Hepatitis C virus (HCV) infection has been estimated in 600,000 subjects in France, with about 80 % of chronic infection. In the latter, anti-HCV antibodies and viral RNA are found together in patients blood. Today, only the use of polymerase chain reaction (PCR) technology allows the diagnosis of HCV chronic infection, confirmed by a positive PCR. However, PCR is a laborious and cost effective method. The aim of this study was to distinguish HCV chronic infection to past-infection or false reactivity only using the serology testing. Therefore, we looked for a correlation between the results of PCR, using the HCV Cobas Amplicor 2.0 assay, and the level of anti-HCV antibodies, assessed by the AxSYM HCV v.3.0 and expressed in signal/cutoff (s/co) ratio. We found using a panel of 200 sera issued from 181 patients, a significant variation of s/co ratios between PCR positive and negative patients (respectively, 87.76 +/- 27.18 vs 10.13 +/- 13.68 s/co, p < 0.0001), only in non treated or previously treated patients, non HIV coinfected, non renal transplanted or haemodialysis patients. An anti-HCV cutoff value at 34 s/co allows a predictive PCR results with 100 % sensitivity and 93.3 % specificity. Thus, for patients having a s/co equal or over 34, a positive PCR was found in 98.1 % of cases, allowing the diagnosis of HCV chronic infection (positive predictive value). Conversely, in patients with less than 34, HCV chronic infection can be excluded in 100 % of cases (negative predictive value). In conclusion, in most cases, the use of anti-HCV quantitative analysis in the AxSYM HCV v.3.0 assay could avoid PCR testing and facilitate the diagnosis of HCV chronic infection.