Introduction
Heroin, cocaine and amphetamines are among the most widely injected drugs,
and complications in injecting drug users (IDUs) resulting in infections
are the most frequent reason for admission to hospital in this group
of patients [1]. Soft tissue infections caused by spore-forming bacteria
in IDUs emerged as a serious problem in the UK in 2000. Cases of infections
due to Clostridium novyi [2], Clostridium botulinum [3,4], Clostridium
tetani [5], Clostridium histolyticum [6], and Bacillus
cereus [7] were subsequently reported in this patient group. The
major risk factors for all these infections was thought to be the availability
of higher purity heroin, and ‘skin-popping’ (subcutaneous
injection) or ‘muscle-popping’ (intramuscular injection)
which is sometimes practised by IDUs when access to veins is lost [2,3,5].
A larger amount of an acidulant, such as citric acid, may be needed to
make higher purity heroin soluble for injection; this is likely to increase
the resulting tissue damage when subcutaneously or intramuscularly injected,
and is thus important for the initiation of a wound infection.
Wound botulism occurs when spores of C. botulinum contaminate
a wound, germinate and produce botulinum neurotoxin in vivo. The symptoms
of botulism are caused by the neurotoxin which blocks the release of acetylcholine
at the neuromuscular junction, resulting in a descending flaccid paralysis.
Patients with botulism typically present with blurred vision, drooping
eyelids, slurred speech, difficulty in swallowing, dry mouth, and muscle
weakness. Patients usually have no fever or loss of sensation and awareness.
If untreated, paralysis may progress to the arms, legs, trunk and respiratory
muscles. If onset is very rapid there may be no symptoms before sudden
respiratory paralysis [8].
Methods
Cases of wound botulism were defined, as outlined elsewhere [9], as illness
resulting from toxin produced by C. botulinum that has infected a wound
producing symptoms including diplopia, blurred vision, bulbar weakness
and symmetric paralysis. Laboratory confirmation was obtained by the detection
of botulinum neurotoxin in serum or wound tissue and/or the isolation of C.
botulinum from a wound [9].
In the United Kingdom (UK), cases of botulism are reported through national
voluntary reporting to the Health Protection Agency (HPA) Centre for Infections
(CfI) and by submission of samples for laboratory confirmation to CfI,
which also receives referred samples from the Republic of Ireland. Laboratory
confirmation is achieved as described elsewhere [10,11,12]. Further clinical
details from affected patients are obtained by administration of a standard
questionnaire to patients by clinicians and microbiologists.
Results
Yearly totals of reports of wound botulism by country in the UK and Republic
of Ireland are shown in Figure 1. No cases were recognised before 2000
and a total of 88 cases were reported between 2000 and 2004. Seventy
cases were in England, 12 in Scotland, 2 in Wales and the remaining
4 in the Republic of Ireland. No cases were reported from Northern
Ireland. All cases occurred in IDUs. The ages were known for 75 of
the 88 cases, and the mean age was 34 years (range 22 to 48). Sixty
one of the cases were in men and 27 in women: in 2004, where information
on gender was provided, 27 were in men and 13 were in women. Details
of clinical presentation, outcomes and drug use will be presented elsewhere
as data collection is ongoing.
Overall, 40 (45%) of the 88 cases were laboratory confirmed by the detection
of botulinum neurotoxin in serum (33 cases), or by the isolation of C. botulinum from
wounds (25 cases). Neurotoxin was detected in serum together with the
isolation of C. botulinum from wounds in 18 of the cases. Neurotoxin
only was detected in the serum of 15 of the cases, and C. botulinum only
was isolated from wounds in the remaining seven cases. Based on the neurotoxin
detected and/or the C. botulinum isolated from the 40 laboratory
confirmed cases, 35 were due to type A, three to type B and two to types
A and B.
During 2004, 36 of the 40 cases reported were in England. Twelve of the
patients in England were laboratory confirmed, and 10 of these cases
were due to type A, one to types A and B, and one to type B. There
was some geographical clustering. with cases concentrated in two regions:
Yorkshire and Humberside, and London [Figure 2].
Discussion
The recognition of wound botulism in injectors of heroin in the UK coincided
with increased recognition of soft tissue infections due to other species
of endospore forming bacteria [2,5,6,7]. It is not clear if the emergence
of these diseases represents the presentation of new diseases in the
UK, or is due to ascertainment bias because of diagnosis of diseases
not previously recognised. However, the recognition of cases of food
and infant botulism together with surveillance systems to capture reports
of cases clearly existed in the UK before the detection of the first
cases in IDUs in 2000. This suggests that, at least for botulism, these
soft tissue infections represent an emerging hazard for this patient
group. There was increased recognition of wound botulism in IDUs in California
in the mid-1990s [13]. The emergence of wound botulism in IDUs in the
United States and the UK may have resulted in part from better recognition
of cases and increased medical surveillance of this group, bit this is
unlikely to be the only explanation for the increase in reported cases.
The outbreaks may also have been due to contamination events of specific
batches of heroin, or they may reflect changes in drug composition or
purity. The availability of ‘black tar’ heroin in the United
States (which differs to that generally used in the UK) was identified
as a contributing factor for the Californian outbreak [13]. No explanation
could be found for the clustering of the wound botulism cases in 2004.
However, this clustering together with an absence of cases in other areas
believed to have high prevalence of IDUs (such as Glasgow and the north
west of England) supports the hypothesis that there was a causal relationship
between the patients. Clustering of cases had not previously occurred
in the north east of England.
A small number of wound botulism cases in IDUs has been reported in several
other European countries. The first cases were reported in Norway in
1997 [14], followed by at least three further cases [15,16]. Between
September 1998 and February 1999, nine cases of wound botulism in IDUs
were identified in Switzerland [17-22], and one in Holland [23]. The
authors have been unable to locate additional case reports amongst IDUs
from other European countries.
Since a major risk factor for all of these soft tissue wound infections
is ‘skin-‘ or ‘muscle-popping’ [2,3,5,13], injection
practices in IDUs are likely to be important, and geographic variations
in these may explain the absence of a similar increase in cases in other
European countries. However, clinicians should suspect botulism in any
patient with an afebrile, descending, flaccid paralysis. Botulinum antitoxin
is effective in reducing the severity of symptoms for all forms of botulism
if administered early in the course of the disease; this should not be
delayed until results of microbiological testing are available. In cases
of wound botulism, antimicrobial therapy and surgical debridement are
important to reduce the organism load and avoid relapse after antitoxin
treatment. C. botulinum is sensitive to benzyl penicillin and
metronidazole. Advice for responding to suspe ct wound botulism is available
on the HPA website [24]. As well as providing information for health
professionals, the HPA website gives advice for preventative measures
to IDUs including the following:
• Smoke rather than inject heroin;
• If IDUs must inject, inject intravenously and not intramuscularly or
subcutaneously;
• Do not share needles, syringes, cookers, or spoons for injection;
• Use as little citric acid as possible;
• If injecting more than one type of drug, inject in separate places;
• If swelling, redness or pain occurs at injection sites, seek medical
advice immediately [24].
At the time of writing (July 2005) a further 20 cases of wound botulism
in IDUs had been reported in the UK during 2005.
Acknowledgements
The authors thank clinical, epidemiological, and microbiological colleagues
for submission of samples and collection of data.
|