Heated Debates: Hot-Water Immersion or Ice Packs as First Aid for Cnidarian Envenomations?

Cnidarian envenomations are an important public health problem, responsible for more deaths than shark attacks annually. For this reason, optimization of first-aid care is essential. According to the published literature, cnidarian venoms and toxins are heat labile at temperatures safe for human application, which supports the use of hot-water immersion of the sting area(s). However, ice packs are often recommended and used by emergency personnel. After conducting a systematic review of the evidence for the use of heat or ice in the treatment of cnidarian envenomations, we conclude that the majority of studies to date support the use of hot-water immersion for pain relief and improved health outcomes.


Introduction
According to some estimates, more than 150 million cnidarian envenomations occur worldwide every year [1]. Adverse reactions range from minor local pain to severe and life-threatening cardiovascular collapse, and can vary depending on the age and health status of the victim, the amount of tentacle contact, and the species involved in the envenomation. However, despite the frequency of cnidarian envenomations and the medical urgency, there is much debate and disagreement between national and international guidelines as to the most effective first-aid measures to reduce pain or pathology from cnidarian stings. Conflicting recommendations confuse medical professionals and the general public and undermine the ability of on-site first responders to provide the best care. Immersion in hot water (~45˝C) is a well documented and commonly recommended therapy for many kinds of marine envenomations, including the painful stings from echinoderms and venomous fish [2]. The mechanism of action is not entirely clear, but there is evidence that marine venoms are heat labile, and thus immersion of the sting area in hot water is thought to inactivate key venom components. Others have postulated that the temperatures needed for such inactivation are much higher (~60˝C), and thus suggest that the pain relief which occurs during hot-water immersion instead results from physiological reactions to heat application (such as increased subcutaneous blood flow).
Conversely, it was once thought that cryotherapy or immersing envenomed areas in ice water might slow the movement of venom towards the core (heart), and thus might lead to better outcomes in deadly envenomations. Experimental evidence does not support the use of cryotherapy or cold-water immersion in reducing pathology or improving clinical outcomes, however, and the practice is no longer recommended [3]. Ice packs are still commonly used in non-life-threatening arthropod envenomations, such as bee stings. In such cases, there is no presumed action on the venom, and rather, relief is believed to occur through the anti-inflammatory effects of cold.
Hot water and ice are both inexpensive and readily available in many locations where cnidarian stings might occur, and thus would make ideal sting treatments if effective. Thus, the purpose of this review is to examine the evidence for the application of either cold (ice) or heat (hot-water immersion or hot showers) in the treatment of cnidarian envenomations. We performed a systematic review of published research to date (as of 25 January, 2016) by searching the following electronic databases: MEDLINE, CINAHL, Cochrane Reviews, and Google Scholar. Search terms were: jellyfish or cubozoan or cnidaria or physalia or "Portuguese man-o-war" or "sea nettle" and sting or envenomation or pain or treatment or heat or "hot-water immersion" or ice. We also examined reference lists from studies and guidelines discovered in our searches for potentially relevant papers missed in our database searches. More than 2000 articles were found via these searches, which were carefully examined and narrowed to the 19 relevant publications and abstracts discussed in Levels I to IV in this paper (not including published abstracts or the additional six case series without heat or cold application which are also discussed in Level III).

Levels of Evidence
The Levels of Evidence used in this review were adapted from the US Agency for Health Care Policy and Research Classification, similar to that reported elsewhere [2,4]. They are as follows: I Evidence from randomized controlled trials a Meta-analysis of randomized controlled trials b Evidence from at least one randomized controlled trial II Evidence from non-randomized controlled trials a Quasi-randomized controlled trials b Non-randomized controlled trials Uncontrolled studies c Expert letters or opinions d Systematic reviews

Level I Evidence: Randomized Controlled/Paired Trials
Three studies were determined to fit the criteria for Level I evidence. However, it is important to highlight the difficulty of obtaining a high-quality randomized trial for the treatment of cnidarian envenomations, particularly when temperature is one of the test conditions. How much tentacle contact/venom delivery occurred, the duration of time between the sting event and when medical care is sought (and thus when the victim is enrolled into the trial), and what measures are performed during that time period (e.g., if the victim rinses the sting site in fresh water, or applies their own remedies prior to approaching a life guard) are all variables that may have a significant impact on clinical outcomes after a sting event and are difficult or impossible to control. Furthermore, there is no adequate way to blind the study participants or administrators when the different treatments are temperature based. That said, summaries of three trials are presented in Table 1. Inclusion of only severe stings and possible allocation bias (baseline differences in severity) Notes: * Indicates p < 0.05; ** Indicates p < 0.001. Two of the Level I studies examined stings from the pan-tropical cubozoan Alatina alata (synonymous with A. moseri, reported as Carybdea alata). Responses to stings from this large box jellyfish can range from mild to severe, including life-threatening Irukandji syndrome. Thomas et al. [5] evaluated the short-term effects of hot or cold pack application as compared to a placebo (less than 15 min post treatment) in 133 swimmers. All participants received vinegar dousing prior to hot or cold pack application. Hot pack application reduced pain scores at 5 and 10 min post treatment, and there was a significantly higher odds ratio (5.2) for the complete cessation of pain with hot pack application compared with placebo. Hot pack application also reduced pain scores when compared directly with cold pack application at 10 min. However, the authors themselves noted that the effect only bordered on clinical significance. In a randomized paired trial on 25 healthy volunteers, Nomura et al. [6] compared hot-water immersion with other common first-aid measures (papain and vinegar). Volunteers were stung on each arm, and one was submerged in hot water and the other received either vinegar or papain. Pain scores were lower with hot-water immersion at 4 and 20 min. However, the use of potential active substances rather than a placebo may have affected the control baseline, and the distraction of two simultaneous stings has been noted as a potential confounding factor [2,3].
The third Level I study was a randomized controlled trial to test the efficacy of hot-water immersion with ice packs for pain relief from Physalia stings (class Hydrozoa). Loten et al. [7] compared the pain reported from 96 sting victims which were randomized to receive either hot-water immersion (49 subjects) or ice packs (47 subjects). A significantly larger percentage of the hot-water group reported less pain at 10 and 20 min of treatment (53% and 87%, respectively). At the halfway interim analysis, the trial was halted as hot-water immersion was determined to be more effective (p = 0.002). The hot-water group did begin the study with a higher degree of initial pain reported, though, which may indicate allocation bias. In addition, only severe stings were admitted into the study, which may also have influenced the results.
A fourth trial which would have been included as Level I evidence was reported in an abstract in Emergency Medicine in 2002 [8], comparing hot showers to ice packs in Physalia envenomations off the coast of Sydney, Australia in a randomized crossover trial. The abstract stated that 54 adults were randomized to receive hot showers or cold pack application (27 each), while only 24 completed crossover. Hot showers reduced total treatment time (11.0 min versus 14.5 min) and led to greater percentage pain reduction (82.1% versus 65.6%) according to the combined results, with 48% of hot-shower patients reporting complete pain cessation (while only 29% reported cessation with ice packs). However, this study lacked follow up, and to our knowledge, complete methods and results were never published in a full, peer-reviewed paper, thus they are not included in Table 1.

Level II Evidence: Non-Randomized Controlled Clinical Trials
No Level II evidence for the use of hot-water immersion or ice packs for cnidarian envenomations was found. One study, which would have been included as Level II evidence, was reported in an abstract in the Journal of Toxicology Clinical Toxicology in 2000 [9] and compared the use of ice or hot-water immersion in a quasi-randomized trial. Adherent tentacles removed with vinegar; on odd days, sting area was immersed in hot water (110˝F/ 43˝C), while on even days, ice packs were applied. Follow up calls at 1, 4, and 24 h. Twenty-four patients completed the trial without protocol violations; 16/16 hot water and 5/8 ice reported pain relief within 60 min. Three patients failed to obtain pain relief with ice packs and subsequently experienced pain relief with immersion in hot water. Hot water was reported as significantly better than ice at relieving pain (p < 0.05). However, this study was never published as a full, peer-reviewed paper, thus it is not included as Level II evidence.

Level III Evidence: Retrospective Case Series
Three retrospective case series were identified where heat application or ice packs were used. The results of these studies are summarized in Table 2. Peca et al. [10] examined 40 cases of Carybdea marsupialis envenomation in individuals admitted to the Accident and Emergency Department of San Giorgio Hospital in Cervia, Italy, in July and August, 1994. No systemic symptomology was found in any of the patients, but all had linear erythematous weals ranging in length from a few centimeters up to 20 centimeters. In eight patients, the sting area was soaked in "warm salty water to deactivate the toxins, which are thermolabile, until resolution of symptoms was obtained, a result achieved in all patients in about 30 min". No statistical analyses were performed, and the temperature of the water was not noted.
Yoshimoto and Yanagihara [11] examined the outcomes of 177 definite or probable cnidarian stings over a five-year period in Hawai'i (1994)(1995)(1996)(1997)(1998). Of the 177, 60 were treated with heat in some form (53 with hot showers, seven with localized hot packs), but only 42% of cases contained sufficent outcome information to be analyzed. Heat was found to be superior to parenteral analgesics (odds ratio of 11.5, p = 0.08), particularly if analyses were restricted to hot showers (odds ratio of 22.0, p = 0.0485). Of the 25 cases treated with heat and in whom outcome information was available, 23 reported relief of pain. When only patients presenting with symtoms of Irukandji syndrome were considered, heat still appeared to be superior to analgesics and benzodiazepines. Heat application was not reported to be deleterious in any of the 60 cases.
Currie et al. [12] examined patients presenting with jellyfish stings at the Royal Darwin Hospital and remote coastal community health clinics in the Northern Territory between 1 April 1991 and 30 May 2004. Ice was applied in 71% of cases, with additional analgesia required almost half (48%) of the time, and parenteral narcotics necessary in almost a third (30%) of cases where ice was used. No statistical analyses were conducted to determine if the application of ice affected clinical outcomes.
In addition, there are six additional case series of cnidarian envenomations, in which neither heat nor ice was applied. Maretić and Russell [13] discussed the outcomes of 55 cases of Anemonia sulcata envenomations from the Adriatic Sea, noting that, for many of the cases, no immediate treatment was given or the treatment was ineffective; vinegar, wine, urine, manure, mustard, ammonia, alcohol, and petrol as common local measures, though the most commonly employed measure was "rubbing a fig over the injured area". No life-threatening cases were reported.
Labadie et al. [14] reported the results of 1079 Physalia physalis stings in patients, who reported to a local poison center in Bordeaux, France from 2008 to 2011. Systemic symptoms were reported in 20% of cases, including gastrointestinal, cardio-respiratory and neurological symptoms. In 2011, 8% of patients presented with severe enough systemic manifestations (such as respiratory distress) to be considered life-threatening by the attending physician, though no fatal cases were reported. The authors noted that hot-water immersion was not proposed as the evidence for its efficacy was for a different species (Physalia utriculus), and the authorities deemed hot-water immersion "too difficult to perform in case of collective accident with numerous patients to manage at the same time".
Four of the additional case series examined patients presenting with Irukandji syndrome; none of cases received ice packs or heat treatments for either the initial sting pathology or systemic symptoms. Little and Mulcahy [15] identified 62 cases of Irukandji syndrome from Cairns emergency departments in 1996. Of these, 61% received opiate analgesia, and 81% had vinegar applied, which appeared to have no adverse effect. Huynh et al. [16] noted that, of 128 patients who were discharged with a diagnosis of "marine stings" from Cairns Base Hospital in Queensland, Australalia, between June 2001 and July 2002, 116 had Irukandji syndrome symptoms. These cases required an average dose of 31 mg of morphine; 36% were discharged directly from the emergency department within eight hours of presentation and 47% were discharged from the emergency department observation ward the next day. Macrokanis et al. [17] found that, of 111 patients with marine stings between January 2001 and July 2003, 88 had symptoms of Irukandji syndrome; 38% of those had vinegar applied before admission to the hospital; 50% were hypertensive upon arrival; and 90% required opiate analgesia. Nickson et al. [18] described the outcomes of 87 Irukandji cases from Australia's Northern Territory between 1990 and 2007. Systemic features were common, including hypertension and electrocardiographic abnormalities, but no fatalities occurred. Vinegar was used in the majority of cases.

Level IV Evidence: Controlled Bench Studies, Expert Opinions/Letters, and Systematic Reviews
Thirteen studies were determined to fit the criteria for Level IV evidence: 1 controlled bench study, three systematic reviews, two uncontrolled studies, and seven expert opinions, which are summarized in Table 3.

Controlled Bench Studies
Yanagihara et al. [19] developed a novel array of ex vivo cnidarian envenomations models with which to test various first-aid measures. In the most sophisticated human tissue model, red blood cells were suspended in an optically clear, low-melting point agarose protected by a skin substitute made from sterilized porcine small intestine. Live tentacles were applied to the skin and allowed to sting spontaneously for 5 min, at which point they were removed and a pack containing either ice water or hot water (45˝C) was applied for 5 min. Size of the lytic zone was calculated after one hour of incubation at 37˝C and compared to controls which received neither treatment. Hot-water treatment significantly reduced the size of the lytic zone (p < 0.05), while ice water had no significant effect.

Systematic Reviews
Three systematic reviews were identified in our search. The first examined evidence for or against hot-water immersion for all marine envenomations [2], finding that "Current published evidence seems to support the use of hot-water immersion in the treatment of non-life threatening marine envenomation, alongside other established first aid measures." Similarly, a systematic review six years later also concluded that the experimental evidence supports the use of heat to inactivate venom components [4], and a similar systematic review for the Cochrane Database found limited support for the use of hot-water immersion and no support for the use of ice to treat cnidarian envenomations [20].

Uncontrolled Trials and Expert Opinions/Letters
Exton et al. [21] provided the first evidence in support of the use of ice in an uncontrolled trial. Physalia stings were classified by severity (severe, moderate or mild) and pain severety (severe, moderate or mild) for 143 sting victims at the beach. For all cases, ice packs were applied for 5-10 min, after which pain was assessed. If pain persisted, an ice pack was applied for an additional 5-10 min, then pain was reassessed. All of the mild pain cases resolved after one ice pack application, while about a quarter of the moderate and over a third of the severe cases required additional treatment. Pain persisted after two ice packs in 25% of the severe cases and 2% of the moderate ones. While ice was credited with symptom relief, there was no placebo or control performed. Thus, it is unclear how many cases would have resolved spontaneously without ice.
Taylor [22] performed a quick experiment using a local Carybdea species at a doctor's seminar at the Busselton Hospital in Western Australia, which was published as a letter to the editor. Five volunteers (doctors and medical students) were stung by dragging tentacles over the forearm, two stings per arm for a total of four stings. Each volunteer then received four treatments, one at each sting site: ice, vinegar, aluminium sulfate, and hot water (approximately 45˝C). Hot water led to 85% pain relief during treatment and 88% continuing pain relief after treatment, while ice relieved pain in 25% of cases during treatment and relief was continued in only 2% of cases after treatment. The simultaneous testing of four treatments may have made pain judgements difficult, however, and no statistics were performed.
The author captured a specimen and allowed himself to be stung out of curiosity. He described a "violent aching pain", which he attempted to relieve with cold water.
Cold water applied to the sting site "was found to rather increase than dimish the effects." Cleland & Southcott 1965 [24] Expert Opinion/Case History Physalia sp.
Case notes from Jack Barnes quoted by authors; Case described as the Physalia sting on the arm of a 21-year-old male. Site was treated by rubbing dry sand and, later, a weak ammonia solution, neither of which seemed to relieve pain; pain spontaneously resolved about 20 min after sting. About 1 hr later, returning to the water led to immediate pain similar to initial sting.
Pain was also "reawakened" 4.5 h after the sting by rubbing the area with ice, producing an "electric feeling," while warm water had no effects.

Burnett 1989 [25] Expert Opinion/Case History Physalia sp.(Hydrozoa)
The author summarizes his experience and recommendations for the treatment of jellyfish stings, citing one particular case where visible erythematous lymphangitic spread could be seen on the extremity of an adult male who had immersed his arm in a 31˝C water bath for 13 min following a Physalia sting.
"The use of local heat is contraindicated since it increases the permeability of the venom. . . . Cold compresses have not been effective against the pain. Pain is reduced as long as cold is applied to the sting site but returns once the extremity is rewarmed." Currie 1994 [26] Expert Opinion Chironex fleckeri (Cubozoa) Review of research on C. fleckeri venom and clinical applications.
States that ice packs provide good pain relief for mild stings, citing the author's unpublished data.
Taylor 2000 [27] Expert Opinon/Case History Tamoya gigantua (Cubozoa) Report of four sting cases, two from the author. For the first three cases, no treatment steps are described. In the last case, the author was stung extensively on the hand and head while snorkeling. He proceeded to apply heat to determine its efficacy.
Hot water instantly relieved pain, but was transient until soaked for 20 min. Hot towels and a hot shower were not effective, so the author soaked his head in a bowl of hot water; a near-complete cessation of pain occured after 20 min.

Tibballs 2006 [28] Expert Opinion Australian Cnidaria
Non-systematic literature review and recommendations for treatment of stings from species found in Australian waters.
Cited Currie 1994 as evidence for ice packs; noted hot water efficacious in box jelly stings for species found outside Australia, and that heat inactivation of venom "has been long known"

Little 2008 [29] Expert Opinion All Cnidaria
Commentary and review of evidence of ice or heat for cnidarian stings, noting that improvement seen in Exton et al. [21] may have been a placebo effect, as suggested by Loten et al. [7]. Hot-water immersion, on the other hand, has been tested in multiple studies, which the author notes all suggested hot water was an effective treatment.
"There is very little evidence to support the use of ice packs for jellyfish stings, and there is increasing evidence that hot-water immersion is more effective in reducing the pain of jellyfish stings. It is time that bodies, such as the ARC, recommend the first aid treatment supported by the evidence." Several experts have described personal successes with ice or heat, or warned of potential complications. The earliest examples found of cold water or ice as a potential treatment contraindicated their use, particularly for Physalia. Bennett [23] found cold-water application increased the symptoms of his own Physalia sting in 1834, while Barnes [24] noted that in the case of one sting, ice "re-awakened" pain symptoms when applied more than four hours after the sting, while warm water had no effects.
Burnett [25] appears to be the first to recommend against the use of heat, citing a single example where visible erythematous lymphangitic spread on the extremity of an adult male who had immersed his arm in a 31˝C water bath for 13 min following a Physalia sting. Burnett thus concluded that "the use of local heat is contraindicated since it increases the permeability of the venom". However, the temperature of the water used was well below body temperature and thus was not "hot water", nor was it in the known range of venom inactivation (somewhere between 40-60˝C, Table 4). Burnett also noted that cold compresses were not effective, as any pain reduction which occurs while the compress is applied "returns once the extremity is rewarmed". He later softened his position on both [30], writing that "additional pain relief can be achieved by cold packs or warm compresses," with the caveats that "care must be taken not to over-chill a small patient nor to apply sufficient heat that vasodilatation opens extra avenues for venom entry to the core of the body". However, fears of negative outcomes from vasodilation due to heat have not been corroborated by the clinical literature (see evidence in Levels I-III; Tables 1-3). Other opinions supportive of ice packs for pain cite unpublished data [26] or opinions of others who cite unpubished data [28], showing how conclusions lacking scientific support are propagate through the literature.
Taylor [27] reported on four stings from the cubozoan Tamoya gigantua, two of which were his own. For the first three cases, no treatment steps are described. In the last case, the author was stung extensively on the hand and head while snorkeling, and thought to test the efficacy of hot-water immersion by soaking his hand and head in water "as hot as he could stand". Hot water instantly relieved pain, but reliefwas transient if the area was removed before 20 min of treatment. Hot towels and a hot shower were not effective.
More recently, Little [29] reviewed many of the studies included in the first three levels of evidence in this paper, and concluded that "there is very little evidence to support the use of ice packs for jellyfish stings, and there is increasing evidence that hot-water immersion is more effective in reducing the pain of jellyfish stings" He also had strong words for the Australian Resucitation Council (ARC) and others who write clinical guidelines, stating that, "It is time that bodies, such as the ARC, recommend the first-aid treatment supported by the evidence".

Mechanism of Action of Heat/Hot-Water Immersion
The prevailing hypothesis as to the mechanism by which heat treats cnidarian envenomations is that the temperatures used inactivate venom components. There is abundant evidence from in vitro and in vivo studies which suggest that cndarian venoms from all classes in the phylum are heat labile (Table 4) [31][32][33][34][35][36][37][38][39][40][41][42][43][44][45]. Some have suggested that the mechanism by which heat provides pain relief is unrelated to inactivation of venom components. The alternative explanation proposed is that hot-water immersion has a direct modulatory effect on pain receptors leading to a reduction in perceived pain [46].
For terrestrial envenomations, hot-water immersion was less commonly recommended (Table 5) [74][75][76][77][78], although there has been some support for the use of targeted, localized heat in the treatment of insect bites and stings [79]. Heat therapy is specifically contraindicated for recluse spider envenomations (Loxoceles spp.), as the dermonecrotic venom factor sphingomyelinase D increases its activity as temperature increases [75]. This group of spiders are the only ones with this particular enzyme, however, so it is unclear if hot-water immersion would be appropriate in other spider envenomations. Preliminary case reports and investigations suggest heat application may aid in pain relief from scorpion stings [76,77] and centipede bites [78]. A prospective cohort study (N = 146) from Germany found that the targeted application of heat (maximum of 51˝C) for 3-6 s using the Bite Away ® device (RIEMSER Pharma GmbH, Greifswald, Germany) led to a reduction visual analog scale scores for pain, swelling and pruritus from mosquito bites and bee and wasp stings [79]. Neither heat application nor ice are recommended for snakebites [80], though we could find no clinical documentation of the use of hot-water immersion. Cohen et al. [81] found that, in a mouse model, hot-water immersion caused increased mortality in one of the two experiements performed with the hemotoxic venom of Agkistrodon piscivorus, contraindicating the use of hot-water immersion for snakebites. Interestingly, they also noted using heat application in their own clinical pathway for snakebite; however, they stated that heat is only applied after antivenom has been administered and provided no information about the clinical efficacy of heat application. Hot-water immersion recommended for envenomations from cnidarians, cone snails, sea urchins, sea stars (noted that may be ineffective for Acanthaster planci), sting rays, and venomous fishes.
No systemic symptoms were noted.

Stingrays Order Myliobatiformes
Hot-water immersion with "water as hot as can be tolderated" for 30-60 min is recommended, even in cases of delayed presentation of symptoms, in addition to removal of foreign bodies as well as antibiotics.  Thirty-three-year-old man stung on finger. Wound washed and sterilized with alcohol, then hand immersed in hot water. Thirty minutes of hot-water immersion lead to relief of pain and numbness.
Ongkili & Phee-Kheng 2013 [73] Case Report Stonefish Order: Scorpaeniformes Family: Synanceiidae Forty-seven-year-old stung on foot. Given 75 mg intramuscular diclofenac, then 5 mg intravenous morphine with no improvement in pain. Soaked foot in hot water at the highest tolerable point via the elbow skin test. Pain score reduced after 15 min of hot-water immersion and diminished at 1 hr. were bitten by mosquitoes, and 8 were stung by bees (5.3%). Swelling, pain and prutitus decreased 10 min after the use of the Bite Away ® medical device, which non-invasively administers concentrated heat (max of 51˝C) to the skin for a choice of 3 or 6 s.

Conclusions
The preponderance of evidence demonstrates that hot-water immersion is a safe and effective method of reducing pain from cnidarian envenomations, and is also associated with improved clinical outcomes. No studies or cases were found where hot-water immersion led to worsened symptoms or poorer clinical outcomes. Fears of negative effects of immersing a stung limb in 45˝C water for 20 min are not supported in the literature; this is not surprising, given that hot-water immersion is considered safe and is recommended for the treatment of stonefish stings and other life-threatening marine envenomations with potential cardiovascular complications.
There are two noted caveats to the use of hot-water immersion in the case of envenomations: temperature safety and temperature maintenance. While there is considerable concern regarding thermal injuries from hot-water immersion, there was only a single recorded case of significant thermal burn from over 200 cases of the use of hot-water immersion reviewed by Atkinson et al. [2]. Similarly, the difficulty of maintaining hot-water at an appropriate temperature on site can be overcome using several methods, including the use of reusable hot packs instead of water, continual hot-water input (such as from a shower or hose at the correct temperature), or an immersion container with better insulatory capabilities. To that end, Lau et al. [54] tested the use of different containers on temperature maintenance, finding that thermal insulators (standard coolers) were able to maintain water temperature effectively for a full 30 min.