par Pembrey, Lucy;Souayah, Hichem ;Levy, Jack ; [et al.]
Référence Journal of medical virology, 69, 2, page (195-201)
Publication Publié, 2003-02
Article révisé par les pairs
Résumé : To investigate whether it is appropriate to assume comparability of hepatitis virus C (HCV)-RNA results across laboratories in multi-centre studies, nine laboratories of the European Paediatric HCV Network participated in an international proficiency study of HCV-RNA assays. A panel of 12 samples of different dilutions and genotypes was sent to each laboratory and tested with qualitative and/or quantitative HCV-RNA assays according to local procedures. Commercial assays were used in seven laboratories and in-house assays in two. All six laboratories in which a commercial qualitative assay was used were proficient, as were four of six runs (in five laboratories) in which a commercial quantitative assay was used. The proficiency of the laboratories where in-house assays were used could not be assessed according to the VQC definition because of differences in the methods used. Overall, there were several false-negative results, but only one false-positive result with a quantitative assay and none with a qualitative assay. The false-negative results may have implications for the diagnosis of infection, and highlight the need for an antibody test to be performed at 18 months to confirm the absence of infection. The results of qualitative assays were generally consistent across laboratories but it was difficult to evaluate and compare the results of quantitative assays. Multivariate analysis of data collected in multi-centre studies should therefore allow for centre and/or assay used. © 2003 Wiley-Liss, Inc.