Investigation and Findings
Identification of Patient
On September 7, 2024, CDC and the Louisiana Department of Health (LDH) were contacted by a clinician regarding an otherwise healthy man aged 18 years from Louisiana, who was being admitted to an intensive care unit with severe pneumonia and respiratory failure that required endotracheal intubation and mechanical ventilation. The patient had developed a cough 1 week earlier. He had worked part-time as a welder for 6 months immediately preceding his illness and had no reported history of smoking, vaping, or excessive alcohol consumption. Blood cultures were positive for B. cereus group bacteria and, because of the patient’s pneumonia, respiratory failure, occupation, and geographic proximity to previously reported cases, welder’s anthrax was suspected (1,3,4) (Table).
Treatment
The patient initially received empiric antibiotic treatment with vancomycin, meropenem, ciprofloxacin, and doxycycline. Obiltoxaximab, a monoclonal antibody anthrax antitoxin directed against the protective antigen of B. anthracis and indicated for treatment of inhalation anthrax, was available from the U.S. Strategic National Stockpile. CDC requested the antitoxin, which was administered to the patient 34 hours after welder’s anthrax was suspected (approximately 1 week after symptom onset). The patient’s condition improved rapidly, and intubation and mechanical ventilation was discontinued 72 hours later. He received continued intravenous antimicrobial therapy (2) and drainage of a pleural effusion. He was discharged with a tailored antibiotic regimen after a 26-day hospitalization. All pulmonary symptoms had resolved by his 3-month follow-up visit.
Clinical Laboratory Analysis
Laboratory Response Network (LRN) polymerase chain reaction (PCR) testing of an isolate obtained from the patient’s blood performed by the Louisiana State Public Health Laboratory confirmed the presence of an anthrax toxin gene. CDC identified the isolate as Bacillus tropicus by LRN algorithm and whole genome sequencing (WGS). Gene comparisons using multilocus sequencing typing indicated sequence type 78 (ST-78) and B. anthracis–associated toxin genes.
Epidemiologic Investigation
After consultation with CDC and CDC’s National Institute for Occupational Safety and Health (NIOSH), LDH interviewed the patient and representatives from his worksite¶ to better understand his exposure history and welding practices. For 6 months preceding his illness onset, the patient worked 4 hours per day, 4 days per week as a welding apprentice in the shipbuilding and repair industry. He conducted shielded metal arc welding** using carbon steel and low-hydrogen carbon steel electrodes both in open areas and confined spaces†† with limited ventilation and minimal use of personal protective equipment (PPE). No fans or ventilation systems were present in the patient’s work area, including within enclosed spaces. Interviews with worksite representatives revealed that employees work on-site for several hours each day, do not consistently wear respirators when welding, and often take breaks and eat lunch in their work area rather than using the designated break area.
Environmental Investigation and Laboratory Analysis
After the patient recovered, LDH requested CDC’s assistance in investigating the environmental source of infection using previously described methods (6). A total of 245 samples were collected from the patient’s worksite, including 95 bulk soil samples from work areas, as well as 98 sterile sponge-stick§§ and 52 macrofoam-swab surface samples from tools or items that he possibly used (6). Samples were also collected from outdoor areas, tools, clothing, buildings, and break areas near the patient’s work area. All samples were stored and transported to CDC, where DNA was extracted and tested for anthrax toxin genes by LRN algorithm (6). A cycle threshold value <40 was considered positive. A second extraction was performed on positive samples. DNA was purified using Agencourt AMPure XP beads according to manufacturer instructions¶¶ for all samples with at least one negative result, and quantitative real-time PCR (qPCR) was repeated. An additional extraction was performed on positive samples and qPCR was repeated. Samples were considered qPCR-positive if two extractions were positive and inconclusive if only one extraction was positive. Culture was performed on all qPCR-positive and inconclusive samples. WGS and phylogenetic analyses were performed on culture isolates (6).
Among the 245 collected samples, 28 (11.4%) were positive by qPCR for anthrax toxin genes (cycle threshold range = 29.0–39.9), consisting of one swab sample from work gloves found on a vessel in an area where the patient worked (not conclusively identified as belonging to the patient), two sponge samples from a handrail and table where the patient worked, and 25 soil samples collected across the worksite, including areas where the patient worked. Five samples (2.0%, all soil) were inconclusive, and the remaining 212 (86.5%) were negative.
One isolate from a positive soil sample was successfully cultured. The genotype was ST-78; all other known ST-78 strains are Bacillus tropicus. WGS indicated 99.998% +/– 0.096% average nucleotide identity among genome sequences from the clinical and environmental isolates using Basic Local Alignment Search Tool (BLAST+),*** an application of the National Center for Biotechnology Information sequence similarity search program. The phylogeny further suggests their high relatedness by the short (horizontal) phylogenetic distances (Figure) and demonstrates clustering with genome sequences derived from patients from Louisiana (GCF_000832805.1, GCA_043275215.1, GCF_002007005.1, and SRR19658610), a zoo animal from Louisiana (GCF_046718865.1), the Louisiana environment (SRR19658609), a patient from Florida (GCF_000688755.1), a patient from Texas (GCF_000789315.1), and a B. cereus group genome with unknown origins (GCF_016027575.1).