ABSTRACT
<p><b>OBJECTIVE</b>To assess the feasibility of using saliva for Schistosomiasis japonica diagnosis.</p><p><b>METHODS</b>Schistosoma japonicum infected animal model was established. Pairs of saliva and serum samples from rabbits and chronic schistosomiasis patients were collected. Anti-schistosoma specific antibodies in saliva and serum were detected by indirect ELISA.</p><p><b>RESULTS</b>The specificities of antibody detection of rabbit saliva and serum were 93% (28/30) and 97% (29/30), respectively, and the sensitivities of antibody detection of rabbit serum and saliva were 100% (24/24) and 88% (21/24), respectively. A significant correlation (r = 0.5307, P = 0.0038 < 0.05) existed between anti-SEA IgG levels in serum and saliva. As with those in serum, anti-SEA IgG levels in saliva could reflect the state of infection and treatment. The sensitivity of antibody detection was 91% (29/32) for patient saliva samples and 100% (32/32) for their sera. 8 samples were positive in 140 normal saliva samples (i.e. 6% false positive rate) and 6 samples were positive in 156 normal serum samples (4% false positive rate). There was a significant correlation (r = 0.4227, P = 0.008 < 0.05) between specific antibodies in saliva and serum.</p><p><b>CONCLUSION</b>The detection of specific antibodies in saliva can be used as a non-invasive immunodiagnosis method of Schistosomiasis japonica.</p>