ABSTRACT
This study assessed the sensitivity and specificity of third-generation serological hepatitis C diagnostic tests from an analysis of the literature. The literature analysis was run using criteria from McMaster University for the assessment of diagnostic tests. The selected studies were grouped according to the type of population at high and low risk for hepatitis C virus (HCV) infection and to the type of reference test. The homogeneity of the sensitivity and the specificity was tested in each group using a Fisher's exact test. Of 132 studies, 10 were selected. When the estimates were homogeneous, summary point estimates and confidence intervals were computed; when the estimates were heterogeneous, subgroup analysis was performed. The sensitivity of third-generation enzyme-linked immunosorbent assay (ELISA3) was estimated at 98.9% (95% CI: 94-100%) in patients with chronic liver disease and at 97.2% (95% CI: 92-99%) in panels of sera. ELISA3 specificity was found at 100% in patients with chronic liver disease. The sensitivity of the third generation recombinant immunoblot assay (RIBA3) was assessed at 78.8% (95% CI: 65-89%) in haemodialysed patients. This analysis provides evidence for the good sensitivity and specificity of ELISA3 assays particularly in high risk patient groups and confirms their use for screening in these populations. Further studies are needed to assess properly RIBA3 in general population and in risk patients.