ABSTRACT
Foot-and-mouth disease virus (FMDV) and classical swine fever virus (CSFV) are highly contagious and can cause great economic losses when introduced into disease-free regions. Accurate estimates of diagnostic specificity (Sp) are important when considering the implementation of surveillance for these agents. The purpose of this study was to estimate diagnostic Sp of a real-time reverse-transcriptase PCR assay developed for detection of FMDV in cattle and domestic swine and CSFV in domestic swine based on non-invasive specimen collection. One thousand and eighty-eight range beef cattle were sampled from thirteen geographic locations throughout Texas. One thousand and one hundred market hogs and cull sows were sampled. Results for both FMDV and CSFV were considered positive if amplification occurred at or before 40 PCR cycles, inconclusive between 40 and 45 cycles and negative otherwise. Ten cattle had nonspecific PCR amplifications for FMDV, but none were classified as positive and only one as inconclusive. Specificity (95% confidence interval) was estimated as 100% (99.7, 100). There were 19 nonspecific PCR amplifications for FMDV in sampled swine with 1 classified as positive, 6 as inconclusive, and 12 as negative. Specificity (95% confidence interval) was estimated as 99.9% (99.5, 100). There were 21 nonspecific PCR amplifications for CSFV, and 1 was classified as positive. Specificity (95% confidence interval) was estimated as 99.9% (99.5, 100). These assays have high Sp, but nonspecific PCR amplifications can occur.
