Probability of Influenza A virus RNA detection at different pooling levels for commonly used sample types in swine breeding herds

Abstract

Pooling samples allows for efficient and cost-effective surveillance of endemic pathogens, enabling broader testing coverage and reducing diagnostic costs. Pooling swine samples for influenza A virus surveillance without negatively impacting sensitivity would depend on the sample type, cycle threshold (Ct value), and dilution level. Therefore, this study aimed to compare the probability of IAV reverse transcription real-time polymerase chain reaction (RT-rtPCR) detection at different pooling levels in family oral fluids, udder wipes, and nasal wipes obtained from an endemic swine breeding herd. Forty-five IAV RT-rtPCR-positive samples (15 family oral fluid, 15 udder wipes, and 15 nasal wipes) obtained from an IAV-positive herd were pooled at different levels (undiluted, 1/3, 1/5, and 1/10). For analysis, samples were categorized into three groups based on the undiluted replicates' mean Ct value: Category A - Ct < 30; Category B - Ct between 30 and 34; Category C - Ct between > 34–38. For all sample types, the probability of influenza A virus RNA detection in Ct category A and family oral fluids and udder wipes in Ct category B did not decrease when the dilution level increased to 1/10. In category C of all sample types, the probability of influenza A virus detection decreased as the dilution level increased. These findings suggest that pooling population-based samples can be an effective strategy for swine surveillance, particularly when working with samples with Ct values below 34, and future research in the field is needed to refine recommendations for field implementation.