Abstract
BACKGROUND: Porcine reproductive and respiratory syndrome virus (PRRSV) and influenza A virus (IAV) are swine pathogens that can significantly impact the performance of post-weaning pigs. While oral fluid (OF) samples are widely used for monitoring these viruses, postmortem tongue fluid (TF) samples present a cost-effective alternative with potential advantages in viral detection. This study aimed to compare the performance of TF and OF samples collected from nursery and finishing pig herds in detecting PRRSV and IAV using reverse transcription-quantitative polymerase chain reaction (RT-qPCR). A Bayesian latent class model was used to estimate diagnostic sensitivity and specificity for TF and OF under the assumption of conditional independence. The study also examined the relationship between mortality rates and RT-qPCR outcomes, the success rate of Sanger sequencing for the PRRSV ORF-5 region, and the effect of pooling daily aggregated TF samples on the probability of PRRSV detection. RESULTS: IAV was detected in 34.9% of OF samples and 30.2% of TF samples, while PRRSV was identified in 67.4% of OF and 53.5% of TF samples. TF samples had a significantly lower mean Ct for PRRSV (29.1) compared to OF samples (32.8) but had a similar Ct (30.9) to OF (29.7) for IAV. The hierarchical latent class Bayesian model estimated the sensitivity and specificity values for OF as 37.3% and 61.7% for IAV, and 64.3% and 35.1% for PRRSV. The estimated sensitivity and specificity values for TF were 33.5% and 66.0% for IAV, and 53.0% and 47.0% for PRRSV. Among 22 matched TF and OF pairs submitted for PRRSV sequencing, 45.5% of OF samples and 63.6% of TF samples were successfully sequenced, with the higher success rate for TF attributed to having lower Ct values. Additionally, mortality rates were notably higher when PRRSV was detected, especially in cases with concurrent IAV detection. Regarding sample pooling, our results indicated that pooling TF samples significantly increased detection probabilities, with a 1/7 dilution achieving a 79% RT-qPCR detection rate, compared to a detection rate of 14.3% when testing a single day's TF sample from a week with only one positive day. CONCLUSION: The findings support the use of TF samples as a viable complement or alternative to OF samples for PRRSV and IAV surveillance in post-weaning pigs when mortalities are available. The cost-efficiency of TF sampling can enhance monitoring compliance, improve early pathogen detection, and facilitate timely responses to emerging threats in swine production. This study advocates for the adoption of TF as a risk-based sampling strategy in nursery and grow-finish settings, complementing live animal samples such as OF, ultimately contributing to better herd health management.