Lesson 1, Volume 16—Anthrax for the Pulmonary Physician

By John G. Bartlett, MD; Tom Inglesby, MD; and Luciana Borio, MD

Objectives

1. Know when to suspect inhalational anthrax.
2. Know how to manage inhalational anthrax.
3. Understand the rationale for antibiotic selection and duration.
4. Understand the epidemiology of anthrax.
5. Understand the diagnostic evaluation.

Key words

anthrax; Bacillus anthracis; bioterrorism; ciprofloxacin; inhalational anthrax; wide mediastinum

Abbreviations

CDC = Centers for Disease Control and Prevention; FDA = Food and Drug Administration; LD₅₀ = median lethal dose

The potential for bioterrorism has been of growing concern for some time due to an understanding of the destructive power of biologic agents combined with the information indicating that a number of nations or states possessed offensive biological weapons programs or were seeking them. A number of biological agents have been identified as category A agents based on their potential to cause widespread illness and death, ease of dissemination or person-to-person transmission, potential for major public health impact, and requirement for special action for public health preparedness. The category A agents are Variola major, Bacillus anthracis, Yersinia pestis, Francisella tularensis, Clostridium botulinum, and the agents of viral hemorrhagic fever; smallpox and anthrax pose the greatest concern.^1,2 Whatever the attitudes were prior to October 3, 2001, the reality of this risk was brought into sharp focus by the epidemic of bioterrorism-related anthrax that was initially detected in a lethal case in Florida³ and subsequently reported in 21 additional suspected or confirmed cases. The great current concern is inhalational anthrax, which is a highly lethal disease with reported mortality rates exceeding 80% even with antibiotic treatment. Thus, inhalational anthrax becomes a component of the differential diagnosis of any patient with a serious, enigmatic respiratory disease and even enters into the differential diagnosis of patients with flu-like illnesses. The purpose of this report is to summarize current recommendations for the diagnosis and treatment of inhalational anthrax based on analyses of cases reported from October 3 to December 15, 2001.

Epidemiology

There are three forms of anthrax with clinical presentations based on the mechanism of microbial transmission: cutaneous due to skin contact, inhalational due to inhalation of the aerosol, and GI due to ingestion of contaminated meat (Table 1). The great majority of naturally occurring anthrax is the cutaneous form, reflecting the occupational hazard associated with contact with anthrax-infected animals or anthrax-contaminated animal products. In the United States, 224 cases of cutaneous anthrax were reported between 1944 and 1994.⁴ The GI form is very uncommon. The inhalational form is also uncommon as a naturally occurring disease; there were 18 cases reported in the United States in the 20th century, and the last reported case was in 1978.⁵ At present, it would now be fair to say that virtually any case of cutaneous anthrax or inhalational anthrax that is not clearly associated with contaminated animal products is probably secondary to a bioterrorist attack.

The epidemiology of the US outbreak in 2001 (Table 2) reflected a distribution pattern of contaminated mail that was largely directed to government officials in Washington, DC, and media representatives in New York City and Florida. Sixteen of the 18 confirmed cases reported through December 15, 2001, reflect exposure to the areas where tainted mail was either processed or received. There are two cases in which this association with mail is suspected but the mechanism of exposure has not been well established; these include one case in a hospital worker in New York City and a second in an elderly patient in Connecticut. Nevertheless, the mail system appears to be the common denominator in nearly all cases. With regard to occupations, postal workers account for 14 of the 22 confirmed and suspected cases, and media personnel account for six. In addition, B anthracis has been identified in the implicated buildings, mail processing systems, and in at least three envelopes. To date, there have been approximately 300 isolates of B anthracis from patients, envelopes, and environments associated with this outbreak (statement from Centers for Disease Control and Prevention [CDC] anthrax meeting; Atlanta, GA; November 2001). All isolates are identical by DNA analysis, indicating a common source and virtually negating any potential role of "background contamination" or naturally occurring disease.

An important factor in pathogenesis, morbidity, mortality, and distribution of B anthracis is particle size. Spores with a particle size of 1 to 5 mm are most efficiently inhaled. Spores of this size would also permit passage through the 10-mm pores of standard envelopes.

Pathogenesis

The mechanisms of disease acquisition for the three forms of anthrax are summarized above and in Table 1. The emphasis here will be on inhalational anthrax, which is most relevant to pulmonary physicians and is also the largest threat in terms of morbidity and mortality. This infection follows inhalation of anthrax spores in sufficient quantity to initiate disease. Again, the particle size is important, with 1 to 5 mm being most efficient for distribution to the alveoli.⁶ The organism is ingested by alveolar macrophages, which transport them via lymphatics to mediastinal lymph nodes, where germination to the toxin-producing vegetative forms may occur in the days or weeks following inhalation.^6-9 The inoculum size necessary to deliver a lethal dose to 50% of exposed persons, or the median lethal dose (LD₅₀), is estimated at 2,500 to 55,000 inhaled spores.¹⁰ This highly quoted figure is poorly understood. First, this is the number that is extrapolated from primate experiments with the assumption that there are minimal species differences between man and animals. Secondly, the LD₅₀ is frequently confused with the minimum infecting dose, although these figures are quite different. The minimum infecting dose is not known nor estimated.

The best information about the pathologic consequences of inhalational anthrax are from the observations in the Sverdlovsk outbreak in Russia in 1979,^11,12 in reports of woolsorters' disease,¹³ and in the experimental animal model.^9,14 The 1979 outbreak in Sverdlovsk represents an accidental release of anthrax spores at the site of an anthrax factory as a component of the Russian bioterrorism effort. The result was an estimated 68 to 105 cases of inhalational anthrax among residents within 2.5 km downwind from the site of release. The incubation period from the time of release to presentation ranged from 2 to 43 days, with a mean of 19.5 days. The median duration of symptoms prior to admission was 3 days, and the median duration of hospitalization to death was 1 day. There appeared to be five survivors and an estimated 50% of patients received antibiotics.

Clinical data are fragmentary because most of the clinical records were destroyed by the KGB, but slide material from 41 autopsies was retained and subjected to reanalysis by an investigating team that has subsequently reported the results.¹² This work showed that the predominant pathologic findings were necrotizing hemorrhagic mediastinitis, pleural effusions with an average of 1,700 mL, and hemorrhagic meningitis in 21 patients (51%). Pneumonitis was sometimes present, but was not a prominent feature. Microbiology studies showed high concentrations of B anthracis with up to 10⁸ cfu/mL in blood and in spinal fluid; however, no cultures were positive after 21 h of antibiotic treatment.¹² Studies in experimental animals tend to confirm these pathologic findings.^9,14 More or less unique was the finding of arteritis with arterial rupture, which presumably accounts for the bloody exudates and the extensive collections of edema fluid in the mediastinum and in skin with cutaneous anthrax that presumably reflects the effect of "edema toxin."¹²

The mortality rate for inhalational anthrax (Table 3) has been historically reported at 80 to 90%. This includes data from the Sverdlovsk epidemic as described above, deaths in 15 of 18 cases (84%) reported in the United States in the 20th century (all three survivors received b-lactam antibiotics),^15,16 and the 2001 outbreak, which caused death in 5 of 11 patients (45%). These data indicate a very high mortality rate after a very brief course of illness, making this one of the most rapidly lethal infections in previously healthy hosts. It also shows some substantial improvement in outcome with the more recent experience that may reflect improved management, but the numbers are small.

Clinical Features

Inhalational anthrax has been historically described as having two stages, but this distinction has not been evident in the 2001 cases. Early illness has been described as flu-like with malaise, fever, headache, sweats, GI symptoms (abdominal pain, nausea, vomiting, and diarrhea), a cough that is usually nonproductive, and myalgias. The pulse is often excessively rapid for the temperature. In contrast to most viral respiratory tract infections, such as influenza, respiratory syncytial virus, parainfluenza, and the common cold, there is no coryza. This first stage lasts an average of about 3 days and is followed by late-stage illness, which is characterized by shock, respiratory failure, and mental status changes.

Clinical and laboratory observations in the first 11 cases of inhalational anthrax in the 2001 US epidemic are summarized in Table 4.^17-21 This shows that the average duration of symptoms prior to hospitalization was 3.5 days (range, 1 to 7 days), which correlates well with the observations in Sverdlovsk (3 days).

With regard to laboratory tests, the CBC usually shows an elevated or high normal WBC (average, 9,500/mm³) with left shift on admission, but the average peak WBC in the first 10 cases was 26,000/mm³. Hemoconcentration is often present, and hematocrits commonly exceed 50%. The most important diagnostic tests are the blood culture, chest radiograph, and chest CT scan.

The admission chest radiograph was abnormal in all 11 US cases, and usually showed the characteristic wide mediastinum (7 cases) and pleural effusions (8 cases) with or without the radiographic appearance of pulmonary infiltrates (7 cases). Although there has been no evidence of pneumonia on postmortem examinations of the recent cases of fatal inhalational anthrax, the radiographic and clinical presentation may be indistinguishable from that of pneumonia.¹⁹ Thus, the presence of pulmonary infiltrates should not be used to rule out the diagnosis of inhalational anthrax. The wide mediastinum may not be easily recognized and is not as specific as commonly thought. Other causes of this finding include tuberculosis, tularemic pneumonia, sarcoid, histoplasmosis, lymphoma, silicosis, tumor, aneurysm, and alveolar proteinosis. Perhaps most helpful is a noncontrast CT scan, which shows a highly characteristic hyperdense lymphadenopathy involving the hilar and/or mediastinal lymph nodes with mediastinal edema. The pleural fluid is characteristically bloody. The diagnosis is usually established with blood culture, which has been uniformly positive when obtained before antibiotic treatment. In one reported case, the diagnosis was established before bacterial growth was detected by the automated blood culture system, when the Gram's stain of the peripheral blood buffy-coat revealed numerous large Gram-positive bacilli.¹⁹ However, it is exceedingly difficult to recover B anthracis in culture from any anatomical site after 24 h, and to our knowledge, has never been achieved when treatment has been given for 24 h. All of these observations are very compatible with the reports from Sverdlovsk.

Microbiology

B anthracis is easy to culture using standard microbiology media with clinical specimens obtained before antibiotic therapy. Clinical specimen processing is recommended for level A (biologic safety level [BSL]-2) laboratories, which represent most hospital laboratories. Work with spores, such as potentially contaminated powder, is best done in a BSL-3 laboratory because of the risk to the microbiologist; this is the recommended setting for culturing Mycobacterium tuberculosis. B anthracis measures 1 to 1.5 x 3 to 5 mm singly or in chains that may look like bamboo stalks. Spores are not usually seen on direct Gram's stain of specimens and usually appear in cultures only when nutrients are depleted. The usual time for growth is 6 to 24 h. For cases in which this information has been available in the current epidemic, blood cultures generally became positive within 6 to 18 h. A tentative microbiologic diagnosis can be made in most level A laboratories on the basis of recovering a Gram-positive bacillus that is spore-forming, nonmotile, nonhemolytic, penicillin-sensitive, and encapsulated. B anthracis has a characteristic morphology on Gram's stain and its presence should never be attributed to contamination; the bacteria forms very long chains when inoculated into blood culture medium and individual bacterial cells are larger than other more commonly isolated Gram-positive bacilli. This is "gumshoe microbiology" that is standard practice for virtually all trained microbiologists. Confirmation of the identification requires polymerase chain reaction, gamma phage lysis, direct fluorescent antibody, or immunohistochemistry tests using reagents that are generally available only in public health or level B laboratories or higher. Thus, referral to a public health laboratory is usually necessary for definitive identification. Specimens appropriate to culture cases include blood, spinal fluid, and pleural fluid or biopsy specimens from any involved site. Serologic tests consist of a screening test to detect antiprotective antigen IgG EIA, which shows a sensitivity of 98.6%, but a specificity of only 79%; specificity is improved by a companion inhibition EIA assay. With inhalational anthrax, there is generally a serologic response with a four-fold titer increase noted at 2 to 3 weeks. The serologic test is not commercially available, so reference to a public health laboratory is necessary.

Treatment

The two major treatments are antibiotic therapy and drainage of pleural effusions.

With respect to antibiotic selection, the Food and Drug Administration (FDA) has approved three drugs for anthrax on the basis of in vitro activity, clinical trials, and experimental studies in primates following inhalation challenge: penicillin, doxycycline, and ciprofloxacin.^6,9 For inhalational anthrax, one of the most important studies used a primate model facing an aerosol challenge of B anthracis spores at a dose of eight LD₅₀s.⁹ In vitro sensitivity tests on B anthracis strains obtained from diverse sources prior to the current epidemic showed good activity of most antibiotics, including penicillin, tetracycline, chloramphenicol, erythromycin, vancomycin, cefazolin, and aminoglycosides. Antibiotics that were inactive in vitro include trimethoprim-sulfamethoxazole and third-generation cephalosporins.^22-24 This is important to note because of their frequent use in empiric treatment of pneumonia. There are published reports by Russian scientists who produced strains of B anthracis resistant to tetracycline and penicillin.²⁵ In consideration of this and other data, the Working Group in Civilian Biodefense made recommendations to initiate therapy for inhalational anthrax with ciprofloxacin (or other fluoroquinolones) with subsequent decisions based on in vitro sensitivity tests of the epidemic strain.⁶ The current epidemic shows all three FDA-approved drugs to be highly active in vitro with minimum inhibitory concentrations of < 0.06 mg/mL. Other drugs that are active in vitro include rifampin, vancomycin, clindamycin, imipenem, and chloramphenicol. Drugs that are less active or inactive include macrolides (clarithromycin and azithromycin), ceftriaxone, sulfamethoxazole, and trimethoprim.⁶

Despite the high activity of penicillin in vitro, the B anthracis strain identified in the October 2001 cases produces an inducible b-lactamase, which could become problematic in patients with a high organism load as with inhalational anthrax. This accounts for the recommendation for ciprofloxacin or doxycycline as the basic component of treatment and prophylaxis for this epidemic strain (Table 5).²⁶ The specific recommendation of the CDC for inhalational anthrax is to administer ciprofloxacin or doxycycline, given IV with either drug given in combination with one or two additional antimicrobial agents. These recommendations apply to virtually all hosts, including adults, children, immunocompromised patients, and pregnant or lactating women.²⁶

The selection of one or two additional antibiotics is an empiric recommendation that is not based on studies in vitro or in vivo. The specific agents to add are also arbitrary, but there should be some specific considerations in this decision: If there is meningitis, drugs that work in pyogenic meningitis and cross the blood-brain barrier (eg, penicillin, ampicillin, chloramphenicol, or rifampin) should be considered; a concern about ciprofloxacin and doxycycline is that neither crosses the blood-brain barrier well and neither has extensive published experience in treatment of pyogenic meningitis. Clindamycin has a special potential advantage in its demonstrated property of shutting down protein synthesis, making it a preferred drug in streptococcal toxic shock syndrome, another toxin-mediated infection.

With regard to choosing between doxycycline and ciprofloxacin, these drugs are both highly active in vitro and work equally well in the animal model⁹; it should be noted that ciprofloxacin is always listed first in the CDC recommendations because these are provided in alphabetical order as a CDC standard for presentation of recommendations. The drugs are considered equally effective for both prophylaxis and therapy. There may be reservations about using either a tetracycline or fluoroquinolone in children, pregnant women, or lactating women because they are not FDA-approved in these groups. This recommendation applies to such patients on a risk-benefit basis and a modest experience that provides some assurance of safety.^27,28

With regard to the long course, the recommendation is to treat inhalational anthrax with combination therapy delivered IV until the patient is clinically stable for a total of 14 to 21 days, and then switch to oral therapy using ciprofloxacin 500 mg bid or doxycycline 100 mg bid to complete a total of 60 days of treatment. The 60-day course is the recommendation of the consensus group, Working Group on Civilian Biodefense,⁶ in 1999, and it deserves emphasis because of concerns about compliance and adequacy of duration with the need for careful posttherapy follow-up. Studies in experimental animals have shown that fatal disease can occur up to 98 days after inhalation challenge,²⁹ that viable spores may persist in mediastinal lymph nodes of primates 100 days after exposure,³⁰ and that 30 days of treatment is not uniformly effective in preventing posttreatment relapse with death.⁹ Because of the uncertainty regarding the adequacy of the 60-day course, the more recent recommendations include the options to stop at 60 days and observe, to continue antibiotics an additional 40 days (for a total of 100 days), or to give vaccine (three doses) plus a 40-day extension.

The Working Group on Civilian Biodefense has noted that the US-licensed anthrax vaccine would be a useful adjunct to antibiotic therapy as postexposure prophylaxis if vaccine were available. The rationale for immunizing those exposed to B anthracis spores is based on studies involving nonhuman primates⁹: None of the animals who received vaccine as an adjunct to a 30-day course of antibiotics developed inhalational anthrax when antibiotics were discontinued, and they were the only group to develop protective antibodies against B anthracis and did not develop disease when rechallenged with a second inhalation dose.

Public Health Issues

There are legal, epidemiologic, and medical reasons for close contact with public health officials, usually through local or state health agencies. The legal issue is that bioterrorism is a federal crime requiring contact with the Federal Bureau of Investigation and other law enforcement agencies, which is usually accomplished through public health contacts. The epidemiologic rationale concerns include establishing the sources of exposure, the extent of spread, and the level of possible environmental contamination. Prevention of inhalational anthrax is the primary goal of prophylaxis, and decisions are largely driven by such epidemiologic investigations. This epidemiologic information is a critical component in assessing the risk for this disease and should strongly help guide clinical decision-making. It is emphasized that 21 of the first 22 cases of confirmed or suspected anthrax in ther 2001 epidemic involved specific locations within buildings or pathways of implicated mail, and 16 of the 18 confirmed cases were employees of the postal system or media personnel. Such information becomes critical in the evaluation of flu-like illnesses where clinical features in the early stages may resemble a number of nonspecific respiratory illnesses, but this early phase also represents the optimal time for therapeutic intervention. Thus, close contact with infection control, infectious diseases specialists, and public health officials is essential, and all become important members of the health-care team when inhalational anthrax is established or suspected.

Conclusions

Inhalational anthrax represents a threat from bioterrorism and introduces new possibilities in the differential diagnosis of patients with flu-like illnesses or previously healthy patients with enigmatic life-threatening illnesses. The specific clues that are most helpful in the diagnosis of inhalational anthrax are the epidemiologic features that suggest exposure combined with any of the following findings associated with an acute febrile illness: chest radiograph showing a wide mediastinum, bloody pleural effusions, chest CT scan showing hyperdense adenopathy plus mediastinal edema, or cultures from any sterile body site (blood, pleural fluid, cerebrospinal fluid) showing Gram-positive bacilli. Treatment consists of ciprofloxacin or doxycycline combined with one or two additional antibiotics with a total duration of treatment of 60 to 100 days. Drainage of pleural effusions may be very beneficial in the management of respiratory failure. Close communication with public health officials is important for legal, epidemiologic, and medical management issues when this diagnosis is suspected or confirmed.

References

1. World Health Organization. Health aspects of chemical and biological weapons. Geneva, Switzerland: World Health Organization, 1970; 98
2. Keim P, Smith KL, Keys C, et al. Molecular investigation of the Aum Shinrikyo anthrax release in Kameido, Japan. J Clin Microbiol 2001; 39:4566–4567
3. Bush LM, Abrams BH, Beall A, et al. Index case of fatal inhalation anthrax due to bioterrorism in the United States. N Engl J Med 2001; 345:1607–1610 [epub ahead of print]
4. Summary of notifiable diseases, 1945-1944. MMWR Morb Mortal Wkly Rep 1994; 43:70–73
5. Brachman PS. Inhalation anthrax. Ann NY Acad Sci 1980; 353:83–93
6. Inglesby TV, Henderson DA, Bartlett JG, et al. Anthrax as a biological weapon: medical and public health management; Working Group on Civilian Biodefense. JAMA 1999; 281:1735–1745
7. Hatch TF. Distribution and deposition of inhaled particles in respiratory tract. Bacteriol Rev 1961; 25:237–240
8. Ross JM. The pathogenesis of anthrax following the administration of spores by the respiratory route. J Pathol Bacteriol 1957; 73:485–495
9. Friedlander AM, Welkos SL, Pitt ML, et al. Postexposure prophylaxis against experimental inhalation anthrax. J Infect Dis 1993; 167:1239–1243
10. Defense Intelligence Agency. Soviet biological warfare threat. Washington, DC: US Department of Defense, 1986; Publication No. DST-161OF-057-86
11. Abramova FA, Grinberg LM, Yampolskaya O, et al. Pathology of inhalational anthrax in 42 cases from the Sverdlovsk outbreak in 1979. Proc Natl Acad Sci USA 1993; 90:2291–2294
12. Grinberg LM, Abramova FA, Yampolskaya OV, et al. Quantitative pathology of inhalational anthrax: I. Quantitative microscopic findings. Mod Pathol 2001; 14:482–495
13. Dalldorf F, Kaufmann AF, Brachman PS. Woolsorters' disease: an experimental model. Arch Pathol 1971; 92:418–426
14. Gleiser CA, Berdjis CC, Harman HA, et al. Pathology of experimental respiratory anthrax in Macaca mulatta. Br J Exp Pathol 1963; 44:416
15. Brachman PS, Plotkin SA, Bumford FH, et al. An epidemic of inhalation anthrax: the first in the twentieth century. Am J Hyg 1960; 72:6–23
16. Brachman PS, Gold H, Plotkin SA, et al. Field evaluation of a human anthrax vaccine. Am J Public Health 1962; 52:632–645
17. Jernigan JA, Stephens DS, Ashford DA, et al. Bioterrorism-related inhalational anthrax: the first 10 cases reported in the United States. Available at http://www.cdc.gov/ncidod/EID/vol7no6/jernigan.htm. Accessed November 2001
18. Update: investigation of bioterrorism-related inhalational anthrax—Connecticut 2001. MMWR Morb Mortal Wkly Rep 2001; 50:1049–1051
19. Borio L, Frank D, Mani V, et al. Death due to bioterrorism-related inhalational anthrax: report of 2 patients. JAMA 2001; 286:2554–2559
20. Mayer TA, Bersoff-Matcha S, Murphy C, et al. Clinical presentation of inhalational anthrax following bioterrorism exposure: report of 2 surviving patients. JAMA 2001; 286:2549–2553
21. Considerations for distinguishing influenza-like illness from inhalational anthrax. MMWR Morb Mortal Wkly Rep 2001; 50:984–986
22. Lightfoot NF, Scott RJ, Turnbull PC. Antimicrobial susceptibility of Bacillus anthracis. Salisbury Med Bull Suppl 1990; 68:95S–98S
23. Odendaal MW, Peterson PM, de Vos V, et al. The antibiotic sensitivity patterns of Bacillus anthracis isolated from the Kruger National Park. Onderstepoort J Vet Res 1991; 58:17–19
24. Doganay M, Aydin N. Antimicrobial susceptibility of Bacillus anthracis. Scand J Infect Dis 1991; 23:333–335
25. Stepanov AV, Marinin LI, Pomerantsev AP, et al. Development of novel vaccines against anthrax in man. J Biotechnol 1996; 44:155–160
26. Update: investigation of bioterrorism-related anthrax and interim guidelines for exposure management and antimicrobial therapy, October 2001. MMWR Morb Mortal Wkly Rep 2001; 50:909–919
27. Schaad UB, abdus Salam M, Aujard Y, et al. Use of fluoroquinolones in pediatrics: consensus report of an International Society of Chemotherapy commission. Pediatr Infect Dis J 1995; 14:1–9
28. American Hospital Formulary Service. AHFS drug information. Bethesda, MD: American Society of Health System Pharmacists, 1996
29. Glassman HN. Industrial inhalation anthrax. Bacteriol Rev 1966; 30:657–659
30. Henderson DW, Peacock S, Belton FC. Observations on the prophylaxis of experimental pulmonary anthrax in the monkey. J Hyg 1956; 54:28–36

Copyright ©2002 American College of Chest Physicians

3300 Dundee Road • Northbrook, Illinois 60062-2348 • (847) 498-1400 • (800) 343-2227
Copyright ©1999-2008 American College of Chest Physicians. All Rights Reserved.
ACCP Privacy Policy | ACCP Web Site Terms of Use