D Robert Dufour1, Mageli Talastas, Maria D A Fernandez, Barbara Harris. 1. Pathology and Laboratory Medicine Service, Veterans Affairs Medical Center and Department of Pathology, The George Washington University Medical Center, Washington, DC, USA. d.robert.dufour@med.va.gov
Abstract
BACKGROUND: Antibodies to hepatitis C virus (anti-HCV) have typically been detected by enzyme immunoassay (EIA). A chemiluminescence assay (CA) for anti-HCV is now commercially available. METHODS: We compared the positive rate for a CA in a HCV screening program for veterans with historical rates obtained with EIA. We also compared results in 2824 samples tested by both methods and assessed the significance of low signal-to-cutoff (S/C) ratios. RESULTS: The frequency of CA-positive results was significantly lower than with EIA (12.6% vs 16.0%; P <0.0001). The frequency of low S/C ratios was also significantly lower with CA (11.5% vs 20.0%; P <0.0001). Among low-positive values, samples positive by CA were significantly less likely to be recombinant immunoblot assay (RIBA)-negative (64% vs 84%; P <0.0005). In parallel testing, results for 111 samples (3.9%) were discrepant between the two assays; all but 6 had low S/C ratios, and confirmatory testing was performed on all but 8 samples. Of 56 EIA-positive, CA-negative samples tested by RIBA, only 1 was positive. Of 24 CA-positive, EIA-negative samples, 62% were RIBA-negative. Using a negative RIBA result as an indication of false-positive anti-HCV results, the positive predictive value of EIA was 93% compared with 98% with CA. HCV RNA was positive in 90% of samples high-positive by both CA and EIA. Only 2 of 30 (7%) low-positive CA samples were RNA-positive. CONCLUSIONS: CA produces fewer false-positive and fewer low-positive results that require confirmatory RIBA testing. The S/C ratio remains useful for characterizing positive results.
BACKGROUND: Antibodies to hepatitis C virus (anti-HCV) have typically been detected by enzyme immunoassay (EIA). A chemiluminescence assay (CA) for anti-HCV is now commercially available. METHODS: We compared the positive rate for a CA in a HCV screening program for veterans with historical rates obtained with EIA. We also compared results in 2824 samples tested by both methods and assessed the significance of low signal-to-cutoff (S/C) ratios. RESULTS: The frequency of CA-positive results was significantly lower than with EIA (12.6% vs 16.0%; P <0.0001). The frequency of low S/C ratios was also significantly lower with CA (11.5% vs 20.0%; P <0.0001). Among low-positive values, samples positive by CA were significantly less likely to be recombinant immunoblot assay (RIBA)-negative (64% vs 84%; P <0.0005). In parallel testing, results for 111 samples (3.9%) were discrepant between the two assays; all but 6 had low S/C ratios, and confirmatory testing was performed on all but 8 samples. Of 56 EIA-positive, CA-negative samples tested by RIBA, only 1 was positive. Of 24 CA-positive, EIA-negative samples, 62% were RIBA-negative. Using a negative RIBA result as an indication of false-positive anti-HCV results, the positive predictive value of EIA was 93% compared with 98% with CA. HCV RNA was positive in 90% of samples high-positive by both CA and EIA. Only 2 of 30 (7%) low-positive CA samples were RNA-positive. CONCLUSIONS: CA produces fewer false-positive and fewer low-positive results that require confirmatory RIBA testing. The S/C ratio remains useful for characterizing positive results.