Multicentre evaluation of the Boehringer Mannheim Hitachi 911 system for homogeneous immunoassays

A multicentre evaluation of the new analyser, Hitachi 911, is reported for three different classes of homogeneous immunoassays (latex assays, immunoprecipitation assays, and CEDIA assays). The evaluation protocol follows ECCLS, IFCC and NCCLS guidelines. Using patient samples and commercial controls, within run and between run coefficients of variation were less than 3% in most cases, but as high as 9.7% for some CEDIA and latex assays. All the assays were linear, either in the reference or the therapeutic range of the analytes. No interference by haemolysis, lipaemia or icterus was observed. The methods were compared with other commercial methods. Coefficients of correlation were higher than 0.94 for all the methods. However, there were differences of slope and intercept for rheumatoid factor, apolipoprotein A-I and apolipoprotein B. On the Hitachi 911, all of the eight methods give precise and accurate results, and compare well with other established methods on immunoassay dedicated analysers. The discrepancies observed could be ascribed to current problems of immunoassay standardization.