Abstract
Restriction fragment length polymorphism (RFLP) was first proposed to classify porcine reproductive and respiratory syndrome virus (PRRSV) in 1998. The primary objective of this study was to identify associations between different PRRSV RFLP types in swine herds in southern Ontario and clinical signs of disease in those herds. Herds included in the study submitted samples to the Animal Health Laboratory at the University of Guelph between September 2004 and August 2007. Each farm owner was surveyed to describe the clinical disease in the herd and the RFLP pattern of an isolate of PRRSV was obtained from a diagnostic sample. The most frequent isolates were RFLP types 1_4 (25.1%), 252 (14.7%), 134 (12%), and 1_2 (7.7%). The distribution of RFLP types in this study was found to be different from a previous investigation in Ontario. Those RFLP types most associated with clinical disease in the farrowing phase of production were 1_4, 1_2, and 134. The only virus type to be significantly associated with disease in the finisher phase was RFLP type 262. During the study period RFLP type 184 emerged in the population in November 2005.
Résumé
En 1998 le polymorphisme de longueur des fragments de restriction (RFLP) a été proposé pour la première fois afin de classifier le virus du syndrome reproducteur et respiratoire porcin (PRRSV). L’objectif premier de la présente étude était d’identifier les associations entre les différents types de RFLP de PRRSV dans les troupeaux porcins dans le sud de l’Ontario et les signes cliniques de maladie dans ces troupeaux. Les troupeaux inclus dans l’étude ont soumis des échantillons au Animal Health Laboratory à l’University of Guelph durant la période de septembre 2004 à août 2007. Chaque propriétaire de ferme était questionné afin qu’il décrive les signes cliniques dans le troupeau et le profil RFLP d’un isolat de PRRSV était obtenu d’un échantillon diagnostique. Les isolats les plus fréquents avaient les types de RFLP 1_4 (25,1 %), 252 (14,7 %), 134 (12 %) et 1_2 (7,7 %). La distribution des types de RFLP dans cette étude était différente d’une enquête antérieure en Ontario. Dans cette dernière, les types de RFLP les plus fréquemment associés avec des signes cliniques dans la phase naisseur de cette production étaient 1_4, 1_2 et 134. Le seul type viral à être associé de manière significative avec la maladie dans la phase de finition était le type RFLP 262. Durant la période d’étude, le type RFLP 184 a émergé dans la population en novembre 2005.
(Traduit par Docteur Serge Messier)
Introduction
Porcine reproductive and respiratory syndrome virus (PRRSV) is responsible for clinical disease throughout every stage of production in a swine farm. The disease is primarily characterized by abortions in sows of more than 100 d gestation, respiratory disease and mortality in nursing and weaned pigs, as well as respiratory disease in older pigs. Porcine reproductive and respiratory syndrome (PRRS) can result in large production losses during a herd outbreak, and without intervention, may result in endemic disease (1). Clinical presentation of PRRS in the individual animal is known to be strain dependent when different viruses are administered to pigs in a controlled environment (2–5). Viruses that are reported to be associated with different clinical disease have also been shown to be genetically different (6,7). However, specific genetic differences in PRRSV associated with pathogenicity are not entirely clear and have only recently been investigated (5,7–10). A previous observational study investigated the association between PRRSV genotype of isolates from an Illinois diagnostic laboratory and clinical disease observed in the herd (11). The authors determined that herds with more similar PRRSV isolates had similar sow mortality, but they did not find associations with other clinical signs of PRRS (11).
Restriction fragment length polymorphism (RFLP) was first proposed to classify PRRSV in 1998 (12). This method employs 3 restriction enzymes, MluI, HincII, and SacII, to cut open reading frame 5 (ORF5) of the PRRSV genome, and viruses are classified on the basis of the resulting cut pattern (12). The classification method proposed by Wesley et al (12) in 1998 was developed using a group of North American PRRSV isolates and is not optimized to differentiate viruses of European or Asian origin (12). Restriction fragment length polymorphism analysis is currently the only standardized method for classifying PRRS viruses. It allows for comparison with previously published findings without access to a library of viruses or sequencing results. Cai et al (13) described the frequency of RFLP types of PRRSV in Ontario from 1998 to 2000. The objectives of the present study were to identify associations between different PRRSV RFLP types in swine herds in southern Ontario and clinical signs of disease in those herds; to describe the distribution of RFLP types in Ontario during the study period; and to determine if the distribution of RFLP types changed over the time of the study period.
Materials and methods
Diagnostic submissions from Ontario swine herds that tested positive for PRRSV by reverse transcription polymerase chain reaction (RT-PCR) at the Animal Health Laboratory (AHL) University of Guelph (Guelph, Ontario) during the period September 1, 2004 to August 31, 2007 were eligible for inclusion in this study. The RT-PCR was performed as described (13) to amplify 433 base pairs of open reading frame 7 (ORF 7) of the PRRSV for cases before December 4, 2006. Samples from cases submitted to the AHL after December 4, 2006 were tested by the Tetracore PRRSV multiplex real time RT-PCR (Tetracore, Rockville, Maryland, USA). The AHL database was searched on one occasion for all PRRSV-positive submissions, referred to as “cases” by AHL, from September 1, 2004 to January 14, 2006. These retrospective cases were eligible only if samples from the original case had been stored by the AHL. The AHL database was searched for PRRSV-positive prospective cases twice a week starting January 15, 2006. For the purposes of this study, the following definitions are applied to be in agreement with previous literature (14,15). The term ownership refers to any premises housing pigs under a single corporate or private ownership. The term premise refers to a single contiguous land parcel with 1 or multiple buildings housing pigs; an ownership may have 1 or multiple premises. The term herd refers to a group of pigs housed at the same premise at the same point in time. The unit of interest for the study was the herd. Herds were eligible for inclusion if the premise had not been included in the study during the previous 30 d. The veterinarian listed on the case report was contacted to determine which premise had been sampled and to obtain contact information for the owner. The owners were contacted and asked if they would like to participate in the study. Permission was obtained for sequencing PRRSV-positive samples in the AHL database. Participating owners or managers were surveyed by telephone. The case and premise information was then entered into a study database in Microsoft Access (Microsoft Corporation, Redmond, Washington, USA).
The telephone survey was completed in approximately 30 minutes. Survey questions included herd demographics and perceptions of clinical signs of PRRS in the herd. The question regarding perceptions of clinical signs of disease was phrased as follows: “At the time of the sample submission to AHL, what clinical problems did you have in the barn? Please indicate with ‘Yes,’ ‘No,’ ‘Don’t know,’ or ‘N/A’ for each clinical problem. If you do not know whether a clinical problem occurred, please respond ‘Don’t know.’ If the problem is not applicable to the barn (for example, if you only have nursery pigs, then the clinical problem ‘abortion’ does not apply), please respond ‘N/A.’” The interviewer then listed each of the clinical signs presented in Table I, and allowed the interviewee to respond to each one. The survey responses were recorded on paper and entered into the study database.
Table I.
Association between clinical signs and wild-type PRRSV compared to vaccine-like PRRSV as measured by odds ratios (OR)
| Clinical signs | Proportion of herds reporting clinical signsa (%) | Wild-type PRRSV RFLP patterns |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 114 | 124 | 132 | 134 | 144 | 182 | 184 | 1_2 | 1_4 | 262 | ||
| Abortion | 139/325 (42.8) | — | — | 14.0b | 8.6b | — | — | 11.2b | 10.2b | 18.2b | — |
| Sows off-feed | 153/329 (46.5) | — | — | 13.0b | 6.3b | — | — | 9.3c | 8.1b | 7.4b | — |
| Stillborn pigs | 131/315 (41.6) | — | — | — | — | — | — | — | — | 3.3b | — |
| Weak-born pigs | 145/319 (45.5) | — | — | 5.2c | 10.9b | 6.4c | — | 23.4c | 8.3b | 8.7b | — |
| Sow/boar mortality | 83/322 (25.8) | — | — | 6.1c | — | 8.0c | — | — | — | 11.2b | — |
| Pre-weaning mortality | 148/319 (46.4) | — | — | — | 6.4b | — | — | — | 6.1b | 4.4b | — |
| Nursery respiratory | 187/309 (60.5) | — | — | — | — | — | — | — | — | — | — |
| Nursery mortality | 202/318 (63.5) | — | — | — | — | — | — | — | — | — | — |
| Finisher respiratory | 118/273 (43.2) | — | — | 0.1c | — | — | — | — | — | — | — |
| Finisher mortality | 123/274 (44.9) | — | — | 0.1c | — | — | — | — | — | — | 19.5b |
The number of herds in the denominator changes for each clinical sign depending on how many herds had the applicable stage of production.
Clinical sign associated with wild-type virus (P ≤ 0.05).
Clinical sign tends to be associated with a wild-type virus (0.05 < P ≤ 0.1).
Reverse transcription polymerase chain reaction was used to amplify ORF5 of PRRSV as previously described (12). The product was sequenced at the Guelph Molecular Supercentre, University of Guelph. From the sequencing results, computer-predicted RFLP typing was performed using the MluI, HincII, and SacII enzymes and Invitrogen Vector NTI Advance 10 software. The RFLP cut patterns were categorized using previously described nomenclature (12), and the information was added to the study database. The RFLP cut patterns are presented as 3 digits representing the results of the 3 enzyme digestions. An underscore ( _ ) is used to represent a cut pattern for the given enzyme that is not consistent with the nomenclature presented by Wesley et al (12).
Sequences were classified on the basis of their RFLP pattern as either wild-type virus or, for those with RFLP patterns 252 and 142, vaccine-like virus (16). Clinical signs of disease in herds where wild-type PRRSV was isolated were compared to clinical signs in herds where vaccine-like virus was isolated. The RFLP patterns shared by < 10 isolates in the dataset were not included in these analyses. Data were analyzed using univariable logistic regression controlling for repeated measures at the premise level by including a random intercept variable at that level. Logistic regression was completed using maximum likelihood based on Gauss Hermite quadrature. A separate model was completed for each clinical sign listed in Table I. In each model, the clinical sign was the binary outcome. Herds where the interviewee responded “Don’t know” or “N/A” for a clinical sign were not included in the model for that clinical sign. The RFLP pattern was tested as a categorical independent variable, with the vaccine-type viruses as the referent group. Pearson residuals and deviance residuals were calculated for each model to investigate outlier covariate patterns.
For each clinical sign listed in Table I, within each RFLP type, the proportion of herds reporting the clinical sign was calculated. The RFLP types were then ranked, from highest to lowest within each clinical sign, by the proportion of herds reporting the clinical sign.
The temporal distributions of all RFLP types were examined by plotting the 60-d moving average of the ratio of the number of cases in an RFLP type divided by the total number of cases in that period. The temporal scan statistic was used to assess temporal clustering within each RFLP type, using the Poisson model in SaTScan software version 7.0.3 (Kuldorff M, Harvard Medical School, Boston, Massachusetts, USA). For the temporal scan statistic, the number of cases per day of an RFLP type was used as the numerator or ‘Case File,’ and the total number of all virus types per day was used as the denominator or ‘Population File.’ Also, 999 Monte Carlo replications were used, time was aggregated into 30-day periods, and the maximum allowable cluster was set to be 50% of the study period. Data were aggregated by 30-day periods to reduce the chance of detecting false clusters of short duration. The analysis was also adjusted for the presence of a log linear trend over time, allowing the software to automatically calculate the trend.
Results
Eight hundred and fifty-nine (859) cases were PRRSV-positive by RT-PCR between September 1, 2004 and January 14, 2006; of these, 116 were included in the study. One thousand and seven (1007) cases were PRRSV-positive by RT-PCR between January 15, 2006 and August 31, 2007; of these, 326 were included in the study. Three hundred and thirty-three (333) premises, distributed across southern Ontario, were included in the study. Four hundred and forty-two (442) surveys and sequences were collected from these premises: 256 premises were sampled once, 54 premises were sampled twice, 15 premises were sampled three times, 7 premises were sampled four times, and 1 premise was sampled five times. Telephone interviews were conducted between January 2006 and December 2007. The distribution of the RFLP patterns that were found in ≥ 10 isolates, is illustrated in Table II. The minor RFLP patterns (those found in < 10 isolates) were, in decreasing order of prevalence, 212, 122, 152, 112, 1_3, 162, 222, 1_1, 1__, 2_2, 113, 141, 164, 181, 183, 214, 251, and 264. The list of clinical signs, the proportion of herds where each clinical sign of PRRS was reported, and the associations between clinical signs and isolated RFLP types are illustrated in Table I. The Pearson and deviance residuals for each model were within the range of -2 and 2; because of this, the effects of individual observations on the models were not further investigated.
Table II.
The distribution of PRRSV with specific restriction fragment length polymorphism (RFLP) patterns in Ontario 2004–2007
| RFLP | Number of sequences identified | Proportion of all sequences (%) |
|---|---|---|
| 1_4 | 111 | 25.12 |
| 252a | 65 | 14.71 |
| 134 | 53 | 11.99 |
| 1_2 | 34 | 7.69 |
| 132 | 24 | 5.43 |
| 142b | 22 | 4.98 |
| 182 | 21 | 4.75 |
| 184 | 16 | 3.62 |
| 144 | 15 | 3.39 |
| 262 | 12 | 2.71 |
| 124 | 11 | 2.49 |
| 114 | 10 | 2.26 |
| Others | 48 | 10.86 |
| Total | 442 | 100 |
Boehringer-Ingelheim Ingelvac® PRRS MLV vaccine strain.
Boehringer-Ingelheim Ingelvac® PRRS ATP vaccine strain.
The percentage of herds reporting clinical signs within each RFLP pattern and the ranks of each RFLP pattern by clinical signs are illustrated in Table III. The frequencies of the 3 most common RFLP types (1_4, 134 and 1_2) compared to the total number of cases is illustrated in Figure 1. Temporal clustering was identified in each of the RFLP patterns tested. In RFLP pattern 1_4, a cluster was identified between September 2005 and July 2007 with a relative risk of 6.1 (P = 0.001). In RFLP pattern 134, a cluster was identified between September 2005 and August 2007 with a relative risk of 7.2 (P = 0.001). In RFLP pattern 1_2 a cluster was identified between September 2006 and April 2007 with a relative risk of 6.2 (P = 0.001). The first case of RFLP type 184 was identified in Nov 2005. This emergence and the temporal distribution of RFLP type 184 is illustrated in Figure 1.
Table III.
The percentage of herds reporting each clinical sign for each RFLP pattern and the rank of the RFLP pattern for frequency of the clinical sign is presented in brackets
| Clinical signs | Vaccine types | 114 | 124 | 132 | 134 | 144 | 182 | 184 | 1_2 | 1_4 | 262 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Abortion | 18 (9) | 33 (8) | 17 (10) | 67 (3) | 61 (4) | 36 (6) | 36 (6) | 75 (1) | 58 (5) | 71 (2) | 11 (11) |
| Sows off-feed | 25 (11) | 56 (6) | 33 (9) | 67 (2) | 58 (4) | 45 (7) | 36 (8) | 75 (1) | 58 (4) | 63 (3) | 33 (9) |
| Stillborn pigs | 40 (7) | 44 (5) | 17 (11) | 61 (1) | 58 (3) | 36 (8) | 27 (10) | 50 (4) | 42 (6) | 61 (1) | 33 (9) |
| Weak-born pigs | 30 (11) | 44 (8) | 33 (9) | 61 (4) | 67 (2) | 55 (6) | 45 (7) | 75 (1) | 56 (5) | 65 (3) | 33 (9) |
| Sow/boar mortality | 14 (10) | 22 (8) | 33 (5) | 44 (3) | 31 (6) | 36 (4) | 22 (8) | 67 (1) | 23 (7) | 47 (2) | 13 (11) |
| Pre-weaning mortality | 40 (7) | 33 (8) | 33 (8) | 56 (4) | 67 (1) | 45 (6) | 55 (5) | 25 (11) | 58 (3) | 60 (2) | 33 (8) |
| Nursery respiratory | 67 (6) | 67 (6) | 70 (5) | 64 (9) | 72 (4) | 67 (6) | 89 (2) | 100 (1) | 57 (11) | 63 (10) | 75 (3) |
| Nursery mortality | 66 (9) | 78 (5) | 70 (8) | 55 (11) | 79 (4) | 80 (3) | 89 (2) | 100 (1) | 71 (7) | 73 (6) | 63 (10) |
| Finisher respiratory | 59 (5) | 33 (10) | 50 (6) | 33 (10) | 37 (9) | 75 (2) | 60 (4) | 75 (2) | 42 (8) | 44 (7) | 78 (1) |
| Finisher mortality | 51 (5) | 17 (11) | 50 (6) | 27 (10) | 47 (8) | 63 (3) | 58 (4) | 75 (2) | 50 (6) | 45 (9) | 89 (1) |
Figure 1.
The temporal distribution of the ratio of the number of cases of the 3 most common types (1_4, 1_2, and 134) and RFLP type 184 over the total number of cases using a 60-day moving average.
Discussion
In this study, RFLP type 1_4 was the most frequently identified type and was associated with clinical signs specifically in the sow herd (Table I). This type also ranked high in clinical signs occurring in the farrowing phase of production (Table III). Type 1_4 was not reported in a 1998–2000 Ontario study (13), but was found in approximately 21% of PRRSV isolates in a 1998–2002 Quebec study (16). Because it was not previously reported in Ontario and is the most frequently isolated type in the current study, an outbreak of this virus type was suspected. An outbreak is defined as 2 or more occurrences of an RFLP type that are epidemiologically linked (17). In the current investigation, proximity in time of similar PRRSV isolates was used to suggest the possibility of epidemiological linkage and therefore an outbreak. In the current study, RFLP type 1_4 did not appear to change in relative frequency across the study period when compared with all virus types identified. The temporal scan statistic did identify a large cluster which included approximately 50% of the study period beginning in September 2005. This cluster was not interpreted as a discrete outbreak of the 1_4 RFLP type because in the 1-year period preceding the cluster, 20.7% of cases were RFLP type 1_4 versus 25.8% during the cluster. It was concluded that this small difference may have indicated an increase over time that was part of a larger outbreak not captured by the data. Because this RFLP type was not found in the 1998–2000 study it is hypothesized that it emerged between 2000 and September 2004 (13). The frequency of all cases of PRRS in the study database was used as a denominator for the temporal scan statistic for each RFLP type. This may have resulted in a lower sensitivity of outbreak detection than if a denominator from the primary population had been used. In the current approach, if there were concurrent outbreaks of several PRRS genotypes, one could mask the other. A recent extension of the RFLP classification system described by Wesley et al (12) is being widely used by North American diagnostic laboratories (Russow K, University of Minnesota, personal communication, 2009). This system would further classify cut patterns of the HINCII enzyme which would create a subgroup within RFLP type 1_4. Further classification of the RFLP type 1_4 might change the results of the current investigation.
In the present study, type 1_2 was associated with the clinical signs of abortion, anorexia, weak-born pigs, and high pre-weaning mortality in the sow herd when compared to herds with vaccine-type virus. Type 1_2 was rarely reported in the Quebec study (16) and was not reported in a previous Ontario study (13). However, type 1_2 was ranked 3rd among wild type viruses only in pre-weaning mortality and was otherwise ranked in the middle range for other clinical signs. This low ranking may indicate that virus type 1_2 is not as clinically important as suggested by the statistical test against the vaccine group. It may be that this virus is associated with increased odds of clinical signs when compared to the vaccine group, but is not different from other wild-type viruses. The temporal distribution of RFLP type 1_2 did not indicate a clear outbreak pattern. The relative frequency of this type did not change greatly over the study period. Type 1_2 possibly emerged recently in Ontario, as it was not found in the 1998–2000 Ontario study (13) but did emerge before the period of the present study. Type 1_2 would be further subclassified by the recent changes to the standard RFLP classification system, and this alternate nomenclature might change the results of the current investigation (Russow K, University of Minnesota, personal communication, 2009).
Virus RFLP type 134 was identified very infrequently in the 1998–2002 Quebec study (16) and represented only 4% of virus isolates in the 1998–2000 Ontario study (13). In contrast, type 134 represented 12% of all viruses in the current study, and steadily increased in relative frequency over the study period. Between January and July 2005, 7.8% of cases were RFLP type 134, while 16.0% of cases were RFLP type 134 between January and July 2007. This increase was interpreted as an outbreak of type 134 because of the consistency of the pattern over time and the near doubling of the relative frequency over the study period.
Type 124 was the second most frequent RFLP type in the 1998–2000 Ontario study, representing 9.4% of isolates (13), and also represented 9% of PRRS viruses in the Quebec study (16). In contrast, type 124 represented only 2.5% of isolates in the present study.
Type 132 which was associated with clinical disease in the farrowing phase of production, in the current study, was rarely isolated in the Quebec study (16) but represented 6.3% in the previous Ontario study (13). In a study of 35 Japanese isolates (1991–1999), 54% were RFLP pattern 132, higher than in the present study, the Quebec study, and the previous Ontario study (18). However, RFLP type 132 was also the only virus type, in the current study, to tend to be associated with absence of clinical disease in any phase of production. Herds with this virus type tended to report less disease in the finisher phase of production than those with the vaccine types.
Type 184 was the most frequent RFLP type in the Quebec study, representing 28% of isolates (16), but was not isolated in the 1998–2000 Ontario study (13), and represented only 3.6% of viruses in the present study. Type 184 was associated with higher odds of abortion than in herds with vaccine-like virus, but was not significantly associated, at P < 0.05, with other clinical signs. Type 184 ranked 1st or 2nd among wild type viruses in all clinical signs except stillborn piglets and pre-weaning mortality. The discrepancy between the ranking and relative odds may be due to the small number of herds with this RFLP type, resulting in insufficient power to detect statistically significant differences. For example, of the 16 herds with RFLP type 184, only 8 had sows on site, 7 had nursery pigs, and 5 had finisher pigs. Only questions pertaining to the phases of production on-site would have been answered, resulting in only a small number of responses to questions concerning clinical signs of PRRS in the herd. Type 184 ranked higher for most clinical signs than did other wild type viruses, suggesting that type 184 may be clinically more important than can be shown statistically. Type 184 also appeared to emerge in the population during November 2005, which may have represented an outbreak of this virus type. However, cases from before September 2004 would have to be investigated to confirm that type 184 was not present between the end of the previous study, ending in 2000, and the beginning of our study (13).
Viruses with RFLP type 262 were not identified in the Quebec study (16) but were identified in 3% of isolates in the previous Ontario study (13). Type 262 was both highly ranked in clinical signs in the finisher phase and was associated with finisher pig mortality compared to herds with vaccine-like virus. Type 262 was the only RFLP type found to be associated with clinical signs in the finisher phase.
In a previous analytic observational study, differences in sow mortality was the only clinical sign of PRRS that was found to be associated with genetic differences (11). In the present study, sow and boar mortality was measured and was the least frequently reported clinical problem. Sow and boar mortality was also only significantly more frequently reported in those herds with type 1_4 than those with the vaccine type. Results of these studies are not directly comparable due to the following reasons. First, the measurements of clinical signs in the previous and present study were both done by interview and were both collected as binary variables. These strategies of data collection are crude and difficult to repeat. Second, the research question and analytical approach differed between the 2 studies. While the previous study investigated associations between differences, our study investigated associations between discrete outcomes and discrete genotypes and therefore comparisons between these studies are not interpretable.
There are several limitations to the present study. The use of a secondary-based study design limited the inclusion of cases that submitted to the participating laboratory. The laboratory was chosen because it receives the largest number of samples for PRRSV PCR-testing in Ontario, and was willing to allow us to access their submission database. A second limitation was the way in which data was collected about clinical disease in the herds. The questions were open to interpretation by the person being interviewed based on what they perceived to be a problem or not. This format was chosen in order to make it possible to conduct the survey over the telephone without the need for collection of production records of varied quality. Third, the use of the RFLP classification system designed by Wesley et al (12) limits the interpretation of the results to this virus classification system. This system was chosen because it is the most common classification system and was being used by the participating laboratory to report PRRSV ORF 5 sequencing results to Ontario veterinarians.
In conclusion, the distribution of PRRSV RFLP types in Ontario has changed since the report from 1998–2000 (13) and showed some similarity to the 1998–2002 Quebec study (16). Furthermore, observed clinical signs in our study were associated with specific RFLP types. Epidemic patterns were observed for two RFLP types, 134 and 184 but could not be ruled out for any of the RFLP types. Data from before the study would have been required to determine whether these virus types had emerged recently.
Acknowledgments
This research was made possible through the financial support of Ontario Pork and the Canada-Ontario Research and Development (CORD) Program. The participation of veterinarians and producers was greatly appreciated as this project required significant industry cooperation. The dedicated cooperation of the Animal Health Laboratory was essential for the completion of PRRS virus identification and the genotyping of the viruses.
References
- 1.Zimmerman J, Benfield DA, Murtaugh MP, et al. Diseases of Swine. 9th ed. Ames, Iowa: Blackwell Publ; 2006. Porcine reproductive and respiratory syndrome virus (Porcine Arterivirus) pp. 387–417. [Google Scholar]
- 2.Halbur PG, Paul PS, Meng XJ, et al. Comparative pathogenicity of nine US porcine reproductive and respiratory syndrome virus (PRRSV) isolates in a five-week-old cesarean-derived, colostrum-deprived pig model. J Vet Diagn Invest. 1996;8:11–20. doi: 10.1177/104063879600800103. [DOI] [PubMed] [Google Scholar]
- 3.Mengeling WL, Vorwald AC, Lager KM, et al. Comparison among strains of porcine reproductive and respiratory syndrome virus for their ability to cause reproductive failure. Am J Vet Res. 1996;57:834–839. [PubMed] [Google Scholar]
- 4.Halbur PG, Paul PS, Frey ML, et al. Comparison of the pathogenicity of two US porcine reproductive and respiratory syndrome virus isolates with that of the Lelystad virus. Vet Pathol. 1995;32:648–660. doi: 10.1177/030098589503200606. [DOI] [PubMed] [Google Scholar]
- 5.Cheon DS, Chae C. Comparison of the pathogenicity of two strains (wild type and vaccine-like) of porcine reproductive and respiratory syndrome virus (PRRSV) in experimentally infected sows. J Comp Pathol. 2004;130:105–111. doi: 10.1016/j.jcpa.2003.09.001. [DOI] [PubMed] [Google Scholar]
- 6.Meng XJ, Paul PS, Halbur PG, et al. Sequence comparison of open reading frames 2 to 5 of low and high virulence United States isolates of porcine reproductive and respiratory syndrome virus. J Gen Virol. 1995;76 (( Pt 12)):3181–3188. doi: 10.1099/0022-1317-76-12-3181. [DOI] [PubMed] [Google Scholar]
- 7.Li Y, Wang X, Jiang P, et al. Genetic analysis of two porcine reproductive and respiratory syndrome viruses with different virulence isolated in China. Arch Virol. 2008;153:1877–1884. doi: 10.1007/s00705-008-0207-6. [DOI] [PubMed] [Google Scholar]
- 8.Zhou L, Zhang J, Zeng J, et al. The 30 amino acid deletion in the Nsp2 of highly pathogenic porcine reproductive and respiratory syndrome virus emerging in China is not related to its virulence. J Virol. 2009;83:5156–5167. doi: 10.1128/JVI.02678-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kim DY, Kaiser TJ, Horlen K, et al. Insertion and deletion in a non-essential region of the nonstructural protein 2 (nsp2) of porcine reproductive and respiratory syndrome (PRRS) virus: Effects on virulence and immunogenicity. Virus Genes. 2009;38:118–128. doi: 10.1007/s11262-008-0303-4. [DOI] [PubMed] [Google Scholar]
- 10.Pei Y, Hodgins DC, Lee C, et al. Functional mapping of the porcine reproductive and respiratory syndrome virus capsid protein nuclear localization signal and its pathogenic association. Virus Res. 2008;135:107–114. doi: 10.1016/j.virusres.2008.02.012. [DOI] [PubMed] [Google Scholar]
- 11.Goldberg TL, Weigel RM, Hahn EC, et al. Associations between genetics, farm characteristics and clinical disease in field outbreaks of porcine reproductive and respiratory syndrome virus. Prev Vet Med. 2000;43:293–302. doi: 10.1016/S0167-5877(99)00104-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wesley RD, Mengeling WL, Lager KM, et al. Differentiation of a porcine reproductive and respiratory syndrome virus vaccine strain from North American field strains by restriction fragment length polymorphism analysis of ORF 5. J Vet Diagn Invest. 1998;10:140–144. doi: 10.1177/104063879801000204. [DOI] [PubMed] [Google Scholar]
- 13.Cai HY, Alexander H, Carman S, et al. Restriction fragment length polymorphism of porcine reproductive and respiratory syndrome viruses recovered from Ontario farms, 1998–2000. J Vet Diagn Invest. 2002;14:343–347. doi: 10.1177/104063870201400415. [DOI] [PubMed] [Google Scholar]
- 14.Cameron AR, Baldock FC. Two-stage sampling in surveys to substantiate freedom from disease. Prev Vet Med. 1998;34:19–30. doi: 10.1016/s0167-5877(97)00073-1. [DOI] [PubMed] [Google Scholar]
- 15.Poljak Z, Dewey CE, Martin SW, et al. Prevalence of and risk factors for influenza in southern Ontario swine herds in 2001 and 2003. Can J Vet Res. 2008;72:7–17. [PMC free article] [PubMed] [Google Scholar]
- 16.Larochelle R, D’Allaire S, Magar R. Molecular epidemiology of porcine reproductive and respiratory syndrome virus (PRRSV) in Quebec. Virus Res. 2003;96:3–14. doi: 10.1016/s0168-1702(03)00168-0. [DOI] [PubMed] [Google Scholar]
- 17.Farrington P, Andrews N. Monitoring the Health of Populations: Statistical Principles and Methods for Public Health Surveillance. Oxford Univer PrUS; 2004. Outbreak detection: Application to infectious disease surveillance; pp. 203–231. [Google Scholar]
- 18.Itou T, Tazoe M, Nakane T, et al. Analysis of open reading frame 5 in Japanese porcine reproductive and respiratory syndrome virus isolates by restriction fragment length polymorphism. J Vet Med Sci. 2001;63:1203–1207. doi: 10.1292/jvms.63.1203. [DOI] [PubMed] [Google Scholar]