To the Editor:
In Australia, snakebite is an infrequent cause for emergency department presentation. Paediatric presentations can present particular challenges, as the local signs of elapid snake envenomation are minimal and systemic features may be nonspecific. The state of Western Australia (WA) occupies about one third of Australia's land mass but has only 1 tertiary pediatric hospital (Princess Margaret Hospital [PMH]) to which the majority of the state's pediatric interhospital transfers and admissions occur. This unique geography and referral patterns present an opportunity to study patterns of illness presentation such as snakebite across the state.
We conducted an ethics committee–approved, structured, retrospective review of the medical records of all pediatric snakebite presentations to PMH from January 1994 to December 2004. Cases were identified by a search of the emergency department and hospital electronic databases using predetermined disease codes. Their medical records were then retrieved and reviewed. In this report we aim to describe the epidemiology and clinical features of pediatric snakebite presentations to PMH during this period. In addition we also reviewed the WA State Coroner's Forensic Pathology Database to gain information on pediatric snakebite deaths during this period. We used the same methodology as a previous published study from PMH 1 that then enabled us to compare our results with that previously reported for the period 1984–1993.
There were 151 presentations for suspected snakebite, 99 from the Perth metropolitan area and the remainder from throughout the state (but predominantly within 100 km of Perth). This geographic distribution is roughly representative of the WA population as a whole. The mean age of those presenting was 6.6 years, 55% were male, and most were bitten on the lower limb (72%). Bites typically occurred in the afternoon and early evening and peaked in late spring and late summer. Of all the clearly bitten patients, 18 were considered definitely envenomed. Of these patients, 13 tested positive for brown snake immunotype on the Australian Snake Venom Detection Kit, 3 for tiger snake, 1 for black snake, and 1 for sea snake. In WA a positive Australian Snake Venom Detection Kit for 1 of the 5 immunotypes is likely to represent venom from the following species: tiger, Notechis scutatus; brown, Pseudonaja nuchalis or Pseudonaja affinis; black, Pseudechis australis (the mulga snake); death adder, Acanthophis antarcticus; and taipan, Oxyuranus scutellatus (bites not reported in WA although theoretically possible). All of the envenoming syndromes seen in these patients were typical for the presumed species involved. The coroner's database recorded 2 deaths in this period, both from death adder (Acanthophis antarcticus) envenomation and occurring in remote areas of WA. Comparison to the previously published data showed no change in mean patient age or bite site (lower vs upper limb), however a small decline was noted in the percentage of males presenting (68%–55%, P = .018).
That only 18 of 151 patients were considered to be definitely envenomed when all of the available information in the medical record was reviewed confirms that most bites were indeed “dry bites” not leading to envenoming. More cases met our predefined clinical or laboratory criteria for envenomation. Common reasons for this misclassification, all of which alone are plausible signs of snake envenomation, included the presence of abdominal pain, vomiting, minor elevations in creatinine kinase, or minimally abnormal clotting profiles. There are a variety of definitions of what constitutes envenomation in a given patient. 2 –6 That this can be confusing to treating clinicians is highlighted by 9 cases in addition to the 18 definite envenomations above that we determined retrospectively were not envenomed but were nevertheless administered antivenom. These cases were usually in remote areas of the state and were cared for by nonexpert medical or nursing staff. We speculate that despite the availability, if sought, of expert toxicological advice by phone, the enormous distances (up to 2000 km) and times (up to 12 hours) involved in transfers to PMH for definitive care may lead to clinicians judging the presentation to be consistent with envenomation and the (incorrect) decision made to give antivenom “just in case” and before transfer for ongoing definitive care.
In conclusion, this study did not detect a change in the patterns of presentation of pediatric snakebite in WA. Most bites are on the lower limb, envenomation is infrequent, and brown snake (presumably the dugite, Pseudonaja affinis, given the geographic locale of the bites) continues to be the most common cause of envenomation. There is ongoing difficulty in using clinical and laboratory criteria in making a diagnosis of elapid snake envenomation in children.