Bites and Stings of medically important venomous arthropods

Published in the International Journal of Dermatology, volume 37, page 481-496, July 1998,

(all photos copyright IJD International Society of Dermatology; Tropical, Geographic and Ecologic. Numbers in the text refer to references listed at the end of the manuscript)

INTRODUCTION

About 75% of the world's animal species are arthropods. While most of these species have no appreciable interaction with humans, a few cause significant medical or agricultural problems. This review is concerned with the small number of medically detrimental species which possess venom that causes morbidity and mortality when humans are bitten or stung. We restrict our discussion to the most important groups of venomous arthropods: the arachnids (spiders, scorpions) and insects of the order Hymenoptera (bees, wasps, ants). These species have a venom apparatus consisting of a gland or pair of glands and a system (fang, sting) meant to inject venom. Beyond the scope of this review are arthropods that employ toxins without delivery systems (i.e, topical noxious repellents) or those which vector disease. This review covers the relevant species worldwide (focusing on North American species), with special reference to the dermatologic expression of the injury.

Toxicologically, there is far more diversity in spider venoms than in the venoms of Hymenoptera and scorpions, so that the correct identification of spider bites has special significance. Spiders are rarely correctly identified by bite victims or their physicians and are readily transported by human commerce out of their native range. There are stellar reviews of the toxicology of arthropod venoms and poisons 1,2, as well as an excellent quick guide to bites and stings of arthropods 3.

BITE VS. STING

In discussing human envenomation incidents, quite often both the lay and medical communities use the words "bite" and "sting" interchangeably, even in the same sentence, to describe one mode of venom injection. In the strict sense, "bite" describes venom injected via structures associated with the mouth such as fangs or mandibles and "sting" connotes the injection of venom via a tapered, posterior structure most accurately called a sting (or colloquially, stinger). Venoms associated with the feeding organs of the animal, the mouth, have evolved for handling prey and therefore are designed to bring about paralysis, widespread tissue destruction and/or death in the victim. Defensive use of these oral venoms is secondary. Venoms associated with posterior stings have evolved primarily for defense and are designed to cause immediate pain such that the victim associates the creature with an unpleasant but probably non-fatal experience and learns to avoid it on future encounters. Both fangs and stings are structures for venom delivery, and have associated venom glands and ducts. "Venomous" refers to any animals that possess venom (not all venomous creatures are dangerous to humans), while "poisonous" refers to those venomous animals whose venoms have a deleterious effect on humans, but also to animals which secrete or sequester deleterious toxins.

THE STINGING HYMENOPTERA

Evolution of the sting

The most common stinging animals are in the insect order Hymenoptera (meaning "veil wings"), which includes the bees, wasps and ants. The most primitive Hymenoptera possess ovipositors to insert eggs into plant tissue. In some parasitic groups this structure and glands associated with it have been modified to inject venom to paralyze other insects which are used as food by the developing larvae. (These parasitic wasps comprise the largest number of species in the Hymenoptera, and are extremely beneficial to agriculture as biological control agents of agricultural pest insects.) The stings of these parasitic wasps are not usually very painful to humans, and their venom has low mammalian toxicity 4,5; it is not designed as a defensive sting, and the wasps which possess it usually flee rather than sting when disturbed. The sting has been specialized as a defensive weapon in other groups of wasps (and the ants and bees which evolved from these wasps). Most wasps and bees are solitary, and do not defend their nests, though they will sting in defense if caught. It is in the social Hymenoptera that we see active defense of the nest, and it is mostly these groups that cause medically significant stings, although a few large species can cause painful and potentially serious stings if captured or stepped on.

Fig. 1. Bees and wasps. Top row (left to right): bumble bee (Bombus sonorus), carpenter bee (Xylocopa californica arizonica). Bottom row (left to right): paper wasp (Polistes fuscatus aurifer), German yellowjacket (Vespula germanica), European honey bee (mix of several Apis mellifera subspecies), Africanized honey bee (Apis mellifera scutellata from Tucson, Arizona). Except the carpenter bee, all insects shown are social, and create nests inhabited mostly by sterile workers. (Photo © P. K. Visscher)

The wasps of medical interest mostly are found in the superfamily Vespoidea which includes yellowjackets and paper wasps (Fig. 1). The sting is a modified ovipositor, only females (queens and sterile workers) can sting, so males are always harmless; however, when grasped, males of some species mimic the female stinging behavior with a quick-thrusting, ventrally-curled abdomen which can cause even entomologists to second-guess their field identification and release them.

Bees and ants are derived evolutionarily from two different lineages of wasps. Bees have given up the carnivorous lifestyle of their wasp ancestors and gather protein from flowers, as pollen. While most bees possess a defensive sting, and will sting if grasped or crushed, only a few social species sting often enough, or have sufficiently venomous stings to be of medical significance. These include the honey bees (genus Apis) and the bumble bees (genus Bombus) (Fig. 1). Among the ants all species are social, but many no longer possess stings. A few species of ants do have medically important stings.

Due to the importance of the stinging Hymenoptera, much research has been done on their venom chemistry and its effects 1,4-6. Local effects of the sting are due to toxins in the venoms, and are for the most part similar within the different groups of Hymenoptera. The more serious effects are systemic responses to allergens in the venom, and are often quite specific, requiring sensitization to the stings of the particular species involved. Hymenopteran stings are able to repel even large vertebrates by causing pain out of proportion to the wound inflicted. They do so by activating pain receptors, "tricking" the victim into thinking that a sting is causing significant tissue damage 4,5. A relative human pain index for an extensive listing of hymenopterans is available 5. The allergenic aspects of hymenopteran venoms probably have evolved for its effects on vertebrates as well 5, an interpretation supported by the observation that the venom of honey bee workers contains major allergens which are absent in the queen honey bees, who use their venom solely against other queens 6.

BEES

Bees constitute the superfamily Apoidea, and are worldwide in distribution. Most bees are solitary, and unlikely to sting, and, even then, most small bees have little venom. Among the social bees, the sting is used in defense of the nest (except that one large group of tropical social bees, the meliponine or stingless bees, no longer have stings).

Bumble bees (Bombus, in the family Apidae) are large bees which make annual nests which may contain up to 200 bees. Although their stings are painful and they aggressively defend their nest, their smaller nest population causes them to pose less threat to humans than honey bees. Bumblebee stings may become more common now that bumblebees colonies are are being cultivated for crop pollination 7. Carpenter bees (Xylocopa, in the family Anthophoridae) are large solitary bees whose nests in wood often bring them into proximity with people, whom they occasionally sting usually when grabbed (Fig. 1). The venoms of bumble bees and carpenter bees have been less well researched than that of honey bees, but are similar, and some venom constituents do cross-react in sensitized individuals 8.

There are 7 species of honey bees ranging in size from the dwarf honey bee, Apis florea, (about 1 cm) to the large Himalayan honeybee, A. laboriosa (about 3 cm). They are native to Europe, Asia and Africa and although all Apis bees share many characteristics, there are differences in basic life history traits such as nest construction, colony size and foraging behavior that distinguish them. All, however, defend their nests with stings; the pain they inflict is relative to the size of the bee. The largest colony populations (up to about 100,000 bees) are found in the common honey bee, Apis mellifera, (Fig. 1) and various subspecies of this species are used worldwide for honey production, with the result that bees of this species are the the most common single source of insect stings (though outnumbered by stings from all wasp species). In recent years, much attention has been focused on the African subspecies, the so-called "killer bee", A. m. scutellata (formerly A. m. adansonii) which was transported to Brazil and has spread throughout the Americas. Progress of the African or Africanized honey bee has been reviewed in south and central America 9,10 and the USA 11.

Fig. 2. The detached sting of a honey bee in human flesh. This characteristic feature readily distinguishes between the sting of a honey bee and a wasp and may be critical in treatment of potentially allergic individuals. (Photo © P. K. Visscher).

One aspect that distinguishes honey bees from the other hymenopterans is that the attachment of their sting is designed to tear and separate from the abdomen when stinging vertebrate tissue (Fig. 2). The detached sting is a useful diagnostic feature for verifying honey bee stings, since the stings of other bees and wasps do not normally detach. People commonly confuse bees and wasps, and it may be critical to discriminate between honey bee and other stings in the case of known allergenic response; there is relatively little cross-reactivity between the major groups of stinging hymenopterans in most sensitive individuals 4.

The adaptation for the separation of bee stings accomplishes three things. First, the detached sting continues to move and embed itself deeply into the flesh. Second, venom continues to be pumped into the wound, reaching more sensitive tissues. Third, the sting produces communicative chemicals called alarm pheromones. The principal component of this pheromone is isopentyl acetate, which smells like bananas. Alarm pheromones evaporate from the surface of the sting, alert other bees, and attract them to a proven vulnerable area. The detachment of the sting appears to be an adaptation for defense specifically against vertebrates; when honey bees sting other insects, the sting remains intact.

Fig. 3. Schematic drawing of the sting apparatus of a honey bee

The sting shaft is composed of a stylet with two lancets which alternately slide back and forth along its length when moved by muscles in the base of the sting (Fig. 3). The lancets are barbed at the tip, and hold in the flesh, so the alternating contractions pull the sting into the flesh. Attached to the lancets are two membranous pump diaphragms; as these move within a venom-filled chamber at the base of the sting, they pump venom down a channel in the sting shaft.

Until recently, virtually all conventional medical advice for removal of bee stings state that one should never pinch the sting during removal as this might squeeze more venom from the attached venom sac through the sting 3,12. In reality, the venom delivered through the sting shaft depends on the pump mechanism, not the venom sac 13. Visscherm et al 14 recently showed that envenomation does not differ if the sting is scraped or pinched off, but that even delays of a few seconds in removing the sting (such as getting an instrument for scraping) lead to substantially greater venom delivery. The advice to victims of bee stings should now be to remove the sting as rapidly as possible, without regard to the method used.

Physiology Of The Sting

The active components of honey bee venom include enzymes, other smaller proteins and peptides, and amines. The principal small proteins and peptides are melittin, apamin, and peptide 401 15,16. Melittin constitutes about 50% of the venom dry weight; it hydrolyzes cell membranes causing changes in permeability and is most responsible for the pain associated with the sting. Peptide 401 is also known as "mast cell degranulating peptide" and causes mast cells to release histamine as they degranulate, setting up an inflammatory reaction. Enzymes include phospholipase A2 (11% of dry weight), which is non-toxic when pure but in concert with melittin is a major hemolytic factor 6. Phospholipase A2 is a major venom allergen and is responsible for inducing IgE-mediated anaphylaxis 4,6,16. Hyaluronidase, which is also common in the venom of other animals (e.g., spiders, snakes), causes changes in cell membranes and is considered the major "spreading factor". It is also the second most common allergen in honey bee venom 6. Honey bee venom also includes some physiologically active amines (histamine, dopamine, norepinephrine) 15,16.

Most fatalities from bee (and wasp) stings occur in hypersensitive individuals; death is most often induced by a single sting, and occurs most often within 1 hour after the sting. The victim is typically over 40 years of age and stung on the head or neck 4,17. Most deaths are caused by respiratory dysfunction with the second most common cause being anaphylaxis; arteriosclerosis may be a compounding factor 4. Large numbers of bee stings can also cause death in non-hypersensitive individuals. The LD50 of bee venom for a human has been estimated to be 500-1500 stings 18,19. Mejia et al 20 cites five people receiving >1000 stings, who manifested acute renal problems, yet four of the five survived.

Dermatologic Expression

The typical dermatologic expression for honey bee venom is a raised white weal with central red spot of about 10 mm which appears a few minutes after the sting, and lasts for about 20 min. Honey bee stings can usually be definitively diagnosed by the presence of the detached sting, which will remain in the wound until removed. There may be edema and pruritis; the initial intense pain will last only minutes and symptoms should resolve in a few days. Patients who have been sensitized by prior stings may display large, local reactions including 10-50 cm edematous swellings forming 4-12 hr post-sting and persisting for 3 days 4. Symptoms, especially edema, are much more pronounced when the sting is delivered to the face and neck. Many victims with edema and pruritis think they are allergic but these are the typical localized symptoms of honey bee envenomation. Occasionally similar large reactions occur at sites not adjacent to the sting site, and these are considered systemic rather than local reactions.

European and Africanized Honey Bees

Africanized honey bees (AHBs) are hybrids between an African subspecies of Apis mellifera and the mosaic of several European subspecies used in New World beekeeping. AHBs and European honey bees (EHBs) are extremely similar (Fig. 1), and can only be reliably distinguished by microscopic examination or molecular genetic methods. Even these give only "probability of Africanization", since the two forms readily interbreed. AHBs and EHBs differ slightly in many behavioral and life history traits such as swarming, foraging, cavity preference size, etc. 21 However, the special concern regarding AHBs is their heightened defensive reaction which can lead to very large numbers of stings. AHBs colonies are more easily disturbed to the point of stinging, sometimes by the presence of a person within tens of meters of their colony. When a colony is disturbed, AHBs are likely to respond in much larger numbers than EHBs, and be more persistent in following a person up to hundreds of meters from the colony. AHBs are slightly smaller than EHBs and also have slightly less venom, though the venom is chemically very similar 22,23.

Because bees generally only sting to defend their nest or if caught, foraging bees on flowers do not pose a threat, whether AHBs or EHBs. When bees of either group are defending a nest, they locate the intruder by cues typical of mammalian predators and attack objects that are darkly-colored, hirsute, moving, sweating, and/or emitting CO2. Once stung, a victim is also located by the alarm pheromones released by the stings.

WASPS

Wasps are an extremely diverse group with worldwide distribution. Most wasps are solitary and harmless, but some social forms exhibit extremely fierce defense of their nests. Wasp stings from all species are the most common human envenomation. Wasps differ from bees in appearing less hairy and often the thin connection between the abdomen and thorax is apparent in wasps (thus a "wasp waist"), while in most social bees and some wasps (e.g. yellowjackets) these two parts, though still narrowly separated, appear more broadly joined and "stocky". In the north temperate zone the most common social wasps are yellowjackets and hornets (Vespula, Dolichovespula and Vespa) and the paper wasps (Polistes) (Fig. 1). In the tropics there is a larger diversity of social wasps, some with very large nests. From a medical viewpoint the stings of all of these wasps are similar, except that hypersensitivity reactions may be quite species specific.

Vespula yellowjacket wasps have spread via human commerce and have become established in many non-native areas (Australia, New Zealand, South America, South Africa) where they have flourished; much effort is now spent trying to eradicate them. Vespa hornets were originally only in the eastern hemisphere, but one European species (Vespa crabro) is now established, though uncommon, in North America. Some yellowjackets are common visitors to picnics and other exposed food, and people are stung when shooing away wasps, walking barefoot near fallen, fruit or swallowing wasps that have crawled into beverage cans. In some cases the abundance of these wasps has closed picnic and camping areas, caused problems in pastry and confectionery shops, prevented pickers from harvesting fruit, and attacked castration wounds and other wounds of livestock.

Paper wasps build smaller umbrella-shaped hanging nests, often on the eaves of houses. Their colonies are much smaller than yellowjackets, and the wasps do not defend them as fiercely, but stings are fairly common since these wasps are in such close contact with humans. In tropical areas, wasps related to the paper wasps form large colonies, and are some of the most fearsome stinging insects.

Physiology Of The Sting

Vespid wasp venoms, like bee venoms, contain enzymes, smaller peptides, and amines, but differ in the particular compounds involved 15,24-26. The collective action of these venom components is to cause pain (by either directly affecting neurons or indirectly by liberating pain-inducing compounds) 5,25. Enzymes include phospholipases, hyaluronidases, and cholinesterases. These are principally responsible for allergic responses to wasp venoms, and are not generally cross-reactive with the bee venom allergens 15,27. The peptides include mastoparans which cause histamine release by degranulating mast cells, and kinins. Histamine, serotonin, and acetylcholine affect neurons and contribute to the pain of these stings.

Dermatologic Expression

The local reaction is edema, erythema (due to increased blood vessel permeability) and pruritis 25,26 which may last several days. Unlike bee stings, the stings of wasps only occasionally remain in the wound 28. In most respects wasp and bee stings are very similar.

ANTS

There are about 10,000 species of ants in the world; they are among the major predators in most terrestrial habitats but, mostly, do not pose a health problem 29. The venom glands in ants have undergone major evolutionary changes and have been modified in some species to serve as pheromones for communication, coordinating the actions of the thousands of individuals that make up an ant society. Also many ant species have no sting. In those that do, the individual stings are generally not very painful or serious. Nevertheless, it is common for tens or hundreds of ants to sting when their colonies are disturbed, and these situations can be of medical significance.

In the USA, the principal ants of concern are fire ants and harvester ants. Red imported fire ants (Solenopsis invicta) from South America have become established in the USA from Florida to Texas and are extending their range 30. In areas where fire ants occur, 30-60% of urban populations are stung annually 31-33. These ants have large colonies, and respond to disturbance with vigorous mass stinging. The ants and stinging events are most common outdoors, but sometimes massive fire ant envenomations can occur indoors. There are also several species of native fire ants in the USA, whose stings are similar but less common. Harvester ants (Pogonomyrmex), which are common in arid areas of the western USA and Mexico 34, have painful stings and large colonies but are less defensive than fire ants and hence, not as problematic.

In Central and South America, in addition to fire ants, ants of the genus Paraponera (25 mm long) cause edematous wounds 35 and are of human health significance as demonstrated by the many common names among the people living within its range (hormiga bala, viente cuatro hora hormiga, munuri, chacha) 29. Schmidt 4 has been stung by a wide spectrum of bees, ants and wasps and states that the sting of P. clavata is excruciatingly painful and causes the greatest pain of any hymenopteran he has experienced. In Australia, the 10-40 mm long bulldog ants (Myrmecia) can cause allergenic responses 36.

Physiology Of The Sting

Unlike the protein rich venom of bees and wasps, most (95%) of the critical components of fire ant venoms are alkaloids (methyl-n-piperidines) which inhibits Na+-K+-ATPase pump of muscle cell membrane resulting in postsynaptic neuromuscular blockade 36,37. Damage is expressed in hemolytic, cytotoxic and neurologic reactions 32,36,37. Allergenic proteins comprise about 0.1% of the venom content and are responsible for the IgE-mediated systemic reactions; anaphylaxis is rare (<1-2%)32,36,38. Verified death has occurred via anaphylactic shock from 1 to less than 150 stings 36,39 but some victims have survived 5,000-10,000 stings without toxic reaction; these extreme cases typically involve incapacitated victims such as an invalid and an alcoholic who used a fire ant nest for a pillow40,41. Even a 5-day old infant sustained about 2000 stings and survived but only with intensive supportive measures 42.

Fig. 4. The pathognomic pustules following a fire ant envenomation. (Photo courtesy of the USDA).

The venom of Pogonomyrmex contains phospholipases A2 and B, lipase, hyaluronidase, acid phosphatase and is one of the most potent hymenopteran venoms 43. The venom causes hemolysis, mass cell degranulation and is 100 times less toxic to arthropods than to mammals, suggesting that it was shaped by natural selection as a defense against mammals.

Dermatologic Expression

The typical reaction to a fire ant sting progresses as follows. 31-33 Initial burning pain at the sting site is followed by formation of a weal about 10 mm in diameter, and accompanied by pruritis and edema. In about 4 h, sterile vesicles form, filled with clear fluid; they become turbid and at about 24 h, form white necrotic pustules. In about 50% of cases a biphasic reaction occurs, consisting of a weal-and flare within 30 min, short retraction of symptoms and then a hypersensitivity reaction involving edema, erythema, pruritis and pain (after 6 hr) which peaks at 24 h 33. Fire ants may sting several times in quick succession as they run across one's skin or bite and then sting as they rotate around the point of mandible attachment, leaving either a line or ring, respectively, of pathognomonic pustules (Fig. 4). Pustules remain for 3-10 days, and may rupture and leave a small scar. If ruptured, secondary infection is a concern.

Stings of Pogonomyrmex and Paraponera are quite algogenic 29. Reactions to Pogonomyrmex stings require 5-30 s for onset before piercing pain is felt. Pain may last several hours and has been described as "ripping muscles or tendons" and like "turning a screw in the flesh" where stung 29. Besides the excruciating pain of Paraponera stings, pain may wax and wane up to 24 h accompanied by parasthesia, vomiting, inflammation and trembling 29.

SPIDERS

There are 34,000 species of spiders described and an estimated total of 170,000 species in the world 44. Of these only a handful are considered to be dangerous. Spiders have an undeserved reputation for being nocturnal ne'er-do-wells, and the fear of spiders (arachnophobia) is very common, varying from mild disgust to a debilitating aversion 45. This is exacerbated by news reports which exaggerate the dangers of spider bites far out of proportion to the actual threat they pose, especially in relation to many of our daily activities. This validation by the media spills over to medical diagnoses where physicians ascribe "potential spider bite" to many lesions of unknown origin (see below). In comparison to the Hymenoptera which usually only elicit fear in direct relation to sting aversion and the insect's physical presence, there is a psychological aspect with spiders where arachnophobia does not require the presence of the spider to cause a negative response. Furthermore, in many cases no negative associative learning event needs to occur to initiate arachnophobia 46.

Whereas the compounds in Hymenoptera venom are similar throughout the group, as well as is the human reactions to stings, the pharmacology of spider venoms is quite variable. For this reason identification of the spider involved is medically significant, and our review below puts more emphasis on identification than it has for Hymenoptera.

WIDOW SPIDERS

Spiders of the genus Latrodectus (of the comb-footed spider family Theridiidae) are worldwide in distribution 47. Females range from 8 to 15 mm in body length, with globose, glabrous abdomens of predominantly shiny black coloration with red markings 48 (although some may be pallid and/or striped 49) and are recognized as dangerous, even deadly, by cultures throughout their range. They are known by many common names such as black widow (North America), shoe-button spider (South Africa), katipo (New Zealand), red-back (Australia), malmignatte and karakurt (Europe). These spiders are nocturnal and build a seemingly haphazard tangle of silk for their web. In nature, they are found in rock outcroppings, and under logs and stones, but they readily adapt to human-altered environments, where they are most commonly found in outbuildings, undisturbed clutter, or primitive toilets where they hide under overhangs by the opening, feasting on the insects attracted to the toilet. Early medical reports of human envenomation by black widows often involved outhouses, and bites often occurred on the genitalia or buttocks 50-52. As indoor plumbing and the shift from rural to urban society increased, the incidences of bites shifted from these sensitive areas to the limbs 53,54. Now, Latrodectus bites occur most frequently when the spider is trapped against human skin either by reaching under objects where the spider is hiding or when putting on clothing, gloves or shoes containing the spider. Latrodectus spiders are generally very timid and only bite as a last defense when they accidentally contact humans. Few bites are reported from immature black widows, whose short fangs may not penetrate the skin, but the changes in appearance as the spiders mature also complicate proper identification. Only one male bite has been documented 55 and, like immatures, males have small venom glands and short fangs. The extensive literature on the effects of human envenomations by Latrodectus are reviewed in 54,56-59.

Fig. 5. Dorsal view of dissected paired venom glands of the western black widow spider, Latrodectus hesperus. (Photo © P. Kirk Visscher)

Physiology of the bite

The venom produced in the Latrodectus venom glands (Fig. 5) contains several protein fractions of which a-latrotoxin is the neurotoxic fraction. It causes massive release of neurotransmitters, changes in ion channels and inhibits vesicle recycling. Symptoms are systemic, spreading through the lymphatic system and usually start about 1-3 h post-bite, whereupon the patient seeks treatment. The most common symptoms are intense pain, rigid board-like stomach muscles (occasionally misdiagnosed as acute abdomen but not tender upon palpation), muscle cramping, malaise, local sweating, nausea, oliguria, vomiting, and hypertension; other symptoms include priapism, fear of dying, brady- or tachycardia, and insomnia 56,60. Although abdominal rigidity is one of the clinically diagnostic symptoms in North American Latrodectus bites, it does not appear to be common in L. hasselti bites in Australia 52,53.

If left untreated, Latrodectus bite symptoms usually last 3-5 days. Calcium gluconate and/or antivenom may be administered to relieve or counteract symptoms 54,57. Prior to antivenom availability, Latrodectus spiders killed approximately 5% of its human victims 61. It has been argued that antivenom should be avoided, since other therapeutic measures can readily counter symptoms and there is chance of allergenic reaction to horse-serum antivenom 60. Because of the efficacy and the relief that it provides (sometimes within minutes 62,63), however, others advocate using Latrodectus antivenom more frequently 56. Deaths have been attributed to pulmonary edema 56. Despite the effects Latrodectus venom has on muscle activity, there is no tendency for pregnant women to abort; typically birth is unremarkable and babies healthy57,62,64.

Dermatologic expression

The bite site may have localized erythema and two fang marks may be seen where the skin was penetrated. Petechial branching from the site may also be visible. In untreated cases, a rash may develop after several days 56,57. Diagnosis is more assured from the definitive systemic symptoms.

FALSE BLACK WIDOWS

Spiders of the genus Steatoda (Family: Theridiidae) are very common throughout the world, most being small and harmless. We include them here because they are very often mistaken for Latrodectus spiders. S. grossa (worldwide) and S. paykulliana (Europe) are slightly smaller than Latrodectus, make a similar-looking web in the same habitat, have a globose abdomen, and are shiny brown: however, they never have an hourglass or other red markings. They are considered to be capable of producing a bite with weak Latrodectus symptoms 65 and one envenomation lead to blistering and fatigue for several days66. Proper spider identification will preclude the overzealous use of Latrodectus antivenom in Steatoda bites.

VIOLIN OR RECLUSE SPIDERS

Members of the spider genus Loxosceles (Family: Sicariidae) are most commonly known as violin, fiddleback and recluse spiders. These spiders are found worldwide, most commonly in the tropics with some species reaching temperate latitudes 67-69. Because many wounds are erroneously attributed to this spider in the USA and because many spiders are misidentified as Loxosceles, we will elaborate on their proper identification.

Fig. 6 . Dorsal view of a brown recluse spider, Loxosceles reclusa, showing the violin pattern on the cephalothorax. (Photo © P. Kirk Visscher)

Loxosceles spiders are medium-sized (6-10 mm body length), rather nondescript with uniformly colored abdomens that can vary from a tawny to dark brown. The legs and body are covered with fine hairs; they have no large leg spines as in many other brown spiders mistaken for them. On the dorsum of the cephalothorax (the first body segment, to which the legs are attached), in many species there is a characteristic darkened violin pattern (Fig. 6) which lends its name to the spider. However, this marking is variable, and other unrelated spiders may have a pattern which can easily be mistaken for the violin. A more consistent method of determining Loxosceles spiders is the eye pattern. Whereas most spiders have eight eyes arranged in two rows of four, Loxosceles spiders have six eyes arranged in pairs (dyads) with one anterior dyad and a lateral dyad on each side (Fig. 7). There are several related non-poisonous spiders that have a similar pattern but they usually have distinct stripes in place of the violin marking and/or additional markings on the abdomen.

Fig. 7 . Anterior view of a brown recluse spider, Loxosceles reclusa, showing the configuration of the 6 eyes in 3 dyads. (Photo © R. S. Vetter)

Loxosceles spiders make a protective silken retreat that occasionally can entangle prey, but they are more often hunters that prowl for prey 70. They can be found in high numbers in human structures 71,72; in South America, it is known as "the spider that lives behind the picture"71. Their nocturnal wanderings place them in contact with humans. Similar to widow spiders, Loxosceles spiders usually bite only when they become trapped next to the victim's skin. Bites occur either when sleeping humans roll onto the spider or put on clothes into which the spider has crawled.

Physiology of bite

Brown recluse bites have been well reviewed 56,58,59,73-77. Bites from Loxosceles can be unremarkable (requiring no care), localized (requiring some care but usually healing without intervention), dermonecrotic (a slow-healing, necrotic ulcerated lesion needing supportive care) or systemic (vascular and renal damage, sometimes life-threatening). Within 10 min of venom injection, there is a constriction of capillaries around the site of the bite. A major venom component is sphingomyelinase D which causes hemolysis. Recluse venom has a strong disruptive effect on endothelial tissue 76,78. Polymorphonucleocytes (PMN) are activated and infiltrate the bite site; in test animals where PMN activity was suppressed, degree of necrosis was lessened 79. A detailed review of the physiological mechanisms at the cellular level can be found in 78. General symptoms are edema, erythema, pruritis, pain at the site, fever. Systemic conditions that might manifest in severe cases are hematoglobinuria, hematoglobinemia, thrombocytopenia, disseminated intravascular coagulation (DIC). Although rare, if death occurs, it is most often from hemolysis, renal failure and DIC; children are most adversely affected due to their small body mass.

Dermatologic expression

This venom is an interesting one from the dermatologic standpoint 73-76. In self-healing wounds, the bite site may not progress past an edematous erythema; these wounds do not become necrotic and non-intrusive care is sufficient. In more serious wounds, a "bull's-eye" wound may form where a central erythematous bleb is separated from a peripheral cyanotic region by a white zone of induration. If the bite becomes violaceous with the first few hours, this usually indicates that severe necrosis may occur 74,75 and more supportive measures are necessary. The initial bleb gives way to ischemia. A central eschar forms, hardens and within 7-14 days the eschar falls out leaving behind an ulcerated depression. The necrosis may continue to spread from the bite site possibly due to an autoimmune response73,74.

Symptoms start 2-6 h after the bite73. By 12-24 h, it is usually apparent if a Loxosceles wound is going to become necrotic; if necrotic symptoms do not express by 48-96 hr, then they will not develop 59. Antivenom was very successful when administered within 24 h 80 but many times, a victim does not seek treatment until after necrosis is well underway (more than 24 h) and therefore, antivenom is less effective. Systemic effects usually take 2-3 days to show symptoms 59. Bites that become systemic do not also become necrotic; it is thought that in necrotic wounds the venom is localized in the tissue whereas in systemic reactions the venom is distributed quickly into the body and without necrotic local effects 59. The wound is usually free of bacterial infection for the first 2-3 days but may be contaminated by patients due to pruritis 74. Recluse venom can exhibit extended necrosis in adipose tissue of thighs, buttocks and abdomen of obese patients; there is also a gravitational flow of the venom effects 59,74,76. Healing can take weeks to months and may leave a unsightly scar.

Previously, extirpation of skin around the wound was considered the best form of treatment but more conservative treatment is now advocated 59,74,76. Extirpation of damaged skin is only recommended in severe cases and only after the limits of the wound are strongly demarcated at 6-8 weeks 75. Medication (Dapsone) used to be prescribed to prevent the emancipation of PMN's, but Dapsone also causes hemolysis and methemoglobinuria, and can be detrimental for those with glucose-6-phosphate dehydrogenase deficiency59,73. Hyperbaric oxygen (HBO) treatments show promise 81,82 but their efficacy has been questioned 83. In recent experiments using piglets challenged with Loxosceles venom, there were no differences using Dapsone, HBO or a combination when compared to controls 84 although only one dosage/exposure level was tested.

HOBO SPIDER

In the last decade, a spider has been implicated in dermonecrotic lesions in the US Pacific Northwest 85-88. This spider, Tegenaria agrestis (Family: Agelenidae), has been called by 5 different common names but recently was bestowed the official common name of "hobo spider" 89. The hobo spider is common around and inside human structures, is 7-14 mm in body length, brownish in color often with a dorsal herringbone abdominal pattern 90. It builds a trampoline-like horizontal mat of silk which retreats into a hole or crack in brick walls or wood piles where the spider awaits prey to fall onto the web 90. It is found from Washington, east to Montana and south to Oregon and northern Utah91. Unlike the spiders mentioned above, hobo spiders can be somewhat aggressive and may bite with little provocation 90. Identification of this spider is quite difficult since there are many other spiders that are similar in habit and coloration; identification should be confirmed by a qualified arachnologist.

Hobo spider venom can cause dermonecrotic lesions with symptoms similar to that of Loxosceles bites but with lesser severity. It is often accompanied by severe headaches lasting 2-7 days and less often by malaise and vision abnormalities 88,90. Severe lesions may take several months to heal and the hobo spider has been implicated as a contributory factor of at least one death 87. Unexplainably, although this spider originates from Europe, it is not considered poisonous there 92 possibly because it is not often found around human structures90,93.

YELLOW SAC SPIDERS

Sac spiders (Family Clubionidae) of the genus Cheiracanthium are found worldwide. They typically are medium-sized (6-10 mm body length), pale yellow with black chelicerae (the anterior structures which house the fangs). The body color can be yellow, greenish, pink or tan dependent on prey recently eaten. In nature, they make silk retreats in curled leaves or in crevices, are nocturnal hunters. These spiders are found often in homes, and bites occur at night when the spiders encounter a sleeping human 94. There are over 100 species worldwide and several have caused envenomations; in the USA, bites have occurred from Massachusetts to California and in Hawaii.95-98.

The bites of Cheiracanthium spiders are accompanied by erythema, edema and pruritis. Many are painful at inception 96-98 and can produce slightly necrotic wounds 99,100 which usually heal in short duration without severe scarring.

BANANA SPIDERS

These are large (30 mm body length) active hunting spiders of the genus Phoneutria (Family: Ctenidae) with bright red hairs on its chelicerae; these spiders live in Central and South American. Whereas most spiders are timid and resort to biting as a last defense, Phoneutria aggressively defends itself, menacingly holding its front pair of legs upward toward the offending stimulus and repositioning itself to face its adversary if a person circles around the spider 72,101. They are found often in rubbish around human dwellings as well as hiding amongst foliage such as banana leaves where they sometimes bite workers harvesting bananas 101.

The bite of this spider is neurotoxic and hence expresses little dermatologic damage. Symptoms include immediate pain, cold sweat, salivation, priapism, cardiac perturbations and occasional death. In vivo studies suggest it is similar in action to a-latrotoxin 56,101.

AUSTRALIAN FUNNEL-WEB SPIDERS

These spiders (Atrax and Hadronyche; Family: Hexathelidae) are large (24-32 mm body length), aggressive and are found along the eastern coast of Australia 56,102. The most dangerous species is the Sydney funnel-web spider, Atrax robustus. These spiders can bite with little provocation, the onset of symptoms can occur within minutes, and before antivenom was available, deaths were recorded within minutes for infants, 2 h for small children and 1 day for adults 103,104. Dermatologic expression is a moot point as the sudden onset of deleterious symptoms predominate.

Symptoms include frothing at the mouth, nausea, vomiting, tingling feeling, dyspynea, profuse sweating, drop in blood pressure, muscular twitching. Deaths occurred due to respiratory complications, pulmonary edema and cardiac arrest 105. On several occasions, the use of tourniquets isolated the venom in a limb until the victim reached medical attention; upon tourniquet removal, systemic envenomation symptoms violently manifested themselves within minutes 103,105,106. Since 1984, antivenom has been available to treat Atrax bites 105, mortality from these spiders is unlikely.

WHITE-TAILED SPIDER

Another Australian native, the white-tailed spider, Lampona cylindrata, (Family Gnaphosidae) has recently been implicated erroneously in envenomations that cause necrotic arachnidism, in part by the introduction of a bacterium via a bite. The spider is 15-20 mm in body length, covered with short dark hairs except for a patch of white hairs on the posteriormost dorsal aspect of the abdomen 102. Although its bite can cause cause minor problems 107, recent work has shown that there is little evidence to link any Australian spider with necrosis caused by Mycobacterium ulcerans108.

SIX-EYED CRAB SPIDERS

In Africa and South America 94,109, spiders of the genus Sicarius (Family Sicariidae) are considered extremely poisonous 94,110. They are very flat, six-eyed spiders (about 17 mm body length) found in remote areas where they bury themselves in sand. Because of this behavior, human contact and subsequent envenomations are rare. Their venom is proteolytic in action and although they are related to the recluse spiders, they appear not to share some characteristic symptoms (edema, erythema) found in Loxosceles envenomations110.

TARANTULAS AND OTHER MYGALOMORPH SPIDERS

Fig. 8. Ventral view of a tarantula, Aphonopelma sp., showing the parallel fang orientation typical of mygalomorph spiders. (Photo © R. S. Vetter)

Many of the large mygalomorph spiders (Family: Theraphosidae), referred to in the western hemisphere as "tarantulas", are virtually harmless despite their imposing size and large fangs (Fig. 8). Although humans do get bitten by these spiders, the rare reports in the medical literature describe mostly innocuous envenomations 56,59,111,112. Because of the increase of large spiders in the pet trade, the greatest risk posed is contact with the urticating hairs on the posterior of the abdomen. An irritated spider brushes the hairs off with it legs causing them to become airborne. Small barbs cause the hairs to become lodged in eyes or nasal cavities causing distress 113,114.

A CALL FOR PARSIMONY IN SPIDER BITE DIAGNOSES

Although some spiders listed above can have deleterious effects, not all bites lead to serious symptoms, and most may in fact be unremarkable. Typically, spiders bite only in defense, only once, and the bite is rarely felt. Multiple "bites" or an initially painful bite would exonerate most spiders as probable culprits. There are many arthropods which actively seek to bite humans, and these seem more logical candidates for suspicion 115.

In one study of 600 suspected "spider bite cases" in southern California, 80% were caused by other arthropods, mostly ticks and reduviid bugs 115,116. Loxosceles spiders have been erroneously blamed for a variety of necrotic wounds of unknown origin, which have often turned out to have other etiological agents (Table 1). It has been estimated that 60% of "alleged brown recluse bites" occur in areas where no Loxosceles spiders are known to exist 116. There are many causative agents of necrotic wounds, many not arthropod in nature, which can be or have been confused for recluse spider bites 74,116,118-120. A call for greater parsimony in spider bite diagnoses has been made more than once 74,88,116,117,121 and excellent guidelines for "verified", "probable" and "possible" spider bite diagnoses have been put forth118.

Table 1. Conditions which can cause necrotic wounds, have been misdiagnosed as or could be confused with "brown recluse spider bite" (from 73,115-117,120)

erythema chronicum migrans (Lyme disease)	Stevens-Johnson syndrome
toxic epidermal necrolysis (Lyells syndrome)	G.C. arthritis dermatitis
erythema multiforme	infected herpes simplex
erythema nodosum	chronic herpes simplex
purpura fulminans	diabetic ulcer
bed sores	poison ivy/oak infection
focal vasiculitis	thromboembolic phenomena
periarteritis nodosa	lymphomatoid papulosis
pyoderma gangrenosum	sporotrichosis
warfarin poisoning	Rocky Mountain Spotted Fever
bite of the pajahuello (Ornithodoros coriaceus, an Argasid tick)

Other causes of necrotic wounds are undoubtedly yet to be discovered. Some may turn out to be due to other spiders. In the late 1800's, some biologists scoffed at the idea that a tiny black widow spider could cause death 50,122, it wasn't until 1957 that the brown recluse spider was proven to be a cause of necrotic lesions in the USA 123, and the hobo spider was implicated in the last 10 years. Even the extremely poisonous Sydney funnelweb spider was not proven to be dangerous until 1927 124. These observations argue that accurate information from physicians can be invaluable in sleuthing out the causes of now-mysterious lesions, but the search for these causes is not helped by facile and unsupported diagnoses of "brown recluse bites".

In rare events where a victim brings in the "smoking gun" of a spider that was observed to bite, or the crushed corpse inside clothing after feeling a pinch or bite, this information is of great value, and one should seek out a trained arachnologist or entomologist for accurate identification. Even badly damaged spider remains can be examined by an arachnologist if not to confirm an identification at least to preclude an erroneous one.

SCORPIONS

Scorpions are distributed throughout the tropical and subtropical belts of the world in habitats ranging from dry deserts to the mountains. Like the other groups of venomous arthropods we have reviewed, only a small fraction of the existing species are known to have venom potent enough to endanger humans. Intriguingly, there are several cases of very potent and equally benign species belonging to the same genus 125. Almost all poisonous scorpions are found in the family Buthidae. Those considered most dangerous are found in the Middle East, Asia and Africa 125.

Fig. 9. Centruroides exilicauda (=sculpturatus) immature. Note the very thin fingers of the pedipalps. Many of the harmless North American scorpions can be differentiated from C. exilicauda because they have much more robust pedipalps. (Photo © P. K. Visscher)

In the USA, only one scorpion is considered to be dangerous, Centruroides exilicauda, (formerly known as C. sculpturatus) and is found mostly in Arizona, spilling over into western New Mexico, southeastern California, Texas and Mexico (Fig. 9). In Mexico, there are 134 scorpion species but only eight (all Centruroides spp.) cause concern 126.

As with most creatures that use venom only for defense, scorpions rarely sting their prey instead using their pedipalps and crushing chelicerae to acquire food.

Physiology of sting

Although poisonous scorpions are sprinkled across several genera taxonomically, the action of the venom is similar 125,127. Scorpion venom contains several polypeptides which interfere with neuronal ionic balance and channel activity. Stings manifest themselves mostly in the peripheral nervous system resulting in symptoms such as intense pain at the sting site, altered heart activity, parasthesia. Death is rare in scorpion stings; and is usually due to respiratory failure. Stings to children are much more dangerous and more often lethal 125. Where antivenom are available, it is very effective in counteracting the effects of the sting. Although scorpion sting used to be a fatal threat in Arizona and Mexico 128, the availability of antivenom has made fatalities very uncommon 126,129.

Dermatologic expression

Because scorpion venom effects are mostly neurological, there is virtually no dermatologic development other than edema and erythema around the sting site 127. Similar to defensive hymenopteran stings, however, the scorpion sting is usually immediately intensely painful as well as forcefully administered so there is a good chance it will be recognized by the victim as the causative agent.

In contrast to other scorpion stings, the sting of C. exilicauda is diagnostic in that the victim is hyperesthetic but never develops edema, erythema or ecchymosis 128,129. Pain usually subsides within 4 h and the total envenomation resolves in 24 h 129. With antivenom, victims are typically asymptomatic with 90 min 128.

CONCLUSIONS

Considering that by far, most of the species with which we share the Earth are arthropods, and that many thousands of these species possess venom, envenomation by arthropods remains remarkably uncommon. This is due to two factors. First, most venoms are used for capturing prey, and humans are too large to be prey for arthropods except parasitic and blood feeding species. Second, those arthropods which use their venom for defense (and most will if pressed) generally do so as a last resort.

The venom of social Hymenoptera (and scorpions) seems to have been evolved to repel large animals, and most of the stings of these insects show similar effects, with a combination of direct tissue disruption and pain-inducing chemicals, and allergens which induce local swelling and sometimes life-threatening systemic reactions.

The venom of spiders shows more variation, and is generally immediately less painful, but more likely to cause serious systemic or local direct toxic effects. The delayed response of these venoms underscores that they were not molded by evolution as deterrents, and also makes an unambiguous association between the responsible spider and the resulting lesions harder to ascertain. In this situation it has been all too common for physicians to ascribe undetermined effects to the "spider bites." The information reviewed here makes clear that this is counterproductive in advancing medical understanding.

Acknowledgment

We thank S. Camazine for comments which improved the manuscript.

REFERENCES

1. Piek T. ed. Venoms of the Hymenoptera: biochemical, pharmacological and behavioural aspects. Orlando, FL: Academic Press, 1986:1-570.

2. Meier J, White J, eds. Handbook of clinical toxicology of animal venoms and poisons. Boca Raton, FL: CRC Press, 1995:1-752.

3. Goddard J. Physician's guide to medically important arthropods. Boca Raton, FL: CRC Press, 1993:1-332.

4. Schmidt JO. Allergy to Hymenoptera venoms. In: Venoms of the Hymenoptera: biochemical, pharmacological and behavioural aspects. Piek T, ed. Orlando, FL: Academic Press, 1986:509-546.

5. Schmidt JO. Hymenopteran venoms; striving toward the ultimate defense against vertebrates. In: Insect defenses: adaptive mechanisms and strategies of prey and predators. Evans DL, Schmidt JO, eds. Albany, NY: State Univ NY Press, 1990:387-419.

6. Banks BEC, Shipolini RA. Chemistry and pharmacology of honey-bee venom. In: Venoms of the Hymenoptera: biochemical, pharmacological and behavioural aspects. Piek T, ed. Orlando, FL: Academic Press, 1986:329-416.

7. Kochuyt AM, Van Hoeyveld E, Stevens EAM. Occupational allergy to bumble bee venom. Clin Exp Allergy 1993; 23:190-195.

8. Piek T. Venoms of bumble-bees and carpenter-bees. In: Venoms of the Hymenoptera: biochemical, pharmacological and behavioural aspects. Piek T, ed. Orlando, FL: Academic Press, 1986:417-424.

9. Taylor OR, Jr. The past and present future spread of Africanized honey bees in the Americas. Bee World 1977; 58:19-30.

10. Taylor OR, Jr. African bees: potential impact in the United States. Bull Entomol Soc Am 1985; 31(4): 15-24.

11. Visscher PK, Vetter RS, Baptista FC. Africanized bees, 1990-1995: Initial rapid invasion has slowed in the U.S. Calif Agric 1997; 51:22-25.

12. Mosbech H. Clinical toxicology of hymenopteran stings. In: Handbook of clinical toxicology of animal venoms and poisons, Meier J, White J, eds. Boca Raton, FL: CRC Press, 1995:349-359

13. Snodgrass RE. Anatomy of the honey bee. Ithaca, NY: Cornell Univ Press, 1956:1-334

14. Visscher PK, Vetter RS, Camazine S. Removing bee stings. Lancet 1996; 348:301-302.

15. Hoffman DR. Hymenopteran venom proteins. Adv Exp Med Biol 1996; 391:169-186.

16. Shipolini RA. Biochemistry of bee venom. In: Handbook of natural toxins, vol 2, Insect poisons, allergens, and other invertebrate venoms, Tu AT, ed. New York, NY: Marcel Dekker Inc. 1984:49-85.

17. Meier J. Biology and distribution of hymenopterans of medical importance, their venom apparatus and venom composition. In: Handbook of clinical toxicology of animal venoms and poisons, Meier J, White J, eds. Boca Raton, FL: CRC Press, 1995:331-348.

18. Camazine S. Hymenopteran stings: reactions, mechanisms, and medical treatment. Bull Entomol Soc Amer 1988; 34:17-21.

19. Michener CD. The Brazilian bee problem. Annu Rev Entomol 1975; 20:399-416.

20. Mejia G, Arbelaez M, Henao JE, et al. Acute renal failure due to multiple stings by Africanized bees. Ann Intern Med 1986; 104:210-211.

21. Winston ML. Killer bees: the Africanized honey bee in the Americas. Cambridge, MA: Harvard Univ Press, 1992:1-162.

22. Nelson DR, Collins AM, Hellmich RL, et al. Biochemical and immunochemical comparison of Africanized and European honeybee venoms. J Allergy Clin Immunol 1990; 85:80-85.

23. Schumacher MJ, Schmidt JO, Egen NB, Dillon KA. Biochemical variability of venoms from individual European and Africanized honeybees (Apis mellifera). J Allergy Clin Immunol 1992; 90:59-65.

24. Nakajima T. Biochemistry of vespid venoms. In: Handbook of natural toxins, vol 2, Insect poisons, allergens, and other invertebrate venoms, Tu AT, ed. New York, NY: Marcel Dekker Inc. 1984:109-133.

25. Nakajima T. Pharamcological biochemistry of vespid venoms. In: Venoms of the Hymenoptera: biochemical, pharmacological and behavioural aspects. Piek T, ed. Orlando, FL: Academic Press, 1986:309-327.

26. Piek T. Pharmacology of hymenoptera venoms. In: Handbook of natural toxins, vol 2, Insect poisons, allergens, and other invertebrate venoms, Tu AT, ed. New York, NY: Marcel Dekker Inc. 1984:135-185.

27. Valentine MD. Anaphylaxis and stinging insect hypersensitivity. JAMA 1992; 268:2830-2833.

28. Mulfinger L, Yunginger J, Styer W, et al. Sting morphology and frequency of sting autotomy among medically important vespids (Hymenoptera: Vespidae) and the honey bee (Hymenoptera: Apidae). J Med Entomol 1992; 29:325-328.

29. Schmidt JO. Chemistry, pharmacology, and chemical ecology of ant venoms. In: Venoms of the Hymenoptera: biochemical, pharmacological and behavioural aspects. Piek T, ed. Orlando, FL: Academic Press, 1986:425-508.

30. Vinson SB. Invasion of the red imported fire ant (Hymenoptera: Formicidae). Amer Entomol 1997; 43:23-39.

31. Adams CT, Lofgren CS. Red imported fire ants (Hymenoptera: Formicidae): frequency of sting attacks on residents of Sumter County, Georgia. J Med Entomol 1981; 18:378-382.

32. deShazo RD, Butcher BT, Banks WA. Reactions to the stings of the imported fire ant. NE J Med 1990; 323:462-466.

33. deShazo RD, Griffing C, Kwan TH, et al. Dermal hypersensitivity reactions to imported fire ants. J Allergy Clin Immunol 1984; 74:841-847.

34. Cole AC, Jr. Pogonomyrmex harvest ants: a study of the genus in North America. Knoxville, TN: Univ Tenn Press, 1968: 1-222.

35. Weber NA. The sting of an ant. Am J Trop Med 1937; 17:765-768.

36. Hoffman DR. Fire ant venom allergy. Allergy 1995; 50:535-544.

37. Fox RW, Lockey RF, Bukantz SC. Neurologic sequelae following the imported fire ant sting. J Allergy Clin Immunol 1982; 70:120-124.

38. Lockey RF. Systemic reactions to stinging ants. J Allergy Clin Immunol 1974; 54:132-146.

39. Hensel AE III, Schutze WH, Lockey RF. Death from imported fire ant, Solenopsis invicta Buren, confirmed by insect identification and autopsy. Ann Allergy 1983; 50:359-360.

40. Diaz JD, Lockey RF, Stablein JJ, Mines HK. Multiple stings by imported fire ants (Solenopsis invicta), without systemic effects. South Med J 1989; 82:775-777.

41. Smith JD, Smith EB. Multiple fire ant stings. Arch Dermatol 1971; 103:438-441.

42. Hardwick WE, Royall JA, Petitt BA, Tilden SJ. Near fatal fire ant envenomation of a newborn. Pediatrics 1992; 90:622-624.

43. Schmidt JO, Blum MS. A harvester ant venom: chemistry and pharmacology. Science 1978; 200:1064-1066.

44. Coddington JA, Levi, HW. Systematics and evolution of spiders. Annu Rev Ecol Syst 1991; 22:565-592.

45. Davey GCL. Characteristics of individuals with fear of spiders. Anxiety Res 1992; 4:299-314.

46. Jones MK, Menzies RG. The etiology of fear of spiders. Anxiety Stress Coping 1995; 8:227-234.

47. Levi HW. The spider genus Latrodectus (Araneae, Theridiidae). Trans Amer Micro Soc 1959; 78:7-43.

48. Kaston BJ. Comparative biology of American black widow spiders. San Diego Soc Natl Hist, Trans 1970; 16:33-82.

49. Lotz LN. Revision of the genus Latrodectus (Araneae: Theridiidae) in Africa. Navors nas Mus Bloemfontein 1994; 10:1-60.

50. Browning CC. Original investigations of spider bites in southern California. So Calif Practitioner 1901; 16:291-300.

51. Kirby-Smith HT. Black widow spider bite. Ann Surg 1942; 115:249-257.

52. Wiener S. Red back spider bite in Australia: an analysis of 167 cases. Med J Aust 1961; 2:44-49.

53. Sutherland SK, Trinca JC. Survey of 2144 cases of red-back spider bites. Med J Aust 1978; 2:620-623

54. Miller TA. Latrodectism: bite of the black widow spider. Amer Fam Physician 1992; 45:181-187.

55. Allred DM. An unusual spider bite. Great Basin Natur 1974; 34:82.

56. White J, Cardoso JL, Fan HW. Clinical toxicology of spider bites. In: Handbook of clinical toxicology of animal venoms and poisons, Meier J, White J, eds. Boca Raton, FL: CRC Press, 1995:261-329.

57. Maretic Z. Latrodectism: variations in clinical manifestations provoked by Latrodectus species of spiders. Toxicon 1983; 21:457-466.

58. Allen C. Arachnid envenomations. Environ Emerg 1992; 10:269-298.

59. Wilson DC, King LE, Jr. Spiders and spider bites. Dermatol Clin 1990; 8:277-286.

60. Moss HS, Binder LS. A retrospective review of black widow spider envenomation. Ann Emerg Med 1987; 16:189-192.

61. Bettini S. Epidemiology of latrodectism. Toxicon 1964; 2:93-101.

62. Russell FE, Marcus P, Streng JA. Black widow spider envenomation during pregnancy: report of a case. Toxicon 1979; 17:188-189.

63. Byrne GC, Pemberton PJ. Red-back spider (Latrodectus mactans hasselti) envenomation in a neonate. Med J Aust 1983; 2:665-666.

64. Handel CC, Izquierdo LA, Curet LB. Black widow spider (Latrodectus mactans) bite during pregnancy. West J Med 1994; 160:261-262.

65. Maretic Z. Araneism: with special reference to Europe. Belgrade, Yugo: Nolit Pub House, 1979:1-255.

66. Rutherford AM, Sutherland SK. Large blister formation after a bite by the common cupboard spider, genus Steatoda. Med J Aust 1989; 151:542.

67. Gertsch WJ, Ennik F. The spider genus Loxosceles in North America, Central America, and the West Indies (Araneae, Loxoscelidae). Bull Amer Mus Natl Hist 1983; 175:264-360.

68. Gertsch WJ. The spider genus Loxosceles in South America (Araneae, Scytodidae). Bull Amer Mus Natl Hist 1967; 136:117-174.

69. Newlands G. A revision of the spider genus Loxosceles Heinecken & Lowe, 1835 (Araneae: Scytodidae) in southern Africa with notes on the natural history morphology. J Entomol Soc So Africa 1973; 38:141-154.

70. Hite JM, Gladney WJ, Lancaster JL Jr, Whitcomb WH. Biology of the brown recluse spider. Univ Arkansas Agric Exp Station Bull #711, 1966:1-26.

71. Schenone H, Rojas A, Reyes H, et al. Prevalence of Loxosceles laeta in houses in central Chile. Am J Trop Med Hyg 1970; 19:564-567.

72. Bücherl W. Biology and venoms of the most important South American spiders of the genera Phoneutria, Loxosceles, Lycosa and Latrodectus. Amer Zool 1969; 9:157-159.

73. Young VL, Pin P. The brown recluse spider bite. Ann Plast Surg 1988; 20:447-452.

74. Wasserman GS, Anderson PC. Loxoscelism and necrotic arachnidism. J Toxicol Clin Toxicol 1983-84; 21:451-472.

75. Anderson PC. Necrotizing spider bites. Am Fam Physician 1982; 26:198-203.

76. Bernstein B, Ehrlich F. Brown recluse spider bites. J Emerg Med 1986; 4:457-462.

77. Williams ST, Khare VK, Johnston GA, Blackall DP. Severe intravascular hemolysis associated with brown recluse spider envenomation. Am J Clin Pathol 1995; 104:463-467.

78. Patel KD, Modur V, Zimmerman GA, et al. The necrotic venom of the brown recluse spider induces dysregulated endothelial cell-dependent neutrophil activation. J Clin Invest 1994; 94:631-642.

79. Smith CW, Micks DW. The role of polymorphonuclear leukocytes in the lesion caused by the venom of the brown spider, Loxosceles reclusa. Lab Invest 1970; 22:90-93.

80. Rees R, Shack RB, Withers E, et al. Management of brown recluse spider bite. Plast Reconstr Surg 1981; 68:768-773.

81. Svendsen FJ. Treatment of clinically diagnosed brown recluse spider bites with hyperbaric oxygen: a clinical observation. J Ark Med Soc 1986; 83:199-204.

82. Strain GM, Snider TG, Tedford BL, Cohn GH. Hyperbaric oxygen effects on brown recluse spider (Loxosceles reclusa) envenomation in rabbits. Toxicon 1991; 29:989-996.

83. Anderson PC. "Brown recluse spider bite: four decades of biomedical progress", 19th Annual National Meeting, Amer Arachnol Soc, Columbia, MO, 21 Jun 1995.

84. Hobbs GD, Anderson AR, Greene TJ, Yealy DM. Comparison of hyperbaric oxygen and dapsone therapy for Loxosceles envenomation. Acad Emerg Med 196; 3:758-761.

85. Vest DK. Necrotic arachnidism in the Northwest United States and its probable relationship to Tegenaria agrestis (Walckenaer) spiders. Toxicon 1987; 25:175-184.

86. Vest DK. Envenomation by Tegenaria agrestis (Walckenaer) spiders in rabbits. Toxicon 1987; 25:221-224.

87. Vest DK. Emergent patterns in the occurrence and severity of probable hobo spider (Tegenaria agrestis) envenomation in humans. Toxicon 1989; 27:84.

88. Fisher RG, Kelly P, Krober MS, et al. Necrotic arachnidism. West J Med 1994; 160:570-572.

89. Breene RG, Dean D, Edwards GB Jr. Common names of arachnids. S Padre Island, TX: American Tarantula Soc, 1995:1-94.

90. Akre RD, Myhre EA. Biology and medical importance of the aggressive house spider, Tegenaria agrestis, in the Pacific Northwest (Arachnida: Araneae: Agelenidae). Melanderia 1991; 47:1-30.

91. Baird CR, Akre RD. Range extension of the aggressive house spider Tegenaria agrestis into southern Idaho, Utah, and Montana during 1992 and 1993. Proc Wash St Ent Soc 1993; 55:996-1000.

92. Preston-Mafham R, Preston-Mafham K. Spiders of the world. Oxford, UK: Facts on File Publ, 1989:1-191.

93. Roberts MJ. Spiders of Britain & Northern Europe. London, UK: HarperCollins Pub, 1995:1-383.

94. Newlands G, Atkinson P. Behavioral and epidemiological considerations pertaining to necrotic araneism in southern Africa. S Afr Med J 1990; 77:92-95.

95. Anonymous. Chiracanthium diversum Koch. Proc Hawaiian Ent Soc 1962; 18:16

96. Furman DP, Reeves WC. Toxic bite of a spider, Chiracanthium inclusum Hentz. Calif Med; 87:114.

97. Krinksy WL. Envenomation of the sac spider Chiracanthium mildei. Cutis 1987; 40:127-129.

98. Minton SA Jr. Poisonous spiders of Indiana and a report of a bite by Chiracanthium mildei. J Indiana St Med Assoc 1972; 65:425-426

99. Spielman A, Levi HW. Probable envenomation by Chiracanthium mildei: a spider found in houses. Am J Trop Med Hyg 1970; 19:729-732.

100. Newlands G, Martindale C, Berson SD, Rippley JJ. Cutaneous necrosis caused by the bite of Chiracanthium spiders. S Afr Med J 1980; 57:171-173.

101. Lucas S. Spiders in Brazil. Toxicon 1988; 26:759-772.

102. Sutherland SK. Treatment of arachnid poisoning in Australia. Aust Fam Physician 1990; 19:47-64.

103. Sutherland SK. The Sydney funnel-web spider (Atrax robustus). 3. A review of some clinical records of human envenomation. Med J Aust 1972; 2:643-647

104. Ingram WW, Musgrave A. Spider bite (arachnidism): a survey of its occurrence in Australia, with case histories. Med J Aust 1933; 2:10-15.

105. Hartman LJ, Sutherland SK. Funnel-web spider (Atrax robustus) antivenom in the treatment of human envenomation. Med J Aust 1984; 141:796-799.

106. Fisher MM, Raftos J, McGuinness RT, et al. Funnel web spider (Atrax robustus) antivenom. 2. Early clinical experience. Med J Aust 1981; 2:525-526.

107. White J, Hirst D, Hender E. 36 cases of bites by spiders, including the white-tailed spider, Lampona cylindrata. Med J Aust 1989; 150:401-403.

108. Atkinson RK, Farrell DJ, Leis AP. Evidence against the involvement of Mycobacterium ulcerans in most cases of necrotic arachnidism. Pathol 1995; 27:53-57.

109. Gerschman de Pikelin BS, Schiapelli RD. Caracteres morfologicos validos en la sistematica del genero Sicarius (Walckenaer 1847) Araneae: Sicariidae. Acta Zool Lilloana 1979; 35:87-96.

110. Newlands G. Preliminary report on the medical importance of Sicarius (Araneae: Sicariidae) and the action of its venom. Mem Instit Butantan 1984; 46:293-304.

111. Lucas SM, Da Silva PI, Jr, Bertani R, Costa Cardoso JL. Mygalomorph spider bites: a report on 91 cases in the state of Sao Paulo, Brazil. Toxicon 1994; 32:1211-1215.

112. Schmidt G. Efficacy of bites from Asiatic and African tarantulas. Trop Med Parasitol 1989; 40:114.

113. Cooke JAL, Miller FH, Grover RW, Duffy JL. Urticaria caused by tarantula hairs. Am J Trop Med Hyg 1973; 22:130-133

114. Waggoner TL, Nishimoto JH, Eng J. Eye injury from tarantula. J Am Optomet Assoc 1997; 68:188-190.

115. Russell FE, Gertsch WJ. Letter to the editor. Toxicon 1982; 21:337-339.

116. Russell FE. A confusion of spiders. Emerg Med 1986; 18:8-13

117. Russell FE, Waldron WG. Letter to editor. Calif Med 1967; 106:248-249

118. Anderson PC. Loxoscelism threatening pregnancy: five cases. Am J Obstet Gynecol 1991; 165:1454-1456

119. Kunkel DB. The myth of the brown recluse spider. Emerg Med 1985; 17:124-128.

120. Rosenstein ED, Kramer N. Lyme disease misdiagnosed as a brown recluse spider bite. Ann Intern Med 1987; 107:782.

121. Anderson PC. Letter to the editor. Toxicon 1982; 20:533

122. Riley CV, Howard LO. A contribution to the literature of fatal spider bites. Insect Life 1889; 1:204-211.

123. Atkins JA, Wingo CW, Sodeman WA. Probable cause of necrotic spider bite in the Midwest. Science 1957; 126:73.

124. Sutherland SK. Venomous creatures of Australia. Melbourne, Aust: Oxford Univ Press 1995:1-128.

125. Keegan HL. Scorpions of medical importance. Jackson, MS: Univ Mississippi Press. 1980:1-140.

126. Dehesa-Davila M. Epidemiological characteristics of scorpion sting in León, Guanajuato, México. Toxicon 1989; 27:281-286.

127. Dehesa-Davila M, Alagon AC, Possini LD. Clinical toxicology of scorpion stings, In: Handbook of clinical toxicology of animal venoms and poisons, Meier J, White J, eds. Boca Raton, FL: CRC Press, 1995:221-238.

128. Curry SC, Vance MV, Ryan PJ, et al. Envenomation by the scorpion Centruroides sculpturatus. J. Toxicol Clin Toxicol 1983-84; 21:417-449.

129. Likes K, Banner W Jr., Chavez M. Centruroides exilicauda envenomation in Arizona. West J Med 1984; 141:634-637.