Tularemia: Current, comprehensive information on pathogenesis, microbiology, epidemiology, diagnosis, treatment, and prophylaxis

Last updated January 2, 2009

Agent
Pathogenesis
Epidemiology
—Reservoirs/Vectors/Modes of Transmission
—Naturally Occurring Tularemia in the United States
—Naturally Occurring Tularemia Worldwide
—Tularemia as a Biological Weapon
Clinical Syndromes
—Overview
—Glandular and Ulceroglandular Tularemia
—Pneumonic Tularemia
—Oculoglandular Tularemia
—Oropharyngeal Tularemia
—Typhoidal Tularemia
Pediatric Considerations
Differential Diagnosis
—Glandular Tularemia
—Ulceroglandular Tularemia
—Pneumonic Tularemia
—Oculoglandular Tularemia
—Oropharyngeal Tularemia
—Typhoidal Tularemia
Laboratory Diagnosis
—Specimen Collection and Transport
—Laboratory Biosafety and Biosecurity Information
—Laboratory Response Network (LRN)
—Standard Tests for Detection of F tularensis
—Additional Tests for Detection, Confirmation, and Characterization of F tularensis
Postexposure Prophylaxis for Tularemia
Treatment of Tularemia
Tularemia Vaccine
Infection Control
Issues Related to Autopsies and Burial
Public Health Reporting and Case Definitions
Images
References

Agent

Microbiologic Characteristics

Tularemia is caused by Francisella tularensis (formerly Pasteurella tularensis). Key microbiologic characteristics include the following (see References: CDC/ASM/APHL 2001: Basic protocols for level A laboratories for the presumptive identification of Francisella tularensis; Cross 2000; Penn 2005; Sneath 1986; Wong 1999).

Tiny, faintly staining, pleomorphic gram-negative rods (0.2-0.5 mcm x 0.7-1.0 mcm); smaller in patient samples than in culture; may be confused with Haemophilus species
Difficult to see by light microscopy in blood, tissue samples, or other specimens that contain significant background material
Nonsporulating, nonmotile
Aerobic (obligatory)
Requires cysteine (or cystine or another sulfhydryl source) for growth (although atypical strains that lack this requirement have been identified [see References: Bernard 1994])
Grows on commercial blood culture media, but does not grow (or grows unreliably) on most other standard agar media
Visible growth on appropriate media requires 2 to 4 days
Weakly catalase-positive (although may be negative), oxidase-negative
Thin lipid-rich capsule
Distinctive cellular fatty-acid profile

Other characteristics of F tularensis include the following:

Wild-type F tularensis strains generally are susceptible to aminoglycosides (streptomycin, gentamicin, kanamycin), tetracyclines, chloramphenicol, and fluoroquinolones.
Francisella tularensis strains generally are resistant to beta-lactam antibiotics, owing in part to beta-lactamase activity.
Organisms can persist for long periods of time in water, mud, and decaying animal carcasses (ie, moist environments).
Ingestion of F tularensis by environmental amebas may affect the bacterial ecology by:

Increasing environmental resistance
Increasing virulence (see References: Berdal 1996)

Subspecies

There are four subspecies of F tularensis. These subspecies can be differentiated by biochemical and molecular tests, and the current taxonomy is as follows (see References: Ellis 2002, Kugeler 2006, Morner 1993, Whipp 2003):

Francisella tularensis subsp tularensis (type A) (see References: Johansson 2004, Farlow 2005, Petersen 2006, Svensson 2004):

Highly infectious, generally more virulent, and more genetically diverse than subsp holarctica
Found primarily in North America
Demonstrates citrulline ureidase activity
Produces acid from glycerol fermentation
Two distinct genetic clades have been identified. The geographic distribution of the two clades in human cases correlates with the distribution of arthropod vectors and rabbit hosts (see References: Farlow 2005):

Clade 1 (aka subpopulation 1, A.I, type A-east) occurs primarily in the central United States, is associated with the distribution of Amblyomma americanum (the Lone Star tick) and Dermacentor variabilis (the American dog tick), and appears to have a high case-fatality rate.
Clade 2 (aka subpopulation 2, A.II, type A-west) occurs primarily in the western United States, is associated with the distribution of Dermacentor andersoni (the Rocky Mountain wood tick) and Chrysops discalis (the deer fly), and has a low case-fatality rate.
Strains of both major clades have been fully sequenced (see References: Larsson 2005, Beckstrom-Sternberg 2007).

Francisella tularensis subsp holarctica (type B): Less virulent than subsp tularensis, does not demonstrate citrulline-ureidase activity, and does not produce acid from glycerol fermentation; three biovars have been identified:

Biovar I: erythromycin sensitive; primarily found in North America, Europe, Siberia, the Far East, and Kazakhstan
Biovar II: erythromycin resistant; primarily found in Eurasia
Biovar japonica: found in Japan

Francisella tularensis subsp mediasiatica: Found in the Central Asian republics of the former Soviet Union (virulence is similar to subsp holarctica); produces acid from glycerol and thus may be confused with subsp tularensis (see References: Whipp 2003)
Francisella tularensis subsp novicida: Considered to be of low virulence and generally causes illness only in immunocompromised hosts (see References: Titball 2003) (note: F tularensis and F novicida traditionally have been considered separate species; however, the current approach is to consider F novicida as a subspecies of F tularensis); produces acid from glycerol and thus may be confused with subsp tularensis [see References: Cross 2000, Hollis 1989, Sjostedt 2003, Titball 2003, Whipp 2003])

Other Francisella species

Other Francisella species may be confused with F tularensis in clinical or environmental samples.
Francisella philomiragia is the only other taxonomically defined species in the genus other than F tularensis. It is halophilic, rarely associated with human disease, and difficult to identify by conventional methods (see References: Ellis 2002, Friis-Moller 2004, Whipp 2003).
Undefined Francisella-like bacteria appear to be common in the environment (see References: Barnes 2005).
Investigators have identified a novel Francisella species in ixodid ticks in some regions. This discovery highlights the need for careful analysis of PCR-based identification (see References: Sreter-Lancz 2008).

Pathogenesis

Virulence factors that contribute to pathogenesis of F tularensis have not been well defined and further studies are needed; however, key points on pathogenesis are outlined below (see References: Ellis 2002, Sjostedt 2003, Titball 2003):

Francisella tularensis is a facultative intracellular pathogen that multiplies predominantly within macrophages. The organisms initially enter macrophages through phagocytosis by a novel process of engulfment within asymmetric pseudopod loops and then disrupt the phagosomal membrane to gain direct access to the cytoplasm (see References: Clemens 2004).
Francisella tularensis virulence is determined in part by the ability of the organisms to replicate within macrophages. Bacteria are released from the macrophage following cell death by apoptosis. A recent study using a murine model demonstrated high levels of F tularensis in the plasma of infected mice. On the basis of this finding, the authors suggest that F tularensis in theblood of infected hostsis taken upby and replicates withinleukocytes and eventually escapes intothe plasma, where itpropagates a cycle ofinfection, escape, and reinfection (see References: Forestal 2007).
The capsule appears to be necessary to protect against serum-mediated lysis but is not required for survival following phagocytosis.
The lipopolysaccharide (LPS) does not exhibit the properties of a classic endotoxin and demonstrates low toxicity in vivo and in vitro, although LPS may have a role in macrophage growth.
The presence of type IV pili appears to be a virulence factor and may be particularly important for F tularensis infections that occur via the peripheral route. Direct repeat-mediated deletion of genes coding for type IV pili results in major virulence attenuation (see References: Forslund 2006).

Pathologic features for the various clinical syndromes caused by F tularensis have been described and are briefly summarized below.

Pneumonic Tularemia

Organisms enter the lungs either through inhalation of infectious aerosols or through hematogenous spread. The infectious dose by the respiratory route is 10 to 50 organisms (see References: Franz 1997, Saslow 1961).

Once in the lungs, the organisms rapidly enter pulmonary macrophages (within minutes) and begin replicating. The explosive replicative capacity of F tularensis appears to be an important factor in virulence associated with pulmonary infection (see References: Malik 2006). An intense accumulation of inflammatory cells, particularly neutrophils and macrophages, can be seen at sites of bacterial replication. The influx of neutrophils appears to play more of a destructive than protective role in the host response.

The following features have been noted for pneumonic tularemia (see References: Lillie 1937, Stuart 1945, Syrjala 1986):

Ulcerative bronchitis and bronchiolitis
Hemorrhagic edema with a nonspecific inflammatory response consisting of lymphocytes, plasma cells, and eosinophils (early in the clinical course)
Discrete nodules with acute suppurative necrosis of lung parenchyma
Alveolar exudates involving mononuclear cells, fibrin, and red blood cells
Nodular, segmental, or lobar consolidation
Caseous or cavitary lesions (later in the clinical course)
Granuloma formation (late in the clinical course)
Pleural fibrinous, fibrinocellular, or fibrinocaseous exudation
Hilar lymphadenopathy

Glandular and Ulceroglandular Tularemia

In both glandular and ulceroglandular tularemia, organisms enter the skin through the bite of infective arthropods, direct contact with infectious materials (such as contaminated carcasses), or percutaneous inoculation with a sharp object (such as a bone fragment from a contaminated carcass).

Organisms can enter through inapparent breaks in the skin surface.
The infectious dose for humans following percutaneous or inhalational inoculation is 10 to 50 organisms (see References: Cross 2000, Penn 2005).
In the ulceroglandular form, the organisms proliferate locally and cause a papule to develop at the site of inoculation within 3 to 5 days after initial exposure (see References: Cross 2000, Penn 2005).

The papule develops as a result of a localized inflammatory response that involves fibrin, neutrophils, macrophages, and T lymphocytes.
The initial inflammatory nidus becomes necrotic and degenerates over the next several days, thereby forming a tender ulcerated lesion at the site of the papule.
The ulcer is typically 2 to 4 cm in diameter and has an irregular and raised border.
A dark scab (which may resemble the characteristic eschar of anthrax) may occur over the area of ulceration.
Organisms spread from the site of inoculation to regional lymph nodes, where they cause necrotizing lymphadenitis surrounded by a neutrophilic and granulomatous inflammatory infiltrate (see References: CDC: Medical examiners, coroners, and biologic terrorism: A guidebook for surveillance and case management). Granulomas may develop in lymph nodes as the inflammatory process progresses; these may eventually coalesce to form abscesses. Follicular hyperplasia and inflammatory cell infiltrates involving predominantly granulocytes often are noted (see References: Sutinen 1986).
Affected lymph nodes may become fluctuant, rupture, and sometimes create draining sinus tracts in the skin.
Organisms may disseminate via hematogenous spread to involve multiple organs, and sepsis syndrome can occur.

In glandular tularemia, regional lymph node involvement occurs, but ulceration at the site of inoculation is absent.

Oculoglandular Tularemia

Organisms gain entry via the conjunctiva.
Superficial necrosis and ulceration of the conjunctiva occur, often accompanied by lymphocytic infiltration. Papules also may be noted (see References: Lillie 1937).
Granulomatous nodules may develop over time (see References: Lillie 1937).
Organisms spread from the conjunctiva to the preauricular, submandibular, or cervical lymph nodes, where they cause focal necrosis and lesions similar to those noted with ulceroglandular tularemia.
Infection most commonly is unilateral.

Oropharyngeal Tularemia

Organisms enter the mucous membrane of the oropharynx following ingestion or inhalation of organisms.
Exudative pharyngitis or tonsillitis usually occurs, and ulcers may develop.
Organisms spread to the cervical lymph nodes where necrosis and suppuration may occur.

Typhoidal Tularemia

Typhoidal tularemia involves a systemic illness without anatomic localization of infection.
Organisms enter the bloodstream through breaks in the skin or through mucous membranes and may affect the lungs and reticuloendothelial organs (ie, lymph nodes, liver, spleen, bone marrow).
Necrotic foci can occur in any involved organ, and caseating granulomas may develop.
Sepsis may occur, leading to shock, organ system failure, adult respiratory distress syndrome, and disseminated intravascular coagulation.

Epidemiology

Reservoirs/Vectors/Modes of Transmission

Reservoirs

Small and medium-sized mammals are the principal natural reservoirs for F tularensis. Examples include (see References: Dennis 1998, Gelman 1961, Hopla 1974):

Lagomorphs (rabbits, hares) (predominantly North America, Europe, Japan)
Aquatic rodents (beaver, muskrats, water voles)
Field voles
Water and wood rats
Squirrels
Lemmings (former Soviet Union, Sweden, Norway)
Meadow and field mice (predominantly former Soviet Union)

Humans, other mammalian species (eg, cats, dogs, cattle, primates), and some species of birds, fish, and amphibians are incidental hosts.

A recent serologic survey of 91 cats in Connecticut and New York found that 12% had antibody to F tularensis (see References: Magnarelli 2007).
An outbreak of tularemia in commercially distributed prairie dogs was recognized in the United States in 2002 (see References: CDC: Outbreak of tularemia among commercially distributed prairie dogs, 2002; Petersen 2004). Serologic testing of potentially exposed persons demonstrated that one person (an animal handler) had a positive F tularensis titer of 1:128 on initial testing that subsequently declined to 1:32 on follow-up testing at 4 and 6 months, suggesting prairie dog–to-human transmission (see References: Avashia 2004).
Outbreaks also have occurred in nonhuman primates housed outdoors. In a German primate facility, 18 of 35 cynomolgus monkeys (Macaca fasicularis) contracted tularemia within a 2-year period; six of the animals died (see References: Matz-Rensing 2007).

Information from studies conducted on Martha's Vineyard suggest that F tularensis can persist in the environment and that persons can acquire infection by engaging in activities that lead to aerosolization (such as lawn mowing, weed-whacking, and using a power blower) (see References: Feldman 2001, Feldman 2003). A recent analysis of sera from a variety of mammals on Martha’s Vineyard found that skunks and raccoons were frequently seroreactive (49% of skunks tested and 52% of raccoons), whereas white-footed mice, cottontail rabbits, deer, rats, and dogs were much more likely to be seronegative (see References: Berrada 2006).

During the fall and winter of 2003, F tularensis was identified on several filters from a biodetection air-monitoring system in Houston, Texas (see References: CIDRAP News 2003). An investigation conducted at that time supported contamination of the filters by naturally occurring F tularensis organisms, although the environmental reservoir was not definitively identified. As with the studies on Martha's Vineyard, these findings support persistence of F tularensis in the environment over time.

Francisella tularensis appears to survive within Acanthamoeba (a relatively ubiquitous protozoa), suggesting that these organisms may serve as a reservoir for F tularensis. Survival within Acanthamoeba may provide a mechanism for F tularensis to persist in the environment (see References: Abd 2003).

Vectors

A number of different arthropod vectors that transmit F tularensis have been identified (see References: Dennis 1998, Hopla 1974).
Primary vectors are ticks (United States, former Soviet Union, and Japan), mosquitoes (former Soviet Union, Scandinavia, and the Baltic region), and biting flies (United States [particularly Utah, Nevada, and California] and former Soviet Union). Examples of specific species include:

Ticks: Amblyomma americanum (Lone Star tick), Dermacentor andersoni (Rocky Mountain wood tick), D variabilis (American dog tick), Ixodes scapularis, Ixodes pacificus, and Ixodes dentatus
Mosquitoes: Aedes cinereus and Aedes excrucians
Biting flies: Chrysops discalis (deer fly), Chrysops aestuans, Chrysops relictus, and Chrysozona pluvialis

Modes of transmission

The average incubation period is 3 to 5 days. Francisella tularensis can be transmitted to humans via various mechanisms:

Bites by infected arthropods (see References: Klock 1973, Markowitz 1985)
Handling of infectious animal tissues or fluids (see References: Young 1969)
Ingestion of contaminated food or water (see References: Greco 1987, Mignani 1988, Reintjes 2002); murine models have confirmed that F tularensis is an effective oral pathogen and may pose a hazard, particularly to immunocompromised individuals if ingested in contaminated food or water (see References: KuoLee 2007)
Possibly direct contact with contaminated soil or water
Inhalation of infectious aerosols, including dust from contaminated hay (see References: Dahlstrand 1971) and aerosols generated by lawn mowing and brush cutting (see References: Feldman 2001, Feldman 2003)
Exposure in the laboratory setting (eg, inhalation of infectious aerosols, handling cultures or other infectious materials, accidental percutaneous exposure) (see References: Overholt 1961, Pike 1976)

Person-to-person transmission has not been documented.

Naturally Occurring Tularemia in the United States

Historical perspective

Colloquial terms for human illness associated with F tularensis include "rabbit fever," "deerfly fever," and "lemming fever."
Most US cases in recent years have been associated with bites from infected arthropods, although rabbits, hares, and other small mammals continue to be major sources of exposure for cases in the southeastern United States (see References: Dennis 1998).
During the 1930s, 2,000 or more cases were reported annually. Since that time, reported case numbers have gradually declined.
During the 1990s, the mean number of cases reported each year was 124 (although this number likely does not reflect actual incidence since many cases either are not reported or are not accurately diagnosed) (see References: CDC: Tularemia—United States, 1990-2000). States with the highest number of reported cases during these years included Arkansas, Missouri, Oklahoma, Kansas, South Dakota, and Montana. Martha's Vineyard also had a high number of identified cases. Reported incidence rates were higher in males than in females, with the highest rates reported for children 5 to 9 years old and persons 75 to 84 years of age.
Traditionally most cases have occurred in rural or semirural environments; cases rarely occur in urban settings.
In the United States, cases occur most commonly between May and August, although cases can occur during any time of year (see References: CDC 2002).
Data from a study of cases in the United States from 1964 through 2004 indicate that type A and type B infections differ in terms of affected populations, anatomic site of isolation, and geographic distribution. Molecular subtyping and pulsed-field gel electrophoresis defined two subpopulations of type A organisms (type A-east, type A-west) that differ in geographic distribution, disease outcome, and transmission (see References: Staples 2006).

Type A-west infections were less severe than either type B or type-A east infections. The case fatality rate for type A-east was 14%, for type B was 7%, and for Type A-west was 0%.
Type A-west infections occurred predominantly in the arid regions of the southwestern United States.
Type B infections clustered along major waterways, including the upper Mississippi River, and areas with high rainfall, such as the Pacific Northwest.
Type A-east infections occurred in Arkansas, Missouri, Oklahoma, and along the Atlantic Coast east of the Appalachians.

Climate change may affect future distribution of tularemia: Modeling studies suggest that, as the climate warms, the southern border of tularemia distributions in the United States will shift northward about 600 km. This could mean a reduced incidence in the south central United States (eg, Louisiana, Mississippi) and a greater incidence in north central states (eg, Michigan to North Dakota) (see References: Nakazawa 2007). Combined environmental and tularemia-incidence data have been used to develop a multivariate logistic regression model for predicting areas that have increased disease risk (see References: Eisen 2008).

Outbreaks

Outbreaks of tularemia occasionally have been recognized in the United States; examples include the following.

Martha's Vineyard, 2000: Fifteen cases of tularemia were reported; 11 patients had primary pneumonic disease and one patient died (see References: Feldman 2001). Illness was caused by F tularensis, type A. Patients were more likely than controls to have used a lawn mower or brush cutters in the 2 weeks before illness onset.
South Dakota, 1984: Twenty cases of glandular tularemia were reported in children and young adults on Crow Creek Indian reservations in the state (see References: Markowitz 1985). Illness was mild (fever, headache, and lymphadenopathy) and was presumably caused by F tularensis, type B. A similar outbreak occurred in 1979 on the Crow Indian Reservation in Montana, where 12 cases were identified (see References: Schmid 1983: Clinically mild tularemia associated with tick-borne Francisella tularensis).
Utah, 1971: Thirty-nine cases of tularemia were reported; most patients contracted illness from the bite of an infected deerfly (see References: Klock 1973). Clinical features included cutaneous ulcers at the site of a bite, lymphadenopathy, fever, chronic malaise, and weakness; all patients recovered. Strain type was not reported.
Vermont, 1968: Forty-seven cases of tularemia were diagnosed in persons who had handled muskrats in the 4 weeks before illness onset (see References: Young 1969). No fatalities were reported, but 14 patients had a severe prostrating illness that lasted an average of 10 days. Strain type was not reported.

Naturally Occurring Tularemia Worldwide

Outside the United States, the incidence of disease is highest in Scandinavian countries and Russia (see References: Dennis 1998).
Tularemia is endemic in neoarctic and paleoarctic regions between the latitudes of 30° and 71° N (ie, North America, Europe, states of the Russian Federation, China, and Japan [see References: Dennis 1998]).
Outbreaks involving a number of countries in Europe and nearby regions have been reported (see References: Christenson 1984, Dahlstrand 1971, Eliasson 2002, Greco 1987, Gurycova 2006, Kantardjiev 2006, Perez-Castrillon 2001, Reintjes 2002, Siret 2006, Syrjala 1985). Highlights include the following:

Sweden periodically has years of heavy tularemia activity. In 2003, 698 cases were reported, with a peak in August (300 cases) and September (164 cases). Other years of heavy activity include 1967, 1970, 1981, and 2000 (see References: Payne 2005). A large outbreak (90 cases) occurred in central Sweden in 2006 (see References: Wik 2006).
An outbreak in the Castilla y Leon region of Spain resulted in more than 500 cases of tularemia in the fall of 2007. The outbreak is thought to have arisen from unusual climatic and environmental circumstances, which produced suitable conditions for propagation of F tularensis. Many of the cases were involved in harvesting and related farm work, which likely contributed to production of aerosols that transported the bacteria (see References: Allue 2008).
Outbreaks occurred in three provinces in northwestern Turkey in February 2004 and again in February 2005 after a 60 year hiatus. Epidemiologic and environmental findings suggested that contaminated water or food was the cause (see References: Celebi 2006, Gurcan 2006).

Tularemia as a Biological Weapon

The following information supports the use of F tularensis as a potential biological weapon (see References: Christopher 1997; Dennis 2001; CDC: Tularemia Fact Sheet; WHO: Health aspects of chemical and biological weapons).

During World War II, the Japanese conducted research on F tularensis as a biological weapon.
During the 1950s and 1960s, the United States developed weapons that could deliver aerosolized F tularensis organisms. The United States government stockpiled weaponized tularemia until stockpiles were destroyed in 1973.
The former Soviet Union also weaponized F tularensis; the Soviet program included development of antibiotic- and vaccine-resistant strains.
In 1969, the World Health Organization estimated that an aerosol dispersal of 50 kg of virulent F tularensis over a metropolitan area with 5 million inhabitants in a developed country would result in 250,000 illnesses, including 19,000 deaths

The most likely form of intentional release for F tularensis organisms would be via infectious aerosols. An aerosol release would be expected to cause the following clinical syndromes:

Many of the patients would present with primary pneumonic tularemia; however, some would present with a nonspecific febrile illness of varying severity (ie, typhoidal tularemia).
Cases of oculoglandular tularemia could occur from eye contamination.
Cases of glandular or ulceroglandular disease could occur through exposure of broken skin to infectious aerosol.
Cases of oropharyngeal disease also could occur through inhalation of organisms.

An outbreak of tularemia caused by a bioterrorist attack would be expected to have the following features:

The incubation period generally correlates with the virulence of the infecting strain; in a bioterrorist attack, a highly virulent strain with a relatively short incubation period likely would be used. Illness onsets would generally occur 3 to 5 days after the initial release but could occur as soon as 1 day and up to 14 days later.
Illness would probably occur in an urban area and not in rural regions (where naturally occurring tularemia would be more prevalent).
Patients would not have risk factors for tularemia exposure (eg, outdoor field work, recent outdoor recreational activity, agricultural exposures, exposure to tissues of potentially infected animals).

In the event of a bioterrorist attack, use of F tularensis strains with enhanced virulence or antimicrobial resistance is of concern; therefore, past experience may not be a valuable predictor of disease severity under such circumstances.

Some animals might serve as sentinels of certain bioterrorism agents, including Bacillus anthracis and Yersinia pestis. While animals are not likely to provide early warning for a bioterrorist event cause by F tularensis, a recent review indicates that animals (such as prairie dogs, other rodents, raccoons, skunks, and cats) could serve as markers for ongoing exposure risk following a tularemia bioterrorist event or could propagate or maintain an epidemic (see References: Rabinowitz 2006).

Clinical Syndromes

Overview

Francisella tularensis infection can cause the following clinical syndromes (see References: Dennis 1998):

Ulceroglandular tularemia (45% to 85% of naturally occurring cases)
Glandular tularemia (10% to 25% of naturally occurring cases)
Pneumonic tularemia (<5% of naturally occurring cases, although outbreaks following inhalational exposure have been reported; secondary pneumonia [often associated with the typhoidal form] occurs relatively frequently and results from hematogenous spread to the lungs)
Oculoglandular tularemia (<5%)
Oropharyngeal tularemia (<5%)
Typhoidal tularemia (<5%; although in outbreaks caused by aerosol exposure, this percentage may be much higher)

Tularemia can range from a mild infection to a severe life-threatening illness.

Before antibiotic therapy was available, the overall case-fatality rate was approximately 7%, although rates as high as 50% were seen with pneumonia and other forms of severe infection (see References: Dennis 2001, Pullen 1945).
Currently, case-fatality rates are low (approximately 2%) (see References: Evans 1985).
Most patients respond rapidly to appropriate antibiotic therapy, with fever and generalized symptoms improving in 24 to 48 hours.
Type A tularemia is more severe than type B, which is generally a mild illness.
One study identified the following factors as associated with a poor outcome (ie, death, relapse, prolonged recovery) (see References: Penn 1987):

Underlying comorbidity (eg, alcoholism, diabetes)
Delay in seeking medical care
Delay in institution of appropriate antibiotic therapy

Clinical features for the major syndromes caused by F tularensis are outlined in the tables below. Initial signs and symptoms can be relatively nondescript and the diagnosis may be missed (see References: Dembek 2003). Two recent cases of human infection with F tularensis were initially diagnosed as herpes simplex or varicella zoster infection (see References: Byington 2008).

Clinical Features of Glandular and Ulceroglandular Tularemia
Feature	Characteristics
Incubation period	3-5 days (range, 1-14 days)
Presenting features^a	—Local painful cutaneous lesion at site of inoculation (papule that ulcerates within a few days) in ulceroglandular form; no cutaneous lesion in glandular form —Tender regional lymphadenopathy —Fever —Constitutional symptoms (chills, malaise, myalgias, arthralgias, headache, anorexia) —Other skin lesions may be noted (erythema nodosum; erythema multiforme–like exanthem on hands, arms, or legs; maculopapular rash, acneiform lesions, urticaria^b) —Clinical features for 39 patients identified during outbreak of predominantly ulceroglandular tularemia associated with exposure to muskrats in Vermont^c: ~Fever (97%) ~Cutaneous ulcers (74%) ~Axillary adenopathy (67%) ~Chills (59%) ~Myalgias (56%) ~Malaise (51%) ~Diaphoresis (28%) ~Epitrochlear adenopathy (25%) ~Nausea and vomiting (8%) ~Pleuritic chest pain (5%) ~Cough (5%) ~Preauricular adenopathy (2%)
Laboratory features	—In one series of 88 patients with tularemia, admission WBCs ranged from 5,000 to 22,000/mm³ (median, 10,400mm³)^d; differential usually normal early in clinical course —Elevated hepatic enzymes and bilirubin may occur^d
Complications	—Suppuration of involved lymph nodes —Secondary pneumonia (31% of patients with ulceroglandular disease in one case series§ and 17% of patients with ulceroglandular or glandular disease in another^e) —Involvement of other organs (via hematogenous spread) —Sepsis syndrome —Illness may be debilitating, with full recovery taking several months —Lymphadenopathy may persist for months^e
Case-fatality rate	—4.4% of 181 patients with ulceroglandular tularemia and 4.3% of 23 patients with glandular tularemia among case series of 225 patients reported from pre-antibiotic era^f —1.6% (2 of 123 patients with ulceroglandular or glandular tularemia) in case series of 165 treated cases occurring in Oklahoma 1979-1885^g —Fatalities usually associated with type A subspecies; type B subspecies less virulent
Abbreviations: WBC, white blood count. ^aSee References: Cross 2000, Dennis 2001, Penn 2005, Sanders 1968, Evans 1985. ^bSee References: Christenson 1984, Cross 2000, Dahlstrand 1971, Evans 1985. ^cSee References: Young 1969. ^dSee References: Evans 1985. ^eSee References: Sanders 1968. ^fSee References: Pullen 1945. ^gSee References: Rhorbach 1991.

Clinical Features of Pneumonic Tularemia^a
Feature	Characteristics
Incubation period	3-5 days (range, 1-14 days)
Presenting features^b	—Patients often present with community-acquired atypical pneumonia nonresponsive to conventional antibiotic therapy —Predominant symptoms include abrupt onset of fever, nonproductive cough, myalgias (particularly low back) —Nausea, vomiting, diarrhea may occur —Illness may be rapidly progressive and severe or may be indolent with progressive weakness and weight loss over several weeks to months —Skin lesions may be noted (erythema nodosum; erythema multiforme–like exanthem on hands, arms, or legs; maculopapular rash; acneiform lesions; urticaria) —Presenting features for 53 Finnish patients with inhalational exposure^c: ~Fever (100%) ~Headache, myalgias, arthralgias ("most") ~Dry cough (45%) ~Retrosternal discomfort, pleural pain, or dyspnea (45%) ~Sore throat (23%) ~Hemoptysis (2%)
Laboratory features	—Radiographic findings for 50 tularemia patients with pleuropulmonary involvement^d: ~Patchy subsegmental air space opacities (74%; unilateral in 54% overall) ~Hilar adenopathy (32%) ~Pleural effusion (30%; unilateral in 20% overall) ~Lobar or segmental opacities (18%; all unilateral) ~Cavitation (16%) ~Oval opacities (8%) ~Cardiomegaly with pulmonary edema pattern (6%; caused by pericarditis in one case) ~Apical infiltrate (4%) ~Empyema and bronchopleural fistula (4%) ~Mediastinal mass (2%; caused by hilar adenopathy) ~Miliary pattern (2%) —In one series of 88 patients with tularemia, admission WBCs ranged from 5,000 to 22,000/mm³ (median, 10,400mm³)^e; differential usually normal early in clinical course —Elevated hepatic enzymes and bilirubin may occur^e —Sputum Gram stain often not helpful in making diagnosis
Complications	—Lung abscesses or cavitary lesions —Adult respiratory distress syndrome^f —Fibrosis and calcifications in affected lung areas or pleura as illness resolves —Granulomatous pleuritis (which may resemble tuberculosis)^g —Empyema with bronchopleural fistula —Involvement of other organs through hematogenous spread —Sepsis syndrome —Meningitis —Pericarditis^d,e —Illness may be debilitating, with full recovery taking several months; relapses have been reported with use of broad-spectrum antibiotics^h
Case-fatality rate	—Fatalities rare with appropriate antibiotic therapy (reported as 2.3% in one case series of 88 patients with tularemia, about half of whom had pulmonary involvement; both deaths occurred in patients with pneumonia^e) —Fatalities usually associated with type A subspecies; type B subspecies less virulent
Abbreviations: WBC, white blood count. ^aPneumonic tularemia may either result from primary inhalational exposure or occur as a secondary process in other forms of tularemia. Similarly, inhalational exposure may cause primary pneumonic tularemia, oropharyngeal tularemia, or typhoidal tularemia without obvious pulmonary involvement. In the past, the term "typhoidal tularemia" was used to refer to infections caused by inhalational exposure, even if pneumonia was the primary manifestation. This resulted in some confusion in the medical literature. Experts have suggested that the term "typhoidal tularemia" not be used in the context of inhalational exposure but rather be used only to describe cases of tularemia with constitutional symptoms or sepsis syndrome and no obvious anatomic focus of disease (see References: Dennis 2001, Syrjala 1986). ^bSee References: Christenson 1984, Cross 2000, Dahlstrand 1971, Evans 1985, Fredricks 1996, Miller 1969, Roy 1989. ^cSee References: Syrjala 1985. ^dSee References: Rubin 1978. ^eSee References: Evans 1985. ^fSee References: Sunderrajan 1985. ^gSee References: Schmid 1983: Granulomatous pleuritis caused by Francisella tularensis: possible confusion with tuberculous pleuritis. ^hSee References: Overholt 1961.

Clinical Features of Oculoglandular Tularemia
Feature	Characteristics
Incubation period	3-5 days (range, 1-14 days)
Presenting features^a	—Multiple painful yellow conjunctival nodules —Conjunctival ulcers —Chemosis —Periorbital and facial edema around affected eye —Extremely tender regional adenopathy involving preauricular, submandibular, or cervical lymph nodes; edema around affected nodes may be present —Patients may present with Parinaud's syndrome (unilateral granulomatous conjunctivitis and enlarged preauricular lymph nodes)† —Constitutional symptoms (fever, chills, malaise, anorexia) —History of minor eye trauma, swimming in potentially contaminated water (possibly a risk factor)^b, or tick exposure may be present with naturally acquired infection
Laboratory features	—Generally unremarkable —Gram stain of conjunctival scrapings may demonstrate organisms, although Gram stain often not helpful^b
Complications	—Suppuration of affected lymph nodes —Sepsis syndrome —Involvement of other organs (through hematogenous spread)
Case-fatality rate	—1 (14.3%) of 7 patients with oculoglandular tularemia among case series of 225 patients reported from pre-antibiotic era^c —Fatalities rare with appropriate antibiotic therapy
^aSee References: Cross 2000, Guerrant 1976, Halperin 1985, Penn 2005. ^bSee References: Halperin 1985. ^cSee References: Pullen 1945.

Clinical Features of Oropharyngeal Tularemia
Feature	Characteristics
Incubation period	3-5 days (range, 1-14 days)
Presenting features^a	—Fever —Constitutional symptoms (chills, malaise, myalgias, arthralgias) —Exudative pharyngitis or tonsillitis —Ulcerations of pharynx, tonsils, soft palate —Stomatitis (less common) —May see pharyngeal membrane suggestive of diphtheria (membrane associated with tularemia does not bleed if removed, unlike diphtheria where removal of membrane reveals bleeding submucosa) —Cervical or retropharyngeal adenopathy (cervical nodes tender to palpation) —Concomitant pneumonia often present —Patients may present with dental abscesses —Findings in 12 patients with pharyngeal involvement in case series of 88 patients^b: ~Erythema (50%) ~Petechiae or hemorrhage (25%) ~Exudate (17%) ~Ulcers (8%) —The most common symptoms among 145 patients in Turkey^c: ~Swelling of the neck (92%) ~Sore throat (92%) ~Fever (90%)
Laboratory features	—Generally unremarkable, although leukocytosis may be present —Among patients in Turkey, ESR was elevated in all, and 79% had an ESR exceeding 55 mm/hr^c
Complications	—Sepsis syndrome —Suppuration of involved lymph nodes —Involvement of other organs (via hematogenous spread) —Illness may be debilitating, with full recovery taking several months
Case-fatality rate	—Fatalities rare with appropriate antibiotic therapy —Fatalities usually associated with type A subspecies; type B subspecies less virulent
Abbreviation: ESR, erythrocyte sedimentation rate. ^aSee References: Tyson 1976, Luotonen 1986, Tunga 2007. ^bSee References: Evans 1985. ^cSee References: Meric 2008.

Clinical Features of Typhoidal Tularemia^a
Feature	Characteristics
Incubation period	3-5 days (range, 1-14 days)
Presenting features^b	—Fever —Constitutional symptoms (chills, malaise, weakness, myalgias, arthralgias) —Prostration —Dehydration —Gastrointestinal symptoms (watery, nonbloody diarrhea; vomiting; abdominal pain) —Skin lesions may be noted (erythema nodosum; erythema multiforme–like exanthem on hands, arms, or legs; maculopapular rash; acneiform lesions; urticaria)
Laboratory features	—In one series of 88 patients with tularemia, admission WBCs ranged from 5,000 to 22,000/mm³ (median, 10,400mm³; differential usually normal)^c —Elevated hepatic enzymes and bilirubin may occur^c —Microscopic pyuria may occur^c
Complications	—Secondary pneumonia (83% of patients with typhoidal disease in one case series^c and 50% in another^d) —Involvement of other organs via hematogenous spread (eg, meningitis,^e hepatitis and jaundice,^f splenic rupture, encephalitis, pericarditis,^c peritonitis, osteomyelitis) —Sepsis syndrome —Rhabdomyolysis^g —Renal failure^d —Illness may be debilitating, with full recovery taking several months, relapses have been reported with use of broad-spectrum antibiotics^h
Case-fatality rate	—50% in one series of 14 patients with typhoidal tularemia among case series of 225 patients reported from pre-antibiotic eraⁱ —6.6% (2 of 30 patients with typhoidal tularemia) in case series of 165 treated cases occurring in Oklahoma 1979-1885^j —Fatalities usually associated with type A subspecies; type B subspecies less virulent
^aIn the past, the term "typhoidal tularemia" was used to refer to infections caused by inhalational exposure, even if pneumonia was the primary manifestation. This has resulted in some confusion in the medical literature. Experts have suggested that the term "typhoidal tularemia" not be used in the context of inhalational exposure but rather be used only to describe cases of tularemia with constitutional symptoms or sepsis syndrome and no obvious anatomic focus of disease (see References: Dennis 2001, Syrjala 1986). ^bSee References: Christenson 1984, Cross 2000, Dahlstrand 1971, Dennis 2001, Evans 1985, Penn 2005, Sanders 1968. ^cSee References: Evans 1985. ^dSee References: Sanders 1968. ^eSee References: Lovell 1986, Stuart 1945: Tularemic meningitis: review of the literature and report of a case with postmortem observations. ^fSee References: Ortego 1986. ^gSee References: Klotz 1987, Penn 1987. ^hSee References: Overholt 1961. ⁱSee References: Pullen 1945. ^jSee References: Rohrbach 1991.

Pediatric Considerations

In general, clinical manifestations of tularemia are similar in children and adults. One report from Arkansas compared type of illness and clinical symptoms between children and adults with naturally occurring tularemia identified in 1983; findings are noted in the following table.

Comparison Between Cases of Tularemia in Children and Adults, Arkansas 1983
Type of Disease or Symptoms	Percentage of Children (N =28)	Percentage of Adults (N = 43)
Type of Disease
Ulceroglandular Glandular Pneumonic Oropharyngeal Oculoglandular Typhoidal Unclassified	45 25 14 4 2 2 6	51 12 18 — — 12 11
Symptoms
Lymphadenopathy Fever Ulcer/papule Pharyngitis Myalgias/arthralgias Nausea/vomiting Headache Cough Diarrhea Conjunctivitis	96 87 45 43 39 35 9 9 4 4	65 21^a 51 — 2 19 5 5 5 —
^aAlthough this percentage was reported for fever among adults in this series of patients, fever generally is a hallmark finding for tularemia. Adapted from Jacobs 1985 (see References).

According to the American Academy of Pediatrics (AAP), streptomycin, gentamicin, and amikacin are recommended for treatment of children (see References: AAP 2006). Other therapies cited by the AAP include tetracycline and chloramphenicol, although relapse rates are higher with these agents.

Ciprofloxacin has recently been shown to be an effective therapy for tularemia in children (see References: Johansson 2000: Ciprofloxacin for treatment of tularemia in children; Johansson 2002). The drug has been approved only for specific indications in patients younger than 18 years old (see References: AAP 2006).

See the section on Treatment of Tularemia for specific drug regimens.

Differential Diagnosis

Differential Diagnosis for Glandular Tularemia
Condition^a,b	Distinguishing Features
Bubonic plague (Yersinia pestis)	—Clinical course often fulminant —Systemic toxicity common
Cat-scratch disease (Bartonella henselae)	—History of contact with cats; usually history of cat-scratch —Indolent clinical course; progresses over weeks —Primary lesion at site of scratch often present (small papule, vesicle)
Mycobacterial infection, including scrofula (Mycobacterium tuberculosis and other Mycobacterium species)	—With scrofula, adenitis occurs in cervical region —Lymph nodes generally painless and nontender —Infections with species other than M tuberculosis more likely to occur in immunocompromised patients
Sporotrichosis (Sporothrix schenckii)	—Lymph nodes generally painless and nontender —Systemic symptoms absent —Painless papulonodular cutaneous lesion usually present distal to involved lymph nodes; secondary cutaneous lesions may occur along lymphatic channels —Patients often have history of contact with soil, plants, or plant products (eg, sphagnum moss, thorned plants such as rose bushes)
Streptococcal or staphylococcal adenitis (Staphylococcus aureus, Streptococcus pyogenes)	—Site of initiating infection often present distal to involved nodes (ie, pustule, infected traumatic lesion) —Involved nodes more likely to be fluctuant
Chancroid (Haemophilus ducreyi)	—Adenitis occurs in inguinal region only —Ulcerative lesion present —History of sexual exposure or activity
Lymphogranuloma venereum (Chlamydia trachomatis)	—Adenitis occurs in inguinal region only —History of sexual exposure 10-30 days previously —Suppuration, fistula tracts common —Although lymph nodes may be somewhat tender, exquisite tenderness usually absent —History of sexual exposure or activity
Primary genital herpes	—Herpes lesions in genital area —Adenitis in inguinal region only —History of sexual exposure or activity
Secondary syphilis (Treponema pallidum)	—Enlarged lymph nodes in inguinal region only —Lymph nodes generally painless and nontender —History of sexual exposure or activity
^aSee References: Butler 1979, Cross 2000, Penn 2005. ^bInfectious causes of generalized lymphadenopathy (eg, cytomegalovirus infection, toxoplasmosis, mononucleosis, lymphoma) also may be considered, depending on the clinical presentation.

Differential Diagnosis for Ulceroglandular Tularemia
Condition	Distinguishing Features
Anthrax (Bacillus anthracis)	—Painless ulcer that develops into black eschar over several days —Extensive non-pitting edema around lesion may occur
Orf (orf virus, a parapox virus)	—Occurs in farm workers —Characterized by pustule that progresses to weeping nodule —Regional adenitis may occur, but not common
Pasteurella infections (Pasteurella multocida)	—History of dog or cat exposure (animal bite or licking of open wound) —Regional lymphadenopathy occurs in 30%-40% of cases
Primary syphilis (Treponema pallidum)	—Characterized by painless ulcer (chancre) in genital area —Lymph nodes generally painless and nontender
Rat-bite fever (Spirillum minus)^a	—Infection caused by S minus occurs in Asia —Maculopapular rash over palms, soles, and extremities 2-4 days after onset of fever
Rickettsialpox (Rickettsia akari)	—Initial presentation involves painless papule which forms black eschar —Generalized maculopapular rash appears 2-3 days later —Regional lymphadenopathy usually present but nontender
Scrub typhus (Orientia tsutsugamushi; formerly Rickettsia tsutsugamushi)	—Zoonotic infection from chigger bites —Occurs in endemic areas (Asia and Western Pacific) —Often associated with a generalized maculopapular rash
Staphylococcal or streptococcal cellulitis (S aureus, S pyogenes)	—May be history of trauma or preexisting lesion at site of infection
^aRat-bite fever caused by Streptobacillus moniliformis (type found in North America and Europe) generally does not result in ulceration at site of bite, is not associated with regional lymphadenopathy, and therefore is not considered in differential diagnosis of ulceroglandular tularemia.

Differential Diagnosis for Pneumonic Tularemia
Condition^a,b	Distinguishing Features
Community-acquired bacterial pneumonia —Mycoplasmal pneumonia (Mycoplasma pneumoniae) —Pneumonia caused by Chlamydia pneumoniae —Legionnaires' disease (Legionella pneumophila or other Legionella species) —Psittacosis (Chlamydia psittaci) —Other bacterial agents (eg, Staphyloccocus aureus, Streptococcus pneumoniae, Haemophilus influenzae, Klebsiella pneumoniae, Moraxella catarrhalis)	—Legionellosis and many other bacterial agents (S aureus, S pneumoniae, H influenzae, K pneumoniae, M catarrhalis) usually occur in persons with underlying pulmonary or other disease or in elderly —Bird exposure with psittacosis —Community outbreaks caused by other etiologic agents less likely to suggest point-source outbreak (as would be seen with intentional release of F tularensis) —Outbreaks of S pneumoniae usually institutional —Community outbreaks of Legionnaires' disease often involve exposure to cooling towers —Gram stain of sputum may be useful in distinguishing agents
Inhalational anthrax (Bacillus anthracis)	—Widened mediastinum and pleural effusions seen on CXR or chest CT —Not true pneumonia; minimal sputum production —Severe and rapidly progressive course; often fulminant and fatal
Pneumonic plague (Yersinia pestis)	—Hemoptysis commonly occurs —Consolidation often noted on CXR early in clinical course (radiographic evidence of pneumonia in patients with tularemia generally not as pronounced early in clinical course) —Severe and rapidly progressive course; often fulminant and fatal
Q fever (Coxiella burnetii)	—Exposure to infected parturient cats, cattle, sheep, goats —May be difficult to distinguish clinically from pneumonic tularemia
Tuberculosis (Mycobacterium tuberculosis)	—More common among elderly or among persons who have lived in tuberculosis-endemic countries (ie, developing world, countries of former Soviet Union)
Viral pneumonia —Influenza —Hantavirus —RSV —CMV	—Influenza generally seasonal (October-March in United States) or involves history of recent cruise ship travel or travel to tropics —Exposure to excrement (urine and feces) of mice with Hantavirus —RSV usually occurs in children (although may be cause of pneumonia in elderly); tends to be seasonal (winter/spring) —CMV usually occurs in immunocompromised patients


Home	Mission & Activities	About Us	Center Support	Contact Us