of the National Committee of Clinical Laboratory Standards (NCCLS, currently the Clinical Laboratory Standards Institute), the National Academy of Clinical Biochemistry (NACB) and the International Organization for Standardization (ISO). However, between-lot variability may result in differences between day-to-day results when the device is continuously used. Therefore, the validation of new lots of reagent strips with a laboratory method is recommended.