Probability of detecting influenza A virus subtypes H1N1 and H3N2 in individual pig nasal swabs and pen-based oral fluid specimens over time.

The probability of detecting influenza A virus (IAV) by virus isolation (VI), point-of-care (POC) antigen detection, and real-time reverse-transcription polymerase chain reaction (rRT-PCR) was estimated for pen-based oral fluid (OF) and individual pig nasal swab (NS) specimens. Piglets (n=82) were isolated for 30 days and confirmed negative for porcine reproductive and respiratory syndrome virus, Mycoplasma hyopneumoniae, and IAV infections. A subset (n=28) was vaccinated on day post inoculation (DPI) -42 and -21 with a commercial multivalent vaccine. On DPI 0, pigs were intratracheally inoculated with contemporary isolates of H1N1 (n=35) or H3N2 (n=35) or served as negative controls (n=12). OF (n=370) was collected DPI 0-16 and NS (n=924) DPI 0-6, 8, 10, 12, 14, 16. The association between IAV detection and variables of interest (specimen, virus subtype, assay, vaccination status, and DPI) was analyzed by mixed-effect repeated measures logistic regression and the results used to calculate the probability (pˆ) of detecting IAV in OF and NS over DPI by assay. Vaccination (p-value<0.0001), DPI (p-value<0.0001), and specimen-assay interaction (p-value<0.0001) were significant to IAV detection, but virus subtype was not (p-value=0.89). Vaccination and/or increasing DPI reduced pˆ for all assays. VI was more successful using NS than OF, but both VI and POC were generally unsuccessful after DPI 6. Overall, rRT-PCR on OF specimens provided the highest pˆ for the most DPIs, yet significantly different results were observed between the two laboratories independently performing rRT-PCR testing.