Consensus Date: 27 October 2006
1 PHLN Summary Laboratory Definition
1.1.1 Definitive Criteriaa Detection of C. botulinum neurotoxin in clinical specimens from a patient with clinical features of the disease
b Detection of C. botulinum neurotoxin from food, consumed by patient with clinical features of the disease, before the onset of clinical botulism.
1.1.2 Suggestive CriteriaDetection of C. botulinum in faeces from a patient with clinical features of botulism is strongly suggestive of the diagnosis even if tests for neurotoxin are negative.
2 IntroductionBotulism is a paralytic disease in which blurred vision, slurred speech, difficulty swallowing, dry mouth and respiratory insufficiency are important features. The various forms of disease are caused by a neurotoxin and are often associated with food consumption. In developed countries the commonest form of disease is infant botulism in which the immature intestine is colonised by C. botulinum after ingestion of the organism. C. botulinum is widespread in the natural environment including soils. It can be found in indoor house dust and may be a natural contaminant of ingested foods; honey being implicated as a possible source in the USA in the 1970s.
Neurotoxin released by intestinal C. botulinum results in the clinical features of infant botulism. Classical botulism develops in older children and adults who consume toxin-contaminated foods, particularly those that have been inadequately processed or preserved. In developed countries these are often home processed or traditional foodstuffs. Botulism is recognised increasingly in adults and older children with intestinal colonisation by C. botulinum following, for example, intestinal surgery. The least common form of botulism is associated with C. botulinum contaminated wounds where the neurotoxin is released into the systemic circulation from the infected soft tissues in a manner similar to tetanus toxin. The differential diagnosis of botulism is limited and includes myasthenia gravis, Eaton-Lambert syndrome, tick paralysis and acute inflammatory polyneuropathy (esp. the Miller-Fisher variant). A specialist neurological opinion will help rule out these possibilities in cases of suspected botulism.
Clostridium species are anaerobic, spore-forming Gram positive bacilli. Several Clostridium species are capable of causing toxin-mediated disease. Botulism is usually caused by neurotoxin-producing strains of Clostridium botulinum. Other Clostridium species (C. baratii, C. noyvii and C. butyricum) can produce neurotoxin and may occasionally cause human disease. Neurotoxin-positive strains of C. botulinum subdivide into three subgroups, the first two of which are associated with human botulism, and the third of which only causes disease in non-human animals (Table 1). Group I is so closely related to C. sporogenes that it can only be distinguished by its ability to produce neurotoxin. Group III is related to C. noyvii. Group I and group II C. botulinum strains differ in that group I contains proteolytic strains.
Definitive laboratory confirmation of botulism is by detection of C. botulinum neurotoxin in clinical specimens from a patient with features of the disease or from food consumed before the onset of clinical botulism. Detection of C. botulinum in faeces from a patient with clinical features of botulism is strongly suggestive of the diagnosis even if tests for neurotoxin are negative. Very few Australian laboratories are able to provide both C. botulinum culture and neurotoxin detection.
Top of page
3 Laboratory Diagnosis of Botulism
3.1 Botulinum neurotoxin detectionA definitive diagnosis of botulism requires detection of C. botulinum neurotoxin (BoNT) in blood or faeces specimens from a patient with features of the disease, or in a specimen of food they consumed before the onset of clinical botulism. BoNT detection is by mouse bioassay and is not available in most clinical laboratories.
3.1.1 Specimen collection and transportBlood and faeces should be collected from the patient as soon after onset of symptoms as possible. The blood is drawn into a plain specimen tube without anticoagulant. Ideally, 15-20mL serum and 25-50g faeces should be collected. 2mL of serum should be obtained from infants and as much faeces as possible.
3.1.2 Specimen preparationa Serum: serum is separated and a series of 0.4mL aliquots dispensed, one for each toxin type and one control. A 0.1 mL aliquot of each antitoxin is then added to give a final volume of 0.5mL. Antitoxin to the commonest types (A, B and E) are used. Type F neurotoxin is very rarely encountered. Antisera for therapeutic use are unsuitable for the neutralisation test. As no shelf life has been recommended for antitoxin once reconstituted, immediate use is advised.
b Faeces: a portion of the faecal specimen is suspended in buffered gelatin. The supernatant is tested with and without antitoxin as in 2.1.2 (a) above after a completion of a multi-step centrifugation process. Non-specific toxicity of faecal supernatant may be seen at a dilution of less than 1:10.
c Other clinical specimens: such as tissue or exudate from suspected wound botulism, should be ground up with sterile mortar and pestle then treated as for faeces in (b) above
d Food: suspect food sources should be treated in accordance with Australian Standard 1766
3.1.3 Specimen inoculationThe 0.4 or 0.5mL specimen/antitoxin mixtures are each inoculated intraperitoneally into laboratory mice, and the animals observed closely over a 4 day period.
3.1.4 Results and interpretationAssessment of the effects of mouse inoculation must be carried out by staff with a valid laboratory animal handling license and experience of the mouse BoNT bioassay. A lethal result may take up to 4 days but neurotoxic effects are often evident to experienced staff after only one day. Serum from patients with acute inflammatory polyneuropathy can produce paralysis in mice. The non-specific toxicity produced in mice inoculated with faecal supernatant at lower dilutions may necessitate repetition of the bioassay with supernatant at higher dilution. BoNT toxicity is usually evident in the mouse bioassay at greater than 1:100 dilution.
3.1.5 TimelineThe mouse bioassay will take a minimum of 4 days to produce a negative result from a clinical or food specimen. Australian guidelines have recommended 5 days. A need to transport specimens, to obtain type-specific antitoxin, or titrate out non-specific mouse toxicity will lengthen the test period.
Top of page
3.1.6 Confidence limitsNo recent statistical evaluation of sensitivity or specificity of the BoNT mouse bioassay is available. The mouse bioassay has been found by some centres to be more sensitive than developmental toxin immunoassays and is thought to detect toxin in around 70% cases. A negative BoNT bioassay does not exclude a diagnosis of botulism in a patient with clinical features of the disease. Although theoretically very specific, the bioassay can produce mouse paralysis when sera from patients with acute inflammatory polyneuropathy have been inoculated.
3.2 C. botulinum isolationDetection of C. botulinum in faeces from a patient with clinical features of botulism is strongly suggestive of the diagnosis even if tests for neurotoxin are negative. BoNT is rarely detected in faeces of patients who have botulism and who are C. botulinum positive. Isolation of C. botulinum from toxin positive specimens can provide useful confirmatory evidence that can be kept indefinitely for subsequent investigation. However, isolation of C. botulinum from toxin negative food specimens is of questionable value given the widespread dissemination of C. botulinum spores in the inanimate environment.
3.2.1 Specimen collectiona Faeces: a portion of the faecal specimen collected for BoNT bioassay can be used for culture.
b Exudate or tissue should be sought from suspected wound botulism cases.
3.2.2 Culture mediaTwo tubes of cooked meat/glucose/starch broth should be inoculated. One should be heated to 80oC to kill vegetative bacteria, then both should be incubated anaerobically at 30oC for 4 days.
3.2.3 Neurotoxin screenThe culture medium is screened for BoNT using the mouse bioassay as described in 2.1 above.
3.2.4 Isolation of C. botulinumC. botulinum is isolated from enrichment media and directly from specimens by streaking onto selective egg yolk media (e.g. CBIA) and incubating the egg yolk plates at 37oC anaerobically. Typical colonies are small and demonstrate lipase but not lecithinase activity.
3.2.5 Confirmation of C. botulinumDefinitive confirmation of the identity of suspected C. botulinum colonies requires demonstration of BoNT production, since group I is biochemically and genetically identical to strains of C. sporogenes and group III closely resembles C. noyvii. The following phenotypic features would be taken as suggestive but not conclusive evidence of C. botulinum: Gram positive bacilli, with oval subterminal spores, anaerobic growth, and lipase activity. Some of the key distinguishing features are given in Table 1 below:
Table 1: Distinguishing features of C. botulinum strains
|C. botulinum||Toxin||Acid from GLU||Acid from MAL||Acid from LAC||Acid from SUC||Principal GLC peaks*||Minor GLC peaks*|
|Group I||A,B,F||+||-/w||-||-||Acetate, Butyrate, Isovalerate||Isobutyrate|
|Group II||C,D||+||V||-||-||Acetate, Propionate, Butyrate|
|Group III||B,E,F||+||+/w||-||+/w||Acetate, Butyrate|
* when grown on PYG or CMC
3.2.6 Confidence limitsThe taxonomic complexity of C. botulinum and related neurotoxin-positive Clostridia dictates that presumptive identification of Clostridium species from patients with suspected botulism should not be used as the sole justification for a diagnosis. The phenotypic and genotypic characteristics of C. botulinum are not specific to this species and raise important questions about its taxonomic status. Without demonstration of BoNT activity, an isolate cannot be considered to be confirmed C. botulinum. Moreover, variations in expression of BoNT activity may render some strains negative by mouse bioassay.
3.3 Quality controlAt present there is no external quality control programme for BoNT detection or C. botulinum isolation from clinical specimens.
3.3.1 Suitable internal controlsThe BoNT mouse bioassay requires a non-neutralised (antitoxin free) serum or faecal supernatant free positive control. A type culture strain of C. botulinum can be used as a positive control on/in primary isolation media, enrichment media and for subsequent confirmatory tests.
3.3.2 Special considerationsThe sporadic nature of botulism and, the technically demanding procedures required for its confirmation mean that there are very few Australian laboratories are able to provide a test service. Performance of the mouse bioassay requires approved animal handling facilities, ethical approval and staff with the appropriate animal handling certification i.e. a vivisection license. Supplies of type-specific diagnostic antisera are sporadic. Competent laboratories may therefore be unable to offer a test service for an unspecified period after using a batch of antisera for a botulism test series.
3.3.3 TimelineIn exceptional circumstances a preliminary positive has been reported positive in only two days. Completion of a set of C. botulinum neurotoxin and culture tests on matched clinical and food samples can take between 10 and 14 days. This period includes assembly of reagents, lengthy anaerobic culture and mouse bioassay procedures. Transportation of specimens, acquisition of type-specific antisera and repetition of mouse bioassay at higher dilutions can prolong the time taken to produce a reliable result.
4 References1 Anon. 1998. Botulism in the United Sates, 1899-1996. Handbook for Epidemiologists, Clinicians and Laboratory Workers. Centers for Disease Control and Prevention, Atlanta, GA.
2 Balows A, Duerden BI. Systematic Bacteriology. 1998. Topley & Wilson’s Microbiology & Microbial Infections. Vol. 2. Collier L, Balows A, Sussman M. Arnold, London, 1998.
3 Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken RH. 1999. Manual of Clinical Microbiology, ASM Press, Washington, D.C., USA.
4 Mandell GL, Bennett JE, Dolin R. 2000. Principles and Practice of Infectious Diseases. 5th Edition. Vol.2. Churchill Livingstone, Edinburgh, UK.