Background and Objectives: Considering the possibility of false-positives in the bovine herpes virus diagnostic tests, this study evaluated
how probable to occur a cross-reaction among bovine herpes virus BoHV1, BoHV5 and BuHV1 is and how this can impact in a prevalence
Methods: 101 bubaline serums were sampled from a slaughterhouse located in Boituva City-SP (Brazil). The search for anti-BoHV1 and
anti-BoHV5 was carried out by virus-neutralization technique (adapted from OIE protocol). On the other hand, the evaluation of the presence
of anti-BuHV1 was developed by ELISA technique (commercial indirect ELISA kit). From the results of the diagnostic, the conditional
probabilities were calculated. For those calculation, each virus was considered reference and compared with the other two species separately.
Results: The condition of higher probability of cross-reaction was P(BoHV1+|BuHV1+) = 0.97, which means the probability of an animal
is positive for BoHV1, given it is infected for BuHV1. On the other hand, the condition of lower probability was P(BuHV1+|BoHV5+) =
0.86, which indicates the probability of an animal is positive for BuHV1, given it is infected for BoHV5. Moreover, if the prevalence of
BuHV1 is higher than 7.8%, at least half of positive cases detected for BoHV5 may indeed be BuHV1. The inverse thinking is considered if
the investigator is interested to estimate the prevalence of BuHV1 where there are cases of BoHV5. In this case, the “threshold” prevalence
of BoHV5 is 5.3%.
Conclusions: The prevalence estimation of BuHV1 is the most influenced by the previous prevalence of BoHV1. In opposition, the
prevalence estimation of BoHV1 is the less influenced by the previous prevalence of BuHV1.