Treatment of snakebite in Australia: gathering the evidence

The World Health Organization has declared snakebite a neglected tropical disease. Although Australia has many of the world’s most venomous snakes, the annual rate of envenoming here is much lower than in some other regions.1 In Australia, rates of snakebite are higher in tropical areas — a prospective study in the Northern Territory estimated the rate of envenoming to be 7.6 per 100 000 people per year.2 Sadly, there were at least three snakebite deaths in Queensland in the summer of 2012–2013.

With the increase in evidence-based guidelines for managing many medical conditions, the publication in this issue of the Journal of an updated guide to the diagnosis and treatment of snakebite in Australia is timely.3 The article from Isbister and colleagues is underpinned by results from the Australian Snakebite Project (ASP), a significant undertaking in which over 100 Australian hospitals have recruited patients bitten by snakes and for which much of the data collection has been from rural hospitals.

Before the ASP, snakebite management recommendations were often based on anecdotal experience. Antivenoms were introduced with no randomised controlled trials. For the first time, we now have a large collection of prospective data, including results of blood investigations, serum venom levels and patient outcomes. While the ASP studies may rank low in the National Health and Medical Research Council hierarchy of evidence, being essentially case series, they provide far greater evidence than was previously available. Of course, with more data come more questions.

A difficulty for many clinicians treating snakebite in rural Australia is the limited access to laboratory investigations. For example, in Cape York medical facilities, if blood samples do not reach the daily flight to Cairns, it can be 24 hours before formal laboratory investigations are performed. Snakebite victims often need to be transferred to Cairns Hospital, at considerable cost. Isbister et al highlight that point-of-care devices for measuring international normalised ratio (INR) give false negative results in patients with venom-induced consumption coagulopathy (VICC). They also argue that the whole blood clotting test (WBCT), which has been previously recommended, is “frequently misleading” in snakebite and that the glass tubes needed for the test are not available in Australia. Evidence for4 and against5 WBCT use in snakebite is based on overseas data, and one Australian study suggests that polycarbonate containers are a potential alternative to glass.6 What we need is evidence regarding the utility of the WBCT in the Australian context and a simple way for clinicians in rural Australia to assess the coagulation status of patients with snakebite.

Questions about dosing of antivenom also remain. Isbister et al state that “One vial of the relevant antivenom is sufficient to bind all circulating venom”,3 yet others suggest, based on clinical experience, that certain groups of snakes may have larger venom loads or other toxins in the venom and require larger doses of antivenom.7 Isbister et al’s recommendations are based on serum venom levels recorded at various times along patients’ clinical courses in the ASP studies. These studies found significant variations in venom levels, ranging from 0.17 to 152 ng/mL for 50 people bitten by tiger snakes,8 0.6 to 624 ng/mL for 15 people bitten by mulga snakes,9 and 0.15 to 210 ng/mL for 131 people bitten by brown snakes.10 In all ASP studies except the brown snake series, no venom was detected in the serum after administration of antivenom (frequently one vial). Of the 115 patients bitten by brown snakes with post-antivenom levels available, only three still had venom present in their serum (albeit low levels, 0.4–0.9 ng/mL).10 Sixteen patients who had each received one vial of brown snake antivenom had no detectable venom. This study also showed that the time until the patient’s INR recovered to < 2 was the same whether one, two or three vials of antivenom were used, supporting the authors’ argument to use one vial.

Brown snake venom can act very quickly — when added to blood in a test tube, a clot will form in less than a minute (Associate Professor Jamie Seymour, School of Public Health and Tropical Medicine, James Cook University, personal communication). There are case reports of envenomed patients presenting to hospital with evidence of bleeding within 20 minutes of being bitten by a brown snake.11,12 Could it be that the venom acts too quickly for the antivenom to stop VICC from occurring, and hence no difference would be seen with one, two or three vials?

Interestingly, an ASP randomised controlled trial investigated whether giving fresh frozen plasma to patients with VICC after they received antivenom shortened the time of recovery of clotting (as measured by an INR < 2).13 Those who received fresh frozen plasma less than 6 hours after the antivenom did not recover as quickly as those who received it more than 6 hours afterwards. This suggests ongoing consumption of clotting factors, which would usually be attributed to free venom, although the authors felt it was due to the presence of “active clotting factors”, as their other brown snake study had shown clearance of venom after administration of antivenom.10

Is it therefore reasonable to assume that serum venom levels reflect the total body burden of venom and the venom’s effect on coagulation or other signs of envenoming, such as paralysis or rhabdomyolysis? In the ASP death adder study, where the median peak serum venom level was 22 ng/mL (interquartile range, 8.5–29 ng/mL), four of 13 patients with detectable venom (range, 0.7–40 ng/mL) had no signs of envenoming (paralysis).14 Although we clearly face some challenges in understanding the correlation between serum venom levels and clinical effects, these levels are probably the best measure we currently have.

Snakebite in Australia is a rare but potentially lethal condition, and most clinicians have limited exposure to it. The recommendations from Isbister and colleagues reflect a major change in the way we should manage snakebite and will hopefully lead to standardisation of care around Australia. Ongoing data collection and study are required to validate and refine these recommendations, which should generate further debate and, in time, new evidence.