Consensus Date: March 2017
1 PHLN Summary Laboratory Definition
1.1.1 Definitive Criteria
- Isolation of Bacillus anthracis from blood, sterile sites, sputum, nasal swabs, wounds, intestinal contents or other clinical specimens.
1.1.2 Suggestive Criteria
- Detection of B. anthracis by nucleic acid test (NAT) covering the genes coding for capsule and virulence factors; OR
- Gram positive square-ended bacilli, non-motile and encapsulated (well shown with polychrome methylene blue stain), from blood or other normally sterile site.
Anthrax is one of the oldest recorded diseases of animals. Anthrax was the first disease of humans and other animals in which the causative agent was definitely demonstrated as a specific microorganism by the French biologist Casimir-Joseph Davaine in 1863 and in 1876 by the German bacteriologist Robert Koch, who isolated the organism in pure culture. It was also the first infectious disease against which a bacterial vaccine was found to be effective, by Louis Pasteur in 1881. These discoveries led to the origin and development of the modern sciences of bacteriology and immunology.
Anthrax is a zoonotic disease that is transmissible to humans through handling or consumption of contaminated animal products. The aetiologic agent of anthrax, Bacillus anthracis, is a spore forming gram-positive bacillus. B. anthracis produces central or paracentral spores without significant swelling of the bacilli. The bacilli stain Gram positive and are typically large and square-ended. They are 1.0 to 1.3 µm by 3.0 to 10 µm in size, and occur singly or in chains of two or three bacilli. B. anthracis spores can remain viable in soil for many years3.
Humans can become infected with B. anthracis by handling products or consuming undercooked meat from infected animals. Infection may also result from inhalation of B. anthracis spores from contaminated animal products such as wool or the intentional release of spores during a bioterrorist attack. Human-to-human transmission has not been reported. Three main forms of anthrax occur in humans: cutaneous, gastro-intestinal, and inhalational3.
Virulent B. anthracis are typically distinguished from closely-related soil Bacillus sp. by the presence of two virulence plasmids, designated pXO1 and pXO2. However in the last decade, B. cereus isolates carrying pXO1 and pXO2-like plasmids containing B. anthracis toxin and /or capsule synthesis genes have been isolated from severe cases of human and primate disease (references http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4892579/, http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594744/). The pXO1 plasmid contains the lef, cya and pag genes, which encode for lethal factor, oedema factor and the protective antigen, respectively. The pXO2 plasmid contains the cap gene, which encodes for the capsule6. Molecular tests aimed at rapidly confirming the identity of an isolate and determining whether the isolate is virulent or not target these genes in particular6. Following infection, the protective antigen combines with, the lethal factor or the oedema factor to produce the lethal toxin and the oedema toxin, respectively. The protective antigen allows the binding of the lethal and oedema factors to the cell membrane and facilitates their passage across the cell membrane where they mediate their toxic effect10.
A. Cutaneous Anthrax
Cutaneous anthrax is the most common form of anthrax and accounts for about 99% of natural occurring anthrax. Cutaneous infections occurs when the bacterium or spore enters a cut or abrasion on the skin, such as when handling contaminated wool, hides, leather or hair products (especially goat hair) from infected animals. The incubation period is between 1-7 days, but may be up to 12 days. Skin infection begins as a raised itchy bump or papule that resembles an insect bite. Within 1-2 days, the bump develops into a fluid-filled vesicle, which ruptures to form a painless ulcer (called an eschar), which is usually 1-3 cm in diameter, with a characteristic black necrotic (dying) area in the centre. Pronounced oedema is often associated with the lesions because of the release of oedema toxin by B. anthracis. Lymph glands in the adjacent area may swell. Approximately 20% of untreated cases of cutaneous anthrax result in death either because the infection becomes systemic or because of respiratory distress caused by oedema in the cervical and upper thoracic regions. Deaths are rare following appropriate antibiotic therapy, with lesions becoming sterile within 24 hours and resolving within several weeks2,3.
B. Gastrointestinal Anthrax
The gastrointestinal form of anthrax may follow the consumption of contaminated meat from infected animals and is characterised by an acute inflammation of the intestinal tract. Gastrointestinal anthrax is more common in endemic areas where people eat the meat of animals that have died suddenly. Gastrointestinal anthrax is essentially cutaneous anthrax occurring in the intestinal tract. Initial signs of nausea, loss of appetite, vomiting, and fever are followed by abdominal pain, vomiting of blood, and severe diarrhoea. The mortality rate is difficult to determine for gastrointestinal anthrax but is estimated to be 25%-60%2,3.
top of page
C. Inhalation Anthrax
This form of anthrax results from inhaling B. anthracis spores, and is most likely following an intentional aerosol release of B. anthracis. After an incubation period of 1-6 days (depending on the number of inhaled spores, can range from 2-43 days), disease onset is gradual and nonspecific. Fever, malaise, and fatigue may be present initially, sometimes in association with a nonproductive cough and mild chest discomfort. These initial symptoms, lasting 1-4 days, are often followed by a short period of improvement (ranging from several hours to days - note: this was not seen in the 2001 US mail inhalational cases), followed by the abrupt development of severe respiratory distress with dyspnoea (laboured breathing), diaphoresis (sweating), stridor (high-pitched whistling respiration), and cyanosis (bluish skin colour). Inhalational anthrax is often associated with gastrointestinal symptoms (see above) and confusion. Haemorrhagic meningitis may occur. Shock and death usually occur within 24-36 hours after the onset of respiratory distress, and in later stages, mortality approaches 100% despite aggressive treatment. Physical findings are usually nonspecific. The chest X-ray typically shows a widened mediastinum, due to lymphadenopathy, and pleural effusions, with or without infiltrates2,3,9.
D. Injection Anthrax
Since 2000, a new form of anthrax has been reported in injecting drug users, linked to use of contaminated heroin. Cases do not present with typical signs of cutaneous anthrax, rather soft tissue involvement around the injection site, often progressing to severe complications. Some cases reported gastrointestinal symptoms such as nausea, vomiting and abdominal pain, while a smaller number experienced neurological symptoms, meningitis, intracranial or subarachnoid haemorrhage. In many cases, anthrax was not suspected or diagnosed with delay, likely contributing to the eight fatalities from 18 published cases. (Reference http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20877).
Most B. anthracis strains are sensitive to a broad range of antibiotics. Penicillin, ciprofloxacin, or doxycycline (but not cephalosporins) are usually recommended for the treatment of anthrax. A combination of antibiotics may be more effective than a single antibiotic in severe disease. To be effective, treatment should be initiated early. If left untreated, the disease is almost always fatal2.
3 Laboratory Diagnosis/Tests
3.1 Culture of clinical specimens for Bacillus anthracis
B. anthracis can be detected by Gram stain of the blood and by blood culture with routine media. Blood cultures collected at the time of presentation or early in the course of illness, before treatment is started, are usually positive9. Only vegetative encapsulated bacilli are present during infection. Spores are not found within the blood, partially because CO2 levels in the body inhibit sporulation. In patients with neurological symptoms, cerebrospinal fluid (CSF) examination may reveal Gram positive bacilli. Culture must be confirmed by an appropriate Reference laboratory using PCR and phage lysis tests.
For optimal isolation of B. anthracis, direct plating onto sheep or horse blood agar (BA) is recommended. This may be direct from clinical specimens or from subcultures of blood culture medium.
In addition, it is recommended that clinical material is also inoculated into a suitable enriched broth, such as brain heart infusion broth or trypticase soy broth2.
For non-sterile site specimens, inoculation onto PLET (Polymixin-Lysozyme-EDTA-Thallous acetate) agar5 is recommended. In addition, cultures overgrown with Proteus sp. can be “cleaned up” by placing some of the subculture material in ether for a minute (approximately). A few drops of the ether suspension on blood agar and PLET will subsequently yield Proteus-free colonies of Bacillus anthracis.
3.1.2 Suitable specimens
Vesicular stage: The organism is best demonstrated in this stage. Gloves should be worn when collecting specimens. Soak two dry sterile swabs in vesicular fluid from a previously unopened vesicle.
Eschar stage: Rotate two swabs beneath the edge of the eschar without removing the eschar2. Alternatively, biopsy material may be submitted.
If the patient is able to produce a stool specimen, stool cultures should be performed.
In later stages of disease, blood cultures will yield the organism if specimens are obtained prior to antibiotic treatment. It is recommended that stool specimens are also inoculated onto PLET agar. In addition, ascitic fluid may yield positive cultures.
If respiratory symptoms are present and sputum is being produced, obtain a specimen for culture and smear.
In later stages of disease (2-8 days post exposure) blood cultures, CSF or pleural fluid may yield the organism, if specimens are drawn before antibiotic treatment2. B. anthracis may be detected directly in clinical specimens by Gram’s stain and/or polymerase chain reaction (PCR) if the diagnosis is strongly suspected. Specimens from internal sites, such as pleural fluid, are not recommended unless absolutely necessary for diagnosis because of the risk to operators collecting the specimen.
NB. Nasal swabs should only be used to support a confirmed exposure to B. anthracis spores, or during an ongoing epidemiological investigation.
3.1.3 Test sensitivity
No mathematical data available.
Depends on the quality and type of the specimen, the type of media chosen, and whether enrichment is used prior to direct plating. There may be enough organisms in the blood to see them on direct smears by Gram’s stain. B. anthracis appears as short chains of 2-4 cells which are encapsulated as evidenced by clear zones around the bacilli. The presence of large encapsulated Gram-positive rods in the blood is strongly presumptive for B. anthracis identification. Antibiotic treatment may inhibit the growth of B. anthracis.
3.1.4 Test specificity
No mathematical data available.
The isolation of encapsulated non-motile Gram-positive rods from a normally sterile site, or from respiratory/faecal samples with a suitable history, confirmed by specific biochemical, phenotypic or molecular techniques. The isolation of Bacillus anthracis is notifiable in all States and Territories of Australia.
top of page
3.1.5 Predictive values
A negative culture does not exclude the diagnosis of anthrax.
3.1.6 Suitable acceptance criteria
After incubation of BA (blood agar) plates for 15-24 hours at 35-37oC, well-isolated colonies of B. anthracis are 2-5 mm in diameter. The flat or slightly convex waxy colonies are irregularly round, with edges that are slightly undulated (irregular, wavy border), and have a ground-glass appearance. There are often comma-shaped projections from the colony edge, producing the "Medusa head" colony. B. anthracis grows best aerobically, but can grow anaerobically. Colonies on BA usually have a tenacious consistency. When teased with a loop, the growth will stand up like beaten egg white1,2.
On PLET agar5, the colonies are smaller (1-2 mm) and have a very distinctive appearance, with a central raised white core and an off-white undulate periphery. After 24 hours incubation, many colonies look like miniature spiral galaxies with an arm coming off from either side, but as the colony grows, they more closely resemble fried-eggs.
NB. PLET agar typically inhibits B. cereus strains.
3.1.7 Suitable internal controls
Properly documented, relevant, quality control program for each type and batch of medium used, e.g. growth of B. anthracis (Sterne strain) on each batch of PLET, with no growth of B. cereus.
3.1.8 Suitable test validation criteria
Culture-based isolation of B. anthracis, confirmed by biochemical, other phenotypic or molecular methods, is the gold standard.
3.1.9 Suitable external QC program
The Royal College of Pathologists Australasia, with support from the Commonwealth Department of Health, have initiated a Biosecurity Quality Assurance Program which covers both Bacillus anthracis and its near neighbours.
3.1.10 Special considerations
In contrast to colonies of B. cereus and B. thuringiensis, colonies of B. anthracis are not -b-haemolytic. However, weak haemolysis may be observed under areas of confluent growth in ageing cultures and should not be confused with b-haemolysis. B. anthracis grows rapidly; heavily inoculated areas may show growth within 6-8 hours and individual colonies may be detected within 12-15 hours. This trait can be used to isolate B. anthracis from mixed cultures containing slower-growing organisms.
3.1.11 Analysis of suspicious powders
A separate protocol has been developed to describe the methodology for the analysis of suspicious powder samples received as a result of a hoax or bioterrorist action for B. anthracis7.
NB. It should be noted that material received as part of a suspected bioterrorism incident will likely consist of a powdered suspension of spores. Whilst such material may be cultured overnight in broth, or directly onto blood agar plates to provide a presumptive result in 24 hours, it presents some difficulties for direct detection by PCR because of the difficulties in extracting DNA from the spore. Testing for spores cannot be validated in Australia because of the unavailability of suitable powdered spore preparations.
3.2 Identification of Bacillus anthracis
There are two levels of action in the identification of Bacillus anthracis:
- Diagnostic laboratories - B. anthracis will generally be detected in routine clinical laboratories first unless all samples are directly diverted to a public health laboratory because of prior notification of a hoax or suspected biological attack. B. anthracis is a Risk Group 3 organism, and diagnostic laboratories should immediately refer all cultures and material to a public health (PHLN) laboratory as soon as there is any suspicion of B. anthracis, and institute appropriate containment and decontamination measures in the laboratory.
Reference laboratories - State or Territory reference laboratories (PHLN) have the appropriate levels of containment for identification, confirmation and molecular characterisation of isolates of B. anthracis.
- There are specific protocols in place for the transfer of clinical material and cultures containing suspect Security Sensitive Biological Agents (SSBA) between diagnostic laboratories and public health reference laboratories. These are regulated under Commonwealth laws relating to SSBA.
3.2.1 Conventional biochemical tests
188.8.131.52 Suitable specimens
A pure culture on solid medium (blood agar or nutrient agar slope).
top of page
Polychrome Methylene Blue (MacFaydean’s stain) is the single easiest test available for presumptive confirmation of B. anthracis in blood films and exudates. The B. anthracis cells stain blue with an obvious thick capsule which stains red surrounding the cells. This stain may also be used to demonstrate capsule formation in cultures grown under 5% CO2 atmosphere on bicarbonate agar.
Virulent strains of B. anthracis are the only organisms that produce rough colonies when grown aerobically and mucoid colonies when grown on a 0.8% sodium bicarbonate medium under a 5% CO2 atmosphere. If the plates are re-incubated aerobically after mucoid colonies have formed, rough outgrowths from the margins will appear, showing reversion to the rough form.
184.108.40.206 Test sensitivity
The test sensitivity will depend on the tests performed. The demonstration of medusa-head colonies with a tenacious consistency, Gram-positive staining bacilli, smooth colonies on 0.8% bicarbonate agar incubated in 5% CO2, and characteristic staining with polychrome methylene blue stain are traditionally considered to be reliable characteristics for confirmation of a B. anthracis isolate.(Table 1) B. anthracis is non-motile in motility medium (note that Bacillus cereus var. mycoides will also give a non-motile reaction). Motility can be tested by inoculating growth from a culture into Trypticase Soy Broth (TSB) for 2 hours at 37oC8. Characterisation of isolates by other phenotypic tests is unreliable and increases the hazard of exposure of laboratory workers.
||Colonies exhibit “tenacity” when teased up with a loop
||Gram positive bacilli with spores - look like railway carriages end-to-end
|Polychrome Methylene Blue stain
||Blue bacilli and spores surrounded by thick red capsule
|Growth on Blood Agar
||No haemolysis in young cultures
|Growth on Bicarbonate agar in CO2
||Smooth mucoid colonies
|Gamma phage lysis
||Inhibition of growth
220.127.116.11 Test specificity
No mathematical data available. The traditional phenotypic characteristics described in 18.104.22.168 may not satisfactorily identify some Bacillus spp resembling B. anthracis, nor do those tests assess the virulence capacity of B. anthracis strains so molecular identification methods are required for final confirmation. Laboratory personnel should be particularly aware that the Sterne strain does not produce a capsule and therefore will not give a typical reaction on Bicarbonate agar.
22.214.171.124 Predictive values
A negative result in one test does not preclude a diagnosis of B. anthracis. Further testing as outlined above may be necessary to confirm the identification. Isolates should conform to the major criteria outlined above.
126.96.36.199 Suitable test criteria
An isolate that exhibits phenotypic and biochemical characteristics consistent with documented reactions for B. anthracis. The major distinguishing features of B. anthracis are typical reaction in polychrome methylene blue stain, non-motility, colony tenacity and production of mucoid smooth colonies on bicarbonate agar incubated in CO2 but rough colonies when incubated aerobically.
Identification can also be confirmed by fatty acid analysis, using the Maldi-TOF(Bioterrorism panel), (requires the security relevant reference library for proper ID. Labs should take care as the normal reference library can result in B.a being identified as Bc. Reference http://jcm.asm.org/content/54/3/764.full. B. anthracis and B. cereus still need additional tests to resolve, even with the biosecurity database.) and gamma phage lysis. There is some data to suggest that the standard Bruker prep will not reliably decontaminate anthrax spores. Specific preps for PC3 level organism are available. Drevinek et al. Letters in Applied Microbiology, 55, 40-46. 2012, Weller et al. PLOS ONE, December 2015.
top of page
188.8.131.52 Suitable internal controls
Test each batch of biochemical substrates with positive and negative control strains.
Results of all testing recorded and the records maintained.
184.108.40.206 Suitable validation criteria
Correct biochemical and phenotypic reactions exhibited by a standard B. anthracis strain.
220.127.116.11 Suitable external QC program
RCPA Biosecurity QAP.
18.104.22.168 Kits/automated systems for biochemical identification
3.3 Molecular Identification and Typing
PCR testing typically relies upon the demonstration of a gene found in the B. anthracis chromosome (but not specific for only B. anthracis) and 2-4 plasmid genes that code for the virulence factors (capsule, protective antigen, lethal factor and oedema factor6) which can be multiplexed. This test can confirm that an isolate is B. anthracis and also distinguish between virulent and non-virulent (i.e. non-capsulated strains lacking pXO2 or non-toxin producing strains lacking pXO1) strains. It should be noted that because there can be a difference in the copy number of the two plasmids (typically pXO1 has been shown to have higher copy numbers per cell than pXO2), a PCR test which shows only one plasmid positive in a multiplex test should be followed up with a singleplex test on all targets. This is particularly important when investigating a potential bioterrorist attack. This molecular assay has only been validated on cultures. Due to the rarity of anthrax cases, this assay has not been validated for direct detection of spores in environmental samples or detection of bacilli in clinical samples.
There have now been a number of reports of Bacillus sp (not B anthracis) possessing either the pXO1-like or pX02-like virulence plasmids. As a result, it has become even more important to use a gene target that will definively identify a Bacillus isolate as Bacillus anthracis. Several chromosomal loci have been suggested, including ,BA813, BA5345, BA5510, rpoB, gvrA,gyrB, saspB, plcR and rRNA genes.
Typing of strains is best performed by whole genome sequencing given the high resolution necessary for distinguishing across the low genetic variability seen in B. anthracis strains. This would be particularly useful if multiple isolates were recorded from different States or Territories at the same time.
Molecular typing of B. anthracis has traditionally been preformed by Multi-locus VNTR analysis (MLVA) which uses the combined power of multiple alleles at several marker loci4. It allows individual strains of anthrax to be assigned to particular genotypes, making geographic tracing a possibility if the strain is a unique genotype associated with one particular area of the world. In addition, although quite complex and time consuming, the methodology is reproducible at different laboratories, enabling rapid comparison of results. Typing is now best performed by whole genome sequencing given the high resolution necessary for distinguishing across the low genetic variability seen in B. anthracis strains. In the event of a number of Anthrax cases in Australia however, cultures would be sent to the Center for Disease Control (CDC) for comparison against their large collection of B. anthracis strains.
The Department of Health has made RAMP (Rapid Analyte Measurement Platform) readers and kits available to a number of PHLN laboratories. These kits are specifically designed to enable rapid presumptive confirmation of anthrax spores. As a result, the usual sample for analysis is white powder preparations received as part of a threat scenario. These kits should not be used on clinical material, firstly because of the lack of spores in clinical material, and secondly because of the danger of false positives caused by some clinical samples, eg. Urine.
Some PHLN labs have access to the GeneXpert system or the Biofire Film Array which can be used with a specific bioterrorism panel to provide rapid molecular confirmation of Bacillus anthracis from powdered spore suspension.
Although serology for B. anthracis is offered in a few overseas laboratories, there are no laboratories in Australia offering this testing. Where serology is considered an option, paired sera are preferred. These should be forwarded to the jurisdictional PHLN laboratory who will undertake to send them to the CDC for testing.
3.5 Immunofluorescence on Tissue
The Centres for Disease Control, Atlanta, Georgia criteria do not include immunofluorescence as a confirmatory test on clinical material, but Direct Fluorescent Antibody reagents are considered to provide confirmatory evidence of anthrax in cultures. However, DFA reagents for both capsule and cell wall polysaccharide are no longer available in jurisdictional PHLN laboratories
top of page
- Murray,P.R., Baron,E.J., Pfaller,M.A., Tenover,F.C. and Yolken,R.H. 1999. Manual of Clinical Microbiology, ASM Press, Washington, D.C
- Lindsey KQ, Morse SA. (eds). 2001. Basic Laboratory Protocols for the Presumptive Identification of Bacillus anthracis. [Online, accessed 23 Jul. 2001] URL: (http://www.cdc.gov/)
- Turnbull PCB. Guidelines for the Surveillance and Control of Anthrax in Humans and Animals. 3rd edition. (1998). World Health Organisation. Department of Communicable Disease Surveillance and Response.
- Keim P, Price LB, Klevytska AM, Smith KL, Schupp JM, Okinaka R, Jackson PJ, Hugh-Jones ME. “Multi-Locus Variable-Number Tandem Repeat Analysis Reveals Genetic Relationships within Bacillus anthracis”. Journal of Bacteriology. May 2000, Vol 182, No. 10., 2928-2936.
- Knisely RF. “Selective Medium for Bacillus anthracis”. Journal of Bacteriology. Sept. 1966. Vol 92. No. 3. 784-786.
- Ramisse V, Patra G, Garrigue H, Guesdon J-L, Mock M. “Identification and characterisation of Bacillus anthracis by multiplex PCR analysis of sequences on plasmids pX01 and pX02 and chromosomal DNA. FEMS Microbiology Letters. 145. 1996. 9-16.
- PHLN Document - White Powder Protocol 05/04/2002.
- Papaparaskevas, J., Houhoula, D.P., Papadimitriou, M., Saroglou, G., Legakis, N.J. & Zerva,L. “Ruling Out Bacillus anthracis”. Emerging Infectious Diseases. April 2004. Vol. 10, No. 4, 732-735.
- Jerrigan JA, Stephens DS, Ashford DA et al. Bioterrorism-related inhalational anthrax: the first 10 cases reported in the United States. Emerg Infect Dis 2001;7:933-944.
- Inglesby, T., O’Toole, T., Henderson, D.A et al. Anthrax as a Biological Weapon, 2002 - Updated Recommendations for Management. JAMA. 2002;287:2236-2252.