Pediatric Tularemia 

A deadly, and long-lasting epidemic (35-40 years), known as the Hittite Plague, traced back to a focus in Canaan along the Arwad-Euphrates trading route, plagued the Eastern Mediterranean in the 14th century BC. Described in Egyptian royal archives, patient symptoms, mode of infection, and geographical focus historically point to the agent as Francisella tularensis. Egypt and Anatolia were spared due to quarantine and political boundaries. Subsequently, wars spread the disease to central Anatolia, from where it was spread to Western Anatolia by the deliberate use of contaminated animals in what is hypothesized to be the first documented incident of biological warfare.^[1]

In more modern times, hare-associated disease compatible with tularemia had been described in Japan as early as 1818, and Homma-Soken provided the first written description of tularemia in 1837 as a febrile illness with generalized lymphadenopathy among people who had eaten infected rabbit meat. McCoy first isolated the causative organism in 1911, following an outbreak of a plaguelike disease among ground squirrels in the extensive reed marshes of Tulare County, California.^[2] In 1912, McCoy and Chapin cultured the organism, which they named Bacterium tularense.^[3]

The first bacteriologically confirmed case of tularemia was an ocular infection reported by Vail in 1914. Wherry and Lamb demonstrated the organism in wild rabbits living near the home of a patient with culture-proven tularemia conjunctivitis.^[4] The genus was renamed Francisella in honor of Edward Francis, whose extensive investigation of tularemia expanded the bacteriologic, syndromic, and epidemiologic understanding of the disease.^[5]

Francisella tularensis is one of the most infectious bacteria known, as it can cause illness in humans with exposure to as few as 10-50 organisms. F tularensis is considered a category A biowarfare agent due to its high infectivity, ease of dissemination, and ability to cause substantial illness and death.^{[6, 7]}

Investigation of F tularensis and its use as a biologic weapon began in World War II during the Japanese occupation of Manchuria. F tularensis was weaponized and stockpiled by the United States until President Nixon terminated the program and stores maintained by the United States military were destroyed. According to former Director of the Soviet Bioweapons program, Dr Ken Alibek, the former Soviet Union produced and maintained strains that were resistant to antibiotics and vaccines.^[7] Due to US concerns regarding its potential use as a biowarfare agent, tularemia was reinstated on the list of reportable diseases in 2000.

Tularemia was quite common in the United States before World War II. Since the 1950s, the incidence had been steadily declining and had remained at fewer than 0.15 cases per 100,000 population since 1965.^[8] Due to this declining incidence, tularemia had been removed from the reportable disease list in 1995, although outbreaks and sporadic cases continued to occur worldwide.^{[9, 10]}

The World Health Organization (WHO) conducted modeling studies in 1970 on the possible use of F tularensis as an aerosolized bioweapon. The WHO estimated that an aerosol dispersal of 50 kg of virulent F tularensis over a metropolitan area with 5 million inhabitants would result in 250,000 incapacitating casualties, including 19,000 fatalities. This dispersal would also result in relapses occurring for many months after the initial exposure and the potential to establish enzootic reservoirs of tularemia in wild animals leading to possible subsequent outbreaks.

F tularensis is a facultative intracellular, aerobic, gram-negative, catalase positive, nonmotile, pleomorphic coccobacillus. It is capable of growing within several different cell types, including macrophages, endothelial cells, and hepatocytes.^{[11, 12]}

Tularemia is a zoonosis caused by the 2 species of F tularensis in the family Francisellaceae. Four subspecies of Francisella tularensis have been described, all of which have been associated with human disease, but only the F tularensis biovar tularensis ( Jellison Type A ) and F tularensis biovar holarctica (F tularensis biovar palaearctica or Jellison Type B) are common causes.

F tularensis biovar tularensis is found predominantly in North America, although cases have been described in Europe, and is an extremely virulent organism. As few as 10-50 organisms may result in disease if inhaled or injected intradermally, whereas oral ingestion would require as many as 10⁸ organisms. F tularensis biovar holarctica is found primarily in Europe and Asia but has also been identified in cases of tularemia in North America. It is from this less virulent strain to humans and rabbits that the live virus vaccine strains are derived. F tularensis biovars novicida and mediasiatica are of low virulence. A Japanese F tularensis biovar holarctica variant japonica has been more recently characterized.

Traditionally, Francisella subspecies have been characterized on the basis of biochemical reactions, growth characteristics, and virulence properties; however, biochemical methods for differentiating between the subspecies have been found imprecise and newer molecular typing methodology has advanced classification of these organisms.

Tularemia is widely distributed; however, it is largely found in the northern hemisphere from 30-71° north latitude. Tularemia is most commonly described in the western and south central regions of the United States and in continental Europe and Asia. Transmission of tularemia to humans most often occurs through an insect bite or contact with contaminated animals or animal products during the processing of animal meat, skin and eating infected animals.^[13] Thorough cooking is felt to lessen the likelihood of transmission.

Transmission has also been described via ingestion of or contact with contaminated water, exposure to contaminated mud, animal bites, and exposure to aerosolized water droplets or dust from contaminated soil or grains. Francisella has been shown to survive for prolonged periods of time in frozen water, mud or animal carcasses.^[10] Carnivores, such as domestic cats, may transiently harbor Francisella in their mouths or on their claws after killing or feeding on infected prey, irrespective of whether they become infected. Human to human spread is not thought to occur. Cases have developed in laboratory workers, who should be notified in advance to safely handle specimens when tularemia is suspected.

F tularensis is capable of infecting hundreds of different invertebrate, aquatic, and terrestrial vertebrate species, including lagomorphs, rodents, ticks, mosquitos, and flies. In any geographic region, usually no more than a dozen mammals are important to its ecology; however, the overall ecology of F tularensis remains poorly characterized, particularly transmission cycles and specific differences between the 4 different subspecies.^[14]

Lagomorphs, including Sylvilagus and Lepus species, have been historically recognized as common sources of transmission, hence the common names wild hare disease and rabbit skinners' disease. In North America, squirrels, muskrat, beavers, and voles have also been identified as natural reservoirs of organism. In the former Soviet Union, in addition to hares, hamsters, voles, water rats and mice have been shown to carry F tularensis. F tularensis live virus strain was demonstrated to infect amoebic cysts, multiply intracellularly in Acanthamoeba castellanii, and that co-culture of both leads to increased Francisella growth, which may provide natural aquatic reservoirs for the transmission of tularemia.

Bloodfeeding flies and arthropods are the most important vectors for tularemia in the United States. Biting flies, such as deerflies (Chrysops discalis), are the predominant vectors in the far western states, while ticks, primarily Amblyomma americanum, Dermacentor andersoni, and Dermacentor variabilis, are important vectors from the Rocky Mountains eastward. In Northern Europe and the former Soviet Union, mosquitos serve as the most important insect vector. Ticks serve as a particularly important reservoir and vector, as at least 13 different species have been found to be infected with F tularensis, and because vertical transmission of the organism transovarially has been demonstrated. The organism may be present in tick feces or saliva and can be inoculated directly or indirectly into the bite wound.

The cell wall of F tularensis possesses high levels of fatty acids and wild strains have an electron-transparent lipid-rich capsule. Loss of this capsule may result in loss of serum resistance and virulence; however, the capsule exhibits no innate immunogenicity or toxicity.

A subcutaneous inoculum of 10 organisms is sufficient to induce disease, whereas an inhalational exposure of 25 organisms may cause a severely debilitating or fatal disease. Over the first 3-5 days after cutaneous exposure, the organism multiplies locally and a papule forms. During the next 2-4 days, the site ulcerates. Organisms spread from the entry site to regional lymph nodes and may disseminate lympho-hematogenously to involve multiple organs. Patients are most likely bacteremic at this time, although it is not usually detected.

Infection produces an acute inflammatory response initially involving local macrophages, neutrophils, fibrin, and then T lymphocytes, epithelioid cells, and giant cells migrate into local necrotic tissue. As the area of necrosis expands, thrombosis of adjacent arteries and veins may occur. Granulomas develop, which may caseate and be mistaken for tuberculosis. Necrotic foci may coalesce to form abscesses. These changes occur in infected sites and have been demonstrated on autopsy in lymph nodes, liver, spleen, bone marrow, and the lungs.^[15] Organisms may remain viable for prolonged periods of time.

During the second to third week of Francisella infection, humoral immunity develops against the organism's carbohydrate antigens. Agglutinating IgM, IgG, and IgA antibodies are seen at this time.^{[16, 12]} Opsonizing IgG and IgM antibodies are also produced, which act in conjunction with complement (C3). B-cell deficient mice have been shown to have impaired clearance of organisms after primary infection with the less virulent vaccine strain of Francisella.^[17]

Alpha/beta T-cell dependent immunity, involving either CD4+ or CD8+ T cells, and directed against protein antigens, has been demonstrated to be necessary for effective eradication of Francisella.^{[11, 12]} Tularemia antigens, whether introduced by natural exposure or vaccination, have been shown to elicit a vigorous and long lasting T_M response in humans. Some of these T_M cells have lytic potential and they demonstrate the ability to enter both intestinal mucosal and nonmucosal lung sites.^[18]

Preformed molecules on the surface of Francisella trigger rearrangements of the host cell cytoskeleton. Macrophage complement receptors interact with complement factor C3 fixed by molecules on the bacterial cell surface and bacteria are internalized by looping phagocytosis. Immediately after phagocytosis, the bacteria are housed in a nonacidified phagosome. Bacteria may degrade the phagosomal membrane and escape into the cytoplasm where they actively multiply. This can lead to cell death and liberation of bacteria. In rodent macrophages, bacterial survival has been shown to be associated with failure of phagosome-lysosome fusion, phagosome acidification, and utilization of host iron.^[19]

Lack or minimal amounts of CD14 on human dendritic cells in the lung appears to contribute to F tularensis evasion of immune response.^[20] Interferon gamma (IFN gamma) and TNF alpha activate macrophages to kill Francisella through the production of reactive nitrogen products such as nitric oxide (NO).^[21] Neutrophils and mononuclear cells have been demonstrated to accumulate at infected liver foci and lyse Francisella -containing hepatocytes, releasing organisms from their relatively protected sequestered environment.^[22] The ability of F tularensis to impair phagocyte function and survive in infected cells is central to its virulence. This intracellular life cycle has been shown to be related to the tightly regulated expression of a series of genes.^{[23, 24]}

In the unique environment of the lung, oxygen-dependent neutrophil killing of wild virulent strains appears to be only partially effective.^[25] Shortly after inhalation, Francisella are found inside cells that typically act as cytokine-producing first responders to infection, including airway macrophages and alveolar epithelial cells. Mouse experiments demonstrate lack of production of pro-inflammatory cytokines, including IL-12p40, TNF, IL-6, and IL-1 alpha. The exact mediators responsible for immunosuppression remain unclear; however, immunomodulatory factors such as TGF-beta and PGE-2 may be actively involved.

Interestingly, approximately 48-72 hours postinfection, a number of cytokines and chemokines, such as IFN-gamma and TNF, are upregulated and levels of proinflammatory mediators such as RANTES, IL-6, and IL-1 beta are detected. Unfortunately, the lung may contain more than 10⁸ CFU bacteria and this upregulation may be too late to prevent death.^{[26, 11]} This late "cytokine storm," similar to that demonstrated in other causes of severe bacterial sepsis, may actually prove harmful to the host, causing capillary leakage, tissue injury, and lethal organ failure.^{[27, 28]}

The incubation period for tularemia depends on the size of the inoculum, but ranges from 1-21 days (average 2-6 d). Individuals with tularemia may be asymptomatic or acutely septic with rapid death. Six clinical forms of tularemia have been identified (see History). Each form is influenced by factors related to the host, organism, and route of transmission and host entry site.

Frequency

United States

Approximately 200 cases of tularemia are reported annually. In the United States, transmission had historically been most frequent from June through August and in December. The summer peak has been felt to be due to insect bites, while the winter peak has been attributed to hunting-associated cases. In more recent years, peak reporting has occurred in late spring and early summer.^[8] From 1990-2000, 56% of cases reported in the United States were from Arkansas, Missouri, Oklahoma, and South Dakota, with high numbers of cases also reported in Massachusetts, Kansas, and Montana.^[19] High incidence rates have been reported by indigenous Alaskans and Native Americans.^[8]

Occupations and avocations associated with an increased risk of tularemia include laboratory work, veterinary practice, farming, landscaping, working with sheep, hunting and trapping, and meat handling.

International

Tularemia is found around the world, distributed within 30-71° northern latitude, but its incidence is unknown.

Mortality/Morbidity

Untreated, the mortality rate from tularemia is 5-30%, with the highest rate occurring with the typhoidal (systemic) form, particularly when accompanied by tularemia pneumonia. The mortality rate also depends on the strain involved; type A is significantly more virulent and is responsible for almost all reported deaths. The mortality rate is less than 1% with appropriate antibiotic therapy.

Race

No racial predilection is reported.

Sex

Male individuals are more frequently affected than female individuals, despite the lack of biologic affinity. This distribution primarily results from increased exposure to specific activities (eg, hunting and skinning animals) and increased occupational vulnerability among male individuals.

Age

Tularemia can be seen in individuals of any age; however, bimodal age peaks have been reported in the last decade, with incidence highest in those aged 5-9 years and those older than 75 years.^[8]