Comparison of direct and indirect methods to maximise the detection of Babesia caballi and Theileria equi infections in Central Southern Italy.

Equine piroplasmosis is a disease of equids, caused by tick-borne apicomplexan protozoan pathogens Babesia caballi and Theileria equi, which, according to the World Organisation for Animal Health (OIE), can be diagnosed by enzyme-linked immunosorbent assay (ELISA), immunofluorescent antibody test (IFAT) and polymerase chain reaction (PCR). The present study was conducted to evaluate and compare the assays available for the diagnosis of equine piroplasmosis. Data employed were obtained from 1300 blood samples collected between 2012-2014 from asymptomatic and symptomatic equines (horses and donkeys) of central-southern regions of Italy and analyzed by ELISA, IFAT, PCR (one commercial and one from literature) and blood smear microscopic examination. Statistical differences of the proportions of positivity for each parasite and group (asymptomatic and symptomatic) among the methods were verified by the z test to identify the most sensitive. The concordance between each pair of methods - for each parasite and within the groups - and trends in detection of suspect samples of four hypothetical diagnostic algorithms using serological and biomolecular assays were evaluated to identify the most suitable laboratory diagnostic workflow. The results of this study highlighted a lower capacity to detect suspect samples of commercial ELISA for B. caballi in all groups when compared to biomolecular methods and IFAT; and of the commercial PCRs in asymptomatic animals, identifying a PCR from literature and IFAT as the best choice for a combined diagnosis. For T. equi, IFAT detected more suspect samples than ELISA, even if the latter showed good performance and some samples were positive only by the ELISA and PCR, indicating that their simultaneous employment is still advantageous. Host-parasite interaction, amino-acid/genetic diversity and differences in detection limits among the assays could be among the reasons in explaining the present results. In view of further studies, ELISA should be used in combination with PCR, that should regularly be included in the laboratory diagnosis to maximise the detection of early infections and support the evaluation of pharmacological treatment.