# 58 - 175 Tularemia

### 175 Tularemia

fluoroquinolone to reduce the need for valve replacement. Treat­
ment is usually given for at least 4–6 months, and clinical endpoints 
for its discontinuation are often difficult to define. Surgery is still 
required for the majority of cases of infection of prosthetic heart 
valves and prosthetic joints.
There is good evidence for the benefit of antimicrobial pro­
phylaxis to reduce the risk of infection after occupational or other 
exposure to Brucella organisms, inadvertent immunization with 
live vaccine intended for use in animals, or exposure to deliberately 
released brucellae. However, there is little evidence to inform the 
duration of prophylaxis or preference for dual or single therapy. 
Based on historical practices, many national guidelines recommend 
the administration of rifampin plus doxycycline for 3 weeks. How­
ever, such regimens are poorly tolerated, and doxycycline mono­
therapy of the same duration may be used instead. Monotherapy 
is the standard recommendation in the United Kingdom but not 
in the United States. Rifampin should be omitted after exposure to 
vaccine strain RB51, which is resistant to rifampin, and replaced 
by another agent such as TMP-SMX in combination with doxycy­
cline. After significant brucellosis exposure, expert consultation is 
advised for women who are (or may be) pregnant.
■
■PROGNOSIS AND FOLLOW-UP
Relapse occurs in up to 30% of poorly compliant patients. Thus, 
patients should ideally be followed clinically for up to 2 years to detect 
relapse, which responds to a prolonged course of the same therapy used 
originally. The general well-being and the body weight of the patient 
are more useful guides than serology to lack of relapse. IgG antibody 
levels detected by the standard agglutination test and its variants can 
remain in the diagnostic range for >2 years after successful treatment. 
Complement fixation titers usually fall to normal within 1 year of cure. 
Immunity is not solid; patients can be reinfected after repeated expo­
sures. Fewer than 1% of patients die of brucellosis. When the outcome 
is fatal, death is usually a consequence of cardiac involvement; more 
rarely, it results from severe neurologic disease. Despite the low mor­
tality rate, recovery from brucellosis is slow, and the illness can cause 
prolonged inactivity, with domestic and economic consequences.
The existence of a prolonged chronic brucellosis state after success­
ful treatment remains controversial. Evaluation of patients in whom 
this state is considered (often those with work-related exposure to bru­
cellae) includes careful exclusion of malingering, nonspecific chronic 
fatigue syndromes, and other causes of excessive sweating, such as alco­
hol abuse and obesity. In the future, the availability of more sensitive 
assays to detect Brucella antigen or DNA may help to identify patients 
with ongoing infection.
■
■PREVENTION
Vaccines based on live attenuated Brucella strains, such as B. abortus 
strain 19BA or 104M, have been used in some countries to protect 
high-risk populations but have displayed only short-term efficacy and 
high reactogenicity. Subunit vaccines have been developed but are of 
uncertain value and cannot be recommended at present. Research in 
this area has been stimulated by interest in biodefense (Chap. S4) and 
may eventually yield new products. The mainstay of veterinary pre­
vention is a national commitment to testing and slaughter of infected 
herds/flocks (with compensation for owners), control of animal 
movement, and active immunization of animals. These measures are 
usually sufficient to control human disease as well. In their absence, 
pasteurization of all milk products before consumption is sufficient to 
prevent nonoccupational animal-to-human transmission. All cases of 
brucellosis in animals and humans should be reported to the appropri­
ate public health authorities.
■
■FURTHER READING
Beeching NJ et al: Brucellosis. BMJ Best Practice, 2023. https://

bestpractice.bmj.com/topics/en-us/911.
Bosilkovski M et al: The current therapeutical strategies in human 
brucellosis. Infection 49:823, 2021.

Centers for Disease Control and Prevention: Brucellosis. 

https://www.cdc.gov/brucellosis/index.html.
Dean AS et al: Clinical manifestations of human brucellosis: A sys­
tematic review and meta-analysis. PLoS Negl Trop Dis 6:e1929, 2012.
Norman FF et al: Imported brucellosis: A case series and literature 
review. Travel Med Infect Dis 14:182, 2016.
Zagorsky P et al: Laboratory diagnosis of human brucellosis. Clin 
Microbial Rev 33:e00073, 2019.
Anders F. Johansson

Tularemia
■
■DEFINITION
Tularemia is a zoonotic disease that can naturally be transmitted from 
vertebrate animals to humans, causing a febrile illness. Synonyms 
include deer-fly fever, rabbit fever, market men’s disease, water-rat 
trappers’ disease, wild hare disease (yato-byo), Francis’ disease, and 
Ohara’s disease.
■
■ETIOLOGY
Tularemia is caused by Francisella tularensis, a small (0.2–0.7 × 0.2 µm), 
facultatively intracellular, aerobic, gram-negative, pleomorphic bac­
terium. F. tularensis is highly virulent for humans and numerous 
mammals, including rodents, hares, and rabbits. It is known to cause 
airborne laboratory-acquired infection; therefore, laboratory work, 
including culturing of the bacterium, must be performed by trained 
staff in biologic safety cabinets under biosafety level 3 conditions. The 
disease tularemia was described by the American researchers McCoy 
and Chapin as a “plague-like disease of rodents” in California in 1911, 
1 year after they cultivated and described the causative bacterium as 
“Bacterium tularense.” Francisella was later named after Edward Fran­
cis, an American bacteriologist who extensively studied the etiologic 
agent and the pathogenesis of tularemia over several decades. There 
are two F. tularensis subspecies of high clinical importance: F. tularensis 
subspecies tularensis (aka Jellison type A) with human disease docu­
mentation exclusively from the North American continent, and sub­
species holarctica (Jellison type B) with human disease documentation 
both from North America and from other parts of the world. A third 
subspecies, F. tularensis subspecies mediasiatica, has been isolated from 
vertebrate animals in Central Asia and Russia, but there is no docu­
mented disease in humans. The division into type A and type B strains 
of F. tularensis is based on minor biochemical test differences, with type 
A strains being significantly more lethal in experimentally infected 
rabbits. Type A but not type B cultures can ferment glycerol and are 
positive in a citrulline ureidase assay. Microbial genome sequencing 
has revealed an average nucleotide identity >99% between type A and 
type B strains of F. tularensis, but current and historic epidemiologic 
studies mirror the experimental findings in rabbits—tularemia mortal­
ity among humans is higher with type A in the United States. However, 
recent research using high-resolution genetic subtyping, combined 
with analyses of disease outcomes, shows there is heterogeneity in mor­
tality risks among different type A genetic varieties, with some being 
less virulent for humans than the type B varieties.
CHAPTER 175
Tularemia
Genetic neighbors of F. tularensis, including Francisella novicida 
(sometimes referred to as a fourth subspecies of F. tularensis with a 
genomic average nucleotide identity of ~97% with other subspecies) 
and the genetically more distant Francisella philomiragia, are oppor­
tunistic pathogens that may cause disease in humans with symptoms 
varying according to the immune status of the patient. Moreover, mod­
ern microbiology methods and intense research efforts have resulted 
in the identification of additional Francisella bacteria, including

pathogens of fish, endosymbiont bacteria of ticks, soil bacteria, and 
bacteria isolated from sea and freshwater, some of which may be 
opportunistic human pathogens. Taxonomists now recognize at least 
nine distinct Francisella species, including Francisella tularensis, the 
causative agent of tularemia.

■
■EPIDEMIOLOGY
F. tularensis infects humans through bites of arthropods functioning as 
disease vectors (e.g., mosquitoes, ticks, tabanid flies), inhaling infec­
tious aerosols, handling infected animals, or ingesting contaminated 
water. The reported human cases represent bridging between endemic 
F. tularensis maintenance in nature, or spillover events from massive 
amplification in infected rodents, hares, or rabbits during disease 
outbreaks among animals (epizootics). The disease is endemic to the 
Northern Hemisphere but has a patchy and geographically uneven dis­
tribution, which is believed to be the result of F. tularensis being main­
tained long-term in smaller geographic areas experiencing repeated 
outbreaks, often at intervals of several years. In the United States, 
reported human cases have dropped sharply from several thousand 
annual cases before 1950 to 200–300 cases annually from 2010–2020, 
with most of these cases being associated with tick bites in Arkansas, 
Kansas, Missouri, and Oklahoma according to Centers for Disease 
Control and Prevention (CDC) statistics. Surveillance data from 2006–
2021 with available F. tularensis culture data showed 66% type A and 
34% type B cases in humans. In some parts of Europe and Asia where 
type B only causes human disease, tularemia was relatively more com­
mon in 2010–2022, with several hundred to >2000 cases in peak years 
in Finland, Sweden, and Turkey, respectively. In the Northern Boreal 
Forest regions of Sweden, Finland, and Russia, tularemia is primarily 
associated with mosquito bites, while tick bites are more common in 
Western and Eastern Europe. Most cases in the United States, as well 
PART 5
Infectious Diseases
Type A Tularemia in North America
Donor and Reservoir
Vector
Reservoir and Vector
Type B Tularemia in North America, Europe, and Asia
Environmental reservoir?
Donor and Reservoir
Reservoir and Vector
Vectors in the Boreal
Forests
FIGURE 175–1  Schematic illustration of the type A and type B disease ecology concepts. (Adapted from P. Keim et al: Molecular epidemiology, evolution, and ecology of 
Francisella. Ann N Y Acad Sci 1105:30, 2007, Figure 11.)

as in Finland and Sweden, occur during summer, likely reflecting 
increased human exposure to blood-feeding arthropods, infected wild­
life mammals, or aerosols from carcasses of infected animals during 
outdoor activities. In contrast to this seasonal pattern, most cases occur 
in the winter months in Turkey, where the predominant infection route 
to humans in some rural areas is nonchlorinated drinking water from 
natural springs that may be contaminated by infected rodents. Colder 
temperatures enhance the survival of the bacterium in water during 
winter. In the Southern Hemisphere, rare human and animal infections 
have been documented from Australia. It is uncertain whether the lack 
of reporting from other areas of the Southern Hemisphere is due to 
lack of disease or from underreporting.
Based on extensive historic field studies performed in the United 
States and the former Soviet Union during the 1950s–1970s, it is 
believed that distinct disease ecologies characterize F. tularensis sub­
species tularensis (type A) and F. tularensis subspecies holarctica (type B), 
respectively. This concept recognizes the association of type A tulare­
mia with rabbits, ticks, domestic cats, and sheep in comparatively dry 
environments, in contrast to the association of type B tularemia with 
streams, ponds, lakes, rivers, and semiaquatic animals such as muskrats 
and beavers (Fig. 175–1).
It remains unknown exactly how the bacterium is maintained in 
nature. F. tularensis is a generalist eukaryotic host-associated microbe 
that can infect many animals and cell types, including amphibians, 
birds, rodents, rabbits, carnivores, and ruminant mammals. Most of 
these animals are accidental hosts or rapidly die from septic disease. 
It is unclear if any can maintain the bacterium for a long duration. 
However, when an animal dies from a tularemia infection with a high 
level of bacteria in their bodies, this can potentially help temporarily 
maintain the bacteria in the environment. Over the long term, decay­
ing infectious tissues might also contribute to the bacteria’s persistence 
Environmental reservoir?

under certain conditions. Another hypothesis for maintenance of F. 
tularensis is persistence in water with protozoa such as amoeba serving 
as host cells.
Since the 1940s, research has shown that F. tularensis can infect 
multiple species of soft and hard ticks, supporting the idea of tularemia 
as an obligate vector-borne disease with ticks as reservoirs. In North 
America, tularemia in humans is mainly tick-borne, but the role of 
ticks in maintaining and spreading the disease geographically is more 
debated compared with other pathogens like Babesia and Borrelia; 
these latter pathogens are better adapted for transmission between the 
different life stages of a tick. Contemporary well-designed experiments 
involving mice and American dog ticks have tested F. tularensis type A 
strains from three main genetic clades and a type B strain. They dem­
onstrated rapid F. tularensis acquisition by adult ticks from bacteremic 
mice, an interrupted blood-feeding behavior, and high transmission 
efficiency to uninfected mice (58–89%), indicating that adult ticks can 
serve as highly effective disease vectors to humans during outbreaks. 
However, ticks of the more immature nymph life stage showed low 
survival after F. tularensis ingestion and low transmission efficiency 
(0–13.5%), indicating that these are ineffective disease vectors. Overall, 
both older and newer studies suggest that F. tularensis infection poses 
a high burden for immature tick stages, and the literature is conflicted 
on the presence of transovarial transmission in ticks. It has been 
suggested that ticks, like other blood-feeding arthropods including 
tabanid flies in North America and mosquitoes in the Boreal Forest 
regions of Northern Europe, are important for transmitting disease to 
humans during tularemia epizootics but may not be obligate for main­
taining F. tularensis. To better understand the zoonotic potential and 
the epidemiology seen in humans, a better understanding of how 

F. tularensis is maintained between outbreaks will be required.
■
■PATHOGENESIS
Natural transmission to humans usually occurs through the skin or by 
inhalation of aerosols containing F. tularensis, but ingesting infected 
food or water also may cause infection. An extremely low dose of 
10–50 bacteria inoculated through microtrauma of the skin or inhaled 
into the lungs suffices to cause disease in humans. In contrast, a dose 
in the range of 106–108 bacteria is required to infect humans and other 
primates by the oral route. Tissue-resident macrophages, including 
alveolar macrophages in the lung and Langerhans cells in the skin, are 
believed to be important in the transport of F. tularensis by blood or 
lymph vessels to lymph nodes, spleen, and bone marrow. In the skin, 
F. tularensis undergoes phagocytosis by neutrophils at initial stage of 
infection. After the bacteria have been taken up by tissue-resident 
macrophages and other local phagocytic cells, F. tularensis escapes 
from the phagosome into the cell cytosol, replicates, leaves the cell, and 
disseminates throughout the host. F. tularensis is a “stealth pathogen,” 
implying that the bacterium is able to manipulate host-cell signaling, 
thus rendering itself a safe milieu inside the host cell suitable for its rep­
lication. In experimental oral infection of mice, there is minimal local 
gastrointestinal pathology despite the large infection dose required. 
It remains unknown what cell types mediate F. tularensis to cross the 
intestinal epithelium to cause fatal systemic infection.
Gross and histopathology findings in nonhuman primates experi­
mentally infected by the inhalation route have shown 0.5- to 1-mm 
necrotic lesions with live bacteria in the spleen at day 4 after the 
infection, and subsequent larger foci in liver and lymph nodes at 
days 5–7, which likely mirror disease events and progression in more 
severe human tularemia cases. At day 6 after infection of nonhuman 
primates, randomly scattered subacute abscesses and pyogranulomas 
6–20 mm in diameter have been observed on the lung surfaces. From 
day 7 and onward, granuloma formation and necrotic lesions in tissues 
surrounded by layers of macrophages, lymphocytes, and giant cells are 
typical reaction patterns in lymph nodes, lungs, spleen, kidney, and 
liver, corroborated by historic reports of similar autopsy findings in 
fatal human tularemia.
The immune invasion strategy of F. tularensis—delaying the immune 
response and permitting rapid systemic distribution—has made stud­
ies of early immune responses challenging. The mechanisms behind 

cell-mediated responses against F. tularensis after a primary infection 
or vaccination by the live vaccine strain of F. tularensis, however, are 
well described and require both B and T cells, resulting in antigenspecific recall T cell responses persisting 25–30 years after infection 
or vaccination. Nevertheless, although the bacterium can manipulate 
and delay them, effective early immune responses occur and involve 
complement, antibodies, neutrophils, inducible NO synthase, phago­
cyte oxidase, and cytokines such as interferon γ, tumor necrosis factor 
α, and interleukin 12.

APPROACH TO THE PATIENT
Clinical recognition of tularemia may be challenging, especially 
if the patient does not spontaneously report fever. Sporadic cases 
of patients seeking health care can easily be misdiagnosed with a 
diversity of other conditions, including lung cancer, head and neck 
cancer, lymphoma, tuberculosis, or another disease with lymphoid 
tissue engagement or a suspected tumor. Tularemia may not be 
suspected before extensive and costly medical investigations have 
already been performed. Awareness of the disease and a high index 
of suspicion is critical. Even in high-endemic areas, patients with 
tularemia often initially are mistakenly managed and treated for 
other conditions, including unspecified viral infection, common 
skin and soft tissue infection, common community-acquired pneu­
monia, or undifferentiated fever or sepsis. A careful disease history 
of a patient with tularemia typically reveals an acute fever onset 
episode with influenza-like illness. If the patient seeks care immedi­
ately, an acute fever onset can be easily recollected; if care is sought 
several weeks after initial symptoms, acute onset of fever may not 
be mentioned spontaneously. The acute illness may be described as 
“a severe virus” with high fever, muscle pain, headache, and other 
constitutional symptoms. Fever may be persistent or intermittent. 
A tularemia-focused disease history should include risk factors 
for exposure to F. tularensis and epidemiologic information (e.g., 
household member or friend having similar symptoms and shar­
ing the same exposure). Exposure risk factors or activities include 
wild animal contact (e.g., rodents, rabbits, hares), contact with or 
inhalation of dust contaminated with their urine or feces, recent 
arthropod bites (tick, fly, or mosquitoes), performing brush cutting 
or grass trimming, handling of wood, handling of hay, and walking 
or hunting in the forest or other natural habitats of wildlife species.
CHAPTER 175
Tularemia
When tularemia is suspected it is important to define patient 
factors with relevance for coping with infection, to define the infec­
tion syndrome, and to decide on an appropriate level for patient 
management:
•	 Consider factors that may increase the risk of severe infection 
(e.g., geriatric frailty, immunosuppression, severe lung and/or 
heart function impairment, other significant comorbidities).
•	 By physical examination, determine any anatomic location and 
extent of inflammation (e.g., presence of pneumonitis, lymph 
node inflammation, skin inflammation including minute ulcers 
or pustules, pharyngitis). Collect information on the rate of pro­
gression and judge the severity of infection.
•	 Determine an appropriate level of care (e.g., localized symptoms 
likely suitable for outpatient care, multiorgan involvement/
hemodynamic instability suggesting inpatient care).
■
■CLINICAL MANIFESTATIONS
The most common initial manifestation of tularemia is an influenzalike illness with high fever. The incubation period is usually 3–5 days 
but may range from 1 to 21 days. Inflamed swollen lymphoid tissue is 
typical for tularemia. The initial infection route determines what lym­
phoid tissue is engaged (Fig. 175–2). If bacteria are acquired through 
skin or oral mucous membranes of the pharynx, palpable enlarged 
tender regional lymph node(s) can result. If bacteria are inhaled, the 
result will be lymph node enlargement in the mediastinum. Different 
recognized clinical forms of tularemia are the same for type A and type 
B tularemia and depend on the initial F. tularensis infection route as 
outlined in Table 175-1.

Preauricular
Cervical, tonsillar,
and supraclavicular
Infraclavicular 
Mediastinal
Axillary
Hilar
Spleen
Epitrochlear
and brachial
Inguinal and
femoral
Popliteal
PART 5
Infectious Diseases
FIGURE 175–2  Characteristic sites of lymphoid tissue engagement in tularemia. 
(From https://training.seer.cancer.gov/lymphoma/anatomy/lymph-nodes.html.)
The primary tularemia ulcer is often a small break in the skin, with 
limited signs of inflammation, and may require careful clinical examination to be detected. In some cases, it may not be detectable and the disease form will then be classified as glandular. Red, tender, swollen lymph 
regional nodes are often the most remarkable clinical examination findings in patients with acute ulceroglandular tularemia (Fig. 175–3).
In oropharyngeal and oculoglandular tularemia, there is more often 
pronounced local inflammation at the primary site of infection with 
acute pharyngitis or conjunctivitis, respectively. The respiratory or 
pneumonic form of tularemia acquired by inhalation of F. tularensis 
may, more often than other clinical forms, be bacteremic and progress 
to severe disease requiring inpatient care, particularly in patients aged 
65 and over and those with comorbidities. In the respiratory/pneumonic 
clinical form of tularemia, onset of dry cough may occur more than 7 
TABLE 175-1  Clinical Forms of Tularemia
FORMa
F. TULARENSIS INFECTION ROUTE
Ulceroglandular or 
glandular
Skin inoculation by blood-feeding arthropods or direct 
contact (touching infected animal or F. tularensis 
contaminated material)
Oropharyngeal
Ingestion of contaminated water or food
Oculoglandular
Touching the eye with contaminated fingers or 
exposure to contaminated aerosol
Respiratory/pneumonic
Inhalation of contaminated aerosol
Typhoidal
Unknown (likely by inhalation, rarely by ingestion)
aAdditional less common but well-described clinical invasive manifestations of 
tularemia exist, including primary tularemia meningitis, endocarditis, and bone and 
joint infections. In patients with immunosuppression, tularemia manifestations may 
be indolent with few focal symptoms and signs.
Source: Reproduced with permission from A Tärnvik (ed): WHO Guidelines on Tularaemia: 
Epidemic and Pandemic Alert and Response. Geneva, World Health Organization Press, 
2007.

FIGURE 175–3  Patient presenting with ulceroglandular tularemia 3 weeks after 
onset of high fever and influenza-like illness. Lymph nodes of the left thigh were red, 
tender, and swollen. A tularemia ulcer was present on the left ankle (not shown). 
(Photo from patient care at Region Västerbotten, Sweden.)
days after onset of the influenza-like illness. Bronchoscopy findings may 
include signs of local tracheitis and pneumonitis, and the lymphoid tissue 
engaged will be in the mediastinum or lung hila. “Typhoidal tularemia” is 
an older medical classification term referring to a disease syndrome like 
typhoid fever with prolonged high fever, fatigue, headache, and nausea. It 
is recommended to avoid using this term, if possible, as it is believed that 
historically, most cases of typhoidal tularemia were acquired by inhalation and were, in fact, respiratory/pneumonic tularemia. For type A tularemia, historic disease descriptions suggest that respiratory/pneumonic 
tularemia or meningitis may also occur as secondary manifestations in 
severe cases of ulceroglandular disease. Data from the United States in 
2006–2021 on 1046 cases with primary disease manifestations available 
showed 47% ulceroglandular, 18% glandular, 17% pneumonic or respiratory, 14% typhoidal, 2% oculoglandular, and 2% oropharyngeal disease 
according to CDC statistics.
Routine biochemical testing of blood generally is not helpful in 
tularemia because findings are unspecific and may or may not reveal 
elevated liver enzymes or red blood cells in the urine. In a recent more 
extensive case series from Sweden of type B tularemia of primarily 
outpatients, white blood cell counts remained within normal limits for 
70% of patients, and the median C-reactive protein level was moderately elevated, peaking at day 7–9 after disease onset at a median of 
78 mg/L (range, 30–130). A recent analysis of 33 patients hospitalized 
with severe respiratory type B tularemia showed peak C-reactive protein levels at 100–400 mg/L among patients with F. tularensis growth 
in blood cultures and 50–220 mg/L among nonbacteremic patients.
■
■DIAGNOSIS
Laboratory confirmation is recommended and may be based on 
recovery of an F. tularensis isolate, F. tularensis–specific antigens or 
nucleic acids, or the detection of F. tularensis–specific antibodies in

blood by serology testing. Appropriate clinical specimens to collect 
for direct detection of F. tularensis by nucleic acids or culture include 
ulcer specimens, lower respiratory tract specimens (sputum/tracheal 
washing/bronchoalveolar lavage), lymph node biopsies or aspirates, 
and blood cultures. Additional clinical specimens with F. tularensis 
growth reported include pharyngeal and ventricular washings. Impor­
tantly, sampling from ulcers should be performed from the edge of 
the wound, preferably selecting a swab with a slightly stiffer stick and 
some rubbing of the tissue so that it becomes somewhat blood-tinged. 
If there is a scab, this should be removed before sampling. If culture 
specimens are sent to a clinical microbiology laboratory, a suspicion of 
tularemia must be clearly communicated to the laboratory in advance 
to minimize the risk of laboratory-acquired infections in lab workers.
Direct detection of F. tularensis–specific nucleic acid in clinical speci­
mens by polymerase chain reaction or another detection method is a 
rapid and reliable diagnostic method. For ulcer specimens, the method 
is comparable with culture in sensitivity and specificity. There is clinical 
experience of F. tularensis nucleic acid detection from other specimen 
types, including sputum, bronchoalveolar lavage, and lymph node biop­
sies/aspirates, but formal scientific evaluation of the diagnostic perfor­
mance is lacking. With increased availability of molecular methods in 
some clinical microbiology laboratories, nucleic acid–based subtyping 
without a prior culture step can be performed directly from clinical 
specimens and may help clinicians to identify F. tularensis varieties 
rapidly. Identification of the subspecies and genetic variety involved in 
tularemia infections is currently limited in most laboratories but may be 
of greatest importance in North America, where type A and type B tula­
remia exist side by side. Cultivation of F. tularensis requires Biosafety 
Level 3 (BSL-3) laboratories, and because F. tularensis is a U.S. Depart­
ment of Health and Human Services (HHS) Tier 1 Select Agent, an 
accurate culture diagnosis in the clinic is challenging. However, culture 
identification provides a definitive diagnosis and allows for subsequent 
genetic methods to identify different F. tularensis varieties with varying 
risks of mortality. Unexpected findings in blood cultures with matrixassisted laser desorption/ionization–time-of-flight (MALDI-TOF) mass 
spectrometry identification of F. tularensis is a scenario that has become 
more common in recent years with the improvement of blood culture 
systems and MALDI-TOF reference libraries. F. tularensis is a slow-growing 
bacterium, and an incubation period of 10 days is recommended for 
optimal sensitivity in blood-culture diagnostics. Serology remains 
commonly used to confirm tularemia and is the mainstay diagnostic in 
many clinical microbiology laboratories. Serology is particularly useful 
in patients when tularemia is suspected as a differential diagnosis at a 
later stage—the antibody responses against F. tularensis are generally 
detectable in the blood of patients 10–20 days post-infection using 
an enzyme-linked immunosorbent assay or a microagglutination test. 
Some commercial serologic tests may detect F. tularensis–specific anti­
bodies as early as 1–2 weeks after disease onset. Importantly, the persis­
tence of F. tularensis antibodies is prolonged, meaning that up to at least 
1 year after acute infection, single titers of antibodies remain above the 
cutoff set by laboratories to detect acute infections. Therefore, a reliable 
serologic diagnosis of tularemia in the acute phase should rely on the 
demonstration of a significant increase of antibody titers between two 
samples taken 2–4 weeks apart.
Chest imaging is useful in respiratory tularemia. Single or multifocal 
consolidation may be present, accompanied by pleural effusion with 
hilar and mediastinal lymphadenopathy (Fig. 175–4).
Computed tomography (CT) scans in the acute phase may reveal 
multiple rounded consolidations, often in the subpleural region and 
sometimes with signs of inclusion necrosis with a diffuse marginal 
zone. Lymph node enlargements in the mediastinal and hilar regions 
on a CT scan may also contain necrotic inclusions. A relatively com­
mon scenario for sporadic cases of tularemia is pulmonary nodules or 
dense consolidations, which can mimic lung cancer, resulting in more 
extensive clinical investigation paths to exclude cancer, including CT 
and positron emission tomography (PET) scans (Fig. 175–5). Endo­
bronchial ultrasound bronchoscopy procedures with biopsy may be 
performed in some cases before tularemia is suspected. The resolution 
of tularemia pathology on chest imaging may be slow and take months.

FIGURE 175–4  Frontal projection chest x-ray with acute respiratory/pneumonic 
tularemia (type B) with consolidation at the right side of the lungs with hilar 
enlargement. Fever and chills started 10 days before. (Photo from patient care at 
Region Västerbotten, Sweden.)
CHAPTER 175
■
■PATHOLOGY
Analyses of tularemia skin lesions of humans infected by skinning of 
rabbits with tularemia have suggested a scenario with accumulation of 
lymphocytes, plasma cells, and neutrophils at the infection site during 
the first week after F. tularensis infection of the skin. Microabscesses with 
neutrophils occur in subcutis along with histiocytes and Langerhans 
cells in dilated lymph vessels. A skin ulcer and granuloma formation 
occur only after the first week, along with lymph node engagement. 
Histopathology findings in other tissues of patients with more advanced 
tularemia and a more extended disease history include irregular micro­
abscesses and granulomas in samples from liver, spleen, kidney, and 
lymph nodes. Autopsy samples from the lungs of humans have shown 
necrotizing pneumonia characterized by abundant fibrin, cellular debris, 
and neutrophils within alveolar walls and alveolar spaces.
Tularemia
TREATMENT
Specific antimicrobial treatment is highly recommended for tula­
remia as outlined in Table 175-2. The antibiotics of choice are cip­
rofloxacin, levofloxacin, gentamicin, or doxycycline. Early specific 
antimicrobial treatment is essential to avoid complications irrespec­
tive of targeting type A or type B tularemia. Several commonly used 
antimicrobials, including all the β-lactam drugs and clindamycin, 
lack treatment efficacy against F. tularensis. Macrolide antibiotics, 
including erythromycin and azithromycin, show an in vitro effect 
against F. tularensis type A and type B varieties known to be pres­
ent in the United States, but globally there are very scarce treatment 
outcome data for these drugs in humans. Generally, macrolide 
antibiotics are not recommended for tularemia because macrolideresistant type B strains are common in Europe and Asia, resulting 
in total lack of treatment effect and overall little experience using 
these drugs for tularemia treatment. Azithromycin, however, may 
be an alternative treatment choice for patients with verified type A 
infection according to recent CDC guidelines.
Streptomycin, classically used to treat tularemia, has been the 
drug of choice due to its high cure rate in a U.S. case series 
reported from 1949–1988, but is now considered a less attractive

A
PART 5
Infectious Diseases
B
C
FIGURE 175–5  Dense consolidation in the left lung of patient with a history of weight 
loss and cough for 3–4 months. A. CT scan showed dense consolidation (2.5 × 2.5 × 
2.0 cm) in the basal left upper lobe extending to the apical part of the lower lobe and 
increased hilar lymphoid tissue. B. Eight days later, PET scan showed increased 
metabolic activity and progress of the consolidation to 5.0 × 4.5 × 3.5 cm. Tissue 
biopsy revealed granuloma formation in the adjacent pleura and no malignant cells. 
C. A tularemia diagnosis was confirmed by serology, the patient was treated with 
ciprofloxacin, and a CT scan performed 2 months later showed resolution of the 
consolidation. (Photos from patient care at Region Västerbotten, Sweden.)
alternative to gentamicin due to its drawbacks, including the need 
for intramuscular administration, limited clinical availability, and 
a higher risk of vestibular and renal toxicity compared with newer 
aminoglycoside antibiotics. Chloramphenicol is another drug with 
a higher risk of adverse effects that, for this reason, is seldom used 
systemically in modern medicine. Chloramphenicol has an in vitro 
effect against F. tularensis and was reportedly effective in treating 
tularemia meningitis in the past.
■
■COMPLICATIONS
Inflamed lymph nodes can progress into an abscess with local and/
or systemic reaction and may occur in about one-third of patients, 
especially if there were delays in F. tularensis–specific antibiotic treatment. An antibiotic treatment delay of >2–3 weeks after disease onset 

TABLE 175-2  Treatment and Postexposure Prophylaxis 
Recommendations for Adult Patients with Tularemia or with Significant 
Exposure to an F. tularensis Aerosol
ADULTS, INCLUDING PREGNANT WOMEN
 
DRUG
DOSEa
DURATION 
(DAYS)
Preferred 
choices
Ciprofloxacin
400 mg every 8 hrs IV or 
750 mg every 12 hrs PO 

Levofloxacin
750 mg every 24 hrs IV 
or PO

Gentamicin
5–7 mg/kg IV or IM per day, 
given every 12 hrs or 24 hrs

Alternative 
choiceb
Doxycyclinec
200 mg loading dose, then 
100 mg every 12 hrs IV 
or PO
14–21
Postexposure 
prophylaxisd
Doxycycline
100 mg PO twice daily

Ciprofloxacin
500 mg every 12 hrs PO

aPatients beginning with IV treatment of ciprofloxacin, levofloxacin, or doxycycline 
can switch to oral administration when clinically indicated. bOther alternative 
choices having less supporting clinical data are high-dose moxifloxacin, ofloxacin, 
amikacin, tobramycin, or plazomycin for 10 days. In verified type A-infection, 
azitromycin may be considered based on in vitro susceptibility analysis. 
cDoxycycline may safely be included as a preferred choice for patients with 
nonsevere disease but is associated with a higher frequency of fever relapse 
and needs a more extended treatment duration. Tetracycline 500 mg every 6 hrs 
IV or PO for 14-21 days is equivalent to doxycycline. dFor low-risk exposures, 
preparedness for early treatment may suffice using daily “fever watch”: the patient 
monitors temperature with instructions to seek immediate treatment if developing a 
fever. Postexposure prophylaxis is recommended for high-risk exposures, including 
performing aerosol-generating procedures with F. tularensis cultures outside a 
biologic safety cabinet or handling infected animals. There are extensive human 
data showing that doxycycline is effective.
Source: Adapted from J Terriquez, C Nelson (eds): Clinical infectious diseases: 
Tularemia: Update on treatment and clinical findings. Clin Infect Dis 78:s1, 2024.
has been statistically associated with lymph node abscess formation. In 
some cases, the progression to a lymph node abscess may occur despite 
ongoing appropriate antibiotic treatment initiated late in the disease 
course. It is unclear if extending the antibiotic treatment duration is of 
clinical benefit. Culture specimens from an abscess at this disease stage 
are typically F. tularensis negative, but an additional 10–day course of 
appropriate antibiotic treatment is often used to avoid further disease 
progression and may be successful. If the clinician considers that there 
is a clear and immediate risk of spontaneous rupture of the skin or 
the mucus membranes in the pharynx with suppuration of pus, open 
surgical drainage or needle aspiration is recommended for treatment. 
A more invasive procedure with total surgical excision may be needed 
in more complicated and long-standing cases. Another complication 
of uncertain genesis occasionally reported by patients includes postinfectious persisting fatigue and disability. Additional complications or 
disease manifestations of tularemia have been reported, including adult 
respiratory distress syndrome, myocarditis, pericarditis, endocarditis, 
meningitis, osteomyelitis, hepatitis, and renal failure.
■
■PROGNOSIS
Prior to the use of antibiotics to treat tularemia, case fatality rates in 
the United States were typically 5%–15% and could be as high as 60% 
for patients with respiratory/pneumonic tularemia. Recent surveillance 
data of tularemia in the United States between 2006 and 2021 show 
that of 903 patients with illness outcome available, 27 patients (3.0%) 
died. Culture of F. tularensis from a clinical specimen was present in 
approximately 50% of cases and associated with significantly reduced 
odds of survival (odds ratio [OR], 0.1; 95% confidence interval [CI], 
0.04–0.4). Treatment with at least one high-efficacy antimicrobial drug 
class (aminoglycoside, fluoroquinolone, or tetracycline) was independently associated with increased odds for survival (adjusted OR, 10.4; 
95% CI, 4.4–24.5 after controlling for disease severity). Notably, previous surveillance data from 1964–2004 considering the modern genetic 
distinction of three epidemiologically important type A genetic varieties in the United States found a 24% fatality outcome for the most lethal