Bacterial Etiology of Necrotic Arachnidism in Black Widow Spider Bites

Spider bites meriting medical intervention on the west coast are surprisingly common. In 2008, there were over 12,500 reported bites on humans, with over 2,500 from widow spiders [1]. (The actual number of bites potentially being far greater because of lack of reporting) [2]. Spider bites are very difficult to treat for physicians, veterinarians, and patients. Diagnosis is also a challenge because the spider culprit is rarely found or caught, and the bite produces a number of nonspecific signs with a long list of differentials.

mm long which is sufficiently long to inject the venom to a dangerous depth. The males, being much smaller, can only inject a minute amount to a lesser depth, thus making them medically insignificant. The actual amount of venom injected per bite, even by a mature female, is very small in actual volume (.02-.03 mg) [7,8]. However, with an extremely high LD50 of 0.002 mg/kg, [7][8][9] even this tiny amount can be fatal.
The venom of Latrodectus contains a multitude of harmful proteinaceous and non-proteinaceous constituents which affect neuromuscular endplates and central nervous system synapses. It even contains 5-hydroxytryptamine (serotonin) [10,11]presumably for the sole purpose of inducing pain [12]. One of the especially harmful proteins in the venom is α-latrotoxin which binds to presynaptic cholinergic receptors of vertebrates and induces massive neurotransmitter exocytosis from central and peripheral nerve terminals [13]. Initially the venom causes small end-plate potentials, but soon results in a complete block in neurotransmission due to depletion of neurotransmitters from terminal synaptic vesicles. This complex action seems to result from glycoproteins and/or gangliosides on the neuromuscular synaptic membranes, thus allowing a channel to form for monovalent cation exchange that locks open. This depolarization promotes calcium-independent release of neurotransmitters down concentration gradients, then inhibits their subsequent reuptake.
The systemic effects and symptoms of black widow envenomation include: severe pain, nausea, tachycardia, myocarditis, hypertension, pyrexia, dyspnea, vomiting, muscle weakness, rigidity, spasm, cramping and tetany, hyperreflexia, paresthesia, salivation, lacrimation, diaphoresis, seizures, anaphylaxis, shock, cardiac arrest, coma and death [9,14,15]. The venom also frequently causes priapism [16], and has a degree of affinity for the facial musculature which can cause extreme facial contortions; this is a condition known as facies latrodectismi [17]. Interestingly, the venom has also been found to have immunosuppressive effects which could potentiate complications such as infection [9,18].
At or near the site of envenomation, lesions can range from rash, erythema, and edema to severe inflammation, infection, and necrosis. Blanching and cyanosis slowly develop at the central core with spreading erythema and progression to dermonecrosis [9].
Necrotic arachnidism is the term for the cutaneous reaction from the spider bite and it is associated with vasodilation, ischemia, and thrombosis with occasional progressive necrotic lesions. While some of the local skin reactions which are manifested as a result of envenomation like edema and erythema are caused directly by constituents of the venom due to (among other things) the lytic action of various enzymes (such as hyaluronidase), it is currently assumed that the dermonecrosis is most likely due to an immune response to the venom, if not enzymes in the venom itself [18].
Our recent findings, however, suggest that the current understanding of necrotic arachnidism may be inaccurate and that several species of pathogenic, antibiotic resistant bacteria on the fangs of the spiders could potentially be inoculated during the spider bite and cause primary cutaneous lesions.
The objectives of this descriptive study were: (1) to determine if there are bacterial species living on the fangs of black widow spiders, (2) if these bacteria are medically significant and associated with disease, and (3) what antibiotics these bacteria are sensitive and resistant to.  10:00 PM and 2:00 AM. Large specimens (above 100mg) with no overt signs of disease were collected into sterile specimen containers then taken back to the lab and examined for missing limbs, external parasites, fungal infections, malnourishment, and dehydration. Any spiders not meeting these criteria were returned to the locations where they were found.

Materials
The spiders were kept in our lab at Western University of Health Sciences College of Veterinary Medicine in individual, autoclaved sterile, 30cm 3 screw top plastic containers. They were provided a 12 hour light/12 hour dark photoperiod, maintained at 24.0+2.0 o C 24h/ day, and provided water through holes in the lid of each container.
In preparation for the fang extraction procedure, we autoclaved surgical packs containing 8 towels, and 4 sets of: bridge forceps, Jacobson micro artery forceps, and Joseph rhinoplasty micro scissors.
The spiders were euthanized using an open drop technique with 100µl isoflurane in a 35ml container resulting in a 7% anesthetic concentration. The spiders were then left in their chambers for 30 minutes to ensure complete and humane euthanasia. Previous studies have shown that isoflurane does not affect bacterial growth [20].

Surgical procedure
Standard surgical aseptic technique was followed for the surgical extraction procedure which included proper scrubbing and gowning with autoclaved surgical robes, caps, masks, and sterile gloves.
To ensure a sterile surgical field, a sterile towel was placed over the microscope base. The sterile surgical pack was then placed over the towel and properly opened revealing the instruments and 8 folded sterile towels. The spiders were placed on the towel under the microscope field in dorsal recumbency. The towels were folded in a manner which provided us with 4 rectangular fields separated by creases. After the fangs were excised, each subsequent spider was placed in a new field to avoid contamination by the previous spider's body. Using the dissecting microscope (Tritech Stereo-Microscope System w/optics and Tritech fiber optic external illuminator) the spiders were carefully manipulated into position by the left fang using the sterile forceps and the fangs were removed using one cut with the microscissors at the level of the base of the chelicerae. This technique prevented any expulsion of endolymph which could contaminate the samples, as well as kept the instruments clean having only touched the fang. When accidental expulsion of endolymph or accidental contamination by the rest of the body occurred, forceps and scissors were exchanged for new sterile ones, and the fangs were discarded. Additionally, instruments were switched out every 10 spiders. After excising the fangs, they were dropped directly into sterile culture media tubes. Once 10 pairs of fangs were dropped into each tube, they were sealed and submitted to Antech Diagnostics (Irvine, CA) for aerobic and anaerobic bacterial cultures and sensitivity testing.

Results
Of the 22 culturettes (each containing 10 pairs of fangs) submitted to Antech Diagnostics over the 3 month period, we received 2 positives for Morganella morganii, 2 positive for Capnocytophaga, 1 positive for Citrobacter amalonaticus, 1 positive for Pantoea agglomerans, 1 positive for Pseudomonas spp. , 7 Positive for Enterococcus spp., and 3 positive for Bacillus spp. (non anthracis). This gives us a total of 17 positives out of 22 samples for a prevalence of 77%.
Capnocytophaga is anaerobic, so sensitivity testing was not performed.
All of the Enterococcus sensitivity testing showed susceptibility to all tested antibiotics (ampicillin, amoxicillin, azithromycin, chloramphenicol, clarithromycin, doxycycline, enrofloxacin, erythromycin, gentamycin (high conc.), marbofloxacin, and streptomycin (High conc.). However, Antech made a point to remark that cephalosporins, and potentiated sulfonamides, are not tested because they are always ineffective against enterococci, and aminoglycosides used alone are also ineffective.
One strain of isolated Enterococcus demonstrated intermediate resistance to chloramphenicol, clarithromycin, erythromycin, and marbofloxacin. Figure 2 depicts a quantitative assessment of the efficacy of antibiotics against the isolated bacteria. As evidenced by the figure, enrofloxacin, marbofloxacin, amikacin, difloxacin, and gentamycin have shown to be overall the most effective antibiotics against these bacteria.
Morganella morganii is one of the most common isolates from snakebite infections, and has also been found in several forms of serious gangrenous skin infections including ecthyma gangrenosum. The necrotizing lesions secondary to bites are often attributes to Morganella [22][23][24][25][26].
Capnocytophaga is another highly pathogenic bacteria which is most frequently found in dog bite wounds. Even minor bites are known to cause serious disease, and infections from this bacteria can lead to several life threatening and fatal syndromes. Capnocytophaga is a common cause of septicemia, gangrene, purpura fulminans, and other necrotizing skin and eye diseases. It has also been shown to cause hemolytic uremic syndrome, and multisystem organ failure [27][28][29][30][31][32][33][34].
Our final pathogenic bacterial isolate, Pseudomonas, is well known to the medical community for causing a wide range of serious conditions from urinary tract infections to meningitis. These include but are not limited to: septic shock, necrotizing skin lesions, and several types of fatal organ infections. Pseudomonas has also been cultured from infected bite wounds from other animal species [39][40][41]. In the medical community, spiders are convenient scapegoats for a variety of symptoms; namely the idiopathic, inflammatory, skin lesion. Over half of all documented cases of necrotic arachnidism in the medical literature are concluded without any substantiating evidence and present from areas where the suspected spider isn't even found [1].
A great number of physicians and veterinarians alike are frustrated at the frequency that patients present with a skin lesion of unknown origin that they or their family physician hastily attributes to "a spider bite." The general consensus within the doctors and researchers that deal with the epidemiology of spider bites is that family physicians and the average public vastly over-diagnose spider bites for just about any idiopathic lesion of the skin [4]. In most cases, when the patient sees a specialist, it is determined that they have a skin or soft tissue infection [2,18].
However, our findings suggest that the primary etiology of spider bite-associated dermonecroses may in fact be bacterial, and these unsubstantiated conclusions are potentially correct. This means many of these "misdiagnosed" spider bite lesions, which turn out to be skin or soft tissue infections, may in fact be accurately diagnosed spider envenomations.
Our results also offer a potential explanation for the traditional unresponsiveness of spider bite lesions to antibiotics since some of the most commonly prescribed antibiotics for skin infections are cephalosporins and β-Lactams [21]-which our findings show have little to no effect on most of these bacteria. Amazingly, many current treatment protocols do not even include antibiotics because their efficacy against spider bites was previously unsubstantiated [15,42,43].

Conclusion
Given our results we opine that fluoroquinolones and aminoglycosides appear to be the most efficacious antibiotics against the spectrum of bacteria found on Black Widow fangs. When possible, these antibiotics should be considered in the treatment protocol for spider envenomations, (especially those from black widows) to potentially help decrease the risk of necrotic arachnidism and secondary infections. Graphical representation of antibiotic efficacy against Black Widow fang bacteria. The 8 isolated bacteria were ranked by pathogenicity and given an arbitrary numerical value starting with 2.0 (for the most pathogenic) and decreasing in increments of 0.2 for each subsequent species. An efficacy score was calculated for each antibiotic by summing the pathogenicity scores for all bacteria the antibiotic was efficacious against, then subtracting the score for any bacteria resistant to that antibiotic and subtracting half the score for any bacteria that showed intermediate resistance. Finally, a confidence interval was calculated by adding the pathogenicity scores for all the bacteria not tested against that particular antibiotic.