Bioterrorism-Related Inhalational Anthrax: The First 10 Cases Reported in the United States

*Centers for Disease Control and Prevention, Atlanta, Georgia, USA; †Emory University School of Medicine, Atlanta, Georgia, USA; ‡Cedars Medical Center, Miami, Florida, USA; §Virtua Health, Mount Holly, New Jersey, USA; ¶Winchester Medical Center, Winchester, Virginia, USA; #Lenox Hill Hospital, New York City, New York, USA; and **Palm Beach County Department of Public Health, West Palm Beach, Florida, USA


Emerging Infectious Diseases. 2001;7(5) 

In This Article


Before this outbreak of bioterrorism-related anthrax, only 18 cases of inhalational anthrax had been reported in the United States in the last century[1,4,9,10,11,12,13,14,15,16,17,18,19,20,21]. The most recent case was in 1976[4]. Most cases were related to exposure to animal products, primarily in textile mills processing goat hair, goatskins, or wool[1]. The clinical characteristics of the 10 cases due to bioterrorism described here share similarities with previously reported cases but have important differences.

Inhalational anthrax has been described as a biphasic clinical illness characterized by a 1- to 4-day initial phase of malaise, fatigue, fever, myalgias, and nonproductive cough, followed by a fulminant phase of respiratory distress, cyanosis, and diaphoresis. Death follows the onset of the fulminant phase in 1 to 2 days[2,22]. The symptoms of the initial phase of inhalational anthrax in the 10 cases caused by bioterrorism (Table 1) were similar to those of the 18 occupationally related cases previously described[1]; however, profound, often drenching sweating was a prominent feature of the current patients and was not emphasized in earlier reports.

Nausea and vomiting were also frequent symptoms of the initial phase in the current bioterrorism-related inhalational cases, suggesting early involvement of the gastrointestinal tract. Gastrointestinal lesions were observed in 39 of 42 total cases of fatal inhalational anthrax associated with the Sverdlovsk outbreak[23]. Most of the lesions in these fatal cases appeared to represent hematogenous spread of B. anthracis to the submucosa of the gastrointestinal tract and did not involve Peyer's patches or, in most cases, mesenteric lymph nodes.

The nondistinctive nature of the initial phase of inhalational anthrax presents a diagnostic challenge. The chest X-ray appeared to be a sensitive indicator of disease in patients with bioterrorismassociated inhalational anthrax, as none of the 10 patients had an initially normal chest X-ray. Multiple abnormalities were noted on initial chest X-ray, including mediastinal widening, paratracheal fullness, hilar fullness, pleural effusions,and parynchemal infiltrates. Two patients had chest X-rays that were interpreted initially as normal, but abnormalities (mediastinal widening with small pleural effusion in one case and perihilar mass versus infiltrate in the other) were detected when the X-rays were reviewed by a radiologist. Pulmonary infiltrates or effusion were initially seen in two cases without evidence of mediastinal widening. Chest CT was helpful in further characterizing abnormalities in the lungs and mediastinum and was more sensitive than chest X-ray in revealing mediastinal lymphadenopathy.

The total WBC count was usually normal or only slightly elevated at the time of initial visit to a health-care provider for patients who sought medical care in the initial phase. However, an elevation in the proportion of neutrophils or band forms was frequently noted and was an early diagnostic clue. During the course of illness, WBC counts increased, sometimes markedly, in most patients. Previous case reports have noted a brief period of improvement between the initial and fulminant phases[1], but this phenomenon was not observed in the current cases.

Blood cultures grew B. anthracis even in the initial phase of the illness in all patients who had not received prior antibiotic therapy. Animal models suggest that primates with inhalational anthrax become bacteremic early in the course of the illness before the fulminant clinical phase develops[24]. In contrast, blood cultures rapidly became sterile after initiation of antibiotic therapy, suggesting that prior antibiotic treatment may substantially decrease the sensitivity of blood cultures as a diagnostic test.

The diagnosis of anthrax was established in three patients without growth of B. anthracis from clinical specimens. In all three of these cases, proper cultures were obtained only after initiation of antibiotic therapy. The diagnosis in these patients was established by a history of exposure or occupational and environmental risk with a clinically compatible syndrome, by the identification of B. anthracis in pleural fluid, pleural biopsy, or transbronchial biopsy specimens by immunohistochemical staining with B. anthracis-specific cell wall and capsular antibodies, or by identifying B. anthracis DNA by PCR on pleural fluid or blood. Serologic data from ELISA available for one patient with inhalational anthrax, also demonstrated a >4-fold increase in levels of serum antibody (IgG) to the PA component of anthrax toxins.

The survival of patients with inhalational anthrax in this series (60%) is higher than previously reported (<15%)[1,3]. All patients received combination antimicrobial therapy with more than one agent active against B. anthracis. The apparent improvement in survival compared with previous cases suggests that the antibiotic combinations used in these patients may have therapeutic advantage over previous regimens. Limited data on treatment of the survivors suggests that early treatment with a fluoroquinolone and at least one other active drug[7] may improve survival. B. anthracis isolates produce a cephalosporinase[7] that inhibits the antibacterial activity of cephalosporins such as ceftriaxone, and cephalosporins should not be used for treatment. Additional antimicrobial susceptibilities of the isolates associated with this outbreak have been previously published[7]. Other explanations for the improved survival rate include earlier recognition and initiation of treatment, better supportive care, differences in the pathogenesis of bioterrorism-related anthrax, differences in susceptibility of the hosts, or a combination of the above.

Pleural effusions were a remarkably consistent clinical feature of inhalational anthrax in this series, occurring in all patients. The pleural effusions were often small on presentation, but in the surviving patients effusions were characterized by progressive enlargement and persistence. Drainage of the pleural cavity was required in seven patients. The characteristics of the pleural fluid in all patients were similar: hemorrhagic, with a high protein concentration and relatively few WBCs. Immunohistochemistry demonstrated large quantities of B. anthracis capsule and cell wall antigens in pleural tissue or pleural fluid cell blocks.

Case 2 was remarkable in that X-ray findings were dominated by large and progressive pulmonary infiltrates, not by mediastinal widening. In fact, there was no mediastinal adenopathy noted on chest CT. Transbronchial biopsies of the patient in Case 2 showed B. anthracis-specific capsular and cell wall antigens in the lung parenchyma. Previous reports have noted bronchopneumonic and pulmonary hemorrhagic infarcts in patients with inhalational anthrax. Abramova et al.[23] described focal hemorrhage and necrotizing anthrax pneumonia in 11 of 42 patients who died with inhalational anthrax. Characteristics of the spore-containing aerosol or individual host factors may influence pathogenesis, so that pulmonary infiltrates may be a more prominent manifestation in some patients with bioterrorism-related inhalational anthrax.

Three patients had supraventricular arrythmias, and three had pericardial effusion on chest CT, one of which may have progressed to tampanade. Pericardial effusions may reflect toxin-related local edema or the local inflammatory process, hemorrhagic necrosis and infarct extending into the pericardium from the mediastinum or may be a direct result of hematogenous spread of B. anthracis.

Anthrax meningitis is a complication of inhalational anthrax and is characteristically hemorrhagic. Pathologic findings are a hemorrhagic leptomeningitis with edema and inflammatory infiltrates[2,23]. Fifty-five percent of patients at Sverdlovsk who died of inhalational anthrax had evidence of meningeal involvement at autopsy. Cerebrospinal fluid was examined in only two patients in our series, and meningitis was documented in one. Three other patients had a history of intermittent confusion, but there were no meningeal signs and lumbar punctures were not performed.

No predominant underlying diseases or conditions were noted in the patients with bioterrorismrelated inhalational anthrax. One patient (Case 5) had a remote history of sarcoidosis but had been free of clinical illness or treatment associated with this diagnosis for the past 25 years. One other case of inhalational anthrax associated with underlying sarcoidosis has been reported[12]. The patient in case 5 also had diabetes mellitus, a condition present in one previously reported case[12]. Smoking did not appear to be a risk factor for inhalational anthrax.

In summary, we describe the clinical presentation of the first 10 cases of bioterrorism-related inhalational anthrax in the United States. The clinical presentation in these patients was variable and often resembled a viral respiratory illness, but the interpretation of the initial symptoms in the context of a possible exposure to B. anthracis often led to an early diagnosis. In contrast to previous reports indicating a death rate >85%[1,3], our series suggests that survival may be markedly improved by combination antimicrobial therapy begun during the initial phase of the illness and by aggressive supportive care (e.g., drainage of pleural effusions). Newer methods of detection such as polymerase chain reaction, immunohistochemistry, and sensitive serologic tests are important adjunctive diagnostic modalities that aid in the diagnosis of B. anthracis infections. Further studies are needed to better define optimal antimicrobial regimens for this disease, to explore the role of adjunctive therapies (e.g., immunoglobulin antitoxin, corticosteroids, and other toxin inhibitors) in the management of inhalational anthrax, and to better understand the pathogenesis of inhalational anthrax associated with intentional release.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.