OVERVIEW: What every practitioner needs to know

Are you sure your patient has Pasteurella infection? What should you expect to find?

  • Pasteurella species most commonly cause skin and soft tissue infections following an animal bite or scratch, typically from a cat or dog. Pain, tenderness, swelling, and erythema often develop and progress rapidly. Localized lymphadenopathy and lymphangitis are common. Spread to adjacent structures can result in tenosynovitis and soft tissue abscess formation. Septic arthritis and osteomyelitis are less common, developing as a result of direct inoculation or by spread from the initial soft tissue site of infection. Fever accompanies soft tissue infection less than half the time. The upper extremities are the most common site of infection, not surprisingly since dog and cat bites mostly occur in the fingers, hands, and arms. The lower extremities are the next most common site of infection, followed by the face and neck.

  • The respiratory tract is the second most common site of Pasteurella infection. A variety of respiratory tract infections have been described including sinusitis, otitis, mastoiditis, tracheobronchitis, and pneumonia with accompanying empyema. Most patients with lower respiratory tract Pasteurella infections are older and have underlying lung conditions such as COPD and bronchiectasis. The symptoms of these infections are not clinically distinguished from other causes of respiratory tract infections. Pneumonia is probably most often focal in appearance but has been described as diffuse in some patients.

  • Other less common Pasteurella infections have included the following (none of which are clinically distinguishing for Pasteurella):


    Continue Reading

    septicemia—typically in the setting of a primary site of infection (e.g. cellulitis, arthritis, peritonitis), occurring in patients with underlying medical conditions, most notably cirrhosis

    endocarditis—rarely reported in native valves; even less commonly reported in prosthetic valves

    central nervous system infections—meningitis (most commonly in infants and patients aged >60 years), brain abscess, and subdural empyema (all rare)

    intra-abdominal infections—appendicitis, spontaneous bacterial peritonitis (in patients with cirrhosis), and peritonitis associated with CAPD

  • Pasteurella skin and soft tissue infections. Physical findings include rapidly spreading edema, erythema, induration, and tenderness at the site of the bite or scratch. Subcutaneous fluctuance may indicate underlying abscess formation. Lymphangitis and regional tender lymphadenopathy are commonly observed. Pain on active or passive movement of fingers, hands, toes, or feet may indicate tenosynovitis. Joint inflammation suggests septic arthritis.

  • Respiratory tract infections. Physical findings are not clinically distinguishing for most sites of infection. Signs of consolidation and pleural fluid accumulation indicate focal pneumonia and empyema.

  • Other sites of infection. Physical findings are not clinically distinguished from disease caused by other pathogens.

How did the patient develop Pasteurella infection? What was the primary source from which the infection spread?

  • Pasteurella infections are most commonly associated with bites or scratches from cats or dogs. Cat bites are thought to represent the highest risk of infection likely due to their small, sharp teeth which more often cause punctures than tears or avulsions and carry a higher incidence of infection. Bites and scratches from a variety of other feline and nonfeline animals have been associated with Pasteurella infections. Infections also have been associated with licking by animals of open wounds.

  • Pasteurella infections have less commonly been described in the absence of bites or scratches, particularly in animal handlers including veterinarians, farmers, and others involved in the livestock industry.

  • A small group of individuals with Pasteurella infections have no clear animal exposure. It is difficult to gauge the relative frequency of such infections.

  • In most clinical settings a history of bites or scratches from an animal, most commonly a dog or cat, should be an important clue as to the possibility of Pasteurella infection. Absent bites or scratches, exposure to animals should raise suspicion of Pasteurella in a patient presenting with rapidly progressing skin and soft tissue infection, particularly among animal handlers.

Which individuals are of greater risk of developing Pasteurella infection?

  • Besides animal exposures, most notably bites and scratches, there are no clear predisposing conditions to skin and soft tissue infections caused by Pasteurella species. For the second most common clinical manifestation of Pasteurella infection, namely respiratory tract infections, pneumonia and empyema are most likely seen in patients with underlying respiratory tract abnormalities, most notably COPD and bronchiectasis. These patients are usually older (aged >60 years). Rare cases of meningitis have been seen in infants and older adults. Rare cases of spontaneous bacterial peritonitis have been seen in cirrhotics.

Beware: there are other diseases that can mimic Pasteurella infection:

Pasteurella infection may be mimicked by:

  • skin and soft tissue infections secondary to an animal or human bites due to polymicrobial flora that may or may not include Pasteurella species

  • cellulitis due to β-hemolytic streptococci or Staphylococcus aureus

  • necrotizing fasciitis due to β-hemolytic streptococci alone or due to polymicrobial infection, the latter particularly in diabetics

What laboratory studies should you order and what should you expect to find?

Results consistent with the diagnosis

  • A peripheral white blood cell count with differential will typically show leukocytosis with a left shift.

  • The erythrocyte sedimentation rate and the C-reactive protein level will be elevated but not otherwise helpful in distinguishing Pasteurella infections from other causes.

Results that confirm the diagnosis

  • Blood cultures are positive for the organism in a minority of patients with skin and soft tissue infections.

  • Cultures of draining wounds and especially cultures taken intraoperatively are most likely to yield a specific diagnosis of Pasteurella infection. The organism can be somewhat fastidious and sometimes confused with other bacteria such as Haemophilus species or Moraxella. Culture of Pasteurellaspecies from any sterile body fluid (e.g. cerebrospinal fluid, peritoneal fluid, or joint fluid) is diagnostic of Pasteurella infection. Respiratory tract culture of Pasteurella species should raise the suspicion of causation, particularly in patients with underlying lung disease who present with lower respiratory tract disease.

What imaging studies will be helpful in making or excluding the diagnosis of Pasteurella infection?

  • A plain radiograph may be helpful in examining bone and soft tissues, particularly if myositis or fasciitis is of concern (evaluate for soft tissue gas formation). $

  • A magnetic resonance imaging scan should be considered for patients with a prolonged course for whom osteomyelitis is a consideration. $$$

($ = 60–125, $$ 125–500, $$$ 500–1,000, $$$$ >1,000)

What consult service or services would be helpful for making the diagnosis and assisting with treatment?

If you decide the patient has Pasteurella infection, what therapies should you initiate immediately?

Patients presenting with simple cellulitis can be managed as outpatients or inpatients depending on the severity of their symptoms. Patients presenting with a rapidly progressing skin and soft tissue infection, especially if febrile or with signs to suggest soft tissue abscess, tenosynovitis, or septic arthritis, should be hospitalized and treated with parenteral antibiotics. Surgery consultation (e.g. general surgery, orthopedic surgery, or plastic surgery) should be considered early on since many of these infections will require surgical debridement in addition to antimicrobial treatment. An infectious diseases physician can assist with antibiotic management, particularly for patients who will require OPAT. Pasteurella infections involving less common sites should be managed as would be typical for these infections with antibiotics directed at Pasteurella species, once identified.

1. Anti-infective agents

If I am not sure what pathogen is causing the infection what anti-infective should I order?

Empiric oral antibiotic treatment should be considered for animal bite wounds prophylactically (discussed below) and utilized in cases that show mild to moderate inflammation. Amoxicillin/clavulanic acid is usually the first choice agent in these situations since it covers not only most Pasteurella isolates but the typical polymicrobial spectrum of organisms associated with bite and scratch wounds that include oral anaerobes, streptococci, and S. aureus. Other oral agents that can be used empirically for penicillin-intolerant patients include moxifloxacin, doxycycline, or the combination of clindamycin with either levofloxacin, TMP/SMX, or an oral second or third generation cephalosporin (e.g. cefuroxime, cefpodoxime, or cefixime) for patients whose penicillin intolerance is not likely immunoglobulin E-mediated.

β-lactamase producing strains of Pasteurella multocida have been isolated from at least five human cases. For that reason susceptibility testing should be routinely performed on all isolates obtained from sterile sources.

For patients with more severe illness requiring hospitalization, first choice empiric therapy should be ampicillin/sulbactam. With the increasing prevalence of community-acquired MRSA, consideration should be given to adding vancomycin empirically as well. For penicillin-intolerant patients without a history of immunoglobulin E-mediated allergy to penicillin, a third or fourth generation cephalosporin (e.g. ceftriaxone, cefotaxime, or cefepime) can be used along with an agent to treat anaerobes such as clindamycin, with or without vancomycin. For patients with a history of anaphylaxis or unknown reaction to penicillin, a fluoroquinolone, doxycycline or TMP/SMX should be substituted for the β-lactam agent.

The optimal duration of antimicrobial treatment has not been well-defined. For skin and soft tissue infections of mild to moderate severity, 7 to 10 days of treatment is probably adequate. For more severe infections with involvement of deeper structures, longer courses are probably sensible. Bone and joint infections should probably be treated for 4 to 6 weeks. For infections involving other organs duration of treatment must be guided by the clinical course.

Treatment options are summarized in Table I.

Table I.
Scenario Antibiotics Dose/Schedule
Empiric—mild to moderate severityFirst line: oralSecond line: oral Amoxicillin/clavulanateMoxifloxacin ORDoxycycline ORClindamycin PLUS    Levofloxacin OR>    TMP/SMX OR>    2nd or 3rd generation cephalosporin* 875mg/125mg twice a day400mg daily100mg twice a day300mg three tmes a day500–750mg daily1 double strength tablet twice a dayDoses vary
Definitive—mild to moderate severity (Pasteurella alone)First line: oralSecond line: oral Penicillin V potassium ORAmoxicillinLevofloxacin ORMoxifloxacin ORDoxycycline ORTMP/SMX OR2nd or 3rd generation cephalosporin* 500mg four times a day875mg twice a day500–750mg daily400mg daily100mg twice a day1 double strength twice a dayDoses vary
Empiric—severe infectionFirst line: parenteralSecond line: parenteral Ampicillin/sulbactam ± Vancomycin3rd or 4th generation cephalosporin* ORLevofloxacin ORMoxifloxacin ORTMP/SMX ORDoxycycline PLUS Clindamycin ± Vancomycin 3g every 6 hoursDose variesDoses vary750mg every 24 hours400mg every 24 hours5–10mg/kg/day (divided every 8 hours)100mg every 12 hours600mg every 8 hoursDose varies
Definitive—severe infection (Pasteurella alone)First line: parenteralSecond line: parenteral Penicillin G ORAmpicillin3rd or 4th generation cephalosporin* ORLevofloxacin ORMoxifloxacin ORTMP/SMX ORDoxycycline 3–4 million units every 4 hours2g every 4 hoursDoses vary750mg every 24 hours400mg every 24 hours5–10mg/kg/day (divided every 8 hours)100mg every 12 hours

* Unless there is a history of immunoglobulin E-mediated allergy.

SMX, sulfamethoxazole; TMP, trimethoprim.

2. Other key therapeutic modalities

  • Given their propensity for rapid progression, spread to deep tissues, and abscess formation, severe Pasteurella skin and soft tissue infections often require urgent and aggressive surgical debridement in addition to antimicrobial therapy.

  • The routine use of antimicrobial treatment as prophylaxis for animal bite wounds is controversial. Prophylaxis should be considered for patients with:

    severe injuries

    likely penetration to the bone and/or joint (especially prosthetic joints)

    immunocompromised patients

    wounds involving the hands

What complications could arise as a consequence of Pasteurella infection?

What should you tell the family about the patient's prognosis?

  • For patients with severe skin and soft tissue infections, particularly involving tendons, joints, or bone, the prognosis for full recovery of function depends upon the extent of disease, the duration of symptoms, and the aggressiveness of medical and surgical intervention. The expected mortality is low.

  • Patients presenting with pneumonia with or without empyema typically are older and have pre-existing lung disease. The combination of a virulent pathogen and an already compromised host results in considerable morbidity and mortality that approaches 30% of cases.

  • Morbidity and mortality of Pasteurella infections involving other organ systems varies by both host and severity of disease. Since many of these patients have underlying medical conditions (e.g. cirrhosis with Pasteurella peritonitis, meningitis in infants) mortality is often quite high.

How do you contract Pasteurella infection and how frequent is this disease?

Pasteurella infections are divided into three groups, epidemiologically: 1) those associated with animal bites and scratches; 2) those associated with animal exposures but without a clear history of an invasive incident; and, 3) those that occur in the absence of any known animal contact.

  • Animal bites/scratches. Skin and soft tissue infections are most often associated with dog or cat bites, more often the latter owing to a higher feline concentration of organisms in the mouth and due to the puncture-type wounds most often seen with cat bites. Licking of nonintact skin such as open sores or wounds also has been associated with Pasteurella infections. A multitude of other animals, including other feline and nonfeline mammals, have been implicated in human Pasteurella infections.

  • Animal exposures (without bites or scratches). Individuals in this group typically have animal contact but no clear animal-related trauma. Most individuals in this group are involved in animal-related professions (e.g. livestock handlers, veterinarians, farmers, etc.). Sites of infection are more varied than skin and soft tissue and include peritonitis, meningitis, septicemia, and endocarditis.

  • No clear animal association. These infections are relatively rare and are similar in scope to the types of infections seen in individuals with animal exposure without bites or scratches. At least for some patients the infection likely arises secondary to pre-existing colonization.

Pasteurella species have a worldwide distribution. Pasteurella species have been cultured from a variety of animal species and are known to cause diseases such as snuffles in rabbits, pneumonia in sheep, and “shipping fever” in cattle. Not surprisingly, cases of Pasteurella infection have been documented following bites and scratches from a number of animal species. Yet cat and dog bites and scratches are the predominant mode of infection. There are no data regarding incidence of infection nor has any seasonal variation of Pasteurella cases been described.

The major means of spread of Pasteurella species is from bites or scratches from animals, particularly cats and dogs. Many animal species are known to be colonized with the organism. Puncture wounds secondary to the small, sharp teeth of cats are thought to represent the highest risk for infection, introducing the organism by direct inoculation into subcutaneous tissues. Bite tear wounds and scratches also carry a risk for infection. Animal licking of nonintact skin has been reported to cause skin and soft tissue infections. Infection tends to spread quickly, disseminating through soft tissue and by lymphatics. Both human to human transmission and vertical transmission have been described. Colonization of the human respiratory tract, particularly in individuals with underlying lung disease, can result in the development of both upper and lower respiratory tract infections including sinusitis, tracheobronchitis, epiglottitis, pneumonia, and empyema.

Most epidemiologic studies of Pasteurella infections have been small case series and individual case reports, focusing on a particular disease manifestation. Weber and colleagues in 1984 reported 34 cases of Pasteurella infection and reviewed the literature. Of their reported and reviewed cases totaling 446, nearly 50% were skin and soft tissue. The remaining major sites of infection included:

  • bone and joint (13%) (septic arthritis, osteomyelitis, bursitis)

  • oral and respiratory tract (14%) (peritonsillar abscess, sinusitis, pharyngitis, epiglottitis, submandibular abscess, tracheobronchitis, pneumonia, empyema)

  • cardiovascular (13%) (bacteremia, endocarditis, mycotic aneurysm, purulent pericarditis, infected vascular graft)

  • central nervous system (5%) (meningitis, brain abscess, subdural empyema)

  • gastrointestinal (4%) (liver abscess,
    SBP, appendiceal abscess, peritonitis, gastroenteritis)

  • genitourinary tract (2%) (cystitis, pyelonephritis, infected ileal loop, renal abscess, vaginitis, cervicitis, Bartholin gland abscess, chorioamnionitis, epididymitis)

  • eyes (1%) (conjunctivitis, corneal ulcer, endophthalmitis)

Of the isolates tested for antimicrobial sensitivity, penicillin, second and third generation cephalosporins, tetracyclines, and chloramphenicol were most consistently active. TMP/SMX also was active. First generation cephalosporins and anti-staphylococcal penicillins were not found to be active. Aminoglycosides were variable. Fluoroquinolones were not yet available at the time of the study.

Zoonotic transmissions are the major means of acquisition of Pastuerella infections, most commonly by the bites and scratches of animals, mostly cats and dogs. In addition to dogs and cats, a sampling of animals that have been reported to cause Pasteurella infections includes lions, pigs, horses, rats, opossums, and turkeys.

What pathogens are responsible for this disease?

The taxonomy of genus Pasteurellahas undergone frequent changes as improved molecular techniques have allowed for more precise speciation. The species, P. multocida is divided now into four subspecies:

  • subspecies multocida(most common isolate overall; recovered from a variety of sites including skin and soft tissue, blood, cerebrospinal fluid, etc.)

  • subspecies septica(less common than multocida, but more closely associated with skin and soft tissue infections)

  • subspecies gallicida

  • subspecies tigris

Other species have been found to cause human disease including P. canis, P. dagmatis, and P. stomatitis. Finally, there are a variety of Pasteurella-related species that cause disease including P. aerogenes, P. bettyae, P. caballi, and other less common species. Virtually all have been isolated from the respiratory tracts of a variety of animals, particularly dogs and cats. (P. canisis unique to dogs.) Colonization of the human respiratory tract has been found in individuals with underlying lung disease. Such individuals usually have household pets.

How do these pathogens cause Pasteurella infection?

Pasteurella multocida consists of five serogroups (A, B, D, E, and F) based on capsular antigens. In addition, 16 serotypes (1 to 16) have been identified based on somatic antigens.

The precise mechanisms by which Pasteurella species cause disease are not fully understood. Once the organism is inoculated it spreads quickly through the soft tissue and proximally along lymphatic vessels. Tenosynovitis and subcutaneous abscess formation are known to develop quickly following the bite or scratch. Direct inoculation into joint or bone has been associated with septic arthritis and osteomyelitis, respectively.

Pasteurella species colonize the mouth and upper respiratory tract of many animals (including humans). They adhere by means of fimbriae. Several toxins produced by virulent P. multocida strains include the following:

  • Pasteurella multocida toxin—a potent mitogen that inhibits dendritic cell migration

  • ToxA protein—induces atrophic rhinitis in pigs

  • polysaccharide capsule—resistance to phagocytosis and complement-mediated killing

  • iron scavenging—ensures an adequate supply of iron necessary to growth and proliferation

What other clinical manifestations may help me to diagnose and manage Pasteurella infection?

Pasteurella should be considered for any patient who presents with a skin and soft tissue infection following an animal bite or scratch. What may help distinguish Pasteurella infection from other bacterial pathogens is its rapid onset and rapid progression with a tendency towards subcutaneous abscess formation and tenosynovitis. Nonetheless, Pasteurella is often one organism cultured as part of a polymicrobial infection following an animal bite or scratch. Certainly in the initial period of treatment, antibiotics should be aimed broadly to include Pasteurella and early surgical intervention should be strongly considered.

The importance of obtaining a detailed history of animal exposures for any patient presenting with an infectious disease is best illustrated by Pasteurella species. Bites and scratches are the most egregious types of exposures and are not commonly missed. Yet contact with animals, particularly licking, at sites of nonintact skin may be overlooked in a patient presenting with a skin and soft tissue infection that ultimately turns out to be Pasteurella. If such an infection were treated as a simple cellulitis with an antibiotic such as an anti-staphylococcal penicillin (e.g. nafcillin, oxacillin) or first generation cephalosporin (cefazolin), aimed at S. aureus and streptococci, Pasteurella would not be covered and the infection could spread due to inadequate coverage. Similarly, mere exposure to pets should at least raise the possibility of Pasteurella infection for a patient presenting with a deep-seated bacterial infection such as meningitis or endocarditis.

The most suggestive clinical findings that might suggest a skin and soft tissue infection secondary to Pasteurella are the rapidly progressive nature of the infection and the findings on exam of tenosynovitis and/or subcutaneous abscess formation. There are no other physical findings of the other types of Pasteurella infection that are clinically distinguishing.

What other additional laboratory findings may be ordered?

No other laboratory tests beyond culture of clinical materials are currently available for the diagnosis of Pasteurella infections.

How can Pasteurella infection be prevented?

The best means of preventing serious Pasteurella skin and soft tissue infections is the possible use of antibiotic prophylaxis for certain high-risk animal bites and scratches. The routine use of antimicrobial treatment as prophylaxis for animal bite wounds is controversial. Prophylaxis should be considered for patients with:

  • severe injuries

  • likely penetration to the bone and/or joint (especially prosthetic joints)

  • immunocompromised patients

  • wounds involving the hands

WHAT'S THE EVIDENCE for specific management and treatment recommendations?

Antony, SJ, Oglesby, KA. “Peritonitis associated with in peritoneal dialysis patients—case report and review of the literature”. Clin Nephrol. vol. 68. 2007. pp. 52-6. (Review of the literature for this uncommon presentation.)

Citron, DM, Warren, YA, Fernandez, HT, Goldstein, MA, Tyrrell, KL, Goldstein, EJ. “Broth microdilution and disk diffusion tests for susceptibility testing of species isolated from human clinical specimens”. J Clin Microbiol. vol. 43. 2005. pp. 2485-8. (Practical guidelines for susceptibility testing of Pasteurella strains.)

Ewing, R, Fainstein, V, Musher, DM, Lidsky, M, Clarridge, J. “Articular and skeletal infections caused by “. South Med J. vol. 73. 1980. pp. 1349-52. (The earliest comprehensive review of bone and joint infections caused by Pasteurella species.)

Goldstein, EJC, Mandell, GL, Bennett, JE, Dolin, R. “Bites”. Mandell, Douglas, and Bennett's Principles and Practice of Infectious Diseases. 2010. (Succinct discussion of animal and human bites with a concise table of recommendations for the management of bite wounds.)

Green, BT, Ramsey, KM, Nolan, PE. ” meningitis: case report and review of the last 11 years”. Scand J Infect Dis. vol. 34. 2002. pp. 213-17. (Review of the literature for this uncommon presentation.)

Harper, M, Boyce, JD, Adler, B. “pathogenesis: 125 years after Pasteur”. FEMS Microbiol Lett. vol. 265. 2006. pp. 1-10. (Concise review of mechanisms of pathogenesis.)

Holst, E, Rollof, J, Larsson, L, Nielsen, JP. “Characterization and distrubution of species recovered from infected humans”. J Clin Microbiol. vol. 30. 1992. pp. 2984-7. (This is the first good study that linked clinical conditions to specific Pasteurella species and subspecies.)

Jorgensen, JH, Hindler, JE. “New consensus guidelines from the Clinical and Laboratory Standards Institute for antimicrobial susceptibility testing of infrequently isolated or fastidious bacteria”. Clin Infect Dis. vol. 44. 2007. pp. 280-6. (Latest guidelines for susceptibility testing that includes Pasteurella species.)

Lion, C, Lozniewski, A, Rosner, V, Weber, M. “Lung abscess due to β-lactamase-producing “. Clin Infect Dis. vol. 29. 1999. pp. 1345-6. (Case report and review of the literature of β-lactamase-producing P. multocida.)

May, BJ, Zhang, Q, Li, LL, Paustian, ML, Whittam, TS, Kapur, V. “Complete genomic sequence of , Pm 70”. Proc Natl Acad Sci U S A. vol. 98. 2001. pp. 3460-5. (Landmark study reporting the complete P. multocida genome with a description of somatic and functional genes.)

Medeiros, I, Saconato, H. “Antibiotic prophylaxis for mammalian bites”. Cochrane Database Syst Rev. vol. 2. 2001. pp. 1-18. (An evidence-based review of the medical literature that assesses the available evidence for antimicrobial prophylaxis following animal bites.)

Migliore, E, Serraino, C, Brignone, C. ” infection in a cirrhotic patient: case report, microbiological aspects and a review of the literature”. Adv Med Sci. vol. 54. 2009. pp. 109-12. (Case report and literature review of around 100 cases of Pasteurella infections in individuals with cirrhosis.)

Nakwan, N, Nakwan, N, Atta, T, Chokephaibulkit, K. “Neonatal pasteurellosis: a review of reported cases”. Arch Dis Child Fetal Neonatal. vol. 94. 2009. pp. F373-6. (Detailed review of 25 cases of neonatal infections, mainly septicemia and meningitis, nearly all with a history of animal exposures; some infections by vertical transmission.)

Nelson, SC, Hammer, GS. ” empyema: case report and review of the literature”. Am J Med Sci. vol. 281. 1981. pp. 43-9. (Review of the literature for this uncommon presentation.)

Szpak, CA, Woodard, BH, White, JO, Zwadyk, P. “Bacterial peritonitis and bacteremia associated with “. South Med J. vol. 73. 1980. pp. 801-3. (Review of case reports of intra-abdominal Pasteurella infections.)

Stevens, DL, Bisno, AL, Chambers, HF. “Practice guidelines for the diagnosis and management of skin and soft tissue infections”. Clin Infect Dis. vol. 41. 2005. pp. 1373-1406. (These are evidence-based guidelines that address the empirical antimicrobial approach to animal bites taking into consideration the likelihood of Pasteurella.)

Weber, DJ, Wolfson, JS, Swartz, MN, Hooper, DC. ” infections: report of 34 cases and review of the literature”. Medicine (Baltimore). vol. 63. 1984. pp. 133-54. (This is the first major review of the literature that defined the epidemiology and clinical spectrum of Pasteurella infections in humans.)