Porcine circovirus type 2–associated disease has been a factor in the health and production of pigs since it was first identified in 1991 and later described in 1997.1 It has become an important emerging infectious disease in the swine industry across the United States. Infection with PCV2 may result in many different clinical syndromes, affecting all body systems.2,3 The mechanism of action by which this pathogen causes various clinical signs and syndromes is complex, multifaceted, and poorly understood. The effects of infection with PCV2 have been described as immunosuppressive; however, immunomodulatory may be a more appropriate description because in peripheral blood, the counts of certain WBCs (lymphocytes) decrease, whereas others (neutrophils and monocytes) increase.4–6 Furthermore, results of some studies7,8 have indicated that this pathogen needs an active immune system to cause disease, whereas results of another investigation9 have not.
Porcine circovirus type 2–associated disease is further complicated by the association of the virus with other pathogens. Although PCV2 can be a primary cause of disease, it rarely acts alone. In retrospective studies10,11 of pigs with PCV2 infection in Korea and the United States, coinfection with 1 or more pathogens was detected in 85% and 98.1% of the pigs that were evaluated, respectively. Also, coinfection with other viral and bacterial pathogens severely worsened the clinical outcome of this disease syndrome.12,13 Because of the positive association of the virus with pathogens such as Salmonella spp and influenza virus, PCV2-in-fected pigs may be rendered more susceptible to these pathogens and could be at increased risk of developing coinfections from human carriers, compared with PCV2-negative swine.10,11,14 The purpose of the study reported here was to identify important infectious agents in PCV2-infected pigs, assess the effects of coinfections on serologic findings and pathologic changes in various tissues (primarily the lungs), and characterize changes in pathogenic interactions in relation to the age of pigs and type of swine production system by construction of disease models.
Materials and Methods
Swine—The pigs included in the study were obtained from a single integrated pig production system. Production flows (each a single-sow source and its associated nurseries and finishing facilities in 1-, 2-, or 3-site management systems15) were chosen on the basis of increased (recent or long-term change) nursery or finishing pig mortality rate. There were 11 farrow-to-finish (1-site management system) flows, 12 farrow-to-feeder (2-site management system) flows, and 18 far-row-to-wean (3-site management system) flows.
From these 41 production flows, 791 pigs were euthanized by use of an overdose of barbiturate administered IV and necropsied. This was completed over a period of approximately 3 months. Within each particular production flow, pigs selected for necropsy were in 1 of 4 age groups; the groups included 3-week-old (early nursery), 9-week-old (late nursery), 16-week-old (early-to-mid finishing), and 24-week-old (late finishing) pigs. For the first 3 age groups, 5 pigs were selected for necropsy from pigs within a single barn; for the 24-week-old group, 3 to 5 pigs were similarly selected.
Pigs were selected on the basis of stage of disease (whether respiratory tract or enteric or both) and were assigned a disease score from 1 to 5. Pigs assigned a disease score of 1 or 2 were well-fleshed animals in the peracute stages of the disease. Pigs assigned a disease score of 3 or 4 were animals in the acute to subacute stage of the disease process and had detectable clinical signs and lower body condition scores, compared with body condition scores of their penmates. Pigs assigned a disease score of 5 were chronically ill with considerable loss of body condition but were not located in a treatment or hospital pen.
All necropsies were completed on site and performed in sequence according to age group (youngest pigs were necropsied first and oldest pigs were necropsied last). Also, within each age group, necropsies were performed in sequence according to disease score (pigs with a disease score of 1 were necropsied first and those with a disease score of 5 were necropsied last). Separate necropsy knives were used for enteric and nonenteric tissues and were cleaned between necropsies. A lung score was assigned to each pig by use of a scoring system based on the extent of gross lung lesions.16 Lung scores were later categorized as low (0% to 10%), midrange (11% to 50%), or high (51% to 100%) levels of lung involvement to group data for modeling. Tissue samples of tonsil, hilar lymph nodes, lungs, heart, spleen, kidneys, liver, brain (3-week-old pigs only), mesenteric lymph nodes, ileum, jejunum, cecum, and colon were collected from each pig; portions of the tissues were fixed for histologic examination. Enteric and nonenteric tissues and a fresh liver sample were kept in separate bags. Furthermore, a vial of whole blood and a sample of feces were collected from each pig and submitted with the collected tissues to the Veterinary Diagnostic Laboratory at the University of Minnesota; all specimens were analyzed on an individual basis.
Diagnostic methods completed for each pig included PCR assays for PRRSV (US and European), PCV2, SIV type A, and Mycoplasma hyopneumoniae. To detect rotavirus, electron microscopic examination of fecal samples was performed. Immunohistochemistry was completed for detection of transmissible gastroenteritis virus. Histologic examination of all tissues was performed, and findings were categorized as PRRSV-associated lung lesions, M hyopneumoniae–associated lung lesions, SIV-associated lung lesions, villous inflammation or atrophy (blunting), and other. The other category included nonenteric or nonlung lesions such as endocarditis; pericarditis; polyserositis; and hepatic, splenic, CNS, and renal lesions. Bacterial culture (with identification of organisms) and antimicrobial susceptibility testing was completed on fresh tissue samples for pathogens such as Salmonella spp, Actinobacillus pleuropneumoniae, Bordetella bronchiseptica, Pasturella multocida, Haemophilus parasuis, Streptococcus suis, Staphylococcus spp, Escherichia coli, Actinobacillus suis, Arcanobacter pyogenes, Lawsonia spp, and spirochetes, among others. Serologic evaluations of each pig included assessment of serum antibodies against PRRSV, M hyopneumoniae, SIV H1N1, and SIV H3N2; all evaluations were performed by use of ELISAs, except the evaluation of anti-SIV H3N2 antibodies, which involved HI. Values for the ELISA and HI tests were entered into low, midrange, and high categories for modeling. Categories for S:P ratios included ranges of 0.00 to 0.49 (low), 0.50 to 0.99 (midrange), and ≥ 1.00 (high). Categories of HI titers included ranges of 0 to 40 (low), 80 to 160 (midrange), and ≥ 320 (high). Enteric parasite screening and identification were also performed on intestinal contents.
Diagnostic reports for 791 pigs were completed and entered into a database. From this pool of pigs, 583 pigs were selected for inclusion in this disease interaction study. Overall, 178 (30.5%) pigs originated from farrow-to-finish (1-site) flows, 175 (30.0%) originated from farrow-to-feeder (2-site) flows, and 230 (39.5%) originated from farrow-to-wean (3-site) flows. There were 157 (26.9%) pigs in the 3-week-old group, 149 (25.6%) pigs in the 9-week-old group, 152 (26.1%) pigs in the 16-week-old group, and 125 (21.4%) pigs in the 24-week-old group. The remaining 208 pigs were excluded for 1 of 2 reasons: variation in age (not exactly 3, 9, 16, or 24 weeks old; n = 83 pigs) or missing lung scores or serologic data (125 pigs). Pigs included in the final analysis had complete data with regard to all 34 disease and production variables to be evaluated. Data from pigs that yielded positive results for PCV2 via the PCR assay were compared with data from pigs that yielded negative results for PCV2 via the PCR assay to detect significant differences in coinfections, lesions, and serologic findings.
Statistical analysis—All statistical analyses were completed by use of computer software.a,b Initial screening of the variables involved bivariate analysis. Traditional 2 × 2 tables on 29 exposure variables were constructed with PCV2 as the outcome variable. From the χ2 analyses, exposure variables with values of P ≤ 0.15 were retained for model fitting. Logistic regression was used to complete model fitting by first applying backward elimination on all retained variables from the bivariate analysis. Next, a simple effects model was constructed. All variables were then tested for potential interactions. Potential confounders (age, type of production site, and disease score) were analyzed in the model. Age and type of production site were determined to be major confounders and were thus retained in the overall model. Disease complexes are known to change with age and vary among types of production systems; thus, the data were analyzed and separate models were constructed for each of these production variables. Initial screening and model fitting were completed as described for each age and type of production system. To obtain ORs and P values for the serologic and lung score categories coded as low, data for these variables were coded in reverse (0 was recoded as 2, 1 remained the same, and 2 was recoded as 0). Variables with a value of P ≤ 0.05 were retained in the models. All models were then analyzedb for fit by plotting the residuals.
Results
In the overall model, pigs that were infected with PCV2 (as determined via PCR assay) were 3.77 times as likely to be infected with M hyopneumoniae as their PCV2-negative counterparts (P < 0.001; Table 1). Infection with SIV type A was also more likely among PCV2-infected pigs (OR = 6.25; 95% CI, 2.40 to 16.21; P < 0.001). Compared with PCV2-negative pigs, PCV2-positive pigs were more than twice as likely to have S:P ratios for PRRSV ≥ 1.0 (OR = 2.12; 95% CI, 1.09 to 4.14). Also, PCV2-positive pigs were significantly (P = 0.023) less likely to have midrange S:P ratios for SIV H1N1 (0.50 to 0.99) than PCV2-negative pigs (OR = 0.47; 95% CI, 0.25 to 0.90). In this overall model, there was an interaction between SIV infection and serum anti-SIV H1N1 S:P ratios, but this term was excluded from the overall model because of lack of significance in the biologically explainable conditional CIs. Both age and production site were confounders and were thus retained in the overall model. Because age and production site were confounders, separate models were then constructed for each age group (3, 9, 16, and 24 weeks) and each type of production system (1-, 2-, and 3-site systems).
Results of logistic regression analysis (overall model) of possible associations between selected clinical and serologic variables and PCV2-positive status among pigs (3 to 24 weeks old) with (n = 363) and without (220) PCV2 infection in 1-, 2-, and 3-site production systems.
| Variable | OR | 95% CI | P value |
|---|---|---|---|
| Infection with Mycoplasma hyopneumoniae | 3.77 | 1.88–7.58 | < 0.001 |
| Infection with SIV type A | 6.25 | 2.41–16.21 | < 0.001 |
| Lung score | |||
| Low (0%–10%) | 0.46 | 0.21–1.04 | 0.063 |
| Midrange (11%–50%) | 1.92 | 1.10–3.33 | 0.021 |
| High (51%–100%) | 2.16 | 0.96–4.85 | 0.063 |
| S:P ratio for PRRSV | |||
| Low (0.00–0.49) | 0.47 | 0.24–0.92 | 0.027 |
| Midrange (0.50–0.99) | 2.03 | 0.98–4.21 | 0.055 |
| High (≥ 1.00) | 2.12 | 1.09–4.14 | 0.027 |
| S:P ratio for SIV H1N1 | |||
| Low | 1.63 | 0.79–3.38 | 0.188 |
| Midrange | 0.47 | 0.25–0.90 | 0.023 |
| High | 0.61 | 0.30–1.27 | 0.188 |
Age and type of production site were determined to be major confounders and were thus retained in the overall model.
On the basis of PCR assay results, the 3-week-old PCV2-positive pigs had more PRRSV coinfection (OR = 3.07; 95% CI, 1.25 to 7.52; P = 0.014), compared with their PCV2-negative counterparts. Compared with PCV2-negative pigs, PCV2-positive pigs were less likely to be coinfected with Salmonella organisms (OR = 0.24; 95% CI, 0.06 to 0.89; P = 0.032). In addition, there were some differences in SIV serologic findings among the 3-week-old pigs (Figure 1). The PCV2-positive pigs were approximately 11 times as likely as PCV2-negative pigs to be coinfected with SIV (OR = 10.98; 95% CI, 2.60 to 46.23; P = 0.001). Among the 3-week-old pigs, the number of PCV2-positive pigs with low S:P ratios for SIV H1N1 was significantly (P = 0.018) greater than the number of PCV2-negative pigs (OR = 6.9; 95% CI, 1.39 to 34.2); however, the number of PCV2-negative pigs with midrange or high S:P ratios for SIV H1N1 was significantly (P = 0.012 and P = 0.018, respectively) greater than the corresponding number of PCV2-negative pigs.
The 9-week-old PCV2-positive pigs had significantly more coinfection with M hyopneumoniae (OR = 4.38; 95% CI, 1.56 to 12.33; P = 0.005) or SIV (OR = 12.94; 95% CI, 2.06 to 81.11; P = 0.006), compared with their PCV2-negative counterparts (Figure 2). Compared with PCV2-negative pigs, PCV2-positive pigs were more likely to be coinfected with S suis (OR = 8.46; 95% CI, 2.28 to 31.43; P = 0.001). In this coinfection model for 9-week-old pigs, production site was a confounder and was thus retained in the model.
The PCV2-positive pigs that were 16 weeks old were 3.7 times as likely as their PCV2-negative counterparts to have other histologic lesions (95% CI, 1.19 to 11.66; P = 0.024). Production site was also a confounder in this model and was retained in the analysis. There were no significant differences in coinfection variables detected in PCV2-positive and -negative pigs that were 24 weeks old.
The data were then separated into far-row-to-finish (1-site production system), farrow-to-feeder (2-site production system), and farrow-to-wean (3-site production system) flows. In the 1-site production flows, H parasuis coinfection was significantly (P = 0.029) less likely among PCV2-positive pigs than among PCV2-negative pigs (OR = 0.35; 95% CI, 0.13 to 0.90). However, coinfection with M hyopneumoniae was 2.34 times as likely in PCV2-positive pigs as in PCV2-negative pigs (95% CI, 1.00 to 5.47; P = 0.05). Lung scores in PCV2-positive pigs were also 2.55 times as likely to be categorized as midrange (11% to 50%) level of lung involvement as were the scores in PCV2-negative pigs (95% CI, 1.19 to 5.48; P = 0.016). Furthermore, PCV2-positive pigs in these 1-site production systems were 3.4 and 5.0 times as likely to have midrange and high S:P ratios for PRRSV, respectively, as their PCV2-negative counterparts (P = 0.017 and P < 0.001, respectively). When age was included in this model, it was a significant confounder and rendered all of these variables nonsignificant (albeit marginally).
In the 2-site production flows, Salmonella organisms, E coli, and rotavirus were detected more frequently in PCV2-negative pigs than in PCV2-positive pigs (P = 0.002, P = 0.002, and P < 0.001, respectively). However, the PCV2-positive pigs were 4.16 times as likely to have other histologic lesions as the PCV2-negative pigs (95% CI, 1.47 to 11.75; P = 0.007). Lung scores were not significant.
In the 3-site production flows, PCV2-positive pigs were 79.89 times as likely to be coinfected with SIV as the PCV2-negative pigs (95% CI, 7.67 to 770.59; P < 0.001; Figure 3). Also, M hyopneumoniae coinfection in PCV2-positive pigs was 9.29 times as great as it was in PCV2-negative pigs (95% CI, 2.05 to 41.97; P = 0.004). Coinfection with A pyogenes was 14.30 times as likely to be detected in PCV2-positive pigs as in PCV2-negative pigs (95% CI, 2.64 to 77.66; P = 0.002). In addition, the PCV2-positive pigs in these 3-site production systems were 14.27 times as likely to have lung scores representing midrange level of lung involvement as PCV2-negative pigs (95% CI, 2.37 to 85.80; P = 0.004; Figure 4). Age was a significant confounder and was thus retained in this model for analysis. Infection with transmissible gastroenteritis virus was not significant in any model, and those values were not included. Results indicated that there were some pigs that were positive for transmissible gastroenteritis virus, but there was no association with PCV2.
Data regarding association of PCV2 infection with and serologic responses to SIV among 3-week-old pigs with (n = 49) and without (108) PCV2 infection. Results represent the percentage of PCV2-positive (gray bars) and -negative (black bars) 3-week-old pigs with detectable SIV and S:P ratios for SIV categorized as low (0.00 to 0.49), midrange (0.50 to 0.99), and high (≥ 1.00). Calculation of ORs in this model revealed that PCV2-infected 3-week-old pigs are 11 times as likely to be infected with SIV and approximately 7 times as likely to have low S:P ratios for SIV (indicative of low concentration of circulating maternal antibody) as their PCV2-negative counterparts.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of PCV2 infection with and serologic responses to SIV among 3-week-old pigs with (n = 49) and without (108) PCV2 infection. Results represent the percentage of PCV2-positive (gray bars) and -negative (black bars) 3-week-old pigs with detectable SIV and S:P ratios for SIV categorized as low (0.00 to 0.49), midrange (0.50 to 0.99), and high (≥ 1.00). Calculation of ORs in this model revealed that PCV2-infected 3-week-old pigs are 11 times as likely to be infected with SIV and approximately 7 times as likely to have low S:P ratios for SIV (indicative of low concentration of circulating maternal antibody) as their PCV2-negative counterparts.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of PCV2 infection with and serologic responses to SIV among 3-week-old pigs with (n = 49) and without (108) PCV2 infection. Results represent the percentage of PCV2-positive (gray bars) and -negative (black bars) 3-week-old pigs with detectable SIV and S:P ratios for SIV categorized as low (0.00 to 0.49), midrange (0.50 to 0.99), and high (≥ 1.00). Calculation of ORs in this model revealed that PCV2-infected 3-week-old pigs are 11 times as likely to be infected with SIV and approximately 7 times as likely to have low S:P ratios for SIV (indicative of low concentration of circulating maternal antibody) as their PCV2-negative counterparts.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of infection with Mycoplasma hyopneumoniae (Myco), SIV, or Streptococcus suis (SS) among 9-week-old pigs with (n = 61) and without (88) PCV2 infection. See Figure 1 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of infection with Mycoplasma hyopneumoniae (Myco), SIV, or Streptococcus suis (SS) among 9-week-old pigs with (n = 61) and without (88) PCV2 infection. See Figure 1 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of infection with Mycoplasma hyopneumoniae (Myco), SIV, or Streptococcus suis (SS) among 9-week-old pigs with (n = 61) and without (88) PCV2 infection. See Figure 1 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of infection with M hyopneumoniae (Myco), SIV, or Arcanobacter pyogenes (AP) among pigs with (n = 138) and without (92) PCV2 infection in 3-site production systems. Results indicate the percentage of PCV2-positive (gray bars) and -negative (black bars) 3-week-old pigs infected with each agent.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of infection with M hyopneumoniae (Myco), SIV, or Arcanobacter pyogenes (AP) among pigs with (n = 138) and without (92) PCV2 infection in 3-site production systems. Results indicate the percentage of PCV2-positive (gray bars) and -negative (black bars) 3-week-old pigs infected with each agent.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of infection with M hyopneumoniae (Myco), SIV, or Arcanobacter pyogenes (AP) among pigs with (n = 138) and without (92) PCV2 infection in 3-site production systems. Results indicate the percentage of PCV2-positive (gray bars) and -negative (black bars) 3-week-old pigs infected with each agent.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of lung scores (LS) among pigs with (n = 138) and without (92) PCV2 infection in 3-site production systems. Results represent the percentage of PCV2-positive (gray bars) and -negative (black bars) pigs with LS categorized as low (0% to 10%), midrange (11% to 50%), and high (51% to 100%) levels of lung involvement.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of lung scores (LS) among pigs with (n = 138) and without (92) PCV2 infection in 3-site production systems. Results represent the percentage of PCV2-positive (gray bars) and -negative (black bars) pigs with LS categorized as low (0% to 10%), midrange (11% to 50%), and high (51% to 100%) levels of lung involvement.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Data regarding association of lung scores (LS) among pigs with (n = 138) and without (92) PCV2 infection in 3-site production systems. Results represent the percentage of PCV2-positive (gray bars) and -negative (black bars) pigs with LS categorized as low (0% to 10%), midrange (11% to 50%), and high (51% to 100%) levels of lung involvement.
Citation: Journal of the American Veterinary Medical Association 230, 2; 10.2460/javma.230.2.244
Discussion
In the present study, results of the epidemiologic analysis of coinfection of PCV2 with other important pathogens in swine were consistent with findings of studies11,17 in which various concurrent infections were frequently identified in pigs with PCVAD. However, by use of logistic regression modeling and assessment of the pathogen interaction potential, we determined that PCV2-negative and -positive pigs were similarly infected with many of the pathogens previously described as important coinfectors. In the overall model, infection with M hyopneumoniae, infection with SIV, midrange lung scores, high S:P ratios for PRRSV, and midrange S:P ratios for SIV H1N1 were all significantly associated with PCV2-positive pigs. However, as hypothesized, the disease status in PCV2-positive pigs changed with age and type of production system. In current swine production systems, available clinical data and conventional wisdom also support the fact that the dynamics of disease, whether associated with enteric, respiratory, cardiovascular, nervous, or reproductive systems, differ with variations in age and type of production system.18,19 Therefore, in our study, several models were constructed for the various age groups and production system types.
Among 3-week-old pigs, the S:P ratios for SIV H1N1 were significantly lower in the PCV2-positive pigs, compared with PCV2-negative pigs. The pigs in this age group should have lingering maternal antibodies given the fact that all the sows from the study farms had received SIV vaccinations before farrowing. In addition, the 3-week-old PCV2-positive pigs were more likely to be infected with SIV than PCV2-negative pigs. Although circulating maternal antibodies against SIV are not fully protective, lower concentrations in young pigs have been associated with increased severity of disease and increased duration of shedding20; this would in turn increase the chances of recovering SIV among a group of 3-week-old pigs. Thus, our finding may have 1 of 3 explanations. First, there may be a primary immunosuppressive effect in the sows that rendered the colostrum deficient in maternal anti-SIV antibody. A second possibility is that the ability of PCV2-positive pigs to passively absorb this antibody was impaired or lacking altogether. Or, given the fact that these young pigs are capable of mounting a rapid immune response, a third possibility may be that the PCV2-positive pigs failed to respond to SIV challenge. Regardless, these pigs may act as a reservoir within a barn or production site and perpetuate disease by infecting immunocompetent pigs when the latter's circulating concentrations of maternal antibodies decrease, thereby increasing the prevalence of SIV among the late nursery and early-to-mid finishing pigs. In the present study, this became apparent in the 9-week-old pigs; in that age group, PCV2-positive pigs were approximately 13 times as likely as PCV2-negative pigs to be concurrently infected with SIV in addition to other pathogens such as M hyopneumoniae and S suis (which fulfills a classic picture of porcine respiratory disease complex). Surprisingly, PRRSV infection was significantly more likely to be present in 3-week-old PCV2-positive pigs, compared with their PCV2-negative counterparts, but this difference was not evident among the 9-week-old pigs.
Differences in concurrent infections were not apparent among the 16-week-old PCV2-positive and -negative pigs. Although none of the respiratory tract or enteric pathogens assessed were associated with the PCV2-positive pigs in this age group, compared with the PCV2-negative pigs, the PCV2-positive pigs were more likely to have other systemic histologic lesions such as endocarditis; pericarditis; polyserositis; or hepatic, splenic, CNS, or renal lesions. This finding may suggest that PCV2 causes most damage early in the life of pigs through immunomodulation and coinfection with other pathogens. However, the virus could be associated with more systemic effects in the early-to-mid finishing phase. By 24 weeks of age, there were no significant differences in pathogen coinfection, histologic lesions, lung scores, or serologic values between the PCV2-positive and -negative pigs.
When the data were analyzed by type of production system, the PCV2-positive pigs in 1-site production systems had significantly greater association of coinfection with M hyopneumoniae, higher proportion of lung scores representative of midrange level of lung involvement, and higher S:P ratios for PRRSV than PCV2-negative pigs, which are common findings in the models constructed for this analysis and consistent with other reports.11,12,21 However, contrary to findings of another investgation10 in which H parasuis was a common coinfector of PCV2-positive pigs, the pigs in 1-site production systems in the present study that were negative for PCV2 had a significantly higher amount of H parasuis coinfection than the pigs that were positive for PCV2. However, the immune response of PCV2-positive pigs is modulated towards effector cells that may destroy invasive bacteria. Therefore, if the immune response involves primarily neutrophils and monocytes, systemic migration of H parasuis would be difficult. It is important to note, however, that retention of age in the 1-site production system model rendered all aforementioned differences in infection status between PCV2-positive and -negative pigs nonsignificant.
In the 2-site production systems, there was no difference in respiratory tract pathogen infections between the PCV2-positive and -negative pigs. However, other histologic lesions were more common in the PCV2-positive pigs. This suggests that infection with PCV2 was associated with various systemic effects in the pigs that were housed in this type of system. Contrary to results of other studies14,22–24 in swine that have suggested associations between infections of enteric pathogens (or pathologic effects of those infections) and PCVAD, infections with Salmonella spp, E coli, and rotavirus were all detected in significantly greater quantities in PCV2-negative pigs housed in 2-site production systems (compared with their PCV2-positive counterparts) in the present study. As previously discussed with H parasuis, infections with bacterial agents that respond to effective innate immune responses or effector cells may be managed, whereas those requiring T-helper-1-cell responses are not. Additionally, the genetics of the pigs (eg, determinants of major histocompatibility complexes) would also likely play an important role in PCVAD. Nevertheless, calculated associations in the disease models indicate that PCV2 (with or without a yet unidentified cofactor or pathogen) is the primary influence on development of PCVAD and not genetics. The inverse association or protective effect modeled with enteric pathogens in PCV2-positive pigs needs further investigation.
The greatest differences in pathogen coinvasion and disease between PCV2-positive and -negative pigs were detected in the 3-site production systems investigated in the present study. In this type of system, M hyopneumoniae, SIV, and A pyogenes were all important coinfectors with PCV2; PCV2-positive pigs were approximately 9, 80, and 14 times as likely to be infected with those agents, respectively, as their PCV2-negative counterparts. Infection with A pyogenes is most likely an indicator of immune response failure and of chronic inflammation stimulated by alveolar macrophages responding to M hyopneumoniae and the more persistent pathogen SIV, all of which will break down normal innate immunity barriers in the lungs. Arcanobacter pyogenes was more likely a secondary or tertiary invader in these pigs. Signs of disease in PCV2-positive pigs may be exacerbated in 3-site systems because of stress and exposure factors associated with the increased amount of handling incurred during loading and transport. Stress-associated changes in heart rate and serum concentrations of cortisol and acute phase proteins have been detected during the processes of loading, unloading, and transport.25–27 Furthermore, it is known that administration of dexamethasone has significantly greater immunosuppressive effects in PCV2-positive pigs than in PCV2-negative pigs.28 On the basis of those findings, PCV2-positive pigs in 3-site production systems may become more susceptible to other respiratory tract pathogens. In the 3-site systems included in the present study, PCV2-positive pigs were 14 times as likely to have lung scores indicative of high level of lung involvement as the PCV2-negative pigs.
The results of our study indicated that PCV2-positive pigs were significantly more likely to have infections with M hyopneumoniae and SIV in the overall analysis that included all ages and types of production systems. Results also indicated that the effects of PCV2 infection may be initiated as early as 3 weeks after farrowing through, as yet uncharacterized, immunomodulation or coinfection with other pathogens. However, initiation of PCVAD syndrome probably begins much earlier. The late nursery PCV2-positive pigs were more likely than their PCV2-negative counterparts to be infected with pathogens that are typically associated with porcine respiratory disease complex. Among 16-week-old pigs, infections with other pathogens did not differ between PCV2-positive and -negative pigs, but the former were affected systemically to a greater extent. There were no significant differences in infections with other pathogens or disease-associated variables between 24-week-old PCV2-positive and -negative pigs. Our data indicated that the extent of the effects of concurrent infection with PCV2 and other pathogens was greatest among pigs in 3-site production systems; this suggests that transport-induced stress and associated exposure to pathogens may contribute to disease exacerbation in pigs processed through those multi-site production systems.
ABBREVIATIONS
| PCV2 | Porcine circovirus type 2 |
| PRRSV | Porcine reproductive and respiratory syndrome virus |
| SIV | Swine influenza virus |
| HI | Hemagglutination inhibition |
| S:P | Sample-to-positive |
| OR | Odds ratio |
| CI | Confidence interval |
| PCVAD | Porcine circovirus–associated disease |
References
- 1.↑
Clark EG. Post-weaning multisystemic wasting syndrome, in Proceedings. 28th Annu Meet Am Assoc Swine Pract 1997;499–501.
- 3.
Allan G, McNeilly F, Krakowa S, et al. PCV-2 infection in swine: more than just postweaning multisystemic wasting syndrome. Vet J 2003;166:222–223.
- 4.
Segales J, Alonso F, Rosell C, et al. Changes in peripheral blood leukocyte populations in pigs with natural postweaning multisystemic wasting syndrome (PMWS). Vet Immunol Immunopathol 2001;81:37–44.
- 5.
Nielsen J, Vincent IE, Botner A, et al. Association of lymphopenia with porcine circovirus type 2 induced postweaning multisystemic wasting syndrome (PMWS). Vet Immunol Immunopathol 2003;92:97–111.
- 6.
Segales J, Domingo M, Chianini F, et al. Immunosuppression in postweaning multisystemic wasting syndrome affected pigs. Vet Microbiol 2004;98:151–158.
- 7.
Krakowka S, Ellis JA, McNeilly F, et al. Activation of the immune system is the pivotal event in the production of wasting disease in pigs infected with porcine circovirus-2 (PCV-2). Vet Pathol 2001;38:31–42.
- 8.
Grasland B, Loizel C, Blanchard P, et al. Reproduction of PMWS in immunostimulated SPF piglets transfected with infectious cloned genomic DNA of type 2 porcine circovirus. Vet Res 2005;36:685–697.
- 9.↑
Ladekjaer-Mikkelsen AS, Nielsen J, Stadejek T, et al. Reproduction of postweaning multisystemic wasting syndrome (PMWS) in immunostimulated and non-immunostimulated 3-week-old piglets experimentally infected with porcine circovirus type 2 (PCV2). Vet Microbiol 2002;89:97–114.
- 10.↑
Kim J, Chung HK, Jung T, et al. Postweaning multisystemic wasting syndrome of pigs in Korea: prevalence, microscopic lesions and coexisting microorganisms. J Vet Med Sci 2002;64:57–62.
- 11.
Pallares FJ, Halbur PG, Opriessnig T, et al. Porcine circovirus type 2 (PCV-2) coinfections in US field cases of postweaning multisystemic wasting syndrome (PMWS). J Vet Diagn Invest 2002;14:515–519.
- 12.
Chae C. Postweaning multisystemic wasting syndrome: a review of aetiology, diagnosis and pathology. Vet J 2004;168:41–49.
- 13.
Opriessnig T, Thacker EL, Yu S, et al. Experimental reproduction of postweaning multisystemic wasting syndrome in pigs by dual infection with Mycoplasma hyopneumoniae and porcine circovirus type 2. Vet Pathol 2004;41:624–640.
- 14.
Ha Y, Jung K, Kim J, et al. Outbreak of salmonellosis in pigs with postweaning multisystemic wasting syndrome. Vet Rec 2005;156:583–584.
- 16.↑
Davies PR, Bahnson PB, Marsh WE, et al. Prevalence of gross lesions in slaughtered pigs—the PigMON database 1990–1993. Research Investment Report. Minneapolis: University of Minnesota Press, 1995.
- 17.
Ellis J, Clark E, Haines D, et al. Porcine circovirus-2 and concurrent infections in the field. Vet Microbiol 2004;98:159–163.
- 18.
Harding JCS, Clark EG, Strokappe JH, et al. Postweaning multisystemic wasting syndrome: epidemiology and clinical presentation. Swine Health Prod 1998;6:249–254.
- 19.
Sibila M, Calsamiglia M, Vidal D, et al. Dynamics of Mycoplasma hyopneumoniae infection in 12 farms with different production systems. Can J Vet Res 2004;68:12–18.
- 20.↑
Easterday BC. Immunologic considerations in swine influenza. J Am Vet Med Assoc 1972;160:645–648.
- 21.
Harms PA, Sorden SD, Halbur PG, et al. Experimental reproduction of severe disease in CD/CD pigs concurrently infected with type 2 porcine circovirus and porcine reproductive and respiratory syndrome virus. Vet Pathol 2001;38:528–539.
- 22.
Jung K, Kim J, Ha Y, et al. The effects of transplacental porcine circovirus type 2 infection on porcine epidemic diarrhoea virus-induced enteritis in preweaning piglets. Vet J 2006;171:445–450.
- 23.
Murakami S, Ogawa A, Kinoshita T, et al. Occurrence of swine salmonellosis in postweaning multisystemic wasting syndrome (PMWS) affected pigs concurrently infected with porcine reproduction and respiratory syndrome virus (PRRSV). J Vet Med Sci 2006;68:387–391.
- 24.
Kim J, Ha Y, Jung K, et al. Enteritis associated with porcine circovirus 2 in pigs. Can J Vet Res 2004;68:218–221.
- 25.
Brown SN, Knowles TG, Wilkins LJ, et al. The response of pigs to being loaded or unloaded onto commercial animal transporters using three systems. Vet J 2005;170:91–100.
- 26.
Apple JK, Kegley EB, Maxwell CV Jr, et al. Effects of dietary magnesium and short-duration transportation on stress response, postmortem muscle metabolism, and meat quality of finishing swine. J Anim Sci 2005;83:1633–1645.
- 27.
Pineiro M, Pineiro C, Carpintero R, et al. Characterisation of the pig acute phase protein response to road transport. Vet J 2006; in press.
- 28.↑
Kawashima K, Tsunemitsu H, Horino R, et al. Effects of dexamethasone on the pathogenesis of porcine circovirus type 2 infection in piglets. J Comp Pathol 2003;129:294–302.
