Porcine reproductive and respiratory syndrome virus (PRRSV) continues to be an outstanding health issue for the global swine production sector. The ability to accurately track the virus and characterize disease status between and within pig populations in today's swine industry is of high importance for the sustainability of the business. Towards that end, the collection of the necessary information to support PRRSV control and elimination programs is dependent upon continuous and reliable monitoring and surveillance systems (MOSS) that need to be both particularly practical and affordable. The issue addressed in this dissertation is the improvement of current PRRSV MOSS employed in breeding herds, through the development of a new sampling method called processing fluids and the assessment and optimization of its applicability for PRRSV RNA and antibody detection in breeding herds. The rational series of studies described below will address the issue pointed above.

Chapter Two introduced the processing fluids method as a new tool for PRRSV monitoring in breeding herds. This field study served as a proof of concept for this new sampling technique. Matching sets of processing fluids and serum samples obtained under field conditions from breeding herds, proved that PRRSV can be detected in processing fluids by reverse transcription quantitative polymerase chain reaction (RT-qPCR) at a higher frequency than the standard method of bleeding 30 pigs and testing the serum in pools of five (83.3% vs 66.6%). Moreover, IgG antibody detection was assessed and confirmed, and ORF5 sequencing was also achievable.

In Chapter Three, the probability of PRRSV RNA detection in processing fluid samples as a function of the within-litter prevalence of the virus was evaluated and compared with that of serum samples obtained from randomly selected piglets (1, 2, 3 or 4 per litter). Results suggest that when within-litter prevalence is ≥ 50% the probability of PRRSV RNA detection in processing fluids would be higher than that of randomly selected individual piglets (i.e., ≥ 99%). Additional comparisons between the processing fluids sampling approach and the standard PRRSV monitoring scheme using 30 serum samples were made through computer simulation (bootstrapping), giving as result an overall probability of PRRSV detection of 100% when using processing fluids vs 92.1% using 30 serum samples. Chapter Three also looked at processing fluids pooling potential, demonstrating that PRRSV detection was 100% achievable in processing fluids within a low prevalence scenario. The test of a massive pool with only 8% PRRSV-positive pigs (i.e., only 67 viremic pigs out of 834 total pigs in the pool) yielded a strong positive RT-qPCR quantification cycle (Cq) response (Cq = 22.0).

The two main objectives of the study described in Chapter Four were, 1) to evaluate the effect of pre-testing conditions of processing fluid samples on RT-qPCR testing results, such as temperatures and time length for sample storage and protocols for nucleic acid extraction and, 2) the adaptation of a commercial PRRSV serum antibody assay (ELISA) for the detection of three different anti-PRRSV antibody (Ab) isotypes (IgM, IgA and IgG) in processing fluids and to establish the test sample-to-positive ratio (S/P) cut-off, i.e., the point that best discriminated positive and negative samples. The two studies within this chapter revealed that, 1) PRRS virus in processing fluid samples would be stable under refrigeration for periods of time of up to 14 days. Also, prolonged exposure to room and higher temperatures would have, correspondingly, a mild and strong detrimental effect on the viral RNA within processing fluid samples, therefore, affecting RT-qPCR testing results. Results also suggested no effect in testing results between the four different commercial RNA purification kits. 2) The results obtained for PRRSV IgG Ab isotype showed perfect discrimination between positive and negative processing fluid samples using a cut-off value of 0.5. Results with the PRRS IgA and IgM ELISA for processing fluids showed low to poor discrimination and therefore would need additional research and optimization.

Chapter Five models the effect of pooling processing fluid samples on the probability of PRRSV RNA detection under a low prevalence scenario and establishes the limit for pooling processing fluids where the probability of PRRSV detection would not fall under 95% threshold. PRRSV RNA detection in pooled processing fluid samples of multiple litters (e.g., 29 to 65), is feasible when at least one pig within a given litter in the pool is positive for PRRSV with a RT-qPCR Cq value  29. Thus, demonstrating that, detection limits for pooling processing fluid samples might be dependent on the magnitude of viremia of the pig(s) within the sample. These findings could be useful for designing accurate and reliable processing fluid-based sampling protocols.