In early 2011, the Centers for Disease Control and Prevention (CDC) invited persons from CDC, the National Institutes of Health, the US Food and Drug Administration (FDA), and university medical centers with expertise in anthrax and critical care medicine to develop an agenda and a list of potential expert panelists to revise the current guidance for prevention and treatment of anthrax. In October 2011 and March 2012, CDC held subject matter–expert panel meetings to review available scientific information and update previous guidelines for prevention and treatment of anthrax. Meeting attendees (see end of article) included clinicians with experience caring for the most recently hospitalized patients with naturally occurring cases of inhalation, gastrointestinal, or injection anthrax and academic and federal civilian and military researchers with anthrax expertise. Specialties represented included internal medicine, pediatrics, obstetrics, infectious disease, emergency medicine, critical care, pulmonology, hematology, and nephrology.

At the meetings, attendees reviewed data from published literature on human anthrax case reports, human and animal clinical trials, anthrax prevention and treatment, and unpublished antitoxin efficacy and safety data. Workgroups were formed to address clinical management, antimicrobial drug use, and antitoxin use. The workgroups drafted changes to previous guidelines, vetted them before the expert panel, and reached general agreement. After the meeting, workgroup members further refined the proposed changes by teleconference and email exchanges. The final draft was distributed to all meeting participants for review and comment. This article summarizes the presentations and discussions of the meeting and updates previous CDC guidelines for prevention and treatment of anthrax in nonpregnant adults.

B. anthracis, the causative agent of anthrax, is a nonmotile spore-forming, gram-positive, rod-shaped bacterium. The spores of B. anthracis, which can remain dormant in the environment for decades, are the infectious form, but vegetative B. anthracis rarely causes disease (1). Spores introduced through the skin lead to cutaneous or injection anthrax; those introduced through the gastrointestinal tract lead to gastrointestinal anthrax; and those introduced through the lungs lead to inhalation anthrax. After entering a human or animal, B. anthracis spores are believed to germinate locally or be transported by phagocytic cells to the lymphatics and regional lymph nodes, where they germinate; or both (6). B. anthracis begins producing toxins within hours of germination (7). Protective antigen (PA) and edema factor (EF) combine to form edema toxin (ET) and PA and lethal factor (LF) combine to form lethal toxin (LT). After binding to surface receptors, the PA portion of the complexes facilitates translocation of the toxins to the cytosol, in which EF and LF exert their toxic effects (8).

The toxins have several physiologic effects. In sedated and mechanically ventilated dogs challenged with 24-h LT infusions to simulate release during anthrax infection, the toxin produced gradual but progressive hypotension associated with reductions in systemic vascular resistance and left ventricular ejection fraction that persisted for ≤72 h (9). Similar challenges with ET produced rapid hypotension associated with sharp and persistent decreases in central venous pressure and systemic vascular resistance. Such changes suggest that both toxins alter the peripheral vasculature, but that LT might also have a direct effect on depressing myocardial function, and ET causes venous and arterial relaxation. Both toxins were associated with progressive renal and hepatic dysfunction because of direct toxic effects or secondarily to hypoperfusion (9).

Gross pathologic lesions observed in non-human primates (NHPs) used in aerosol challenge models of inhalation anthrax include edema, congestion, hemorrhage, and necrosis in the lungs and mediastinum. Splenitis and necrotizing or hemorrhagic lymphadenitis involving the mediastinal, tracheobronchial, and other lymph nodes are common (10). Primary pulmonary lesions, including those of pneumonia, are occasionally observed. Meningeal involvement ranging from edema, congestion, hemorrhage, and necrosis to suppurative or hemorrhagic meningitis, usually secondary to hematogenous spread from other types of anthrax, occurs in ≤77% of animals studied (10). Autopsy findings for persons who died from inhalation anthrax in Sverdlovsk and in the United States (3) are consistent with findings from the NHP studies. Persons who died had extensive amounts of serosanguinous fluid in pleural cavities and edema and hemorrhage of the mediastinum and surrounding soft tissues, and 48% had cerebral edema, 21% had ascites, 17% had pericardial effusions, and 14% had petechial rash. Mediastinal lymph nodes and spleen also showed hemorrhage and necrosis (11,12).

The approach to prevention and treatment of anthrax differs from that for other bacterial infections. The production of toxin, potential for antimicrobial drug resistance, frequent occurrence of meningitis, and presence of latent spores must be taken into account when selecting postexposure prophylaxis (PEP) or a combination of antimicrobial drugs for treatment of anthrax. In addition, most data on which these antimicrobial drug options are based predate the availability of many of the antimicrobial drugs discussed and are drawn from limited animal studies.

Among patients with inhalation anthrax during 1900–2005, antimicrobial drug combination therapy was more likely to be curative than antimicrobial drug monotherapy (26). There is also a theoretical benefit for combined use of bactericidal and protein synthesis inhibitor agents. Bactericidal agents have immediate killing effect. However, the high rates of illness and death seen with anthrax are caused, in part, by B. anthracis exotoxin production. In vitro toxin production is inhibited earlier by protein synthesis inhibitors than by bactericidal agents (29); this characteristic is associated with clinical benefit in streptococcal toxic shock and clostridial sepsis. Combination antimicrobial drug regimens were used in all 8 survivors of inhalation anthrax during 2001–2012; five survivors were also treated with a protein synthesis inhibitor (4,26,30). Patients hospitalized for systemic anthrax should be immediately treated with a combination of broad-spectrum intravenous antimicrobial drug treatment pending confirmatory test results because any delay may prove fatal.

Naturally occurring B. anthracis has variable β-lactam resistance, particularly to cephalosporins (31). Thus, cephalosporins are contraindicated. In addition, multidrug resistance in naturally occurring B. anthracis infection has been reported (32) and can be induced in vitro (33). This finding increases concern that a B. anthracis strain could be resistant to ≥1 recommended antimicrobial drug options or that use of β-lactams for a strain that was originally β-lactam susceptible could lead to development of β-lactam-resistance during prophylaxis, if adherence to treatment is poor, as was observed among postal workers in 2001 (34). Another concern is that β-lactam use could induce resistance during treatment, especially with the high bacterial loads characteristic of systemic anthrax. Thus, penicillin-based antimicrobial drug use requires a high index of suspicion for emergence of resistance.

Meningitis and hemorrhagic brain parenchymal infection has been observed in ≤50% of human anthrax cases (16,26). Thus, meningitis must be considered in all cases of systemic anthrax, and antimicrobial drugs used to treat possible meningitis must have good penetration of the central nervous system (CNS).

The presence of a spore form of B. anthracis requires prolonged antimicrobial drug prophylaxis. Incubation periods ≤43 days have been observed in humans (2) and 58 days in NHPs (35), and viable spores have been detected in lungs of NHPs for ≤100 days after aerosol exposures (36). Thus, persons exposed to aerosolized B. anthracis are presumed to be at risk for inhalation anthrax from ungerminated spores retained in their lungs after the initial exposure, including patients treated for any form of anthrax who were exposed to aerosolized spores. In addition, if antimicrobial drug treatment is initiated soon after exposure, animal studies suggest the acquired immune response might be blunted and not be protective (35). This finding suggests the need to continue antimicrobial drug therapy for 60 days to clear germinating organisms.

Well-timed and effective PEP can potentially save thousands of lives (Technical Appendix Table 2). PEP of asymptomatic persons should ideally start as soon as possible after exposure because its effectiveness decreases with delay in implementation. In 2009, the US Advisory Committee on Immunization Practices recommended 60 days of antimicrobial drug prophylaxis for immediate protection and a 3-dose series of Anthrax Vaccine Adsorbed (AVA) BioThrax (Emergent Biosolutions Inc., Rockville, MD, USA) for long-term protection after exposure to anthrax (37). To ensure adequate and continued protection, everyone exposed to aerosolized B. anthracis spores should receive a full 60 days of PEP antimicrobial drugs, whether they are unvaccinated, partially vaccinated, or fully vaccinated.

Empiric treatment for anthrax in which anthrax meningitis is suspected or cannot be ruled out should include ≥3 antimicrobial drugs with activity against B. anthracis; ≥1 drug should have bactericidal activity, ≥1 should be a protein synthesis inhibitor, and all should have good CNS penetration (Technical Appendix Table 3). Intravenous combination treatment for systemic anthrax with possible meningitis should be provided for ≥2 weeks or until the patient is clinically stable, whichever is longer. Given the high mortality rate associated with meningitis, some expert panelists favored 3 weeks of treatment for patients in whom meningitis could not be ruled out.

Intravenous ciprofloxacin is preferred as the primary bactericidal component in the treatment of systemic disease on the basis of efficacy in NHP infection models and recent use for anthrax cases. Levofloxacin and moxifloxacin are considered equivalent alternatives to ciprofloxacin. The fluoroquinolones have adequate CNS penetration (16) and there are no reports of natural resistance.

The carbapenem class of antimicrobial drugs is highly resistant to β-lactamases and provides good CNS penetration. Meropenem is preferred as the second antimicrobial drug in the combination antimicrobial drug regimen for anthrax meningitis. If meropenem is unavailable, doripenem and imipenem/cilastatin are considered equivalent alternatives. Imipenem/cilastatin is associated with increased seizure risk (38) and should be used with caution in patients with suspected meningitis. If the B. anthracis strain is susceptible to penicillin (MIC <0.125 µg/mL), penicillin G or ampicillin are acceptable alternatives to carbapenems.

At least 1 antimicrobial drug that inhibits protein synthesis should be used to reduce exotoxin production. Linezolid is preferred as the first-line protein synthesis inhibitor. It is preferred over clindamycin because it is likely to provide better CNS penetration (39), although randomized controlled trials on treatment for CNS infections with either agent are lacking. However, linezolid toxicity issues must be taken into consideration. Myelosuppression, peripheral and optic neuropathies, and serotonin syndrome have been reported in patients receiving linezolid (40). Linezolid should be used cautiously in patients with pre-existing myelosuppression. In patients receiving monoamine oxidase inhibitors or serotonin reuptake inhibitors, the benefit of linezolid treatment should be weighed against the risk for serotonin toxicity and an alternative should be considered. If patients experience visual impairment, prompt ophthalmic evaluation is recommended. If patients have contraindications to linezolid use or it is unavailable, clindamycin is an acceptable alternative. Rifampin, although not a protein synthesis inhibitor, has been widely used for its synergistic effect with a primary drug and could also be used in this capacity if linezolid or clindamycin are unavailable. The protein synthesis inhibitor chloramphenicol has good CNS penetration and has historically been used to successfully treat anthrax. Where available, it could be an acceptable alternative if linezolid, clindamycin, and rifampin are unavailable. Doxycycline should not be used if meningitis is suspected because it does not adequately penetrate the CNS.

With the following 4 exceptions, antimicrobial drug options for patients with systemic anthrax are similar to those for patients with suspected meningitis or when meningitis cannot be ruled out (Technical Appendix Table 4). First, treatment should include ≥2 antimicrobial drugs with activity against B. anthracis; ≥1 should have bactericidal activity and ≥1 should be a protein synthesis inhibitor. Second, initial intravenous combination treatment should be given for ≥2 weeks or until the patient is clinically stable, whichever is longer. Third, if the B. anthracis strain is susceptible to penicillin, then penicillin G is considered equivalent to the fluoroquinolone options for primary bactericidal treatment. Fourth, treatment with antimicrobial drugs that have good CNS penetration is not a crucial factor. Thus, meropenem is recommended as an acceptable alternative option than as a first-line antimicrobial drug, and vancomycin is also an acceptable alternative. Clindamycin and linezolid are considered equivalent first-line choices for protein synthesis inhibitors. Doxycycline is added as an alternative protein synthesis inhibitor option if linezolid or clindamycin are contraindicated or unavailable.

Once patients with systemic illness who were exposed to aerosolized spores have completed initial combination treatment, they should be transitioned to single-agent oral treatment to prevent relapse from surviving B. anthracis spores. Antimicrobial drug options are the same as those for PEP (Technical Appendix Table 2).

Uncomplicated cutaneous anthrax has been successfully treated with a single oral antimicrobial drug (Technical Appendix Table 5). Oral fluoroquinolones (ciprofloxacin, levofloxacin, and moxifloxacin) and doxycycline are equivalent first-line agents. Clindamycin is an alternative option if fluoroquinolones and doxycycline are contraindicated or unavailable. Given the long history of successful treatment of localized uncomplicated cutaneous anthrax with penicillin, amoxicillin and penicillin VK are also alternative therapeutic options if the isolate is known to be susceptible to penicillin. However, adequate dosages must be used because of the potential for development of drug resistance during treatment with subtherapeutic dosing.

Duration of treatment for localized or uncomplicated cutaneous disease depends on the B. anthracis exposure source. If naturally acquired (e.g., animals with anthrax, products such as hides from animals with anthrax), a 7–10-day course of antimicrobial drugs is sufficient. If bioterrorism-related exposure or an aerosol exposure is suspected, the recommendation is the same as that for PEP (Technical Appendix Table 2) because the patient is likely to have also inhaled spores.

In case series of patients with cutaneous anthrax who received no treatment during 1890–1907 or antiserum during 1903–1907 (Technical Appendix Table 6), 23.7% of the 26,567 untreated patients died. In contrast, 11.8% of the 305 patients who received antiserum died. Although there are far fewer data for patients with inhalation anthrax, in the pre-antibiotic era, there were 3 survivors among 6 patients treated with antiserum (26).

There are currently 2 antitoxins in the CDC Strategic National Stockpile: raxibacumab (GlaxoSmithKline, London, UK) and Anthrax Immune Globulin Intravenous (AIGIV) (Cangene Corporation, Winnipeg, Manitoba, Canada). Both antitoxins inhibit binding of PA to anthrax toxin receptors and translocation of the 2 primary toxins (LT and ET) into cells. Raxibacumab is a recombinant, fully humanized, IgG1λ monoclonal antibody. AIGIV is a human polyclonal antiserum made from plasma of persons immunized with AVA, which might have some direct effect on LF and EF.

Biodefense experts currently place B. anthracis at or near the top of the list for potential threat agents: it is available and easy to disseminate, and systemic anthrax infection has a high mortality rate. Before 2001, mortality rates for patients with inhalation anthrax approached 90%. Since that time, 8 (53%) of 15 known patients with inhalation anthrax have survived. These survivors were given an early diagnosis, began combination antimicrobial drug treatment to eradicate the bacteria and inhibit toxin production, and had aggressive pleural effusion management. The principles of prevention and clinical care provided in this report are based on currently available data and the lessons learned from recent anthrax events.

There are many opportunities for improving prevention of and treatment for anthrax. Some evidence suggests the duration of PEP antimicrobial drug use could be decreased when given in conjunction with AVA. Further research could provide sufficient evidence to decrease the length of the antimicrobial drug component of PEP. Efforts are ongoing to develop point-of-care assays. Assays that quantify toxin levels or antibody response could enable earlier diagnosis and treatment and help identify optimal timing for antitoxin administration. Pharmacokinetic data for antimicrobial drugs used to treat anthrax are limited and almost no pharmacokinetic data for CNS are available, but such data are needed for dosing recommendations. Animal studies are needed to evaluate synergistic or antagonistic activity in multiple antimicrobial drug combinations; to study the survival effect of corticosteroids or other adjunctive treatments for meningitis, edema, and vasopressor-resistant shock; and to determine whether currently recommended antitoxin dosing is optimal. Currently available antitoxins counteract extracellular toxins. Studies are needed to assess whether intracellular toxin inhibitors, some of which are FDA approved for other indications, are efficacious in animal models of anthrax infection. Such agents might be of use in patients in the fulminant phase of disease. When research needed to change the anthrax mortality rate is conducted, B. anthracis will become a less competent, and therefore less desirable, bioweapon.