TYPE Original Research PUBLISHED 24 November 2022 DOI 10.3389/fvets.2022.953918 Porcine reproductive and OPEN ACCESS respiratory syndrome prevalence and processing fluids EDITED BY Ioannis Magouras, City University of Hong Kong, Hong Kong SAR, China REVIEWED BY use for diagnosis in United Yosuke Sasaki, Meiji University, Japan Jean-Pierre Frossard, States breeding herds Animal and Plant Health Agency, United Kingdom Mariana Kikuti1 , Carles Vilalta1,2,3 , Juan Sanhueza1,4 , *CORRESPONDENCE Cesar A. Corzo Claudio Marcello Melini1 and Cesar A. Corzo1* corzo@umn.edu 1 Department of Veterinary Population Medicine, University of Minnesota, Saint Paul, MN, SPECIALTY SECTION United States, 2 Unitat mixta d’Investigació IRTA-UAB en Sanitat Animal, Centre de Recerca en Sanitat This article was submitted to Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), Barcelona, Spain, Veterinary Epidemiology and 3 Instituto de Investigación y Tecnología Agroalimentaria (IRTA), Programa de Sanitat Animal, Centre Economics, de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), a section of the journal Barcelona, Spain, 4 Departamento de Ciencias Veterinarias y Salud Pública, Facultad de Recursos Frontiers in Veterinary Science Naturales, Universidad Católica de Temuco, Temuco, Chile RECEIVED 26 May 2022 ACCEPTED 05 September 2022 PUBLISHED 24 November 2022 Introduction: Processing fluids have been recently adopted by the U.S. swine CITATION industry as a breeding herd PRRS monitoring tool due to their increased Kikuti M, Vilalta C, Sanhueza J, Melini CM and Corzo CA (2022) representativeness of animals within the herd. Here, we use the Morrison Swine Porcine reproductive and respiratory Health Monitoring Project (MSHMP) database, representative of ∼50% of the syndrome prevalence and processing U.S. swine breeding herd, to describe processing fluids submissions for PRRS fluids use for diagnosis in United States breeding herds. diagnosis and their relation to PRRS prevalence and time to stability over time Front. Vet. Sci. 9:953918. between 2009 and 2020. doi: 10.3389/fvets.2022.953918 Methods: An ecological time series Poisson regression modeling the number COPYRIGHT © 2022 Kikuti, Vilalta, Sanhueza, Melini of status 1 farms and weekly percentage of processing fluids submissions and Corzo. This is an open-access for PRRS diagnosis was done. Time to stability was calculated for sites that article distributed under the terms of the Creative Commons Attribution detected a PRRS outbreak within the study period and modeled through a License (CC BY). The use, distribution proportional hazards mixed effect survival model using production system as or reproduction in other forums is a random-effect factor and epiweek as a panel variable. permitted, provided the original author(s) and the copyright owner(s) Results: Processing fluids diagnosis submissions increased starting in 2017. are credited and that the original The difference between each year’s highest and lowest weekly prevalence publication in this journal is cited, in accordance with accepted academic averaged 10.9% between 2009 and 2017, whereas it averaged 5.0% in 2018– practice. No use, distribution or 2020 period. Each year’s lowest weekly prevalence ranged from 11.3 to 19.5% reproduction is permitted which does not comply with these terms. in 2009–2017 and from 22.4 to 29.2% in 2018–2020. We also detected an increasing proportion of breeding sites that did not reach stability within 1 year of reporting an outbreak (chi-square for trend p < 0.0001). The total time to stability was not associated with the region of the country in which the site was located, the site’s air filtration status, its PRRS status before the outbreak, or the different statuses a site achieved to be classified as stable, when accounting for the production system in the multivariate model. However, a higher proportion of system-wide processing fluids use was associated with increased time to stability. Discussion: Altogether, the temporal concurrence of processing fluids used for PRRS virus monitoring suggests that the adoption of this sampling strategy Frontiers in Veterinary Science 01 frontiersin.org Kikuti et al. 10.3389/fvets.2022.953918 may help explain the changes observed in PRRS status 1 prevalence since 2018, although further studies are still needed. KEYWORDS disease outbreak, porcine reproductive and respiratory syndrome, epidemiology, swine diseases, epidemiological monitoring Introduction represented on the farm’s monitoring program, this strategy is potentially more sensitive in detecting PRRS, particularly at Porcine reproductive and respiratory syndrome (PRRS), lower within-farm prevalence. caused by Betaarterivirus suid 1 (PRRSV-1) and Betaarterivirus PRRS prevalence in breeding herds, or the proportion of the suid 2 (PRRSV-2) (1), has been a burden on the U.S. swine breeding herds that are classified as PRRS-positive and weaning industry for over three decades, causing a major economic PRRS-virus-positive piglets at a given point in time, is a product impact (2). Efforts have been made in the last decade to share of both the incidence (number of new herds that detect a new data and conduct analyses of different levels to control, prevent, PRRS strain) and the time that a herd requires to control the and eliminate it (3). Herd surveillance of PRRS is a crucial within-herd transmission of a wild-type or vaccine strain. This activity to understand the dynamics of transmission, but also is commonly known as time to stability. Once the herd starts to control and eliminate the virus from a production system. to consistently produce PRRS-negative pigs (i.e., there is an For breeding herds, a standardized terminology was proposed absence of clinical signs and a lack of detectable viremia in based on serum samples of wean-age pigs (4). Blood collection weaning-age pigs for at least 90 days), the herd is known to from 30 to 60 randomly selected due-to-wean piglets for reverse have reached stability (4). Many veterinarians and producers transcriptase polymerase chain reaction (RT-PCR) testing was also opt to pursue elimination (i.e., negative PRRS shedding and the industry standard up until recently to detect the presence of exposure status), in which case reaching stability can be one of the virus, assuming this age group accurately represented PRRS the steps toward elimination. The adoption of more sensitive status in the breeding herd (4). Since then, new monitoring monitoring tools, such as the molecular testing (e.g., RT-PCR) strategies have arisen, such as the use of processing fluids to of processing fluid samples for PRRS-virus RNA detection, can screen the newborn pig population in breeding herds, leading potentially increase our ability to detect disease occurrence. to an updated classification system including this new sampling Thus, we aimed to describe the temporal correlation between methodology (5). the use of processing fluids for PRRS virus detection and the Processing fluids are the serosanguineous exudate obtained overall PRRS prevalence in the U.S. breeding herd population, from tails and testicles as part of castration and tail-docking as well as the average time it took farms to reach stability over practices. This sampling methodology has become widely the studied period. adopted in the U.S. swine industry since it is convenient, easy to adapt to daily farm chores, reliable, and cost-effective, and Methods the sample can be collected from a large number of piglets to monitor for PRRS in breeding herds (6, 7). This sample type is An ecological study was designed to investigate the potentially more sensitive in detecting the PRRS virus when only association between PRRS prevalence in breeding herds and a few animals are infected. The increased sensitivity to detect if the use of processing fluids for PRRS diagnosis. The studied PRRS is present in a given farm when using processing fluids population was all the production systems participating in may rely on the increased representativeness, given this sample the Morrison Swine Health Monitoring Project (MSHMP), is generated through routine practices in a large number of 3– which represents ∼50% of the U.S. swine-breeding herd (3). 5 day-old pigs. The number of litters that can be aggregated to The MSHMP collects data generated by participating systems’ reduce testing costs depends on the cycle threshold (Ct) value of routine monitoring for PRRS and other infectious diseases. the positive sample being diluted. The lower the Ct-value (i.e., For PRRS, participating systems report changes (i.e., outbreak, the higher the viremia in the individual animals), the higher stability, elimination) in the health status for each of their the level of aggregation that can be achieved. An aggregated breeding farms weekly according to the American Association of sample of 40 litters is predicted to detect a PRRS-virus positive Swine Veterinarian PRRS breeding herd classification guidelines litter when the Ct-value of that positive sample is below 30 (8). proposed in 2011 (4) for the duration of this study. Briefly, Testing aggregated processing fluid samples reduces diagnosis status 1 indicates herds that are positive unstable in which cost while optimizing labor compared to blood collection of within-herd viral transmission and shedding is present. Status 30–60 due-to-wean piglets (7). Because more animals are being 2 represents seropositive breeding herds classified as positive Frontiers in Veterinary Science 02 frontiersin.org Kikuti et al. 10.3389/fvets.2022.953918 stable with a sustained lack of viremia in weanling pigs. Statuses diagnosis that were processing fluids per production system per 3 and 4 are herds classified as provisional negative or negative, year. Sites’ contribution times were censored in December 31, where no viral shedding exists but seropositive (in the case of 2020 for sites that were still unstable. Factors associated with status 3) or seronegative (in the case of status 4) animals are time to stability with a p ≤ 0.10 in the bivariate model were present at the farm. Farms in status 2 were further classified into included in the full multivariate model. A stepwise backwards status 2fvi (field virus inoculation) or 2vx (modified-live virus elimination process was done using p ≤ 0.05 as a cutoff to vaccine use) according to the elimination or control strategy define the final model. All statistical analysis was conducted adopted. Therefore, the weekly PRRS prevalence, defined as the using STATA 17.0 (11). percentage of farms in each status category from July 2009 to December 2020, was calculated on a weekly basis according to participant reports and was plotted over time. Results Most (75%, 30/40) participating companies conduct their PRRS-virus diagnosis for at least a fraction of their sites A median of 889 sites were followed weekly throughout in two main Veterinary Diagnostic Laboratories (VDLs), the study period (a minimum of 364 sites and a maximum University of Minnesota (UMN) and Iowa State University of 1,051 sites). The lowest median weekly number of sites (ISU). Data regarding the sample type submitted for PRRS- that contributed was 365 in 2009 and the highest was 1,033 virus diagnosis by RT-PCR were obtained for all submissions in 2019. A decrease in PRRS statuses 2 and 2fvi weekly from MSHMP participants between January 2015 and December prevalence has been observed since 2013, accompanied by 2020 from both UMN and ISU VDLs. Data from both an increase in status 2vx prevalence (Figure 1). This change VDLs were combined, and the percentage of sample types in the pattern regarding status 2 and its sub-classifications submitted monthly throughout the studied period was plotted. remained practically unchanged for the remainder of the studied Additionally, an ecological time series Poisson regression period. Regarding the weekly prevalence of status 1 sites, a modeling the number of status 1 farms and the weekly seasonal pattern of higher prevalence in winter months going percentage of processing fluids submissions for PRRS diagnosis into spring and lower prevalence in summer and fall months was done (9) using the number of total sites enrolled in MSHMP from 2009 to late 2013 was observed (Supplementary Table 1). that were sharing weekly health statuses as exposure. A second Maximum and minimum weekly prevalence for each year model added a flexible spline function to account for seasonality during this period ranged from 19.2 to 31.5% and 10.7 to and long-term trends in the status 1 prevalence with 52 knots. 15.6%, respectively. The percentage difference between the Both models were compared based on Akaike information highest and lowest weekly prevalence in each year was 5.9% in criterion (AIC) and the best model’s deviance residuals were 2009, in which monitoring started in the middle of the year, plotted over the studied time period. and ranged from 12.9 to 17.1% for 2010–2013 with a median Time to stability was calculated for sites that detected a PRRS yearly difference of 14.1% throughout this period. In 2014, outbreak within the study period. It comprised the number however, status 1 weekly prevalence showed less pronounced of weeks from when the breeding herd status was changed peaks and valleys compared to the observed in previous years, to status 1 for the first time because of an outbreak to when as shown by the 5.0% difference between the highest (19.2%) it was reclassified into status 2, 2fvi, 2vx indicating it had and lowest (14.2%) weekly prevalence for the year. For 2015– reached a stable status. Time to stability for farms that had a 2017, the median difference between the yearly highest and PRRS outbreak (i.e., sites that transitioned to status 1) during lowest weekly prevalence increased to 10.6%, ranging from 2015-2016 (before processing fluids testing was available) were 8.4 to 10.9%, illustrating that peaks and valleys returned to compared to those in which the outbreak occurred during 2018– being more clearly defined throughout the period. However, 2020 (after processing fluids testing was available) using the since 2018, the seasonality of status 1 prevalence has not Wilcoxon rank–sum test. The percentage of farms that did not been as clearly defined as in previous years. Not only did the reach stability within 1 year according to the year in which the median yearly difference between the higher and lower weekly PRRS outbreak was detected was compared using chi-square for prevalence decrease to 5.0%, but 2018–2020 were also the years trend. Additionally, each farm’s time to stability from farms that with the highest lowest weekly prevalence. The lowest weekly experienced a PRRS outbreak from January 2015 to December prevalence observed in 2018, 2019, and 2020 was 23.8, 22.1, and 2020 (a time in which PCR data was also available) was modeled 26.3%, respectively. through a proportional hazards mixed effect survival model The median monthly PRRS RT-PCR submissions during using production system as a random-effect factor and epiweek the study period was 7,936, ranging from 6,815 to 9,397. as a panel variable according to: (1) the region of the country, as Based on this dataset, processing fluids specimens started defined by the Swine Health Information Center (SHIC) (10); (2) being submitted in August 2017 and became frequently used air filtration status; and (3) system-wide processing fluids use, during 2018–2020, comprising 15.2 to 26.5% of all diagnostic defined by the overall proportion of samples submitted for PRRS specimen submissions for that period (Figure 2). Since August Frontiers in Veterinary Science 03 frontiersin.org Kikuti et al. 10.3389/fvets.2022.953918 FIGURE 1 PRRS prevalence of United States sow herd status from 1,140 herds participating in the Morrison Swine Health Monitoring Project beginning July 01, 2009. 2018, a median of 1,544 (ranging from 1,106 to 1,958) (IQR): 19–48], before processing fluids started being used, than processing fluids samples were submitted monthly for PRRS between 2017 and 2020 (median 35 weeks; IQR: 22–56; p = RT-PCR detection. 0.002), after they were adopted. Similarly, the chi-square for Prevalence of status 1 farms over time, as well as the trend for the proportion of sites that did not reach stability weekly percentage of processing fluids submissions over time, within 1 year was significant (p=0.01). Of all sites that detected are plotted in Supplementary Figure 1. The naïve Poisson model a PRRS outbreak in 2015, 21.7% did not reach stability within to assess status 1 sites and the percentage of processing fluids use 1 year. This percentage decreased to 17.8% in 2016, and then association without adjustment for long-term patterns showed increased to 23.5, 31.4, 28.4, and 27.7% in 2017, 2018, 2019, and an incidence rate ratio (IRR) of 1.009 (95% CI 1.007–1.01, p 2020, respectively. < 0.001), suggesting a slight increase in the number of status Factors associated with each farm’s time to stability are 1 farms when the percentage of processing fluid use is higher described in Table 1. In the bivariate analysis, region, system- (AIC = 1,715.029). However, after adding the spline function to wide processing fluids use, and status when stable were account for long-term trends, this association does not remain associated with time to stability at p ≤ 0.10 and selected for statistically significant (IRR = 0.999; 95% CI 0.998–1.001; p inclusion in the multivariate model. In the multivariate model, = 0.36) and the model is improved (AIC = 1,696.557). The however, only processing fluids use remained associated with deviance residual is shown in Supplementary Figure 2, which time to stability. The higher the system-wide use of processing suggests that this model’s performance was poorer earlier in the fluids in the year the outbreak was detected, the lower the study period. rate to reach stability (i.e., higher time to stability) was found. A total of 1,436 herds had a PRRS outbreak, of which 1,203 Contrasts of the global mean of time to stability by each level of reached stability within the studied period. The median time categorical variables were also assessed (Supplementary Table 2); to stability was lower in sites that detected the PRRS outbreak however, no difference between each level and the global mean between 2015 and 2016 [median 32 weeks; interquartile range was observed. Frontiers in Veterinary Science 04 frontiersin.org Kikuti et al. 10.3389/fvets.2022.953918 FIGURE 2 Percentage of sample type specimens submitted for PRRS RT-PCR diagnosis at the UMN and ISU Veterinary Diagnostic laboratories amongst MSHMP participants. Discussion processing fluids. Instead, we used system-wide processing fluids use as a proxy. However, we hypothesize that it is Given that the PRRS yearly incidence was lower from July possible that farms at low within-herd prevalence were being 2018 to June 2020 (average cumulative incidence of 24.2%) than misclassified as stable (e.g., false stability) prior to the adoption from July 2014 to June 2018 (average cumulative incidence of a more sensitive sampling strategy. Nonetheless, this temporal of 29.9%) (12), and that overall positivity in PRRS RT-PCR correlation does not necessarily mean causation, and other submissions was within the expected range (13), a reasonable factors might also be involved in this change in prevalence hypothesis for the increase in prevalence of status 1 farms pattern. Particularly since while the prevalence of status 1 farms during 2018 onwards is that farms that had PRRS outbreaks plateaued from late 2019, the frequency of processing fluids use are remaining in status 1 for longer periods of time than did not significantly change from mid-2018. Additionally, it is in previous years. This is supported by the PRRS prevalence important to note that due-to-wean piglets might yield positive pattern observed, in which the cyclical decrease in prevalence PRRS RT-PCR results even after two consecutive negative 4- is no longer clearly observed during the summer and fall week batches processing fluids results, thus monitoring should seasons, and by the significant chi-square for trend in the not rely solely on this monitoring strategy (14). percentage of sites that do not reach stability within 1 year. Other factors must be considered as additional factors However, these are crude estimates with no adjustments made explaining the increased prevalence, particularly because PRRS for production system or other factors that could potentially occurrence and processing fluids use were not consistently influence disease reporting. associated throughout the different ecological analysis. For The temporal concurrence of processing fluids used for example, the introduction of new strains in already infected PRRS virus monitoring suggests that the adoption of this herds will contribute to them remaining in category 1. Specific monitoring strategy may help to explain the changes observed virus strains may have developed the capacity to persist for in PRRS status 1 prevalence since 2018. This is supported by longer periods of time at the population level, making the time to the positive association found between a higher system-wide stability longer than expected. Viral strains associated with PRRS proportion of processing fluids submitted for PRRS diagnosis outbreaks were not assessed in this study. The new L1C 144 and a higher time to stability. An important limitation of this PRRS variant (15) only emerged as a significant health problem analysis was that we were not able to discern what laboratory in late 2020 and remains restricted to one region of the U.S. tests were responsible for each site’s status change, and whether Thus, we expect its effect in PRRS prevalence trends would be the farms with a higher time to stability were the ones using minimal in this study. However, the countrywide outbreaks of Frontiers in Veterinary Science 05 frontiersin.org Kikuti et al. 10.3389/fvets.2022.953918 TABLE 1 Bivariate proportional hazards mixed effect survival model than the ones that used field-virus inoculation (19). However, results between time to stability and investigated predictors, using although we did not assess the role of interventions during the production system as a random-effect factor. unstable period, sites that were classified as 2fvi had a higher n Bivariate rate in reaching stability than sites that were classified as 2 in HR 95% CI p-value the bivariate analysis, but that association was not sustained in the multivariate model. Additionally, a previous study with a Region similar dataset showed a shorter time to stability in farms that 2 283 1 – – experienced PRRS outbreaks from a previous status 2fvi, 3, or 4 3 682 0.71 0.50–1.01 0.06 compared to sites previously at status 2vx (20). Here, previous 4 38 0.53 0.27–1.05 0.07 status 2fvi was only marginally associated with a higher rate to 5 359 1.06 0.66–1.72 0.80 reach stability compared to sites previously in status 2vx, but this Air filtration was also not sustained in the multivariate model. None 479 1 – – This study not only describes the industry-wide adoption Partial 33 0.87 0.52–1.46 0.61 of processing fluids as a monitoring tool for PRRS diagnosis, Year-Round 180 1.24 0.94–1.62 0.13 but also a temporal correlation between its adoption and an Processing fluids use* 1,137 0.16 0.08–0.34 <0.001 increased PRRSV prevalence and time to stability in recent years. Status when stable Despite this, additional studies are necessary to ascertain the 2 170 1 – – direct relationship between processing fluids use and longer time 2fvi 460 1.43 1.09–1.86 0.01 to stability. 2vx 573 1.10 0.89–1.35 0.38 Previous status Data availability statement 2fvi 378 1.22 0.99–1.51 0.07 3 40 1.15 0.79–1.68 0.47 The datasets presented in this article are not readily available 4 210 1.04 0.85–1.28 0.70 because the data that support the findings of this study are 2vx 750 1 – – generated and owned by the production systems participating CI, confidence interval. in this study and is available on request from the corresponding *Production system-wide proportion of submitted samples that were processing fluids in author. The data are not publicly available due to privacy the year the PRRS outbreak was detected. restrictions. Requests to access the datasets should be directed to corzo@umn.edu. porcine epidemic diarrhea virus (PED) in 2013 and 2014 (16, 17) could have affected PRRS detection, since all efforts were focused on the detection, control, and elimination of this new disease Ethics statement that heavily affected the U.S. swine industry. There are ∼40% No ethical approval was required as this article describes of the breeding herds in category 3–4 (18), which is certainly data collected through routine veterinary diagnosis by the farms’ a good indicator of practitioners and producers continuing to veterinarians as part of standard care. work toward maintaining naïve herds and working toward a naïve status. Additionally, the adoption of processing fluids and the decision to pursue elimination may vary widely according Author contributions to the production system or perceived disease pressure within a given region. To account for this effect on the farms’ MK and CAC contributed to the conception and design time to stability, production system was added as a random of the study. MK performed data analysis with critical effect in our model and region was assessed as a possible contributions from CV, JS, and CMM. MK wrote the first draft of associated factor. Nevertheless, that might not have captured the manuscript. All authors contributed to manuscript revision, all the nuances regarding these factors that could affect time read, and approved the submitted version. to stability. Lastly, stability is considered to be a herd moving from status 1 into status 2, 2fvi, 2vx. However, for both statuses 2fvi (field- Funding virus inoculation) and 2vx (modified-live virus vaccine usage), replacement gilts and in some cases sows and weanling pigs The Morrison Swine Health Monitoring Project is a continue to be exposed to PRRS virus intentionally. A previous Swine Health Information Center (SHIC, www.swinehealth.org) study demonstrated that sites that used modified-live vaccine funded project, Project # 20-172 SHIC (MK, CV, JS, CMM, as an outbreak intervention strategy achieved stability later and CAC). The funders had no role in study design, data Frontiers in Veterinary Science 06 frontiersin.org Kikuti et al. 10.3389/fvets.2022.953918 collection and analysis, decision to publish, or preparation of organizations, or those of the publisher, the editors and the the manuscript. reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Acknowledgments The authors would like to thank MSHMP participating pig Supplementary material production systems and swine practitioners, as well as UMN and ISU Veterinary Diagnostic Laboratories for their willingness to The Supplementary Material for this article can be share information. found online at: https://www.frontiersin.org/articles/10.3389/ fvets.2022.953918/full#supplementary-material Conflict of interest SUPPLEMENTARY FIGURE 1 Weekly prevalence of status 1 farms and weekly percentage of processing fluids submissions over time. The authors declare that the research was conducted in the SUPPLEMENTARY FIGURE 2 absence of any commercial or financial relationships that could Deviance residual of the Poisson model with adjustment for long-term be construed as a potential conflict of interest. patterns with spline function. SUPPLEMENTARY TABLE 1 Lowest, median and highest weekly PRRS prevalence by year. ∗ 2009 Publisher’s note weekly prevalence comprise data from July to December only. SUPPLEMENTARY TABLE 2 All claims expressed in this article are solely those of the Contrasts of global mean of time to stability by each level of categorical authors and do not necessarily represent those of their affiliated variables. References 1. Walker PJ, Siddell SG, Lefkowitz EJ, Mushegian AR, Adriaenssens J Vet Diagnostic Investig. (2019) 31:625–8. doi: 10.1177/1040638719 EM, Dempsey DM, et al. Changes to virus taxonomy and the 852999 statutes ratified by the international committee on taxonomy of 9. Bhaskaran K, Gasparrini A, Hajat S, Smeeth L, Armstrong B. Time series viruses 2020. Arch Virol. (2020) 165:2737–48. doi: 10.1007/s00705-020- regression studies in environmental epidemiology. Int J Epidemiol. (2013) 42:1187– 04752-x 95. doi: 10.1093/ije/dyt092 2. Holtkamp D, Kliebenstein J, Neumann E, Zimmerman J, Rotto 10. Swine Health Information Center. Rapid Response Program. (2022). H, Yoder T, et al. Assessment of the economic impact of porcine Available online at: https://www.swinehealth.org/rapid-response-to-emerging- reproductive and respiratory syndrome virus on United States pork disease-program/ (accessed September 11, 2022). producers. J Swine Heal Prod. (2013) 21:72–84. doi: 10.31274/ans_air- 180814-28 11. StataCorp. Stata Statistical Software: Release 17. StataCorp (2021). 3. Perez AM, Linhares DCL, Arruda AG, VanderWaal K, 12. Morrison Swine Health Monitoring Project. PRRS Cumulative Incidence Machado G, Vilalta C, et al. Individual or common good? Beginning July 01, 2009. Morrison Swine Health Monitoring Project (2021). Voluntary data sharing to inform disease surveillance systems in Available online at: https://vetmed.umn.edu/centers-programs/swine-program/ food animals. Front Vet Sci. (2019) 6:194. doi: 10.3389/fvets.2019. outreach-leman-mshmp/mshmp/mshmp-prrs-figures 00194 13. Trevisan G, Linhares DCL, Magalhaes E, Linhares L, Crim B, Dubey P, 4. Holtkamp DJ, Polson DD, Torremorell M, Morrison B, Classen DM, Becton et al. Swine Disease Reporting System: Report #40. Swine Health Information L, et al. Terminology for classifying swine herds by porcine reproductive and Center (2021). Available online at: https://www.swinehealth.org/domestic-disease- respiratory syndrome virus status. J Swine Heal Prod. (2011) 19:44–56. Available surveillance-reports/ (accessed June 2, 2021). online at: https://www.aasv.org/shap/issues/v19n1/v19n1p44.pdf 14. Trevisan G, Jablonski E, Angulo J, Lopez WA, Linhares 5. Holtkamp D, Torremorell M, Corzo CA, Linhares DCL, Almeida MN, Yeske DCL. Use of processing fluid samples for longitudinal P, et al. Proposed modifications to porcine reproductive and respiratory syndrome monitoring of PRRS virus in herds undergoing virus elimination. virus herd classification. J Swine Heal Prod. (2021) 29:261–70. Available online at: Porcine Health Manag. (2019) 5:18. doi: 10.1186/s40813-019- https://www.aasv.org/shap/issues/v29n5/v29n5p261.html 0125-x 6. Vilalta C, Sanhueza J, Alvarez J, Murray D, Torremorell M, Corzo 15. Kikuti M, Paploski IAD, Pamornchainavakul N, Picasso-Risso C, Schwartz C, et al. Use of processing fluids and serum samples to characterize M, Yeske P, et al. Emergence of a new lineage 1c variant of porcine reproductive porcine reproductive and respiratory syndrome virus dynamics in 3 day- and respiratory syndrome virus 2 in the United States. Front Vet Sci. (2021) old pigs. Vet Microbiol. (2018) 225:149–56. doi: 10.1016/j.vetmic.2018. 8:752938. doi: 10.3389/fvets.2021.752938 09.006 16. American Association of Swine Veterinarians. UMN Case Report. American 7. Lopez WA, Angulo J, Zimmerman J, Linhares D. Porcine reproductive and Association of Swine Veterinarians (2013). Available online at: https://www.aasv. respiratory syndrome monitoring in breeding herds using processing fluids. J org/aasv%20website/Resources/Diseases/PED/PEDVBiosecurity.pdf (accessed Swine Heal Prod. (2018) 26:146–50. Available online at: https://www.aasv.org/shap/ January 20, 2007). issues/v26n3/v26n3p146.html 17. Morrison Swine Health Monitoring Project. PED EWMA analysis for 8. Vilalta C, Baker J, Sanhueza J, Murray D, Sponheim A, Alvarez years 2013–2021. Morrison Swine Health Monitoring Project (2022). Available J, et al. Effect of litter aggregation and pooling on detection of online at: https://vetmed.umn.edu/centers-programs/swine-program/outreach- porcine reproductive and respiratory virus in piglet processing fluids. leman-mshmp/mshmp/ped-charts Frontiers in Veterinary Science 07 frontiersin.org Kikuti et al. 10.3389/fvets.2022.953918 18. Morrison Swine Health Monitoring Project. PRRS prevalence of sow herd programs to control PRRSv in sow herds. Prev Vet Med. (2014) 116:111– status beginning July 01, 2009. Morrison Swine Health Monitoring Project (2021). 9. doi: 10.1016/j.prevetmed.2014.05.010 Available online at: https://vetmed.umn.edu/centers-programs/swine-program/ 20. Sanhueza J, Vilalta C, Corzo C, Arruda A. Factors affecting porcine outreach-leman-mshmp/mshmp/mshmp-prrs-figures reproductive and respiratory syndrome virus time-to-stability in breeding herds 19. Linhares DCL, Cano JP, Torremorell M, Morrison RB. Comparison in the midwestern United States. Transbound Emerg Dis. (2018) 66:823– of time to PRRSv-stability and production losses between two exposure 30. doi: 10.1111/tbed.13091 Frontiers in Veterinary Science 08 frontiersin.org