Pasteurella Species

Description of the Pathogen

The family Pasteurellaceae includes the genera Pasteurella, Haemophilus, Actinobacillus, Aggregatibacter, and Mannheimia, among others. On the basis of the genomic sequence, it appears that Pasteurella and Haemophilus diverged approximately 270 million years ago. DNA hybridization separates the Pasteurella spp. into two groups: (1) Pasteurella sensu stricto and (2) Pasteurella -related spp., with the latter more closely related to Haemophilus and Aggregatibacter ( Table 228.1 ). Species of the genus Pasteurella are nonmotile, facultatively anaerobic, gram-negative coccobacilli measuring 1 to 2 µm in length. Organisms grow in culture on a variety of commercial media, including sheep blood and chocolate agar, but not usually on MacConkey agar media. They are fastidious and can be difficult to isolate and identify from nonsterile specimens such as sputum. In one study matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF) accurately identified 59 of 66 Pasteurella isolates and had the advantage of rapidity. Most strains are catalase, oxidase, and indole positive and produce acid from sucrose. The most common human isolates belong to the P. multocida group and appear as smooth, iridescent, blue colonies on growth media. Encapsulated isolates typically appear mucoid. Strain differences of P. multocida have traditionally been identified on the basis of capsular antigens that define 5 serogroups (A–F) and somatic antigens that define 16 serovars (1–16). Strains in groups B, E, and F have been rarely associated with human infection. Of the remaining strains (A, D), group A strains have been more frequently isolated as respiratory tract colonizers or pathogens, whereas non-A strains have been isolated more frequently from nonrespiratory tract specimens, including blood, cerebrospinal fluid, and abscess fluid. Ribosomal DNA and other sequence-based molecular techniques offer better resolution than serotyping and have now superseded it for identification and typing of Pasteurella. Because of their sometimes fastidious nature, specific guidelines for methodologies to accurately determine minimal inhibitory concentrations (MICs) of Pasteurella spp. have been published.

Epidemiology

On the basis of case reports and case series of infected patients, Pasteurella spp., particularly P. multocida, appear to have a worldwide distribution. For most Pasteurella spp. the principal reservoir is in animals. P. multocida has been isolated from the upper respiratory tracts of a variety of animals, including dogs, cats and other felines, pigs, and a wide variety of domestic and wild animals. Dogs and cats have particularly high colonization rates. In most cases carriage is asymptomatic, although both upper and lower respiratory tract infections and septicemia are well known to occur in animals. Although the reservoirs of most of the non- multocida spp. ( P. canis, P. stomatis, P. dagmatis, P. aerogenes, and P. pneumotropica ) are likely animals, notable is Pasteurella bettyae, a cause of neonatal infection and genitourinary infection in adults, whose reservoir is not well defined. Respiratory tract colonization by P. multocida in humans is well known to occur. In most cases colonized patients have underlying upper or lower respiratory tract diseases, including chronic sinusitis, chronic obstructive pulmonary disease (COPD), or bronchiectasis. Most colonized patients have a history of household or domesticated animal contact.

Broadly speaking, human infection with Pasteurella can be divided into three types: infection occurring after animal bites, usually from dogs or cats; infection occurring after other animal exposures; and infection with no known animal contact. Infection after animal bites is the most commonly reported clinical setting for the organism (see Chapter 315 ).

Among animal bites, dog bites are most common, followed by cat bites. Approximately 15% to 20% of dog bite wounds and greater than 50% of cat bite wounds become infected. The higher incidence of infection after cat bites probably results from the fact that cat teeth are thinner and more commonly result in puncture wounds, which are known to carry a higher risk of infection. Pasteurella spp. are the most commonly isolated pathogens from dog and cat bites, present in 50% and 75% of cases, respectively; followed by Staphylococcus and Streptococcus spp. Francis and associates, studying bite-related P. multocida infections in Oregon between 1962 and 1972, noted that 76% were the result of cat bites, and the remaining 24% were from dog bites. The difference in incidence of Pasteurella infections in dog and cat bites also may reflect the higher rate of upper respiratory colonization in cats. Pasteurella infections have also been reported after bites from a variety of other animals, including pigs, rats, lions, opossums, and rabbits. In addition to bites, Pasteurella infections have also been reported after dog and cat scratches and from the licking of open wounds by these animals.

Pasteurella infections are well known to develop in patients exposed to animals but without a history of bites or scratches. These include skin and soft tissue infections, bone and joint infections, pneumonia, meningitis, endocarditis, and septicemia. Persons at risk for infection from animal exposure include veterinarians, farmers, livestock handlers, pet owners, and food handlers. Although the route of infection in many reported cases is not clear, most have been presumed to result from inadvertent direct inoculation of organisms or from upper respiratory tract colonization, with subsequent dissemination to the target organ or organs. In one case report a patient with underlying bronchiectasis and diabetes was hospitalized with pneumonia caused by P. multocida . Pulsed-field gel electrophoresis of her sputum isolate, compared with P. multocida isolated from her dog's pharynx, yielded identical patterns, strongly implicating her dog as the source of her infection.

In a significant proportion of Pasteurella cases, no known animal exposure or contact can be identified. In 1970 Hubbert and associates reviewed what was then the world's literature; they identified 72 reported cases of P. multocida infection unrelated to bites and described 136 additional cases. In 16% of the reviewed cases and 31% of their additional cases, no animal exposure or contact could be identified. Once again, the spectrum of infectious complications was wide, similar to that described for patients with nonbite animal exposures. In a more recent literature review of 156 Pasteurella bacteremia adult cases, 20% did not have animal exposures reported. Vertical transmission of P. multocida has been described rarely.

Pathogenesis

Despite Pasteurella spp. having been long established as human pathogens, the precise mechanisms of pathogenicity remain uncertain. In animals virulent P. multocida strains adhere to mucosal epithelial cells in the upper respiratory tract, particularly in the tonsils, and in some cases they are mediated by fimbriae. Several virulence factors have been described in Pasteurella spp. Much attention has focused on P. multocida toxin (PMT). PMT, produced by strains A and D, is a potent mitogen that has been shown to activate a number of intracellular signaling cascades, resulting in a multitude of deleterious effects. Specifically, PMT has been shown to inhibit the migratory response of dendritic cells, thereby impairing immune surveillance. Along with ToxA protein, also produced by P. multocida, PMT is associated with progressive atrophic rhinitis in pigs. P. multocida lipopolysaccharide (LPS) is another virulence factor, proven by the fact that strains with truncated LPS have been shown to be attenuated in virulence. In addition, most virulent Pasteurella strains produce polysaccharide capsules that confer many possible mechanisms of pathogenicity, including resistance to desiccation, promotion of adherence, and resistance to phagocytosis and complement-mediated killing. Finally, binding of transferrin by some pathogenic Pasteurella strains has been demonstrated and may be a mechanism used by the bacteria to ensure an iron supply necessary for growth. Indeed, on the basis of the genomic sequence data, 2.5% of the entire genome encodes for proteins involved in iron acquisition.

The humoral response to P. multocida infection has been characterized. Antibodies to somatic and capsular antigenic determinants develop within 2 weeks after clinical infection. Capsular antibodies are more long-lasting than somatic antibodies. The precise role for such antibodies in host defense in humans is not clear. Studies in healthy breeders whose livestock suffered from pasteurellosis show a high rate of seropositivity to Pasteurella , illustrating the limitations of serology for the diagnosis of active Pasteurella infection in humans.

Clinical Manifestations

Most reported Pasteurella infections in humans are caused by P. multocida and involve skin and soft tissues. Although both subspecies multocida and septica cause skin and soft tissue infections as a result of animal bites and scratches, subspecies multocida has also been found to cause respiratory tract infections and systemic infection. Other species have been described much less commonly ( Table 228.2 ). Beyond skin and soft tissues, Pasteurella can uncommonly cause chronic respiratory tract infection in patients with underlying pulmonary disease and invasive infection syndromes, such as bacteremia, endocarditis, and meningitis. In a retrospective case-control study seeking to identify risk factors for invasive Pasteurella infection, only advanced age was found to be significant in multivariate analysis, but it is thought that immunocompromised hosts are at enhanced risk.