Accuracy of HCV-RNA PCR tests for diagnosis or exclusion of vertically acquired HCV infection.

The aim of the study was to estimate the sensitivity, specificity, positive (PPV) and negative (NPV) predictive values, and likelihood ratios for HCV-RNA PCR tests for the early diagnosis or exclusion of HCV infection in vertically exposed children. Data were included for children with confirmed HCV infection status from a European multi-center study. Confirmation was dependent on antibody status at or beyond 18 months, the 'gold standard' measure of infection status against which the use of qualitative HCV-RNA PCR tests was assessed. Of the 547 children included in this analysis, 193 were HCV-infected and 354 were not. Sensitivity of the HCV-RNA PCR test was low at birth (22%), but increased to 85% by 6 months. Specificity of RNA PCR was constant over age at 98%. The PPV of the PCR test rose from 33% at birth to 78% at 9 months of age, while NPV ranged from 96% to 99%. The high positive likelihood ratios from 1 month of age indicate strong evidence to diagnose infection but the negative likelihood ratios were consistent with weak evidence to exclude infection. The results suggest that the first qualitative HCV-RNA PCR test should be delayed until after the first month of life given the low sensitivity in the first few weeks. Although a negative test result after this time indicates probable absence of infection, this should be confirmed with a negative anti-HCV antibody test between 9 and 15 months of age as negative PCR results can be observed in infected children with fluctuations in viremia. Copyright 2005 Wiley-Liss, Inc.