Low predictive value of seroprevalence of Toxoplasma gondii in cattle for detection of parasite DNA

Abstract

The role of beef in human infections with Toxoplasma gondii is not clear. To get a better understanding of the value of seroprevalence as an indication of the role of beef in human infections with T. gondii we studied the seroprevalence of T. gondii in Dutch cattle and analysed the correlation between detection of antibodies and parasitic DNA. An indirect ELISA was developed and used to test a sample of the Dutch cattle population. Since validation of the ELISA was hampered by a lack of sufficient bovine reference sera, the results were analysed in two different ways: using a cut-off value that was based on the course of the OD in 27 calves followed from birth until 16 months of age, and by fitting a mixture of two normal distributions (binormal mixture model) to the log-transformed ODs observed for the different groups of cattle in the study population. Using the cut-off value, the seroprevalence was estimated at 0.5% for white veal, 6.4% for rosé veal and 25.0% for cattle. However, using the frequency distributions the prevalences were higher: 1.9% for white veal, 15.6% for rosé veal and 54.5% for cattle. Next, for 100 cattle the results with two different serological assays (ELISA and Toxo-Screen DA) were compared with detection of parasites by our recently developed sensitive magnetic capture PCR. Toxoplasma gondii DNA was detected in only two seronegative cattle. This discordance demonstrates that seroprevalence cannot be used as an indicator of the number of cattle carrying infectious parasites. Demonstrating parasitic DNA in seronegative cattle and not in seropositive cattle suggests that only recent infections are detectable. Whether beef from these PCR-positive cattle is infectious to humans remains to be studied.