Final Diagnosis -- Infection with Pasteurella and Enterobacteracae




Dog and cat bites account for approximately 1% of all emergency room visits annually in the United States. The majority of dog bites usually occur in children with the highest rate seen in boys between the ages of 9 and 10 years. Most of the incidents are caused by animals that are known to the victim (5). During the visit to the emergency room it is important to perform a complete physical examination to document the presence or absence of wounds with associated signs of wound infection or joint penetration.

Wound infections following dog bites are caused by a variety of microorganisms. Isolates recovered from clinical cultures of dog bite wounds are usually complex and include Staphylococcal Aureus, alpha, beta, gama hemolytic Streptococcus, several genera of gram-negative organisms and a number of anaerobic microorganisms that are usually part of the normal mouth flora of dogs. One study of 107 patients with dog or cat bites used a reference laboratory for all specimens and recovered a median of five isolates per culture (range 0 - 16) as compared to a median of one isolate (range 0 - 5) in clinical laboratories (1). In the case presented above 3 isolates were recovered from the dog bite wound (2 aerobic and one anaerobic) (Figs 1, 2, 3 and 4).

Patients who have Pasteurella inoculated via a bite wound develop rapid onset of pain, erythema and edema locally. This may occur within hours of the bite or maybe delayed by several days. Broadly speaking human infection with Pasteurella species can be divided into 3 types (i) infection occurring after animal bites (usually dogs or cats); (ii) infection occurring after animal exposure (vetenarians, farmers, live stock handlers, pet owners and food handlers); (iii) infection with no known animal contact (usually lung infections). Most reported cases of Pasteurella infections in humans are caused by P.multocida and involve skin and soft tissue with occasional development of regional lymphadenopathy. Beyond skin and soft tissue, other sites of infection are uncommon and have been the subject of individual case reports or small case series (they include bone and joint infections, CNS infections, septicemia, endocarditis, respiratory tract infections and intra-abdominal infections).

Pasteurella species is one of the most common pathogens isolated from dog and cat bites (50 and 75% respectively), since they inhabit the oral cavity and gastrointestinal tract of these animals (5). In the case presented above Pasteurella multocida was isolated from the dog bite wound. Although found more often in cat bites, Pasteurella species have been isolated from half of dog bites contradicting the impression that this is an uncommon pathogen in dog bites (2).

Pasteurella was first isolated from birds with cholera in 1878 and two years later characterized by Pasteur. The Pasteurella species belongs to the family of Pasteurellaeceae. On a 24-hour blood agar plate, P. multocida are 1-2 mm in diameter and nonhemolytic. It appears as smooth, iridescent, blue and watery mucoid colonies. However the definitive diagnosis of P.multocida is done by indole production, ornithine decarboxylase and acidification of maltose and sucrose (3). They are non-motile, facultative anaerobic gram-negative coccobacilli measuring 1 to 2 m in length (Fig. 5). DNA hybridization studies have determined that they are closely related to Actinobacillius species. The majority of strains that are fermentative are indole, catalase, oxidase and sucrose positive. Many of the pathogenic isolates are also encapsulated. The organism grows in culture in a variety of commercially available media, which include sheep blood, and chocolate agar media. In order to identify the Pasteurella species the oxidase reaction is the most reliable (3). Holst and associates characterized and speciated 159 strains of Pasteurella recovered from clinical specimens of 146 patients over a period of 3 years. The majority of infections were caused by 5 different species or sub species - P. multocida ssp multocida, P. multocida ssp septica, Pasteurella canis, Pasteurella stomatis and Pasteurella dogmatis (2).

The specific mechanism of the pathogenesis of Pasteurella has been studied in animals. They are believed to adhere to the mucosal epithelium of the upper respiratory tract especially the tonsils. The adherence in some of the cases is mediated by fimbriae, which is especially noted in the toxigenic strains (3). The Pasteurella species also produces several virulence factors and toxins (e.g. leukotoxin). They therefore impair cellular response and stimulate inflammatory response. In addition most virulent strains of Pasteurella produce polysaccharide capsules which are antiphagocytic and inhibit intracellular killing by neutrophils (3).

The humoral response to P.mulocida has been characterized. Antibodies to somatic and capsular antigen determinants develop within 2 weeks of clinical infection. Capsular antibody is more long lasting than somatic antibodies. However the precise role of these antibodies in human beings is not clear (3).

The other organism isolated from the dog bite wound in the above case was an Enterobacter cloacae. The Entrobacter organisms are mobile and are less encapsulated. They do not produce H2 S on triple sugar iron medium and are indole and methyl red negative. They are VP positive and can grow in the presence of KCN. They also use citrate as the sole carbon source and can ferment lactose. Decarboxylation of the diamino acids, lysine, arginine and ornithine are used to differentiate various species of Enterobacteracae ( E. cloacae, E. aerogenoes, E. agglomerans and K. pneumoniae) (9).

While Enterobacter sp are most often recognized as opportunistic pathogens of hospitalized patients especially those treated with antibiotics and associated with burn wounds, respiratory and urinary tract infections. Typically, nosocomial Enterobacter cloacae strains are multiple antibiotic-resistant. Isoates associated with animals and soil are typically not multiply resistant. The susceptibility pattern of the E. cloacae isolated from this patient strongly suggests that this strain was of animal origin.

The sensitivity of E.cloacae isolate from the above case to various antibiotics was tested by Kirby Bauer method and showed the following:

The above sensitivity report suggest that this is clearly an animal isolate, because nosocomial isolates are typically more resistant.

In the management of dog bites, cultures of fresh uninfected appearing wounds are not helpful and should not be performed. Studies have shown that cultures obtained immediately following a bite injury correlate poorly with subsequent cultures taken when evidence of clinical infection is present (4).

Management of uninfected bite wounds consists of rabies prophylaxis and tetanus immunizations (if indicated) (10). Antimicrobial prophylaxis after animal bites has remained a controversial subject. A few small trials have been done but none large enough to reach significant conclusions to unequivocally speak of the efficacy of such therapy. However the use of antibiotic prophylaxis for dog bite wounds is controversial. One meta analysis of eight randomized trials found an absolute benefit of one infection prevented for every 14 patients treated (6). Nonetheless most experts recommend a short course of oral antimicrobial therapy with Ampicillin-sulbactam or amoxicillin-clavulanate especially following cat bites because of the high incidence of infection (50%) (7).

When giving antibiotics therapy should be directed against the polymicrobial infection which frequently occurs following animal bite wounds. Clinical evidence of wound infection after a few hours following a bite injury is indicative of the presence of P.multocida among the infecting flora. In such instances some of the authors suggest that cultures are unnecessary and polymicorbial coverage directed against the mouth flora of dogs and cats should be administered empirically (8).

In the above case the patient required treatment with oral antibiotics (Amoxicillin-clavulanate) and Bacitracin cream because of early signs of local infection and situation of the bite which was on the face. A sensitivity panel was done for E.cloacae (above table) and MICU for P.multocida (sensitive to pencillin) yielded a reading of 0.19mg/ml. The patient recovered uneventfully.


  1. Talan DA, Citron DM, Abrahamian FM, Moran GJ, Goldstein EJC: Bacteriologic analysis of infected Dog and Cat Bites. New Eng J Med 1999; 14:1 - 12.
  2. Holst E, rolkof J, Larsson L. Characterization and distribution of Pasturella species recovered from infected humans. J. Clin Microbiol 1992; 30: 2984-2987.
  3. Murray PR, Jobaron E, Pfaller MA, Tenover FC, Yolken RH. Manual of clinical Microbiology. 7th ed. ASM.
  4. Goldstein EJC. Bite wounds and infection. Clin Infect Dis 1992; 14:633.
  5. Weiss HB, Friedman DI, Coben JH. Incidence of Dog bites injuries treated in emergency departments. JAMA 1998; 279:51.
  6. Cummings P. Antibiotics to prevent infection in patients with dog bite wounds: A meta -analysis of randomized trials. Ann Emerg Med 1994;23:535.
  7. Callaham M. Prophylactic antibiotics in dog bite wounds: Nipping at the heels of progress. Ann Emerg Med 1994; 23:577.
  8. Stucker FJ, Shaw GY, Boyd S. Management of animal and human bites in the head and neck. Arch Otolaryngol Head Neck Surg 1990; 116:789.
  9. Koneman EW, Allen SD, Janda WM, Schreckenberger PC, Winn WC. Diagnostic Microbiology. 5th ed. Lippincott Williams & Wilkins. 1997.
  10. Rabies prevention United states,1999. Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 1999;48 (RR-1), January 8,1999.

Contributed by Leena Lourduraj, MD, Eileen Driscoll, MT(ASCP), and A. William Pasculle, ScD

Case IndexCME Case StudiesFeedbackHome