Shiga toxin-producing Escherichia coli O157:H7 outbreak associated with school field trips at a farm animal exhibit-Tennessee, September-October 2023.

AIMS: In October 2023, the Tennessee Department of Health identified an outbreak of Shiga toxin-producing Escherichia coli (STEC) O157:H7 infections among elementary school students who attended school field trips to the same farm animal exhibit. Our aim was to determine STEC source and prevent additional illnesses by initiating epidemiologic, laboratory and environmental investigations. METHODS AND RESULTS: We identified cases using laboratory-based surveillance and by surveying caregivers of children who attended the exhibit. Probable cases were defined as illness with abdominal cramps or diarrhoea after attendance; confirmed cases were laboratory-confirmed STEC infection in an attendee or household contact. A site visit was conducted, and event organizers were interviewed. Human stool, animal faeces and environmental samples were tested for STEC O157:H7 by real-time polymerase chain reaction (PCR), culture and whole-genome sequencing (WGS). Approximately 2300 elementary school students attended the animal exhibit during 2 days. Field trip activities included contact with different farm animal species, drinking pasteurized milk outside animal enclosures and eating lunch in a separate building onsite. We received survey responses from 399 caregivers for 443 (19%) animal exhibit attendees. We identified 9 confirmed and 55 probable cases with illness onset dates during 26 September to 12 October. Seven children aged 1-7 years were hospitalized. Four children aged 1-6 years experienced haemolytic uraemic syndrome; none died. Laboratory testing identified STEC O157:H7 by culture from eight human stool samples with 0-1 allele difference by WGS. Three environmental samples had Shiga toxin (stx 2) genes detected by PCR, but no STEC isolates were recovered by culture. CONCLUSIONS: This is the largest reported STEC O157:H7 outbreak associated with an animal exhibit in Tennessee. We identified opportunities for educating school staff, event organizers and families about zoonotic disease risks associated with animal contact and published prevention measures.

Prevalence of gabapentin in drug overdose postmortem toxicology testing results.

BACKGROUND: The goal of this study was to establish and compare baseline data on the prevalence of gabapentin identified through postmortem toxicology testing among drug overdose decedents in several geographically diverse states/jurisdictions with differing levels of drug overdose fatality burdens in 2015. METHODS: Death certificates and postmortem toxicology result reports from five U.S. jurisdictions were used to identify residents who died from drug overdoses in year 2015 and to calculate prevalence rates of gabapentin in postmortem toxicology by jurisdiction. RESULTS: On average, 22% of all drug overdose decedents in our study tested positive for gabapentin. The percentage of gabapentin-positive overdose deaths varied significantly among jurisdictions: 4% in Northeast Tennessee, 7% in Maricopa County, 15% in West Virginia, 20% in North Carolina, and 41% in Kentucky (p < 0.0001). Among the drug overdose decedents who tested positive for opioids (including heroin), 26% also tested positive for gabapentin, with significant variation among states/jurisdictions (p < 0.0001). There was a significant difference in the gender distribution among drug overdose decedents who tested positive for gabapentin (46% male) vs. those who tested negative for gabapentin (65% male) (p < 0.0001). In Kentucky, gabapentin was listed as a contributing drug on the death certificate in 40% of the overdose deaths with gabapentin-positive toxicology; in North Carolina this percentage was 57%. CONCLUSIONS: Routine gabapentin postmortem testing and linking of death certificate, medical examiner, coroner, toxicology, and prescription history data will provide more reliable information on the extent of gabapentin misuse, diversion, and implications for clinical care.

Comparative effectiveness of antibiotic treatment strategies for pediatric skin and soft-tissue infections.

OBJECTIVE: To compare the effectiveness of clindamycin, trimethoprim-sulfamethoxazole, and ß-lactams for the treatment of pediatric skin and soft-tissue infections (SSTIs). METHODS: A retrospective cohort of children 0 to 17 years of age who were enrolled in Tennessee Medicaid, experienced an incident SSTI between 2004 and 2007, and received treatment with clindamycin (reference), trimethoprim-sulfamethoxazole, or a ß-lactam was created. Outcomes included treatment failure and recurrence, defined as an SSTI within 14 days and between 15 and 365 days after the incident SSTI, respectively. Adjusted models stratified according to drainage status were used to estimate the risk of treatment failure and time to recurrence. RESULTS: Among the 6407 children who underwent drainage, there were 568 treatment failures (8.9%) and 994 recurrences (22.8%). The adjusted odds ratios for treatment failure were 1.92 (95% confidence interval [CI]: 1.49-2.47) for trimethoprim-sulfamethoxazole and 2.23 (95% CI: 1.71-2.90) for ß-lactams. The adjusted hazard ratios for recurrence were 1.26 (95% CI: 1.06-1.49) for trimethoprim-sulfamethoxazole and 1.42 (95% CI: 1.19-1.69) for ß-lactams. Among the 41 094 children without a drainage procedure, there were 2435 treatment failures (5.9%) and 5436 recurrences (18.2%). The adjusted odds ratios for treatment failure were 1.67 (95% CI: 1.44-1.95) for trimethoprim-sulfamethoxazole and 1.22 (95% CI: 1.06-1.41) for ß-lactams; the adjusted hazard ratios for recurrence were 1.30 (95% CI: 1.18-1.44) for trimethoprim-sulfamethoxazole and 1.08 (95% CI: 0.99-1.18) for ß-lactams. CONCLUSIONS: Compared with clindamycin, use of trimethoprim-sulfamethoxazole or ß-lactams was associated with increased risks of treatment failure and recurrence. Associations were stronger for those with a drainage procedure.

Children with asthma hospitalized with seasonal or pandemic influenza, 2003-2009.

OBJECTIVE: To describe the characteristics and clinical courses of asthmatic children hospitalized with seasonal or 2009 pandemic H1N1 influenza and compare complications by influenza type. METHODS: During the 2003-2009 influenza seasons and the 2009 pandemic, we conducted surveillance of 5.3 million children aged 17 years or younger for hospitalization with laboratory-confirmed influenza and identified those with asthma (defined as those aged 2-17 years with a history of asthma in their medical record or a discharge code for acute asthma exacerbation or status asthmaticus). We collected data from medical records on medical history and clinical course; data on asthma severity and control were not routinely collected. RESULTS: During the 2003-2009 influenza seasons, 701 (32%) of 2165 children hospitalized with influenza had asthma; during the 2009 pandemic, 733 (44%) of 1660 children had asthma. The median age of the asthmatic children was 7 years, and 73% had no additional medical conditions. Compared with asthmatic children with seasonal influenza, a higher proportion with 2009 pandemic H1N1 influenza required intensive care (16% vs 22%; P=.01) and were diagnosed with pneumonia (40% vs 46%; P=.04), whereas equal proportions had respiratory failure (5% vs 5%; P=.8) and died (1% vs 1%; P=.4). More asthmatic children with influenza A (seasonal or pandemic) had diagnoses of asthma exacerbations compared with those with influenza B (51% vs 29%; P<.01). CONCLUSIONS: The majority of asthmatic children hospitalized with influenza have no additional medical conditions. Complications such as pneumonia and need for intensive care occur in a substantial proportion, highlighting the importance of influenza prevention through vaccination among asthmatic children.

Seasonal and 2009 pandemic influenza A (H1N1) virus infection during pregnancy: a population-based study of hospitalized cases.

We sought to describe characteristics of hospitalized reproductive-aged (15-44 years) women with seasonal (2005/2006 through 2008/2009) and 2009 pandemic influenza A (H1N1) virus infection. We used population-based data from the Emerging Infections Program in 10 US states, and compared characteristics of pregnant (n = 150) and nonpregnant (n = 489) seasonal, and pregnant (n = 489) and nonpregnant (n = 1088) pandemic influenza cases using χ(2) and Fisher's exact tests. Pregnant women represented 23.5% and 31.0% of all reproductive-aged women hospitalized for seasonal and pandemic influenza, respectively. Significantly more nonpregnant than pregnant women with seasonal (71.2% vs 36.0%) and pandemic (69.7% vs 31.9%) influenza had an underlying medical condition other than pregnancy. Antiviral treatment was significantly more common with pandemic than seasonal influenza for both pregnant (86.5% vs 24.0%) and nonpregnant (82.0% vs 55.2%) women. Pregnant women comprised a significant proportion of influenza-hospitalized reproductive-aged women, underscoring the importance of influenza vaccination during pregnancy.

Hospitalized patients with 2009 pandemic influenza A (H1N1) virus infection in the United States--September-October 2009.

Given the potential worsening clinical severity of 2009 pandemic influenza A (H1N1) virus (pH1N1) infection from spring to fall 2009, we conducted a clinical case series among patients hospitalized with pH1N1 infection from September through October 2009. A case patient was defined as a hospitalized person who had test results positive for pH1N1 virus by real-time reverse-transcription polymerase chain reaction. Among 255 hospitalized patients, 34% were admitted to an intensive care unit and 8% died. Thirty-four percent of patients were children <18 years of age, 8% were adults ≥ 65 years of age, and 67% had an underlying medical condition. Chest radiographs obtained at hospital admission that had findings that were consistent with pneumonia were noted in 103 (46%) of 255 patients. Among 255 hospitalized patients, 208 (82%) received neuraminidase inhibitors, but only 47% had treatment started ≤ 2 days after illness onset. Overall, characteristics of hospitalized patients with pH1N1 infection in fall 2009 were similar to characteristics of patients hospitalized with pH1N1 infection in spring 2009, which suggests that clinical severity did not change substantially over this period.

Adult hospitalizations for laboratory-positive influenza during the 2005-2006 through 2007-2008 seasons in the United States.

BACKGROUND: Rates of influenza-associated hospitalizations in the United States have been estimated using modeling techniques with data from pneumonia and influenza hospitalization discharge diagnoses, but they have not been directly estimated from laboratory-positive cases. METHODS: We calculated overall, age-specific, and site-specific rates of laboratory-positive, influenza-associated hospitalization among adults and compared demographic and clinical characteristics and outcomes of hospitalized cases by season with use of data collected by the Emerging Infections Program Network during the 2005-2006 through 2007-2008 influenza seasons. RESULTS: Overall rates of adult influenza-associated hospitalization per 100,000 persons were 9.9 during the 2005-2006 season, 4.8 during the 2006-2007 season, and 18.7 during the 2007-2008 season. Rates of hospitalization varied by Emerging Infections Program site and increased with increasing age. Higher overall and age-specific rates of hospitalization were observed during influenza A (H3) predominant seasons and during periods of increased circulation of influenza B. More than 80% of hospitalized persons each season had > or =1 underlying medical condition, including chronic cardiovascular and metabolic diseases. CONCLUSIONS: Rates varied by season, age, geographic location, and type/subtype of circulating influenza viruses. Influenza-associated hospitalization surveillance is essential for assessing the relative severity of influenza seasons over time and the burden of influenza-associated complications.

Influenza-associated pneumonia in children hospitalized with laboratory-confirmed influenza, 2003-2008.

BACKGROUND: Pneumonia is one of the most common complications in children hospitalized with influenza. We describe hospitalized children with influenza-associated pneumonia and associated risk indicators. METHODS: Through Emerging Infections Program Network population based surveillance, children aged <18 years hospitalized with laboratory confirmed influenza with a chest radiograph during hospitalization were identified during the 2003-2008 influenza seasons. A case with radiologically confirmed influenza-associated pneumonia was defined as a child from the surveillance area hospitalized with: (1) laboratory-confirmed influenza and (2) evidence of new pneumonia on chest radiograph during hospitalization. Hospitalized children with pneumonia were compared with those without pneumonia by univariate and multivariate analysis. RESULTS: Overall, 2992 hospitalized children with influenza with a chest radiograph were identified; 1072 (36%) had influenza-associated pneumonia.When compared with children hospitalized with influenza without pneumonia, hospitalized children with influenza-associated pneumonia were more likely to require intensive care unit admission (21% vs. 11%, P < 0.01), develop respiratory failure (11% versus 3%, P < 0.01), and die(0.9% vs. 0.3% P 0.01). In multivariate analysis, age 6 to 23 months(adjusted OR: 2.1, CI: 1.6 -2.8), age 2 to 4 years (adjusted OR: 1.7, CI:1.3-2.2), and asthma (adjusted OR: 1.4, CI: 1.1-1.8) were significantly associated with influenza-associated pneumonia. CONCLUSIONS: Hospitalized children with influenza-associated pneumonia were more likely to have a severe clinical course than other hospitalized children with influenza, and children aged 6 months to 4 years and those with asthma were more likely to have influenza-associated pneumonia.Identifying children at greater risk for influenza-associated pneumonia will inform prevention and treatment strategies targeting children at risk for influenza complications.

Burden of seasonal influenza hospitalization in children, United States, 2003 to 2008.

OBJECTIVES: To estimate the rates of hospitalization with seasonal influenza in children aged <18 years from a large, diverse surveillance area during 2003 to 2008. STUDY DESIGN: Through the Emerging Infections Program Network, population-based surveillance for laboratory-confirmed influenza was conducted in 10 states, including 5.3 million children. Hospitalized children were identified retrospectively; clinicians made influenza testing decisions. Data collected from the hospital record included demographics, medical history, and clinical course. Incidence rates were calculated with census data. RESULTS: The highest hospitalization rates occurred in children aged <6 months (seasonal range, 9-30/10 000 children), and the lowest rates occurred in children aged 5 to 17 years (0.3-0.8/10 000). Overall, 4015 children were hospitalized, 58% of whom were identified with rapid diagnostic tests alone. Forty percent of the children who were hospitalized had underlying medical conditions; asthma (18%), prematurity (15% of children aged <2 years), and developmental delay (7%) were the most common. Severe outcomes included intensive care unit admission (12%), respiratory failure (5%), bacterial coinfection (2%), and death (0.5%). CONCLUSIONS: Influenza-associated hospitalization rates varied by season and age and likely underestimate true rates because many hospitalized children are not tested for influenza. The proportion of children with severe outcomes was substantial across seasons. Quantifying incidence of influenza hospitalization and severe outcomes is critical to defining disease burden.

Acute flaccid paralysis: the spectrum of a newly recognized complication of West Nile virus infection.

OBJECTIVES: Acute flaccid paralysis (AFP) has recently emerged as a major central nervous system complication associated with West Nile virus (WNV) infection. The spectrum of clinical presentations of AFP in WNV infection and its sequelae have not been well-studied. METHODS: We describe three patients with AFP due to WNV infection and review the clinical presentations of 56 patients with this complication derived from published studies. RESULTS: Patients with AFP and WNV presented with a spectrum of illness ranging from single extremity paralysis to quadriparalysis with cranial nerve involvement. Patients commonly developed respiratory failure (54%) and bladder dysfunction (22%). While fever was nearly universal (92%), signs of meningismus were less common (17%). Cerebrospinal fluid (CSF) analysis generally revealed a modest pleocytosis, and imaging studies were not diagnositic. Persistent neurologic impairment occurred in all survivors; overall mortality rate was high (22%) and was associated with both the extent of paralysis and advanced age. CONCLUSION: AFP in the setting of WNV is associated with significant mortality and long-term morbidity.

Case series: outbreak of conversion disorder among amish adolescent girls.

OBJECTIVE: Outbreak investigations are challenging in a cross-cultural context, and outbreaks of psychiatric disease are rare in any community. We investigated a cluster of unexplained debilitating illness among Amish girls. METHOD: We reviewed the medical records of cases, consulted with health care providers, performed active case finding, administered open-ended and structured interviews, and met with Amish community members. A case-patient was defined as a resident of the county who was bedridden for more than 3 weeks with otherwise unexplained systemic weakness and anorexia from January 2000 to February 2002. RESULTS: Five case-patients were identified. All were Amish girls aged 9 to 13 years. All five had debilitating voluntary motor deficits, anorexia, and weight loss. Four experienced neck weakness with inability to hold up their heads. Thorough medical evaluations failed to identify an organic etiology. All five patients met the diagnostic criteria for conversion disorder. Substantial social conflict within the Amish community preceded illness onset. Family behavioral interventions were recommended. Three months after the investigation, four of five patients demonstrated some improvement. CONCLUSIONS: Clinicians should be aware of the potential for outbreaks of psychogenic illness and work to address the challenges of developing effective intervention strategies, particularly in the cross-cultural context.

Photokeratitis and UV-radiation burns associated with damaged metal halide lamps.

BACKGROUND: Damaged metal halide lamps are known to cause outbreaks of photokeratitis and UV-radiation burns among children and adults, which has prompted the Food and Drug Administration, Rockville, Md, to publish consumer recommendations to prevent such injuries. We investigated 3 outbreaks of photokeratitis and UV-radiation burns in gymnasiums associated with failure to heed these recommendations. OBJECTIVE: To determine the cause of the outbreaks and promote interventions to prevent further injuries. DESIGN AND SETTING: A cohort study of persons exposed to damaged metal halide lamps during the index outbreak in a community gymnasium and a descriptive epidemiologic study of 2 subsequently identified outbreaks in other gymnasiums. PARTICIPANTS: A total of 273 persons potentially exposed during events in 3 gymnasiums. MAIN OUTCOME MEASURE: Photokeratitis with onset within 12 hours of the event. The intensity of UV radiation was measured, and an occupational exposure standard applied. RESULTS: Investigation of the index outbreak identified 18 (approximately 3%) persons who met our case definition for photokeratitis. The median incubation period was 7 hours, and health care visits were reported by 11 persons (61%). Of the 18 patients, 17 (94%) were seated in the back of the gymnasium. Among 37 persons sitting in this high-risk area, the attack rate was 46%. Only 1 (9%) of 11 persons wearing glasses or contact lenses with UV-radiation protection in the high-risk area developed photokeratitis (relative risk, 0.15; P =.01). The safe occupational exposure limit in the high-risk area was 10 to 15 minutes, but exposures of 1 to 3 hours were reported. Prevention recommendations had not been instituted at any of the 3 facilities. CONCLUSIONS: Injuries from metal halide lamps are avoidable, but prevention recommendations may not be widely observed. All facilities using metal halide lamps in areas where children may be exposed should follow the Food and Drug Administration recommendations; amending the National Electric Code may be warranted.

Childhood immunization rates before and after the implementation of Medicaid managed care.

OBJECTIVE: To evaluate trends in childhood immunization coverage after implementation of Medicaid managed care in Tennessee (TennCare) in 1994. DESIGN: Before-and-after study using the Tennessee Department of Health annual cross-sectional survey of children aged 24 months. PATIENTS: A mean of 1663 children per year who were randomly sampled during 1986-1999. MAIN OUTCOME MEASURE: Completion rate for recommended immunizations by the age of 24 months or younger. RESULTS: A total of 23 044 children were included. The proportion of children continuously enrolled in Medicaid from age 1 to 24 months increased slightly with TennCare. Among children enrolled, immunization rates increased considerably before TennCare (1986-1993) and continued to increase after TennCare (1994-1999), albeit less dramatically. Immunization coverage was significantly lower for children enrolled compared with children not enrolled in fee-for-service Medicaid. Among children enrolled in fee-for-service Medicaid, black children were more likely to be inadequately immunized than white children (40% vs 26%; relative risk [RR], 1.56; 95% confidence interval [CI], 1.40-1.73). These gaps were nearly eliminated after TennCare. An increased proportion of children enrolled in TennCare received immunizations in the private sector. Among children enrolled in fee-for-service Medicaid, those receiving immunizations entirely in the private sector were more likely to have incomplete immunization status than children immunized entirely in the public sector (27% vs 21%; RR, 1.28; 95% CI, 1.20-1.37). Under TennCare and after implementation of the Vaccines for Children program in Tennessee, the difference was not significant. CONCLUSIONS: Overall, TennCare had no discernible negative effect on immunization rates in Tennessee and perhaps contributed to decreasing the immunization gap between children enrolled and children not enrolled in Medicaid and between black and white children.

Outbreak of joint and soft-tissue infections associated with injections from a multiple-dose medication vial.

An outbreak of Staphylococcus aureus joint and soft-tissue infections occurred after therapeutic injections in an outpatient setting. A physician performed intra-articular or soft-tissue injections on 17 patients in August 2001, and 5 (29%) were subsequently hospitalized for infections at the site. S. aureus was isolated from 4 patients, and all 4 isolates were indistinguishable by pulsed-field gel electrophoresis. Of 10 patients injected with lidocaine and triamcinolone, 5 (50%) developed infections, compared with 0 of 7 patients injected with triamcinolone only (P=.04). A multiple-dose vial (MDV) of lidocaine was likely contaminated with S. aureus. A possible contributing factor was refrigeration after the use of MDVs of lidocaine; the manufacturer recommends storage at room temperature. An in vitro study of S. aureus in MDVs of lidocaine revealed prolonged survival at refrigerator temperatures. This outbreak highlights the importance of strict attention to aseptic procedures and carefully following manufacturers' instructions when using MDVs.

Pseudomonas aeruginosa and Serratia marcescens contamination associated with a manufacturing defect in bronchoscopes.

BACKGROUND: Several outbreaks and pseudo-outbreaks of Pseudomonas aeruginosa and Serratia marcescens infections associated with bronchoscopy have been reported. We conducted an investigation of P. aeruginosa and S. marcescens isolates related to bronchoscopy at a community hospital. METHODS: We reviewed the records of all bronchoscopic procedures at the community hospital from July to October 2001. Environmental samples were obtained. Pulsed-field gel electrophoresis (PFGE) was performed on isolates of P. aeruginosa. RESULTS: From July 1 to October 31, 2001, 66 bronchoscopic procedures were performed in 60 patients, and 43 specimens were obtained for bacterial culture; 20 of the specimens (47 percent) were positive for P. aeruginosa. Six (30 percent) of the specimens that were positive for P. aeruginosa also yielded S. marcescens. All 20 P. aeruginosa isolates were associated with procedures performed with three of four new bronchoscopes from the same manufacturer. Contrary to manufacturing specifications, the biopsy-port caps on all four bronchoscopes were easily removable, and P. aeruginosa was cultured from the biopsy ports of the three implicated bronchoscopes. The PFGE patterns of P. aeruginosa isolates from the bronchoscopes, patients, and two environmental samples were indistinguishable. One patient was hospitalized with P. aeruginosa pneumonia 11 days after bronchoscopy. The manufacturer reported a design change instituted in 1997, and production problems may have resulted in the distribution of bronchoscopes that did not meet specifications. CONCLUSIONS: We documented contamination of bronchoscopes with P. aeruginosa and S. marcescens and possible infection of patients at a community hospital as a result of the inadequate disinfection of bronchoscopes because of a manufacturing defect.

Outbreak of aseptic meningitis associated with echovirus 13.

BACKGROUND: Before 2001, echovirus 13 accounted for only 65 of approximately 45,000 reported enteroviral isolates in the United States. During spring 2001, several outbreaks of echovirus 13 meningitis occurred, primarily affecting children. We investigated a large outbreak in Shelby County, TN, to determine the characteristics and clinical manifestations of echovirus 13 meningitis. METHODS: We identified cases of aseptic meningitis at a children's hospital from April through August 2001 by reviewing discharge records. For patients with laboratory-confirmed echovirus 13 meningitis, we reviewed charts and interviewed parents. RESULTS: We identified 303 hospitalizations caused by aseptic meningitis at the children's hospital from April through August. Hospitalizations peaked in May. Twenty-six percent of hospitalized patients were infants age <4 months; 63% were male. Hospitalization rates were 3 times greater among black children than among white children (140 vs. 47 per 100,000). Echovirus 13 was isolated from specimens from 37 (80%) of 46 patients with positive viral cultures. Of those with laboratory-confirmed echovirus 13, 35 (95%) had fever, 26 (70%) had vomiting, 20 (54%) had headache, 16 (43%) had stiff neck and 16 (43%) had irritability. No sequelae or deaths were identified. CONCLUSIONS: Echovirus 13 emerged as a predominant strain of enterovirus associated with aseptic meningitis in the United States in 2001. In this outbreak echovirus 13 meningitis appeared to be clinically indistinguishable from aseptic meningitis caused by other enteroviruses.

Bioterrorism-related anthrax surveillance, Connecticut, September-December, 2001.

On November 19, 2001, a case of inhalational anthrax was identified in a 94-year-old Connecticut woman, who later died. We conducted intensive surveillance for additional anthrax cases, which included collecting data from hospitals, emergency departments, private practitioners, death certificates, postal facilities, veterinarians, and the state medical examiner. No additional cases of anthrax were identified. The absence of additional anthrax cases argued against an intentional environmental release of Bacillus anthracis in Connecticut and suggested that, if the source of anthrax had been cross-contaminated mail, the risk for anthrax in this setting was very low. This surveillance system provides a model that can be adapted for use in similar emergency settings.

Fatal inhalational anthrax in a 94-year-old Connecticut woman.

We describe the 11th case of bioterrorism-related inhalational anthrax reported in the United States. The presenting clinical features of this 94-year-old woman were subtle and nondistinctive. The diagnosis was recognized because blood cultures were obtained prior to administration of antibiotics, emphasizing the importance of this diagnostic test in evaluating ill patients who have been exposed to Bacillus anthracis. The patient's clinical course was characterized by progression of respiratory insufficiency, pleural effusions and pulmonary edema, and, ultimately, death. Although her B anthracis bacteremia was rapidly sterilized after initiation of antibiotic therapy, viable B anthracis was present in postmortem mediastinal lymph node specimens. The source of exposure to B anthracis in this patient is not known. Exposure to mail that was cross-contaminated as it passed through postal facilities contaminated with B anthracis spores is one hypothesis under investigation.