Redback Antivenom… what is everyone RAVEing on about?

There has been ongoing controversy about the efficacy of redback spider antivenom in recent times, since the initial RAVE trial suggested that there was no difference between IV and IM antivenom, and in fact, maybe no benefit over placebo. This conclusion was drawn from the finding that IV antivenom performed no better than IM at improving symptoms, and following IM administration, the antivenom was not detectable in the blood, and could therefore be acting as a placebo:

However, our laboratory findings that RBS antivenom is not detectable in the systemic circulation following IM administration suggests that this study may be equivalent to a placebo controlled trial of IV antivenom, and that RBS antivenom by any route provides little benefit over placebo.

The authors have followed this up now with the RAVE II trial: Isbister2014-RAVE2title

Study objective

Latrodectism is the most important spider envenomation syndrome worldwide. There remains considerable controversy over antivenom treatment. We aimed to investigate whether antivenom resulted in resolution of pain and systemic effects in patients with latrodectism who received standardized analgesia.


In a multicenter randomized placebo-controlled trial of redback spider antivenom for latrodectism, 224 patients (>7 years) with a redback spider bite and severe pain, with or without systemic effects, were randomized to receive normal saline solution (placebo) or antivenom after receiving standardized analgesia. The primary outcome was a clinically significant reduction in pain 2 hours after trial medication compared with baseline. A second primary outcome for the subgroup with systemic features of envenomation was resolution of systemic features at 2 hours. Secondary outcomes were improved pain at 4 and 24 hours, resolution of systemic features at 4 hours, administration of opioid analgesics or unblinded antivenom after 2 hours, and adverse reactions.


Two hours after treatment, 26 of 112 patients (23%) from the placebo arm had a clinically significant improvement in pain versus 38 of 112 (34%) from the antivenom arm (difference in favor of antivenom 10.7%; 95% confidence interval −1.1% to 22.6%; P=.10). Systemic effects resolved after 2 hours in 9 of 41 patients (22%) in the placebo arm and 9 of 35 (26%) in the antivenom arm (difference 3.8%; 95% confidence interval −15% to 23%; P=.79). There was no significant difference in any secondary outcome between antivenom and placebo. Acute systemic hypersensitivity reactions occurred in 4 of 112 patients (3.6%) receiving antivenom.


The addition of antivenom to standardized analgesia in patients with latrodectism did not significantly improve pain or systemic effects.

Where to from here?

From my brief reading, the antivenom group had a higher rate of pain reduction at 2 hours, by a greater degree, and required less rescue opiates, but none of these differences reached statistical significance. The authors note that the study was powered to detect a difference of 20% (Number needed to treat (NNT) ≤ 5), with the argument that since this treatment is for analgesia only (it is not life-saving), then the risks associated with the antivenom across a larger NNT would outweigh the benefits (a previous number needed to harm (NNH) = 2). Interestingly, pain was not resolved in quite a large number of cases in both groups. That leaves us with the question, do we now continue to give redback antivenom?

What does Prof Julian White think about it?

I thought it would be interesting to hear what our resident local toxinologist, Prof Julian White, has to say about the findings of this trial, and invited him to comment.

These are his thoughts:

Would you go to a Rave party? Some might consider me a bit old now, and conservative, so perhaps the youngsters out there might not want to listen to my opinion, but my advice is to stay away from potentially hazardous activities/ideas as rave parties. But is this relevant to the RAVE2 study, you ask? Maybe, depending on your point of view. For me, being a medical doctor is, at the core, about achieving the best outcomes for patients considered as a group, not just as isolated individual cases. My experience with latrodectism over several decades is that in significant cases of envenoming, with distressing regional or generalised symptoms, the patient is in very significant distress and urgently seeks some relief from that distress. This is especially true for young children (interestingly a group excluded from RAVE2). If we do nothing, what is the risk? Well, death is a very low risk, it seems; there are 13 recorded deaths from latrodectism in Australia prior to the introduction of RBSAV in 1956 and none since and detail is lacking on the pre-1956 deaths, so we can’t be sure if they were direct envenoming deaths. Therefore any treatment we give is not really to save life, it is to reduce distress, or the period of distress. Without treatment, in cases of significant regional or generalised latrodectism, how long will the period of significant distress be? Here the data is wooly in my view, but if we look at experience with latrodectism globally, then it is reasonable to suggest it is at least 6 hours, but in most cases 12+ hours, extending out to days, even weeks in some cases. Seeing a patient who has suffered latrodectism for days before seeking definitive treatment is not rare. What definitive treatments could we consider? Leaving antivenom aside for the moment, a number of treatments have been tried around the world and, mostly, been found wanting. IV calcium gluconate infusion was favoured in the US but is now a discredited treatment, showing no real benefits. Relaxants such as benzodiazepenes have been tried and are still favoured by some, notably in North America, where latrodectism appears to have a more prominent muscle spasm element. Analgesia is the most obvious treatment, because pain is the most common severe effect in latrodectism. There is considerable published experience indicating that oral analgesia is poorly effective for latrodectism and that opioid parenteral analgesia is often of limited effectiveness. Of course, analgesia just attempts to modify the symptom, it does nothing to counteract the cause (venom), so you have to keep using it until the venom is removed from the body by other means. Venom will be removed over time by normal homeostatic mechanisms, but the time involved will certainly be many hours and often extend over several days or longer. Another important point about analgesia as a treatment for latrodectism is that it is just targeting pain and will make no difference to the other venom effects, most notably profuse sweating. Anyone who has treated latrodectism will know that profuse sweating is both common and quite distressing for patients. So, what about antivenom? Antivenom is equine IgG antibody against whole Latrodectus venom. There is clear evidence that the presumed principal toxin group, the alpha-latrotoxins, are targeted by antivenom. Does the antivenom target the active site of the toxins? That is less clear, but just by targeting the toxin it will assist clearance, therefore should reduce the period of toxin exposure, even if the active site is not the binding site for antivenom. Exactly where the toxins target within the body is less certain, in humans, but the latrotoxins are neuroexcitatory toxins, stimulating the nervous system. They are quite unlike snake paralytic neurotoxins; they do not cause their effect by paralysing synapses and they do not cause flaccid paralysis. What is clear is that their target is not in the blood stream, but in the extravascular space. It follows that the important antivenom concentration level is not in the blood, but in the extravascular target tissues, specifically synapses. It is quite unclear if measured antivenom levels in blood have any relevance to the clinical effectiveness of antivenom in latrodectism. There are several issues with antivenom use for latrodectism. (1) Does the antivenom reach the toxin target sites effectively? (2) Does the antivenom effectively neutralise the clinical effects of the toxins? (3) Which administration routes are effective for antivenom? (4) What is the hazard rate for antivenom and how does this relate to a risk/benefit analysis of effectiveness as treatment? The evidence to answer each of these questions is either incomplete, or contradictory. There is no clear cut answer. What is clear is that large scale anecdotal or retrospective studies of latrodectism in Australia appear to indicate that antivenom is effective at treating symptoms, is effective IM as well as IV, and is safe (no recorded deaths from treatment). Even outside Australia, experience with other anti-latrodectus antivenoms suggests they are at least reasonably safe. In the US a single reported death due to anaphylaxis to antivenom has produced fear that the cure is worse than the disease, but this greatly overemphasises the risk. At a rough estimate, globally at least 100,000 people have been treated with anti-latrodectus antivenom, and there is just one reported death. The RAVE studies have assumed that if antivenom is not detected in blood then it cannot be working. From this they extrapolated that if the effectiveness of IM was similar to IV, then antivenom didn’t work. This provided the ethical basis for a trial of antivenom IV versus placebo. We don’t know exactly how antivenom works in latrodectism so how can we be sure that a lack of blood levels after IM injection indicates IM is ineffective? There is a huge experience in Australia with RBSAV given IM as a seemingly effective treatment. Could it be that we just don’t understand enough about the molecular mechanisms of latrodectism to explain why IM might work? Giving antivenom IV certainly makes some sense, because it ensures rapid distribution of antivenom. I now prefer to give RBSAV IV for this reason, plus I think it can be less stressful for children and more controlled. If you get a major adverse reaction (anaphylaxis) with IV administration, you can at least stop administration of more antivenom while you control that reaction. The IM route does not offer that option. I can state with personal certainty that I have seen RBSAV be associated with rapid (30 min to 2 hrs) reduction in symptoms of latrodectism on many occasions, entirely consistent with the pre-RAVE literature. I have seen this apparent response in both adults and children, and in particular, young children where a placebo effect seems less likely. The pain resolves. The distress resolves. The sweating resolves. The hypertension resolves. I hesitate to characterise this as “magical”, but that is a word frequently used by colleagues and patients, witnessing/experiencing the apparent effect of RBSAV. None of this is an RCT. None of it is therefore top notch evidence based. What if we follow the RAVE2 recommendations and stop using antivenom? In RAVE2 the authors note that 75% of cases had poor control of pain, despite opioid analgesia. Does this indicate opioid analgesia is an effective treatment by itself? I would argue it suggests the opposite, that opioid treatment is ineffective in latrodectism. So why did RAVE2 find that antivenom didn’t work? Maybe it was the endpoints. The antivenom group had more rapid response than the placebo group, but at 24 hrs both were similar. Latrodectism does not continue indefinitely. If you wait long enough it will resolve. The medical issue is one of shortening the period to resolution. It would appear that antivenom, even in RAVE2, had some success in achieving that. What about resolution of other symptoms such as sweating? RAVE2 didn’t measure that effectively. What about antivenom dose? Here there may be a problem with RAVE2, with discrepancy about the actual dose used, based on information provided in the paper. BOTTOM LINE: I recommend that a well established and apparently effective treatment for latrodectism, namely RBSAV, should still be considered the standard of care, until and if further independent studies confirm the RAVE2 study and a more effective treatment protocol is discovered and proven, because RAVE2 shows that the current alternative to RBSAV, opioid analgesia, is poorly effective. I hope this rather lengthy discussion is what you were hoping for. Regards, Julian White

And for those coming to visit Australia:

A Clinician's Guide to Australian Venomous Bites and Stings

If you would like a copy of this book, you can ask CSL to send you one (for free). This is open to medical practitioners registered with AHPRA. Contact CSL and just state that you would like a copy and cite your AHPRA registration number, and a book will soon appear in your letterbox.

Share your thoughts