Untersuchungen zur Gerätevarianz und Chargenstabilität von Point-of-care-Testing Blutzuckermessgeräten

The guidelines of the German “Bundesärztekammer” for quality assurance of quantitative laboratory analyses (RiliBÄK) makes demands on the accuracy of POCT devices with unit-use-reagent for glucose determination when used in the medical field. Instruments for self-monitoring of blood glucose (SMBG) are routinely being used by diabetic patients. In addition, these instruments are frequently used for Point-of-Care-Testing (POCT) in hospitals. Because thereby immediate therapeutic decisions are made, the POC devices must deliver reliable and exact measuring results. The results of many evaluation studies have already been published including inter- and intra-assay imprecision, comparison with laboratory methods and different sources of interferences. Only a few studies on differences between individual instruments and test strip lot to lot differences have been published so far. Therfore here the influence of different instruments of the same type and company (n=10) as well as different test strip lots (n=3) on the measurement of blood sugar was investigated. The six instruments for self-monitoring of glucose tested were HemoCue Glucose 201+ (HemoCue AB), Ascensia Contour and Ascensia Elite XL (both Bayer Vital GmbH), Accu-Chek Sensor (Roche Diagnostics GmbH), MediSense Precision Xtra (Abbott Laboratories) and OneTouch Ultra (Firma LifeScan) with corresponding test strips or microcuvettes for the intra-assay imprecision, the instrument to instrument imprecision and the lot to lot imprecision. Overall, more than 10.000 measurements of EDTA-whole blood samples were done with glucose concentrations from 30 to 400 mg/dl. The coefficients of variation was calculated to determine precision. The determination of the intra-assay imprecision showed relative standard deviation between 1.3% and 4.8% for all instruments in the whole measurement range of glucose concentration. The determination of instrument to instrument imprecision for the measurement range from 30 to 400 mg/dl provided similarly good results for all analyzers tested with a slightly better performance of Ascensia Elite XL. The mean imprecisions were between 2.1% and 3.7%. The determination of lot to lot imprecision revealed mean relative standard deviations between 1.8% and 3.4% for glucose concentrations from 30 to 400 mg/dl. The microcuvettes of the HemoCue Glucose 201+ Blutglucoseanalyzer showed the best test result. Total imprecisions were between 2.5 and 4.8% for all instruments in the range from 60 to 400 mg/dl, with the HemoCue Glucose 201+ Analyzer showing the best total imprecision of 2.5%. With the increasing number of POCT instruments in hospitals in the future these differences have to be considered in the total quality management of POCT next to accuracy, inter- and intra-assay imprecision and linearity of an individual analyzer.