Background

Each year, envenomation by pit viper snakes (Family Viperidae, subfamily Crotalinae, genera Crotalus, Agkistrodon, and Sistrurus) causes at least 2,700 people to seek hospital treatment in the United States. About half of these patients receive antivenom[1]. In October 2000, the United States Food and Drug Administration (US FDA) approved a Fab antivenom product for crotaline snakebite. Compared with equine-derived whole-IgG antivenom, Crotalidae Polyvalent Immune Fab (Ovine) (CroFab™, Protherics, Nashville, TN; hereafter, FabAV) is thought to convey a reduced risk of acute and delayed-type hypersensitivity reactions[2].

The US FDA approved FabAV based on two clinical trials, both of which excluded patients with severe envenomation.[3–5] The reason for this exclusion was equipoise: at the time the trials were conducted (1993–96), treating life-threatening venom effects with investigational antivenom in lieu of a proven standard therapy was considered unethical. As a result of the trial design, the US FDA approved FabAV, "for the management of patients with minimal or moderate North American crotalid envenomation"[3].

Wyeth Pharmaceuticals announced in 2001 that it would cease production of equine antivenom[6]. It appears that the last lot of equine antivenom expired in April, 2007, and no other antivenom has been approved for treating crotaline snakebite[7]. Therefore, at the present time, there is no approved antivenom therapy for severe crotaline snakebite available in the United States.

Available data suggest that FabAV is being widely used to treat severe envenomations. The 2006 report of the American Association of Poison Control Centers' National Poison Data System (NPDS) lists 2,768 victims of crotaline snake bite treated in a health care facility, of whom five died, 1,528 had moderate to severe toxicity, and 1,239 had minimal or no clinical effects[1]. FabAV was administered to 1,359 patients during this period. In addition, several cohort studies and many case reports have described the use of FabAV to treat snake bite victims of all severities.

We analyzed the English language medical literature to characterize the reported response to FabAV therapy of patients suffering severe crotaline envenomation.

Methods

We searched PubMed, Ovid MEDLINE, and EMBASE to identify all published articles containing primary data about North American crotalid envenomations treated with FabAV. All searches were performed on July 28, 2008, using the search strategy listed in Table 1. All article types were considered, including prospective clinical trials, cohort and non-cohort case series, single case reports, review articles, editorials, commentaries, published abstracts, and letters-to-the-editor. Citation lists from the three databases were imported into a ProCite database (ProCite® version 5.0 for Windows, ISI ResearchSoft, Philadelphia, PA, USA), and duplicate articles were removed. We then performed an internal search to deselect articles outside the scope of the search. The terms searched were rat(s), mouse, mice, rabbit(s), cellular, in vivo, and in vitro.

Table 1 Search terms
Full size table

A single medical toxicologist (EJL) then reviewed each title and abstract. Articles judged to have no likelihood of containing data about FabAV administration for North American crotalid envenomations (e.g. reports that clearly referred only to the treatment of African, Asian, or Australian snakebite) were excluded from further review.

Two supplemental sources complimented the computerized literature search. Our article list was cross-referenced with the topic-specific article files of Protherics, Inc, the manufacturer of FabAV. The authors also searched the bibliographies of all included articles, major textbooks in emergency medicine and medical toxicology, and our institutional article files for additional primary source materials. These supplemental searches yielded some publications prior to 1999.

Two board-certified medical toxicologists with experience in the management of snakebite (EJL, THS) independently reviewed the full-text version of each article, identified all cases meeting the a priori definition of severe envenomation (see below), and extracted case-level data to a standardized form. This form consisted of two pages for data concerning hypersensitivity reactions (collected for a separate manuscript), one page for summary data about any severe envenomation cases, and one additional page for each severely envenomated patient reported. Training on the abstraction step consisted of one hour of instruction, followed by focused review of eleven practice abstractions. The reviewers were not blinded to the intent of the study. Data were entered into a Microsoft Access database (Microsoft Access 2003®, Microsoft Corporation, Redmond, WA, 2003) and double verified to ensure accuracy of data entry. Disagreement between the abstractors was identified and resolved by consensus.

Included cases were reviewed to identify duplicate publication; where necessary, study authors were contacted to determine if cases reported in different publications were one and the same.

Data were analyzed using SAS (v. 9.1, SAS Institute, Cary, NC). Descriptive statistics were used; confidence intervals were not presented due to small sample sizes.

The definition of "severe" envenomation was that used in the US FDA-approved prescribing information for FabAV (Table 2). This standard definition contains several areas of ambiguity, which were resolved a priori as follows:

Table 2 Definitions of Envenomation Severity from the FabAV Prescribing Information[3]
Full size table

The standard definition of local tissue effects classifies, "swelling, pain, or ecchymosis involving less than a full extremity," as "moderate," and, "swelling, pain or ecchymosis involving more than an entire extremity," as "severe." We included cases involving swelling, pain, or ecchymosis involving exactly an entire extremity in the "severe" group.

The standard definition of neurotoxicity classifies, "oral paresthesias or unusual tastes," as "moderate," but has no criteria for severe neurotoxicity. We considered severe muscle weakness, difficulty speaking or swallowing, and fasciculations remote from the bite site (sometimes a sign of impending paralysis) to be signs of severe neurotoxicity.

The phrase, "coagulation parameters are abnormal, with serious bleeding or severe threat of bleeding," in the standard definition required more precise definition. Using published criteria, we considered "severe threat of bleeding" to be present if the platelet count was less than 50,000 cells/mm3, if the fibrinogen was less than 50 mg/dl (1.5 micromol/L), or if the international normalized ratio (INR) or protime ratio were > 5.0[8] If protime was reported without INR or normal range data, a protime > 50 seconds was considered to represent severe threat of bleeding.

The standard definition does not specify a time point at which severity is assessed. In order to mirror clinical practice, we graded the severity of envenomation based on the initial presentation, i.e. the patient's clinical condition during the first six hours after presentation for care. Cases that were of minimal or moderate severity on initial presentation, but that developed one or more features of severe envenomation many hours or days later, were counted as "minimal" or "moderate" based on the severity of the initial presentation.

In some cases, authors reported clinical manifestations such as, "severe tachycardia," or, "hypotension," without reporting numeric values. We counted these cases as, "severe," if the author of the report described them as, "severe."

The Snakebite Severity Score (SSS) is a validated tool for assessing crotaline snake envenomation[9]. A SSS ≥ 7 is generally considered to represent severe envenomation. Although clinicians' global assessment of severity was used to determine eligibility in the FabAV premarketing trials, serial calculations of SSS were made in all patients as the primary efficacy outcome of the premarketing trials and at least one other report[4, 5, 10] Although the SSS is a composite measure of severity, rather than a unique venom effect, serial calculations of SSS were the only case-level data presented about response to treatment in some reports[4, 5, 10]. Therefore, when data were available, we calculated the SSS for each case reported and included those with a SSS ≥ 7 in the "severe" group.

In this report, we defined "initial control" of a specific venom effect, a priori, as cessation of progression of local tissue effects (pain and swelling), and as complete resolution of systemic effects (hypotension, neurotoxicity, and medically important bleeding). In accordance with previous research, initial control of coagulopathy was defined as definite improvement in coagulopathy or platelet count tests, as appropriate, combined with the absence of systemic bleeding[5, 8, 11–13]. "Initial control of the envenomation syndrome" was defined a priori as simultaneous initial control of all specific venom effects experienced by that particular patient.

After initial control of the envenomation syndrome is achieved, the manufacturer of FabAV recommends administration of 2-vial doses of FabAV, given at 6, 12, and 18 hours ("maintenance therapy"), in order to prevent recurrent or delayed onset venom effects[3, 5]. We scored a patient as having received maintenance therapy if the patient received at least three doses of FabAV, each consisting of two or more vials, within a 24-hour period after initial control was reached.

Annual reports of the American Association of Poison Control Centers' NPDS, and its predecessor, the Toxic Exposure Surveillance System (TESS), contain robust statistical data about poisonings in general, and brief (< 100 word) case abstracts about fatal poison exposures. Because they do not focus on snakebites, NPDS/TESS reports were expected neither to be identified by the search strategy, nor to contain sufficient detail to be included in the main analysis. Nonetheless, they remain a relevant, important source of peer-reviewed data. Therefore, we hand-searched the 2000–2006 annual reports for any reported fatalities that occurred after FabAV administration[1, 14–19].

Results

Article and subject identification

The search strategy identified 147 unique publications regarding FabAV. The article selection process is depicted in Figure 1. A total of 22 articles contained at least one case of severe envenomation by a North American crotaline snake treated with FabAV. These articles describe 30 apparently-unique cases of severe envenomation. From this group, five cases did not contain sufficient data about the clinical course after FabAV administration to judge whether the manifestations of severe envenomation responded to therapy, and one case was determined to be included in two different series; these cases were therefore excluded[10, 20–24]. The remaining 24 cases from 19 published reports are presented in Table 3[4, 10, 13, 20, 22, 25–38].

Figure 1
figure 1

Article identification and selection process. This information is also presented as an attached file.

Full size image
Table 3 Published cases of severe envenomation treated with FabAV
Full size table

Five cohort studies and fourteen non-cohort studies were identified. Two of the cohort studies collected data prospectively, two collected data retrospectively, and one used both prospective and retrospective data collection. All of the non-cohort studies were of retrospective design. Seven severely envenomated patients were reported in the cohort studies, and 17 severely envenomated patients were described in the non-cohort studies.

Initial response to FabAV therapy

Sixty-five specific severe venom effects were reported in these 24 patients. The initial response to FabAV treatment for these specific severe venom effects was: improved/resolved, 50 effects (77%); no improvement, 11 effects (17%); not reported, 4 effects (6%). All of the 22 specific venom effects (100%) experienced by the seven patients in the cohort studies improved after FabAV administration. In contrast, of the 43 specific venom effects experienced by the patients in the non-cohort studies, 28 effects improved or resolved (65%), 11 effects did not improve (26%), and the response of 4 effects were not reported (9%).

Initial control of the envenomation syndrome was achieved in 12 patients (50%), not achieved in 9 patients (38%), and not fully reported in 3 patients (13%). All seven (100%) of the severely envenomated patients in the cohort studies achieved initial control. Once again, the response to initial therapy was not as good in the patients reported in non-cohort studies. Among these 17 patients, initial control of the envenomation syndrome was achieved in 5 patients (29%), not achieved in 9 patients (53%), and incompletely documented in 3 patients (18%). The median dose of FabAV used to achieve initial control of the envenomation syndrome in these 12 patients was 6 vials (range: 4 – 18 vials).

Persistent severe venom effects

One or more persistent severe venom effects were reported in 0/7 patients reported in cohort studies (0%), and in 9/17 patients in non-cohort reports (53%). These cases are summarized in Table 4[27–30, 34, 36, 38]. These effects consisted of limb swelling, limb pain, soft tissue bleeding, thrombocytopenia, neurotoxicity, or compartment syndrome. Response to therapy was not reported for four patients, summarized in Table 5[26, 27, 32, 35].

Table 4 Specific severe venom effects that persisted after FabAV administration
Full size table
Table 5 Specific severe venom for which response to therapy was not documented
Full size table

Recurrence or delayed onset of severe venom effects

Recurrence or delayed-onset of at least one severe venom effect was reported in 12 patients, including 3/7 patients (43%) in the cohort studies, described in Table 6[4, 10, 13, 27, 28, 31, 32, 34–38]. Maintenance therapy was administered to two patients (17%), not administered to 9 patients (75%), and not documented for one patient (8%). With one exception (recurrent limb pain and swelling), all cases of recurrence or delayed onset of severe venom effects involved defibrination (with or without prothrombin time elevation) and/or thrombocytopenia, and were clinically occult. Although these events were judged a priori to represent "a severe threat of bleeding," none of the 11 patients (0%) with recurrent or delayed-onset hematologic venom effects developed bleeding.

Table 6 Recurrence or delayed onset of severe venom effects
Full size table

Permanent sequelae of envenomation

Few publications assessed and reported long-term outcomes. Therefore, the available data are inadequate to describe the long term outcomes after crotaline snakebite treated with FabAV. No published manuscripts described death following FabAV administration were identified in the literature search.

Reports to the US National Poison Data System

The TESS/NPDS data include 21 deaths due to snakebite reported to participating US poison control centers from 2000 – 2006[1, 14–19]. Of these, five patients received FabAV prior to death; two additional patients received unspecified antivenom. These cases are summarized in Table 7. Five patients presented in extremis and died of cerebral anoxia and/or multisystem organ failure; the other two patients died from complications of substance abuse.

Table 7 Reports of death after FabAV administration reported to the US National Poison Data System, 2000–2006
Full size table

Discussion

Physicians in the United States treating victims bitten by rattlesnakes, cottonmouth and copperhead snakes, and pygmy rattlesnakes no longer have access to an antivenom that is licensed and approved to treat severely envenomated victims. The previous standard therapy, whole IgG antivenom, is no longer available; the currently-available antivenom, FabAV, was tested and approved only for use in mildly and moderately envenomated patients. Those patients with severe envenomation – hypotension, severe hematologic effects, and/or severe limb findings – are clinical "orphans." Data from the American Association of Poison Control Centers suggest that, when faced with the choice of off-label administration of FabAV or supportive care only, treating physicians most often choose to administer FabAV to severely envenomated patients[39]. It is difficult to conceive of a placebo-controlled trial of FabAV in severe snakebite; to our knowledge, no such study has been conducted.

A significant body of published literature describes the overall clinical experience with FabAV; this study identified 147 relevant articles, of which 19 articles provided case-level data about severely envenomated patients treated with FabAV.

Out of approximately ten published cohort studies of FabAV-treated patients, we were able to identify seven patients from five reports who met our a priori definition of severe envenomation. All seven of these patients demonstrated good initial response to FabAV therapy.

Severe snakebite is sometimes associated with the need for intubation, either due to airway edema or as part of supportive care of a patient in shock. Our review found three cases of severely envenomated patients who required intubation. One case involved a patient who was intubated for venom-induced periglottic edema, received FabAV, and was successfully extubated the next day[34]. A second case involved a patient who developed multisystem organ failure after deliberate intravenous injection of rattlesnake venom in a suicide attempt. The indication for his intubation, which occurred prior to FabAV therapy, was profound shock and gastrointestinal hemorrhage. These problems responded quickly to FabAV therapy; the patient successfully self-extubated on the third day of hospitalization[33]. We judged both these cases to demonstrate successful treatment of the venom effect, "need for intubation," with FabAV. The third case involved a patient who was intubated due to progressive neurotoxicity due to envenomation by an unknown rattlesnake. Administration of FabAV failed to prevent the need for intubation, and significant neurotoxicity progressed even after aggressive FabAV therapy[36].

Recurrence and delayed onset of severe venom effects are a known complication of snakebite, whether treated with FabAV or whole-IgG antivenom[5, 40, 41]. Cases involving both severe and initially-minor envenomation have been previously reported[41]. Three of the seven patients reported in the cohort studies developed recurrence phenomena. Two of these cases involved recurrent defibrination syndrome without bleeding; neither patient received maintenance FabAV therapy. Maintenance therapy has been shown in a randomized controlled trial to prevent early recurrence of local tissue venom effects. Use of maintenance therapy to prevent recurrent coagulopathy is based on strong pharmacokinetic arguments[5, 12, 37, 42]. The third patient developed recurrent limb pain and swelling despite maintenance therapy.

Unfavourable outcomes, including severe venom effects that were refractory to the FabAV doses given, delayed-onset severe venom effects, and recurrence phenomena, were all reported more commonly in case reports and other non-cohort studies than in cohort studies. The difference was statistically significant (P = 0.005 for initial control, Fisher's Exact test). This is as expected; the presentation of novel results ("originality factor") increases the likelihood that a report will be accepted for presentation at a scientific meeting, and that it will subsequently be published as a full-length manuscript[43].

It is clear from these reports that, in some patients, defibrination and/or thrombocytopenia do not respond to large doses of FabAV. However, based on the observation that no reports describe medically significant bleeding that began after FabAV administration, the risk of bleeding in this situation is probably low. Subsequent to this structured literature review, a case report of fatal cerebral hemorrhage associated with recurrent defibrination in a FabAV-treated patient has been reported[44]. The authors of this report could find no other cases of significant spontaneous bleeding in 34 published cases of recurrent coagulopathy. A large cohort study of FabAV-treated patients, followed to resolution with detailed biochemical characterization including venom antigenemia, would be valuable to address this important question. Until that time, the recommendations made by Yip continue to represent the best available guidance in this area, with the caveat that, beyond a certain point, administration of additional FabAV to patients with refractory hemostatic dysfunction is unlikely to be beneficial[8].

The recommended dosing of FabAV is 4- to 6-vial aliquots, repeated as needed until the desired clinical effect is achieved. In the premarketing studies of mild and moderately envenomated patients, the median dose of FabAV used to achieve initial control of the envenomation syndrome was 6 vials (range: 3 to 12 vials)[4, 5]. In this review of severely envenomated patients, initial control of severe venom effects was achieved after a median dose of 6 vials (range: 4 to 18 vials) was administered. Some patients received extraordinary doses in response to persistent or recurrent severe venom effects; whether patients benefited from doses in excess of 18 vials, excluding maintenance therapy, is unclear.

The NPDS contains information about 15,917 crotaline snake envenomations treated in a health care facility from 2000 to 2006; 21 of these patients (0.13%) died. Five fatality reports describe death that occurred after FabAV administration; another two patients received unspecified antivenom prior to death. No deaths appear to be caused by an adverse reaction to FabAV. Although it is difficult to make broad conclusions based on the sparse descriptions in these reports, the lack of any clear cases of treatment-failure associated death is reassuring. The fact that the fatal case reported by Kitchens and Eskin did not appear in the NPDS reports underscores one weakness of US poison-center based data, which rely on voluntary reporting[44].

In addition to issues surrounding retrospective data collection and publication bias, this report is limited by the lack of a comparison group. Published reports of severely envenomated North American crotalid victims managed without antivenom are extremely rare; therefore, it is difficult to differentiate the response to FabAV therapy from the natural history of untreated snakebite.

Because of the number of patients reported in cohort series is small, we cannot estimate the true rate of treatment failures, death, and other uncommon but important adverse events in the larger population of severely envenomated patients treated with FabAV.

No study reported in this series examined long-term outcomes.

One prior study, reported only in abstract form, has evaluated the use of FabAV in treatment of severe crotalid envenomation[45]. In that retrospective review of poison center cases, 9.3% of crotalid envenomations were judged to be severe. Initial control was achieved in 57% of severe cases, using a mean dose of 10.5 vials of FabAV.

As with any review, the conclusions of this report are limited by the available literature. To our knowledge, the largest cohort study of Crotaline snake bite victims treated with FabAV reported 93 cases[46]. A much larger, multi-center cohort study would be extremely useful to better define the answers to unresolved management issues.

Conclusion

In this integrative review of the published literature, treatment of severely-envenomated crotalid snake victims with FabAV was generally associated with good short-term outcomes. Persistent, recurrent, or delayed-onset venom effects, particularly thrombocytopenia and defibrination, were observed in several patients, but no patient developed bleeding complications after receiving FabAV. FabAV therapy was well-tolerated. FabAV therapy appears to be appropriate in the management of severely envenomated crotalid snake victims.