Version August 2025
INTRODUCTION —
Tularemia is a zoonotic infection caused by Francisella tularensis, an aerobic and fastidious gram-negative bacterium. Human infection occurs following contact with infected animals or invertebrate vectors. Synonyms include Francis disease, deer-fly fever, rabbit fever, market men disease, water-rat trappers disease, wild hare disease (yato-byo), and Ohara disease [1]. The clinical manifestations of Francisella infection may range from asymptomatic illness to septic shock and death, in part depending on the virulence of the infecting strain, portal of entry, inoculum, and the immune status of the host [1].
The clinical manifestations, diagnosis, treatment, and prevention of tularemia will be reviewed here. The microbiology, pathogenesis, and epidemiology of infection due to F. tularensis are discussed separately. (See "Tularemia: Microbiology, epidemiology, and pathogenesis".)
CLINICAL MANIFESTATIONS
Initial nonspecific symptoms — Tularemia usually starts with nonspecific systemic symptoms, including fever, chills, anorexia, and malaise, which occur approximately 3 to 5 days (range 1 to 21 days) following exposure. The onset of these symptoms can be either abrupt or indolent. The fever may abate after a few days but then quickly return. Other nonspecific symptoms include headache, fatigue, soreness in the chest or muscles, abdominal pain, emesis, or diarrhea. In some patients, these systemic symptoms may have waned by the time of evaluation.
When patients do come to medical attention, they usually have specific clinical manifestations associated with one of the six major clinical forms of tularemia, depending on the portal of entry [2]:
●Ulceroglandular tularemia
●Glandular tularemia
●Oculoglandular tularemia
●Pharyngeal (oropharyngeal) tularemia
●Pneumonic tularemia
●Typhoidal tularemia
These syndromes are discussed separately in detail below, although overlapping manifestations may be present.
Some features may vary by geographic location. Pulse-temperature dissociation (eg, relative bradycardia in the setting of a fever) was reported in only 5 percent of patients in a series from Sweden but was reported in 42 percent of patients in a series from the United States [2,3]. It is unclear if this difference is related to subtype; F. tularensis subspecies holarctica predominates in Sweden, whereas both subspecies tularensis and holarctica occur in the United States, but the study from the United States did not specify the infecting subtype. The different subtypes (ie, subspecies) of F. tularensis are discussed elsewhere. (See "Tularemia: Microbiology, epidemiology, and pathogenesis", section on 'Francisella species'.)
Clinical syndromes
Ulceroglandular disease — Ulceroglandular disease, characterized by a skin lesion and associated adenopathy, is the most common and most easily recognizable form of tularemia [1,2,4]. As an example, in a review of 1163 tularemia cases reported to the United States Centers for Disease Control and Prevention (CDC) between 2006 and 2021, ulceroglandular disease was the most common clinical form, reported in 42 percent [5]. In an earlier review of 190 tularemia cases in Missouri between 2000 and 2007, it was the most common clinical form among adults but was second to glandular disease among children [6]. Among 177 tularemia patients identified in France between 2008 and 2017, ulceroglandular disease was also the most common presentation; all were infected with F. tularensis subspecies holarctica [7]. (See "Tularemia: Microbiology, epidemiology, and pathogenesis".)
Patients with ulceroglandular disease usually report recent animal contact or exposure to potential insect vectors (particularly ticks). They typically present with fever and a single erythematous ulcerative lesion at the site of inoculation (eg, the site of a tick bite) (picture 1). Ulcers on the hands and arms are more common following animal exposures; ulcers on the head or neck, trunk, perineum, and legs are more common following tick exposures. Occasionally, more than one skin lesion may be present [2].
Affected patients also have tender regional lymphadenopathy, which can occur before, at the same time, or shortly after the appearance of the skin lesion. Adenopathy involving cervical or occipital nodes is more common in children than adults [8], and the associated ulcers may be hidden in the scalp. The overlying skin of the node can be erythematous, as observed in 19 percent of cases in a series of 215 Swedish patients with infection due to the less-virulent F. tularensis subspecies holarctica [3]. A "sporotrichoid" presentation, or subcutaneous nodules along the draining lymphatics, has also been described in some patients with tularemia [9]. However, frank lymphangitis is not usually seen; its presence should suggest the uncommon complication of bacterial superinfection of the skin ulcer.
Suppuration of affected lymph nodes is a relatively common complication and may occur despite antibiotic therapy, particularly among patients with delayed recognition and treatment. In a review of tularemia cases in Missouri, 15 of 81 patients (19 percent) with lymphadenopathy required incision and drainage of suppurative nodes [10]. Recurrent lymph node suppuration despite treatment has been described in a patient with tularemia who had been treated with an anti-tumor necrosis factor (TNF) agent and methotrexate [11]. Suppurated, fluctuant nodes warrant surgical or needle drainage. (See 'Adjunctive management' below.)
Patients seeking medical attention relatively late in the course of disease may have adenopathy with little or no fever and only evidence of a healed skin lesion.
Glandular disease — Glandular tularemia refers to tender regional lymphadenopathy involving single or multiple nodes, in the absence of an identifiable skin lesion. It is a relatively common presentation of tularemia. Among 1163 tularemia cases reported to the CDC between 2006 and 2021, glandular disease was the second most common presentation, reported in 16 percent [5]. In an earlier series from Missouri, it was the most common among children (44 percent of cases) [6]. Among 177 tularemia patients identified in France between 2008 and 2017, glandular disease was also the second most common presentation [7].
Glandular disease is transmitted via the same mechanism as ulceroglandular disease, and the clinical features of the associated adenopathy are the same, but in glandular disease, there is no evident lesion at the site of inoculation. Suppurative lymph nodes can also occur with glandular disease. (See 'Ulceroglandular disease' above.)
Oculoglandular disease — Oculoglandular tularemia refers to infection involving the eye and accounts for a small percentage of tularemia cases.
It occurs when F. tularensis gains access to the conjunctiva, either via splashing infected material into the eye, rubbing the eyes with contaminated fingers, or by infected aerosols. Eye symptoms are usually unilateral and include pain, photophobia, and increased tearing. Eye examination demonstrates conjunctival erythema with edema and vascular engorgement. Some patients may have conjunctival purulence, small conjunctival ulcers or nodules, and periorbital erythema and/or edema [12]. Tender regional adenopathy may be present in the preauricular, postauricular, cervical, and submandibular regions. Parinaud oculoglandular syndrome specifically refers to conjunctivitis in one eye and swollen lymph nodes in front of the ear on the same side; F. tularensis is one cause of this syndrome. (See 'Differential diagnosis' below.)
Complications include corneal ulceration and dacryocystitis. Tularemia also has been associated with other less common ocular manifestations, including a case of unilateral uveitis [13,14].
Suppurative lymph nodes can also occur with oculoglandular disease. (See 'Ulceroglandular disease' above.)
Pharyngeal (oropharyngeal) disease — Pharyngeal tularemia involves the mouth and throat and accounts for a small percentage of cases in the United States. However, pharyngeal disease accounts for a larger percentage of cases in other parts of the world, particularly related to use of well water or during outbreaks in the setting of war or natural disaster [15].
It results from an oropharyngeal portal of infection, usually ingestion of contaminated food or water. Transmission can also occur from oral exposure to contaminated droplets or by hand-to-mouth exposure (eg, in the setting of finger contamination from crushing ticks or handling contaminated animals).
The major symptoms are fever, severe sore throat, and swelling in the neck [16]. Examination demonstrates an exudative pharyngitis and tonsillitis, cervical lymph node enlargement, and occasionally pharyngeal or tonsillar ulcers. Pre-parotid and retropharyngeal lymph nodes also may be enlarged and tender. In addition, a pharyngeal membrane mimicking diphtheria can occur [17].
Pneumonic disease — Pneumonic tularemia refers to a clinical presentation dominated by pulmonary involvement. Pneumonic disease is more common in adults but can affect any age group. Of 1163 cases of tularemia reported to the CDC between 2006 and 2021, pneumonic disease accounted for 16 percent [5,18]. Pneumonic disease caused by F. tularensis subspecies tularensis (prevalent in North America) is generally more severe than that caused by subspecies holarctica (prevalent in other parts of the world) [19,20]; however, subspecies holarctica may also cause severe pneumonia, particularly in immunocompromised patients [21].
Pneumonic disease can be categorized as primary or secondary, based on the route of transmission.
●Primary pneumonic disease results from direct inhalation of the organism into the lungs. Occupations at particular risk for primary disease include farmers, landscapers, and laboratory workers.
Following the initial nonspecific symptoms (eg, fever, headache, malaise, myalgias, nausea, and anorexia), fevers, chest pain, and cough with scant sputum production become more pronounced [22,23]. Patients sometimes complain of substernal or pleuritic chest pain. As an example, in one series of 128 patients with tularemia in Oklahoma, all 32 patients with pneumonic disease had fever upon presentation (range 101.5 to 107°F) [24]. Most (72 percent) had nonproductive cough, 31 percent had pleuritic chest pain, and 25 percent had a pleural effusion.
Findings on chest examination include rales, signs of consolidation, and a friction rub or evidence of pleural fluid.
Early after inhalational exposure, the chest radiograph may be normal, but abnormalities usually develop as respiratory findings become more prominent [22]. Common radiographic changes include peribronchial infiltrates, lobar consolidation, pleural effusion, and hilar adenopathy. Rounded infiltrates and cavitation from pneumonic tularemia are uncommon, although the presence of nodular infiltrates with a pleural effusion should raise concern for tularemic pneumonia or pneumonic plague. (See "Clinical manifestations, diagnosis, and treatment of plague (Yersinia pestis infection)", section on 'Clinical manifestations'.)
●Secondary pneumonic disease results from hematogenous spread to the lung and can complicate any of the other major forms of tularemia [2,24].
The clinical presentation of secondary pneumonic tularemia is varied.
Secondary pneumonia can present with bilateral disease, involvement of the lower lobes, and/or with miliary disease. There can be pulmonary infiltrates, pleural effusion, or both. In one series, some patients with secondary lung involvement had abnormal chest radiographs but no clinical evidence of pneumonia [2]. Pulmonary nodules, pleural effusion, and mediastinal adenopathy have been described in a patient with typhoidal tularemia who had been treated with the anti-TNF agent infliximab [25].
Pleural effusions in pneumonic tularemia are exudative with a lymphocytic predominance and may have an elevated adenosine deaminase level [26]. Pleural or lung biopsies can demonstrate granuloma formation and therefore be confused with pulmonary tuberculosis [1,26]. Empyema requiring decortication has been reported [10].
Respiratory failure requiring mechanical ventilation and the adult respiratory distress syndrome can result from either primary or secondary pneumonic tularemia. In one series of 128 patients with tularemia, the 32 patients with pneumonic disease had a higher fatality rate compared with those without pulmonary involvement. They were also older, less likely to recall a vector exposure, more likely to require hospitalization and have a longer hospital stay, and more likely to have positive cultures (from blood or other specimens) [24].
Typhoidal disease — Typhoidal tularemia is a systemic febrile illness without prominent regional adenopathy or other localizing signs that does not fit another major form of the disease. Typhoidal disease is a common presentation in certain locations. Of 1163 cases of tularemia reported to the CDC between 2006 and 2021, typhoidal disease was the fourth most common form, reported in 13 percent [5]. However, it was the most common tularemia presentation among cases reported in Arkansas from 2009 through 2013 and was particularly frequent among older patients [27].
Typhoidal disease may result from any portal of entry, but the source is usually inapparent at the time of presentation. Affected patients often have chronic underlying conditions.
The clinical presentation ranges from acute sepsis to a chronic febrile illness. Major symptoms include fever, chills, anorexia, headache, myalgias, sore throat, abdominal pain, and diarrhea. Prominent physical findings may include evidence of intravascular volume depletion, mild pharyngitis, and diffuse abdominal tenderness.
Occasionally, localizing findings can be present. A clinical presentation with predominant abdominal symptoms has been referred to as "abdominal tularemia," potentially from ingestion of the pathogen; mesenteric adenopathy can be present. Enlargement of the liver and spleen is more likely to be detectable with a longer duration of illness. Pulmonary involvement secondary to hematogenous spread is seen in up to 45 percent of cases, although some of these cases could have been subclinical pneumonic disease that was not recognized until later [1]. (See 'Pneumonic disease' above.)
Potential laboratory findings in severe typhoidal tularemia include elevated creatine phosphokinase (CPK), myoglobinuria, hyponatremia, and renal failure.
Presentation in immunocompromised patients — Immunocompromised patients with tularemia usually have fever with or without any of the nonspecific symptoms described above (see 'Clinical syndromes' above). They may be more likely to present with pneumonic or typhoidal illness.
A review of 17 immunocompromised individuals with tularemia reported fever in 94 percent, sweats or fatigue in 36 percent, respiratory symptoms in 41 percent, and abdominal symptoms in 24 percent [28]. Eight patients (48 percent) presented with pneumonic tularemia, 5 (29 percent) had typhoidal tularemia, and only 4 (24 percent) had ulceroglandular or glandular disease.
Other features
Secondary skin manifestations — Secondary skin changes are common in all forms of tularemia, reported in up to 50 percent in some series, and are often misdiagnosed or overlooked [3,4,29-31]. These secondary eruptions are usually maculopapular, vesiculopapular, erythema multiforme, erythema nodosum, or urticarial; some have been mistaken for varicella or drug eruptions [30]. Sweet syndrome (acute febrile neurophilic dermatosis, which typically presents as tender, edematous, and inflamed papules, plaques, and nodules) also has been reported to occur with tularemia [31]. More than one type of eruption can occur in the same patient [32].
Laboratory findings — Routine laboratory tests are nonspecific. The white blood cell count may be low, normal, or elevated. Other nonspecific findings may include elevated inflammatory markers, low platelet count, low serum sodium, abnormal liver enzymes, evidence of rhabdomyolysis or myoglobinuria, and pyuria.
Complications — If untreated, tularemia can cause prolonged fever, weight loss, adenopathy, and debility that can last for weeks or months [17]. Even with appropriate treatment, some patients will have a lengthy recovery following tularemia.
Patients with prolonged tularemia often complain of fatigue and lassitude, and may have anorexia, weakness, and weight loss. Neuropsychiatric complaints include headache, difficulty concentrating, and disturbed sleep [33]. Many of these patients have had suppurative lymph nodes, a common complication when lymph nodes are involved (see 'Ulceroglandular disease' above). Risk factors for a poor outcome include older age, serious underlying disease, a delay in correct diagnosis, prolonged symptoms prior to treatment, pneumonic or typhoidal disease, renal failure, and inadequate antibiotic treatment [24,34].
Other complications include sepsis, renal failure, rhabdomyolysis, and hepatitis [2,34]. Rarely, F. tularensis infection may cause otitis media and mastoiditis, endocarditis, pericarditis, myocarditis, meningitis, osteomyelitis, peritonitis, granulomatous hepatitis, splenic hematoma, spontaneous splenic rupture, aortitis, or prosthetic joint infection [1,7,35-41]. All four patients identified in one literature review of F. tularensis endocarditis initially presented with typhoidal disease [37]. F. tularensis subspecies holarctica infection of a bioprosthetic valve occurred in a patient presenting with prolonged fever and a resolving skin lesion [42,43]. One report described a patient whose only manifestation of tularemia was pericarditis; the diagnosis was made serologically [44].
Meningitis, reported with ulceroglandular and typhoidal disease, can develop 3 to 30 days after the onset of illness and cause a cerebrospinal fluid mononuclear cell pleocytosis with low glucose and high protein [45-48]. Meningitis developed in a patient with fever and rash after he ran his lawn mower over a dead rabbit [47]. Other rare neurological manifestations attributed to tularemia include Guillain-Barré syndrome and isolated cranial nerve abnormalities [49,50].
Potential bioterrorism use — F. tularensis is a Tier 1 bioterrorism agent (ie, of highest concern for bioterrorism use), as classified by the CDC, in part because of its low infectious dose, high associated mortality, and potential for easy dissemination.
A bioterrorist attack with F. tularensis would most likely employ aerosolization of the organism to do the most harm to the most people [51]. Such an attack would most likely result in an outbreak of inhalational tularemia three to five days later, marked by an acute, undifferentiated febrile illness with predominant manifestations of pneumonia, pleuritis, and hilar lymphadenopathy [51].
Rapid recognition and reporting of a possible bioterrorist event due to tularemia are thus difficult clinical challenges. The possibility would be suggested by clustered cases of pneumonic or typhoidal disease, particularly in urban areas in patients without the expected epidemiologic exposures to animals, insects, or environmental activities.
DIAGNOSIS
Clinical suspicion — Tularemia should be suspected in patients with a compatible clinical syndrome and epidemiologic risk factors. Because laboratory confirmation may be delayed, the initial diagnosis of tularemia is often made presumptively when the patient's presentation is both clinically and epidemiologically consistent and other causes have been excluded.
Specific clinical features that should prompt consideration for tularemia include:
●Regional lymphadenopathy, particularly if associated with an inoculation site
●Conjunctivitis accompanied by local lymphadenopathy
●Severe pharyngitis that is unresponsive to penicillin and undiagnosed after routine testing
●Persistent systemic febrile illness that is undiagnosed after routine testing
●Community-acquired pneumonia that is unresponsive to standard antibiotic therapy and undiagnosed after routine testing
●Nodular infiltrates plus a pleural effusion on chest imaging
When these clinical features are observed in the setting of a history of animal (particularly wild animal) exposure or insect bites, the possibility of tularemia is greater, and making the presumptive diagnosis is reasonable. In particular, people who are farmers, veterinarians, hunters, park service employees, landscapers, meat handlers, or laboratory workers are at increased risk for exposure. The patient's location, activities, and travel history should also inform the likelihood of tularemia. Tularemia has been reported throughout the northern hemisphere but is less common in the United Kingdom. Human Francisella infections have also been reported in Australia. In the United States, tularemia is most commonly reported in the south-central states, the Pacific Northwest, and parts of Massachusetts (figure 1). Clusters of cases, particularly of infections consistent with pneumonic tularemia in the absence of typical exposures, should raise suspicion for the possibility of a bioterrorism event. (See "Tularemia: Microbiology, epidemiology, and pathogenesis", section on 'Epidemiology'.)
The diagnosis of tularemia requires a high index of suspicion, as the exposure history or epidemiologic risk may not be evident, and certain signs, such as fever or lesions around an inoculation site, may have abated by the time of presentation.
Meningitis is a rare complication of tularemia. Patients with suspected tularemia who have progressive headache, signs of meningeal irritation, or altered mental status warrant evaluation for meningitis with lumbar puncture. (See "Clinical features and diagnosis of acute bacterial meningitis in adults", section on 'Cerebrospinal fluid examination'.)
Microbiologic diagnosis
Approach — The diagnosis of tularemia can be confirmed either by a positive culture or by serology. When tularemia is clinically suspected, serology for F. tularensis should be submitted at the time of presentation and again at least two to four weeks after presentation. This is because it takes at least two weeks after infection for antibodies to Francisella to be detectable, and diagnostic rises in convalescent antibody titers do not appear until at least two to four weeks after the onset of symptoms. The diagnosis is confirmed with a fourfold or greater change in titer from the initial to convalescent serology. Relevant patient specimens (such as a swab of an ulcer, blood, specimens from fluctuant or necrotic lymph nodes, respiratory specimens, and tissue biopsies) should also be sent for culture with specific instructions to the laboratory that tularemia is suspected (so laboratory workers can take appropriate precautions to avoid getting ill from processing specimens); cultures are diagnostic if positive but do not rule out the possibility of tularemia if negative. Molecular and immunohistochemical tests on such specimens can rapidly identify F. tularensis while awaiting confirmation, but these tests may not be specific and are not considered confirmatory. (See 'Serology' below and 'Culture' below and 'Molecular and other testing' below.)
In the event of a suspected bioterrorist event using F. tularensis, steps to alert the appropriate authorities should be taken. In the United States, this includes infection control personnel and health care facility administration, the Federal Bureau of Investigation field office, local police, the Centers for Disease Control and Prevention (CDC), and medical emergency services [52]. Specimens can be submitted to a specialized laboratory in the Laboratory Response Network for assistance with rapid diagnosis. (See "Tularemia: Microbiology, epidemiology, and pathogenesis".)
Serology — The diagnosis of tularemia is confirmed serologically by detecting a fourfold or greater change in titers of antibodies to F. tularensis between acute and convalescent serum specimens [53]. Tube agglutination titers of 1:160 or higher or microagglutination titers of 1:128 or higher are considered positive [53]. The results of serologic testing should always be interpreted in the context of the clinical suspicion for tularemia. Serologic studies should be performed only in patients in whom tularemia is a realistic possibility; they should not be used as a screening test among febrile patients.
A diagnostic increase in antibody titer generally occurs two to four weeks after the onset of symptoms. Antibody titers are not reliably positive until after at least two weeks of infection, so they are rarely helpful in the acute setting. Both immunoglobulin (Ig)M and IgG antibodies appear together following the initial infection, and both antibody titers may remain elevated for years after an infection. Thus, a single positive titer is supportive of the diagnosis but may also result from an old infection [54]. Serologic assays for tularemia can cross-react with heterophile antibodies and antibodies to other gram-negative organisms such as Brucella or Legionella, but cross-reactions are typically positive at a low, nondiagnostic titer [1].
In the United States, serologic studies have historically been performed using a tube agglutination or microagglutination assay; however, enzyme-linked immunosorbent assays (ELISAs) have more recently become more common. ELISAs are more sensitive but less specific than agglutination assays and are qualitative, meaning diagnosis cannot be confirmed by fourfold increase in titer. Indirect immunofluorescence assays (IFAs) are also occasionally used but are less specific than the other two options. Serologic tests cannot identify the specific infecting F. tularensis subspecies. Investigational methods to improve serodiagnostic testing are being pursued [55,56].
Culture — Positive cultures can confirm the diagnosis of tularemia. Thus, it is appropriate to submit relevant clinical specimens for culture when tularemia is suspected. Depending on the clinical presentation, relevant specimens include blood, lymph node drainage or biopsy specimens, skin lesion swabs or biopsy specimens, pleural fluid, respiratory specimens (eg, bronchial washings), and pharyngeal or ocular swabs. The laboratory should be notified prior to submitting specimens for culture to optimize growth conditions as well as to take proper precautions to reduce the risk of infection among laboratory personnel. (See "Tularemia: Microbiology, epidemiology, and pathogenesis", section on 'Laboratory precautions'.)
When F. tularensis is seen on Gram stain of clinical specimens, it appears as weakly stained, tiny gram-negative coccobacilli (picture 2). Routine cultures may be negative because the organism is slow-growing and fastidious. In addition, many routine solid media do not contain cysteine, which F. tularensis requires for growth. Culture growth is facilitated by use of supportive media. Clinicians should be aware that laboratories that use matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MS) for isolate identification may not be able to identify or may misidentify F. tularensis; although successfully identified by MALDI-TOF systems in research laboratories, F. tularensis is not included in commercial databases labeled for diagnostic use [57].
Other details regarding the growth and identification of F. tularensis are found elsewhere. (See "Tularemia: Microbiology, epidemiology, and pathogenesis", section on 'Laboratory features'.)
Molecular and other testing — Polymerase chain reaction (PCR) assays can be performed on human specimens for the rapid presumptive diagnosis of tularemia while awaiting confirmation. These tests are generally not available for routine testing in most hospital or clinical labs but in the United States can be obtained through state public health laboratories and the Laboratory Response Network. Appropriate samples may include ulcer swabs, lymph node aspirates, blood, respiratory specimens (eg, bronchial washings), pleural fluid, spinal fluid, or tissue biopsies. Francisella PCR assays offer the potential advantages of a more rapid diagnosis than serologic tests or cultures, a greater sensitivity than smears or cultures, a more limited exposure of laboratory personnel to the potential hazards of processing cultures, and the availability of the basic methodology in many clinical laboratories. PCR can also be useful for the diagnosis of patients with prolonged illness and in those already given antibiotic treatment.
Sequence-based PCR methods, including metagenomic sequencing, have also been successfully used to diagnose tularemia [58]. However, clinicians should be aware that some commonly used molecular methods, including 16S sequencing, may be unable to distinguish F. tularensis from the closely related opportunistic pathogen Francisella novicida.
Other specific techniques for the rapid presumptive diagnosis of tularemia have been developed, including direct fluorescent antibody (DFA) staining of clinical specimens and immunohistochemical staining of tissue [1,53]. However, these methods are not commercially available.
Rapid tests to detect multiple potential bioterrorism pathogens simultaneously are being explored, as are genomic, proteomic, metabolomic, and immunologic methods for diagnosis.
DIFFERENTIAL DIAGNOSIS —
The differential diagnosis of tularemia is broad and depends on the predominant clinical syndrome.
●Fever and lymph node enlargement (ulceroglandular and glandular disease) – Many other infectious and noninfectious etiologies can cause fever and regional lymphadenopathy. Important infectious etiologies include streptococcal or staphylococcal lymphadenitis, cat scratch disease (Bartonella infection), sporotrichosis, toxoplasmosis, fungal or mycobacterial infection, Spirillum minus rat bite fever, anthrax, plague, syphilis, and other sexually transmitted infections [1]. Staphylococcal and streptococcal infections are more common than tularemia and often include frank cellulitis and perhaps purulence.
As with tularemia, cat scratch disease, sporotrichosis, toxoplasmosis, S. minus rat bite fever, anthrax, and plague are also associated with recent exposure to the outdoors, animals, or insects; all are uncommon infections. Skin ulcers are more common with tularemia and anthrax than with cat scratch disease or plague, a necrotic ulcer with surrounding induration and edema strongly favors anthrax, and the rapid onset of tender buboes suggests plague.
Noninfectious causes, such as malignancy or a necrotic spider bite, can also cause similar symptoms [59].
The evaluation of regional lymphadenopathy is discussed elsewhere. (See "Evaluation of peripheral lymphadenopathy in adults", section on 'Evaluation' and "Peripheral lymphadenopathy in children: Etiology", section on 'Localized lymphadenopathy'.)
●Conjunctival disease (oculoglandular disease) – When patients present with unilateral conjunctivitis associated with swollen lymph nodes in front of the ear on the same side (Parinaud oculoglandular syndrome), other potential etiologies include cat scratch disease (Bartonella infection) and herpes simplex infection. Other more common causes of conjunctivitis include adenoviral infection and pyogenic bacterial infection. (See "Infectious conjunctivitis", section on 'Specific types of infectious conjunctivitis'.)
●Severe pharyngitis (pharyngeal disease) – More common causes of pharyngitis are adenovirus, infectious mononucleosis, and streptococcal pharyngitis. (See "Evaluation of acute pharyngitis in adults", section on 'Infectious causes'.)
●Pneumonia, pulmonary infiltrates (pneumonic disease) – Clinical symptoms and radiographic findings are not sufficiently specific to distinguish tularemic pneumonia from other causes of community-acquired pneumonia. Among patients with apparent community-acquired pneumonia who have negative cultures and fail to respond to routine therapy, other diagnoses to consider include Coxiella infection, psittacosis, mycobacterial infection, pulmonary mycoses, and pneumonic plague [1]. (See "Nonresolving pneumonia", section on 'Misdiagnosis of pathogens'.)
Pneumonic tularemia may also be mistaken for lung cancer, particularly when infectious causes are not considered and positron emission tomography (PET)/computed tomography (CT) scans are positive [60].
●Fever of unknown origin (typhoidal disease) – The differential diagnosis of fever of unknown origin is broad. Other culture-negative systemic infections without localizing features include typhoid fever, brucellosis, Coxiella infection, tick-born relapsing fever, culture-negative endocarditis, malaria, rickettsioses, anaplasmosis, ehrlichiosis, and viral illnesses. (See "Fever of unknown origin in adults: Etiologies".)
TREATMENT —
Antimicrobial therapy (table 1) should be administered promptly to all patients with suspected or confirmed tularemia. Although spontaneous resolution of infection in the absence of specific treatment has been recorded [2], early effective treatment is associated with less morbidity. Since the introduction of effective antibiotics (in particular streptomycin), historical mortality rates from tularemia have decreased from as high as 60 percent in severely ill patients with pneumonic or typhoidal disease to less than 5 percent overall [19,24,51,54].
Active antibiotics — Antimicrobials with well-established clinical efficacy include the fluoroquinolones ciprofloxacin and levofloxacin, doxycycline, tetracycline, the aminoglycosides streptomycin and gentamicin, and chloramphenicol. These agents exhibit achievable minimal inhibitory concentrations (MICs) when tested using a standardized in vitro method against F. tularensis [61-64]. Natural resistance to aminoglycosides, fluoroquinolones, or tetracycline in human isolates has not been reported [64]. However, the majority of 29 human isolates of F. tularensis subspecies holarctica in Spain were resistant to tigecycline [65]. In a study from France, there was no fluoroquinolone resistance among 42 F. tularensis subspecies holarctica isolates and no molecular evidence of DNA gyrase mutations (which would confer fluoroquinolone resistance) among 82 tissue samples from patients with tularemia, including those who had a suboptimal outcome with fluoroquinolone treatment [66].
Other agents do not have a routine role in treatment. Beta-lactams have been associated with clinical failure despite favorable in vitro susceptibilities in some cases [67]. Although successful use of azithromycin and erythromycin has been reported and isolates sampled in the United States have low MICs to erythromycin [64], they are not considered first-line therapy, and resistant strains are prevalent in parts of Europe and Russia [1,54,68,69].
Regimen selection — Our approach to regimen selection depends on the severity of infection, as below. In general, this treatment approach is based on observational data, which primarily consist of case series and reports [5,70,71]. No prospective controlled clinical trials have compared the efficacy of different drug regimens in recent decades or clearly defined the optimal duration of therapy for tularemia.
Drug classes that are highly effective for tularemia are the aminoglycosides, fluoroquinolones, and tetracyclines. Overall, use of an agent from one of these classes has been associated with better survival compared with use of other agents (adjusted odds ratio [OR] 10.4, 95% CI 4.4-24.5 in one analysis of over 900 cases reported in the United States) [5].
Given that natural resistance to these agents has not been reported, obtaining susceptibility testing is not clinically necessary when planning to use one of them.
Severe infection — For patients with severe infection, we suggest gentamicin (given intramuscularly or intravenously). Streptomycin (given intramuscularly or intravenously) is also an appropriate option but is not readily available in many locations. Aminoglycosides are the drugs of choice for such patients, as there is the most successful clinical experience with these agents. Severe infection includes prolonged or extensive systemic symptoms prior to therapy, sepsis with or without renal failure in any form of tularemia, typhoidal tularemia, and pneumonic tularemia with dyspnea or respiratory distress. For patients who cannot take aminoglycosides, a fluoroquinolone is an alternative.
We also use an aminoglycoside or fluoroquinolone when empiric treatment for tularemia is indicated in patients with an uncertain diagnosis who require hospitalization. (See 'Adults' below.)
Patients with rare complications, such as meningitis or endocarditis, are treated initially with combination therapy, ideally in consultation with an expert in infectious diseases. This is discussed separately. (See 'Meningitis or endocarditis' below.)
The doses are outlined in the table (table 1). The duration of aminoglycoside treatment is generally 7 to 10 days; for children, some experts suggest 10 to 14 days [72]. Ultimately, the duration should be tailored to clinical signs and symptoms, including resolution of fever, and should be extended (eg, to 14 days) for especially severe cases or for patients whose response to treatment is delayed. Adults who respond to therapy can also transition to an oral regimen to complete a course of treatment. (See 'Adults' below.)
Aminoglycosides have been associated with higher survival and cure rates and fewer relapses when treatment is initiated in a timely manner. As an example, the fatality rates among patients with tularemia who received aminoglycoside monotherapy was 1 percent (2 of 178 patients) in a systematic review of global cases published between 1993 and 2023 [71] and 0 percent (0 of 69 patients) in an analysis of cases reported to the Centers for Disease Control and Prevention (CDC) in the United States from 2006 to 2021 [5]. In contrast, fatality rates among those who received therapy that did not contain a fluoroquinolone, aminoglycoside, or tetracycline ranged from 8 to 18 percent. In an earlier review of case reports and series, the cure rate among 244 patients who received streptomycin for tularemia was 97 percent, with no relapses (table 1) [70]. Among the 36 patients who received gentamicin, the cure rate was 86 percent, and there were two relapses. Subsequent case series have reported similar or higher cure rates with gentamicin [10,73]. Extended-interval (eg, once-daily) gentamicin dosing has been used successfully and is a more convenient option for outpatient therapy [22,74,75].
Streptomycin has traditionally been the preferred aminoglycoside because of extensive experience supporting its use, its high efficacy, and the fact that, in the United States, it is approved for the treatment of tularemia by the US Food and Drug Administration [70]. However, gentamicin is equally effective and more readily available than streptomycin. In addition, timely blood levels are usually more readily obtained for gentamicin than streptomycin, and gentamicin has less vestibular toxicity. Thus, gentamicin has become the preferred aminoglycoside for the treatment of tularemia [72].
Some experts have recommended that severe disease be managed with a combination of an aminoglycoside and a fluoroquinolone, although this has not been proven to be superior to an aminoglycoside alone [76].
Mild or moderate infection
Adults — Initial oral treatment is reasonable for adult patients who can be managed reliably as outpatients and for hospitalized patients without severe disease. For such patients, we suggest an oral fluoroquinolone (eg, ciprofloxacin or levofloxacin) for 7 to 10 days or doxycycline for 14 to 21 days. An oral agent may also be appropriate to complete treatment in patients who responded to initial parenteral therapy. Doses are listed in the table (table 1).
Ciprofloxacin and levofloxacin are highly active in vitro. Both agents have been used successfully in all tularemia syndromes, including pneumonic tularemia, although there is more published experience with ciprofloxacin [5,10,15,71,77-79]. The fatality rates among patients with tularemia who received fluoroquinolone monotherapy were 1.2 percent (2 of 167 patients) in a systematic review of global cases of tularemia published between 1993 and 2023 [71] and 1.3 percent (2 of 150 patients) in an analysis of United States cases (including both subspecies tularensis and holarctica) reported to the CDC with sufficient clinical information from 2006 to 2021 [5]. In contrast, fatality rates among those who received therapy that did not contain a fluoroquinolone, aminoglycoside, or tetracycline ranged from 8 to 18 percent. Cure rates are also good with fluoroquinolones [10,15,80]. In the study of oropharyngeal tularemia in Turkey, 27 of 53 patients (49 percent) who were treated with doxycycline had treatment failure compared with 25 percent failure rates for aminoglycosides and fluoroquinolones; however, most patients in this case series (82 percent) experienced treatment delays of >14 days [15]. In another series that included 22 patients who received ciprofloxacin as initial therapy for tularemia, treatment was successful in 21 (95 percent); additionally, ciprofloxacin resulted in cure in 30 of 34 patients who received it after nonresponse to other regimens [80].
Doxycycline is another good option and has the advantage of activity against other tick-borne infections in cases where the diagnosis is not initially clear. Tetracyclines have also been associated with low fatality rates [5,71]. Some studies suggest that the effectiveness of doxycycline is comparable to that of fluoroquinolones [81]. However, the data are mixed, and in some case series, tetracyclines have been associated with a higher rate of treatment failure than fluoroquinolones [3,15,80]. In the study of oropharyngeal tularemia in Turkey, 27 of 53 patients who were treated with doxycycline had treatment failure [15]. In another series that included 13 patients treated with at least 14 days of doxycycline (alone or in combination with ineffective agents), 15 percent had treatment failure [10]. In contrast, the rate of treatment failure with doxycycline in a series of tularemia cases from France was 10 percent among 90 patients and lower than that with fluoroquinolones, although variable dosing of fluoroquinolones may have impacted outcomes [7]. Tetracyclines have also been associated with relapse and/or need for repeated drainage procedures [70,71,82].
Children — For children with mild or moderate infection, we suggest a fluoroquinolone (ciprofloxacin or levofloxacin) for 10 days or doxycycline for 21 days. Gentamicin is an alternative option; some experts note that the recommended duration is 10 to 14 days but that 7 days is likely sufficient for children with mild disease [72]. Doses are listed in the table (table 1).
Although aminoglycosides have traditionally been first-line therapy for children with tularemia of any severity, and the American Academy of Pediatrics recommends gentamicin as the drug of choice, increasing evidence suggests good outcomes with ciprofloxacin [10,30,71,83], which does not require parenteral administration and is less associated with nephrotoxicity. In a systematic review of global cases of tularemia published between 1993 and 2023, 32 children received monotherapy with a fluoroquinolone (mainly ciprofloxacin), of whom 10 and 19 percent warranted lymph node aspiration or excision, respectively [71]. Recurrent fever occurred in 6 percent. Data for doxycycline in children are more limited, and efficacy is extrapolated from evidence in adults, as discussed elsewhere (see 'Adults' above). Some experts suggest that doxycycline not be employed for definitive treatment because of the paucity of data [72].
Meningitis or endocarditis — Meningitis and endocarditis are rare complications of tularemia. Both should be managed in consultation with an expert in infectious diseases.
●For adults and children with tularemic meningitis, we suggest an aminoglycoside combined with doxycycline or ciprofloxacin [72]. A combination regimen is preferred because cerebrospinal fluid levels of aminoglycosides may be erratic [72]. In general, the duration of treatment for tularemic meningitis is 14 to 21 days but should be tailored to clinical signs and symptoms, including resolution of fever.
Studies have reported successful treatment of tularemic meningitis with streptomycin plus chloramphenicol as well as streptomycin plus doxycycline, gentamicin plus doxycycline, and gentamicin plus ciprofloxacin [45,48,84,85]. However, chloramphenicol should only be used if ciprofloxacin or doxycycline cannot be given and it is available for immediate use. Chloramphenicol should not be used to treat other forms of tularemia because other highly effective options are safer.
●Endocarditis should also be managed initially with a combination regimen of an aminoglycoside plus a fluoroquinolone. Studies informing the optimal treatment of F. tularensis endocarditis are extremely limited, but case reports, including cases of prosthetic valve endocarditis, have documented favorable outcomes with two weeks of gentamicin plus a fluoroquinolone followed by another two to four weeks of a fluoroquinolone [37,42,43].
Specific circumstances
Pregnancy — Tularemia in pregnancy may be associated with prematurity or fetal loss, although the extent of the risk is unknown, and healthy newborns without adverse effects from maternal tularemia also have been reported [17,86-92].
We suggest gentamicin, ciprofloxacin, or levofloxacin as first-line agents for pregnant patients with tularemia. Gentamicin and ciprofloxacin have been effective in a small number of cases [87,89]. Although gentamicin and fluoroquinolones have the potential for fetal toxicity, treatment during pregnancy likely outweighs these risks given the potential associations between tularemia and adverse pregnancy outcomes. Azithromycin was also effective in a pregnant patient in Arkansas, and a prolonged course of azithromycin was effective in a patient in France where erythromycin-sensitive strains of F. tularensis subspecies holarctica predominate [69,88]. F. tularensis subspecies tularensis is generally susceptible to macrolides, whereas only certain strains of F. tularensis subspecies holarctica are susceptible. Antibiotic use in pregnancy is discussed in detail elsewhere. (See "Prenatal care: Patient education, health promotion, and safety of commonly used drugs", section on 'Antibiotics'.)
Immunosuppression — Immunosuppressed patients with tularemia have an increased risk of treatment failure or relapse, and their optimal antibiotic management is unknown. We prefer an aminoglycoside for treatment of tularemia in immunocompromised patients, as used in patients with severe infection (see 'Severe infection' above). Case reports of tularemia patients with various underlying immunosuppressing conditions have documented successful treatment with gentamicin, a fluoroquinolone, or doxycycline, alone or in combination (table 1) [11,25,28,77,93-99]. Treatment may need to be prolonged beyond the usual recommended duration in patients who are slow to respond.
Bioterrorism event — Treatment of tularemia from a bioterrorism event depends on the numbers of ill patients [51]. Contained attacks generally allow individual medical management, whereas resources may be more constrained in the setting of mass casualties.
Individuals infected as a result of a bioterrorism attack can be managed using the same aminoglycoside or oral agents discussed above (table 1) (see 'Severe infection' above and 'Mild or moderate infection' above). However, due to the risk of engineered resistance, we recommend initial treatment with two distinct classes of antimicrobials until additional information on the released F. tularensis strain(s) is available [51].
For mass casualties, oral regimens should be prioritized for most individuals but intravenous or intramuscular regimens can be used for patients with severe infection [51].
Adjunctive management — Even with appropriate antibiotic therapy, enlarged lymph nodes can progress to fluctuance and suppuration. In such cases, incision and drainage of the lymph node are warranted.
Debridement and drainage are also warranted for empyema in the setting of pneumonic tularemia. (See "Epidemiology, clinical presentation, and diagnostic evaluation of parapneumonic effusion and empyema in adults".)
Relapses — Relapses can occur following any regimen but are more common when tetracyclines (bacteriostatic antimicrobials) are used for fewer than 14 days. Retreatment with the initial agent used is reasonable since resistance in clinical isolates has not been reported. If doxycycline was used initially, it can be used again but for a longer time (such as 21 days); alternatively, the patient can be retreated with an aminoglycoside or a fluoroquinolone.
This approach may not be effective for strains released as part of a bioweapon attack, however, since organisms can be engineered for resistance to commonly used agents [51]. (See 'Bioterrorism event' above.)
PREVENTION
Minimizing exposure — Preventive measures include behavioral strategies to minimize the risk of exposure to the organism:
●Not using bare hands to skin or dress wild animals
●Avoiding sick or dead animals
●Wearing masks, eye protection, and gloves when disposing of dead animals, including those brought home by cats and other pets
●Wearing clothing that covers exposed skin and that is tight at the wrists and ankles
●Using insect repellents that are also effective against ticks
●Removing ticks promptly
●Only drinking potable water
●Adequately cooking wild meats
Institutional infection control — Standard precautions are adequate for hospitalized patients with tularemia; person-to-person transmission is extremely rare and has not been reported via the respiratory route. Whenever tularemia is suspected or proven, microbiology laboratory and autopsy personnel handling patient specimens should be notified so that they can take precautions to minimize the likelihood of exposure to F. tularensis. (See "Tularemia: Microbiology, epidemiology, and pathogenesis", section on 'Laboratory precautions'.)
Post-exposure prophylaxis — The type of exposure, the length of time since the exposure, and the patient's concerns should be considered when making the decision about giving post-exposure prophylaxis.
●Exposure to naturally occurring tularemia – We typically do not suggest post-exposure prophylaxis following natural exposure. This includes a tick bite [1,2] or close contact with an individual (including a patient) with tularemia [1,51]. Following such low-risk exposures, we monitor patients for fever or other symptoms of tularemia without prophylaxis. However, laboratory workers, autopsy personnel, or other workers who are exposed through nonintact skin, mucosal surfaces, or aerosols to materials contaminated with F. tularensis may have a higher risk of infection. For such exposure, post-exposure prophylaxis is a reasonable option if initiated early (eg, within a week of exposure), and we individualize the decision to use it based on extent of exposure and underlying comorbidities. Individuals who have had tularemia in the past do not require antibiotic prophylaxis with subsequent exposures. Although recurrent infections have been documented, most recurrences are clinically mild ulceroglandular disease. If post-exposure prophylaxis is used, the regimens are the same as those discussed below.
●Potential bioterrorism event – We suggest post-exposure prophylaxis for adults and children who are exposed to F. tularensis in a bioterrorism event and identified during the early incubation period (eg, one week). Individuals who are only identified after others have become symptomatic with tularemia should be monitored without prophylaxis and preemptively treated if they develop consistent symptoms [51].
If given, post-exposure prophylaxis should be administered as soon as possible following the exposure:
•Regimens for adults are ciprofloxacin 500 mg orally twice daily for 7 days, levofloxacin 500 mg orally once daily for 7 days, or doxycycline 100 mg orally twice daily for 10 to 14 days [51,100].
•Regimens for children are ciprofloxacin 15 mg/kg orally twice daily (not to exceed 1 g daily) for 7 days, levofloxacin 8 mg/kg orally twice daily for those <50 kg (not to exceed 500 mg daily) and 500 mg orally once daily for those ≥50 kg for 7 days, or doxycycline 2.2 mg/kg orally twice daily for those <45 kg and 100 mg orally twice daily for those ≥45 kg for 10 to 14 days [51,72].
Data on post-exposure prophylaxis are limited to small human and animal challenge studies in which antibiotics administered within 24 to 72 hours of experimental inhalational exposure prevented clinical illness [51,101,102].
Investigational vaccines — No tularemia vaccine is currently available. A vaccine prepared from the live vaccine strain (LVS) of F. tularensis subspecies holarctica that was previously used is no longer available because of concerns about its unknown mechanisms of attenuation, stability, and production. Research to develop a new vaccine is actively investigating subunit vaccines, new live strains with defined attenuation, and an improved LVS vaccine [65,103-106].
SUMMARY AND RECOMMENDATIONS
●Microbiology and transmission – Tularemia is a zoonotic infection caused by Francisella tularensis, an aerobic and fastidious gram-negative bacterium. Human infection occurs following contact with infected animals or invertebrate vectors. (See "Tularemia: Microbiology, epidemiology, and pathogenesis".)
●Clinical syndromes – Tularemia usually has an abrupt onset of nonspecific symptoms, such as fever, chills, headache, and malaise following an incubation period of three to five days. Patients usually present with clinical features associated with one of six major syndromes (with occasional overlap) depending on the portal of entry:
•Ulceroglandular tularemia – This is the most common manifestation. Affected patients present with fever and a single erythematous papulo-ulcerative lesion with a central eschar (picture 1) accompanied by tender regional lymphadenopathy. They usually report recent animal handling or exposure to potential insect vectors (particularly ticks). (See 'Ulceroglandular disease' above.)
•Glandular tularemia – This manifests as enlargement of a single or multiple regional lymph nodes in the absence of an identifiable skin lesion. It occurs more often in children than adults. (See 'Glandular disease' above.)
•Oculoglandular and pharyngeal tularemia – These are less common forms of tularemia that result from organism invasion through the conjunctivae and oral mucosa, respectively. Eye findings are usually unilateral and include pain, photophobia, and conjunctival erythema with regional adenopathy; the major features of pharyngeal tularemia are an exudative pharyngitis and tonsillitis with cervical node enlargement. (See 'Oculoglandular disease' above and 'Pharyngeal (oropharyngeal) disease' above.)
•Pneumonic tularemia – This can result from direct inhalation of F. tularensis (primary) or bacteremic spread to the lung (secondary). In primary disease, cough is typically nonproductive. Common findings on chest imaging include peribronchial infiltrates, lobar consolidation, pleural effusion, and hilar adenopathy. Rounded infiltrates and cavitation are uncommon. (See 'Pneumonic disease' above.)
•Typhoidal tularemia – This is a systemic febrile illness without prominent regional adenopathy or other clear localizing signs that does not fit another major form of the disease. The clinical presentation ranges from acute sepsis to a chronic febrile illness. (See 'Typhoidal disease' above.)
●Skin findings – Secondary skin rashes are an underappreciated and relatively common manifestation in all forms of tularemia. These secondary eruptions are usually maculopapular, vesiculopapular, erythema multiforme, erythema nodosum, or urticarial; some have been mistaken for varicella or drug eruptions. (See 'Secondary skin manifestations' above and 'Complications' above.)
●When to suspect tularemia – Tularemia should be suspected in patients with features compatible with one of these clinical syndromes and epidemiologic risk factors. A history of animal (particularly wild animal) exposure or insect bites should heighten suspicion for tularemia; the patient's location (figure 1), activities, and travel history should also inform the likelihood of tularemia. Because laboratory confirmation may be delayed, the initial diagnosis is often made presumptively. Clusters of cases, particularly of infections consistent with pneumonic tularemia in the absence of typical exposures, should raise suspicion for the possibility of a bioterrorism event. (See 'Clinical suspicion' above.)
●Diagnosis – When tularemia is clinically suspected, serology for F. tularensis should be submitted at the time of presentation and again at least two to four weeks later. A single-tube agglutination titer of 1:160 or higher or a single microagglutination titer of 1:128 or higher is supportive of the diagnosis; a fourfold or greater change in titer between acute and convalescent serum specimens is confirmatory. Specimens should also be sent for culture with specific instructions to the laboratory that tularemia is suspected; cultures are diagnostic if positive but do not rule out the possibility of tularemia if negative. (See 'Microbiologic diagnosis' above.)
●Treatment – Antibiotics are the cornerstone of therapy and should be administered promptly to patients with documented or suspected tularemia. Adjunctive management includes incision and drainage of suppurated lymph nodes. (See 'Treatment' above.)
•Severe infection – For adults and children with severe infection, we suggest gentamicin (Grade 2C). Streptomycin has traditionally been the preferred aminoglycoside because of greater experience with this agent, but gentamicin is also effective, is more readily available, and has less associated vestibular toxicity. Fluoroquinolones are a reasonable alternative for severe disease. (See 'Severe infection' above.)
•Mild infection – For adults or children with mild or moderate infection, we suggest an oral fluoroquinolone (eg, ciprofloxacin) or doxycycline (Grade 2C). (See 'Mild or moderate infection' above.)
•Doses and durations – These are listed in the table (table 1).
●Prevention – Preventive measures include behavioral strategies to minimize the risk of exposure (eg, avoiding sick or dead animals, not using bare hands to skin game, tick or insect bite prevention). Following natural exposure, we suggest monitoring for clinical illness rather than antibiotic post-exposure prophylaxis (Grade 2C). However, post-exposure prophylaxis (with fluoroquinolones or doxycycline) is reasonable for adults with high-risk exposure, such as laboratory workers with nonintact skin, mucosal surface, or inhalational exposure to materials contaminated with F. tularensis. (See 'Prevention' above.)
