Tularemia As a Biological Weapon Medical and Public Health Management

CONSENSUS STATEMENT

Tularemia as a Biological Weapon Medical and Public Health Management

David T. Dennis, MD, MPH Objective The Working Group on Civilian Biodefense has developed consensus- Thomas V. Inglesby, MD based recommendations for measures to be taken by medical and public health professionals if tularemia is used as a biological weapon against a civilian population. Donald A. Henderson, MD, MPH Participants The working group included 25 representatives from academic medi- John G. Bartlett, MD cal centers, civilian and military governmental agencies, and other public health and Michael S. Ascher, MD emergency management institutions and agencies. Edward Eitzen, MD, MPH Evidence MEDLINE databases were searched from January 1966 to October 2000, using the Medical Subject Headings Francisella tularensis, Pasteurella tularensis, bio- Anne D. Fine, MD logical weapon, biological terrorism, bioterrorism, biological warfare, and biowar- Arthur M. Friedlander, MD fare. Review of these references led to identification of relevant materials published prior to 1966. In addition, participants identified other references and sources. Jerome Hauer, MHS Consensus Process Three formal drafts of the statement that synthesized infor- Marcelle Layton, MD mation obtained in the formal evidence-gathering process were reviewed by mem- Scott R. Lillibridge, MD bers of the working group. Consensus was achieved on the final draft. Joseph E. McDade, PhD Conclusions A weapon using airborne tularemia would likely result 3 to 5 days later in an outbreak of acute, undifferentiated febrile illness with incipient pneumonia, pleu- Michael T. Osterholm, PhD, MPH ritis, and hilar lymphadenopathy. Specific epidemiological, clinical, and microbiologi- Tara O’Toole, MD, MPH cal findings should lead to early suspicion of intentional tularemia in an alert health system; laboratory confirmation of agent could be delayed. Without treatment, the Gerald Parker, PhD, DVM clinical course could progress to respiratory failure, shock, and death. Prompt treat- Trish M. Perl, MD, MSc ment with streptomycin, gentamicin, doxycycline, or ciprofloxacin is recommended. Prophylactic use of doxycycline or ciprofloxacin may be useful in the early postexpo- Philip K. Russell, MD sure period. Kevin Tonat, DrPH, MPH JAMA. 2001;285:2763-2773 www.jama.com for the Working Group on Civilian Biodefense infected through diverse environmen- CONSENSUS METHODS tal exposures and can develop severe and The working group comprised 25 rep- I know of no other infection of animals communicable to man that can be acquired from sometimes fatal illness but do not trans- resentatives from academic medical sources so numerous and so diverse. In short, mit infection to others. The Working centers, civilian and military govern- one can but feel that the status of tularemia, Group on Civilian Biodefense consid- mental agencies, and other public health both as a disease in nature and of man, is one ers F tularensis to be a dangerous poten- and emergency management institu- of potentiality. R. R. Parker1 tial biological weapon because of its ex- tions. This group followed a specified treme infectivity, ease of dissemination, process in developing a consensus state- ULAREMIA, A BACTERIAL ZOONO- and substantial capacity to cause illness ment. MEDLINE databases from Janu- sis, is the subject of this fifth ar- and death.8-11 ary 1966 to October 2000 were searched ticle in a series providing rec-

ommendations for medical and Author Affiliations: National Center for Infectious Dis- (Drs Fine and Layton), and Kroll Associates (Mr Hauer), Tpublic health management following use eases, Centers for Disease Control and Prevention, New York, NY; ican Inc, Eden Prairie, Minn (Dr Oster- of various agents as biological weapons Atlanta, Ga (Drs Dennis, Lillibridge, and McDade); Cen- holm); and Office of Emergency Preparedness, Depart- 2-5 ter for Civilian Biodefense Studies, Johns Hopkins Uni- ment of Health and Human Services, Rockville, Md (Dr of terrorism. The causative agent of tu- versity Schools of Medicine (Drs Inglesby, Bartlett, and Tonat). laremia, Francisella tularensis, is one of Perl) and Public Health (Drs Henderson, O’Toole, and Ex Officio Participants in the Working Group on Ci- Russell), Baltimore, Md; Viral and Rickettsial Diseases vilian Biodefense are listed at the end of this article. the most infectious pathogenic bacteria Laboratory, California Department of Health Services, Corresponding Author and Reprints: David T. Den- known, requiring inoculation or inha- Berkeley (Dr Ascher); US Army Medical Research Insti- nis, MD, MPH, Division of Vector-Borne Infectious Dis- tute of Infectious Diseases, Ft Detrick, Md (Drs Eitzen, eases, National Center for Infectious Diseases, Cen- lation of as few as 10 organisms to cause Friedlander, and Parker); Bureau of Communicable ters for Disease Control and Prevention, PO Box 2087, disease.6,7 Humans become incidentally Disease, New York City Health Department Fort Collins, CO 80522 (e-mail: [email protected]).

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON using the Medical Subject Headings 1950s and 1960s, the US military devel- larensis and other potential biological Francisella tularensis, Pasteurella tu- oped weapons that would disseminate warfare agents to better protect the US larensis, biological weapon, biological ter- F tularensis aerosols10; concurrently, it military, including protocols on decon- rorism, bioterrorism, biological war- conducted research to better under- tamination, prophylaxis, clinical rec- fare, and biowarfare. Review of the stand the pathophysiology of tulare- ognition, laboratory diagnosis, and bibliographies of these references led mia and to develop vaccines and anti- medical management.29 The CDC op- to identification of relevant materials biotic prophylaxis and treatment erates a national program for bioter- published prior to 1966. In addition, regimens. In some studies, volunteers rorism preparedness and response that participants identified other pub- were infected with F tularensis by di- incorporates a broad range of public lished and unpublished references and rect aerosol delivery systems and by ex- health partnerships.30,31 sources for review. posures in an aerosol chamber.10 A live The first draft of the consensus state- attenuated vaccine was developed that EPIDEMIOLOGY ment was a synthesis of information ob- partially protected against respiratory Geographic Distribution tained in the formal evidence-gathering and intracutaneous challenges with the and Human Exposures process. Members of the working group virulent SCHU S-4 strain of F tularen- Tularemia occurs throughout much of were asked to make written comments sis,6,7 and various regimens of strepto- North America and Eurasia.15,21,22,32 In on this first draft in May 1999. Subse- mycin, tetracyclines, and chlorampheni- the United States, human cases have quent revised drafts were reviewed and col were found to be effective in been reported from every state except edited until full consensus of the work- prophylaxis and treatment.25-27 By the Hawaii; however, most cases occur in ing group was achieved. late 1960s, F tularensis was one of sev- south-central and western states (es- eral biological weapons stockpiled by the pecially Missouri, Arkansas, Okla- HISTORY AND POTENTIAL US military.10 According to Alibeck, a homa, South Dakota, and Montana).33-35 AS A BIOLOGICAL WEAPON large parallel effort by the Soviet Union In Eurasia, the disease is also widely en- Tularemia was first described as a plague- continued into the early 1990s and re- demic, although the greatest numbers like disease of rodents in 1911 and, sulted in weapons production of F tu- of human cases are reported from shortly thereafter, was recognized as a po- larensis strains engineered to be resis- northern and central Europe, espe- tentially severe and fatal illness in hu- tant to antibiotics and vaccines.24 cially Scandinavian countries and those mans.12 Tularemia’s epidemic potential In 1969, a World Health Organiza- of the former Soviet Union.36,37 Tula- became apparent in the 1930s and 1940s, tion expert committee estimated that an remia is almost entirely a rural dis- when large waterborne outbreaks oc- aerosol dispersal of 50 kg of virulent ease, although urban and suburban ex- curred in Europe and the Soviet F tularensis over a metropolitan area with posures occasionally do occur.38-41 Union13-15 and epizootic-associated cases 5 million inhabitants would result in Throughout its range, F tularensis is occurred in the United States.16,17 As well, 250000 incapacitating casualties, includ- found in widely diverse animal hosts F tularensis quickly gained notoriety as ing 19000 deaths.28 Illness would be and habitats and can be recovered from a virulent laboratory hazard.18,19 Public expected to persist for several weeks and contaminated water, soil, and vegeta- health concerns impelled substantial disease relapses to occur during the en- tion.15,20-22,32 A variety of small mam- early investigations into tularemia’s ecol- suing weeks or months. It was assumed mals, including voles, mice, water rats, ogy, microbiology, pathogenicity, and that vaccinated individuals would be squirrels, rabbits, and hares, are natu- prevention.19-22 only partially protected against an aero- ral reservoirs of infection. They ac- Francisella tularensis has long been sol exposure. Referring to this model, the quire infection through bites by ticks, considered a potential biological Centers for Disease Control and Preven- flies, and mosquitoes, and by contact weapon. It was one of a number of agents tion (CDC) recently examined the ex- with contaminated environments. Al- studied at Japanese germ warfare re- pected economic impact of bioterrorist though enzootic cycles of F tularensis search units operating in Manchuria be- attacks and estimated the total base costs typically occur without notice, epizo- tween 1932 and 194523; it was also ex- to society of an F tularensis aerosol otics with sometimes extensive die- amined for military purposes in the attack to be $5.4 billion for every 100000 offs of animal hosts may herald out- West. A former Soviet Union biologi- persons exposed.9 breaks of tularemia in humans.16,22,42,43 cal weapons scientist, Ken Alibeck, has The United States terminated its bio- Humans become infected with F tu- suggested that tularemia outbreaks af- logical weapons development pro- larensis by various modes, including fecting tens of thousands of Soviet and gram by executive order in 1970 and, bites by infective arthropods,42,44-47 han- German soldiers on the eastern Euro- by 1973, had destroyed its entire bio- dling infectious animal tissues or flu- pean front during World War II may logical arsenal.10 Since then, the US ids,17,48,49 direct contact with or inges- have been the result of intentional use.24 Army Medical Research Institute of In- tion of contaminated water, food, or Following the war, there were continu- fectious Diseases has been responsible soil,13,20,40,50,51 and inhalation of infec- ing military studies of tularemia. In the for defensive medical research on F tu- tive aerosols.43,52-56 Persons of all ages

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON and both sexes appear to be equally (F tularensis biovar palaearctica) [type symptoms rather than specific mani- susceptible to tularemia. Certain ac- B]), most of whom acquired infection festations of pneumonia; enlargement tivities, such as hunting, trapping, while doing farm work that created con- of hilar nodes was the principal radio- butchering, and farming, are most likely taminated aerosols. Case exposures and graphic finding in these cases.54 to expose adult men. Laboratory disease onsets occurred during a period workers are especially vulnerable to in- of months but peaked during the win- Inhalational Tularemia Following fection, either by accidentally inocu- ter, when rodent-infested hay was Use as a Biological Weapon lating themselves or by inhaling aero- being sorted and moved from field stor- Although F tularensis could be used as solized organisms.18,22,56-58 Ordinary age sites to barns. Among 140 serologi- a weapon in a number of ways, the work- exposures during examination of an cally confirmed cases thought to have ing group believes that an aerosol re- open culture plate can cause infec- been infected by inhalation, most had lease would have the greatest adverse tion. Although F tularensis is highly in- typical acute symptoms of fever, fa- medical and public health conse- fectious and pathogenic, its transmis- tigue, chills, headache, and malaise; only quences. Release in a densely popu- sion from person to person has not been 14 (10%) of confirmed patients had lated area would be expected to result documented. symptoms of pneumonia, such as dys- in an abrupt onset of large numbers of pnea and chest pains. Patients gener- cases of acute, nonspecific febrile ill- Incidence ally responded well to tetracycline, and ness beginning 3 to 5 days later (incu- The worldwide incidence of tularemia is no deaths were reported. Inhalational tu- bation range, 1-14 days), with pleuro- not known, and the disease is probably laremia in the United States has in- pneumonitis developing in a significant greatly underrecognized and underre- volved only single cases or small clus- proportion of cases during the ensuing ported. In the United States, reported ters of cases, variously resulting from days and weeks. Public health authori- cases have dropped sharply from sev- laboratory exposures,18,56,57 disturbance ties would most likely become aware of eral thousand per year prior to 1950 to of contaminated animal carcasses,38,39,41 an outbreak of unusual respiratory dis- less than 200 per year in the 1990s.33-35 and suspected infective environmental ease in its early stages, but this could be Between 1985 and 1992, 1409 cases and aerosols.41,52 Cases of inhalational tula- difficult to distinguish from a natural 20 deaths were reported in the United remia in the United States are thought outbreak of community-acquired infec- States, for a mean of 171 cases per year to be due mostly to the more virulent tion, especially influenza or various and a case-fatality rate of 1.4%.34 Per- F tularensis biovar tularensis (type A) and atypical pneumonias. The abrupt onset sons in all age groups were affected, but usually follow an acute and severe course, of large numbers of acutely ill persons, most were children younger than 10 with prominent pneumonitis. Some the rapid progression in a relatively high years and adults aged 50 years or older. cases, however, have radiographic evi- proportion of cases from upper respira- Of 1298 cases for which information on dence of pleuropneumonia with mini- tory symptoms and bronchitis to life- sex was available, 942 (72.6%) oc- mal or absent respiratory signs on physi- threatening pleuropneumonitis and sys- curred in males, and males outnum- cal examination.39,41,52 temic infection affecting, among others, bered females in all age groups. Most Although airborne F tularensis would young, previously healthy adults and cases occur in June through September, be expected to principally cause pri- children should, however, quickly alert when arthropod-borne transmission is mary pleuropneumonic infection, some medical professionals and public health most common.17,35,59 Cases in winter exposures might contaminate the eye, authorities to a critical and unexpected usually occur among hunters and trap- resulting in ocular tularemia; pen- public health event and to bioterror- pers who handle infected animal car- etrate broken skin, resulting in ulcero- ism as a possible cause (TABLE 1). Un- casses.17,35,48 In the United States, cases glandular or glandular disease; or cause til the etiology became clear, clinicians are mostly sporadic or occur in small oropharyngeal disease with cervical would need to work closely with epide- clusters34,35,49; in Eurasia, waterborne, ar- lymphadenitis. In the aforementioned miologists and diagnostic laboratories to thropod-borne, and airborne outbreaks Swedish outbreak, conjunctivitis was differentiate the illness from various involving hundreds of persons have been reported in 26% of 140 confirmed cases community-acquired pneumonias and reported.40,43,44,51,53-55 and an infected ulcer of the skin was to determine if it could have resulted reported in nearly 12%; pharyngitis was from use of one of several potential bio- Natural Occurrences of reported in 31% and oral ulcers in about terrorism weapons agents, such as those Inhalational Tularemia 9% of the cases; and 32% of these pa- causing tularemia, plague, anthrax, or The largest recorded airborne tulare- tients had various exanthemas, such as Q fever.2,4,29 mia outbreak occurred in 1966-1967 in erythema multiforme and erythema no- In general, tularemia would be ex- an extensive farming area of Sweden.43 dosum.43 Tularemia outbreaks arising pected to have a slower progression of This outbreak involved more than 600 from similar agricultural exposures have illness and a lower case-fatality rate than patients infected with strains of the been reported from Finland,53 mostly either inhalational plague or anthrax. milder European biovar of F tularensis presenting with general constitutional Plague would most likely progress very

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON

from glycerol, and does not demon- Table 1. Diagnosis of Inhalational Tularemia Following Use of a Biological Weapon strate citrulline ureidase activity. In Eu- Clinical Findings rope and Asia, all human tularemia is Sudden onset of acute febrile illness, progressing in some patients to pharyngitis, bronchiolitis, pneumonitis, pleuritis, hilar lymphadenitis. Complications of overwhelming untreated infection thought to be caused by the milder type may lead to sepsis and inflammatory response syndrome. B strains, although recent studies there Epidemiology have identified naturally occurring F tu- Point-source outbreak pattern; likely urban, nonagricultural setting. Unexpected severe respiratory larensis related to F tularensis biovar tu- illness in otherwise healthy persons. Risk related to degree of exposure with no differences in 63,64 susceptibility by age or sex. larensis. A few rapidly growing strains of F tularensis have been recov- Microbiology ered from the blood of immunocom- Small, gram-negative coccobacilli in direct stain of respiratory secretions. Sputum, tracheobronchial secretions, and blood should be cultured using cysteine-enriched medium. Antimicrobial promised patients not showing serore- susceptibility of isolates should be determined. Direct fluorescent antibody stain is first-line, rapid activity to F tularensis.65 identification procedure at reference laboratories. Polymerase chain reaction and antigen detection procedures may also provide rapid identification. Microagglutination assay can detect Transformed plasmids have been en- serum antibodies beginning 10 days after illness onset. Virulence testing and molecular genetic gineered to express chloramphenicol characterizations are performed at specialized laboratories. and tetracycline resistance in F tularen- Pathology sis.66 Virulent, streptomycin-resistant Histological findings of acute suppurative necrosis followed by granulomatous reactions. Target F tularensis strains have been exam- organs include lungs, lymph nodes, spleen, liver, and kidney. ined in biowarfare agent studies both Radiology in the United States and the Soviet Peribronchial infiltrates leading to bronchopneumonia in 1 or more lobes, often accompanied by 24,27,56 pleural effusion and enlarged hilar nodes. Signs may be absent or minimal, with only 1 or several Union. Although F tularensis viru- small, discrete pulmonary infiltrates, or scattered granulomatous lesions of lung parenchyma or lence factors are poorly understood and pleura. characterized,67,68 it is possible that strain virulence could be enhanced rapidly to severe pneumonia, with co- be similar across sex and age groups, through laboratory manipulation. pious watery or purulent sputum pro- and risk would be related to degree of duction, hemoptysis, respiratory insuf- exposure to the point source (Table 1). PATHOGENESIS AND ficiency, sepsis, and shock.4 Inhalational An outbreak of inhalational tularemia CLINICAL MANIFESTATIONS anthrax would be differentiated by its in an urban setting should trigger a high Pathogenesis characteristic radiological findings of level of suspicion of an intentional Francisella tularensis can infect hu- prominent symmetric mediastinal wid- event, since all reported inhalational tu- mans through the skin, mucous mem- ening and absence of bronchopneumo- laremia outbreaks have occurred in ru- branes, gastrointestinal tract, and lungs. nia.2 As well, anthrax patients would be ral areas. It is a facultative intracellular bacte- expected to develop fulminating, toxic, rium that multiplies within macro- and fatal illness despite antibiotic treat- MICROBIOLOGY AND phages.68,69 The major target organs are ment.29 Milder forms of inhalational tu- VIRULENCE FACTORS the lymph nodes, lungs and pleura, laremia could be clinically indistin- Francisella tularensis is a small, non- spleen, liver, and kidney.19,20,49,70-72 Un- guishable from Q fever; establishing a motile, aerobic, gram-negative cocco- treated, bacilli inoculated into skin or diagnosis of either would be problem- bacillus. It has a thin lipopolysaccha- mucous membranes multiply, spread to atic without reference laboratory test- ride-containing envelope and is a hardy the regional lymph nodes and further ing. Presumptive laboratory diag- non–spore-forming organism that sur- multiply, and may then disseminate to noses of plague or anthrax would be vives for weeks at low temperatures in organs throughout the body. Bactere- expected to be made relatively quickly, water, moist soil, hay, straw, and de- mia may be common in the early phase although microbiological confirma- caying animal carcasses.21,22,60,61 Fran- of infection. The initial tissue reaction tion could take days. Isolation and iden- cisella tularensis has been divided into to infection is a focal, intensely suppu- tification of F tularensis using routine 2 major subspecies (biovars) by viru- rative necrosis consisting largely of laboratory procedures could take sev- lence testing, biochemical reactions, and accumulations of polymorphonuclear eral weeks. epidemiological features.62 Francisella leukocytes, followed by invasion of Once a substantial cluster of cases of tularensis biovar tularensis (type A) may macrophages, epithelioid cells, and lym- inhalational tularemia had been iden- be highly virulent in humans and ani- phocytes. Suppurative lesions become tified, epidemiological findings should mals, produces acid from glycerol, dem- granulomatous, and histopathological suggest a bioterrorist event. The abrupt onstrates citrulline ureidase activity, and examination of the granulomas shows onset and single peak of cases would is the most common biovar isolated in a central necrotic, sometimes caseat- implicate a point-source exposure with- North America.22,60 Francisella tularen- ing zone surrounded by a layer of epi- out secondary transmission. Among ex- sis biovar palaearctica (type B) is rela- thelioid cells, multinucleated giant cells, posed persons, attack rates would likely tively avirulent, does not produce acid and fibroblasts in a radial arrange-

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON ment, typical of other granulomatous inhalation, especially when there are Figure 1. Cervical Lymphadenitis in a conditions, such as tuberculosis and signs of pleuropneumonic disease; this Patient With Pharyngeal Tularemia sarcoidosis.20,70,71 usage can be misleading and has been Monkeys that inhaled the virulent discouraged.54,75 SCHU S-4 strain of F tularensis (type A) The onset of tularemia is usually developed acute bronchiolitis within 24 abrupt, with fever (38°C-40°C), head- hours of exposure to 1-µm particles and ache, chills and rigors, generalized body within 48 hours of exposure to 8-µm par- aches (often prominent in the low ticles.73 By 72 hours following chal- back), coryza, and sore throat. A pulse- lenge, inflammation was present in peri- temperature dissociation has been noted bronchial tissues and alveolar septa. in as many as 42% of patients.49 A dry Bronchopneumonia was most pro- or slightly productive cough and sub- nounced in animals exposed to the sternal pain or tightness frequently oc- smaller particles and was characterized cur with or without objective signs of Patient has marked swelling and fluctuant suppura- by tracheobronchial lymph node en- pneumonia, such as purulent sputum, tion of several anterior cervical nodes. Infection was largement and reddish, firm, 0.2- to 0.5- dyspnea, tachypnea, pleuritic pain, or acquired by ingestion of contaminated food or wa- cm-diameter discrete inflammatory le- hemoptysis.7,19,26,70,74 Nausea, vomit- ter. Source: World Health Organization. sions scattered throughout the lungs. In ing, and diarrhea sometimes occur. the absence of treatment, the disease pro- Sweats, fever and chills, progressive der, generally has an indolent charac- gressed to pneumonic consolidation and weakness, malaise, anorexia, and weight ter, and may be covered by an eschar. organization, granuloma formation, and loss characterize the continuing ill- Typically, one or more regional affer- eventual chronic interstitial fibrosis. ness. Studies of volunteers have shown ent lymph nodes may become en- Humans with inhalational expo- that F tularensis aerosol exposures can larged and tender within several days sures also develop hemorrhagic inflam- incapacitate some persons in the first of the appearance of the papule. Even mation of the airways early in the course 1 or 2 days of illness, and significant im- with antibiotic treatment, the affected of illness, which may progress to bron- pairment in performing tasks can con- nodes may become fluctuant and rup- chopneumonia.54 Histopathological ex- tinue for days after antibiotic treat- ture. In oculoglandular tularemia, amination of affected lungs shows al- ment is begun.76 In untreated tularemia, which follows direct contamination veolar spaces filled with an exudate of symptoms often persist for several of the eye, ulceration occurs on the mononuclear cells. Pleuritis with ad- weeks and, sometimes, for months, usu- conjunctiva, accompanied by pro- hesions and effusion and hilar lymph- ally with progressive debility. Any form nounced chemosis, vasculitis, and adenopathy are common radiological of tularemia may be complicated by he- regional lymphadenitis. Glandular tu- and pathological findings.70,72 matogenous spread, resulting in sec- laremia is characterized by lymphade- ondary pleuropneumonia, sepsis, and, nopathy without an ulcer. Clinical Manifestations rarely, meningitis.74,77 Oropharyngeal tularemia is acquired The primary clinical forms of tulare- Prior to the advent of antibiotics, the by drinking contaminated water, ingest- mia vary in severity and presentation overall mortality from infections with ing contaminated food, and, some- according to virulence of the infecting the more severe type A strains was in times, by inhaling contaminated drop- organism, dose, and site of inoculum. the range of 5% to 15%, and fatality lets or aerosols.14,20,36,43,50,51,79 Affected Primary disease presentations include rates as high as 30% to 60% were re- persons may develop stomatitis but more ulceroglandular, glandular, oculoglan- ported for untreated pneumonic and se- commonly develop exudative pharyn- dular, oropharyngeal, pneumonic, ty- vere systemic forms of disease.72,78 Cur- gitis or tonsillitis, sometimes with ul- phoidal, and septic forms.19,20,49,70,72,74,75 rently, the overall case-fatality rate of ceration. Pronounced cervical or retro- The term typhoidal tularemia has been reported cases in the United States is pharyngeal lymphadenopathy may occur used to describe illness in tularemia pa- less than 2%.34,49 Type B infections are (FIGURE 1).74,79 tients with systemic infections mani- rarely fatal. Tularemia pneumonia can be the festing as fever and other constitu- In ulceroglandular tularemia, the direct result of inhaling contaminated tional signs without cutaneous or form that typically arises from han- aerosols or be secondary to hematog- mucosal membrane lesions or re- dling a contaminated carcass or follow- enous spread from a distal site. An gional lymphadenitis. Sometimes, these ing an infective arthropod bite, a local aerosol release of F tularensis would be patients present with prominent gas- cutaneous papule appears at the inocu- expected to result in acute illness with trointestinal manifestations, such as di- lation site at about the time of onset of signs and symptoms of 1 or more of arrhea and pain. Confusion is created generalized symptoms, becomes pus- pharyngitis, bronchiolitis, pleuropneu- when typhoidal tularemia is used to de- tular, and ulcerates within a few days monitis, and hilar lymphadenitis, scribe the illness in patients infected by of its first appearance. The ulcer is ten- accompanied by various manifesta-

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON

more lobes, and often accompanied by cians encountering patients with Figure 2. Chest Radiograph of a Patient With Pulmonary Tularemia pleural effusions and hilar lymphade- findings of atypical pneumonia, pleu- nopathy (FIGURE 2).72,75 Signs may, ritis, and hilar lymphadenopathy. Iden- however, be minimal or absent, and tification of F tularensis in clinical speci- some patients will show only 1 or sev- mens may be missed or delayed for days eral small, discrete pulmonary infil- or weeks when procedures for routine trates or scattered granulomatous microbiological screening of bacterial lesions of lung parenchyma or pleura. pathogens are followed, and it is un- Although volunteers challenged with likely that a serendipitous laboratory aerosols of virulent F tularensis (type identification would be the sentinel A) regularly developed systemic symp- event that alerted authorities to a ma- toms of acute illness 3 to 5 days fol- jor bioterrorism action. lowing exposure, only 25% to 50% of Physicians who suspect inhala- participants had radiological evidence tional tularemia should promptly col- of pneumonia in the early stages of lect specimens of respiratory secre- infection.7,26 On the other hand, pul- tions and blood and alert the laboratory monary infection can sometimes rap- to the need for special diagnostic and idly progress to severe pneumonia, safety procedures. Francisella tularen- 72,80 Infiltrates in left lower lung, tenting of diaphragm, prob- respiratory failure, and death. Lung sis may be identified by direct exami- 75 ably caused by pleural effusion, and enlargement of abscesses occur infrequently. nation of secretions, exudates, or bi- left hilum. Source: Armed Forces Institute of Pathology. Typhoidal tularemia is used to de- opsy specimens using direct fluorescent scribe systemic illness in the absence antibody or immunohistochemical of signs indicating either site of inocu- stains.81-83 By light microscopy, the lation or anatomic localization of in- organism is characterized by its small Box. Clinicians Caring for fection. This should be differentiated size (0.2µmϫ0.2-0.7 µm), pleomor- Patients With Suspected from inhalational tularemia with pleu- phism, and faint staining. It does not Tularemia Should ropneumonic disease.54,75 show the bipolar staining characteris- Immediately Contact Their: Tularemia sepsis is potentially se- tics of Yersinia pestis,4 the agent of (1) Hospital epidemiologist or vere and fatal. As in typhoidal tulare- plague, and is easily distinguished from infection control practitioner and mia, nonspecific findings of fever, ab- the large gram-positive rods character- (2) Local or state health depart- dominal pain, diarrhea, and vomiting istic of vegetative forms of Bacillus ments may be prominent early in the course anthracis (FIGURE 3).2 Microscopic Consult your local telephone op- of illness. The patient typically ap- demonstration of F tularensis using erator, the telephone directory un- pears toxic and may develop confu- fluorescent-labeled antibodies is a rapid der “governmental listings,” or the sion and coma. Unless treated promptly, diagnostic procedure performed in des- Internet at http://www.cdc.gov/other septic shock and other complications ignated reference laboratories in the Na- .htm#states or http://www.astho.org of systemic inflammatory response syn- tional Public Health Laboratory Net- /state.html drome may ensue, including dissemi- work; test results can be made available If the local and state health de- nated intravascular coagulation and within several hours of receiving the ap- partments are unavailable, contact bleeding, acute respiratory distress syn- propriate specimens if the laboratory is the Centers for Disease Control and drome, and organ failure.80 alerted and prepared. Suspicion of in- Prevention at (970) 221-6400 or halational tularemia must be promptly http://www.cdc.gov/ncidod/dvbid DIAGNOSIS /dvbid.htm reported to local or state public health Tularemia in humans occurs infre- authorities so timely epidemiological quently, resulting in a low index of di- and environmental investigations can agnostic suspicion among clinicians and be made (BOX). tions of systemic illness. Inhalational laboratorians. Since rapid diagnostic Growth of F tularensis in culture is exposures, however, commonly result testing for tularemia is not widely avail- the definitive means of confirming the in an initial clinical picture of systemic able, the first indication of intentional diagnosis of tularemia.60,81 Francisella illness without prominent signs of res- tularemia might follow recognition by tularensis can be grown from pharyn- piratory disease.7,43,53,56 The earliest public health authorities of a cluster- geal washings, sputum specimens, and pulmonary radiographic findings of ing of acute, severe respiratory illness even fasting gastric aspirates in a high inhalational tularemia may be peri- with unusual epidemiological fea- proportion of patients with inhala- bronchial infiltrates, typically advanc- tures (Table 1). Suspicion of tulare- tional tularemia.56 It is only occasion- ing to bronchopneumonia in 1 or mia might be triggered in alert clini- ally isolated from the blood. Fran-

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON

Figure 3. Gram Stain Smears of the Agents of Anthrax (Bacillus anthracis), Plague (Yersinia pestis), and Tularemia (Francisella tularensis), Demonstrating Comparative Morphology, Size, and Staining Characteristics

A B C

A, B anthracis is a large (0.5-1.2 µm ϫ 2.5-10.0 µm), chain-forming, gram-positive rod that sporulates under certain conditions (Gram stain of organism from culture; original magnification ϫ250); B, Y pestis is a gram-negative, plump, non–spore-forming, bipolar-staining bacillus that is approximately 0.5-0.8 µm ϫ 1-3 µm (Gram stain of smear from infected tissue; original magnification ϫ250); C, F tularensis is a small (0.2 µm ϫ 0.2-0.7 µm), pleomorphic, poorly staining, gram-negative coc- cobacillus (Gram stain of organism from culture; original magnification ϫ500) (inset, direct immunofluorescence of smear of F tularensis; original magnification ϫ400. Sources: A and B, Sherif Zaki, Centers for Disease Control and Prevention; C, Armed Forces Institute of Pathology. cisella tularensis grows best in cysteine- established, results of antigen detection Figure 4. Francisella tularensis Growth at enriched broth and thioglycollate and polymerase chain reaction analyses 72 Hours After Inoculation broth and on cysteine heart blood can be obtained within several hours of agar, buffered charcoal-yeast agar, and receipt of isolates. Typically, serum an- chocolate agar. Selective agar (such as tibody titers do not attain diagnostic lev- chocolate agar selective for Neisseria els until 10 or more days after onset of gonorrhea isolation) may be useful illness, and serology would provide mini- when culturing materials from non- mal useful information for managing an sterile sites, such as sputum. Inocu- outbreak. Serological confirmation of lated media should be incubated at cases, however, may be of value for fo- 37°C. Although growth may be visible rensic or epidemiological purposes. Most as early as 24 to 48 hours after inocu- laboratories use tube agglutination or mi- lation, growth may be delayed and cul- croagglutination tests that detect com- These Francisella tularensis colonies show character- tures should be held for at least 10 days bined immunoglobulin M and immu- istic opalescence on cysteine heart agar with sheep before discarding. Under ideal condi- noglobulin G.84,85 A 4-fold change in titer blood (cultured at 37°C for 72 hours). Source: Cen- tions, bacterial colonies on cysteine- between acute and convalescent serum ters for Disease Control and Prevention. enriched agar are typically 1 mm in specimens, a single titer of at least 1:160 diameter after 24 to 48 hours of incu- for tube agglutination or 1:128 for mi- the United States, a live attenuated vac- bation and 3 to 5 mm in diameter by croagglutination is diagnostic for F tu- cine derived from the avirulent live vac- 96 hours.60,81 On cysteine heart agar, F larensis infection. Information on refer- cine strain has been used to protect tularensis colonies are characteristi- ence diagnostic testing and shipping/ laboratorians routinely working with cally opalescent and do not discolor the handling of specimens can be obtained F tularensis; until recently, this vac- medium (FIGURE 4). from state public health laboratories and cine was available as an investiga- Antigen detection assays, polymer- from the Division of Vector-Borne In- tional new drug.89 It is currently un- ase chain reaction, enzyme-linked im- fectious Diseases, CDC, Fort Collins, der review by the US Food and Drug munoassays, immunoblotting, pulsed- Colo (telephone: [970] 221-6400; e-mail: Administration (FDA), and its future field gel electrophoresis, and other [email protected]). availability is undetermined. specialized techniques may be used to In a retrospective study of civilians identify F tularensis and to characterize VACCINATION working with F tularensis at a US Army strains.84-87 These procedures are usu- Beginning in the 1930s, the Soviet research facility, the incidence of acci- ally performed only in research and ref- Union used a live attenuated vaccine to dental acute inhalational tularemia erence laboratories, however. In labora- immunize tens of millions of persons among laboratorians declined from 5.70 tories where advanced methods are living in tularemia-endemic areas.88 In cases per 1000 person-years of risk at

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON

a time when a killed vaccine was in use dren.90,94,96,97 Treatment with cipro- Table 2. Working Group Consensus Recommendations for Treatment of Patients to 0.27 cases per 1000 person-years of floxacin should be continued for 10 With Tularemia in a Contained Casualty risk after introduction of the live vac- days. In persons beginning treatment Setting* cine.58 Although the incidence of ul- with parenteral doxycycline, cipro- Contained Casualty Recommended Therapy ceroglandular disease remained un- floxacin, or chloramphenicol, therapy Adults changed in the 2 periods, signs and can be switched to oral antibiotic ad- Preferred choices symptoms were considered milder ministration when clinically indi- Streptomycin,1gIMtwice daily Gentamicin, 5 mg/kg IM or IV once daily† among those who received the live vac- cated. Very limited experiences in treat- Alternative choices cine. In volunteer studies, the live at- ing tularemia patients with ␤-lactam Doxycycline, 100 mg IV twice daily Chloramphenicol, 15 mg/kg IV 4 times tenuated vaccine did not protect all re- and macrolide antibiotics have been re- daily† cipients against aerosol challenges with ported, and treatment failures have oc- Ciprofloxacin, 400 mg IV twice daily† virulent F tularensis.7,26 curred.98 Use of ␤-lactam and macro- Children Correlates of protective immunity ap- lide antibiotics in treating tularemia is Preferred choices pear about 2 weeks following natural neither FDA-approved nor recom- Streptomycin, 15 mg/kg IM twice daily (should not exceed 2 g/d) infection or vaccination. Given the short mended by the working group. Gentamicin, 2.5 mg/kg IM or IV 3 times incubation period of tularemia and in- Children. In children, streptomycin daily† Alternative choices complete protection of current vac- or gentamicin is recommended by the Doxycycline; if weight Ն45 kg, 100 mg IV cines against inhalational tularemia, working group as first-line treatment in twice daily; if weight Ͻ45 kg, give 2.2 mg/kg IV twice daily vaccination is not recommended for a contained casualty situation (Table 2). Chloramphenicol, 15 mg/kg IV 4 times postexposure prophylaxis. The work- Doxycycline, ciprofloxacin (Յ1 g/d), and daily† ing group recommends use of the live chloramphenicol can be used as alter- Ciprofloxacin, 15 mg/kg IV twice daily†‡ vaccine strain only for laboratory per- natives to aminoglycosides. Fluoroqui- Pregnant Women sonnel routinely working with F tu- nolones have been reported to cause car- Preferred choices Gentamicin, 5 mg/kg IM or IV once daily† larensis. tilage damage in immature animals and Streptomycin,1gIMtwice daily are not FDA-approved for use in chil- Alternative choices TREATMENT Doxycycline, 100 mg IV twice daily dren. However, short courses of these Ciprofloxacin, 400 mg IV twice daily† Contained Casualty agents have not been associated with ar- *Treatment with streptomycin, gentamicin, or ciprofloxa- Situation thropathy in pediatric patients, and the cin should be continued for 10 days; treatment with doxycycline or chloramphenicol should be continued for 14-21 Adults. In a contained casualty situa- potential risks of their use must be days. Persons beginning treatment with intramuscular tion, in which logistics permit indi- weighed against their benefits in treat- (IM) or intravenous (IV) doxycycline, ciprofloxacin, or chlor- 96,99,100 amphenicol can switch to oral antibiotic administration vidual medical management, the working serious infections. when clinically indicated. †Not a US Food and Drug Administration–approved use. ing group recommends parenteral ‡Ciprofloxacin dosage should not exceed 1 g/d in children. antimicrobial therapy for tularemia Mass Casualty Situation (TABLE 2). Streptomycin is the drug Doxycycline and ciprofloxacin, admin- of choice.49,74,90,91 Gentamicin, which istered orally, are the preferred choices Table 3. Working Group Consensus is more widely available and may be for treatment in the mass casualty set- Recommendations for Treatment of Patients used intravenously, is an acceptable al- ting, for both adults and children With Tularemia in a Mass Casualty Setting 49,74,90-93 and for Postexposure Prophylaxis* ternative. Treatment with (TABLE 3). The ciprofloxacin dosage for Mass Casualty Recommended Therapy aminoglycosides should be continued children should not exceed 1 g/d. In a Adults for 10 days. Tetracyclines and chlor- mass casualty situation, the working Preferred choices amphenicol are also used to treat tula- group believes the benefits to children Doxycycline, 100 mg orally twice daily remia49,74,90; however, relapses and pri- from short courses of doxycycline or Ciprofloxacin, 500 mg orally twice daily† mary treatment failures occur at a fluoroquinolones (Table 3) outweigh Children higher rate with these bacteriostatic the risks of their use. Preferred choices agents than with aminoglycosides, and Since it is unknown whether drug- Doxycycline; if Ն45 kg, give 100 mg orally twice daily; if Ͻ45 kg, give 2.2 mg/kg they should be given for at least 14 days resistant organisms might be used in a orally twice daily to reduce chance of relapse.27,74,90 Fluo- bioterrorist event, antimicrobial sus- Ciprofloxacin, 15 mg/kg orally twice daily†‡ roquinolones, which have intracellu- ceptibility testing of isolates should be Pregnant Women lar activity, are promising candidates conducted quickly and treatments al- Preferred choices Ciprofloxacin, 500 mg orally twice daily† for treating tularemia. Ciprofloxacin, tered according to test results and clini- Doxycycline, 100 mg orally twice daily which is not labeled for use in tulare- cal responses. *One antibiotic, appropriate for patient age, should be cho- mia, has been shown to be active against Antibiotics for treating patients in- sen from among alternatives. The duration of all rec- 94 95 ommended therapies in Table 3 is 14 days. F tularensis in vitro and in animals fected with tularemia in a bioterror- †Not a US Food and Drug Administration–approved use. and has been used to successfully treat ism scenario are included in a na- ‡Ciprofloxacin dosage should not exceed 1 g/d in children. tularemia in both adults and chil- tional pharmaceutical stockpile

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON maintained by the CDC, as are venti- developed symptomatic tularemia af- orous shaking, growing cultures in lators and other emergency equip- ter antibiotic treatment was stopped. volume, animal studies) require BSL-3 ment needed to respond to situations In the unlikely event that authori- conditions.102 When F tularensis is pre- of large numbers of critically ill per- ties quickly become aware that an F tu- sumptively identified in a routine BSL-2 sons that strip local and state re- larensis biological weapon has been clinical laboratory (level A), speci- sources.30 used and are able to identify and reach mens should be forwarded to a BSL-3 exposed persons during the early in- laboratory (level B) (eg, a state public Management of Special Groups cubation period, the working group rec- health laboratory) for confirmation of Pregnant Women. In a contained casu- ommends that exposed persons be pro- agent and other studies, such as anti- alty situation, short courses of gentami- phylactically treated with 14 days of oral microbial susceptibility testing.11 Bod- cin are likely to pose a low risk to fetuses doxycycline or ciprofloxacin (Table 3). ies of patients who die of tularemia when used to treat tularemia in preg- In a circumstance in which the weapon should be handled using standard pre- nant women (Table 2). Rare cases of fetal attack has been covert and the event is cautions. Autopsy procedures likely to nerve deafness and renal damage have discovered only after persons start to cause aerosols, such as bone sawing, been reported with other aminoglyco- become ill, persons potentially ex- should be avoided. Clothing or linens sides but have not been reported with posed should be instructed to begin a contaminated with body fluids of pa- gentamicin. The benefits of gentamicin fever watch. Persons who develop an tients infected with F tularensis should in treating pregnant women with tula- otherwise unexplained fever or flulike be disinfected per standard precau- remia are expected to outweigh any illness within 14 days of presumed ex- tions protocols.101 potential risk to fetuses. In a mass casu- posure should begin treatment as out- alty situation, oral ciprofloxacin is con- lined in Tables 2 and 3. ENVIRONMENTAL sidered the best alternative to gentami- In the laboratory, persons who have DECONTAMINATION cin for pregnant women (Table 3). had potentially infective exposures to AND PROTECTION Immunosuppressed Persons. There F tularensis should be administered oral Under natural conditions, F tularensis is scant experience in treating tulare- postexposure antibiotic prophylaxis if may survive for extended periods in a mia in immunocompromised patients. the risk of infection is high (eg, spill, cold, moist environment. The work- However, considering the greater occur- centrifuge accident, or needlestick). If ing group lacks information on sur- rence in immunocompetent patients of the risk is low, exposed persons can be vival of intentionally dispersed par- tularemia relapses and treatment fail- placed on a fever watch and treated if ticles but would expect a short half- ures following use of bacteriostatic an- they develop symptoms. life due to desiccation, solar radiation, timicrobial agents compared with Postexposure prophylactic antibi- oxidation and other environmental fac- aminoglycosides, streptomycin or gen- otic treatment of close contacts of tu- tors, and a very limited risk from sec- tamicin should be used when possible to laremia patients is not recommended ondary dispersal. In circumstances of treat patients with known immune dys- since human-to-human transmission of a laboratory spill or intentional use in function in either contained casualty or F tularensis is not known to occur. which authorities are concerned about mass casualty situations (Table 2). an environmental risk (eg, inanimate INFECTION CONTROL surfaces wet with material thought to POSTEXPOSURE ANTIBIOTIC Isolation is not recommended for tu- contain F tularensis), decontamina- RECOMMENDATIONS laremia patients, given the lack of hu- tion can be achieved by spraying the Persons beginning treatment with strep- man-to-human transmission. In hos- suspected contaminant with a 10% tomycin, gentamicin, doxycycline, or pitals, standard precautions101 are bleach solution (1 part household ciprofloxacin in the incubation period recommended by the working group for bleach and 9 parts water). After 10 min- of tularemia and continuing treat- treatment of patients with tularemia. utes, a 70% solution of alcohol can be ment daily for 14 days might be pro- Microbiology laboratory personnel used to further clean the area and re- tected against symptomatic infection. should be alerted when tularemia is duce the corrosive action of the bleach. In studies of aerosol challenge with in- clinically suspected. Routine diagnos- Soap water can be used to flush away fective doses of the virulent SCHU S-4 tic procedures can be performed in bio- less hazardous contaminations. Per- strain of F tularensis, each of 8 volun- logical safety level 2 (BSL-2) condi- sons with direct exposure to powder or teers given oral dosages of tetracy- tions. Examination of cultures in which liquid aerosols containing F tularensis cline, 1 g/d for 28 days, and each of 8 F tularensis is suspected should be car- should wash body surfaces and cloth- volunteers given tetracycline, 2 g/d for ried out in a biological safety cabinet. ing with soap water. Standard levels of 14 days, were fully protected when Manipulation of cultures and other ac- chlorine in municipal water sources treatment was begun 24 hours follow- tivities involving infectious materials should protect against waterborne in- ing challenge.27 Two of 10 volunteers with a potential for aerosol or droplet fection.60 Following an urban release, given tetracycline, 1 g/d for only 5 days, production (centrifuging, grinding, vig- the risk to humans of acquiring tula-

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON remia from infected animals or arthro- The views, opinions, assertions, and findings con- ring in laboratory workers. Public Health Rep. 1922; tained herein are those of the authors and should not 37:392-413. pod bites is considered minimal and be construed as official US Department of Defense or 19. Simpson WM. Tularemia (Francis’ disease). Ann could be reduced by educating the pub- US Department of Army positions, policies, or deci- Intern Med. 1928;1:1007-1059. sions unless so designated by other documentation. 20. Francis E. A summary of present knowledge of lic on simple avoidance of sick or dead Additional Articles: This article is the fifth in a series tularemia. Medicine. 1928;7:411-432. animals and on personal protective entitled Medical and Public Health Management Fol- 21. Hopla CE. The ecology of tularemia. Adv Vet Sci measures against biting arthropods. lowing the Use of a Biological Weapon: Consensus Comp Med. 1974;18:25-53. Statements of the Working Group on Civilian Biode- 22. Jellison WL. Tularemia in North America. Mis- fense. See references 2 through 5. soula: University of Montana; 1974:1-276. ADDITIONAL RESEARCH Acknowledgment: We thank May C. Chu, PhD, CDC, 23. Harris S. Japanese biological warfare research on for assistance with laboratory diagnostic aspects of tu- humans: a case study of microbiology and ethics. Ann Simple, rapid, and reliable diagnostic laremia, and Edward B. Hayes, MD, CDC, for assis- N Y Acad Sci. 1992;666:21-52. tests that could be used to identify per- tance with clinical and epidemiological aspects of tu- 24. Alibek K. Biohazard. New York, NY: Random sons infected with F tularensis in the laremia. House; 1999:29-38. 25. McCrumb FR Jr, Snyder MJ, Woodward TE. Stud- mass exposure setting need to be de- ies on human infection with Pasteurella tularensis: veloped. Further methods should be de- REFERENCES comparison of streptomycin and chloramphenicol in the prophylaxis of clinical disease. Trans Assoc Am Phy- signed to rapidly define the molecular 1. Parker RR. Recent studies of tick-borne diseases sicians. 1957;70:74-80. genetic characteristics of organisms, es- made at the United States Public Health Service Labo- 26. McCrumb FR Jr. Aerosol infection in man with Pas- ratory at Hamilton, Montana. In: Proceedings of the teurella tularensis. Bacteriol Rev. 1961;25:262-267. pecially as they may relate to engi- Fifth Pacific Congress; 1934:3367-3374. 27. Sawyer WD, Dangerfield HG, Hogge AL, Cro- neered attributes, such as enhanced 2. Inglesby TV, Henderson DA, Bartlett JG, et al, for zier D. Antibiotic prophylaxis and therapy of airborne the Working Group on Civilian Biodefense. Anthrax virulence and resistance to antimicro- tularemia. Bacteriol Rev. 1966;30:542-548. as a biological weapon: medical and public health man- 28. Health Aspects of Chemical and Biological Weap- agement. JAMA. 1999;281:1735-1745. bial agents or normally lethal environ- ons. Geneva, Switzerland: World Health Organiza- 3. Henderson DA, Inglesby TV, Bartlett JG, et al, for tion; 1970:105-107. mental conditions. Complete sequenc- the Working Group on Civilain Biodefense. Smallpox 29. Franz DR, Jahrling PB, Friedlander AM, et al. Clini- as a biological weapon: medical and public health man- ing and analysis of the genome of cal recognition and management of patients ex- agement. JAMA. 1999;281:2127-2137. posed to biological warfare agents. JAMA. 1997;278: natural strains of F tularensis would pro- 4. Inglesby TV, Dennis DT, Henderson DA, et al, for 399-411. the Working Group on Civilian Biodefense. Plague as vide an archival base for understand- 30. Khan AS, Morse S, Lillibridge S. Public health pre- a biological weapon: medical and public health man- ing genetic variants, functions of genes, paredness for biological terrorism in the USA. Lancet. agement. JAMA. 2000;283:2281-2290. 2000;356:1179-1182. and mechanisms of action useful in de- 5. Arnon SA, Schecter R, Inglesby TV, et al, for the 31. Tucker JB. National health and medical services veloping means to protect against F tu- Working Group on Civilain Biodefense. Botulinum toxin as a biological weapon: medical and public health man- response to incidents of chemical and biological ter- larensis. Research is also needed to de- agement. JAMA. 2001;285:1059-1070. rorism. JAMA. 1997;278:362-368. velop accurate and reliable procedures 6. Saslaw S, Eigelsbach HT, Wilson HE, Prior JA, Carhart 32. Hopla CE, Hopla AK. Tularemia. In: Beran GW, S. Tularemia vaccine study, I: intracutaneous chal- Steele JH, eds. Handbook of Zoonoses. 2nd ed. Boca to rapidly detect F tularensis in envi- lenge. Arch Intern Med. 1961;107:121-133. Raton, Fla: CRC Press; 1994:113-126. ronmental samples. 7. Saslaw S, Eigelsbach HT, Prior JA, Wilson HE, Carhart 33. Centers for Disease Control and Prevention. Sum- S. Tularemia vaccine study, II: respiratory challenge. mary of notifiable diseases, United States, 1997. New technologies should be ex- Arch Intern Med. 1961;107:134-146. MMWR Morb Mortal Wkly Rep. 1998;46:71-80. plored for developing active (eg, DNA- 8. World Health Organization. Health Aspects of 34. Dennis DT. Tularemia. In: Wallace RB, ed. Maxcy- based) or passive (eg, monoclonal an- Chemical and Biological Weapons. Geneva, Switzer- Rosenau-Last Public Health and Preventive Medi- land: World Health Organization; 1970:75-76. cine. 14th ed. Stamford, Conn: Appleton & Lange; tibody–based) vaccines for rapid 9. Kaufmann AF, Meltzer MI, Schmid GP. The eco- 1998:354-357. preexposure or postexposure protec- nomic impact of a bioterrorist attack: are prevention 35. Boyce JM. Recent trends in the epidemiology of and post-attack intervention programs justifiable? tularemia in the United States. J Infect Dis. 1975;131: tion. Emerg Infect Dis. 1997;2:83-94. 197-199. 10. Christopher GW, Cieslak TJ, Pavlin JA, Eitzen EM. 36. Ta¨rnvik A, Sandstro¨m G, Sjo¨stedt A. Epidemio- Ex Officio Participants in the Working Group on Ci- Biological warfare: a historical perspective. JAMA. 1997; logical analysis of tularemia in Sweden 1931-1993. vilian Biodefense: George Counts, MD, CDC; Mar- 278:412-417. FEMS Immunol Med Microbiol. 1996;13:201-204. garet Hamburg, MD, former assistant secretary for 11. Centers for Disease Control and Prevention. Bio- 37. Pollitzer R. History and Incidence of Tularemia planning and evaluation, Department of Health and logical and chemical terrorism: strategic plan for pre- in the Soviet Union: A Review. Bronx, NY: Institute Human Services (DHHS); Robert Knouss, MD, Office paredness and response: recommendations of the CDC for Contemporary Russian Studies, Fordham Univer- of Emergency Preparedness, DHHS; Brian Malkin, Esq, Strategic Planning Workgroup. MMWR Morb Mor- sity; 1967:1-103. formerly with the FDA; and Stuart Nightingale, MD, tal Wkly Rep. 2000;49(RR-4):1-14. 38. Halsted CC, Klasinghe HP. Tularemia pneumo- Office of the Assistant Secretary for Planning and Evalu- 12. Francis E. Tularemia. JAMA. 1925;84:1243- nia in urban children. Pediatrics. 1978;4:660-662. ation, DHHS. 1250. 39. Martone WJ, Marshall LW, Kaufmann AF, Hobbs Funding/Support: Funding for this study primarily was 13. Karpoff SP, Antononoff NI. The spread of tula- JH, Levy ME. Tularemia pneumonia in Washington, provided by each participant’s institution or agency. remia through water as a new factor in its epidemi- DC. A report of three cases with possible common- The Johns Hopkins Center for Civilian Biodefense Stud- ology. J Bacteriol. 1936;32:243-258. source exposures. JAMA. 1979;23:2315-2317. ies provided travel funds for 5 of the group. 14. Silchenko VS. Epidemiological and clinical features 40. Rogutsky SV, Khramtsov MM, Avchinikov AV, et Disclaimers: In some instances, the indications, dos- of tularemia caused by waterborne infection. Zh Mik- al. Epidemiological investigation of an outbreak of tu- ages, and other information in this article are not con- robiol Epidemiol Immunobiol. 1957;28:788-795. laremia in the Smolensk region. Zh Mikrobiol sistent with current approved labeling by the US Food 15. Gelman AC. The ecology of tularemia. In: May (Moscow). 1997;2:33-37. and Drug Administration (FDA). The recommenda- JM, ed. Studies in Disease Ecology. New York, NY: 41. McCarthy VP, Murphy MD. Lawnmower tulare- tions on use of drugs and vaccine for uses not ap- Hafner Publishing Co; 1961:89-108. mia. Pediatr Infect Dis J. 1990;9:298-299. proved by the FDA do not represent the official views 16. Jellison WL, Kohls GM. Tularemia in Sheep and 42. Klock LE, Olsen PF, Fukushima T. Tularemia epi- of the FDA nor of any of the federal agencies whose Sheep Industry Workers in Western United States. demic associated with the deerfly. JAMA. 1973;226: scientists participated in these discussions. Unlabeled Washington, DC: US Public Health Service; 1955:1- 149-152. uses of the products recommended are noted in the 17. Public health monograph 28. 43. Dahlstrand S, Ringertz O, Zetterberg. Airborne tu- sections of this article in which these products are dis- 17. Francis E. Sources of infection and seasonal inci- laremia in Sweden. Scand J Infect Dis. 1971;3:7-16. cussed. Where unlabeled uses are indicated, informa- dence of tularemia in man. Public Health Rep. 1937; 44. Christenson B. An outbreak of tularemia in the tion used as the basis for the recommendation is 52:103-113. northern part of central Sweden. Scand J Infect Dis. discussed. 18. Lake GC, Francis E. Six cases of tularemia occur- 1984;16:285-290.

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF TULAREMIA AS A BIOLOGICAL WEAPON

45. Warring WB, Ruffin JS. A tick-borne epidemic of 66. Pavlov VM, Mokrievich, Volkovoy K. Cryptic plas- of tularemia in different clinical forms and severities tularemia. N Engl J Med. 1946;234:137-140. mid pFNL10 from Francisella novicida-like F6168: the of the disease. J Infect Dis. 1986;153:142-145. 46. Ohara Y, Sato T, Homma M. Arthropod-borne base of plasmid vectors for Francisella tularensis. FEMS 85. Bevanger L, Macland JA, Naess AI. Agglutinins tularemia in Japan: clinical analysis of 1,374 cases ob- Immunol Med Microbiol. 1996;13:253-256. and antibodies to Francisella tularensis outer mem- served between 1924 and 1996. J Med Entomol. 1998; 67. Sandstro¨m G, Sjo¨stedt A, Johansson T, Kuoppa brane antigens in the early diagnosis of disease dur- 35:471-473. K, Williams JC. Immunogenicity and toxicity of lipo- ing an outbreak of tularemia. J Clin Microbiol. 1988; 47. Markowitz LE, Hynes NA, de la Cruz P, et al. Tick- polysaccharide from Francisella tularensis LVS. FEMS 26:433-437. borne tularemia: an outbreak of lymphadenopathy in Microbiol Immunol. 1992;105:201-210. 86. Grunow R, Splettstoesser W, McDonald S, et al. children. JAMA. 1985;254:2922-2925. 68. Ta¨rnvik A. Nature of protective immunity to Fran- Detection of Francisella tularensis in biological speci- 48. Young LS, Bicknell DS, Archer BG, et al. Tulare- cisella tularensis. Rev Infect Dis. 1989;11:440-451. mens using a capture enzyme-linked immunosor- mia epidemic, Vermont, 1968: forty-seven cases linked 69. Fortier AH, Green SJ, Polsinelli T, et al. Life and bent assay, an immunochromatographic handheld as- to contact with muskrats. N Engl J Med. 1969;280: death of an intracellular pathogen: Francisella tu- say, and a PCR. Clin Diagn Lab Immunol. 2000:7: 1253-1260. larensis and the macrophage. Immunol Ser. 1994;60: 86-90. 49. Evans ME, Gregory DW, Schaffner W, McGee ZA. 349-361. 87. Higgins JA, Hubalek Z, Halouzka J, et al. Detec- Tularemia: a 30-year experience with 88 cases. Medi- 70. Pullen RL, Stuart BM. Tularemia: analysis of 225 tion of Francisella tularensis in infected mammals and cine. 1985;64:251-269. cases. JAMA. 1945;129:495-500. vectors using a proble-based polymerase chain reac- 50. Jellison WL, Epler DC, Kuhns E, Kohls GM. Tula- 71. Lillie RD, Francis EI. The pathology of tularaemia tion. Am J Trop Med Hyg. 2000;62:310-318. remia in man from a domestic rural water supply. Pub- in man (Homo sapiens). In: The Pathology of Tula- 88. Sjo¨stedt A, Ta¨rnvik A, Sandstro¨m G. Francisella lic Health Rep. 1950;65:1219-1226. raemia. Washington, DC: US Government Printing Of- tularensis: host-parasite interaction. FEMS Immunol 51. Mignani E, Palmieri F, Fontana M, Marigo S. Ital- fice; 1937:1-81. National Institute of Health Bulletin Med Microbiol. 1996;13:181-184. ian epidemic of waterborne tularaemia. Lancet. 1988; No. 167. 89. French GR, Plotkin SA. Miscellaneous limited- 2:1423. 72. Stuart BM, Pullen RL. Tularemic pneumonia: Re- use vaccines. In: Plotkin S, Mortimer EA, eds. Vac- 52. Teutsch SM, Martone WJ, Brink EW, et al. Pneu- view of American literature and report of 15 addi- cine. Philadelphia, Pa: WB Saunders; 1999:728-733. monic tularemia on Martha’s Vineyard. N Engl J Med. tional cases. Am J Med Sci. 1945;210:223-236. 90. Enderlin, G, Morales L, Jacobs RF, Cross TJ. Strep- 1979;301:826-828. 73. White JD, Rooney JR, Prickett PA, Derrenbacher tomycin and alternative agents for the treatment of 53. Syrja¨la¨ H, Kujala P, Myllyla¨ V, Salminen A. Air- EH, Beard CW, Griffith WR. Pathogenesis of experi- tularemia: review of the literature. Clin Infect Dis. 1994; borne transmission of tularemia in farmers. Scand J mental respiratory tularemia in monkeys. J Infect Dis. 19:42-47. Infect Dis. 1985;17:371-375. 1964;114:277-283. 91. Jacobs RF, Narain JP. Tularemia in children. Pe- 54. Syrja¨la¨ H, Sutinen S, Jokinen K, Nieminen P, Tuu- 74. Cross JT, Penn RL. Francisella tularensis diatr Infect Dis. 1983;2:487-491. ponen T, Salminen A. Bronchial changes in airborne (tularemia). In: Mandell GL, et al. eds. Principles and 92. Mason WL, Eigelsbach HT, Little SF, et al. Treat- tularemia. J Laryngol Otol. 1986;100:1169-1176. Practice of Infectious Diseases. Philadelphia, Pa: ment of tularemia, including pulmonary tularemia, with 55. Puntigam F. Erkrakungen an torakalen formen der Churchill Livingstone; 2000:2393-2402. gentamicin. Am Rev Respir Dis. 1980;121:39-45. tulara¨mia bei arbeitnehmern in Zuckerfabriken. Z Hyg. 75. Avery FW, Barnett TB. Pulmonary tularemia: a re- 93. Cross JT, Schutze GE, Jacobs RF. Treatment of tu- 1960;147:162-168. port of five cases and consideration of pathogenesis laremia with gentamicin in pediatric patients. Pediatr 56. Overholt EL, Tigertt WD, Kadull PJ, et al. An analy- and terminology. Am Rev Respir Dis. 1967;95:584- Infect Dis J. 1995;14:151-152. sis of forty-two cases of laboratory-acquired tulare- 591. 94. Syrja¨la¨ H, Schildt R, Ra¨isa¨inen S. In vitro suscep- mia. Am J Med. 1961;30:785-806. 76. Alluisi EA, Beisel WR, Bartonelli PJ, Coates GD. tibility of Francisella tularensis to fluoroquinolones and 57. Pike RM. Laboratory-associated infections: sum- Behavioral effects of tularaemia and sandfly fever in treatment of tularemia with norfloxacin and cipro- mary and analysis of 3921 cases. Health Lab Sci. 1976; man. J Infect Dis. 1973;128:710-717. floxacin. Eur J Clin Microbiol Infect Dis. 1991;10: 13:105-114. 77. Stuart BM, Pullen RL. Tularemic meningitis: re- 68-70. 58. Burke DS. Immunization against tularemia: analy- view of the literature and report of a case with post- 95. Russell P, Eley SM, Fulop MJ, Bell DL, Titball RW. sis of the effectiveness of live Francisella tularensis vac- mortem observations. Arch Intern Med. 1945;76:163- The efficacy of ciprofloxacin and doxycycline against cine in prevention of laboratory-acquired tularemia. 166. tularemia. J Antimicrob Chemother. 1998;41:461- J Infect Dis. 1977;135:55-60. 78. American Public Health Association. Tularemia. 465. 59. Centers for Disease Control and Prevention. Sum- In: Chin J, ed. Control of Communicable Diseases 96. Limaye AP, Hooper CJ. Treatment of tularemia mary of notifiable diseases, United States, 1994. Manual. Washington, DC: American Public Health As- with fluoroquinolones: two cases and review. Clin In- MMWR Morb Mortal Wkly Rep. 1995;43:3. sociation; 2000:532-535. fect Dis. 1999;29:922-924. 60. Bell JF. Tularemia. In: Steele JH, ed. CRC Hand- 79. Amoss HL, Sprunt DH. Tularemia: review of lit- 97. Johansson A, Berglund L, Gothefors L, et al. Cipro- book Series in Zoonoses. Vol 2. Boca Raton, Fla: CRC erature of cases contracted by ingestion of rabbit and floxacin for treatment of tularemia in children. Pedi- Press; 1980:161-193. the report of additional cases with a necropsy. JAMA. atr Infect Dis J. 2000;19:449-453. 61. Pomanskaia LA. The survival times of the organ- 1936;106:1078-1080. 98. Cross JT, Jacobs RF. Tularemia: treatment fail- isms of tularaemia on grain and straw. J Microbiol Epi- 80. Sunderrajan EV, Hutton J, Marienfeld D. Adult res- ures with outpatient use of ceftriaxone. Clin Infect Dis. demiol Immunobiol. 1957;28:597-603. piratory distress syndrome secondary to tularemia pneu- 1993;17:976-980. 62. Wong JD, Shapiro DS. Francisella. In : Murray PR, monia. Arch Intern Med. 1985;145:1435-1437. 99. Quinolones. In: AHFS Drug Information 1999. ed. Manual of Clinical Microbiology. 7th ed. Wash- 81. Centers for Disease Control and Prevention. Bethesda, Md: American Society of Health-System ington, DC: ASM Press; 1999:647-651. Basic laboratory protocols for the presumptive iden- Pharmacists; 1999:670-684. 63. Johansson A, Ibrahim A, Goransson I, et al. Evalu- tification of Francisella tularensis. Available at: http:// 100. American Academy of Pediatrics. Antimicrobi- ation of PCR-based methods for discrimination of Fran- www.bt.cdc.gov/Agent/Tularemia/tularemia20010417 als and related therapy. In: Peter G, ed. Red Book 2000: cisella species and subspecies and development of a .pdf. Accessed April 20, 2001. Report of the Committee on Infectious Diseases. 25th specific PCR that distinguishes the two major subspe- 82. White JD, McGavran MH. Identification of Pas- ed. Elk Grove Village, Ill: American Academy of Pe- cies of Francisella tularensis. J Clin Microbiol. 2000; teurella tularensis by immunofluorescence. JAMA. diatrics; 2000:645-646. 38:4180-4185. 1965;194:180-182. 101. Garner JS. Guideline for isolation precautions in 64. Gurycova D. First isolation of Francisella tularen- 83. Guarner J, Greer PW, Bartlett J, Chu MC, Shieh hospitals. Infect Control Hosp Epidemiol. 1996;17: sis subspecies tularensis in Europe. Eur J Epidemiol. WJ, Zaki SR. Immunohistochemical detection of Fran- 51-80. 1998;14:797-802. cisella tularensis in formalin-fixed paraffin- 102. US Department of Health and Human Services. 65. Clarridge JE III, Raich TJ, Sjo¨sted A, et al. Char- embedded tissue. Appl Immunohistochem Mol Mor- Laboratory biosafety level criteria. In: Richmond JY, acterization of two unusual clinically significant Fran- phol. 1999;7:122-126. McKinney RW, eds. Biosafety in Microbiological and cisella strains. J Clin Microbiol. 1996;34:1995- 84. Syrja¨la¨ H, Koskela P, Ripatti T, Salminen A, Herva Biomedical Laboratories. 4th ed. Washington, DC: 2000. E. Agglutination and ELISA methods in the diagnosis Dept of Health and Human Services; 1999:17-52.

Downloaded from jama.ama-assn.org by guest on April 25, 2012