Anthrax As a Biological Weapon, 2002 Updated Recommendations for Management

Home , 2001 anthrax attacks, Ames strain, Necro (rapper)

CONSENSUS STATEMENT

Anthrax as a Biological Weapon, 2002 Updated Recommendations for Management

Thomas V. Inglesby, MD Objective To review and update consensus-based recommendations for medical Tara O’Toole, MD, MPH and public health professionals following a Bacillus anthracis attack against a civilian population. Donald A. Henderson, MD, MPH Participants The working group included 23 experts from academic medical cen- John G. Bartlett, MD ters, research organizations, and governmental, military, public health, and emer- Michael S. Ascher, MD gency management institutions and agencies. Edward Eitzen, MD, MPH Evidence MEDLINE databases were searched from January 1966 to January 2002, using the Medical Subject Headings anthrax, Bacillus anthracis, biological weapon, Arthur M. Friedlander, MD biological terrorism, biological warfare, and biowarfare. Reference review identified Julie Gerberding, MD, MPH work published before 1966. Participants identified unpublished sources. Jerome Hauer, MPH Consensus Process The first draft synthesized the gathered information. Written comments were incorporated into subsequent drafts. The final statement incorpo- James Hughes, MD rated all relevant evidence from the search along with consensus recommendations. Joseph McDade, PhD Conclusions Specific recommendations include diagnosis of anthrax infection, in- Michael T. Osterholm, PhD, MPH dications for vaccination, therapy, postexposure prophylaxis, decontamination of the environment, and suggested research. This revised consensus statement presents new Gerald Parker, PhD, DVM information based on the analysis of the anthrax attacks of 2001, including develop- Trish M. Perl, MD, MSc ments in the investigation of the anthrax attacks of 2001; important symptoms, signs, and laboratory studies; new diagnostic clues that may help future recognition of this Philip K. Russell, MD disease; current anthrax vaccine information; updated antibiotic therapeutic consid- Kevin Tonat, DrPH, MPH erations; and judgments about environmental surveillance and decontamination. for the Working Group on Civilian JAMA. 2002;287:2236-2252 www.jama.com Biodefense

F THE BIOLOGICAL AGENTS to several locations via the US Postal Ser- tory and to 18 occupational exposure that may be used as weap- vice. Twenty-two confirmed or suspect cases in the United States during the ons, the Working Group cases of anthrax infection resulted. 20th century. Information about the po- on Civilian Biodefense Eleven of these were inhalational cases, tential impact of a large, covert attack identifiedO a limited number of organ- of whom 5 died; 11 were cutaneous cases using B anthracis or the possible effi- isms that, in worst case scenarios, could (7 confirmed, 4 suspected).2 In this ar- cause disease and deaths in sufficient ticle, these attacks are termed the an- Author Affiliations: The Center for Civilian Biode- numbers to gravely impact a city or re- thrax attacks of 2001. The conse- fense Strategies (Drs Inglesby, O’Toole, Henderson, Bartlett, and Perl) and the Schools of Medicine (Drs gion. Bacillus anthracis, the bacterium quences of these attacks substantiated Inglesby, Bartlett, and Perl) and Public Health (Drs that causes anthrax, is one of the most many findings and recommendations in O’Toole and Henderson), Johns Hopkins University, Department of Health and Human Services (Drs Ascher, serious of these. the Working Group on Civilian Biode- and Russell and Mr Hauer), Baltimore, and US Army Several countries are believed to have fense’s previous consensus statement Medical Research Institute of Infectious Diseases, (Drs 3 Eitzen, Friedlander, and Parker), Frederick, Md; Cen- offensive biological weapons pro- published in 1999 ; however, the new ters for Disease Control and Prevention, Atlanta, Ga grams, and some independent terrorist information from these attacks war- (Drs Hughes, McDade, and Gerberding); Center for groups have suggested their intent to use rant updating the previous statement. Infectious Disease Research and Policy, University of Minnesota School of Public Health, Minneapolis (Dr biological weapons. Because the possi- Before the anthrax attacks in 2001, Osterholm); and the Office of Emergency Prepared- bility of a terrorist attack using bioweap- modern experience with inhalational ness, Department of Health and Human Services, Rock- ville, Md (Dr Tonat). ons is especially difficult to predict, de- anthrax was limited to an epidemic in Corresponding Author and Reprints: Thomas V. tect, or prevent, it is among the most Sverdlovsk, Russia, in 1979 following Inglesby, MD, Johns Hopkins Center for Civilian Bio- 1 defense Strategies, Johns Hopkins University, Can- feared terrorism scenarios. In Septem- an unintentional release of B anthracis dler Bldg, Suite 830, 111 Market Pl, Baltimore, MD ber 2001, B anthracis spores were sent spores from a Soviet bioweapons fac- 21202 (e-mail: [email protected]).

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF ANTHRAX AS A BIOLOGICAL WEAPON cacy of postattack vaccination or thera- orders. However, some nations contin- tent to which they have been successful peutic measures remains limited. Poli- ued offensive bioweapons develop- is not reported.13 cies and strategies continue to rely ment programs despite ratification of In the anthrax attacks of 2001, B an- partially on interpretation and extrapo- the BWC. In 1995, Iraq acknowledged thracis spores were sent in at least 5 let- lation from an incomplete and evolv- producing and weaponizing B anthra- ters to Florida, New York City, and ing knowledge base. cis to the United Nations Special Com- Washington, DC. Twenty-two con- mission.7 The former Soviet Union is firmed or suspected cases resulted. All CONSENSUS METHODS also known to have had a large B of the identified letters were mailed The working group comprised 23 rep- anthracis production program as part from Trenton, NJ. The B anthracis resentatives from academic medical cen- of its offensive bioweapons program.8 spores in all the letters were identified ters; research organizations; and gov- A recent analysis reports that there is as the Ames strain. The specific source ernment, military, public health, and clear evidence of or widespread asser- (provenance) of B anthracis cultures emergency management institutions and tions from nongovernmental sources used to create the spore-containing agencies. For the original consensus alleging the existence of offensive bio- powder remains unknown at time of statement,3 we searched MEDLINE da- logical weapons programs in at least this publication. tabases from January 1966 to April 1998 13 countries.6 It is now recognized that the origi- using Medical Subject Headings of an- The anthrax attacks of 2001 have nal Ames strain of B anthracis did not thrax, Bacillus anthracis, biological heightened concern about the feasibil- come from a laboratory in Ames, Iowa, weapon, biological terrorism, biological ity of large-scale aerosol bioweapons at- rather from a laboratory in College Sta- warfare, and biowarfare. Reference re- tacks by terrorist groups. It has been tion, Tex. Several distinct Ames strains view identified work published before feared that independent, well-funded have been recognized by investigating 1966. Working group members identi- groups could obtain a manufactured scientists, which are being compared fied unpublished sources. weapons product or acquire the exper- with the Ames strain used in the at- The first consensus statement, pub- tise and resources to produce the mate- tack. At least 1 of these comparison lished in 1999,3 followed a synthesis of rials for an attack. However, some ana- Ames strains was recovered from a goat the information and revision of 3 drafts. lysts have questioned whether “weapons that died in Texas in 1997.14 We reviewed anthrax literature again grade” material such as that used in the Sen Daschle’s letter reportedly had 2 in January 2002, with special atten- 2001 attacks (ie, powders of B anthracis gofB anthracis containing powder; the tion to articles following the anthrax at- with characteristics such as high spore quantity in the other envelopes has not tacks of 2001. Members commented on concentration, uniform particle size, low been disclosed. The powder has been a revised document; proposed revi- electrostatic charge, treated to reduce reported to contain between 100 bil- sions were incorporated with the work- clumping) could be produced by those lion to 1 trillion spores per gram15 al- ing group’s support for the final con- not supported by the resources of a na- though no official analysis of the consensus document. tion-state. The US Department of De- centration of spores or the chemical The assessment and recommenda- fense recently reported that 3 defense em- composition of the powder has been tions provided herein represent our best ployees with some technical skills but published. professional judgment based on cur- without expert knowledge of bioweap- The anthrax attacks of 2001 used 1 rent data and expertise. The conclu- ons manufactured a simulant of B an- of many possible methods of attack. The sions and recommendations need to be thracis in less than a month for $1 mil- use of aerosol-delivery technologies in- regularly reassessed as new informa- lion.9 It is reported that Aum Shinrikyo, side buildings or over large outdoor ar- tion develops. the cult responsible for the 1995 re- eas is another method of attack that has lease of sarin nerve gas in a Tokyo sub- been studied. In 1970, the World Health HISTORY OF CURRENT THREAT way station,10 dispersed aerosols of an- Organization16 and in 1993 the Office of For centuries, B anthracis has caused thrax and botulism throughout Tokyo at Technology Assessment17 analyzed the disease in animals and serious illness least 8 times.11 Forensic analysis of the potential scope of larger attacks. The in humans.4 Research on anthrax as a B anthracis strain used in these attacks 1979 Sverdlovsk accident provides data biological weapon began more than 80 revealed that this isolate most closely on the only known aerosol release of B years ago.5 Most national offensive matched the Sterne 34F2 strain, which anthracis spores resulting in an epi- bioweapons programs were termi- is used for animal vaccination pro- demic.18 nated following widespread ratifica- grams and is not a significant risk to hu- An aerosol release of B anthracis tion or signing of the Biological Weap- mans.12 It is probable that the cult at- would be odorless and invisible and ons Convention (BWC) in the early tacks produced no illnesses for this and would have the potential to travel many 1970s6; the US offensive bioweapons other technical reasons. Al Quaeda also kilometers before dissipating.16,19 Aero- program was terminated after Presi- has sought to acquire bioweapons in its sol technologies for large-scale dissemi- dent Nixon’s 1969 and 1970 executive terrorist planning efforts although the ex- nation have been developed and tested

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

have reduced drastically the animal rally occurring cases of inhalational an- Figure 1. Gram Stain of Blood in Culture 24 Media mortality from the disease. How- thrax in the United States since 1976, ever, B anthracis spores remain preva- the occurrence of a single case is now lent in soil samples throughout the cause for alarm. world and cause anthrax cases among herbivores annually.22,25,26 MICROBIOLOGY Anthrax infection occurs in hu- B anthracis derives from the Greek word mans by 3 major routes: inhalational, for coal, anthrakis, because of the black cutaneous, and gastrointestinal. Natu- skin lesions it causes. B anthracisisan rally occurring inhalational anthrax is aerobic, gram-positive, spore-forming, now rare. Eighteen cases of inhala- nonmotile Bacillus species. The non- tional anthrax were reported in the flagellated vegetative cell is large (1-8 µm United States from 1900 to 1976; none long, 1-1.5 µm wide). Spore size is ap- Gram-positive bacilli in long chains (original were identified or reported thereafter. proximately 1 µm. Spores grow readily magnificationϫ20). Enlargement shows typical Most of these cases occurred in special- on all ordinary laboratory media at 37°C, “jointed bamboo-rod” appearance of Bacillus anthracis (original magnificationϫ100). Reprinted from risk groups, including goat hair mill or with a “jointed bamboo-rod” cellular ap- Borio et al.36 wool or tannery workers; 2 of them pearance (FIGURE 1) and a unique were laboratory associated.27 “curled-hair” colonial appearance. Ex- Cutaneous anthrax is the most com- perienced microbiologists should be able by Iraq7 and the former Soviet Union8 mon naturally occurring form, with an to identify this cellular and colonial mor- Few details of those tests are avail- estimated 2000 cases reported annually phology; however, few practicing mi- able. The US military also conducted worldwide.26 The disease typically fol- crobiologists outside the veterinary com- such trials over the Pacific Ocean in the lows exposure to anthrax-infected ani- munity have seen B anthracis colonies 1960s. A US study near Johnston Atoll mals. In the United States, 224 cases of beyond what they may have seen in pub- in the South Pacific reported a plane cutaneous anthrax were reported be- lished material.37 B anthracis spores ger- “sprayed a 32-mile long line of agent tween 1944 and 1994.28 One case was re- minate when they enter an environ- that traveled for more then 60 miles be- ported in 2000.29 The largest reported ment rich in amino acids, nucleosides, fore it lost its infectiousness.”20 epidemic occurred in Zimbabwe be- and glucose, such as that found in the In 1970, the World Health Organi- tween 1979 and 1985, when more than blood or tissues of an animal or human zation estimated that 50 kg of B anthra- 10000 human cases of anthrax were re- host. The rapidly multiplying vegeta- cis released over an urban population ported, nearly all of them cutaneous.30 tive B anthracis bacilli, on the contrary, of 5 million would sicken 250000 and Although gastrointestinal anthrax is will only form spores after local nutri- kill 100000.16 A US Congressional Of- uncommon, outbreaks are continu- ents are exhausted, such as when an- fice of Technology assessment analy- ally reported in Africa and Asia26,31,32 fol- thrax-infected body fluids are exposed sis from 1993 estimated that between lowing ingestion of insufficiently to ambient air.22 Vegetative bacteria have 130000 and 3 million deaths would fol- cooked contaminated meat. Two dis- poor survival outside of an animal or hu- low the release of 100 kg of B anthra- tinct syndromes are oral-pharyngeal man host; colony counts decline to being cis, a lethality matching that of a hy- and abdominal.31,33,34Little informa- undetectable within 24 hours following drogen bomb.17 tion is available about the risks of di- inoculation into water.22 This contrasts rect contamination of food or water with the environmentally hardy proper- EPIDEMIOLOGY OF ANTHRAX with B anthracis spores. Experimental ties of the B anthracis spore, which can Naturally occurring anthrax in hu- efforts to infect primates by direct gas- survive for decades in ambient condi- mans is a disease acquired from con- trointestinal instillation of B anthracis tions.37 tact with anthrax-infected animals or spores have not been successful.35 Gas- anthrax-contaminated animal prod- trointestinal infection could occur only PATHOGENESIS AND CLINICAL ucts. The disease most commonly oc- after consumption of large numbers of MANIFESTATIONS curs in herbivores, which are infected vegetative cells, such as what might be Inhalational Anthrax after ingesting spores from the soil. found in raw or undercooked meat from Inhalational anthrax follows deposi- Large anthrax epizootics in herbi- an infected herbivore, but experimen- tion into alveolar spaces of spore- vores have been reported.21 A pub- tal data is lacking. bearing particles in the 1- to 5-µm lished report states that anthrax killed Inhalational anthrax is expected to range.38,39 Macrophages then ingest the 1 million sheep in Iran in 194522; this account for most serious morbidity and spores, some of which are lysed and de- number is supported by an unpub- most mortality following the use of B stroyed. Surviving spores are trans- lished Iranian governmental docu- anthracis as an aerosolized biological ported via lymphatics to mediastinal ment.23 Animal vaccination programs weapon. Given the absence of natu- lymph nodes, where germination oc-

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF ANTHRAX AS A BIOLOGICAL WEAPON curs after a period of spore dormancy of sudden death in severe anthrax infec- these virulence factors, B anthracis has variable and possibly extended duration.48 The molecular target of lethal and a capsule that prevents phagocytosis. tion.35,40,41 The trigger(s) responsible for edema factors within the affected cell Full virulence requires the presence of the transformation of B anthracis spores is not yet elucidated.49 In addition to both an antiphagocytic capsule and the to vegetative cells is not fully understood.42 In Sverdlovsk, cases occurred Figure 2. Pathogenesis of Bacillus anthracis from 2 to 43 days after exposure.18 In experimental infection of monkeys, fatal 40 Exposure disease occurred up to 58 days and 98 Spore Phagocytosis MACROPHAGE 43 Inhalational days after exposure. Viable spores were Spore Germinates demonstrated in the mediastinal lymph Cutaneous nodes of 1 monkey 100 days after ex- Gastrointestinal Capsule Bacillus Proliferates posure.44 Vegetative Bacterium Once germination occurs, clinical Bacillus symptoms follow rapidly. Replicating Cytolysis B anthracis bacilli release toxins that lead to hemorrhage, edema, and necrosis.32,45 In experimental animals, once B anthracis Entry of Toxin Components Into Cell Virulence Factors toxin production has reached a criti- Protective Anthrax Toxin CELL TARGET cal threshold, death occurs even if ste- Toxin Components Antigen Receptor (ATR) rility of the bloodstream is achieved Protective Antigen (PA) 1 Protective Antigen Attaches to ATR with antibiotics.27 Extrapolations from Lethal Factor (LF) Edema Factor (EF) animal data suggest that the human 2 Protease Cleaves PA Antiphagocytic LD50 (ie, dose sufficient to kill 50% of PA PA Capsule 20 63 persons exposed to it) is 2500 to 55000 46 Other inhaled B anthracis spores. The LD10 3 PA-ATR Complexes was as low as 100 spores in 1 series of Form Heptamer Endosome monkeys.43 Recently published extrapolations from primate data suggest that 4 LF and/or EF Lethal Bind to Heptamer as few as 1 to 3 spores may be suffi- Factor cient to cause infection.47 The dose of spores that caused infection in any of Edema 5 Endocytosis Factor the 11 patients with inhalational an- Translocation 6 LF thrax in 2001 could not be estimated Calmodulin although the 2 cases of fatal inhala- EF tional anthrax in New York City and ATP

Connecticut provoked speculation that Cleavage of MAPKKs and cAMP the fatal dose, at least in some indi- Possibly Other Target Proteins viduals, may be quite low. A number of factors contribute to the Possible Pathogenic Mechanisms of B anthracis pathogenesis of B anthracis, which Inhibition of Immune Function Increase in cAMP and Alterations makes 3 toxins— protective antigen, le- Inhibition of Phagocytosis (LF, EF, Capsule) in Local Water Homeostasis (EF) thal factor, and edema factor—that com- Possible Inhibition of Induced Release of Cytolysis (LF) bine to form 2 toxins: lethal toxin and Proinflammatory Mediators (LF, EF) Release of Cellular Contents, Including Proinflammatory Mediators IGURE Dysregulation of Signal Transduction Pathways edema toxin (F 2). The protec- (eg, Tumor Necrosis Factor α, tive antigen allows the binding of le- Inhibition of MAPK and Possibly Other Pathways (LF) Interleukin 1β, Reactive Oxygen thal and edema factors to the affected Activation of Oxidative Burst Pathway (LF) Intermediates) and Lysosomal Enzymes Possible Inhibition (EF) and/or cell membrane and facilitates their sub- Other sequent transport across the cell mem- Possible Activation (LF) of Transcription Factors brane. Edema toxin impairs neutrophil function in vivo and affects water The major known virulence factors of B anthracis include the exotoxins edema toxin (PA and EF) and lethal homeostasis leading to edema, and le- toxin (PA and LF) and the antiphagocytic capsule. Although many exact molecular mechanisms involved in the pathogenicity of the anthrax toxins are uncertain, they appear to inhibit immune function, interrupt in- thal toxin causes release of tumor ne- tracellular signaling pathways, and lyse cell targets causing massive release of proinflammatory mediators. ATP crosis factor ␣ and interleukin 1 ␤, fac- indicates adenosine triphosphate; cAMP, cyclic adenosine monophosphate; MAPKK, mitogen-activated protein kinase kinase; and MAPK, mitogen-activated protein kinase. tors that are believed to be linked to the

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

showed that all patients had hemor- were nonspecific. This stage of illness Table 1. Initial Symptoms, Physical Findings, and Test Results in Patients With Inhalational rhagic thoracic lymphadenitis, hemor- lasted from hours to a few days. In some Anthrax Following US Anthrax Attacks in rhagic mediastinitis, and pleural effu- patients, a brief period of apparent re- October and November 2001* sions. About half had hemorrhagic covery followed. Other patients pro- Symptoms (N = 10) meningitis. None of these autopsies gressed directly to the second, fulmi- Fever and chills 10 showed evidence of a bronchoalveolar nant stage of illness.4,27,56 Sweats, often drenching 7 Fatigue, malaise, lethargy 10 pneumonic process although 11 of 42 This second stage was reported to have Cough, minimal 9 patient autopsies had evidence of a fo- developed abruptly, with sudden fever, or nonproductive Nausea or vomiting 9 cal, hemorrhagic, necrotizing pneu- dyspnea, diaphoresis, and shock. Mas- Dyspnea 8 monic lesion analogous to the Ghon sive lymphadenopathy and expansion of Chest discomfort or 7 pleuritic pain complex associated with tuberculo- the mediastinum led to stridor in some 50 57,58 Myalgias 6 sis. These findings are consistent with cases. A chest radiograph most of- Headache 5 other human case series and experimen- ten showed a widened mediastinum con- Confusion 4 57 Abdominal pain 3 tally induced inhalational anthrax in sistent with lymphadenopathy. Up to Sore throat 2 animals.40,51,52 A recent reanalysis of pa- half of patients developed hemorrhagic Rhinorrhea 1 thology specimens from 41 of the Sverd- meningitis with concomitant meningis- Physical Findings lovsk patients was notable primarily for mus, delirium, and obtundation. In this Fever Ͼ37.8°C 7 Tachycardia, heart rate 8 the presence of necrotizing hemor- second stage, cyanosis and hypoten- Ͼ100/min rhagic mediastinitis; pleural effusions av- sion progressed rapidly; death some- Hypotension, Ͻ110 mm Hg 1 eraging 1700 mL in quantity; meningi- times occurred within hours.4,27,56 Laboratory Results tis in 50%; arteritis and arterial rupture In the 20th-century series of US cases, White blood cell count, median 9800 × 103/µL in many; and the lack of prominent the mortality rate of occupationally ac- Differential neutrophilia, Ͼ70% 7 Neutrophil band forms, Ͼ5% 4† pneumonitis. B anthracis was recov- quired inhalational anthrax was 89%, but Elevated transaminases, 9 ered in concentrations of up to 100 mil- the majority of these cases occurred be- SGOT or SPGT Ͼ40 U/L‡ Hypoxemia, alveolar-arterial 6 lion colony-forming units per milliliter fore the development of critical care 53 oxygen gradient Ͼ30 mm Hg in blood and spinal fluid. units and, in most cases, before the ad- on room air oxygen In animal models, physiological se- vent of antibiotics.27 At Sverdlovsk, it had saturation Ͻ94% Metabolic acidosis 2 quelae of severe anthrax infection have been reported that 68 of the 79 pa- Elevated creatinine, 1 included hypocalcemia, profound hy- tients with inhalational anthrax died.18 Ͼ 1.5 mg/dL (132.6 µmol/L) poglycemia, hyperkalemia, depres- However a separate report from a hos- Chest X-ray Film Findings sion and paralysis of respiratory cen- pital physician recorded 358 ill with 45 Any abnormality 10 Mediastinal widening 7 ter, hypotension, anoxia, respiratory dead; another recorded 48 deaths among 54,55 59 Infiltrates or consolidation 7 alkalosis, and terminal acidosis, sug- 110 patients. A recent analysis of avail- Pleural effusion 8 gesting that besides the rapid admin- able Sverdlovsk data suggests there may Chest Computed Tomographic Findings§ istration of antibiotics, survival might have been as many as 250 cases with 100 Any abnormality 8 improve with vigilant correction of elec- deaths.60 Sverdlovsk patients who had Mediastinal lymphadenopathy, 7 widening trolyte disturbances and acid-based im- onset of disease 30 or more days after Pleural effusion 8 balance, glucose infusion, and early me- release of organisms had a higher re- Infiltrates or consolidation 6 chanical ventilation and vasopressor ported survival rate than those with *This table was adapted with permission from Jernigan, et al.61 administration. earlier disease onset. Antibiotics, an- †Five persons had laboratory results measuring neutro- Historical Data. Early diagnosis of in- tianthrax globulin, corticosteroids, me- phil band forms. ‡SGOT indicates serum glutamic oxalacetic transami- halational anthrax is difficult and re- chanical ventilation, and vaccine were nase; SGPT, serum glutamic pyruvic transaminase. §Eight persons had computed tomographic scan results. quires a high index of suspicion. Prior used to treat some residents in the af- to the 2001 attacks, clinical informa- fected area after the accident, but how tion was limited to a series of 18 cases many were given vaccine and antibiot- 3 toxin components.37 An additional reported in the 20th century and the ics is unknown, nor is it known which factor contributing to B anthracis patho- limited data from Sverdlovsk. The clini- patients received these interventions or genesis is the high concentration of bac- cal presentation of inhalational an- when. It is also uncertain if the B an- teria occurring in affected hosts.49 thrax had been described as a 2-stage thracis strain (or strains) to which pa- Inhalational anthrax reflects the na- illness. Patients reportedly first devel- tients was exposed were susceptible to ture of acquisition of the disease. The oped a spectrum of nonspecific symp- the antibiotics used during the out- term anthrax pneumonia is misleading toms, including fever, dyspnea, cough, break. However, a community-wide in- because typical bronchopneumonia does headache, vomiting, chills, weakness, tervention about the 15th day after ex- not occur. Postmortem pathological abdominal pain, and chest pain.18,27 posure did appear to diminish the studies of patients from Sverdlovsk Signs of illness and laboratory studies projected attack rate.60 In fatal cases, the

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF ANTHRAX AS A BIOLOGICAL WEAPON interval between onset of symptoms and previous cuts or abrasions made one es- 30,67 Figure 3. Lesion of Cutaneous Anthrax death averaged 3 days. This is similar to pecially susceptible to infection. Ar- Associated With Microangiopathic Hemolytic the disease course and case fatality rate eas of exposed skin, such as arms, Anemia and Coagulopathy in a in untreated experimental monkeys, hands, face, and neck, were the most 7-Month-Old Infant which have developed rapidly fatal dis- frequently affected. In Sverdlovsk, cu- ease even after a latency as long as 58 taneous cases occurred only as late as days.40 12 days after the original aerosol re- 2001 Attacks Data. The anthrax at- lease; no reports of cutaneous cases ap- tacks of 2001 resulted in 11 cases of in- peared after prolonged latency.18 halational anthrax, 5 of whom died. After the spore germinates in skin tis- Symptoms, signs, and important labo- sues, toxin production results in local ratory data from these patients are listed edema. An initially pruritic macule or in TABLE 1. Several clinical findings from papule enlarges into a round ulcer by the 1 cm the first 10 patients with inhalational an- second day. Subsequently, 1- to 3-mm thrax deserve emphasis.36,61-66 Malaise vesicles may appear that discharge clear By hospital day 12, a 2-cm black eschar was present in the center of the cutaneous lesion. Reprinted from and fever were presenting symptoms in or serosanguinous fluid containing Freedman et al.63 all 10 cases. Cough, nausea, and vom- numerous organisms on Gram stain. As iting were also prominent. Drenching shown in FIGURE 3, development of a sweats, dyspnea, chest pain, and head- painless, depressed, black eschar fol- The only published case report of ache were also seen in a majority of pa- lows, often associated with extensive cutaneous anthrax from the attacks of tients. Fever and tachycardia were seen local edema. The anthrax eschar dries, 2001 is notable for the difficulty in rec- in the majority of patients at presenta- loosens, and falls off in the next 1 to 2 ognition of the disease in a previously tion, as were hypoxemia and eleva- weeks. Lymphangitis and painful lymph- healthy 7-month-old, the rapid pro- tions in transaminases. adenopathy can occur with associated gression to severe systemic illness Importantly, all 10 patients had ab- systemic symptoms. Differential diag- despite hospitalization, and clinical normal chest x-ray film results: 7 had nosis of eschars includes tularemia, scrub manifestations that included microan- mediastinal widening; 7 had infil- typhus, rickettsial spotted fevers, rat bite giopathic hemolytic anemia with renal trates; and 8 had pleural effusions. fever, and ecthyma gangrenosum.68 Non- involvement, coagulopathy, and hypo- Chest computed tomographic (CT) infectious causes of eschars include natremia.63 Fortunately, this child scans showed abnormal results in all 8 arachnid bites63 and vasculitides. recovered, and none of the cutaneous patients who had this test: 7 had me- Although antibiotic therapy does not cases of anthrax diagnosed after the diastinal widening; 6, infiltrates; 8, pleu- appear to change the course of eschar 2001 attacks were fatal. ral effusions. formation and healing, it does decrease Data are insufficient to identify fac- the likelihood of systemic disease. With- Gastrointestinal Anthrax tors associated with survival although out antibiotic therapy, the mortality rate Some think gastrointestinal anthrax oc- early recognition and initiation of treat- has been reported to be as high as 20%; curs after deposition and germination of ment and use of more than 1 antibi- with appropriate antibiotic treatment, spores in the upper or lower gastroin- otic have been suggested as possible fac- death due to cutaneous anthrax has been testinal tract. However, considering the tors.61 For the 6 patients for whom such reported to be rare.4 rapid transit time in the gastrointesti- information is known, the median pe- Following the anthrax attacks of nal tract, it seems more likely that many riod from presumed time of exposure 2001, there have been 11 confirmed or such cases must result from the inges- to the onset of symptoms was 4 days probable cases of cutaneous anthrax. tion of large numbers of vegetative ba- (range, 4-6 days). Patients sought care One case report of cutaneous anthrax cilli from poorly cooked infected meat a median of 3.5 days after symptom on- resulting from these attacks has been rather than from spores. In any event, set. All 4 patients exhibiting signs of ful- published (Figure 3).63 This child had the oral-pharyngeal form of disease re- minant illness prior to antibiotic ad- no reported evidence of prior visible sults in an oral or esophageal ulcer and ministration died.61 Of note, the cuts, abrasions, or lesions at the site of leads to the development of regional incubation period of the 2 fatal cases the cutaneous lesion that developed. lymphadenopathy, edema, and sep- from New York City and Connecticut The mean incubation period for cuta- sis.31,33 Disease in the lower gastrointes- is not known. neous anthrax cases diagnosed in 2001 tinal tract manifests as primary intesti- was 5 days, with a range of 1 to 10 days, nal lesions occurring predominantly in Cutaneous Anthrax based on estimated dates of exposure the terminal ileum or cecum,50 present- Historically, cutaneous anthrax has to B anthracis–contaminated letters. Cu- ing initially with nausea, vomiting, and been known to occur following the taneous lesions occurred on the fore- malaise and progressing rapidly to deposition of the organism into skin; arm, neck, chest, and fingers.69 bloody diarrhea, acute abdomen, or sep-

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

ries in the LRN can diagnose bioweap- Table 2. Diagnosis of Inhalational Anthrax Infection* ons pathogens. Several preliminary Category Findings diagnostic tests for B anthracis can be Epidemiology Sudden appearance of several cases of severe acute febrile illness with fulminant course and death performed in hospital laboratories us- or ing routine procedures. B anthracis is Acute febrile illness in persons identified as being at risk following a specific attack (eg, those in the 2001attacks: postal workers, a gram-positive, nonhemolytic, encap- members of the news media, and politicians and their staff) sulated, penicillin-sensitive, spore- Diagnostic tests Chest radiograph: widened mediastinum, infiltrates, pleural effusion forming bacillus. Confirmatory tests Chest computed tomographic scan: hyperdense hilar and mediastinal nodes, mediastinal edema, infiltrates, pleural effusion such as immuno-histochemical stain- Thoracentesis: hemorrhagic pleural effusions ing, gamma phage, and polymerase Microbiology Peripheral blood smear: gram-positive bacilli on blood smear chain reaction assays must still be per- Blood culture growth of large gram-positive bacilli with preliminary identification of Bacillus species† formed by special reference laborato- Pathology Hemorrhagic mediastinitis, hemorrhagic thoracic lymphadenitis, ries in the LRN. hemorrhagic meningitis; DFA stain of infected tissues The determination of individual pa- *See Table 1 for list of febrile illness symptoms and signs. tient exposure to B anthracis on the ba- †Most rapid assays are available only at laboratories participating in the Laboratory Response Network. sis of environmental testing is complex due to the uncertain specificity and sen- Figure 4. Chest Radiograph and Computed Tomography (CT) Image sitivity of rapid field tests and the difficulty of assessing individual risks of ex- A B posure. A patient (or patients) seeking medical treatment for symptoms of inhalational anthrax will likely be the first evidence of a clandestine release of B anthracis as a biological weapon. The appearance of even a single previously healthy patient who becomes acutely ill with nonspecific febrile illness and symptoms and signs consistent with those listed in Table 1 and whose condition rapidly deteriorates should receive A, Portable chest radiograph of 56-year-old man with inhalational anthrax depicts a widened mediastinum (white arrowheads), bilateral hilar fullness, a right pleural effusion, and bilateral perihilar air-space disease. prompt consideration for a diagnosis of B, Noncontrast spiral CT scan depicts an enlarged and hyperdense right hilar lymph node (white arrowhead), anthrax infection. The recognition of cu- bilateral pleural effusions (black arrowheads), and edema of the mediastinal fat. Reprinted from Mayer et al.66 taneous cases of anthrax may also be the first evidence of an anthrax attack.70 sis. Massive ascites has occurred in some DIAGNOSIS The likely presence of abnormal find- cases of gastrointestinal anthrax.34 Ad- TABLE 2 lists the epidemiology, diag- ings on either chest x-ray film or chest vanced infection may appear similar to nostic tests, microbiology, and pathol- CT scan is diagnostically important. Al- the sepsis syndrome occurring in ei- ogy for a diagnosis of inhalational though anthrax does not cause a clas- ther inhalational or cutaneous an- anthrax infection. Given the rarity of sic bronchopneumonia pathologically, thrax.4 Some authors suggest that ag- anthrax infection, the first clinical or it can cause widened mediastinum, mas- gressive medical intervention as would laboratory suspicion of an anthrax ill- sive pleural effusions, air broncho- be recommended for inhalational an- ness must lead to early initiation of grams, necrotizing pneumonic lesions, thrax may reduce mortality. Given the antibiotic treatment pending con- and/or consolidation, as has been noted difficulty of early diagnosis of gastroin- firmed diagnosis and should provoke above.36,55,56,61,64-66 The result can be hy- testinal anthrax, however, mortality may immediate notification of the local or poxemia and chest imaging abnormali- be high.4 Postmortem examinations in state public health department, local ties that may or may not be clinically dis- Sverdlovsk showed gastrointestinal sub- hospital epidemiologist, and local or tinguishable from pneumonia. In the mucosal lesions in 39 of 42 patients,50 state public health laboratory. In the anthrax attacks of 2001, each of the first but all of these patients were also found United States, a Laboratory Response 10 patients had abnormal chest x-ray to have definitive pathologic evidence of Network (LRN) has been established film results and each of 8 patients for an inhalational source of infection. There through a collaboration of the Associa- whom CT scans were obtained had ab- were no gastrointestinal cases of an- tion of Public Health Laboratories normal results. These included wid- thrax diagnosed in either the Sverd- and the CDC (details are available at: ened mediastinum on chest radio- lovsk series or following the anthrax at- http://www.bt.cdc.gov/LabIssues/index graph and effusions on chest CT scan tacks of 2001. .asp). Currently 81 clinical laborato- (FIGURE 4). Such findings in a previ-

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF ANTHRAX AS A BIOLOGICAL WEAPON ously healthy patient with evidence of of inhalational anthrax in the 2001 se- ill patient but may be useful for epide- overwhelming febrile illness or sepsis ries. If cutaneous anthrax is suspected, miologic purposes. would be highly suggestive of ad- a Gram stain and culture of vesicular Postmortem findings are especially vanced inhalational anthrax. fluid should be obtained. If the Gram important following an unexplained The bacterial burden may be so great stain is negative or the patient is taking death. Thoracic hemorrhagic necrotiz- in advanced inhalational anthrax in- antibiotics already, punch biopsy should ing lymphadenitis and hemorrhagic nec- fection that bacilli are visible on Gram be performed, and specimens sent to a rotizing mediastinitis in a previously stain of peripheral blood, as was seen laboratory with the ability to perform im- healthy adult are essentially pathogno- following the 2001 attacks. The most munohistochemical staining or poly- monic of inhalational anthrax.50,58 Hem- useful microbiologic test is the stan- merase chain reaction assays.69,70 Blood orrhagic meningitis should also raise dard blood culture, which should show cultures should be obtained and antibi- strong suspicion of anthrax infec- growth in 6 to 24 hours. Each of the 8 otics should be initiated pending con- tion.32,50,58,75 However, given the rarity of patients who had blood cultures ob- firmation of the diagnosis of inhala- anthrax, a pathologist might not iden- tained prior to initiation of antibiotics tional or cutaneous anthrax. tify these findings as caused by anthrax had positive blood cultures.61 How- Nasal swabs were obtained in some unless previously alerted to this possi- ever, blood cultures appear to be ster- persons believed to be at risk of inha- bility. ilized after even 1 or 2 doses of antibi- lational anthrax following the anthrax If only a few patients present con- otics, underscoring the importance of attacks of 2001. Although a study has temporaneously, the clinical similar- obtaining cultures prior to initiation of shown the presence of B anthracis ity of early inhalational anthrax infec- antibiotic therapy (J. Gerberding, oral spores in nares of some monkeys fol- tion to other acute febrile respiratory communication, March 7, 2002). If the lowing experimental exposure to B an- infections may delay initial diagnosis laboratory has been alerted to the pos- thracis spores for some time after ex- although probably not for long. The se- sibility of anthrax, biochemical test- posure,72 the predictive value of the verity of the illness and its rapid pro- ing and review of colonial morphol- nasal swab test for diagnosing inhala- gression, coupled with unusual radio- ogy could provide a preliminary tional anthrax in humans is unknown logical findings, possible identification diagnosis 12 to 24 hours after inocu- and untested. It is not known how of B anthracis in blood or cerebrospi- lation of the cultures. Definitive diag- quickly antibiotics make spore recov- nal fluid, and the unique pathologic nosis could be promptly confirmed by ery on nasal swab tests impossible. One findings should serve as an early alarm. an LRN laboratory. However, if the patient who died from inhalational an- The index case of inhalational anthrax clinical laboratory has not been alerted thrax had a negative nasal swab.36 Thus, in the 2001 attacks was identified be- to the possibility of anthrax, B anthra- the CDC advised in the fall of 2001 that cause of an alert clinician who sus- cis may not be correctly identified. Rou- the nasal swab should not be used as a pected the disease on the basis of large tine procedures customarily identify a clinical diagnostic test. If obtained for gram-positive bacilli in cerebrospinal Bacillus species in a blood culture ap- an epidemiological purpose, nasal swab fluid in a patient with a compatible proximately 24 hours after growth, but results should not be used to rule out clinical illness, and as a result of the sub- some laboratories do not further iden- infection in a patient. Persons who have sequent analysis by laboratory staff who tify Bacillus species unless specifically positive nasal swab results for B an- had recently undergone bioterrorism requested. This is because the isola- thracis should receive a course of post- preparedness training.65 tion of Bacillus species most often rep- exposure antibiotic prophylaxis since resents growth of the common con- a positive swab would indicate that the VACCINATION taminant Bacillus cereus.71 Given the individual had been exposed to aero- The US anthrax vaccine, named an- possibility of future anthrax attacks, it solized B anthracis. thrax vaccine adsorbed (AVA), is an in- is recommended that routine clinical Antibodies to the protective antigen activated cell-free product, licensed in laboratory procedures be modified, so (PA) of B anthracis, termed anti-PA IgG, 1970, and produced by Bioport Corp, B anthracis is specifically excluded af- have been shown to confer immunity in Lansing, Mich. The vaccine is li- ter identification of a Bacillus species animal models following anthrax vacci- censed to be given in a 6-dose series. bacteremia unless there are compel- nation.73,74 Anti-PA IgG serologies have In 1997, it was mandated that all US ling reasons not to do so. If it cannot been obtained from several of those in- military active- and reserve-duty per- be excluded then the isolate should be volved in the 2001 anthrax attacks, but sonnel receive it.76 The vaccine is made transferred to an LRN laboratory. the results of these assays are not yet pub- from the cell-free filtrate of a nonen- Sputum culture and Gram stain are lished. Given the lack of data in hu- capsulated attenuated strain of B an- unlikely to be diagnostic of inhalational mans and the expected period required thracis.77 The principal antigen respon- anthrax, given the frequent lack of a to develop an anti-PA IgG response, this sible for inducing immunity is the pneumonic process.37 Gram stain of spu- test should not be used as a diagnostic PA.26,32 In the rabbit model, the quan- tum was reported positive in only 1 case test for anthrax infection in the acutely tity of antibody to PA has been corre-

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF ANTHRAX AS A BIOLOGICAL WEAPON lated with the level of protection against The Institute of Medicine (IOM) re- A. Henderson, oral communication, experimental anthrax infection.78 cently published a report on the safety February 2002). Preexposure vaccination with AVA and efficacy of AVA,84 which con- The use of AVA was not initiated has been shown to be efficacious against cluded that AVA is effective against in- immediately in persons believed to have experimental challenge in a number of halational anthrax and concluded that been exposed to B anthracis during the animal studies.78-80 A similar vaccine was if given with appropriate antibiotic 2001 anthrax attacks for a variety of rea- shown in a placebo-controlled human therapy, it may help prevent the devel- sons, including the unavailability of vac- trial to be efficacious against cutane- opment of disease after exposure. The cine supplies. Subsequently, near the end ous anthrax.81The efficacy of postex- IOM committee also concluded that of the 60-day period of antibiotic pro- posure vaccination with AVA has been AVA was acceptably safe. Committee phylaxis, persons deemed by investigat- studied in monkeys.40 Among 60 mon- recommendations for new research in- ing public health authorities to have been keys exposed to 8 LD50 of B anthracis clude studies to describe the relation- at high risk for exposure were offered spores at baseline, 9 of 10 control ani- ship between immunity and quantita- postexposure AVA series (3 inocula- mals died, and 8 of 10 animals treated tive antibody levels; further studies to tions at 2-week intervals, given on days with vaccine alone died. None of 29 ani- test the efficacy of AVA in combina- 1, 14, and 28) as an adjunct to pro- mals died while receiving doxycy- tion with antibiotics in preventing in- longed postexposure antibiotic prophy- cline, ciprofloxacin, or penicillin for 30 halational anthrax infection; studies of laxis. This group of affected persons was days; 5 developed anthrax once treat- alternative routes and schedules of ad- also offered the alternatives of continu- ment ceased. The remaining 24 all died ministration of AVA; and continued ing a prolonged course of antibiotics or when rechallenged. The 9 receiving monitoring of reported adverse events of receiving close medical follow-up with- doxycycline for 30 days plus vaccine at following vaccination. The committee out vaccination or additional antibiot- baseline and day 14 after exposure did did not evaluate the production pro- ics.87 This vaccine is licensed for use in not die from anthrax infection even af- cess used by the manufacturer. the preexposure setting, but because it ter being rechallenged.40 A recently published report85 ana- had not been licensed for use in the post- The safety of the anthrax vaccine has lyzed a cohort of 4092 women at 2 mili- exposure context, it was given under been the subject of much study. A re- tary bases from January 1999 to March investigational new drug procedures. cent report reviewed the results of sur- 2000. The study compared pregnancy The working group continues to con- veillance for adverse events in the De- rates and adverse birth outcomes be- clude that vaccination of exposed per- partment of Defense program of 1998- tween groups of women who had been sons following a biological attack in con- 2000.82 At the time of that report, vaccinated with women who had not junction with antibiotic administration 425976 service members had received been vaccinated and the study found for 60 days following exposure provide 1620793 doses of AVA. There were that anthrax vaccination with AVA had optimal protection to those exposed. higher rates of local reactions to the vac- no effect on pregnancy or adverse birth However, until ample reserve stock- cine in women than men, but “no pat- outcomes. piles of vaccine are available, reliance terns of unexpected local or systemic ad- A human live attenuated vaccine has must be placed on antibiotic adminis- verse events” were identified.82 A recent been produced and used in countries tration. To date, there have been no review of safety of AVA anthrax vacci- of the former Soviet Union.86 In the reported cases of anthrax infection among nation in employees of the United States Western world, live attenuated vac- those exposed in the 2001 anthrax attacks Army Medical Research Institute of In- cines have been considered unsuit- who took prophylactic antibiotics, even fectious Diseases (USAMRIID) over the able for use in humans because of safety in those persons not complying with the past 25 years reported that 1583 per- concerns.86 complete 60-day course of therapy. sons had received 10722 doses of AVA.83 Current vaccine supplies are lim- Preexposure vaccination of some per- One percent of these inoculations (101/ ited, and the US production capacity sons deemed to be in high-risk groups 10722) were associated with 1 or more remains modest. Bioport is the single should be considered when substan- systemic events (defined as headache, US manufacturing facility for the tial supplies of vaccine become avail- malaise, myalgia, fever, nausea, vomit- licensed anthrax vaccine. Production able. A fast-track program to develop ing, dizziness, chills, diarrhea, hives, an- has only recently resumed after a halt recombinant anthrax vaccine is now orexia, arthralgias, diaphoresis, blurred required the company to alter produc- under way. This may lead to more plen- vision, generalized itching, or sore tion methods so that it conformed to tiful vaccine stocks as well as a prod- throat). The most frequently reported the US Food and Drug Administration uct that requires fewer inoculations.88 systemic adverse event was headache (FDA) Good Manufacturing Practice Studies to evaluate intramuscular vs (0.4% of doses). Local or injection site standard. Bioport has a contract to subcutaneous routes of administra- reactions were reported in 3.6%. No produce 4.6 million doses of vaccine tion and less frequent dosing of AVA long-term sequelae were reported in this for the US Department of Defense that are also under way. (J. Hughes, oral series. cannot be met until at least 2003 (D. communication, February 2002.)

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

THERAPY fection with ciprofloxacin has not been intravenously with 2 or more antibiot- Recommendations for antibiotic and studied in humans, animal models sug- ics active against B anthracis had a greater vaccine use in the setting of an aerosol- gest excellent efficacy.40,56,93 In vitro data chance of survival.61 Given the limited ized B anthracis attack are conditioned suggest that other fluoroquinolone an- number of persons who developed in- by a very small series of cases in hu- tibiotics would have equivalent effi- halational anthrax, the paucity of com- mans, a limited number of studies in ex- cacy although no animal data using a parative data, and other uncertainties, it perimental animals, and the possible ne- primate model of inhalational anthrax remains unclear whether the use of 2 or cessity of treating large numbers of are available.92 Penicillin, doxycycline, more antibiotics confers a survival ad- casualties. A number of possible thera- and ciprofloxacin are approved by the vantage, but combination therapy is a peutic strategies have yet to be ex- FDA for the treatment of inhalational reasonable therapeutic approach in the plored experimentally or to be submit- anthrax infection,56,89,90,94 and other face of life-threatening illness. Another ted for approval to the FDA. For these antibiotics are under study. Other drugs factor supporting the initiation of com- reasons, the working group offers con- that are usually active in vitro include bination antibiotic therapy for treat- sensus recommendations based on the clindamycin, rifampin, imipenem, ment of inhalational anthrax is the best available evidence. The recommen- aminoglycosides, chloramphenicol, van- possibility that an engineered strain of dations do not necessarily represent uses comycin, cefazolin, tetracycline, lin- B anthracis resistant to 1 or more anti- currently approved by the FDA or an of- ezolid, and the macrolides. biotics might be used in a future attack. ficial position on the part of any of the Reports have been published of a Some infectious disease experts have also federal agencies whose scientists par- B anthracis strain that was engineered advocated the use of clindamycin, cit- ticipated in these discussions and will to resist the tetracycline and penicillin ing the theoretical benefit of diminish- need to be revised as further relevant in- classes of antibiotics.95 Balancing con- ing bacterial toxin production, a strat- formation becomes available. siderations of treatment efficacy with egy used in some toxin-mediated Given the rapid course of symptom- concerns regarding resistance, the streptococcal infections.98 There are no atic inhalational anthrax, early antibi- working group in 1999 recommended data as yet that bear specifically on this otic administration is essential. A delay that ciprofloxacin or other fluoroqui- question. Central nervous system pen- of antibiotic treatment for patients with nolone therapy be initiated in adults etration is another consideration; doxy- anthrax infection may substantially with presumed inhalational anthrax in- cycline or fluoroquinolone may not reach lessen chances for survival.89,90 Given the fection.3 It was advised that antibiotic therapeutic levels in the cerebrospinal difficulty in achieving rapid microbio- resistance to penicillin- and tetracy- fluid. Thus, in the aftermath of the an- logic diagnosis of anthrax, all persons in cline-class antibiotics should be as- thrax attacks, some infectious disease au- high-risk groups who develop fever or sumed following a terrorist attack un- thorities recommended preferential use evidence of systemic disease should start til laboratory testing demonstrated of ciprofloxacin over doxycycline, plus receiving therapy for possible anthrax in- otherwise. Once the antibiotic suscep- augmentation with chloramphenicol, rif- fection as soon as possible while await- tibility of the B anthracis strain of the ampin, or penicillin when meningitis is ing the results of laboratory studies. index case had been determined, the established or suspected. There are no controlled clinical stud- most widely available, efficacious, and The B anthracis isolate recovered ies for the treatment of inhalational least toxic antibiotic was recom- from patients with inhalational an- anthrax in humans. Thus, antibiotic regi- mended for patients requiring treat- thrax was susceptible to all of the an- mens commonly recommended for em- ment and persons requiring postexpo- tibiotics expected in a naturally occur- pirical treatment of sepsis have not been sure prophylaxis. Since the 1999 ring strain.97 This isolate showed an studied. In fact, natural strains of B an- consensus statement publication, a inducible ␤-lactamase in addition to a thracis are resistant to many of the an- study96 demonstrated the develop- constitutive cephalosporinase. The im- tibiotics used in empirical regimens for ment of in vitro resistance of an iso- portance of the inducible ␤-lactamase sepsis treatment, such as those regi- late of the Sterne strain of B anthracis is unknown; these strains are highly mens based on the extended-spectrum to ofloxacin (a fluoroquinolone closely susceptible to penicillin in vitro, with cephalosporins.91,92 Most naturally oc- related to ciprofloxacin) following sub- minimum inhibiting concentrations less curring B anthracis strains are sensitive culturing and multiple cell passage. than .06 µg/mL. A theoretical concern to penicillin, which historically has been Following the anthrax attacks of 2001, is that this sensitivity could be over- the preferred anthrax therapy. Doxycy- the CDC97 offered guidelines advocat- come with a large bacterial burden. For cline is the preferred option among the ing use of 2 or 3 antibiotics in combina- this reason, the CDC advised that pa- tetracycline class because of its proven tion in persons with inhalational an- tients with inhalational anthrax should efficacy in monkey studies56 and its ease thrax based on susceptibility testing with not be treated with penicillin or amoxi- of administration. Other members of this epidemic strains. Limited early informa- cillin as monotherapy and that cipro- class of antibiotics are suitable alterna- tion following the attacks suggested that floxacin or doxycycline be considered tives. Although treatment of anthrax in- persons with inhalational anthrax treated the standards based on in vitro activ-

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

Table 3. Recommended Therapy for Inhalational Anthrax Infection in the Contained Casualty Settinga,b Category Initial IV Therapy c,d Duration Adults Ciprofloxacin, 400 mg every 12 h IV treatment initiallye before switching to oral antimicrobial therapy or when clinically appropriate: Doxycycline, 100 mg every 12 hf Ciprofloxacin 500 mg twice daily and or 1 or 2 Additional antimicrobialsd Doxycycline 100 mg twice daily Continue oral and IV treatment for 60 dj Children Ciprofloxacin, 10-15 mg/kg every 12 hg,h IV treatment initiallye before switching to oral antimicrobial therapy or when clinically appropriate: Doxycyclinef,I for those aged Ciprofloxacin 10-15 mg/kg every 12 hh Ͼ8 y and weight Ͼ45 kg: 100 mg every 12 h; or Ͼ8 y and weight Յ45 kg: 2.2 mg/kg every 12 h; Doxycyclinei for those aged Յ8 y: 2.2 mg/kg every 12 h Ͼ8 y and weight Ͼ45 kg: 100 mg twice daily and Ͼ8 y and weight Յ45 kg: 2.2 mg/kg twice daily 1 or 2 Additional antimicrobialsd Յ8 y: 2.2 mg/kg 2 daily Continue oral and IV treatment for 60 d j Pregnant womenk Same for nonpregnant adults IV treatment initially before switching to oral antimicrobial therapy when clinically appropriateb; oral therapy regimens are the same for nonpregnant adults Immunocompromised persons Same for nonimmunocompromised adults and children aThis table is adapted with permission from Morbidity and Mortality Weekly Report.97 For gastrointestinal and oropharyngeal anthrax, use regimens recommended for inhalational anthrax. bCiprofloxacin or doxycycline should be considered an essential part of first-line therapy for inhalational anthrax. cSteroids may be considered as an adjunct therapy for patients with severe edema and for meningitis based on experience with bacterial meningitis of other etiologies. dOther agents with in vitro activity include rifampin, vancomycin, penicillin, ampicillin, chloramphenicol, imipenem, clindamycin, and clarithromycin. Because of concerns of constitutive and inducible ␤-lactamases in Bacillus anthracis, penicillin and ampicillin should not be used alone. Consultation with an infectious disease specialist is advised. eInitial therapy may be altered based on clinical course of the patient; 1 or 2 antimicrobial agents may be adequate as the patient improves. fIf meningitis is suspected, doxycycline may be less optimal because of poor central nervous system penetration. gIf intravenous (IV) ciprofloxacin is not available, oral ciprofloxacin may be acceptable because it is rapidly and well absorbed from the gastrointestinal tract with no substantial loss by first-pass metabolism. Maximum serum concentrations are attained 1 to 2 hours after oral dosing but may not be achieved if vomiting or ileus is present. hIn children, ciprofloxacin dosage should not exceed 1 g/d. iThe American Academy of Pediatrics recommends treatment of young children with tetracyclines for serious infections (ie, Rocky Mountain spotted fever). jBecause of the potential persistence of spores after an aerosol exposure, antimicrobial therapy should be continued for 60 days. kAlthough tetracyclines are not recommended during pregnancy, their use may be indicated for life-threatening illness. Adverse effects on developing teeth and bones of fetus are dose related; therefore, doxycycline might be used for a short time (7-14 days) before 6 months of gestation. The high death rate from the infection outweighs the risk posed by the antimicrobial agent. ity, efficacy in the monkey model, and In experimental animals, antibiotic or antibiotic intolerance. For cutane- FDA approval. therapy during anthrax infection has ous lesions associated with extensive In the contained casualty setting (a prevented development of an immune edema or for cutaneous lesions of the situation in which a modest number of response.40,95 This suggests that even if head and neck, clinical management patients require therapy), the work- the antibiotic-treated patient survives an- should be conservative as per inhala- ing group supports these new CDC anthrax infection, the risk of recurring tional anthrax treatment guidelines in tibiotic recommendations97 (TABLE 3) disease may persist for a prolonged pe- Table 3. Although previous guidelines and advises the use of intravenous an- riod because of the possibility of de- have suggested treating cutaneous antibiotic administration. These recom- layed germination of spores. There- thrax for 7 to 10 days,32,71 the working mendations will need to be revised as fore, we recommend that antibiotic group recommends treatment for 60 days new data become available. therapy be continued for at least 60 days postexposure in the setting of bioterror- If the number of persons requiring postexposure, with oral therapy replac- ism, given the presumed concomitant in- therapy following a bioterrorist attack ing intravenous therapy when the pa- halational exposure to the primary aero- with anthrax is sufficiently high (ie, a tient is clinically stable enough to take sol. Treatment of cutaneous anthrax mass casualty setting), the working oral medication. generally prevents progression to sys- group recognizes that combination drug Cutaneous anthrax historically has temic disease although it does not pre- therapy and intravenous therapy may been treated with oral penicillin. For rea- vent the formation and evolution of the no longer be possible for reasons of lo- sons articulated above, the working eschar. Topical therapy is not useful.4 gistics and/or exhaustion of equip- group recommends that oral fluoroqui- In addition to penicillin, the fluoro- ment and antibiotic supplies. In such nolone or doxycycline in the adult dos- quinolones and the tetracycline class of circumstances, oral therapy may be the age schedules described in TABLE 5 be antibiotics, other antibiotics effective in only feasible option (TABLE 4). The used to treat cutaneous anthrax until an- vitro include chloramphenicol, clinda- threshold number of cases at which tibiotic susceptibility is proven. Amoxi- mycin, extended-spectrum penicillins, combination and parenteral therapy be- cillin is a suitable alternative if there are macrolides, aminoglycosides, vancomy- come impossible depends on a variety contraindications to fluoroquinolones or cin, cefazolin, and other first-genera- of factors, including local and re- doxycycline such as pregnancy, lactat- tion cephalosporins.91,99 The efficacy of gional health care resources. ing mother, age younger than 18 years, these antibiotics has not yet been tested

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

Table 4. Recommended Therapy for Inhalational Anthrax Infection in the Mass Casualty Setting or for Postexposure Prophylaxis* Alternative Therapy if Strain Duration After Category Initial Oral Therapy† Is Proved Susceptible Exposure, d Adults Ciprofloxacin, 500 mg orally every 12 h Doxycycline, 100 mg orally every 12 h‡ 60 Amoxicillin, 500 mg orally every 8 h§ Children Ciprofloxacin, 20-30 mg/kg per d orally taken Weight Ն20 kg: amoxicillin, 500 mg orally every 8 h§ 60 in 2 daily doses, not to exceed 1 g/d࿣ Weight Ͻ20 kg: amoxicillin, 40 mg/kg taken orally in 3 doses every 8 h§ Pregnant women¶ Ciprofloxacin, 500 mg orally every 12 h Amoxicillin, 500 mg orally every 8 h§ 60 Immunosuppressed persons Same as for nonimmunosuppressed adults and children *Some of these recommendations are based on animal studies or in vitro studies and are not approved by the US Food and Drug Administration. †In vitro studies suggest ofloxacin (400 mg orally every 12 hours, or levofloxacin, 500 mg orally every 24 hours) could be substituted for ciprofloxacin. ‡In vitro studies suggest that 500 mg of tetracycline orally every 6 hours could be substituted for doxycycline. In addition, 400 mg of gatifloxicin or monifloxacin, both fluoroquinolones with mechanisms of action consistent with ciprofloxacin, taken orally daily could be substituted. §According to the Centers for Disease Control and Prevention recommendations, amoxicillin is suitable for postexposure prophylaxis only after 10 to 14 days of fluoroquinolones or doxycycline treatment and then only if there are contraindications to these 2 classes of medications (eg, pregnancy, lactating mother, age Ͻ18 years, or intolerance of other antibiotics). ࿣Doxycycline could also be used if antibiotic susceptibility testing, exhaustion of drug supplies, adverse reactions preclude use of ciprofloxacin. For children heavier than 45 kg, adult dosage should be used. For children lighter than 45 kg, 2.5 mg/kg of doxycycline orally every 12 hours should be used. ¶See “Management of Pregnant Population” for details. in humans or animal studies. The work- Table 5. Recommended Therapy for Cutaneous Anthrax Infection Associated With ing group recommends the use of these a Bioterrorism Attack* antibiotics only to augment fluoroqui- Category Initial Oral Therapy† Duration, d‡ nolones or tetracyclines or if the pre- Adults Ciprofloxacin, 500 mg twice daily† 60 ferred drugs are contraindicated, not or available, or inactive in vitro in suscep- Doxycycline, 100 mg twice daily† Children§ Ciprofloxacin, 10-15 mg/kg every 12 h 60 tibility testing. B anthracis strains ex- (not to exceed 1 g/d)† hibit natural resistance to sulfamethoxa- or zole, trimethoprim, cefuroxime, Doxycycline for those aged§ Ͼ8 y and weight Ͼ45 kg: 100 mg every 12 h cefotaxime sodium, aztreonam, and Ͼ8 y and weight Յ45 kg: 2.2 mg/kg every 12 h ceftazidime.91,92,99 Therefore, these anti- Յ8 y: 2.2 mg/kg every 12 h biotics should not be used. Pregnant women࿣ Ciprofloxacin, 500 mg twice daily 60 or Pleural effusions were present in all Doxycycline, 100 mg twice daily of the first 10 patients with inhala- Immunocompromised Same for nonimmunocompromised adults and children tional anthrax in 2001. Seven needed persons drainage of their pleural effusions, 3 re- *This table is adapted with permission from the Morbidity and Mortality Weekly Report.98 Cutaneous anthrax with signs of systemic involvement, extensive edema, or lesions on the head or neck require intravenous therapy, and a mul- 69 quired chest tubes. Future patients tidrug approach is recommended (Table 3). †Ciprofloxacin or doxycycline should be considered first-line therapy. Amoxicillin can be substituted if a patient cannot with inhalational anthrax should be ex- take a fluoroquinolone or tetracycline class drug. Adults are recommended to take 500 mg of amoxicillin orally 3 pected to have pleural effusions that will times a day. For children, 80 mg/kg of amoxicillin to be divided into 3 doses in 8-hour increments is an option for completion of therapy after clinical improvement. Oral amoxicillin dose is based on the need to achieve appropriate likely require drainage. minimum inhibitory concentration levels. ‡Previous guidelines have suggested treating cutaneous anthrax for 7 to 10 days, but 60 days is recommended for bioterrorism attacks, given the likelihood of exposure to aerosolized Bacillus anthracis. Postexposure Prophylaxis §The American Academy of Pediatrics recommends treatment of young children with tetracyclines for serious infections (eg, Rocky Mountain spotted fever). Guidelines for which populations ࿣Although tetracyclines or ciprofloxacin is not recommended during pregnancy, their use may be indicated for life- threatening illness. Adverse effects on developing teeth and bones of a fetus are dose related; therefore, doxycycline would require postexposure prophy- might be used for a short time (7-14 days) before 6 months of gestation. laxis to prevent inhalational anthrax following the release of a B anthracis aerosol as a biological weapon will need to posure to a B anthracis aerosol. There- severe allergic reactions related to the be developed by public health officials fore, for postexposure prophylaxis, we medications, but no persons required depending on epidemiological circum- recommend the same antibiotic regi- hospitalization because of an adverse stances. These decisions would re- men as that recommended for treat- drug reaction.101 Many persons did not quire estimates of the timing, loca- ment of mass casualties; prophylaxis begin or complete their recommended tion, and conditions of the exposure.100 should be continued for at least 60 days antibiotic course for a variety of rea- Ongoing case monitoring would be postexposure (Table 4). Preliminary sons, including gastrointestinal tract in- needed to define the high-risk groups, analysis of US postal workers who were tolerance, underscoring the need for to direct follow-up, and to guide the ad- advised to take 60 days of antibiotic pro- careful medical follow-up during the pe- dition or deletion of groups requiring phylaxis for exposure to B anthracis riod of prophylaxis.101 In addition, given postexposure prophylaxis. spores following the anthrax attacks of the uncertainties regarding how many There are no FDA-approved postex- 2001 showed that 2% sought medical at- weeks or months spores may remain la- posure antibiotic regimens following ex- tention because of concern of possible tent in the period following discontinu-

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF ANTHRAX AS A BIOLOGICAL WEAPON ation of postexposure prophylaxis, per- to 21 days of fluoroquinolone or doxy- an alternate to ciprofloxacin in preg- sons should be instructed to report cycline administration because of the nant women. If doxycycline is used in immediately flulike symptoms or fe- concern about the presence of a ␤-lac- pregnant women, periodic liver func- brile illness to their physicians who tamase.102 In a contained casualty set- tion testing should be performed. No should then evaluate the need to ini- ting, the working group recommends adequate controlled trials of penicillin tiate treatment for possible inhala- that children with inhalational an- or amoxicillin administration during tional anthrax. As noted above, postex- thrax receive intravenous antibiotics pregnancy exist. However, the CDC rec- posure vaccination is recommended as (Table 3). In a mass casualty setting and ommends penicillin for the treatment an adjunct to postexposure antibiotic as postexposure prophylaxis, the work- of syphilis during pregnancy and prophylaxis if vaccine is available. ing group recommends that children re- amoxicillin as a treatment alternative ceive oral antibiotics (Table 4). for chlamydial infections during preg- Management of Special Groups The US anthrax vaccine is licensed nancy.94 According to CDC recommen- Consensus recommendations for spe- for use only in persons aged 18 to 65 dations, amoxicillin is suitable postex- cial groups as set forth herein reflect the years because studies to date have been posure prophylaxis or treatment of clinical and evidence-based judgments conducted exclusively in this group.77 inhalational anthrax in pregnancy only of the working group and at this time No data exist for children, but based on after 14 to 21 days of fluoroquinolone do not necessarily correspond with FDA- experience with other inactivated vac- or doxycycline administration.102 approved use, indications, or labeling. cines, it is likely that the vaccine would Ciprofloxacin (and other fluoroqui- Children. It has been recommended be safe and effective. nolones), penicillin, and doxycycline that ciprofloxacin and other fluoroqui- Pregnant Women. Fluoroquino- (and other tetracyclines) are each ex- nolones should not be used in children lones are not generally recommended creted in breast milk. Therefore, a younger than 16 to 18 years because of during pregnancy because of their known breastfeeding woman should be treated a link to permanent arthropathy in ado- association with arthropathy in adoles- or given prophylaxis with the same an- lescent animals and transient arthropa- cent animals and small numbers of chil- tibiotic as her infant based on what is thy in a small number of children.94 How- dren. Animal studies have discovered no most safe and effective for the infant. ever, balancing these risks against the evidence of teratogenicity related to cipro- Immunosuppressed Persons. The risks of anthrax infections caused by an floxacin, but no controlled studies of antibiotic treatment or postexposure engineered antibiotic-resistant strain, the ciprofloxacin in pregnant women have prophylaxis for anthrax among those working group recommends that cipro- been conducted. Balancing these pos- who are immunosuppressed has not floxacin be used as a component of com- sible risks against the concerns of anthrax been studied in human or animal mod- bination therapy for children with inha- due to engineered antibiotic-resistant els of anthrax infection. Therefore, the lational anthrax. For postexposure strains, the working group recom- working group consensus recom- prophylaxis or following a mass casu- mends that pregnant women receive mends administering antibiotics in the alty attack, monotherapy with fluoro- ciprofloxacin as part of combination same regimens recommended for im- quinolones is recommended by the therapy for treatment of inhalational munocompetent adults and children. working group97 (Table 4). anthrax (Table 3). We also recommend The American Academy of Pediat- that pregnant women receive fluoroqui- INFECTION CONTROL rics has recommended that doxycy- nolones in the usual adult dosages for There are no data to suggest that pa- cline not be used in children younger postexposure prophylaxis or mono- tient-to-patient transmission of an- than 9 years because the drug has re- therapy treatment in the mass casualty thrax occurs and no person-to-person sulted in retarded skeletal growth in in- setting (Table 4). The tetracycline class transmission occurred following the an- fants and discolored teeth in infants and of antibiotics has been associated with thrax attacks of 2001.18,67 Standard bar- children.94 However, the serious risk of both toxic effects in the liver in preg- rier isolation precautions are recom- infection following an anthrax attack nant women and fetal toxic effects, mended for hospitalized patients with supports the consensus recommenda- including retarded skeletal growth.94 all forms of anthrax infection, but the tion that doxycycline, instead of cipro- Balancing the risks of anthrax infec- use of high-efficiency particulate air fil- floxacin, be used in children if antibi- tion with those associated with doxy- ter masks or other measures for air- otic susceptibility testing, exhaustion cycline use in pregnancy, the working borne protection are not indicated.103 of drug supplies, or adverse reactions group recommends that doxycycline There is no need to immunize or pro- preclude use of ciprofloxacin. can be used as an alternative to cipro- vide prophylaxis to patient contacts (eg, According to CDC recommenda- floxacin as part of combination therapy household contacts, friends, cowork- tions, amoxicillin was suitable for treat- in pregnant women for treatment of in- ers) unless a determination is made that ment or postexposure prophylaxis of halational anthrax. For postexposure they, like the patient, were exposed to possible anthrax infection following the prophylaxis or in mass casualty set- the aerosol or surface contamination at anthrax attacks of 2001 only after 14 tings, doxycycline can also be used as the time of the attack.

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

In addition to immediate notifica- indoor environments, can rapidly de- a common-source epidemic,3,60 and it tion of the hospital epidemiologist and liver high concentrations of spores to is possible to account for virtually all state health department, the local hos- persons in the vicinity.104 In some cir- confirmed cases having occurred within pital microbiology laboratories should cumstances, indoor airflows, activity pat- the area of the plume on the day of the be notified at the first indication of an- terns, and heating, ventilation, and air accident. Moreover, if secondary aero- thrax so that safe specimen processing conditioning systems may transport solization had been important, new under biosafety level 2 conditions can spores to others parts of the building. cases would have likely continued well be undertaken as is customary in most Following the period of primary aero- beyond the observed 43 days. hospital laboratories.56 A number of dis- solization, B anthracis spores may settle Although persons working with ani- infectants used for standard hospital on surfaces, possibly in high concen- mal hair or hides are known to be at in- infection control, such as hypochlo- trations. The risk that B anthracis spores creased risk of developing inhalational rite, are effective in cleaning environ- might pose by a process of secondary or cutaneous anthrax, surprisingly few mental surfaces contaminated with aerosolization (resuspension of spores occupational exposures in the United infected bodily fluids.22,103 into the air) is uncertain and is likely States have resulted in disease. During Proper burial or cremation of hu- dependent on many variables, includ- the first half of the 20th century, a sig- mans and animals who have died being the quantity of spores on a sur- nificant number of goat hair mill work- cause of anthrax infection is important face; the physical characteristics of the ers were heavily exposed to aerosolized in preventing further transmission of the powder used in the attack; the type of spores. Mandatory vaccination became disease. Serious consideration should be surface; the nature of the human or me- a requirement for working in goat hair given to cremation. Embalming of bod- chanical activity that occurs in the af- mills only in the 1960s. Prior to that, ies could be associated with special fected area and host factors. many unvaccinated person-years of high- risks.103 If autopsies are performed, all re- A variety of rapid assay kits are avail- risk exposure had occurred, but only 13 lated instruments and materials should able to detect B anthracis spores on en- cases of inhalational anthrax were re- be autoclaved or incinerated.103 The CDC vironmental surfaces. None of these kits ported.27,54 One study of environmental can provide advice on postmortem pro- has been independently evaluated or en- exposure, conducted at a Pennsylvania cedures in anthrax cases. dorsed by the CDC, FDA, or Environ- goat hair mill, showed that workers in- mental Protection Agency, and their haled up to 510 B anthracis particles of DECONTAMINATION functional characteristics are not at least 5 µm in diameter per person per Recommendations for decontamina- known.105 Many false-positive results 8-hour shift.54 These concentrations of tion in the event of an intentional aero- occurred following the anthrax at- spores were constantly present in the en- solization of B anthracis spores are based tacks of 2001. Thus, any result using vironment during the time of this study, on evidence concerning aerosolization currently available rapid assay kits does but no cases of inhalational anthrax techniques, predicted spore survival, en- not necessarily signify the presence of occurred. vironmental exposures at Sverdlovsk and B anthracis; it is simply an indication Field studies using B anthracis–like among goat hair mill workers, and en- that further testing is required by a cer- surrogates have been carried out by US vironmental data collected following the tified microbiology laboratory. Simi- Army scientists seeking to determine anthrax attacks of 2001. The greatest risk larly, the sensitivity and false-negative the risk of secondary aerosolization. to humans exposed to an aerosol of B an- rate of disease kits are unknown. One study concluded that there was no thracis spores occurs when spores first At Sverdlovsk, no new cases of in- significant threat to personnel in areas are made airborne, the period called pri- halational anthrax developed beyond 43 contaminated by 1 million spores per mary aerosolization. The aerobiological days after the presumed date of re- square meter either from traffic on as- factors that affect how long spores release. None were documented during phalt-paved roads or from a runway main airborne include the size of the dis- the months and years afterward, de- used by helicopters or jet aircraft.106 A persed particles and their hydrostatic spite only limited decontamination and separate study showed that in areas of properties.100 Technologically sophisti- vaccination of 47000 of the city’s 1 mil- ground contaminated with 20 million cated dispersal methods, such as aero- lion inhabitants.59 Some have ques- Bacillus subtilis spores per square me- sol release from military aircraft of large tioned whether any of the cases with on- ter, a soldier exercising actively for a quantities of B anthracis spores manipu- set of disease beyond 7 days after release 3-hour period would inhale between lated for use in a weapon, are poten- might have represented illness follow- 1000 and 15000 spores.107 tially capable of exposing high num- ing secondary aerosolization from the Much has been written about the bers of victims over large areas. Recent ground or other surfaces. It is impos- technical difficulty of decontaminat- research by Canadian investigators has sible to state with certainty that sec- ing an environment contaminated with demonstrated that even “low-tech” de- ondary aerosolizations did not occur in B anthracis spores. A classic case is the livery systems, such as the opening of en- Sverdlovsk, but it appears unlikely. The experience at Gruinard Island, Scot- velopes containing powdered spores in epidemic curve reported is typical for land. During World War II, British mili-

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF ANTHRAX AS A BIOLOGICAL WEAPON tary undertook explosives testing with chanical systems of mail processing in full expert analysis of the contami- B anthracis spores. Spores persisted and a specific building would affect the risk nated environment and the anthrax remained viable for 36 years following of disease acquisition in a worker han- weapon used in the attack and be made the conclusion of testing. Decontami- dling a contaminated letter in that fa- in consultation with experts on envi- nation of the island occurred in stages, cility. It is still uncertain what the mini- ronmental remediation. If vaccines were beginning in 1979 and ending in 1987 mum dose of spores would be to cause available, postexposure vaccination when the island was finally declared infection in humans although it may might be a useful intervention for those fully decontaminated. The total cost is theoretically be as few as 1 to 3 spores.47 working in highly contaminated areas, unpublished, but materials required in- The mechanisms of disease acquisi- because it could further lower the risk cluded 280 tons of formaldehyde and tion in the 2 fatal inhalational anthrax of anthrax infection. 2000 tons of seawater.108 cases in New York City and in Con- In the setting of an announced al- Following the anthrax attacks of necticut remain unknown although it leged B anthracis release, such as the se- 2001, substantial efforts were under- is speculated that disease in these 2 ries of anthrax hoaxes occurring in many taken to decontaminate environmen- cases followed the inhalation of small areas of the United States in 1998111 and tal surfaces exposed to B anthracis numbers of spores present in some following the anthrax attacks of 2001, spores. Sections of the Hart Senate of- manner in “cross-contaminated” mail. any person coming in direct physical fice building in Washington, DC, con- The discovery of B anthracis spores contact with a substance alleged to be taminated from opening a letter laden in a contaminated letter in the office of containing B anthracis should thor- with B anthracis, were reopened only Sen Daschle in the Hart office build- oughly wash the exposed skin and ar- after months of decontamination pro- ing led the Environmental Protection ticles of clothing with soap and wa- cedures at an estimated cost of $23 mil- Agency to conduct tests in this office ter.112 In addition, any person in direct lion.109 Decontamination efforts at many to assess the risk of secondary aerosol- physical contact with the alleged sub- other buildings affected by the an- ization of spores. Prior to the initia- stance should receive postexposure anthrax attacks of 2001 have not yet been tion of decontamination efforts in the tibiotic prophylaxis until the sub- completed. Hart building, 17 blood agar gel plates stance is proved not to be B anthracis. Prior to the anthrax attacks of 2001, were placed around the office and nor- The anthrax attacks of 2001 and new re- there had been no recognition or sci- mal activity in the office was simu- search104 have shown that opening let- entific study showing that B anthracis lated. Sixteen of the 17 plates yielded ters containing substantial quantities of spores of “weapons grade” quality B anthracis. Although this experiment B anthracis spores in certain conditions would be capable of leaking out the did not allow conclusions about the spe- can confer risk of disease to persons at edges of envelopes or through the pores cific risk of persons developing an- some distance from the location of where of envelopes, with resulting risk to the thrax infection in this context, it did the letter was opened. For this reason, health of those handling or processing demonstrate that routine activity in an when a letter is suspected of contain- those letters. When it became clear that environment contaminated with B an- ing (or proved to contain) B anthracis, the Florida case of anthrax was likely thracis spores could cause significant immediate consultation with local and caused by a letter contaminated with spore resuspension.110 state public health authorities and the B anthracis, assessment of postal work- Given the above considerations, if an CDC for advised medical management ers who might have handled or pro- environmental surface is proved to be is warranted. cessed that letter showed no illness.69 contaminated with B anthracis spores in When the anthrax cases were discov- the immediate area of a spill or close Additional Research ered, each was linked to a letter that had proximity to the point of release of B an- Development of a recombinant an- been opened. At first, there was no evi- thracis biological weapons, the work- thrax vaccine that would be more eas- dence of illness among persons han- ing group believes that decontamina- ily manufactured and would require dling or processing unopened mail. This tion of that area would likely decrease fewer doses should remain a top pri- fact influenced the judgment that the risk of acquiring anthrax by second- ority. Rapid diagnostic assays that could persons handling or processing un- ary aerosolization. However, as has been reliably identify early anthrax infec- opened B anthracis letters were not at demonstrated in environmental decon- tion and quickly distinguish from other risk. These judgments changed when tamination efforts following the an- flulike or febrile illnesses would be- illness was discovered in persons who thrax attacks of 2001, decontamina- come critical in the event of a large- had handled or processed unopened let- tion of buildings or parts of buildings scale attack. Simple animal models for ters in Washington, DC. Much re- following an anthrax attack is techni- use in comparing antibiotic prophylac- mains unknown about the risks to per- cally difficult. For these reasons, the tic and treatment strategies are also sons handling or processing unopened working group would advise that deci- needed. Operational research to bet- letters containing B anthracis spores. It sions about methods for decontamina- ter characterize risks posed by envi- is not well understood how the me- tion following an anthrax attack follow ronmental contamination of spores,

Downloaded from jama.ama-assn.org by guest on April 25, 2012 MANAGEMENT OF ANTHRAX AS A BIOLOGICAL WEAPON particularly inside buildings, and re- necticut, 2001. MMWR Morb Mortal Wkly Rep. 2001; 26. Brachman P, Friedlander A. Anthrax. In: Plotkin 50:1077-1079. S, Orenstein W, eds. Vaccines. 3rd ed. Philadelphia, search on approaches to minimize risk 3. Inglesby TV, Henderson DA, Bartlett JG, et al. An- Pa: WB Saunders Co; 1999:629-637. in indoor environments by means of air thrax as a biological weapon. JAMA. 1999;281:1735- 27. Brachman P, Friedlander A. Inhalation anthrax. 1745. Ann N Y Acad Sci. 1980;353:83-93. filters or methods for environmental 4. Lew D. Bacillus anthracis. In: Mandell GL, Ben- 28. Summary of notifiable diseases, 1945-1994. cleaning following a release are also nett JE, Dolin R, eds. Principles and Practice of Infec- MMWR Morb Mortal Wkly Rep. 1994;43:70-78. needed. A better understanding of the tious Disease. New York, NY: Churchill Livingstone 29. Human anthrax associated with an epizootic Inc; 1995:1885-1889. among livestock. MMWR Morb Mortal Wkly Rep. genetics and pathogenesis of anthrax, 5. Christopher G, Cieslak T, Pavlin J, Eitzen E. Bio- 2001;50:677-680. as well as mechanisms of virulence and logical warfare: a historical perspective. JAMA. 1997; 30. Myenye K, Siziya S, Peterson D. Factors associ- 278:412-417. ated with human anthrax outbreak in the Chikupo and immunity, will be of importance in the 6. Monterey Institute for International Studies chemi- Ngandu villages of Murewa district in Mashonaland prospective evaluation of new thera- cal and biological weapons resource page. Chemical East Province, Zimbabwe. Cent Afr J Med. 1996;42: peutic and diagnostic strategies. Novel and Biological Weapons. Monterey, Calif: Monterey 312-315. Institute for International Studies; 2001. Available at: 31. Sirisanthana T, Nelson K, Ezzell J, Abshire T. Se- therapeutic approaches with promise, http://cns.miis.edu/research/cbw/possess.htm. rological studies of patients with cutaneous and oral- such as the administration of competi- 7. Zilinskas RA. Iraq’s biological weapons. JAMA. 1997; pharyngeal anthrax from northern Thailand. Am J Trop 278:418-424. Med Hyg. 1988;39:575-581. tors against the protective antigen com- 8. Alibek K, Handelman S. Biohazard: The Chilling 32. Friedlander A. Anthrax. In: Zajtchuk R, Bellamy plex,113 should also be tested in ani- True Story of the Largest Covert Biological Weapons R, eds. Textbook of Military Medicine: Medical As- Program in the World. New York, NY: Random House; pects of Chemical and Biological Warfare. Washing- mals and developed where evidence 1999. ton, DC: Office of the Surgeon General, US Dept of supports this. Recent developments 9. Miller J. A germ-making plant. New York Times. the Army; 1997:467-478. such as the publishing of the B anthra- September 4, 2001; A1 33. Sirisanthana T, Navachareon N, Tharavichitkul P, 10. Public Health Service Office of Emergency Pre- Sirisanthana V, Brown AE. Outbreak of oral- cis genome and the discovery of the paredness. Proceedings of the Seminar on Respond- pharyngeal anthrax. Am J Trop Med Hyg. 1984;33: crystalline structure of the lethal and ing to the Consequences of Chemical and Biological 144-150. Terrorism; Washington, DC: US Dept of Health and 34. Dutz W, Saidi F, Kouhout E. Gastric anthrax with edema factor could hold great clinical Human Services; 1995. massive ascites. Gut. 1970;11:352-354. hope for both the prevention and treat- 11. WuDunn S, Miller J, Broad W. How Japan germ 35. Lincoln R, Hodges D, Klein F, et al. Role of the 114 terror alerted world. New York Times. May 26, 1998; lymphatics in the pathogenesis of anthrax. J Infect Dis. ment of anthrax infection. A1, 6. 1965;115:481-494. Ex Officio Participants in the Working Group on Ci- 12. Keim P, Smith K, Keys C, Takahashi H, Kurata T, 36. Borio L, Frank D, Mani V, et al. Death due to bio- vilian Biodefense: George Curlin, MD, National In- Kaufmann A. Molecular investigation of the Aum Shin- terrorism-related inhalational anthrax. JAMA. 2001; stitutes of Health, Bethesda, Md; Margaret Ham- rikyo anthrax release in Kameido, Japan. J Clin Mi- 286:2554-2559. burg, MD, Nuclear Threat Initiative, Washington, DC; crobiol. 2001;39:4566-4567. 37. Williams R. Bacillus anthracis and other spore Stuart Nightingale, MD, Office of Assistant Secretary 13. Washington File. New CIA report documents forming bacilli. In: Braude AI, Davis LE, Fierer J, eds. for Planning and Evaluation, DHHS, Washington, DC; global weapons proliferation trends. Washington, DC: Infectious Disease and Medical Microbiology. Phila- William Raub, PhD, Office of Public Health Prepared- US Dept of State; February 1, 2002. Available at: http:// delphia, Pa: WB Saunders Co; 1986:270-278. ness, DHHS, Washington, DC; Robert Knouss, MD, usinfo.state.gov/products/washfile. Accessed Febru- 38. Druett H, Henderson D, Packman L, Peacock S. Office of Emergency Preparedness, DHHS, Rockville, ary 1, 2002. Studies on respiratory infection. J Hyg. 1953;51:359- Md; Marcelle Layton, MD, Office of Communicable 14. Enserink M. Microbial genomics: TIGR begins as- 371. Disease, New York City Health Department, New York, sault on the anthrax genome. Science. 2002;295: 39. Hatch T. Distribution and deposition of inhaled NY; and Brian Malkin, formerly of FDA, Rockville, Md. 1442-1443. particles in respiratory tract. Bacteriol Rev. 1961;25: Funding/Support: Funding for this study primarily was 15. Kennedy H. Daschle letter bombshell billions of 237-240. provided by each participant’s institution or agency. anthrax spores. New York Daily News. October 31, 40. Friedlander AM, Welkos SL, Pitt ML, et al. Post- Disclaimers: In many cases, the indication and dos- 2001: 5. exposure prophylaxis against experimental inhala- ages and other information are not consistent with cur- 16. Health Aspects of Chemical and Biological Weap- tion anthrax. J Infect Dis. 1993;167:1239-1242. rent approved labeling by the US Food and Drug Ad- ons. Geneva, Switzerland: World Health Organiza- 41. Ross JM. The pathogenesis of anthrax following ministration (FDA). The recommendations on the use tion; 1970. the administration of spores by the respiratory route. of drugs and vaccine for uses not approved by the FDA 17. Office of Technology Assessment, US Congress. J Pathol Bacteriol. 1957;73:485-495. do not represent the official views of the FDA or of Proliferation of Weapons of Mass Destruction. Wash- 42. Hanna PC, Ireland JA. Understanding Bacillus an- any of the federal agencies whose scientists partici- ington, DC: US Government Printing Office; 1993. thracis pathogenesis. Trends Microbiol. 1999;7:180- pated in these discussions. Unlabeled uses of the prod- Publication OTA-ISC-559. 182. ucts recommended are noted in the sections of this 18. Meselson M, Guillemin J, Hugh-Jones M, et al. 43. Glassman H. Industrial inhalation anthrax. Bac- article in which these products are discussed. Where The Sverdlovsk anthrax outbreak of 1979. Science. teriol Rev. 1966;30:657-659. unlabeled uses are indicated, information used as the 1994;266:1202-1208. 44. Henderson DW, Peacock S, Belton FC. Observa- basis for the recommendation is discussed. The views, 19. Simon J. Biological terrorism: preparing to meet tions on the prophylaxis of experimental pulmonary opinions, assertions, and findings contained herein are the threat. JAMA. 1997;278:428-430. anthrax in the monkey. J Hyg. 1956;54:28-36. those of the authors and should not be construed as 20. Regis E. The History of America’s Secret Germ 45. Smith H, Keppie J. Observations on experimen- official US Department of Health and Human Ser- Warfare Project. New York, NY: Random House; 1999. tal anthrax. Nature. 1954;173:869-870. vices, US Department of Defense, or US Department 21. Kohout E, Sehat A, Ashraf M. Anthrax: a con- 46. Soviet Biological Warfare Threat. Washington, DC: of Army positions, policies, or decisions unless so des- tinuous problem in south west Iran. Am J Med Sci. Defense Intelligence Agency, US Dept of Defense; ignated by other documentation. 1964;247:565. 1986. Publication DST-161OF-057-86. Acknowledgment: The working group wishes to thank 22. Titball R, Turnbull P, Hutson R. The monitoring and 47. Peters CJ, Hartley DM. Anthrax inhalation and le- Jeanne Guillemin, PhD, Matthew Meselson, PhD, detection of Bacillus anthracis in the environment. Soc thal human infection. Lancet. 2002;359:710-711. Timothy Townsend, MD, Martin Hugh-Jones, MA, Appl Bacteriol Symp Ser. 1991;20:9S-18S. 48. Dixon TC, Meselson M, Guillemin, J, Hanna PC. VetMB, MPH, PhD, and Philip Brachman, MD, for their 23. l’agriculture Pdmd. Premiere Partie. I-la fievre char- Anthrax. N Engl J Med. 1999;341:815-826. review and commentary on the originally published bonneuse en Iran.1-historique—especes atteintes; 49. Friedlander AM. Microbiology: tackling anthrax. manuscript, and Molly D’Esopo for her efforts in the 1946. Located at: Archives De L’institute D’hessarek, Nature. 2001;414:160-161. preparation of the revised manuscript. Teheran, Iran. 50. Abramova FA, Grinberg LM, Yampolskaya O, 24. Pienaar U. Epidemiology of anthrax in wild ani- Walker DH. Pathology of inhalational anthrax in 42 mals and the control on anthrax epizootics in the Kruger cases from the Sverdlovsk outbreak in 1979. Proc Natl REFERENCES National Park, South Africa. Fed Proc. 1967;26:1496- Acad Sci U S A. 1993;90:2291-2294. 1. Carter A, Deutsch J, Zelicow P. Catastrophic ter- 1591. 51. Dalldorf F, Kaufmann AF, Brachman PS. Wool- rorism. Foreign Aff. 1998;77:80-95. 25. Dragon D, Rennie R. The ecology of anthrax sorters’ disease. Arch Pathol. 1971;92:418-426. 2. Investigation of bioterrorism-related anthrax: Con- spores. Can Vet J. 1995;36:295-301. 52. Gleiser CA, Berdjis CC, Harman HA, Gochenour

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

WS. Pathology of experimental respiratory anthrax in 74. Welkos S, Little S, Friedlander A, Fritz D, Fellows 94. American Hospital Formulary Service. AHFS Drug Macaca Mulatta. Br J Exp Pathol. 1963;44:416-426. P. The role of antibodies to Bacillus anthracis and an- Information. Bethesda, Md: American Society of Health 53. Grinberg LM, Abramova FA, Yampolskaya OV, thrax toxin components in inhibiting the early stages System Pharmacists; 1996. et al. Quantitative pathology of inhalational anthrax, of infection by anthrax spores. Microbiology. 2001; 95. Stepanov AV, Marinin LI, Pomerantsev AP, Starit- I. Mod Pathol. 2001;14:482-495. 147(pt 6):1677-1685. sin NA. Development of novel vaccines against an- 54. Dahlgren CM, Buchanan LM, Decker HM, et al. 75. Brachman P. Anthrax. In: Hoeprich PD, Jordan MC, thrax in man. J Biotechnol. 1966;44:155-160. Bacillus anthracis aerosols in goat hair processing mills. Ronald AR, eds. Infectious Diseases. Philadelphia, Pa: 96. Choe C, Bouhaouala S, Brook I, Elliott T, Knud- Am J Hyg. 1960;72:24-31. JB Lippincott; 1994:1003-1008. son G. In vitro development of resistance to ofloxa- 55. Walker JS, Lincoln RE, Klein F. Pathophysiologi- 76. Anthrax vaccine, military use in Persian Gulf re- cin and doxycycline in Bacillus anthracis Sterne. An- cal and biochemical changes in anthrax. Fed Proc. 1967; gion [press release]. Washington, DC: US Dept of De- timicrob Agents Chemother. 2000;44:1766. 26:1539-1544. fense; September 8, 1998. 97. Investigation of bioterrorism-related anthrax and 56. Franz DR, Jahrling PB, Friedlander A, et al. Clini- 77. Michigan Department of Public Health. Anthrax interim guidelines for exposure management and an- cal recognition and management of patients ex- Vaccine Adsorbed. Lansing: Michigan Dept of Public timicrobial therapy, October 2001. MMWR Morb Mor- posed to biological warfare agents. JAMA. 1997;278: Health; 1978. tal Wkly Rev. 2001;50:909-919. 399-411. 78. Pitt M, Little S, Ivins B, et al. In vitro correlate of 98. Stevens DL, Gibbons AE, Bergstron R, Winn V. 57. Vessal K, Yeganehdoust J, Dutz W, Kohout E. Ra- immunity in an animal model of inhalational anthrax. The Eagle effect revisited. J Infect Dis. 1988;158:23- diologic changes in inhalation anthrax. Clin Radiol. J Appl Microbiol. 1999;87:304. 28. 1975;26:471-474. 79. Ivins BE, Fellows P, Pitt ML, et al. Efficacy of stan- 99. Lightfoot NF, Scott RJ, Turnbull PC. Antimicro- 58. Albrink WS, Brooks SM, Biron RE, Kopel M. Hu- dard human anthrax vaccine against Bacillus anthra- bial susceptibility of Bacillus anthracis. Salisbury Med man inhalation anthrax. Am J Pathol. 1960;36:457- cis aerosol spore challenge in rhesus monkeys. Salis- Bull. 1990;68:95-98. 471. bury Med Bull. 1996;87:125-126. 100. Perkins WA. Public health implications of air- 59. Guillemin J. Anthrax: The Investigation of a Deadly 80. Fellows P, Linscott M, Ivins B, et al. Efficacy of a borne infection. Bacteriol Rev. 1961;25:347-355. Outbreak. Berkeley: University of California Press; human anthrax vaccine in guinea pigs, rabbits, and 101. Update: adverse events associated with an- 1999. rhesus macaques against challenge by Bacillus anthrax prophylaxis among postal employees: New Jer- 60. Brookmeyer R, Blades N, Hugh-Jones M, Hender- thracis isolates of diverse geographical origin. Vac- sey, New York City, and the District of Columbia met- son D. The statistical analysis of truncated data: ap- cine. 2001;20:635. ropolitan area, 2001. MMWR Morb Mortal Wkly Rep. plication to the Sverdlovsk anthrax outbreak. Biosta- 81. Brachman PS, Gold H, Plotkin SA, Fekety FR, Wer- 2001;50:1051-1054 tistics. 2001;2:233-247. rin M, Ingraham NR. Field evaluation of human an- 102. Interim recommendations for antimicrobial pro- 61. Jernigan J, Stephens D, Ashford D, Omenaca C, thrax vaccine. Am J Public Health. 1962;52:632- phylaxis for children and breastfeeding mothers and et al. Bioterrorism-related inhalation anthrax: the first 645. treatment of children with anthrax. MMWR Morb 10 cases reported in the United States. Emerg Infect 82. Surveillance for adverse events associated with an- Mortal Wkly Rep. 2001;50:1014-1016. Dis. 2001;7:933-944. thrax vaccination. MMWR Morb Mortal Wkly Rep. 103. American Public Health Association. Anthrax. In: 62. Barakat LA, Quentzel HL, Jernigan JA, et al. Fa- 2000;49:341-345. Benson AS, ed. Control of Communicable Diseases tal inhalational anthrax in a 94-year-old Connecticut 83. Pittman P, Gibbs P, Cannon T, Friedlander A. An- Manual. Washington, DC: American Public Health As- woman. JAMA.2002;287:863-868. thrax vaccine. Vaccine. 2001;20:972-978. sociation; 1995:18-22. 63. Freedman A, Afonja O, Chang M, et al. Cutane- 84. Committee to Assess the Safety and Efficacy of 104. Kournikakis B, Armour SJ, Boulet CA, Spence M, ous anthrax associated with microangiopathic hemo- the Anthrax Vaccine, Medical Follow-Up Agency. The Parsons B. Risk assessment of anthrax threat letters. lytic anemia and coagulopathy in a 7-month-old in- Anthrax Vaccine: Is It Safe? Does It Work? Washing- Defence Research Establishment Suffield. September fant. JAMA. 2002;287:869-874. ton, DC: Institute of Medicine, National Academy Press. 2001. Available at: http://www.dres.dnd.ca/Meetings 64. Mina B, Dym JP, Kuepper F, et al. Fatal inhala- March 2002. Available at: http://www.iom.edu/iom /FirstResponders/tr01-048_annex.pdf. tional anthrax with unknown source of exposure in a /iomhome.nsf/WFiles/Anthrax-8-pager1FINAL /$file 105. Use of onsite technologies for rapidly assessing 61-year-old woman in New York City. JAMA. 2002; /Anthrax-8-pager1FINAL.pdf environmental Bacillus anthracis contamination on sur- 287:858-862. 85. Wiesen AR, Littell CT. Relationship between faces in buildings. MMWR Morb Mortal Wkly Rep. 65. Bush LM, Abrams BH, Beall A, Johnson CC. In- prepregnancy anthrax vaccination and pregnancy and 2001;50:1087. dex case of fatal inhalational anthrax due to bioter- birth outcomes among US Army women. JAMA. 2002; 106. Chinn KS. Reaerosolization Hazard Assess- rorism in the United States. N Engl J Med. 2001;345: 287:1556-1560. ment for Biological Agent-Contaminated Hardstand 1607-1610. 86. Turnbull PC. Anthrax vaccines. Vaccine. 1991;9: Areas. Life Sciences Division, Dugway Proving Ground, 66. Mayer TA, Bersoff-Matcha S, Murphy C, et al. Clini- 533-539. Utah: US Dept of the Army; 1996. Publication DPG/ cal presentation of inhalational anthrax following bio- 87. Statement by the Department of Health and JCP-96/012. terrorism exposure. JAMA. 2001;286:2549-2553. Human Services regarding additional options for 107. Resnick IG, Martin DD, Larsen LD. Evaluation 67. Pile JC, Malone JD, Eitzen EM, Friedlander A. An- preventive treatment for those exposed to inhala- of Need for Detection of Surface Biological Agent Con- thrax as a potential biological warfare agent. Arch In- tional anthrax [news release]. Washington, DC: US tamination: Dugway Proving Ground, Life Sciences tern Med. 1998;158:429-434. Dept of Health and Human Services; December 18, Division, US Dept of the Army; 1990:1-35. Publica- 68. Kaye E, Kaye K. Fever and rash. In: Eugene Braun- 2001. tion DPG-FR-90-711. wald, Anthony S. Fauci, Kurt J. Isselbacher, et al, eds. 88. The Counter Bioterrorism Research Agenda of the 108. Manchee RJ, Stewart WD. The decontamina- Harrison’s Textbook of Medicine. New York, NY: National Institute of Allergy and Infectious Diseases tion of Gruinard Island. Chem Br. July 1988;690- McGraw-Hill; 2001. for CDC Category A Agents. Washington, DC: Na- 691. 69. Investigation of bioterrorism-related anthrax and tional Institute of Allergy and Infectious Diseases; Feb- 109. Hsu SS, Cost of anthrax cleanup on Hill to top interim guidelines for clinical evaluation of persons with ruary 2002. Available at: http://www.niaid.nih.gov $23 million, EPA says. Washington Post. March 7, possible anthrax. MMWR Morb Mortal Wkly Rep. /dmid/pdf/biotresearchagenda.pdf. 2002: A7. 2001;50:941-948. 89. Barnes J. Penicillin and B anthracis. J Pathol Bac- 110. Altman L. New tests confirm potency of an- 70. Carucci JA, McGovern TW, Norton SA, et al. Cu- teriol. 1947;194:113-125. thrax in Senate office building. New York Times. De- taneous anthrax management algorithm. J Am Acad 90. Lincoln R, Klein F, Walker J, et al. Successful treat- cember 11, 2001:B6. Dermatol. 2001; Nov 21. Available at: http://www ment of monkeys for septicemic anthrax. Antimicrob 111. Bioterrorism alleging use of anthrax and in- .harcourthealth.com/scripts/om.dll/serve Agents and Chemother—1964. Washington, DC: terim guidelines for management—United States, ?arttype=full&article=a121613. American Society for Microbiology; 1965:759-763. 1998. MMWR Morb Mortal Wkly Rep. 1999;48: 71. Penn C, Klotz S. Anthrax. In: Gorbach S, Bartlett 91. Odendaal MW, Peterson PM, de Vos V, Botha AD. 69-74. J, Blacklow N, eds. Infectious Diseases. Philadelphia, The antibiotic sensitivity patterns of Bacillus anthra- 112. Medical Response to Biological Warfare and Ter- Pa: WB Saunders Co; 1998:1575-1578. cis isolated from the Kruger National Park. Onderste- rorism. Gaithersburg, Md: US Army Medical Re- 72. Hail A, Rossi C, Ludwig G, Ivins B, Tammariello poort J Vet Res. 1991;58:17-19. search Institute of Infectious Diseases, Centers for Dis- R, Henchal E. Comparison of noninvasive sampling sites 92. Doganay M, Aydin N. Antimicrobial susceptibil- ease Control and Prevention, and US Food and Drug for early detection of Bacillus anthrax spores from rhe- ity of Bacillus anthracis. Scand J Infect Dis. 1991;23: Administration; 1998. sus monkeys after aerosol exposure. Mil Med. 1999; 333-335. 113. Mourez M, Kane R, Mogridge J, et al. Design- 164:833-837. 93. Kelly D, Chulay JD, Mikesell P, Friedlander A. Se- ing a polyvalent inhibitor of anthrax toxin. Nat Bio- 73. Pitt M, Little S, Ivins B, et al. In vitro correlate of rum concentrations of penicillin, doxycycline, and cipro- technol. 2001;19:958-961. immunity in a rabbit model of inhalational anthrax. floxacin during prolonged therapy in rhesus mon- 114. Friedlander AM. Microbiology: tackling an- Vaccine. 2001;19:4768-4773. keys. J Infect Dis. 1992;166:1184-1187. thrax. Nature. 2001;414:160-161.

Downloaded from jama.ama-assn.org by guest on April 25, 2012 LETTERS floxacin is only approved for “inhalational anthrax (postexpo- who weigh less than 20 kg in a mass casualty setting or for postexposure prophy- sure)” and is not approved by the FDA for the treatment of in- laxis is “80 mg/kg to be taken orally in 3 divided doses every 8 h.” halational anthrax.” At this time, then, clinicians have no options The footnote marked by a section mark (§) in Table 4 should read as follows: “Ac- cording to the CDC recommendations for the bioterrorist attacks in 2001, in which that have been approved by the FDA for the treatment of in- B anthracis was susceptible to penicillin, amoxicillin was a suitable alternative for halational anthrax. In the absence of FDA approval for any spe- postexposure prophylaxis in infants, children, and women who were pregnant or cific treatment for inhalational anthrax, clinicians must rely on who were breastfeeding. Amoxicillin was also a suitable alternative for completion of 60 days of antibiotic therapy for patients in these groups with cutaneous other sources of guidance regarding treatment recommenda- or inhalational anthrax whose clinical illness had resolved after treatment with a tions for this disease process. ciprofloxacin- or doxycycline-based regimen (14-21 days for inhalational or com- plicated cutaneous anthrax; 7-10 days for uncomplicated cutaneous anthrax). Such Dr Tice recommends consideration of a loading dose of doxy- patients required prolonged therapy because they were presumably exposed to cycline and ciprofloxacin in the treatment of inhalational an- aerosolized B anthracis.” thrax. We do not believe there is sufficient evidence to sup- In Table 5 on page 2247, the pediatric dosage of ciprofloxacin for treatment of port changing our recommendations to include these cutaneous anthrax infection should be “15 mg/kg per dose taken orally every 12 recommendations. Tetracyclines exhibit persistent time- h (maximum of 500 mg per dose).” Pediatric doxycycline dosage should be based on weight (ie, Ͼ or Յ 45 kg) not age. dependent bactericidal effects; the time above minimum inhibitory concentration (MIC) predicts therapeutic out- The most current versions of Tables 3, 4, and 5 are available online at: http://jama .ama-assn.org/cgi/content/full/287/17/2236. come.1 Fluoroquinolone antibiotics, on the other hand, exhibit persistent concentration-dependent killing with persis- The textual changes are as follows: On page 2245, the sentence “Penicillin, doxycycline, and ciprofloxacin are approved by the FDA for the treatment of inhala- tent effects; the ratio of the area under the curve to the MIC tional anthrax infection,56,89,90,94 and other antibiotics are under study” should read, predicts therapeutic outcome.2 These factors are more impor- “Penicillin and doxycycline are approved by the FDA for the treatment of anthrax.56,89,90,94 Although neither penicillin, doxycycline, nor ciprofloxacin are spe- tant clinically than steady state levels of these drugs. In addi- cifically approved by the FDA for the treatment of inhalational anthrax, these drugs tion, we are aware of no information that suggests improve- may be useful when given in combination with other antimicrobial drugs.” ment in clinical outcome using loading doses of these classes On page 2247, the sentence in the “Postexposure Prophylaxis” section of the text of antibiotics, and the therapeutic efficacy of the standard rec- that says, “There are no FDA-approved postexposure antibiotic regimens follow- ommended dosing regimen for these antibiotics (the same regi- ing exposure to a B anthracis aerosol” should read, “Ciprofloxacin, doxycycline, and penicillin G procaine are approved by the FDA for postexposure prophylaxis mens that appear in our consensus paper) have been demon- of inhalational anthrax.” strated in numerous clinical settings. Until more data regarding On page 2248 in the “Children” subsection, the sentence that begins “According improvement in clinical outcomes following the use of load- to CDC recommendations ...”should read “According to the CDC recommen- ing doses for these antimicrobials exists, we are reluctant to dations for the bioterrorist attacks in 2001, in which B anthracis was susceptible propose any changes in the guidelines. to penicillin, amoxicillin was a suitable alternative for postexposure prophylaxis in infants and children (Table 4).” In the “Pregnant Women” subsection, the sen- Thomas Inglesby, MD tence that begins, “According to the CDC recommendations ...”should read, “According to the CDC recommendations for the bioterrorist attacks in 2001, in Tara O’Toole, MD, MPH which B anthracis was susceptible to penicillin, amoxicillin was a suitable alterna- Johns Hopkins Center for Civilian Biodefense Strategies tive for postexposure prophylaxis in women who were pregnant or who were breast- Baltimore, Md feeding (Table 4).”

1. Craig W, Pharmacokinetic/pharmacodynamic parameters: rationale for anti- bacterial dosing of mice and men. Clin Infect Dis. 1998;26:1-10. 2. Forrest A, Nix DE, Ballow CH, Goss TF, Birmingham MC, Schentag JJ. Phar- macodynamics of intravenous ciprofloxacin in seriously ill patients. Antimicrob Agents CME ANNOUNCEMENT Chemother. 1993;37:1073-10781. CME Hiatus: July Through December 2002 CME from JAMA/Archives Journals will be sus- CORRECTION pended between July and December 2002. Beginning in early 2003, we will offer a new online CME pro- Incorrect Wording: Subsequent to the publication of the Consensus Statement entitled “Anthrax as a Biological Weapon, 2002: Updated Recommendations for gram that will provide many enhancements: Management,” published in the May 1, 2002, issue of THE JOURNAL (2002;287: 2236-2252), the authors wish to make available the following updates based on • Article-specific questions information from the US Food and Drug Administration and the Centers from Dis- • Hypertext links from questions to the relevant ease Control and Prevention (CDC). content In Table 3 on page 2246, the pediatric dosage of ciprofloxacin for “Initial IV [in- • Online CME questionnaire travenous] Therapy” for inhalational anthrax in the contained casualty setting should read, “10 mg/kg every 12 h (maximum of 400 mg per dose)” and subsequent • Printable CME certificates and ability to access total oral therapy under “Duration” should be “15 mg/kg per dose taken orally every CME credits 12 h (maximum of 500 mg per dose).” The doxycycline dosages for children should be based on weight (ie, Ͼ or Յ 45 kg) and not on age. We apologize for the interruption in CME and hope In Table 4 on page 2247, the pediatric dosage of ciprofloxacin for “Initial Oral that you will enjoy the improved online features that Therapy” of inhalational anthrax infection in the mass casualty setting or for post- will be available in early 2003. exposure prophylaxis should read, “15 mg/kg per dose taken orally every 12 h (maximum of 500 mg per dose).” The correct dosage of amoxicillin for children

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