Anthrax and Microbiology of B. anthracis

OVERVIEW: What every clinician needs to know

Pathogen name and classification

Bacillus anthracis is a gram-positive, spore-forming rod in the family Bacillaceae, Bacillus cereus group (B. anthracis, B. cereus, B. thuringiensis, B. mycoides, B. pseudomycoides, B. weihenstaphenensis).

What is the best treatment?

  • Cutaneous anthrax: localized, no evidence for systemic illness or dissemination

    Oral ciprofloxacin or doxycycline for 7-10 days; continue antimicrobials for 60 days if bioterrorism-related or possibility of aerosolized exposure

  • Systemic anthrax: includes inhalational, gastrointestinal, meningeal, injectional, bacteremic cases and cutaneous cases with systemic illness, extensive edema, or involvement of the head/face/neck

    Intravenous ciprofloxacin, levofloxacin, or doxycycline; may be changed to oral formulation when clinically indicated (i.e., patient able to take oral medications)

    Clindamycin or other protein synthesis inhibitors may be added to impair toxin synthesis.

    Multi-drug therapy is recommended mainly because of the risk of bio-engineered resistance.

    Continue antimicrobial therapy for 60 days.

    In cases in which meningitis is diagnosed or suspected, add at least 1 agent with effective CNS penetration and activity against B. anthracis (e.g., meropenem). Note that penicillin may be used in this role; the susceptibility of the infecting strain is confirmed.

    Vaccination with AVA is recommended.

    Drainage of extravascular fluid collections is recommended (e.g., thoracentesis for pleural fluid, paracentesis for ascites) as these may be toxin reservoirs.

    Administration of anthrax immune globulin (AIG, available from the CDC) may provide additional benefits beyond antimicrobial therapy.

    The role of raxibacumab (a monoclonal antibody against the protective antigen component of anthrax toxin) is unknown but may also provide benefit if administered very early in the course of systemic anthrax.

    The roles of steroid therapy for anthrax meningitis or malignant edema remain undefined.

  • Third-generation cephalosporins and trimethoprim-sulfamethoxazole should be avoided. Engineered resistance is easy to achieve and should be a consideration in patients with potential bioterrorism-related exposure.

    Class A (bla1) and class B (bla2) beta lactamases are encoded on the B. anthracis chromosome. B. anthracis also encodes an arylamine N-acetyltransferase [(BACAN)NAT1] that acetylates and inactivates sulfamethoxazole, and the dihydrofolate reductase of this bacterium is inherently resistant to trimethoprim. Mutations in the type II topoisomerase genes have been associated with high-level fluoroquinolone resistance in two strains made resistant via serial passage. The mechanisms underlying resistance to other antimicrobials have not been investigated to date.

    Susceptibility testing at a reference clinical microbiology laboratory within the CDC Laboratory Response Network (LRN) is the best method for detecting resistance. Both automated microbroth dilution and Etest agar gradient diffusion produce comparable results.

    Adjunctive anti-toxin therapy using immunoglobulins is also available: anthrax immune globulin (available via the CDC), raxibacumab (a human monoclonal antibody against protective antigen; Human Genome Sciences, Rockville, MD)

How do patients contract this infection, and how do I prevent spread to other patients?

  • Epidemiology

    Although there has not been a seasonal variation in anthrax incidence described in humans, the protean symptoms of inhalational anthrax can be mistaken for influenza. It is prudent to maintain clinical suspicion for anthrax during seasonal influenza outbreaks.

    The main determinant of infection risk is the presence of B. anthracis spores in the direct environment of the host. The specific interaction of the host with the environment determines the route of exposure and type of clinical disease.

    aerosol exposuresà inhalational anthrax

    contact exposureà cutaneous anthrax

    ingestion exposureàgastrointestinal anthrax

    injection exposureà complicated skin/soft tissue anthrax.

    Based on non-human primate studies, the estimated lethal dose of inhaled spores for humans ranges from 2500 to 760,000.

    Anthrax remains predominantly a zoonotic infection in areas of the world with a primarily agriculture-based economy. In these areas, the incidence of anthrax is inversely related to the extent of livestock vaccination.

    Anthrax is endemic in many areas of the world, including East, Central, and Southeast Asia; the Indian sub-continent; Indonesia; sub-Saharan Africa; Mexico; and parts of Central and South America. Sporadic cases continue to occur in the United States, particularly in parts of New Mexico, southwestern Texas, and the Midwest.

    Bioterrorism-related anthrax is, by definition, a non-prevalent infection.

    The exact incidences of human anthrax infections are unknown, mainly because most cases of human anthrax occur in settings where anthrax is endemic and not formally studied for epidemiological purposes. The World Anthrax Data Site ( provides an excellent global view of both veterinary and human anthrax disease activity.

  • Infection control issues

    The vegetative bacillary form of B. anthracis is not readily transmitted between infected people and others. However, a person with anthrax has clearly been exposed to the infectious spore form of B. anthracis and should be considered contagious until appropriately evaluated. A determination should be made regarding likely sources of exposure, as the source helps guide decisions regarding decontamination and the use of personal protective equipment (PPE).

    Endemic or natural epidemic exposures, such as those that occur in the agricultural or veterinary setting, are unlikely to pose a public health threat in the healthcare setting, and universal precautions should be applied. Outside these settings, bioterrorism-related exposures should be considered. In this situation, the patient may be a continued source for spore exposure because of contamination of clothing and personal articles, and these items should be handled as biohazardous material.

    Decontamination protocols should be in place for rapid triage and disposition of patients potentially exposed to B. anthracis via bioterrorism. Once the patient has been appropriately decontaminated, universal precautions should be observed while caring for them, but no additional precautions are necessary because of the negligible risk of person-to-person transmission.

    Vaccination using the current, FDA-approved anthrax vaccine adsorbed (AVA) is recommended for specific groups based on level of risk for exposure to B. anthracis spores.

    Pre-exposure vaccination is for groups at risk of spore exposure despite available measures for reducing exposure. Examples include laboratory personnel with repeated exposure to fully virulent B. anthracis spores as part of their research, military personnel at risk of exposure to spores, or people performing environmental investigations and/or remediation of spaces that may contain the spores. Pre-exposure vaccination consists of a total of five AVA doses given intramuscularly (im) at start; 4 weeks; and 6, 12, and 18 months, followed by annual booster doses.

    Post-exposure prophylaxis (PEP) is for previously unvaccinated people with spore exposure in addition to post-exposure antimicrobial prophylaxis. PEP vaccination consists of three doses of AVA given at start and 2 and 4 weeks, followed by doses at 6, 12, and 18 months.

  • Post-exposure anti-infective prophylaxis is recommended for anyone with potential aerosol exposure to B. anthracis spores. Currently, the recommended duration of PEP is 60 days, using one of the following antimicrobial agents:

    First line: ciprofloxacin, doxycycline

    Second line: levofloxacin

    Alternatives: penicillin G procaine, amoxicillin; note that beta-lactam susceptibility may be compromised in a bio-engineered strain of B. anthracis, and penicillin resistance can occur in naturally acquired strains. For these reasons, beta lactams are not recommended as first line agents for the prophylaxis or treatment of B. anthracis, unless the drug susceptibility of the infecting strain is known.

    Anti-infective prophylaxis should be done in conjunction with post-exposure vaccination.

What host factors protect against this infection?

  • Components of both the innate and adaptive arms of the immune system participate in protection against infection with B. anthracis. However, B. anthracis utilizes several mechanisms to overcome these protective responses and cause disease. The most important of these mechanisms appears to be production of anthrax toxin, leading to effective paralysis of both innate and adaptive immune responses. In fact, administration or inherent production of antibodies directed against protective antigen (PA), a component of anthrax toxin, both provide a significant measure of protection against anthrax. This protection forms the basis for the efficacy of AVA as an element of PEP.

  • People with increased exposure to B. anthracis spores are at higher risk for contracting this infection. Little is known about human-specific factors that may predispose a person to infection with B. anthracis.

    Animal models of experimental anthrax have demonstrated roles for complement deficiency, Toll-like receptor 2 (TLR2) and TLR4 expression, MyD88 signaling, Nod-like receptor (NLR) recognition, and CXC chemokine production in the immune response to B. anthracis.

    It is likely that people traditionally considered immunocompromised, such as those with organ transplants, HIV/AIDS, or hematological malignancies, are at increased risk for infection with B. anthracis, but this hypothesis waits testing in the clinical arena.

  • The histopathology of inhalational anthrax has been extensively described from humans infected during an accidental release of B. anthracis from a Soviet bioweapon facility in Sverdlovsk in 1979. These patients had hemorrhagic lymphadenitis and hemorrhagic mediastinitis, and cases in which dissemination occurred often had hemorrhagic meningitis and hemorrhagic lesions in the submucosa of the gastrointestinal tract. More recent cases of inhalational anthrax from the 2001 outbreak in the United States confirmed observations from Sverdlovsk of extravascular fluid accumulation, especially in the form of pleural effusions. Cutaneous anthrax demonstrates ulceration, edema, coagulative necrosis, vasculitis, perivascular inflammation, and tissue hemorrhage. Gastrointestinal (GI) anthrax pathology has not been systematically described in the published literature, but surgical findings from patients with GI anthrax often show ulceration of mucosal layers along the GI tract, mesenteric lymphadenitis, bowel edema, and hemorrhage.

    The pathogenesis of inhalational anthrax is the most thoroughly studied aspect of this disease. The pathogenesis of other forms of anthrax, such as primary/meningeal, cutaneous, gastrointestinal, and injectional disease, has not been as thoroughly investigated.

    Inhalational anthrax begins with the phagocytosis of spores by alveolar macrophages and dendritic cells in the lung.

    These spores are transported by these cells to regional mediastinal lymph nodes and undergo germination into the bacillary form.

    Shortly after germination, bacilli encapsulate and begin producing anthrax toxin. Early toxin production is critical for promoting survival of bacilli within professional phagocytes.

    Bacilli also express several virulence factors that promote escape from phagosomes and, ultimately, the phagocyte. These factors include a cholesterol-dependent cytolysin, anthrolysin O, and several phospholipases.

    Once the bacilli have escaped from phagocytes, they begin replicating and gain entry into the systemic circulation, where they disseminate to other organ systems.

    Bacilli multiply to very high densities in the blood (~10 log10 cfu/mL) and continue to produce anthrax toxin. The toxin components [protective antigen (PA), edema factor (EF), and lethal factor (LF)] work together in many different target cell types to disable the host immune response, damage host tissues, and promote survival and dissemination of B. anthracis. Some examples of mechanisms utilized by B. anthracis to cause disease include:

    Lethal toxin (LT=PA+LF) causes endothelial cell dysfunction and death, macrophage apoptosis, and interferes with dendritic cell antigen presentation.

    B. anthracis secretes two metallo-proteases, Npr599 and InhA, that cleave von Willebrand factor and ADAMTS13, resulting in coagulopathy.

    Edema toxin (ET=PA+EF) induces vascular leakage, perivascular inflammatory infiltrates, and edema formation via several mediators, including arachidonic acid derivatives, neurokinins, and histamine.

    Both ET and LT inhibit neutrophil actin-based motility, leading to impaired chemotaxis and phagocytosis of bacilli.

    Both ET and LT cause cardiovascular dysfunction, hypotension, and shock, with associated mortality, in animal models.

What are the clinical manifestations of infection with this organism?

  • The main forms of anthrax are cutaneous, gastrointestinal (GI), inhalational (IA), injection-related, and meningeal (sometimes referred to as “primary” when it occurs in isolation). Cutaneous anthrax is the most common form and is associated with the lowest mortality, whereas anthrax meningitis, either isolated or occurring with another form of the disease, is almost always fatal.

    Cutaneous anthrax typically manifests on exposed skin surfaces (i.e., the face, extremities, and less commonly on the trunk) as a painless, ulcerating papule. The lesion may begin as a papule, but evolves with central ulceration, eschar formation, and peri-lesional edema. There may be vesicles that precede the ulcer or are present as satellite lesions. The eschar is usually very dark and “coal-like,” and this appearance is responsible for the species name “anthracis,” derived from the Greek word for coal (
    Figure 1). Systemic signs and symptoms, such as fever, leukocytosis, and hypotension, may be present, especially in severe forms of this disease. Malignant edema is one severe form of facial cutaneous anthrax that may present with airway compromise. Untreated cutaneous anthrax may lead to dissemination, septic shock, and death, but appropriate antimicrobial therapy leads to clinical cure in the vast majority of cases.

    GI anthrax is usually acquired by eating the meat from an animal that died from anthrax. There are two main forms of GI anthrax: oropharyngeal and intestinal.

    Oropharyngeal anthrax may present as a severe ulcerative pharyngitis, with airway compromise as a major concern. This disease mimics diphtheria in many respects, with pseudomembrane formation and systemic toxicity.

    The clinical presentation of intestinal anthrax is that of gastroenteritis, with nausea, vomiting, diarrhea, abdominal pain, fevers, chills, and malaise. There may also be evidence for GI hemorrhage, with hematemesis, melena, or hematochezia. A notable feature of GI anthrax is the development of ascites, sometimes in massive quantities. Disease onset after ingestion of the implicated source is fairly rapid, occurring between 15 and 72 hours in one series.

    Inhalational anthrax (IA) is often described as a biphasic illness, with the initial phase presenting as an influenza-like illness (ILI) characterized by malaise, cough, fever, and chills, followed by rapid progression to the next phase of acute respiratory failure, hypotension, altered mentation, multisystem organ failure, and death. Unlike ILI, rhinorrhea and/or nasal congestion are not common features of IA. Conversely, most patients with ILI do not experience dyspnea, whereas this is a common feature of IA.

    Chest radiographs of patients with IA may reveal a widened mediastinum, pleural effusions, and/or alveolar infiltrates (
    Figure 2).

    Injection-related anthrax is newly described, with the most extensive clinical experience coming from injection drug use (IDU) among patients in Glasgow, Scotland. The main source of B. anthracis spores in these patients was contaminated heroin imported from central Asia. Inoculation of spores into the subcutaneous tissues or muscle during IDU resulted in the development of severe soft tissue infections (i.e., cellulitis, myositis, and abscess).

    Clinical features include soft tissue edema, local excessive bruising or hemorrhage at the injection site, local pain at the injection site (
    Figure 3).

    Most patients present with rapid onset of clinical sepsis, with hypotension, coagulopathy, and multi-organ failure.

    C-reactive protein may be normal or lower than expected for the severity of illness.

    Injection-related anthrax may mimic necrotizing fasciitis (NF), but surgical exploration does not reveal NF. The Scottish series revealed fat necrosis, massive tissue edema, and excessive hemorrhage in the operative bed.

    Delayed or secondary lesions may appear despite initial debridement and stabilization of the patient.

    Anthrax meningitis complicates 10-50% of systemic anthrax cases (i.e., non-cutaneous) and may occur as the sole presentation of anthrax, in a form often referred to as primary anthrax meningitis. This form carries the highest mortality (>95%).

    Clinical presentation includes acute bacterial meningitis with fever, headache, confusion, neck stiffness, seizures, and focal motor deficits. Signs of meningismus may be absent. A petechial rash may be seen and mistaken for meningococcemia.

    The cerebrospinal fluid analysis typically reveals a neutrophil-predominant pleocytosis, elevated protein, and hypoglycorrhachia, and the Gram stain may show large “boxcar”-shaped gram-positive rods growing in chains (Figure 4).

Figure 1.

Cutaneous anthrax

Figure 2.

Chest radiograph with mediastinal widening

Figure 3.

Injectional anthrax

Figure 4.

Gram stain of B. anthracis in cerebrospinal fluid

What common complications are associated with infection with this pathogen?

  • Little is known about the long-term complications associated with anthrax. Short-term complications relate to the species.

How should I identify the organism?

  • Bacillus anthracis is readily cultivated from clinical specimens. With the exception of cutaneous anthrax, blood cultures are useful for the diagnosis of anthrax because of the propensity of this organism to disseminate from primary site of infection. Specimens obtained from the site of primary clinical involvement will likely yield B. anthracis in culture as well. The clinical microbiology laboratory must be notified of the suspicion for B. anthracis, as many Bacillus species are part of the normal flora and specific testing at a specialized reference laboratory is needed to identify B. anthracis at the species level.

  • Gram staining is the best staining technique. Gram-positive rods, commonly described as “fat,” “large,” or “boxcar”-shaped, growing in chains of varying length are the morphology by microscope. Under certain growth conditions, terminal spores may also be seen.

  • Clinical specimens should be plated on sheep blood agar (SBA) to identify non-hemolytic, non-motile Bacillus colonies. Polymyxin, lysozyme, ethylenediaminetetraacetic acid, thallium acetate (PLET) medium is a selective medium for B. anthracis, but further confirmatory tests are still required to identify the organisms at the species level.

  • Colonies are grey-white or white, flat, non-hemolytic on SBA, non-motile, and may have “tailing” (extension of the colony along the line of plate streaking). Colonies are “tacky” when teased with a loop. When grown in a 5-20% CO2 atmosphere on blood agar with 0.7% bicarbonate, capsule production can be seen with strains that harbor the pXO2 plasmid. These colonies appear mucoid, and the capsule can be visualized using McFadyean’s polychrome methylene blue staining.

  • Biochemical assays have limited utility when trying to differentiate B. anthracis from other Bacillus species and may not be needed at all when other techniques are employed.

    16S rRNA sequencing: This cannot differentiate between B. cereus and B. anthracis, as they share identical 16S rRNA sequences. However, sequencing of the intergenic spacer region between 16S and 23S rRNA genes readily differentiates between B. anthracis and members of the B. subtilis group (B. subtilis, B. megaterium, B. atrophaeus).

    Nucleic acid amplification testing (NAAT): PCR-based assays (PCR, real-time PCR, multiplex PCR) may be used to identify targets that help identify B. anthracis, such as the toxin genes (pag, cya, and lef) present on the pXO1 plasmid, and cap loci on the pXO2 plasmid. However, pXO1/2 have been identified in other Bacillus species, such as B. thuringiensis and B. cereus. Some of these strains were isolated from humans with clinical disease. Other amplification targets (e.g., rpoB, gyrA, BA5510) present on the chromosome have been used, often in combination with primers for targets on the virulence plasmids. NAAT offers the advantages of excellent sensitivity, specificity, and rapid assay time but currently are not well-standardized and may only be useful for cultured organisms and clinical samples.

    Gamma phage lysis: The PlyG lysin of gamma phage mediates specific bacteriolysis of B. anthracis. Gamma phage lysis has high specificity but requires specialized resources and is used to identify B. anthracis in reference laboratories.

    Immunological methods: A variety of immunoassays are available for the detection of toxin components (PA, LF, EF). Some have been used to detect and quantify these proteins in clinical samples, such as serum, ascites, and pleural fluid. A direct fluorescence assay for both capsule and cell wall polysaccharide is used as a confirmatory test by reference laboratories. There are two immunoassays with FDA approval for in vitro diagnostic use: an enzyme-linked immunosorbent assay (QuickELISA™ Anthrax-PA Kit, Immunetics, Boston, MA) and an immunochromatographic test (RedLine Alert™, Tetracore, Rockville, MD). Immunetics reports a sensitivity of 100% and specificity greater than 99% for the ELISA-based assay; this assay detects anti-PA antibodies in serum. The Tetracore assay is for use on bacterial colonies grown on SBA and has a reported sensitivity of 98.6% and a specificity of 100%.

    B. anthracis grows rapidly (overnight) on sheep- or horse-blood agar, as well as brain-heart infusion agar, and other nutrient-rich agars.

    Culture techniques remain the gold standard with respect to the routine isolation and identification of B. anthracis from clinical specimens. The sensitivity of routine culture has not been prospectively compared to other techniques that may have greater sensitivity, such as NAAT.

  • PCR-based techniques are very helpful for the rapid detection of B. anthracis in both clinical and environmental samples. There are no FDA-approved commercially available NAAT platforms available to private laboratories. The Mayo-Roche Rapid Anthrax Test is a rapid-cycle real-time PCR (RT-PCR) test that utilizes LightCycler™ technology, with a reported sensitivity/specificity of 100/100%. Idaho Technology (Salt Lake City, UT) produces a RT-PCR system (Joint Biological Agent Identification and Diagnostic System, JBAIDS) for the Department of Defense that is highly portable and provides rapid detection capabilities for multiple biothreat agents, including B. anthracis. This system is designed for environmental detection. The CDC also utilizes RT-PCR technology for rapid identification of B. anthracis in clinical specimens, and this resource is available via the CDC Laboratory Response Network (LRN).

  • Matrix is another method for identifying the pathogen.

How does this organism cause disease?

  • Bacillus anthracis is not known to colonize hosts prior to infection.

  • Two main virulence factors are largely responsible for anthrax: the poly-D-glutamic acid capsule and the anthrax toxin.

    Capsule is required for full virulence in animal models and inhibits phagocytosis of bacilli.

    Anthrax toxin is composed of three separate proteins, known as protective antigen (PA), edema factor (EF), and lethal factor (LF).

    Anthrax toxin is classified as an AB toxin, with EF and LF as the active (A) moieties, and PA as the binding (B) component.

    LF is a zinc-dependent metalloprotease that cleaves most members of the mitogen activated protein kinase kinase (MAPKK) family, inactivating these kinases and preventing effective MAPK activation.

    EF is a Ca2+/calmodulin-activated adenylate cyclase that rapidly converts ATP into cyclic AMP (cAMP). Cyclic AMP is a second messenger that affects many cellular activities via interaction with membrane channels, activation of protein kinase A (PKA), and activation of Epac (exchange protein activated by cAMP).

    Effects of LT and ET: lethal toxin (LT) is the term used to describe the combination of PA+LF, whereas edema toxin (ET) is the term for PA+EF; LT and ET are artificial constructs used to experimentally evaluate the effects of LF separate from EF, as these two enzymes are present together during infection and intoxication of the host. See
    Table I.

    LT is lethal to many animal species when tested in experimental models, but the lethal dose 50 (LD50) varies among and within species. ET is also lethal in experimental animal models, but with a significantly increased LD50 compared to LT for most species.

Table I.
Induces vascular leakage via prostanoids, histamine, neurokinins Induces vascular leakage via ketotifen-sensitive mechanism
Inhibits macrophage cell cycle at G1/G0 interface Induces macrophage and endothelial cell apoptosis
Inhibits platelet aggregation Activates inflammasome
Severe hypotension Impaired cardiac contractility
Inhibits phospholipase A2 expression in alveolar macrophages Impaired glucocorticoid receptor trans-activation
Rapid lethality in mice Rapid lethality in Fisher rats, some murine strains
Impaired endothelial chemotaxis Impaired neutrophil oxidative burst, actin-based motility
Impaired cytokine production from dendritic cells Enhanced VCAM-1 expression on endothelial cells
Impaired neutrophil oxidative burst Reduced platelet aggregation
  • Other virulence factors that play a role in anthrax pathogenesis include:

    Anthrolysin O: a cholesterol-dependent cytolysin, role in bacillary escape from phagosomes, pore formation in membranes of leukocytes leading to cytolysis

    BslA: S-layer adhesion, critical role in adhesion/invasion of epithelial cells, endothelial cells of blood-brain barrier

    Secreted neutral proteases (Npr599, InhA): proteolytic cleavage of vWF, degradation of tight junctions in blood-brain barrier

    Phosholipases (A, B, C): escape from the phagolysosome

    bNOS (for bacterial nitric oxide synthase): critical for NO-induced bacterial catalase expression, inhibition of Fenton reaction, favors survival of germinating spore within macrophages

    Iron acquisition systems: B. anthracis has several mechanisms for acquiring iron, a nutrient absolutely required for survival and growth. Petrobactin and bacillibactin are siderophores, whereas IsdX1 and IsdX2 are secreted hemophores.

  • Different virulence factors favor survival, dissemination, and replication of B. anthracis at different stages during the infection:

    Spore germination: bNOS and anthrax toxin both inhibit macrophage-mediated destruction of germinating spores. Petrobactin is needed at this stage for early iron acquisition from the macrophage.

    Bacillary escape from phagocytes: facilitated by ALO, phospholipases C

    Prevention of bacillary phagocytosis: Capsule formation significantly impairs opsonization and phagocytosis of bacilli, whereas neutrophil function (phagocytosis, oxidative burst, chemotaxis) is severely impaired by anthrax toxin. ALO may lyse paralyzed leukocytes, also preventing phagocytosis.

    Dissemination: Bacilli replicate rapidly in blood, utilizing iron acquisition factors to mobilize host iron for growth. Anthrax toxin and secreted neutral proteases facilitate movement of bacteria in and out of the vasculature by inducing vascular leakage. BslA mediates adhesion of bacilli to cell surfaces.

    Hemorrhage: Anthrax toxin impairs platelet aggregation via multiple mechanisms, whereas Npr599 and InhA cleave vWF and ADAMTS13, processes that favor coagulation factor consumption and poor platelet-fibrin clot formation and maturation.

    Death: The exact mechanism underlying host death via anthrax remains unclear, but many processes contribute. These include depression of cardiac contractility, hypotension, loss of preload due to vascular leakage and hemorrhage, reduced organ perfusion with decreased oxygen delivery, and organ dysfunction. Massive bacillemia may contribute to poor organ perfusion by increasing blood viscosity.

  • A general note regarding treatment recommendations for anthrax: No human controlled clinical trials exist, nor will these trials ever be conducted due to the nature of anthrax as a highly mortal, but uncommon human infection. As a consequence, treatment recommendations are based largely on expert opinion and data derived from animal models, including studies in non-human primates.

WHAT’S THE EVIDENCE for specific management and treatment recommendations?

(The authors of this source used broth microdilution and time-kill studies to investigate the antimicrobial susceptibility of 7 B. anthracis isolates to fluoroquinolones (gatifloxacin, levofloxacin), penicillin, meropenem, and rokitamycin. Using Staphylococcus aureus breakpoints, all of the isolates were susceptible to these antimicrobials. All of these drugs were bactericidal at 24 hours of incubation and concentrations >4xMIC. Levofloxacin was bactericidal at 1xMIC at 24 hours and at 4xMIC at 12 hours.)

(The authors of this source used broth microdilution and time-kill studies to evaluate the activity of a large number of antimicrobials against two attenuated strains of B. anthracis (B. anthracis Sterne and the vaccine strain ST-1). They found that ciprofloxacin had a MIC of 0.03mg/L for both strains. Ceftriaxone was the least active antimicrobial, with a MIC of 8.0mg/L. Three drugs (quinupristin/dalfopristin, rifampicin, and moxifloxacin) demonstrated very rapid killing, whereas chloramphenicol failed to kill these strains.)

(The authors of this source studied 40 clinical isolates from anthrax cases in Turkey using agar dilution. They observed the following MICs: ciprofloxacin 0.06mg/L, levofloxacin 0.12mg/L, doxycycline 0.03mg/L, and penicillin G 0.016mg/L.)

(The authors of this source induced broad antimicrobial resistance in two B. anthracis strains (Sterne, ST-1) via serial passage with increasing concentrations of antimicrobials. This included increased MICs for fluoroquinolones, tetracyclines, beta lactams, vancomycin, clindamycin, linezolid, macrolides, and quinupristin/dalfopristin.)

(The authors of this source utilized broth microdilution and time-kill studies to examine combination versus individual antimicrobial activity against the Sterne and ST-1 strains. They used fractional inhibitory concentrations (FICs) to determine synergistic, indifferent, or antagonistic combination effects. The only tested combination found to have synergistic activity for both strains was rifampin PLUS clindamycin, whereas telithromycin PLUS amoxicillin demonstrated synergy versus the Sterne strain only. Combinations of ciprofloxacin and another drug (tetracycline, penicillin G, clarithromycin, clindamycin, quinupristin/dalfopristin, rifampin, or linezolid) demonstrated indifference, with no antagonism for either strain.)

(The authors of this source used a hollow-fiber model to mimic human pharmacokinetics and examine differences in activities of antimicrobials against spore-forming and non-spore-forming B. anthracis strains, since current antimicrobials are only active against vegetative bacilli and these are the toxin-synthesizing forms responsible for disease.)

(The authors of this source examined serum antimicrobial concentrations in rhesus monkeys being treated for experimental inhalational anthrax. They found serum peak/trough concentrations as follows: penicillin 2.7/0.8 mg/L, doxycycline 1.31/0.26mg/L, and ciprofloxacin 1.22/0.14mg/L. These values exceeded the MICs for 90% of the infecting strains.)

(The authors of this source exposed rhesus monkeys to about 8 LD50 of Ames strain spores via aerosol, followed by prophylaxis with antimicrobials (penicillin, ciprofloxacin, doxycycline), vaccine alone (days 1 and 15 postexposure), or doxycycline PLUS vaccine. Antimicrobials were started on day 1 after exposure and continued for 30 days. Mortality: 90% of untreated controls died; 80% of "vaccine only" monkeys died; penicillin 30% died; ciprofloxacin 11% died; doxycycline 10% died; doxycycline PLUS vaccine 0% died.)

(The authors of this source infected C57BL/6J mice subcutaneously with LD90 dose of Sterne strain spores and treated at various time points with doxycycline. Any delay in doxycyline administration beyond time 0 of the infection resulted in mortality, with the percent mortality increasing with the delay.)

(The authors of this source exposed guinea pigs to fully virulent spores (Vollum strain or ATCC 6605) via the intranasal route, then initiated post-exposure prophylaxis 24 hours later using one of several antimicrobials: ciprofloxacin, tetracycline, erythromycin, cefazolin, or trimethoprim/sulfamethoxazole (TMP/SMX). Neither cefazolin nor TMP/SMX prevented death; all of the other antimicrobials were 100% effective at preventing death during treatment (total of 14 days). All of the erythromycin-treated guinea pigs died after stopping this drug; about 20% of the tetracycline group survived after cessation; 50-100% of the ciprofloxacin group survived, depending on drug dose. Extension of antimicrobial therapy to 30 days did not fully protect guinea pigs from death after discontinuation of ciprofloxacin or tetracycline, but addition of a PA-based vaccine to antimicrobials provided full protection after discontinuation.)

(The authors of this source infected DBA/2 mice with 10 LD50 of Sterne strain spores intraperitoneally, then treated them with ciprofloxacin (50mg/kg subcutaneously daily on days 1-9 postinfection) +/- anti-PA antibody. Survival at day 10 post-infection was about 60% with ciprofloxacin alone, about 30% with antibody alone (untreated about 25%), but was 100% with the combination.)

(The authors of this source infected BALB/c mice via aerosol with 50-100 LD50 of fully virulent Ames strain spores then treated the mice with ciprofloxacin (30mg/kg every 12 hours) or doxycycline (40mg/kg every 6 hours) for either 14 or 21 days. Sixty-day survival for 14 days therapy was 80% for ciprofloxacin, 90% for doxycycline; for 21 day therapy, ciprofloxacin 100%, doxycycline 70%. Mortality for untreated mice was 100%.)

(The authors of this source studied two groups of rhesus monkeys exposed to a mean aerosol Ames strain spore dose of 442 LD50. Group 1 began ciprofloxacin therapy 1-2 hours after exposure (post-exposure prophylaxis, PEP), whereas group 2 received ciprofloxacin for 10 days after becoming bacteremic (the infection group). Group 1 monkeys had 100% survival while receiving ciprofloxacin, but only 2 monkeys survived after discontinuation of ciprofloxacin. In group 2, 30% of the monkeys died during ciprofloxacin therapy, but the remaining 70% survived, even after discontinuation.)

(The authors of this source exposed New Zealand White rabbits to 150 LD50 Ames strain spores via aerosol and treated them with 5 days of levofloxacin at intravenous doses of 12.5mg/kg or 25mg/kg daily. Treatment was started on detection of PA antigenemia, resulting in 70 and 90% survival at day 28 post-infection.)

(The authors of this source observed that treatment of bacteremic (≥105 cfu/mL) guinea pigs or rabbits with ciprofloxacin or doxycyline cured these animals, and addition of anti-PA antibodies to antimicrobial therapy cured animals with high level bacteremia (about 106 cfu/mL).)

(The authors of this source exposed common marmosets to 100 LD50 Ames strain spores via aerosol then treated them with ciprofloxacin 17.5mg/kg po twice daily for 10 days, beginning on the day of exposure. Survival was 100% during antimicrobial therapy, but 33% of the primates died after discontinuation.)

(The authors of this source examined the efficacy of post-exposure prophylaxis using three different rabbit antisera (anti-LF, anti-PA, and anti-Sterne strain) in guinea pigs exposed to 25 LD50 of Vollum strain spores. Rabbit anti-PA sera provided a dose-dependent survival benefit when given up to 24 hours pre- or post-exposure, with the degree of protection at each dose declining rapidly with further delays in administration post-exposure.)

(The authors of this source investigated the protective efficacy of a human monoclonal anti-PA antibody (5H3) in a lethal toxin-infused rat model of anthrax sepsis. 5H3 was administered at various points after the initiation of a 24-hour LT infusion intravenously in Sprague-Dawley rats, and hemodynamic parameters were monitored. 5H3 improved mean blood pressure when compared to placebo at all points studied, but the benefits of 5H3 diminished with increasing delay from the start of LT infusion.)

(The authors of this source investigated the efficacy of raxibacumab, a human anti-PA IgG1λ monoclonal antibody, as a prophylactic or therapeutic agent in rabbits and cynomolgus macaques after aerosol exposure to fully virulent B. anthracis. Raxibacumab protected both rabbits and monkeys in a dose-dependent manner from death at 14 days (rabbits) or 28 days (monkeys) when given prophylactically for aerosol exposure to 100 LD50 spores. Raxibacumab also provided a modest survival benefit to both rabbits and monkeys when given as treatment (with no concurrent antimicrobial) at the first signs of systemic illness. This benefit was dose-dependent and most pronounced at an intravenous dose of 40mg/kg once. This dose was studied in healthy human volunteers and produced no significant adverse effects; the mean half-life of raxibacumab in human volunteers was 20.44+/-6.46 days.)