Pasteurella infections

INTRODUCTION — Pasteurella are small gram-negative coccobacilli that are primarily commensals or pathogens of animals. However, these organisms can cause a variety of infections in humans, usually as a result of cat scratches, or cat or dog bites or licks.

Pasteurella infections will be reviewed here. Other infections related to dog and cat bites or scratches are discussed separately. (See "Animal bites (dogs, cats, and other mammals): Evaluation and management" and "Zoonoses: Dogs" and "Zoonoses: Cats".)

MICROBIOLOGY — The members of the genus Pasteurella are small, nonmotile, nonspore-forming, gram-negative organisms. In Gram-stained specimens, they generally appear as a single bacillus, often with bipolar staining, but may also be seen in pairs or short chains (picture 1) [1].

Pasteurella spp are facultatively anaerobic and grow well at 37°C on 5 percent sheep blood (the preferred culture medium), chocolate, or Mueller-Hinton agar; growth is uncommon on MacConkey's agar. Colonies of Pasteurella spp are 1 to 2 mm in diameter after 24 hours of growth at 37°C and are opaque and grayish [2]. A slight greening underneath the colonies may be noted. Most strains recovered from clinical specimens are catalase, oxidase, indole, sucrose, and ornithine decarboxylase positive. The indole-positive species exhibit a mouse-like odor. Media containing vancomycin, clindamycin, and/or amikacin have been used to select for Pasteurella [1]. Potential bacterial virulence factors include the capsule, lipopolysaccharide, sialidases, hyaluronidase, surface adhesins, iron acquisition proteins [3], and the Pasteurella multocida toxin [4].

Human infections have been reported from P. multocida (the most common pathogen and type species for the genus, which includes P. multocida subsp multocida, P. multocida subsp septica, and P. multocida subsp gallicida), Pasteurella canis, Pasteurella dagmatis, and Pasteurella stomatis [2,5]. All are associated with dogs and cats. P. multocida isolates are classified based on a combination of capsular polysaccharide serotyping, which distinguishes isolates into one of five capsular serogroups (ie, A, B, D, E, and F); serotypes A and D account for most human disease [6]. Related species include Pasteurella aerogenes, Pasteurella bettyae, Pasteurella caballi, and Pasteurella pneumotropica [1]. Polymerase chain reaction plus sequence-based ribotyping analysis using universal primers for the 16sRNA gene or rpoB gene sequencing have now superseded phenotypic methods for the identification, characterization, and differentiation of P. multocida and other Pasteurella spp [2,6]. Matrix-assisted laser desorption ionization-time-of-flight mass spectrometry can also be used for accurate identification [2]. The complete genome sequence of an avian clone of P. multocida [7] and the type strain P. multocida subsp multocida ATCC 43137 have been determined [7].

EPIDEMIOLOGY

Animal hosts — Pasteurella spp are found worldwide [3]. P. multocida is a component of the normal upper respiratory tract flora of fowl and mammals. Cats and dogs have the highest carriage rate of P. multocida at 70 to 90 percent and 20 to 50 percent, respectively [8]. Other Pasteurella spp can be found in the oral cavity of a variety of animals, including dogs (P. canis), cats and dogs (P. dagmatis, P. stomatis), pigs and hamsters (P. aerogenes), and horses (P. caballi).

Pasteurella can cause a variety of diseases in animals, such as fowl cholera in domestic fowl, shipping fever in cattle (P. multocida is a cause of secondary infection following parainfluenza infection), hemorrhagic septicemia in cattle and lambs, fibrinous pneumonia in cattle, snuffles in rabbits, and other focal infections [9].

Transmission — Wild and domestic animals are the main reservoirs for human infection [2,9]. Pasteurella spp are the most frequent isolates from dog bites and cat scratches and bites (approximately 50 and 75 percent, respectively) [10-12]. Human infection with Pasteurella also may result from bites of other felines (eg, tigers, lions, panthers, cougars), horses, pigs, rats, rabbits, wolves, and other animals [13]. (See "Animal bites (dogs, cats, and other mammals): Evaluation and management".)

Although most human infections are caused by dog or cat bites, nonbite transmission has been described [13], such as infection resulting from animal licks [14-17]. Licking of distal areas of nonintact skin by cats and dogs have led to seeding of proximal prosthetic joints such as a total knee [18] or hip [19] arthroplasty. Kissing animals has been reported as the source in Pasteurella in several cases [20]. Sharing food items with a dog (popsicles) has led to P. multocida bacteremia [21]. Indirect transmission through fomites has also been suggested, as illustrated by cases of infants with severe P. multocida infections linked to contact with pet saliva from a pacifier or a sibling's finger [22,23]. A recent review suggested that nonbite transmission was more common in older individuals and those with comorbidities and resulted in more life-threatening infections than bites [13].

Molecular analysis has documented animal-to-human transmission both by direct and indirect contact. P. multocida pneumonia in a 75-year-old woman was linked to her dog by pulsed-field electrophoresis analysis [24]. Meningitis in an immunocompetent 25-year-old woman was linked to her dog by matrix-assisted laser desorption ionization-time-of-flight mass spectrometry [25]. P. multocida subsp septica meningitis in a one-day-old boy was linked to the family cats by repetitive-sequence-based polymerase chain reaction [26].

Human-to-human horizontal transmission has also been documented, including transmission via close contact with colonized individuals and transmission via contaminated blood products [27,28]. Vertical transmission can occur via transplacental infection, endometritis, or genital tract colonization [23,29].

Risk factors for severe disease — Severe infections can occur in previously healthy individuals. However, individuals with impaired host defenses or immunocompromising conditions (including those with cirrhosis, hematologic malignancies, or solid organ transplantation) are at particular risk. Additionally, serious infections are more likely at the extremes of life (ie, in neonates [29] or with advanced age [30]).

In a study that compared features of P. multocida infections with or without history of an animal bite, infection without a bite history was associated with more severe comorbidity and immunocompromising conditions, as well as more severe disease, including bacteremia, need for intensive care unit management, and mortality [30]. An immunocompromising condition has been reported as a risk factor for hospitalization following a cat [31] or dog bite [32]. Comorbidities that increase the risk of particular Pasteurella infections include a prosthetic joint for septic arthritis, underlying lung disease for pneumonia, and peritoneal dialysis for peritonitis.

CLINICAL MANIFESTATIONS — Pasteurella can cause serious soft tissue infections and, less commonly, septic arthritis, osteomyelitis, sepsis, and meningitis, particularly in infants [33] and other immunocompromised hosts [32,34]. Infections with P. multocida can be divided into three categories (table 1) [35]:

●Soft tissue infections including cellulitis following animal bites or scratches; bites and scratches can also result in abscesses and necrotizing soft tissue infections

●Bone and joint infections

●Respiratory infections, usually in the setting of chronic pulmonary disease

●Serious invasive infections often unrelated to animal bites, such as meningitis, intra-abdominal infection, endocarditis, or ocular infection

Soft tissue infections — Clinical soft tissue infections with P. multocida usually occur after cat bites, cat scratches, or dog bites [12,15,35-37], but may also occur following cat or dog licks of nonintact skin [14-16]. P. multocida wound infections characteristically have a very rapid development of an intense inflammatory response. Most patients develop symptoms within 24 hours of the initial injury, and as early as three hours after a cat bite [35].

Pain and swelling are prominent. Purulent drainage is noted in about 40 percent of patients, lymphangitis in about 20 percent, and regional adenopathy in 10 percent. Cellulitis often occurs within 24 to 48 hours [35,38,39]. Necrotizing fasciitis may also occur [15,40]. (See "Necrotizing soft tissue infections".)

Bone and joint infections — P. multocida can cause septic arthritis and/or osteomyelitis. Most cases of septic arthritis involve a cat or dog bite distal to the involved joint, without direct penetrating injury to the joint; approximately one-third of cases are not preceded by an animal bite or scratch. Septic arthritis without osteomyelitis most commonly involves a single joint, usually the knee. There is a predilection for involvement of prosthetic joints or joints damaged by rheumatoid arthritis or degenerative joint disease [18,41-44]. More than 50 percent of patients who have septic arthritis also have altered host defenses from diabetes mellitus, corticosteroids, or alcoholism. (See "Septic arthritis in adults".)

Osteomyelitis results from either local extension of soft tissue infection or direct inoculation into the periosteum by the cat or dog bite. Osteomyelitis usually is preceded by significant wound infection and cellulitis. Cat bites are more likely to lead to bone infection than are dog bites, presumably because the small, sharp teeth of the cat more easily penetrate the periosteum.

The combination of Pasteurella septic arthritis and osteomyelitis almost always occurs after cat bites of the hand, resulting in osteomyelitis of the phalanx and interphalangeal arthritis.

Respiratory infections — P. multocida can cause a variety of upper and lower respiratory tract infections including glossitis, pharyngitis, sinusitis, otitis media, mastoiditis, epiglottitis, tracheobronchitis, pneumonia, empyema, and lung abscess [30,35,45-49]. The majority of patients with P. multocida respiratory tract infections have underlying lung disease such as chronic obstructive lung disease or chronic bronchitis [49]. As an example, chronic pulmonary infection has been reported in a patient with bronchiectasis [50].

Pneumonia is the most common pleuropulmonary infection caused by P. multocida, with most cases occurring in older adults [35,51]. Pneumonia may be a primary event or secondary to bacteremic spread [49]. In one series of 108 patients with Pasteurella pleuropulmonary infection, 49 had pneumonia, 37 tracheobronchitis, 25 empyema, and 3 lung abscess [52]. Bacteremia was found in 55 percent of the patients in whom blood cultures were obtained. The overall mortality in this series was 29 percent. An underlying disease was found in 93 percent of patients.

The clinical course of Pasteurella respiratory tract infections is nonspecific. Common symptoms include fever, malaise, dyspnea, and pleuritic chest pain [35,45]. Approximately two-thirds of patients have fever. The onset of the illness may be gradual or abrupt. In patients with pneumonia, chest examination often reveals localized findings such as dullness, rhonchi, or wheezes. The chest radiograph most commonly demonstrates lobar consolidation, but multilobar and diffuse infiltrates have been described.

Other serious invasive infections — Pasteurella spp can cause a variety of other serious invasive infections such as meningitis, bacteremia, endocarditis, and peritonitis. Sepsis and septic shock may result from P. multocida infection [53-55] and present with purpura fulminans [56,57]. Serious invasive infection has been reported after exposure to animals in the absence of a bite or a scratch [13,14,54,58-64]. In addition, P. multocida can rarely be associated with in utero infection leading to fetal death [65], acute suppurative thyroiditis [66], or urinary tract infection [67]. P. multocida infections of aortic and/or femoral vascular grafts [68-70] and hemodialysis grafts [71] have been reported.

Most of the uncommon manifestations of P. multocida infection have been noted in case reports (table 1):

●Meningitis — P. multocida meningitis is a disease of the extremes of life [14,17,20,23,25,26,28,35,58-62,72,73]. Fifty percent of cases have involved infants younger than one year of age, and 30 percent have involved adults older than 60 years [35,60,74]. Mechanisms by which P. multocida can cause central nervous system infection include direct inoculation with an animal bite (eg, a perforating bite to an infant's skull), contamination from contiguous infected wounds after trauma or neurosurgery, extension from an adjacent infected site by spread through lymphatics or veins, and bacteremic seeding of the meninges [74,75].

The cerebrospinal fluid (CSF) generally contains polymorphonuclear leukocytes with a low glucose and elevated protein, similar to other pyogenic infections. Gram stain of the CSF is positive in 80 percent of patients, but the organisms are frequently misidentified as Haemophilus influenzae or Neisseria meningitidis. (See "Health care-associated meningitis and ventriculitis in adults: Clinical features and diagnosis".)

●Bacteremia — Bacteremia with P. multocida usually accompanies a localized infection, most commonly cellulitis. Bacteremia occurs in 25 to 50 percent of patients with pneumonia, meningitis, and septic arthritis due to P. multocida. However, it may also occur in the absence of a localized site of infection [35,53].

Many patients with bacteremia have evidence of a predisposing condition, such as significant liver dysfunction, diabetes mellitus, solid organ transplantation, malignancy, or advanced age [34,35,53,76-82]. As an example, in a comprehensive review of the literature (119 cases), a major comorbidity was reported in 67 percent of patients with P. multocida bacteremia [81]. On multivariate analysis, the only factor associated with mortality was a major comorbidity. However, sepsis and septic shock have also been described in previously healthy individuals [83,84].

The overall incidence of bacteremia is low. A population-based study from Queensland, Australia reported an overall annual incidence of bacteremia of 3.3 per million residents [82]. The incidence was highest in more recent years. The overall all-cause case fatality rates have ranged from 9 to 31 percent [81,82].

●Endocarditis — Endocarditis is a rare complication of P. multocida sepsis, with about 30 cases of P. multocida endocarditis reported in the literature [85-87]. Risk factors include older age, male sex, and comorbidities, especially heart disease [87]. Infection may involve native or prosthetic valves.

●Abdominal infections — Intra-abdominal infections caused by P. multocida include peritonitis and appendicitis [76,88-95].

Peritonitis due to P. multocida, especially in the setting of continuous ambulatory peritoneal dialysis, has been reported [88-91,96-99]. In these cases, P. multocida was isolated along with more traditional intra-abdominal pathogens. Gram stain of the peritoneal fluid was negative in approximately one-half of the cases.

Cats were the source of almost all peritoneal infections. In most of the cases associated with peritoneal dialysis (approximately two-thirds), the dialysis tubing had been punctured by the cat.

DIAGNOSIS

Clinical suspicion and evaluation — Pasteurella spp are the first organisms to consider in any patient who presents with a soft tissue infection following cat scratches, or cat or dog bites or licks. Such infections are characterized by rapid onset and an intense inflammatory response, similar to group A streptococcal (Streptococcus pyogenes) soft tissue infections. (See 'Soft tissue infections' above.)

Cat bite wounds tend to penetrate deeply, with a higher risk of associated osteomyelitis, tenosynovitis, and septic arthritis than dog bites. Additionally, septic arthritis in abnormal or prosthetic joints may follow distal animal bites or scratches. Clinicians should maintain a low threshold of suspicion for deeper infection in these scenarios. The possibility of Pasteurella spp should also be considered when patients with impaired host defenses (eg, liver insufficiency, diabetes mellitus, malignancy) present with serious infections (eg, pulmonary infections, meningitis, or spontaneous bacterial peritonitis) and a history of animal contact.

In these clinical infections, culture of the relevant clinical specimen is warranted. In particular, in wounds from animal scratches or bites, Gram stain and aerobic and anaerobic cultures from the depth of the puncture or laceration should be obtained, if possible. Further evaluation of animal bite wounds is discussed in detail elsewhere. (See "Animal bites (dogs, cats, and other mammals): Evaluation and management".)

Microbiological diagnosis — The diagnosis of P. multocida infection is made by isolation of the organism in culture. The organism grows readily on sheep blood agar, which is routinely used in the clinical laboratory. (See 'Microbiology' above.)

P. multocida may be isolated from respiratory tract samples of patients with pleuropulmonary disease. Because P. multocida respiratory tract infections are rare and have no distinctive characteristics, P. multocida may not be suspected as the infecting pathogen. In addition, Pasteurella may be mistaken for H. influenzae, Moraxella catarrhalis, Neisseria spp, or Acinetobacter spp on sputum Gram stain. The organism may also be misidentified initially on Gram stains of other deep culture specimen, such as cerebrospinal fluid. However, when recovered in culture, Pasteurella should be recognized as a serious pathogen and the patient treated accordingly.

TREATMENT — Specific treatment for suspected or known P. multocida infection should, in general, use one of the antibiotic options recommended by the Clinical and Laboratory Standards Institute (CLSI) for susceptibility testing, as described below [100]. For monomicrobial infections, penicillin is the drug of choice.

Antibiotic options — Pasteurella spp, including P. multocida, are usually susceptible to a number of antibiotics (table 2) [101-106]. These include penicillin, amoxicillin-clavulanate, piperacillin-tazobactam, doxycycline, fluoroquinolones (eg, levofloxacin, moxifloxacin), third- or later-generation cephalosporins (eg, cefpodoxime, cefixime, ceftriaxone, ceftaroline), carbapenems (eg, imipenem, meropenem), and trimethoprim-sulfamethoxazole. There are no clinical trials documenting or evaluating the efficacy of different antibiotic agents for Pasteurella infections, specifically. Recommendations on antibiotic selection are based on observational evidence from case reports and series and the expected in vitro susceptibility pattern of Pasteurella species.

For monomicrobial infections due to Pasteurella, penicillin is often the drug of choice because of its narrow spectrum, low cost, general safety, and extensive experience with its use. Other first-line agents include ampicillin, amoxicillin, and cefuroxime. Although penicillin resistance is uncommon, it has been described in a subset of cases. In a report of 44 patients with P. multocida infections cultured between 2000 and 2014, of the 32 isolates that had β-lactamase testing reported, 16 percent were β-lactamase-positive [30]. Presence of a beta-lactamase was not associated with the type or severity of infection [30].

Clinical failures have been noted in patients treated with oral erythromycin, semi-synthetic penicillins (eg, oxacillin, dicloxacillin), first-generation cephalosporins (eg, cephalothin, cephalexin, cefadroxil), and clindamycin. These agents have poor in vitro activity against P. multocida and should be avoided. Emergent antibiotic resistance (ie, to penicillin G, ampicillin, tetracycline) has been reported in a patient with chronic pulmonary infection [50].

For soft tissue bite wound infections, susceptibility testing is generally not necessary. However, in light of reports of beta-lactamase production, testing of isolates from normally sterile sources (blood cultures, deep tissue, implanted prosthetic devices) and respiratory specimens for beta-lactamase activity may be warranted, particularly in immunocompromised patients [100]. Additional reasons to perform susceptibility testing include [107]:

●Persistent infection

●Clinical failure of initial antibiotic therapy

●Allergy to or intolerance of the drugs of choice

●Possible resistance to a drug that is being considered for therapy

When indicated, antibiotic susceptibility testing should follow the guidelines of the CLSI [100]. P. multocida susceptibility break points have been provided for the following agents:

●Penicillins (penicillin, ampicillin, amoxicillin, amoxicillin-clavulanate)

●Cephalosporins (ceftriaxone)

●Quinolones (levofloxacin, moxifloxacin)

●Tetracyclines (tetracycline, doxycycline)

●Macrolides (erythromycin, azithromycin)

●Others (chloramphenicol, trimethoprim-sulfamethoxazole)

Animal bite wound infections

Empiric antimicrobial and general management — Empiric antimicrobial therapy, whether administered by mouth or intravenously, should be directed at the polymicrobial infection that frequently occurs after animal bites, including Pasteurella spp and also oral anaerobes, streptococci, Capnocytophaga canimorsus, and Staphylococcus aureus. We suggest amoxicillin-clavulanate for initial empiric therapy of animal-associated wounds for those who can take oral therapy. Infection with P. multocida resistant to amoxicillin-clavulanate is rare but has been reported [108]. Intravenous therapy is warranted in patients who cannot tolerate or fail oral therapy, or who have evidence for sepsis, metastatic spread, or deeper infections (such as septic arthritis or osteomyelitis). We suggest ampicillin-sulbactam for intravenous empiric therapy for patients with mild infection and agents with broader coverage (eg, piperacillin-tazobactam, a carbapenem, or a third-generation cephalosporin plus metronidazole) for patients with more severe infections. Doses and other potential regimens are listed in the tables (table 3 and table 4) and discussed in detail elsewhere. (See "Animal bites (dogs, cats, and other mammals): Evaluation and management", section on 'Management'.)

Other general management of P. multocida soft tissue infection associated with an animal bite includes evaluation for complications (including neurovascular or other deep injury), irrigation of large wounds, and determination of the need for tetanus or rabies postexposure prophylaxis [36]. These issues are also discussed in detail elsewhere. (See "Animal bites (dogs, cats, and other mammals): Evaluation and management" and "Tetanus" and "Rabies immune globulin and vaccine".)

Definitive therapy — Definitive antimicrobial therapy should be based upon the results of wound cultures.

Skin and soft tissue infection — Most soft tissue infections following an animal bite are polymicrobial [10], and the definitive antibiotic regimen should cover other pathogens that are isolated on wound culture in addition to Pasteurella (table 3). Amoxicillin-clavulanate generally is appropriate to complete therapy for soft tissue infections following an animal bite. Even if cultures cannot be acquired or Pasteurella is not isolated on culture, we still use a regimen that includes coverage against Pasteurella, and avoid monotherapy with cephalexin, clindamycin, or macrolides.

When Pasteurella is isolated, susceptibility testing is generally not necessary given the typical susceptibility pattern of the organism. However, because of reports of beta-lactamase-producing isolates, we favor beta-lactamase testing for high-risk populations (ie, immunocompromised hosts). If Pasteurella is the only pathogen isolated, treatment with penicillin VK (50 mg/kg per day divided in four doses for children beyond the newborn period; maximum adult dose 500 mg per dose four times per day [total dose 2 g per day]) is appropriate if beta-lactamase production has not been identified. Amoxicillin (25 to 50 mg/kg per day divided in three doses for children beyond the newborn period; 500 mg three times daily for adults) is also an effective agent, and for children who use oral suspensions of medications, its taste may be better tolerated. Amoxicillin-clavulanate is appropriate in the setting of beta-lactamase production. Options for penicillin-allergic patients are discussed elsewhere. (See 'Penicillin-allergic patients' below.)

Therapy for local infections is usually 7 to 10 days, and for more severe infections, 10 to 14 days [33]. Wound drainage or debridement may be necessary. Most cases of Pasteurella soft tissue infection following an animal scratch or bite resolve with appropriate oral antibiotics and wound drainage, when indicated.

Septic arthritis and osteomyelitis — When Pasteurella is isolated in cases of septic arthritis or osteomyelitis, we favor susceptibility testing to evaluate for beta-lactamase production. If there is no beta-lactamase, these infections can be treated with high doses of parenteral antibiotics (eg, penicillin G, 3.5 to 4.0 million units every 4 hours for adults or 200,000 units/kg per day in 4 to 6 divided doses for children, with a maximum daily dose of 24 million units). In the presence of beta-lactamase production or other coinfecting pathogens, ampicillin-sulbactam, piperacillin-tazobactam, broad-spectrum cephalosporins (eg, ceftriaxone), and quinolones (eg, levofloxacin) are other options. Options for penicillin-allergic patients are discussed elsewhere. (See 'Penicillin-allergic patients' below.)

Septic arthritis often warrants multiple aspirations or debridement of the joint for control of the infection. In prosthetic joint involvement, removal of the prosthesis is usually required for resolution.

Antimicrobial therapy is generally continued for four to six weeks for bone and joint infections [33]. In some cases, if clinically appropriate, an oral regimen can be used to complete the course following improvement on the initial parenteral regimen. This decision is similar to that for bone and joint infection in general and is discussed in detail elsewhere. (See "Hematogenous osteomyelitis in children: Management", section on 'Switch to oral therapy' and "Septic arthritis in adults", section on 'Duration' and "Nonvertebral osteomyelitis in adults: Treatment", section on 'Duration of therapy' and "Bacterial arthritis: Treatment and outcome in infants and children", section on 'Oral therapy'.)

Other invasive infections — Most positive cultures of blood, deep tissues, or respiratory tract yield only Pasteurella in isolation [30]. We favor testing Pasteurella isolates from such sterile or respiratory sites for beta-lactamase production. For isolates that do not produce a beta-lactamase, high-dose intravenous penicillin G can be used for therapy. Alternatives for beta-lactamase-producing infections include ampicillin-sulbactam, piperacillin-tazobactam, and broad-spectrum cephalosporins (eg, ceftriaxone). For patients with meningitis, antibiotic selection should also take into account penetration of the agent into the cerebrospinal fluid, and we typically recommend a third generation cephalosporin (eg, ceftriaxone). Options for penicillin-allergic patients are discussed elsewhere. (See 'Penicillin-allergic patients' below.)

Data on the management of these other Pasteurella infections are limited to case reports. Duration of therapy is generally the same as for other infections at the same site. For example, pneumonia is generally treated for five to seven days in the absence of bacteremia, and for at least 14 days in the setting of bacteremia. Gram-negative rod meningitis warrants 21 days of therapy and repeated lumbar puncture to ensure improvement. Success with six weeks of therapy has been reported with Pasteurella endocarditis [85]; however, valve replacement may be required for cure.

Durations of therapy longer than those listed here may be warranted in the case of slow response or particularly severe disease.

Penicillin-allergic patients — Alternative regimens for patients with penicillin allergies depend on the history and type of allergy.

For patients with reactions to penicillin that are not immediate hypersensitivity reactions (ie, not hives or anaphylaxis) to penicillin, broad spectrum cephalosporins (eg, cefixime or cefpodoxime for oral therapy and ceftriaxone for parenteral therapy) are alternatives. These drugs are active against Pasteurella species in vitro, but there is limited clinical experience using them for the treatment of Pasteurella [33].

For patients with a history of immediate hypersensitivity reactions to penicillin, the clinical approach to deciding on the use of cephalosporins or carbapenems is discussed in detail elsewhere. (See "Allergy evaluation for immediate penicillin allergy: Skin test-based diagnostic strategies and cross-reactivity with other beta-lactam antibiotics".)

Options include using cephalosporins or carbapenems with a graded challenge (ie, starting at a very low dose and administering a 10-fold increased dose at intervals until the therapeutic dose is reached) or after desensitization to the chosen agent with the assistance of an allergy specialist. P. multocida has been reported to be highly susceptible to carbapenems (imipenem, meropenem, ertapenem) [101,103,106], but there is limited experience using these agents. Otherwise, nonbeta-lactam alternatives that typically have in vitro activity against Pasteurella include aztreonam, quinolones (eg, levofloxacin or moxifloxacin), trimethoprim-sulfamethoxazole, and doxycycline.

PROGNOSIS — The prognosis for Pasteurella infections depends upon the site of infection and underlying medical conditions.

●Most soft tissues infections resolve with appropriate oral antibiotics and wound drainage, when indicated.

●Approximately 50 percent of patients with osteomyelitis experience slow healing, nonunion, joint fusion, limitations of motion, or residual deformity.

●Functional outcome of P. multocida infections of the hand is poor [109].

●Mortality in Pasteurella meningitis, bacteremia, and endocarditis are approximately 25 percent, 30 percent, and 30 percent, respectively.

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Beyond the Basics topic (see "Patient education: Animal and human bites (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Microbiology and risk factors – Pasteurella are small Gram-negative coccobacilli that can cause a variety of infections in humans, usually as a result of cat scratches, or cat or dog bites or licks. Pasteurella multocida is the most common pathogenic species. Severe infections can occur in previously healthy individuals. However, individuals with immunocompromising conditions (including those with cirrhosis, hematologic malignancies, and solid organ transplantation) are at particular risk. (See 'Epidemiology' above.)

●Cutaneous manifestations – Pasteurella wound infections are characterized by the rapid development of an intense inflammatory response within 24 hours of the initial injury. Pain and swelling are prominent. Purulent drainage, lymphangitis, and regional lymphadenopathy also may be present. (See 'Soft tissue infections' above.)

●Infections other than of the skin – Pasteurella can also cause septic arthritis, usually at a joint proximal to an animal bite, and/or osteomyelitis, either from local extension of soft tissue infection or direct inoculation. Other invasive infections include pneumonia and other respiratory tract infections, meningitis, bacteremia, endocarditis, and peritonitis (table 1). (See 'Bone and joint infections' above and 'Respiratory infections' above and 'Other serious invasive infections' above.)

●Diagnosis – Pasteurella infection should be considered in any patient who presents with a soft tissue infection following cat scratches, or cat or dog bites or licks. The diagnosis of Pasteurella infection is made by isolation of the organism in culture; the organism grows readily on sheep blood agar. (See 'Diagnosis' above.)

●Management

•Wound management – Treatment of Pasteurella soft tissue infection associated with an animal bite should begin with management of the bite wound.

•Empiric antimicrobial selection – Empiric antimicrobial therapy for infected animal wounds should include an agent active against P. multocida and other potential copathogens. For patients with mild infection, we suggest amoxicillin-clavulanate for oral therapy and ampicillin-sulbactam for intravenous therapy (for those unable to tolerate oral therapy) (Grade 2C). For patients with more severe infection, we suggest empiric therapy with an agent with broader coverage (such as piperacillin-tazobactam, a carbapenem, or a third-generation cephalosporin with metronidazole) (Grade 2C). Doses and other potential regimens are listed in the tables (table 3 and table 4). (See 'Empiric antimicrobial and general management' above.)

•Definitive antimicrobial therapy – Definitive therapy for Pasteurella infection should be based upon the results of wound cultures. Susceptibility testing of isolates in soft tissue infections is generally not necessary. We perform beta-lactamase testing in infection in immunocompromised hosts and on isolates from normally sterile sources and respiratory specimens. (See 'Antibiotic options' above and 'Definitive therapy' above.)

For definitive therapy of monomicrobial infections due to Pasteurella, we suggest penicillin (Grade 2C). Amoxicillin, ampicillin, beta-lactam-beta-lactamase inhibitor combinations, extended spectrum cephalosporins, quinolones, trimethoprim-sulfamethoxazole, doxycycline, and azithromycin are potential alternatives (table 2). Semi-synthetic penicillins (eg, oxacillin or dicloxacillin), first-generation cephalosporins, clindamycin, and erythromycin have poor activity and should be avoided. (See 'Definitive therapy' above and 'Antibiotic options' above and 'Penicillin-allergic patients' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges William A Rutala, PhD, MPH, who contributed to an earlier version of this topic review.