SENSITIVITY OF HIV ANTIBODY DETECTION IN SALIVA

To assess the sensitivity and specificity of HIV antibody detection in saliva we tested matched serum and saliva samples from HIV-infected a nd uninfected individuals. Saliva specimens were collected by two diff erent devices of the Salivette system and stored at different temperat ures. Samples were tested for HIV antibodies by two commercially avail able enzyme-linked immunosorbent assays (ELISAs; Wellcome, Biotest). H IV antibodies were detected in 98.5% (Wellcome) and 97.8% (Biotest) of the saliva samples (standard Salivettes) from 135 seropositive indivi duals. Using the Salivettes flavoured with citric acid the sensitivity was only 22.9%. No reactions in ELISA were found in saliva from HIV-s eronegative individuals. Salivary HIV-specific IgA was detected in 90% of seropositive individuals. All positive saliva samples stored at ro om temperature were still reactive after 20 days; of those stored at 3 7-degrees-C 23 out of 24 were positive when retested on day 5. Sensiti vity of HIV antibody detection in saliva samples dried onto filter pap er was 100% when a minimum of 100 mul of saliva was used. HIV antibody testing in saliva is an efficient tool for large scale epidemiologica l studies when standard Salivettes are used for sample collection. Sal iva samples can be stored in Salivettes or dried onto filter paper for several days at room temperature and under tropical conditions (37-de grees-C).