High risk of false positive results in a widely used diagnostic test for detection of the porcine reproductive and respiratory syndrome virus (PRRSV).

During 2003 and 2004, increasing numbers of positive PRRSV RT-PCR results were reported from herds negative for PRRSV infection. Interestingly, three herds represent nucleus herds with no animal contacts from outside and without clinical symptoms of PRRS until now. Since these positive results that were obtained using a PCR protocol adapted to routine laboratory conditions could not be reproduced with other PRRSV specific RT-PCRs, controlled negative and positive samples were used to examine this phenomenon. A RT-PCR assay for detection and differential diagnosis of the European and North American genotypes of the porcine reproductive and respiratory syndrome virus (PRRSV) according to the method previously published by Oleksiewicz et al. [Oleksiewicz, M.B., Botner, A., Madsen, K.G., Storgaard, T., 1998. Sensitive detection and typing of porcine reproductive and respiratory syndrome virus by RT-PCR amplification of whole viral genes. Vet. Microbiol. 64, 7-22] was investigated in parallel to another recently published method [Pesch, S., 2003. Etablierung einer Nachweismethode für die zwei Genotypen von dem porcine reproductive and respiratory syndrome virus (PRRSV) und ein Beitrag zu seiner molekularen Epidemiologie. Thesis, Institute of Virology, Faculty of Veterinary Medicine, University of Leipzig]. A panel of 228 clinical samples sent in for PRRSV routine diagnostics served as test panel. It was found that both methods have similar analytical sensitivity. However, the primers published by Oleksiewicz were shown to yield a very high proportion of false positive results under routine diagnostic laboratory conditions, i.e. they resulted in RT-PCR products with non-PRRSV sequences, that were indistinguishable from truly positive reagents in standard gel electrophoresis settings. The reason for and possible implications of this finding as well as the risk of modifying published methods without control are discussed.