Comparison of fifteen commercial assays for detecting Varicella Zoster virus IgG with reference to a time resolved fluorescence immunoassay (TRFIA) and the performance of two commercial assays for screening sera from immunocompromised individuals

The performance of fifteen, commercially available, VZV IgG assays and an "in house" indirect immunofluorescence (IF) assay has been compared to a reference VZV IgG time resolved immunofluorescence assay (VZV TRFIA). A panel of 273 VZV TRFIA IgG positive serum samples and 136 VZV TRFIA IgG susceptible sera, collected from a number of UK hospitals was used. Irrespective of the interpretation of equivocal results the most sensitive assays were Dade Behring EIA (97.4%), "in house" IF (95.2%), Human EIA (92.3%) and Becton Dickinson latex agglutination (94.1%). The least sensitive assays were Virion EIA (69.6%), Diesse EIA (68.9%) and Diasys EIA (68.5%). The least sensitive (<70%) assays all had >99.0% specificity whereas the most sensitive assays had lower specificities; for example, Dade Behring EIA had a specificity of 69.9% when equivocals were treated as VZV IgG negative. For some assays e.g. Dade Behring EIA there were major discrepancies between our findings and those reported by the manufacturer which may reflect the constitution of the panel(s) of sera used for evaluation or the reference method adopted or the choice of cut-off criteria (particularly relevant to our findings for the Behring EIA). Care must be taken to choose an assay with high specificity in order to accurately assess the need for vaccination or immunoprophylaxis; however, high sensitivity is preferable to prevent inappropriate and expensive treatment.