Back

Evaluation of a new combined hepatitis C antigen/antibody assay for routine HCV testing of patient samples

Abstract number: 1733_1379

Vermeersch P., Van Meensel B., Van Ranst M., Lagrou K.

Objective: To evaluate a new combined hepatitis C antigen/antibody assay (HCV Ultra, Bio-Rad) in patient samples that were borderline positive or positive with AxSYM anti-HCV EIA (version 3, Abbott) in a routine hospital setting. This test was shown to have a smaller window phase for the detection of acute HCV infection compared to anti-HCV assays that only detect antibodies.

Methods: The performance of HCV Ultra was determined in 257 sera that were borderline positive (S/CO = 0.8–1.0) or positive (S/CO > 1.0) on AxSYM. The group of positive sera consisted of 82 of the 2408 in-house sera tested over a 2.5 month period and 175 sera referred for confirmatory testing. We also tested 18 sera that were negative on AxSYM. All sera were tested with Monolisa Plus Anti-HCV EIA (version 2, Bio-Rad). Sera that were AxSYM S/CO > 1.0 and Monolisa S/CO > 3 were considered positive. Otherwise immunoblot analysis was performed with INNO-LIA HCV Score (Innogenetics). When INNO-LIA did not allow a conclusion (indeterminate), the sample was tested with PCR for the presence of HCV RNA when enough serum was available.

Results: All 118 sera that were positive with both AxSYM and Monolisa were positive with HCV Ultra. The results of the 111 other sera that were not positive with both AxSYM and Monolisa are shown in table 1. A significant number of sera were undetermined with INNO-LIA (13.5%). These 15 sera were excluded for the calculation of the performance of HCV Ultra. In 13 of these samples, HCV Ultra gave a correct result. The sensitivity and specificity of HCV Ultra on AxSYM borderline and positive sera was 99% and 95%, respectively. The 18 sera that were negative on AxSYM were also negative on Monolisa and Ultra. All sera that were HCV Ultra S/CO ≥2.5 were from HCV-positive patients. The positive predictive value for in-house samples was 0% for AxSYM borderline positive sera and 54% for AxSYM positive sera. The only serum that was negative with HCV Ultra and positive with INNO-LIA was from a patient with normal liver enzymes and was negative with PCR. The 4 sera that were positive with HCV Ultra and negative with INNO-LIA could be from seroconverters as was confirmed by PCR in the only patient from which serum was available.

Table 1. Results in samples that were positive with AxSYM and Monolisa Plus

HCV Ultra\INNO-LIANegativePositiveUndetermined
Negative78112
Positive4133

Conclusion: The performance of HCV Ultra in sera that were AxSYM borderline or positive was excellent. Confirmation testing with INNO-LIA has little or no added value in sera tested with HCV Ultra.

Session Details

Date: 31/03/2007
Time: 00:00-00:00
Session name: European Society of Clinical Microbiology and Infectious Diseases
Subject:
Location: ICC, Munich, Germany
Presentation type:
Back to top