Bites attributed to the brown recluse spider, Loxosceles reclusa Gertsch & Mulaik, are frequently reported by medical personnel throughout Florida, whereas the extensive arachnological evidence contradicts the alleged widespread occurrence of Loxosceles spiders in the state. We compared reports of brown recluse spider bites made by medical personnel from a 6-yr Florida poison control center database to the known verifications of Loxosceles spiders from 100 yr of Florida arachnological data. Medical personnel diagnosed 124 brown recluse spider bites from 31 of Florida's 67 counties in 6 yr. In contrast, only 11 finds of ≈70 Loxosceles spiders have been made in 10 Florida counties in 100 yr. Florida does not have sufficient widespread populations of Loxosceles spiders to warrant consideration of brown recluse spider envenomation as a probable etiology of dermonecrosis. Florida health care would improve if medical personnel would consider the multitude of other etiologies that manifest in dermonecrosis.
The brown recluse spider, Loxosceles reclusa Gertsch & Mulaik, was not described as a North American source of necrotic skin lesions until 1957 (Atkins et al. 1957). However, since that time, the spider's threat to humans has been exaggerated such that brown recluse spider bites are now diagnosed throughout North America in states or regions where the spiders are extremely rare or have never been found (Russell 1986, Vetter 2000, Edwards 2001, Vetter and Bush 2002a, Vetter et al. 2003, Bennett and Vetter 2004). Paradoxically, despite this infamy, the documented information regarding its distribution is highly variable among American states. Recently, Wendell (2003) published a review regarding envenomation by brown recluse spiders; he pointed out that the paucity of definitive distribution information in concert with physician unfamiliarity with the spider's distribution in portions of the southeastern United States where the spider is not endemic can lead to overdiagnosis of bites.
Populations of L. reclusa are documented from southeastern Nebraska to the southernmost counties of Ohio, south into most of Texas to northern Georgia and the westernmost tip of South Carolina (Gertsch and Ennik 1983, Vetter 2000; R.S.V., unpublished data). Two uncommon, non-native species have very sporadically been found in the United States: Loxosceles rufescens (Dufour) (Mediterranean origin) and Loxosceles laeta (Nicolet) (South American origin). Florida is considered to be almost exclusively outside the endemic range of L. reclusa (except possibly the westernmost panhandle counties), yet bite diagnoses are commonplace throughout the state and have been made for decades. In endemic Loxosceles areas, individual homes can support dozens to thousands of Loxosceles spiders without envenomation of inhabitants (Schenone et al. 1970, Vetter and Barger 2002), suggesting benign behavior and extremely low risk. For Florida bite diagnoses to be correct, there should be substantial evidence of Loxosceles spider populations: homeowners and alleged bite victims would often submit Loxosceles spiders for verification, and arachnologists would be familiar with localities where the spiders could be reliably collected.
We attempt to demonstrate that bites from Loxosceles spiders are over-reported in Florida by comparing reports of bite diagnoses to the extensive arachnological information that is available in the state. We expect this article to fill current gaps of knowledge and give Florida medical personnel more tools with which to provide better assessment of necrotic skin lesions.
Materials and Methods
Information regarding the distribution of Loxosceles spiders in Florida was acquired from the taxonomic genus revision (Gertsch and Ennik 1983) as well as publications regarding Florida spider fauna involving Loxosceles spiders (Banks 1904, Whitcomb and Wallace 1972, Edwards 2001). Additional information is based on the experience of G.B.E. who has been curator of Arachnida and Myriapoda at the Florida State Collection of Arthropods for 25 yr (with 31 yr of Florida arachnology experience), curating an estimated 100,000+ Florida-collected spiders dating back to 1915. As part of state duties, G.B.E. has identified hundreds of spider submissions from the general public (many submitted as suspected brown recluses), has identified thousands of Florida spiders collected in various faunal surveys, investigates alleged discoveries of Loxosceles in Florida, and has written a state agency publication regarding the Loxosceles situation in Florida (Edwards 2001). In addition, because of its subtropical fauna consisting of many species unique to the continental United States, Florida has been subject to long-term, extensive arachnological sampling.
Unpublished information regarding brown recluse spider bites as reported to all three poison centers in the Florida Poison Information Center Network (FPICN) was acquired from Dr. Jay Schauben, director of the FPICN Data Center. Interpretation of FPICN data was accomplished through correspondence and conversations with Dr. Schauben, by using the standard categories of coding as designated by the American Association of Poison Control Centers. In total, 844 Florida brown recluse spider bite reports from 1997 to 2002 were represented in this database.
The most critical data used in this study were based upon the 124 reports categorized as "in (enroute to) HCF" (i.e., health care facility) signifying brown recluse spider bite diagnoses made by medical personnel. Data are presented here as number of reports by county and age demographics of patients. Dr. Schauben could not estimate the proportion of the total Florida brown recluse spider bite diagnoses represented by the FPICN data.
The remaining 720 reports were segregated from data analysis because they were either "referred to HCF" (198 cases) or "managed on site (nonhealth care facility)" (522 cases). The former refers to persons contacting the FPICN for information regarding alleged brown recluse spider bites where symptoms warranted medical attention (without assessment of etiologic accuracy nor corroboration by medical personnel); the concerned persons were encouraged to seek medical attention. The latter refer to cases where persons informed the FPICN of alleged Loxosceles bites, which were not treated at medical facilities (but most often at home) and probably had no assessment by medical personnel. Although this information is decidedly less useful for this study, it does have significance in that patients presenting histories of alleged brown recluse spider bites could incorrectly influence physician assessment of lesions.
Corroborative Spider and Bite Misdiagnoses Information.
Corroborative information was generated from several ongoing Loxosceles spider research projects in the United States. R.S.V. is identifying arachnids considered to be Loxosceles spiders submitted by the general public (Florida data augmented by G.B.E.). This study started in California in 1992; it was expanded nationwide in 2000, with 1,300 arachnids submitted as of December 2003 and is scheduled to terminate in December 2004. The information from this study compares Florida submissions to those from four endemic Loxosceles states (Texas, Missouri, Kansas, and Oklahoma). Also ongoing, R.S.V. is recording medical misdiagnoses of brown recluse spider bites from contacts nationwide. Additional information was provided by L.F.J. who was a practicing physician in Florida for 21 yr.
Loxosceles spiders have been verified in only 10 of Florida's 67 counties dating from 1904 (Table 1). L. reclusa spiders have been verified in five counties, mostly in the northern portion of the state, predominantly interceptions of single specimens (two found in cars, one on a ship). L. rufescens was found in two counties, and one home in Polk County harbored L. laeta. Undetermined juvenile Loxosceles have been found in at least two other buildings. Therefore, as far as can be determined, Loxosceles spiders have been historically verified in approximately seven buildings in Florida, only two or so of which were homes.
The 124 FPICN brown recluse spider bite reports made by medical personnel occurred in 31 counties (with two of unknown county origin) of which only six counties have had verifications of Loxosceles spiders (Fig. 1). Counties where brown recluse spider bites were diagnosed are scattered geographically throughout the state but predominate in the peninsular area (Fig. 1). Demographics included 70 males (mean age 32.5 ± 17.5 yr, range 0–70) and 54 females (mean age 30.6 ± 19.8 yr, range 0–77) (Fig. 2). The 720 FPICN reports of Loxosceles bites made by nonmedical personnel originated from 52 of Florida's 67 counties (77.6%).
In the ongoing study of suspected Loxosceles spiders, of 74 spiders submitted from Florida (from 42 cities in 25 counties), none were Loxosceles. Most were wolf (Lycosidae); huntsman, Heteropoda venatoria (L.) (Sparassidae); or crevice spiders, Kukulcania hibernalis (Hentz) (Filistatidae). In comparison, from endemic Loxosceles states, Loxosceles spiders were frequently submitted: Texas, 64% (74 recluses of 116 spiders submitted); Missouri, 88% (42 of 48); Kansas, 83% (38 of 46); and Oklahoma, 77% (23 of 30). Of the 1,300 arachnids submitted nationwide, no homeowner from outside the native range of Loxosceles spiders has yet submitted an L. reclusa spider.
Only one Florida person diagnosed with a brown recluse spider bite has ever submitted a Loxosceles spider (and the veracity of spider involvement was questioned by the attending physicians; Edwards 2001). Of Florida brown recluse spider bite misdiagnoses referred to us by physicians, two cases were later reevaluated as pyoderma gangrenosum and one as melanoma.
Reports of brown recluse spider bites are far more common than are verifications of Loxosceles spiders in Florida. In a 6-yr period, the FPICN recorded 124 brown recluse spider bite assessments by medical personnel in nearly one-half of Florida's counties. From our collective experience of several decades with many contacts, consultations and referrals from diverse professions (medical, legal, news media, and pest control), we feel that this represents a small percentage of the actual number of Florida bite diagnoses made annually. During this same 6-yr period, only five discoveries of Loxosceles spiders were made in Florida. In 100 yr of arachnological evidence with tens of thousands of spiders collected and identified by arachnologists in Florida, only 11 finds of ≈70 Loxosceles spiders have been verified (more than one-half from one home) in 10 counties (Table 1). This paucity of Loxosceles spiders is in spite of 1) relocation of thousands of Americans from endemic Loxosceles regions, 2) relocation of thousands of immigrants from foreign countries with Loxosceles populations, 3) relocation by military personnel and equipment, and 4) continual interstate commerce. As an example of the low probability of Florida Loxosceles spider bite, during summer 2001, one Kansas home without envenomation of inhabitants produced more brown recluse spiders per week for each of 7 wk (Vetter and Barger 2002) than the total number of Loxosceles spiders ever documented in Florida. Despite the limited evidence of Loxosceles spiders, belief in their presence is widespread among the Florida general populace as alleged bites (without medical personnel assessment) have been reported from 52 counties. Physicians relying too heavily on patients' accounts may misdiagnose many necrotic lesions as spider bites. This also contributes to the overall public impression that there are large populations of brown recluse spiders in Florida.
Diagnoses of brown recluse spider bites in the absence of Loxosceles spiders are common throughout North America. A recent study regarding California, Oregon, Washington, and Colorado shows historical records of only 35 L. reclusa or L. rufescens spiders, whereas 216 brown recluse spider bite diagnoses were referred to the authors in 41 mo in these four states (Vetter et al. 2003). Although California harbors several native desert and one non-native Loxosceles species, 95% of the California bite diagnoses originated from coastal or northern cities where no Loxosceles specimens have ever been verified (Vetter et al. 2003). A Colorado medical article claims that brown recluse spiders cause an annual medical problem in the state (Mara and Myers 1977); however, a statewide study has identified >26,000 Colorado spiders with only two Loxosceles being recovered from one building (Vetter et al. 2003). In South Carolina, a survey of 940 physicians revealed 478 brown recluse spider bite diagnoses made in 1990 (Schuman and Caldwell 1991). Gaddy and Morse (1985) list brown recluse spiders from only three localities in South Carolina; however, they make the overspeculative conclusion that the spider "probably occurs statewide" based on their erroneous statement that brown recluses are "now found from coast to coast," which they are not (Gertsch and Ennik 1983; Vetter 2000, 2003). Wendell (2003) states that brown recluse spiders are uncommon in South Carolina, and recent information (R.S.V., unpublished data) verifies that they exist in the westernmost tip of the state, which is the easternmost border of the spider's native distribution in the southeastern United States. Nonetheless, the number of annual bite diagnoses for South Carolina is very large considering the scant evidence of the spider. Several bite diagnoses have even been made by physicians in such unlikely Loxosceles localities as Canada and Alaska (Vetter and Bush 2002a, Vetter et al. 2003, Bennett and Vetter 2004), yet no L. reclusa spiders (and only three specimens of L. laeta) have been found north of the southern Canadian border (Vetter and Bush 2002a).
Although uncommon in Florida, as well as elsewhere, necrotic skin lesions of obscure origin should not be considered rare. Determination of their etiology is usually difficult and often impossible. Infection with various bacteria and other microbes; chemical insult; allergic reactions; cutaneous manifestation of numerous unrecognized systemic conditions such as autoimmune phenomena; as well as vasculopathy due to diabetes, arteriosclerosis, hypertension, advanced age, and debilitation for any reason, can sometimes be implicated but are difficult to substantiate. Being subtropical, most of Florida presents an elevated risk of skin conditions seldom seen elsewhere and is home to numerous biting and stinging arthropods, which are generally considered capable of producing necrotic lesions. Florida's large geriatric population might be expected to contribute to an increased incidence of nonhealing or delayed-healing wounds; this is not, however, reflected in the data presented here (Fig. 2). Although it is possible that some patients are travelers or visitors from endemic Loxosceles areas, similar to other nonendemic states, overdiagnosis is a more plausible explanation for alleged Florida brown recluse spider bites than is transient human activity. However, considering that dozens to thousands of Loxosceles spiders have been collected in individual homes without envenomation in endemic Loxosceles areas, that homeowners from endemic states often submit Loxosceles spiders for verification, and that Florida does not support any widespread populations of Loxosceles spiders, brown recluse spider bites are not a likely cause of necrotic skin lesions in Florida.
The brown recluse spider has garnered far more attention than it deserves outside its endemic area in North America. Unless it can be proven that widespread populations of Loxosceles spiders are firmly established, Florida medical personnel should focus on the varied spectrum of etiologies of dermonecrosis (Vetter et al. 2003, Isbister and Whyte 2004) rather than blame a spider that is virtually nonexistent in the state. Medical personnel who continue to diagnose such bites based solely on clinical examination of lesions will proliferate erroneous information among the medical community and general public. Too often this diagnosis causes both patient and physician to have "brown recluse tunnel vision," preventing them from considering other more tenable etiologies, thereby delaying proper diagnosis and treatment. This misconception is an obstacle to better health care and can increase the use of inappropriate and potentially dangerous therapies (Bryant and Pittman 2003, Vetter and Bush 2004), which may be ineffective against the actual medical condition. This can cause superfluous medical litigation (Kunkel 1985, Vetter 2003), induce unwarranted arachnophobia, and cause patients to erroneously and recklessly eradicate spiders from households where exposure to pesticide may pose a much greater risk to occupants than spiders ever would.
Finally, for several decades in North America and in other parts of the world, the medical community has interchangeably used the terms brown recluse spider bite, loxoscelism, and necrotic arachnidism to describe a wide spectrum of necrotic dermal lesions of unknown, uncertain, or indistinct cause. This has lead to many instances of erroneous attribution of necrotic lesions to spiders (Kemp 1990, Moaven et al. 1999, Roche et al. 2001, Osterhoudt et al. 2002, Vetter and Bush 2002b, Vetter et al. 2003, Isbister and Whyte 2004) due in part to arachnophobia as well as the need of patients and physicians to attribute an easily recognizable, external cause to idiopathic ulcers (Isbister 2004). Physicians should diagnose idiopathic necrotic ulcer when the etiology is uncertain. This is an accurate description of the clinical presentation of such a lesion, as opposed to presumptively incriminating spiders as an unprovable and often inaccurate etiology for a necrotic dermal lesion (White 2003).
We thank Jay Schauben of the Florida Poison Information Center Network who provided the brown recluse spider envenomation report database and interpretation of the data categories, without which this manuscript would not have been possible. We also thank two anonymous reviewers for comments that improved the manuscript.