[An inter-laboratory study of anti-HCV antibody detection in saliva samples]

The purpose of this study was to evaluate the efficacy of anti-hepatitis C virus (HCV) antibody detection in the saliva samples of 108 drug users in an inter-laboratory study.

Between January and June 2001, 108 subjects in Lille, Metz and Lens received a test to detect anti-HCV antibodies in their saliva. Two consecutive saliva samples were taken in each subject (Salivette system, Sarstedt). An HCV serology (Axsym HCV 3.0, Abbott) was also performed and serum HCV RNA detection by Amplicor HCV 2.0 (Roche) was performed when HCV serology was positive. Sixty three patients had a negative HCV serology, 45 had a positive HCV serology, and 31 of these had positive HCV RNA as well. Tests for the detection of the anti HCV antibody in saliva samples were performed as a blind study in both the Lille and the Thionville laboratories.

The sensitivity of saliva anti-HCV antibody tests was respectively 71% (32/45) and 78% (35/45) in Lille and Thionville. In the event of positive HCV viremia, the sensitivity was respectively 90% (28/31) and 93% (29/31). The specificity was respectively 97% (61/63) and 98.5% (62/63). Results from the two laboratories agreed for 101 saliva tests while discrepancies were found in 7 (Kappa Concordance Coefficient: 0.85).

This study confirms, in a large, unselected population sample, that anti-HCV antibody detection tests in saliva allow the detection of 90% of viremic HCV-antibody-positive patients with excellent specificity. The simplicity and reproductibility of this technique makes it a precious tool for epidemiological studies.