Wound botulism associated with subcutaneous drug useBMJ 2002; 325 doi: https://doi.org/10.1136/bmj.325.7371.1020 (Published 02 November 2002) Cite this as: BMJ 2002;325:1020
- A F A Merrison, specialist registrar ()a,
- K E Chidley, specialist registrarb,
- J Dunnett, consultantb,
- K A Sieradzan, consultanta
- a Department of Neurology, Frenchay Hospital, Bristol BS16 1LE
- b Department of Anaesthetics, Frenchay Hospital
- Correspondence to: A F A Merrison
- Accepted 22 January 2002
Wound botulism in drug users is a rare but important cause of weakness and neuropathy
Botulism has long been considered a foodborne infection, the causative agent, Clostridium botulinum usually being transmitted in preserved foods. Yet C botulinum is widely distributed in soil and may be transmitted through wounds. An important distinction is that infections from food are limited to the amount of toxin ingested whereas in infections from wounds the toxin can be produced in situ until infection is eliminated from the wound.1
A new mode of presentation is occurring in drug users, the organism entering at sites of subcutaneous injection. The first reported case due to subcutaneous injection was in New York in 1982.2 As drug use has increased so has the number of people exposed to C botulinum. Over 90% of cases of wound botulism have been reported in the United States (75% in California), with only a small number of cases in the United Kingdom and elsewhere in Europe. 1 3–6
Botulism manifests as an acute descending paralysis, with involvement of autonomic and cranial nerves. C botulinum produces a potent exotoxin that binds irreversibly to the presynaptic membrane causing failure of transmission at the neuromuscular junction, autonomic ganglia, and parasympathetic nerve terminals. Diagnosis is based on clinical and neurophysiological findings, serology, and by identifying the organism or toxin. An antitoxin is available, but treatment remains largely supportive. We report a case of wound botulism in a subcutaneous heroin user who posed diagnostic difficulties.
A 50 year old man presented with a two week history of a flu-like illness and four days of dysarthria, dysphagia, shortness of breath, and neck discomfort. On the day of admission he developed diplopia and generalised weakness and collapsed at home. He was an intravenous and subcutaneous drug user (mainly heroin), and he was positive for hepatitis C.
In the emergency department he was restless, and his verbal response was indistinct. He had generalised weakness, depressed reflexes, and flexor plantars. Within 12 hours he was hypotensive, oliguric, tachypnoeic, increasingly drowsy, and confused. Within 18 hours arterial blood gases showed severe respiratory failure. He was intubated, ventilated, and transferred to the intensive care unit.
Neurological examination 48 hours later showed that he was conscious and able to obey commands but profoundly weak. He had complete ptosis, limitation of eye movements in all directions, and nystagmus on lateral gaze. His pupils were 4 mm in diameter and responded sluggishly to light. He had global facial weakness and severe bulbar dysfunction. He had symmetrical, mainly proximal, weakness with no movement at the shoulder, hip, or knee. Reflexes were depressed and plantars flexor.
Investigations showed a neutrophilia, increased concentration of C reactive protein, and normal routine biochemistry. Blood cultures grew Staphylococcus aureus. A drug screen was positive for opiates and benzodiazepine. A chest x ray film showed consolidation of the right lower lobe. An echocardiogram appeared normal. Computed tomography and magnetic resonance imaging of the head gave normal results as did computed tomography of the cervical spine. Lumbar puncture showed an opening pressure of 12 cm of water (reference range 5-20 cm of water), no white cells, 17 red cells, normal glucose levels, protein 0.49 g/l (reference range 0.2-0.4 g/l), and negative bacteriology and virology results. A Tensilon test was weakly positive.
He was treated with antibiotics, thiamine, and B vitamins. Neurophysiological studies could not be performed on the intensive care unit where he was admitted, and it was not possible to transfer him to the regional neurosciences centre until 10 days after admission. He was therefore treated empirically with intravenous immunoglobulin, the presumed diagnosis being Miller-Fisher variant of the Guillain-Barré syndrome.
He remained systemically stable but unchanged neurologically. On day 4 several raised, erythematous areas were noted on his buttocks.
On day 10 he was transferred to the neurosciences centre. Neurophysiological studies showed normal sensory responses, noticeably attenuated motor responses, normal conduction velocity, sparse fibrillation potentials on needle electromyelography, and no decrement or facilitation on repetitive stimulation. These are classic neurophysiological findings in botulism.7 Botulinum antitoxin was given.
On day 14, ultrasonography confirmed abscesses in his buttocks; these were surgically drained. On day 18 C botulinum and its toxin were identified in the blood and abscesses. Further antibiotics were given.
Seven months later there was noticeable improvement in limb power, but he still needed help getting from bed to chair. He had partial ptosis and limitation of eye movements. Moderate facial and bulbar weakness remained. He required minimal ventilatory support (continuous positive airway pressure ventilation) through a tracheostomy for 10 months. He is now able to function independently.
Botulism, in particular botulism due to wounds, is rare. If a diagnosis of Miller-Fisher variant of the Guillain-Barré syndrome is considered but the pattern of weakness and neurophysiological findings do not support this then botulism should be considered. This is particularly relevant for drug users, and a search for wound sites should be undertaken.
In this case features that were atypical for the Guillain-Barré syndrome were the sluggish response of the pupils to light (pupils may be fixed in mid-position or fail to accommodate in 50% of cases) and the unusual nature of the ophthalmoplegia (particularly the severe ptosis; table). The moderately increased concentration of protein in the cerebrospinal fluid and the mildly positive Tensilon test also suggested an alternative diagnosis.
In support of a diagnosis of botulism, the weakness was predominantly proximal (most patients present with descending weakness), there was noticeable facial weakness, diplopia, and bulbar dysfunction. Reflexes were depressed (as in 50% of cases), and sensory and cognitive functions were preserved. Gastrointestinal disturbance was not a dominant feature as would be expected in foodborne but not wound botulism. Other features of botulism (not apparent here) include malaise, sore throat, fever, and autonomic dysfunction (cardiovascular, paralytic ileus, urinary retention).
Early recognition of botulism is essential. Patients receiving antitoxin early have shorter hospital stays and fewer days on ventilatory support.8 Wound sites should be cleared to prevent further in situ production of toxin. It is recommended that antitoxin be given before wound debridement to combat released toxin. Benzylpenicillin or metronidazole should also be given.
In California there are over 18 000 intravenous drug users. Of 102 cases of wound botulism in California between 1951 and 1998, 101 were in drug users.1 Risk factors identified in 26 cases studied between 1996 and 1998 were the use of black tar heroin (heroin containing impurities such as dirt and shoe polish) and subcutaneous but not intravenous injection. Unclean injection sites and the sharing of needles did not increase risk.3 Over 80% of C botulinum in the drug users was type A, the remainder type B.1 Type A is more aggressive than type B; in one study over 80% of patients required ventilatory support, compared with 40% of cases during the last outbreak of foodborne botulism in the United Kingdom.9 With improved intensive care services mortality has fallen to 7%, and the duration of hospital stay ranges from days to months. Neurophysiological changes may persist for years, and functional recovery may be prolonged.
Botulism is a life threatening disease. As the use of drugs (including subcutaneous use) increases in the United Kingdom there may be a concomitant increase in cases of botulism.
C Jewkes and F Joseph cared for the patient while in hospital.
Contributors: AFAM wrote the first draft of the paper; KEC, JD, and KAS contributed to revisions. JD first considered the diagnosis of botulism in this case.
Competing interests None declared.