On October 9, 2001, a letter containing anthrax spores was mailed from
New Jersey to Washington, DC. The letter was processed at a major postal facility
in Washington, DC, and opened in the Senate's Hart Office Building on October
15. Between October 19 and October 26, there were 5 cases of inhalational
anthrax among postal workers who were employed at that major facility or who
handled bulk mail originating from that facility. The cases of 2 postal workers
who died of inhalational anthrax are reported here. Both patients had nonspecific
prodromal illnesses. One patient developed predominantly gastrointestinal
symptoms, including nausea, vomiting, and abdominal pain. The other patient
had a "flulike" illness associated with myalgias and malaise. Both patients
ultimately developed dyspnea, retrosternal chest pressure, and respiratory
failure requiring mechanical ventilation. Leukocytosis and hemoconcentration
were noted in both cases prior to death. Both patients had evidence of mediastinitis
and extensive pulmonary infiltrates late in their course of illness. The durations
of illness were 7 days and 5 days from onset of symptoms to death; both patients
died within 24 hours of hospitalization. Without a clinician's high index
of suspicion, the diagnosis of inhalational anthrax is difficult during nonspecific
prodromal illness. Clinicians have an urgent need for prompt communication
of vital epidemiologic information that could focus their diagnostic evaluation.
Rapid diagnostic assays to distinguish more common infectious processes from
agents of bioterrorism also could improve management strategies.
On October 9, 2001, a letter containing anthrax spores was mailed to
a US senator's office in Washington, DC.1 That
letter was processed at the Brentwood postal facility, one of the largest
in the Washington, DC, metropolitan area, and eventually opened by the senator's
staff 6 days later. As of October 30, 2001, 5 postal workers who worked at
or handled bulk mail from that facility had been hospitalized with confirmed
inhalational anthrax. Two of them died shortly after admission to area hospitals.
These 2 fatal cases are reported here to educate clinicians about the clinical
presentation and course of illness of inhalational anthrax following exposure
to anthrax spores.
On October 16 (day 1), a 47-year-old male postal worker who worked in
the mail sorting area of the Brentwood facility developed nausea, abdominal
pain, and "flulike" symptoms. He attributed his symptoms to "food poisoning"
and continued to work despite ongoing symptoms. On October 20 (day 5), while
in church, he had a brief, self-limited syncopal episode. By the time paramedics
arrived, he felt better. He went home, did not eat, and immediately went to
bed, unlike his usual routine. He reported to work that evening and on October
21 (day 6), at 2 AM, he developed worsening nausea, vomiting, abdominal pain,
and profuse sweating and drove to an emergency department. On arrival, he
also complained of lightheadedness and diaphoresis and denied dyspnea and
chest pain. He had no significant medical history. His only chronic medical
problem was asthma, and salmeterol was his only medication.
The patient's temperature was 36.1°C, blood pressure was 82/59 mm
Hg, pulse was 95/min, respirations were 18/min, and oxygen saturation was
99% in room air. Physical examination was unremarkable. Laboratory data revealed
mild leukocytosis and hemoconcentration. Chest radiograph (Figure 1) showed a subtle and ill-defined area of increased density
in the right subhilar region; follow-up with computed tomography (CT) was
suggested. After treatment with intravenous fluids, promethazine, and famotidine,
his symptoms resolved. At 4 AM, vital signs were as follows: temperature,
36.6°C; blood pressure, 109/68 mm Hg; pulse, 86/min; and respirations,
20/min. He was discharged to home at approximately 5 AM with a presumptive
diagnosis of gastroenteritis and instructions to see his primary care physician
the following day.
Throughout the day of October 21 (day 6), the patient's gastrointestinal
symptoms progressed and he developed dyspnea and diaphoresis. On October 22
(day 7) at 4:45 AM, his wife found him slumped in the bathroom. He was taken
to a hospital by ambulance. On arrival, he reported nausea, vomiting, and
lightheadedness and denied dyspnea or chest pain but was ill-appearing and
in respiratory distress. Temperature was 35.5°C, blood pressure was 76/46
mm Hg, pulse was 152/min, respirations were 28/min, and oxygen saturation
was 96% in room air. Physical examination revealed regular tachycardia, decreased
breath sounds at the lung bases, diffuse wheezes and rhonchi, mild abdominal
tenderness, and mottled and cool skin.
A working diagnosis of inhalational anthrax was made because of reports
in the media that 2 postal workers had been hospitalized in the Washington,
DC, metropolitan area with inhalational anthrax.2
Laboratory data (Table 1) showed
leukocytosis, hemoconcentration, azotemia, and an elevated amylase level.
Chest radiographs (Figure 1) revealed
progressive pulmonary infiltrates and, compared with the radiograph of October
21, new pleural effusions and a widened mediastinum. A Gram stain of the admission
peripheral blood buffy coat revealed numerous large, gram-positive bacilli
(Figure 2). At that time, a Gram
stain was requested on blood that had been incubated for 1 hour in BD Bactec
(BD, Sparks, Md) culture media. The long chains of gram-positive bacilli (Figure 3A) and the "jointed bamboo-rod" appearance
(Figure 3B) were considered typical
of Bacillus anthracis. Anthrax was confirmed by the
Centers for Disease Control and Prevention (CDC) in the following days3 (the CDC confirms B anthracis
in culture when 2 tests are positive: polymerase chain reaction and gamma
Despite prompt initiation of therapy with one 500-mg dose of parenteral
levofloxacin; penicillin G, 3 million U every 4 hours; one 2-g dose of ceftriaxone;
and one 600-mg dose of rifampin, the patient developed progressive hypoxemic
respiratory failure and marked hypotension. Mechanical ventilation and vasopressor
therapy was instituted. Gram stain of an endotracheal aspirate showed few
white blood cells and rare large gram-positive bacilli. A cotton swab from
the nose was obtained for culture at admission, plated on sheep blood agar,
and read at 24 hours, and was negative for B anthracis.
The patient developed abdominal distention; a CT scan with intravenous contrast
revealed moderate ascites and small-bowel intramural pneumatosis (Figure 4A). A chest CT image revealed parenchymal
infiltrates, pleural effusions, and a widened mediastinum (Figure 4B). On October 22 (day 7), 5 hours after admission, the
patient developed ventricular tachycardia followed by refractory bradycardia
and asystole and died despite attempts at cardiac resuscitation.
At autopsy, there were multifocal hemorrhagic foci in the mediastinum
with extension along hilar and parenchymal bronchi and blood vessels. There
was hilar lymphadenopathy (2.5 cm) and hemorrhagic necrotizing hilar lymphadenitis;
microscopic examination showed effacement of the nodal architecture, acute
hemorrhagic necrosis, abundant karyorrhectic debris, and infiltration of polymorphonuclear
leukocytes. Brown and Brenn stain of the hilar soft tissue and lymph nodes
showed abundant gram-positive bacilli. There was no gross or microscopic evidence
of pneumonia but there were large serous pleural effusions bilaterally (right,
250 cc; left, 500 cc).
There was a large amount of ascites (2500 cc). The small bowel was edematous
and multifocal mesenteric soft tissue hemorrhage was present. A portion of
the ileum showed hemorrhage and necrotizing infection without visible mucosal
ulceration. Microscopic examination of the involved section of the ileum showed
serosal and submucosal acute and chronic inflammation extending into the lamina
propria. Brown and Brenn stain of the ileum showed abundant gram-positive
bacilli. The mesenteric lymph nodes, terminal ileum, and large bowel were
grossly unremarkable. Brain and cerebrospinal fluid were not examined. The
cause of death was certified as inhalational anthrax. The manner of death
was certified as homicide.
On October 17 (day 1), a 55-year-old male postal worker who worked as
a distribution clerk in the mail sorting area of the Brentwood facility and
who had hypertension, diabetes mellitus, and remote history of sarcoidosis
sought outpatient medical attention for myalgias, weakness, and fever of 1
day's duration. He was diagnosed as having a viral syndrome and sent home.
His symptoms persisted. On October 21 (day 5) at 6 AM, he presented to an
emergency department (of a different hospital than that of patient 1) complaining
of severe dyspnea, myalgias, weakness, retrosternal chest pressure, high fever,
chills, nausea, vomiting, and a cough productive of scant greenish sputum.
On initial physical examination, the patient was alert, well-appearing,
and in no acute respiratory distress. Temperature was 38.9°C, blood pressure
was 119/73 mm Hg, pulse was 150/min, respirations were 20/min, and oxygen
saturation was 94% in room air. The remainder of the examination was notable
for rales and decreased breath sounds at the right lung base and an irregular
tachycardia. Electrocardiogram showed atrial fibrillation.
Laboratory data (see Table 1)
revealed marked leukocytosis, profound hemoconcentration, metabolic acidosis,
hyperglycemia, azotemia, and hypoxemia. The admission chest radiograph is
shown in Figure 5A.
Because the health care staff were aware through media reports that
2 postal workers had been hospitalized in the local metropolitan area with
inhalational anthrax, the patient was admitted with a diagnosis of suspected
inhalational anthrax. He was given prompt treatment with one 500-mg dose of
parenteral levofloxacin. Within 13 hours of admission, he developed marked
hypotension requiring institution of vasopressors and progressive hypoxemic
respiratory failure requiring endotracheal intubation. Shortly thereafter,
he developed bradycardia and asystole and died despite attempts at cardiac
resuscitation. The case was reported to the medical examiner.
A Gram stain of blood incubated in BD Bactec culture media (Figure 5B) was recognized to have gram-positive
rods the next morning, 24 hours after the patient's admission. The CDC subsequently
confirmed the organism to be B anthracis.3
At autopsy, there was marked acute hemorrhagic mediastinitis, hemorrhagic
necrotizing lymphadenitis of the hilar and mediastinal lymph nodes (3-5 cm
in diameter), and bilateral serosanguinous pleural effusion (right, 1300 cm3; left, 700 cm3). Bilateral acute subpleural hemorrhage
was present at the hilum of both lungs with extension along the adventitia
of the intraparenchymal branches of the pulmonary arteries. Microscopic examination
of the lymph nodes revealed effacement of the nodal architecture, acute hemorrhagic
necrosis, infiltration by polymorphonuclear leukocytes, and numerous bacilli.
There was no evidence of pneumonia or meningitis. The cause of death was certified
as inhalational anthrax. The manner of death was certified as homicide.
Inhalational anthrax presents with nonspecific symptoms that cannot
be distinguished from many more common diseases based on early clinical manifestations
or routine laboratory tests. Both patients in this report sought medical care
for apparently mild, nonspecific illnesses and were sent home. Only after
the news media reported cases of inhalational anthrax involving 2 postal workers
from the local mail facility did these patients' physicians consider the possibility
that they could have inhalational anthrax. At that point, the patients had
been ill for 7 days (patient 1) and 5 days (patient 2). The courses of their
illness are similar to that reported in an outbreak of anthrax that occurred
in Sverdlovsk in 1979.4 Despite aggressive
medical therapy, both patients developed rapidly progressive disease and died.
These patients received antimicrobial therapy at the discretion of their
physicians, before the CDC released formal guidelines on October 26, 2001.3 These guidelines recommend combination therapy for
inhalational anthrax and complicated cutaneous anthrax. Ciprofloxacin or doxycycline
are recommended in conjunction with another active antimicrobial drug, such
as rifampin or clindamycin. Even though all isolates tested were susceptible
to penicillin, β-lactamases were identified in these isolates, and penicillin
monotherapy for treatment of systemic infection is not recommended. Susceptibility
testing also revealed intermediate sensitivity to ceftriaxone and presence
of a cephalosporinase. Cephalosporins, therefore, are not indicated for the
treatment of B anthracis infection.
These 2 cases emphasize that in the event of serious outbreaks of infectious
diseases, such as those that occurred following the handling of anthrax-containing
letters at the Brentwood postal facility, rapid communication of epidemiologic
data to front-line medical care providers (especially emergency physicians
and primary care clinicians) is essential so they may initiate appropriate
diagnostic procedures and therapies. Efforts should be made to enhance communications
systems between public health agencies and clinicians.5
A presumptive diagnosis of B anthracis can
be made readily by most microbiology laboratories. The organism grows readily,
can be safely evaluated in a biosafety level 2 facility, has characteristic
Gram stain morphology, and should be suspected anytime a nonmotile, nonhemolytic
gram-positive rod is identified in the setting of a compatible clinical syndrome.
Communicating presumptive diagnostic information throughout the public health
and clinical community must be a priority. Nasal swab cultures cannot reliably
rule out exposure to B anthracis, as stated by the
CDC3 and exemplified by patient 1, who, despite
severe clinical disease, had a negative nasal swab culture. Whether the culture
would have been positive if a Dacron swab had been used, if a more rigorous
culturing technique was used, or if a deeper site had been sampled is unknown.
These 2 case reports also emphasize the importance of having microbiology
laboratory capacity to more conclusively identify anthrax and other diseases
that may be caused by bioterrorism in the clinical settings where these diseases
will present. Rapid diagnostic tests to distinguish early anthrax infection
from other diseases of similar clinical appearance should be a high priority
on a national research agenda to respond to the threat of bioterrorism. There
are distressingly few health care facilities that can provide comprehensive
diagnostic services and expert consultative support.
The clinical presentations of these 2 patients provide key information
about the clinical presentations and clinical courses of these homicides.
Clinicians throughout the country in the full range of health science disciplines
must rapidly increase their knowledge about agents of bioterrorism and work
collaboratively with governmental, academic, and private organizations to
ensure that their communities have the resources and expertise necessary to
manage these assaults expeditiously and efficiently.
Register and get free email Table of Contents alerts, saved searches, PowerPoint downloads, CME quizzes, and more
Subscribe for full-text access to content from 1998 forward and a host of useful features
Activate your current subscription (AMA members and current subscribers)
Purchase Online Access to this article for 24 hours
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 104
Customize your page view by dragging & repositioning the boxes below.
and access these and other features:
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.