Evaluation of hepatitis C antibody testing in saliva specimens collected by two different systems in comparison with HCV antibody and HCV RNA in serum.

Two different ELISA assays, the Ortho HCV 3.0 ELISA (Ortho Diagnostics Systems) and the Mono-Lisa anti-HCV Plus (Sanofi Diagnostics Pasteur) were evaluated for the detection of hepatitis C virus (HCV) antibody in saliva samples. Specimens were collected from 152 individuals who participated in a longitudinal cohort study on HIV infection, and who used illicit drugs. Saliva specimens were collected using two different systems: Salivette (Sarstedt) and Omni-Sal (Saliva Diagnostic Systems). Saliva specimens were tested following modified protocols by both ELISAs, and the results were compared with serum specimens that were tested according to the instructions of the manufacturer. Serum samples of 102 (67%) participants were positive by both assays, and 50 persons were negative for HCV antibody. A total of 99 of the 102 serum specimens were confirmed as positive using Ortho Riba HCV 3.0 (Ortho Diagnostics System) and Deciscan HCV (Sanofi Diagnostics Pasteur), and 3 yielded discrepant results. As no cut-off level is known for testing saliva samples by ELISA, 3 different levels were chosen: mean (M) + 1 standard deviation (SD), M + 2 SD, and M + 3 SD of the optical densities of saliva tests of the 50 HCV serum antibody negative persons. At a level of M + 1 SD and M + 2 SD the Salivette/Mono-Lisa combination gave the greatest proportion of HCV antibody positive saliva specimens obtained from the 102 HCV serum antibody positive participants, 88% and 79%, respectively. Differences between the various collection systems and assay combinations were not significant statistically. In 76 of the 102 persons with HCV antibodies in serum, HCV RNA was detected in serum. Salivary presence of HCV RNA, however, could not be demonstrated. The results show that the assays compared are unsuitable for diagnostic use, but the sensitivities of the assays are acceptable for use in epidemiological studies. Copyright 2001 Wiley-Liss, Inc.