Original research

November and December, 1999

Profiling Mycoplasma hyopneumoniae in farms using serology and a nested PCR technique

Maria Calsamiglia, DVM, PhD; Carlos Pijoan, DVM, PhD; Gary J. Bosch, DVM

Calsamiglia M, Pijoan C, Bosch GJ. Profiling Mycoplasma hyopneumoniae in farms using serology and a nested PCR technique. Swine Health Prod. 1999;7(6):263-268. Also available as a PDF (156k).

MC, CP: Clinical and Population Sciences, College of Veterinary Medicine, 385 AnSci/VetMed, 1988 Fitch Avenue, St. Paul, Minnesota 55108; email: pijoa001@tc.umn.edu. GB: Pharmacia & Upjohn Animal Health

Also online: "Critical Review" by Eileen L. Thacker


Objective: To profile herds experiencing Mycoplasma hyopneumoniae pneumonia using both a nested polymerase chain reaction (PCR) technique and serology and to compare both techniques.

Methods: Five commercial farms with a history of mycoplasmal pneumonia participated in the study. Samples were taken from different age groups during a single visit, starting at 5, 6, 7, or 8 weeks of age and up to nearly marketing age. Ten animals per age group (five different age groups per farm) were randomly selected and nasal swabs and blood samples taken. DNA from nasal swabs was extracted and the nested PCR to detect M. hyopneumoniae infected animals was performed. Sera were tested for antibodies to M. hyopneumoniae. The estimated antigen and antibody prevalence per age group was calculated, and plotted with their corresponding confidence intervals. Differences between both tests were measured using c2 analysis.

Results: Serum and bacterium profiles for each farm were obtained. The nested PCR showed a higher proportion of infected animals in the early spread of the microorganism than serology, and provided more accurate information on the dynamics of infection.

Implications: The nested PCR detects M. hyopneumoniae earlier then serology. Moreover, detecting M. hyopneumoniae from nasal swabs offers new and valuable information in helping interpret serum profiles and figuring out when animals get infected and how and where M. hyopneumoniae is being transmitted. This information will potentially allow for better timing in the application of medication, vaccination, or management strategies.


Keywords: swine, M. hyopneumoniae, nested PCR, serum profiles, bacterium profiles.

Received: February 3, 1999
Accepted: July 27, 1999

Mycoplasma hyopneumoniae is one pathogen involved in enzootic pneumonia.1 Although control of the disease is possible through vaccination,2 medication,2 or management strategies,3,4 timing of these treatments is critical for their effectiveness.

Until recently, serological tests were the only commonly available diagnostic tools to detect exposure to microorganism in live animals and to obtain a better understanding of the dynamics of the disease (Sitjar M, et al. Proc IPVS Cong, 1994;p133). However, these serological tests have some limitations and are difficult to interpret. Serology detects the onset of seroconversion, not the onset of infection. Time to seroconversion after exposure to M. hyopneumoniae varies. In several experimental challenge studies, antibodies were detected at 2 weeks postinfection by ELISA in the first animals that seroconverted.5-7 However, not all animals seroconverted at the same time, and it was not until 4-5 weeks after challenge that all animals had detectable serum antibodies. In contact-exposed piglets, antibodies are usually detected later. In a study where contact pigs were exposed to intranasally infected pigs, pigs with antibodies were initially seen 5 weeks after contact, and it was not until 9 weeks after contact that all animals had seroconverted.8 In field infections, time of seroconversion appears also to be delayed. In a study where antibodies against M. hyopneumoniae from several groups were serially measured and the thorax of each pig radiographed, it was concluded that the peak severity of lung lesions was associated with seroconversion, although the latter was delayed approximately 6-8 weeks.9

At present, the ELISA is considered the most useful test for serology. It can detect all classes of immunoglobulins, gives quantitatively measurable results, and is very sensitive. When first developed,10 the ELISA contained protein antigens from sodium dodecyl sulfate (SDS)-solubilized M. hyopneumoniae cells, which resulted in nonspecific reactions.11 The specificity of the ELISA has been improved12-14 by using the neutral detergent Tween 20 for extraction of membrane proteins from M. hyopneumoniae whole cells, thereby reducing the content of high (> 90 kD) and low (<31 kD) molecular weight proteins.15 It has been shown14 that cross reactions with Mycoplasma flocculare are minimal, especially in natural infections, although they still exist. Therefore, the results of the Tween 20 ELISA must be interpreted with caution, especially when the optical densities are between 0.200 and 0.260. These readings can be observed in early or late infections, but could also be due to cross reactions with M. flocculare.

Recently, a nested PCR that is able to detect M. hyopneumoniae from nasal swabs has been described.16 Other one-step PCRs for M.hyopneumoniae have been described,17-20 but they are apparently not able to consistently detect the microorganism from nasal swabs. The primers have been tested with most mycoplasmas and acholeplasmas described in swine (Mycoplasma hyorhinis, Mycoplasma flocculare, Mycoplasma hyosynoviae, Mycoplasma hyopharyngis Mycoplasma buccale, Actinobacillus axanthum, Actinobacillus granularum, Actinobacillus laidlawii) as well as several avian mycoplasmas. The primers of the PCR for M. hyopneumoniae did not amplify DNA from any of these microorganisms.

The primers have been also tested in live animals, in a challenge study using M. hyopneumoniae-free pigs. Before challenge, all animals were swabbed and shown to be mycoplasma negative by PCR.16 Although PCR is a very promising tool to profile herds to determine the carrier state of incoming gilts and establish the health status of different farms from which pigs are going to be commingled, the significance of obtaining a positive result from a nasal swab is still unknown.

The objective of this study was to profile five herds experiencing respiratory disease related with mycoplasma using the nested PCR technique and serology.

Materials and methods


Five farms with an ongoing M. hyopneumoniae disease problem, as diagnosed by the veterinarian of the farm, were selected for the study:

  • Farm A was a three-site production system, with all-in-all-out (AIAO) management by room in the finishers. The sow herd was negative to M. hyopneumoniae according to serology and history. The mycoplasma outbreak was seen in the finishing barns. No vaccination nor medication for M. hyopneumoniae was being used.
  • Farm B was a 1200-sow herd, two-site production system, with wean-to-finish buildings managed AIAO by building. The pigs started coughing at around 16 weeks of age, and continued until they went to market. A vaccination protocol with M+Pac(TM) M. hyopneumoniae bacterin (Schering Plough; Madison, New Jersey) was being used. A 1-mL dose was administered subcutaneously at 5 and 7 weeks of age.
  • Farm C was a three-site production system with piglets commingled from ten different sow herds, each of them containing 2000 head. The nurseries and finishers were managed AIAO by site. Weaning age was 19 days on average. The mycoplasma-related cough was seen 50 days after entering the finishers (around 16 weeks of age). No vaccines against M. hyopneumoniae were being used.
  • Farm D was a 6000-sow herd, three-site production system that was managed AIAO by building and that weaned at 16-17 days of age. Respiratory signs were seen at around 18 weeks of age. A vaccination protocol with Suvaxyn Respifend(TM) (Fort Dodge; Fort Dodge, Iowa) was being used: the first 2-mL dose was administered intramuscularly (IM) at 7 weeks of age and the second one at 9 weeks.
  • Farm E was a three-site production system. Weaned pigs were commingled in the nurseries from 10 different sow herds, each of them with 2500 head. Nurseries and finishers were managed AIAO by site. Slight respiratory clinical signs, affecting approximately 10%-20% of the pigs, were seen by the end of the finishing stage. No vaccination or medication protocols were being used.


Samples were taken from different age groups during a single visit, starting with nursery-aged pigs (approximately 6 weeks of age) and then sampling at 4-week age intervals(at 6, 10, 14, 18, and 22 weeks of age). Ten animals per age group were restrained with a snare and nasal swabs and blood samples were taken. Nasal samples were taken using swabs with transport media (Culturette(TM), Becton Dickinson). The swab was introduced into the nostrils, reaching deeply into the turbinates. Nasal swabs were kept moistened with the transport media and sent overnight on ice to the laboratory, where they were further processed.

Culture was not attempted because it lacks sensitivity in nasal swab samples17 and is more time consuming than the nested PCR. Blood samples were used for serology and nasal swabs for detecting the microorganism with the nested PCR. DNA extracted from a pure culture of M. hyopneumoniae was used as positive control and double-distilled water submitted to the same extraction protocol as negative control.


Serology samples were sent to Iowa State University Diagnostic Laboratory, where they were assayed with the Tween 20 ELISA, as previously described.16 This test does not differentiate vaccinated from infected animals.

DNA extraction

Nasal swabs were resuspended in 350 mL of sterile PBS. DNA was extracted as previously described.18 Briefly, suspensions were boiled for 5 minutes, DNA was extracted with phenol:chloroform, and precipitated with ethanol and sodium acetate (pH 5.2, final concentration of 0.5M). DNA was resuspended in 40 mL of double-distilled water.

Nested PCR

Polymerase chain reaction was performed using primers and conditions previously described.18 Two sets of primers from the 16S ribosomal gene were used. The outer primer pair consisted of:

  • a forward primer: 5&#180- GAG CCT TCA AGC TTC ACC AAG A- 3&#180(nucleotide positions 212-233), and
  • a reverse primer: 5´- TGT GTT AGT GAC TTT TGC CAC C- 3´ (nucleotide positions 839-860).

The inner primer pair consisted of:

  • a forward primer: 5'&#180- ACT AGA TAG GAA ATG CTC TAG T-3´ (nucleotide positions 463-484), and
  • a reverse primer: 5&#180- GTG GAC TAC CAG GGT ATC T-3´ (nucleotide positions 797-815).

Five µL of the DNA preparation were used as PCR templates. Five µL of the DNA preparation were used as PCR templates in the first reaction and 0.5 mL of the product were used for the second reaction. The amplifications were performed in a 25-mL reaction mixture containing 0.2 mM concentration of each primer, 20 pmol of each nucleotide (Boehringer Mannheim; Germany), 1xPCR- buffer (Boehringer Mannheim), 5% glycerol, 3mM MgCl2 (Boehringer Mannheim), and 1U of TaqDNA polymerase (Boehringer Mannheim). Both reactions where performed in a thermocycler (GeneAmp PCR system 2400, Perkin-Elmer; Branchburg, New Jersey) and required the same conditions: 30 cycles, denaturation at 94 degrees C for 30 seconds, annealing at 60 degrees C for 45 seconds, and extension at 72 degrees C for 30 seconds.

Seven-µL aliquots of the amplified samples were analyzed by electrophoresis in a 1% agarose gel with 0.5 mg per mL of ethidium bromide, and then visualized and photographed (Eagle Eye, Stratagene; La Jolla, California).

Analysis of results

Individual samples were tested separately, but because animals were not identified, the serology and nested PCR results of each individual animal could not be determined.

c2 analysis was used to compare the proportion of pigs that were positive to the PCR test versus the serologic test in each farm, at each sampling period. The data on the individual farms were also collapsed and the overall differences between both tests evaluated. The latter was done by comparing the tests over time using a stratified analysis, Mantel- Haenszel c2, as well as at each specific sampling period with a c2 analysis. Odds ratios were calculated using the Mantel-Haenszel procedure.


Farm A

The nurseries appeared to have a low prevalence of M. hyopneumoniae, both by PCR and serology (Figure 1). There were no statistical differences detected in the proportion of positive pigs between the two tests until the 14-week age group. When pigs entered the finishers they rapidly became infected: the 14-week-old group already showed 90% of positive animals by PCR, and a high percentage of positive animals was also observed in the 18-week-old group. Seroconversion was not detected until much later, in the 22- to 24- week-old group, where a high proportion of PCR- and serology-positive animals were observed.

Farm B

Piglets entered the wean-to-finish barn with only a few animals positive to M. hyopneumoniae by the nested PCR (Figure 2). The number of PCR-positive pigs steadily increased with age, until a critical mass of infected pigs was reached and mycoplasma-related coughing appeared. The antibodies detected until 14 weeks of age were probably due to vaccination, and seroconversion due to infection was not detected until the 18-week-old group.

Farm C

Nursery pigs seemed to have a low prevalence of M. hyopneumoniae, which could only be detected by PCR (Figure 3). An increasing proportion of PCR-positive animals were detected over time, reaching a peak at 16 weeks of age, which coincided with the onset of cough. In the finishers, the proportion of positive animals detected by PCR was greater (P<.05) than the one obtained with serology, until the 20-week-old group. In the latter group seroconversion was detected for the first time, and the proportion of serology-positive animals was greater than the PCR positive animals (P<.05) in the 24-week-old group.

Farm D

Pigs in the nurseries had relatively large numbers of positive animals, which kept increasing in the finisher, where mycoplasma-related coughing appeared (Figure 4). The antibodies detected at 10 weeks of age were probably due to vaccination. Antibodies due to M.hyopneumoniae infection appeared at 18 weeks of age.

Farm E

A few nursery pigs were positive to M. hyopneumoniae PCR (Figure 5). When the pigs entered the finishers, the proportion of PCR-positive pigs remained at relatively low numbers. From 18-24 weeks of age, an abrupt increase in the detection of M. hyopneumoniae in nasal swabs occurred, which coincided with the appearance of mycoplasma-related cough. Increasing antibody titers were observed at 18 and 24 weeks of age. Although PCR detected equal or greater proportion of positive pigs than serology in all sampled groups, significant differences between serology and PCR where only observed in the 24-week-old group.

Across all five herds, the nested PCR was able to detect more positive animals than serology in the 14- to 16-week-old and 18- to 20-week-old groups (Figure 6). A larger proportion of PCR-positive versus serology-positive animals was also detected in the first two age groups in the finishers (P<.05). The chance of finding a positive animal by PCR was 27.5 times higher than using serology in the 14- to 16-week age group, and five times higher in the 18-20 week age group. In the 14- to 16-week-old group, the odds ratio of the PCR-positive versus serology-positive was 27.85, and in the 18- to 20-week-old group the odds ratio was 5.17. In the oldest age group, no differences were detected between serology and PCR (P>.05). Overall, there was a 3.44 times greater chance of finding a positive animal with the nested PCR technique than with serology.


This study was conducted in order to compare the nested PCR technique with serology in the profiling of herds. Traditionally, serum profiles have used 10-15 animals of different age groups.9,22,23 Although these sample sizes are not sufficient to make an accurate estimate of the prevalence of positive animals in each age group, a rough estimate of the dynamics of infection can be drawn. The sample size required to estimate a prevalence varies with the expected prevalence, the level of confidence, and the desired accuracy. If either low or high prevalences are expected, 25 animals suffice to obtain an estimate with a 90% confidence interval and an accuracy of 10%. When the expected prevalence approaches 50%, larger samples should be taken: approximately 70 animals. Because of this, traditional seroprofiles, as well as the PCR profiles performed in this study, should not be used to estimate prevalence. They are, however, very useful for obtaining a picture of the farm's disease dynamics that can be used to establish control strategies.

This study showed that the nested PCR is a valuable technique in profiling farms with M. hyopneumoniae. It was found that the proportion of the pigs testing positive via PCR was higher than those found by serology, specially during the early phase of the spread of the organism. However, no statistical differences were found between PCR and serology in the nursery groups, presumably because of the antibodies detected in the vaccinated groups, the low prevalence of positive animals, and the small sample size used in the study. In older animals (22-24 weeks of age), a larger proportion of animals had seroconverted, and therefore PCR and serology did not differ significantly.

Detecting the microorganism has some advantages over serology: dynamics of infection are better evaluated, and therefore more accurate information can be obtained regarding when the animals get exposed to the microorganism. This information will allow the best timing for administering medication, vaccination, and/or management strategies. As seen in this study, time to seroconversion appeared to be variable: in Farm E, low mycoplasma prevalence was found until the 24-week group, where 80% of the animals tested PCR positive and 30% were serologically positive. This increase coincided with the appearance of respiratory clinical signs. In Farm C, there was a large increase in PCR-positive animals between the 12- and 16-week-old groups. Clinical signs appeared at 16 weeks of age but seroconversion was not detected until 4 weeks later. The low prevalence of positive animals in the nurseries may have proven insufficient to trigger clinical signs or seroconversion in this farm, and mycoplasma-related cough was not evident until a large percentage of animals were PCR positive. In Farm A, there was a sharp increase of positive animals from 8-14 weeks of age, but seroconversion was not detected until the 22-24 week age group. This apparent delay of seroconversion in naturally infected pigs has been previously described, and different intervals have been reported, ranging from 4 weeks to 9 weeks following contact exposure.9-11 Because of this variability, it is difficult to determine time of infection from serology. The variability may be a measure of infection dose, since this dose appears to play a role in lesion severity and amount of shedding by the infected animal9 and therefore probably also in the time of seroconversion of contact pigs.

The slower spread of the microorganism does not explain entirely the delay of seroconversion detected in field cases or contact exposed animals. Clinical signs and pneumonic lesions have been found to be associated.24 Studies have shown that the interval from onset of clinical signs or peak pneumonic lesions to seroconversion is increased in field cases compared to what is found in experimental challenge experiments.11,24 In an intratracheal experimental infection, coughing started at 2 weeks postinfection and peaked at 5 weeks, and seroconversion was detected at 3-4 weeks after infection--2 weeks after the coughing began.24 Conversely, in a field study from a farm infected with M. hyopneumoniae, seroconversion was not detected until 6-8 weeks after the onset of peak pneumonia.11 Also, in that latter study, a trend to earlier seroconversion was observed when the pneumonic peak was higher. Therefore, magnitude of peak appears to affect not only the magnitude of seroconversion, but also the time to seroconversion after infection.

Because of the delay and variability of seroconversion after infection, it is difficult to extrapolate the moment of infection from the serology results. The nested PCR could be used to assist in determining the moment of infection and therefore to optimize the timing of the vaccination and medication protocols.

The two vaccinated herds showed increasing, and then later decreasing antibody titers shortly after vaccination. However, the proportion of positive pigs was not as high as expected, and the decay and disappearance of antibodies was rapid. This could be due to the nature of the bacterins used, the serological test, or inadequate administration or timing of the vaccines. However, this does not imply that the vaccines were ineffective, since the relationship between antibodies and protective immune response against M. hyopneumoniae is still under debate. Some researchers have suggested that the humoral immune response is protective,25,26 whereas others have proposed that it is the cellular immune response.27 It was also observed that vaccination did not prevent presence of M. hyopneumoniae in the nose, as detected by PCR from nasal swabs. This is in accordance with other studies, where vaccination did not prevent the isolation of M. hyopneumoniae from pigs.2 However, no conclusions regarding the efficacy of the vaccines could be drawn, because no unvaccinated control groups were available in the vaccinated farms.

An important limitation of this study is that it was a cross-sectional study, not a prospective study. Therefore, the profiles obtained do not show the evolution of a same group of pigs during their life, but the status of disease in each age group at a point in time. This has the disadvantage of not always reflecting the evolution of infection rate with age. In the case of stable herds, these studies are fairly adequate in showing the evolution of disease throughout the production system. This is not the case, however, in unstable herds, or in offsite herds, where each group is different and disease varies with each group.

The nested PCR also has some limitations: it is a qualitative, not quantitative test. The nested PCR will give a positive result if there are at least 80 microorganisms in the sample.16 In this study, only the number of positive animals in a particular age group was detected, but the bacterial load that these animals were carrying could not be determined. This latter information can be important in assessing the potential of these animals to spread disease, and establishing the significance of finding such a positive animal. However, if the organism is detected, even if it is present only in low numbers, there is a risk that the animal can be shedding and will infect other animals. A quantitative test would better explain and/or predict the slower or more rapid spread of mycoplasma in a group of pigs.

The usefulness of the nested PCR for mycoplasma is also limited at this time by the fact that it is not known what percentage of animals have to be positive in order to reach the critical mass for clinical signs to appear. In this study, the number of positive animals appeared to be related to appearance of clinical signs and later to seroconversion. However, a threshold of positive animals for clinical signs to occur could not be established, especially since the sample size was inadequate for properly measuring prevalence. It is also possible that the "critical mass" varies with the strain involved, bacterial load, management factors, other pathogens present, number and density of animals, susceptibility, medication, and vaccination, so that each farm has its own "threshold." There are studies suggesting that the presence of M. hyopneumoniae in a farm does not necessarily imply an outbreak of disease.5,28 However, none of these studies has shown that M. hyopneumoniae can remain in large populations of intensively housed pigs for several years without producing evidence of respiratory disease.

Further studies, involving more animals per age group--including farms without disease--and quantitating the amount of bacterial load, could give more insight and information on this putative "critical mass" and the effect and significance of carrier animals on disease.


  • The nested PCR is able to detect M. hyopneumoniae earlier in infection than serology can.
  • Detecting M. hyopneumoniae from nasal swabs offers new and valuable information in helping interpret serum profiles, determining when animals get infected, and how and where M. hyopneumoniae is being transmitted.
  • This information has the potential to allow for better timing in the application of medication, vaccination, or management strategies.


This work was partially supported by Pharmacia & Upjohn, United States.


1. Whittlestone P. Enzootic Pneumonia of pigs (EPP). Adv Vet Sci Comp Med. 1973; 17:1-55.

2. Ross RF. Mycoplasmal Diseases. In: Leman AD, et al, eds. Diseases of Swine, 6th Ed. Ames, Iowa: Iowa State University Press; 1986:469-483.

3. Clark LK, Scheidt AB, Armstrong CH, Knox K, Mayrose VB. The effect of all-in/all-out management on pigs from a herd with enzootic pneumonia. Vet Med. 1991;86:946-951.

4. Clark LK, Armstrong CH, Freeman MJ, Scheidt AB, Sands-Freeman L, Knox K. Investigating the transmission of Mycoplasma hyopneumoniae in a swine herd with enzootic pneumonia. Vet Med. 1991;86:543-550.

5. Messier S, Ross RF, Paul PS. Humoral and cellular immune responses of pigs inoculated with Mycoplasma hyopneumoniae. Am J Vet Res. 1990;51:52-58.

6. Suter M, Kobisch M, Nicolet J. Stimulation of immunoglobulin-containing cells and isotype-specific antibody response in experimental Mycoplasma hyopneumoniae infection in specific-pathogen-free pigs. Infect Imm. 1985;49:615-620.

7. Piffer IA, Young TF, Petenate A, Ross RF. Comparison of complement fixation test and enzyme-linked immunosorbent assay for detection of early infection with Mycoplasma hyopneumoniae. Am J Vet Res. 1984;45:1122-1126.

8. Armstrong CH, Freeman MJ, Sands-Freeman L, Lopez-Osuna M, Young T, Runnel LJ. Comparison of the enzyme-linked immunosorbent assay and the indirect hemagglutination and complement fixation tests for detecting antibodies to Mycoplasma hyopneumoniae. Can J Comp Med. 1983;47:464-470.

9. Sitjar M, Noyes EP, Simon X, Pijoan C. Relationships among seroconversion to Mycoplasma hyopneumoniae, lung lesions, and production parameters in pigs. Swine Health Prod. 1996;4:273-277.

10. Bruggmann S, Keller H, Bertschinger HU, Engberg B. Quantitative detection of antibodies to Mycoplasma suipneumoniae in pigs' sera by enzyme-linked immunosorbent assay. Vet Rec. 1977; 101:109-111.

11. Armstrong CH, Freeman MJ, Sands LL. The enzyme-linked immunosorbent assay (ELISA) for diagnosing mycoplasmal pneumonia of swine. Am Assoc Vet Lab Diagn. 1978;21:377-390.

12. Kazama S, Yagihashi T, Seto K. Preparation of Mycoplasma hyopneumoniae antigen for the enzyme-linked immunosorbent assay. Can J Vet Res. 1989;53:176-181.

13. Armstrong CH, Sands-Freeman LL, Freeman MM. Serological, pathological and cultural evaluations of swine infected experimentally with Mycoplasma flocculare. Can J Vet Res. 1987;51:185-188.

14. Bereiter M, Young TF, Joo HS, Ross RF. Evaluation of the ELISA and comparison to the complement fixation test and radial immunodiffusion enzyme assay for detection of antibodies against Mycoplasma hyopneumoniae in swine serum. Vet Microbiol. 1990;25:177-192.

15. Nicolet J, Paroz P, Bruggmann S. Tween 20 soluble proteins of Mycoplasma hyopneumoniae as antigen for an enzyme linked immunosorbent assay. Res Vet Sci. 1980;29:305-309.

16. Calsamiglia M, Pijoan C, Trigo A. Application of a nested polymerase chain reaction assay to detect Mycoplasma hyopneumoniae from nasal swabs. J Vet Diagn Invest. 1999;11:246-251.

17. Artiushin S, Stipkovits L, Minion FC. Development of polymerase chain reaction primers to detect Mycoplasma hyopneumoniae. Mol Cell Probes. 1993;7:381-385.

18. Blanchard B, Kobisch M, Bove JM, Saillard C. Polymerase chain reaction for Mycoplasma hyopneumoniae in tracheobronchiolar washings from pigs. Mol Cell Probes. 1996;10:15-22.

19. Harasawa R, Koshimizu K, Takeda O, Uemori T, Asada K, Kato I. Detection of Mycoplasma hyopneumoniae by the polymerase chain reaction. Mol Cell Probes. 1991;5:103-109.

20. Mattsson JG, Bergstrom K, Wallgreen P, Johansson KE. Detection of Mycoplasma hyopneumoniae in nose swabs from pigs by in vitro amplification of the 16S rRNA gene. J Clin Microbiol. 1995;33:893-897.

21. Goodwin RFW. Isolation of Mycoplasma suipneumoniae from the nasal cavities and lungs of pigs affected with enzootic pneumonia or exposed to this infection. Res Vet Sci. 1972;13:262-267.

22. Freese W. Serology: An aid in the diagnosis of certain respiratory diseases. Agri-Practice. 1990; 11(6):12-15.

23. Sitjar M. Relacion entre lesiones neumonicas y seroconversion a diferentes patogneos respiratorios. (Master's dissertation): Barcelona, Universitat Autonoma de Barcelona, 1996.

24. Kobisch M, Blanchard B, Le Potier MF. Mycoplasma hyopneumoniae infection in pigs: Duration of the disease and resistance to reinfection. Vet Res. 1993;24:67-77.

25. Lam KM, Switzer WP. Mycoplasmal pneumonia of swine: Active and passive immunizations. Am J Vet Res. 1971;32:1737-1741.

26. Messier S , Ross RF, Paul PS. Humoral and cellular immune responses of pigs inoculated with Mycoplasma hyopneumoniae. Am J Vet Res. 1990;51:52-58.

27. Tajima M, Yagihashi T, Nunoya T, Takeuchi A, Ohashi F. Mycoplasma hyopneumoniae infection in pigs immunosuppressed by thymectomy and treatment with antithymocyte serum. Am J Vet Res. 1984;45:1928-1932.

28. Goodwin RFW. The phenomenon of suppressed respiratory disease in the litters of older sows. Vet Rec.1965;77:383-387.