Infants born to mothers who seroconverted for toxoplasmosis during pregnancy are at risk of sequelae. In the case of a negative work-up at birth, congenital infection can be ruled out only by monitoring the disappearance of maternal immunoglobulin G (IgG) transmitted through the placenta, which can be achieved by regular blood sampling during the first year. To alleviate the discomfort of this follow-up, we developed an indirect enzyme-linked immunosorbent assay to detect specific IgG diffusing passively from the blood through the gingival epithelium by collecting oral fluid on microsponges. To assess the feasibility of the test, 212 patients were first enrolled. Levels of specific IgG in oral fluid were significantly higher in seropositive (n = 195) than in seronegative (n = 17) patients (mean optical densities, 1.145 +/- 0.99 versus 0.092 +/- 0.127; P < 0.0001). In a population of 93 patients < 15 months of age born to mothers who displayed toxoplasmic infection during pregnancy, 70 were free of congenital infection and were followed up until their serology turned negative, and 23 were congenitally infected. The same patterns of IgG were observed in the oral fluid and sera in each group. Using a cutoff of 0.04 (optical density value), the sensitivity and specificity of the test were 67.9% and 80.3%, respectively, and the probability of not having a congenital infection when the test on oral fluid was negative was 99%. Although the performance of the test needs to be improved, oral fluid sampling appears to be a promising tool for monitoring infants with suspected congenital toxoplasmosis.