Haemophilus Parasuis

Author: Simone Oliveira, Newport Laboratories


Haemophilus parasuis is still one of the main causes of nursery mortality in most U.S. herds [1]. Mortality rates due to Haemophilus parasuis can be as high as 10% [2], which makes this agent one of the most costly pathogens in swine production. Although a few herds experience nursery mortality solely due to H. parasuis, disease caused by this agent may occur at the same time as other bacterial and viral infections. Streptococcus suis and Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) are two of the agents that are frequently isolated from pigs showing polyserositis due to H. parasuis [3]. The epidemiology of Streptococcus suis infection is very similar to H. parasuis, which may explain why these two agents often affect nursery pigs at 4 to 6 weeks after weaning [4]. PRRSV infection, however, has a very distinct role in H. parasuis infection. A recent study suggests that pigs colonized by a virulent H. parasuis strain are susceptible to development of systemic infection by this agent following PRRSV challenge [5]. Although the interaction between PRRSV and H. parasuis is now evident both experimentally and in the field, the mechanisms involved in such interaction are still unclear.


  • Discuss features regarding diagnosis and epidemiology of Haemophilus parasuis.
  • Describe methods of control of Haemophilus parasuis in the nursery.

Etiological agent (cause)

Haemophilus parasuis is the etiological agent (cause) of the syndrome currently known as “Glässer’s disease”. This organism is an early colonizer of the upper respiratory tract and may be normally isolated from the nasal cavity, tonsil, and trachea of healthy pigs. Although non-pathogenic strains predominate in the upper respiratory tract [1], some animals may harbor virulent strains that can cause systemic disease characterized by fibrinous polyserositis, arthritis, and meningitis [6]. Fifteen serovars of H. parasuis have been reported [7], and there is extensively strain variation within serovar groups [1].

Clinical signs and lesions

The clinical signs characteristic of H. parasuis infection are certainly not unique to this agent, and a differential diagnosis with other pathogens such as S. suis is necessary. The clinical presentation of H. parasuis systemic infection may vary with the virulence of the strain and the immune status of the pig. Three main presentations may be observed in the field: super-acute, acute, and chronic. Super-acute infections are characterized by sudden death with lack of clinical signs and gross lesions in most cases. Some animals may show a slight increase of fluids in the pericardial sac, pleural, and abdominal cavities. Fibrin may or may not be observed. In these cases, H. parasuis may be isolated from the blood, which confirms septicemia. Super-acute infections usually occur in naive herds (H. parasuis-free) or due to infection by a highly virulent strain. Acute infections are more commonly observed in the field and are characterized by fibrinous polyserositis (Figure 1). Clinical signs will usually appear 2-3 days following infection. Affected pigs may present fever (106 – 107 º F), labored breathing (abdominal breathing), coughing, swollen joints, and central nervous system (CNS) signs, which are characterized by lateral decubitus, paddling, and trembling. When samples are properly collected, H. parasuis can be easily isolated from these animals. Acute infection may also be characterized by development of either pneumonia or CNS signs. These clinical presentations may be associated with specific groups of strains. Some strains seem to have a tropism to the brain, causing only CNS signs. Other strains may be found in lungs causing pneumonia without systemic infection. Further studies are necessary to clarify the association between specific strains and clinical presentations in the field. Chronically affected animals (6-7 weeks after weaning) are survivals of the peak of nursery mortality. These animals probably received antibiotic treatments and were able to survive systemic infection. In most chronically affected animals, H. parasuis can no longer be isolated from lesions. These animals usually present poor growth performance throughout late nursery and early finisher and may die from complications of fibrosis in the thoracic cavity.3, 6

Diagnosis and strain characterization

The diagnosis of H. parasuis systemic infection is based on the association between clinical history and isolation of the agent from characteristic lesions. Successful isolation of H. parasuis from clinical samples can be achieved by sampling acutely affected, non-treated animals. Isolation is easier when clinically affected animals are euthanized and fresh samples are submitted to a diagnostic laboratory as soon as possible. Sample collection may be performed using a sterile swab containing Stuart media. It is very important to collect the fibrin on the surface of affected organs, as H. parasuis will be mostly concentrated in this material. Tissues may also be collected for isolation, and should be submitted in separate sterile bags to the laboratory. It is known that different strains can affect one pig at the same time. Special care should be taken to separate brain tissue or swabs from other samples, as some H. parasuis strains seem to have a tropism to the brain [3]. When isolation is negative, polymerase chain reaction (PCR) may be used to detect the presence of H. parasuis in tissues and swabs [8]. PCR is useful to define the role of H. parasuis in nursery mortality, but isolation is still necessary for further characterization of strains by serotyping and genotyping. A new technique, more sensitive and specific, has been recently developed to serotype H. parasuis. Indirect hemaglutination (IHA) is now the technique of choice for H. parasuis serotyping. It apparently reduces the percentage of non-typable isolates compared with traditional serotyping by agar gel precipitation test (AGPT) [9]. Genotyping by Enterobacterial Repetitive Intergenic Consensus-based PCR or ERIC-PCR has been extensively used to characterize and compare H. parasuis field strains [1,10]. This technique is more discriminatory than serotyping and can be used to detect prevalent strains in affected herds (Figure 2). Both techniques (serotyping by IHA and genotyping by ERICPCR) are important for the development of control strategies.


Colonization of piglets by H. parasuis occurs a few days after birth, possible through nose to nose contact with sows and gilts. Gilts appear to shed fewer organisms than sows [11]. If this is true, gilt piglets are at higher risk of developing systemic infection by virulent H. parasuis strains after weaning. Another factor that appears to influence colonization of piglets by H. parasuis is weaning age. Pigs weaned at 14 days tend to have lower levels of colonization compared with piglets weaned at 28 days [11]. Piglets that were colonized in the farrowing house in the presence of maternal immunity may develop active immunity against virulent strains and be protected against systemic infection after weaning and commingling. Piglets that were not colonized prior to weaning are naive animals that may develop systemic infection when commingled with colonized pigs. Systemic infection usually happens around 4 to 6 weeks after weaning, when maternal immunity is no longer protective. This hypothesis drives today’s control strategies to reduce nursery mortality due to H. parasuis infection, as it will be discussed in the control section [12].

Another interesting feature of the epidemiology of H. parasuis is the fact that no matter how many sources are commingled in the nursery, only a few prevalent strains (usually 2-3) will predominate in the herd [1]. Several within herd studies using genotyping by ERIC-PCR have confirmed this fact [1,4,13]. Another interesting feature of the epidemiology of H. parasuis between herds is the fact that although many herds are affected by the same serovar group, each herd has a different prevalent strain. This was observed in herds affected by strains from serovars 2, 4, and 12 [1].

Co-infection with other pathogens

Figure 1A
Figure 1B

As mentioned before, S. suis has a very similar within-herd epidemiology compared with H. parasuis. Both organisms are early colonizers of the respiratory tract and the hypothesis for colonization dynamics for S. suis appear to be similar to that described for H. parasuis [12]. Disease in the nursery caused by S. suis seems to be also associated with the commingling of colonized and non-colonized piglets. Systemic infection by H. parasuis and S. suis occur at the same time in the nursery, and sometimes, these organisms may coinfect nursery pigs. Clinical signs are very similar. It is very common to isolate S. suis from pigs showing fibrinous polyserositis and H. parasuis from pigs showing only CNS signs without other systemic lesions. Streptococcus suis is mostly isolated from found dead pigs, while H. parasuis is easily isolated from clinically affected, non-treated, euthanized pigs. Testing of clinical samples by PCR to detect H. parasuis may help on the differential diagnosis with S. suis infections [3].

For many years, field experiences have suggested that PRRSV and H. parasuis may have a synergistic role in nursery mortality. However, several studies failed to prove the interaction between these two agents [14]. A recent study demonstrated that sentinel pigs that were housed in the same airspace as pigs challenged by the intra-tracheal route with a virulent H. parasuis strain developed pneumonia by this agent following PRRSV challenge. This study is the best evidence to date that these two agents may in fact interact and act synergistically [5]. The mechanisms involved in the interaction between PRRSV and H. parasuis are still unclear. One hypothesis is that the destruction of alveolar macrophages (innate immune response) by PRRSV allows potentially virulent H. parasuis strains to proliferate in the lung and cause pneumonia.


The correct diagnosis of H. parasuis systemic infection in nursery and the evaluation of the withinherd epidemiology is critical for control of nursery mortality caused by this agent. Sample collection from clinically affected, non-treated, euthanized pigs is very important. If isolation is not accomplished, PCR can help to define if H. parasuis is involved in nursery mortality. When isolation is accomplished, further characterization of isolates by serotyping and genotyping is recommended. Both commercial of autogenous products can be used to control nursery mortality. Due to the poor cross-protection between different serovars, it is recommended that the commercial product selected for vaccination contains the serovars affecting the herd. Many H. parasuis isolates are non-typable, and finding a matching commercial product becomes a challenge. In these cases, autogenous vaccines may be a good option for disease control. Selection of isolates to be included in autogenous vaccines should be based on the following factors: serovar, genotype (identification of prevalent strains), site of isolation (systemic isolates from pericardium, pleura, peritoneum, joints, or brain), and date of isolation (isolates recovered within the past 2 years are preferred). Vaccination schedule varies with the peak of mortality in each herd. When systemic infection by H. parasuis is observed 1-3 weeks after weaning (or sometimes in the farrowing house), sow vaccination at 4 and 2 weeks prior to farrowing may be an option. Some herds opt to vaccinate sows and piglets, and although there are some concerns regarding maternal immunity interference in piglet vaccination, a few studies suggest that better results may be achieved when sows and piglets are vaccinated. The majority of swine herds experience the peak of mortality due to H. parasuis infection at 4 to 6 weeks after weaning. This period corresponds to the decrease of maternal immunity. In these cases, piglet vaccination at weaning and 2 weeks later may control mortality [1,3,6]..

Figure 2

Recently, controlled exposure of 5-day old piglets to a low dose of live, virulent H. parasuis using the oral route has been used as an alternative measure to control nursery mortality caused by this agent. Although this technique may reduce nursery mortality by 50%, it does have some inherent risks, especially in herds with active PRRSV infection [2,4-5]. A recent study evaluated the role of sow vaccination and PRRSV infection on the safety of H. parasuis controlled exposure. Sow vaccination with a killed bacterin containing the same H. parasuis strain used for colonization reduced the risk of clinical signs and systemic infection following piglet exposure. Pigs from non-vaccinated sows showed coughing and swollen joints following exposure to a low dose of live H. parasuis, and one pig from this group developed H. parasuis systemic infection 2 days after weaning. Pigs from vaccinated and non-vaccinated sows that were exposed to H. parasuis at 5 days old were equally protected against homologous challenge by the intratracheal and intraperitoneal routes at 4 weeks of age. These results suggest that controlled exposure promoted the development not only of local immune response (protection against respiratory challenge), but also of systemic immune response (protection against intraperitoneal challenge). Pigs from both groups were protected against systemic infection by the strain used for controlled exposure even following PRRSV challenge. Sentinel pigs from the negative control group that were housed in the same air space as pigs challenged with H. parasuis developed pneumonia by this agent following PRRSV challenge. These results suggest that PRRSV may predispose pigs harboring virulent H. parasuis strains to develop systemic infection by this agent [5].


Correct diagnosis of H. parasuis systemic infection in the nursery is key for control of this costly pathogen. It is very important to consider the differential diagnosis with S. suis infections and to evaluate the role of PRRSV concurrent infections in the nursery. Detection of H. parasuis in clinical samples may be performed by PCR, although isolation is still recommended for further characterization of isolates by serotyping and genotyping. Control measures should be designed based on the timing of mortality in each specific herd. Definition of serovars and the number of strains involved in nursery mortality are also important for selection of control measures. Commercial products should contain serovars similar to those present in the herd. Autogenous products should contain representative isolates from each group of strains affecting the herd. Controlled exposure should be used with caution, since clinical signs may occur in exposed piglets and systemic infection by the H. parasuis strain used for exposure may happen following stressful events such as weaning. When using this alternative control measure, one should keep in mind that pigs from herds experiencing active PRRSV infection in the farrowing house and in the nursery may be at greater risk of systemic infection by H. parasuis.


1. Oliveira, S., Blackall, PJ, Pijoan, C. Characterization of the diversity of Haemophilus parasuis field isolates by serotyping and genotyping. Am J Vet Res 2003;64(4):435-442.

2. Oliveira, S, Pijoan, C., Morrison, R. Comparison of Haemophilus parasuis control in the nursery using vaccination and controlled exposure. Journal of Swine Health and Production. (Accepted). 2004.

3. Oliveira, S, Pijoan, C, Morrison, R. Role of Haemophilus parasuis in nursery mortality. Proceedings of the Allen D. Leman Swine Conference, Minneapolis USA, 2002. p. 111.

4. Oliveira, S, Batista, L, Torremorell, M, Pijoan, C. Experimental colonization of piglets and gilts with systemic strains of Haemophilus parasuis and Streptococcus suis to prevent disease. Can J Vet Res 2002;65(3):161-167.

5. Oliveira, S, Mahlberg, J, Simonson, R. Safety of controlled exposure to Haemophilus parasuis: The role of sow vaccination and PRRSV infection. Submitted for oral presentation at the 18th Congress of the International Pig Veterinary Society, Hamburg. 2004.

6. Oliveira, S, Pijoan, C. Haemophilus parasuis: new trends on diagnosis, epidemiology and control. Review article. Vet. Microbiol. 2004;99(1):1-12

7. Kielstein, P, Rapp-Gabrielson, VJ. Designation of 15 serovars of Haemophilus parasuis on the basis of immunodiffusion using heat-stable antigen extracts. J. Clin. Microbiol. 1992;30(4):826-865.

8. Oliveira, S, Galina, L, Pijoan, C. Development of a PCR test to diagnose Haemophilus parasuis infections. J Vet Diag Invest 200;13:495-501.

9. Del Rio, WL, Gutiérrez, CB, Ferri, EFR. Value of indirect hemagglutination and coagglutination tests for serotyping Haemophilus parasuis. J. Clin. Microbiol. 2003;41(2):880-882.

10. Rafiee, M, Bara, M, Stephens, CP, Blackall, PJ. Application of ERIC-PCR for the comparison of isolates of Haemophilus parasuis. Aust. Vet. J. 2000;78(12):846-849.

11. Kirkwood, RN, Rawluk, SA, Cegielski, AC, Otto, AJ. Effect of pig age and autogenous sow vaccination on nasal mucosal colonization of pigs by Haemophilus parasuis. J. Swine Health Prod. 2001;9(2):77-79.

12. Pijoan, C, Torremorell, M, Solano, G. Colonization patterns by the bacterial flora of young pigs. Proc Am Assoc Swine Pract. 1997. pp 463-464.

13. Oliveira, S, Ruiz, A, Pijoan, C. Phenotypic and genotypic characterization of Haemophilus parasuis isolates involved in a multi-farm outbreak. Proceedings of the 16th Congress of the International Pig Veterinary Society, Melbourne; 2000. p.530.

14. Solano, GI, Bautista, E, Molitor, TW, Segales, JM, Pijoan, C. Effect of porcine reproductive and respiratory syndrome virus infection on the clearance of Haemophilus parasuis by porcine alveolar macrophages. Can J Vet Res 1998;62:251-256.