Recurrent Coagulopathy After Antivenom Treatment of Crotalid Snakebite

Gregory M. Bogdan, PhD, Rocky Mountain Poison and Drug Center; Richard C. Dart, MD, PhD, Denver Health Authority, Denver, Colo; Steve C. Falbo, the University of Colorado Health Sciences Center, Denver, Colo; Jude McNally, RPh, and Daniel Spaite, MD, Arizona Poison and Drug Information Center, Tucson, Ariz

South Med J. 2000;93(6) 

Abstract and Introduction

Background: We studied whether recurrence of coagulopathy, defined as the return of a coagulation abnormality after initial normalization, occurred after the use of antivenin (Crotalidae) polyvalent.
Methods: A retrospective, blinded, descriptive analysis of 354 consecutive cases of North American crotalid snake envenomation was done. Inclusion criteria were documented clinical evidence of crotalid snakebite, presence of a coagulopathy (platelet count <150,000/mm3, prothrombin time above normal, or fibrinogen level <150 mg/dL), and treatment with antivenin (Crotalidae) polyvalent.
Results: Of 112 cases with a coagulopathy extending beyond 6 hours after envenomation, 31 had sufficient coagulopathy testing to detect recurrence. Fourteen of these patients (45%) had recurrence of coagulopathy, and two cases were severe (fibrinogen level 29 mg/dL; platelet count 36,000/mm3).
Conclusion: Recurrence of coagulopathy after envenomation by North American crotalid snakes may occur after use of antivenin (Crotalidae) polyvalent and can result in severe coagulation abnormalities.

Poisonous snakebites in the United States predominantly involve the crotalid snakes -- rattlesnakes, cottonmouths, and copperheads. Crotalid snakes have complex venoms that are composed of numerous components ranging from small metal complexes and peptides to complex proteins with molecular weights in excess of 100,000 daltons.[1] Nicole and Raphael aptly described snake venom as a "mosaic of antigens."[2]

Crotalid snake venom components are capable of injuring nearly all organs. The most common effect is local tissue destruction, but severe coagulation abnormalities are common, and other systemic effects may occur (ie, hypotension, pulmonary edema, cardiovascular collapse, renal failure).[3-5] The coagulation defect produced by crotalid snake venom may involve platelets, fibrinogen, and thrombin or any combination of these elements. The course of coagulation abnormality may persist more than a week.[6]

After the introduction of antivenin (Crotalidae) polyvalent (Wyeth-Ayerst, Philadelphia, Pa) in 1954, specific treatment was available for crotalid envenomation. It remains the only antivenom commercially available for the treatment of crotalid snakebite in the United States. Antivenin (Crotalidae) polyvalent consists of partially purified equine serum from horses immunized with crotalid snake venoms. The mechanism of action is thought to be direct binding of venom components by venom-specific IgG contained in the antivenom. Treatment of crotalid snakebite in the United States typically includes the use of 5 to 20 vials or more of antivenom. Unfortunately, use of antivenin (Crotalidae) polyvalent has been plagued by frequent allergic reactions, leading to the development of a new investigational antivenom produced in sheep. This new antivenom uses manufacturing procedures to minimize antigenic protein content, thereby decreasing allergic reactions associated with its use.[7]

However, a prospective, randomized trial of the new antivenom revealed that recurrence of coagulopathy was observed in several patients. (Recurrence is defined as the return of an abnormal coagulation test result after at least one normal test result was obtained.)[8] The discovery of this phenomenon of coagulopathy recurrence raised concerns that it may also occur after the use of antivenin (Crotalidae) polyvalent. The purpose of our study was to use a previously established data base of crotalid snakebite cases to determine whether the phenomenon of coagulopathy recurrence has occurred after the use of antivenin (Crotalidae) polyvalent.

Methods

The records reviewed for our study included 354 cases of North American crotalid snake envenomation reported to the Arizona Poison and Drug Information Center (Tucson, Ariz) from July 1985 to September 1992. All calls coded as snakebites by the specialist in poison information were enrolled in the data base. The standard procedure was to monitor all cases closely until the patient's discharge from the health care facility. Case information used in our analysis consisted of a completed standard Toxic Exposure Surveillance System form, the patient's hospital records, or both. Since our analysis of the course of coagulation test results was conceived after the data base was created, the specialists were unaware of the study question at the time the case was managed and entered into the data base.

To form the data base, each record was abstracted by a single research assistant using a standardized data collection form for demographic data (age, sex, bite site, and geographic area where bite occurred), coagulation test results (platelet count, prothrombin time [PT], and serum fibrinogen), and information concerning antivenin (Crotalidae) polyvalent use (time and amount administered). The research assistant was unaware of the study question.

A subset of the data base was formed of all cases in which a coagulation abnormality was documented at any time. Criteria for inclusion of a case were (1) documented historical and clinical evidence of crotalid snakebite, (2) presence of a coagulopathy (platelet count <150,000/mm3, PT above normal for the reporting lab -- generally 10.2 to 13.8 seconds or within 2 seconds of control -- or fibrinogen <150 mg/dL), and (3) use of antivenin (Crotalidae) polyvalent. Since our purpose was to document the course of coagulopathy, cases were excluded if only one result for any of these tests was reported or if the test results did not span a period of at least 6 hours from the time of envenomation. Finally, since coagulopathy recurrence depends on the presence of a coagulation abnormality that returns to normal before becoming abnormal again, exclusion/inclusion criteria were then reviewed for each case to determine whether at least one additional test was done after the patient's coagulation status normalized.

Data were entered using Excel 7.0 software (Microsoft Corp, Redmond, Wash) and imported into InStat 3.0 (GraphPad Software Inc, San Diego, Calif) for statistical analysis. Categorical variables were presented as frequency of occurrence and were compared by Fisher's Exact Test. Continuous variables were presented as mean ± SD and comparisons done with the use of Student's t test. An alpha level of .05 was the criterion for statistical significance.

Results

Of the 354 cases from July 1985 through September 1992, 157 cases had at least one abnormal coagulation test. Of these, 112 patients had one or more abnormal results on platelet, PT, or fibrinogen determinations made at least 6 hours after envenomation and receipt of antivenom treatment. However, 81 of the 112 cases were excluded due to insufficient subsequent testing to detect recurrence (defined as at least one test result after normalization of the aberrant coagulation test). Therefore, 31 cases were analyzed for recurrence of coagulopathy. Comparison of the 31 included and 81 excluded patients revealed no differences in terms of sex, mean age, geographic region of bite occurrence, anatomic bite site, or use of antivenom (Table 1).

The 31 enrolled cases were similar to those in previous reports of snakebite. Patients were typically young men bitten on the hand (Table 1). The average age was 23.3 ± 17.6 years (range, 1 to 76 years). Bites occurred primarily on the upper (finger, hand, arm) or lower (toe, foot, calf/shin, thigh) limbs. Cases were enrolled from throughout the United States, but most came from the Southwest. Most patients were bitten by rattlesnakes; however, the precise species of crotalid snake involved was not usually identified (Table 2).

Of the 31 patients qualifying for analysis, coagulation tests showed recurrence in 14 patients (45%) (Table 3). The mean time to onset of recurrence was 51 hours (range, 17 to 238 hours). Ten cases (71%) of recurrence consisted of decreased platelets, and four cases (29%) involved coagulation factor abnormalities (eg, abnormal PT or serum fibrinogen). In nine cases (64%), the coagulation abnormalities during recurrence were the same type of abnormalities that were initially recorded. The severity of both initial and recurrent coagulopathy ranged from minimal to severe (Table 3). The worst cases of coagulopathy recurrence involved a fibrinogen level of 29 mg/dL and a platelet count of 36,000/mm3.

The average amount of antivenom administered in these 31 patients was 15 ± 8 vials (mean ± SD). No patient with recurrence received additional antivenom after the development of recurrence.

Discussion

Our results indicate that coagulation abnormalities are a common manifestation of crotalid snakebite among cases reported to the Arizona Poison and Drug Information Center. Many patients had at least one coagulation test, and 157 patients had at least one abnormal coagulation test. However, few patients had more than one coagulation assessment documented, and few had a repeat coagulation assessment, even after an abnormality had been documented. The clinicians involved apparently did not consider recurrence of coagulopathy a risk or, if they considered recurrence possible, chose not to assess it.

Our results indicate that such optimism may not be warranted. Among patients treated with crotalid antivenom in whom repeat laboratory testing was done, recurrence was not prevented. Recurrence of coagulopathy was documented in 14 (45%) of all patients in whom adequate testing was done (ie, at least one additional coagulation test after normal coagulation tests had been documented). The most common coagulation abnormality involved in recurrence was thrombocytopenia, but fibrinogen level and PT became abnormal in some patients.

In two patients, the repeat tests revealed serious coagulation abnormalities. The laboratory results of the other 12 patients appeared less dangerous, but recurrence was apparent. Since most of these patients had only one test after coagulation test results returned to normal, more serious coagulation abnormalities may have occurred but may have been missed due to the infrequent sampling. Of greater concern is that 81 patients (72%) receiving antivenom treatment and having a known coagulation abnormality had no repeat testing. It seems likely that a number of recurrences were missed due to the lack of further coagulation testing.

Until recently, neither the indexed medical literature nor standard textbooks on the topic of crotalid snakebite addressed the topic of recurrence.[9-11] However, related information suggests that conditions for coagulopathy recurrence may be present in victims of crotalid snakebite. Venom-induced anticoagulation may persist for a week or more when antivenom is not used.[6] Persistent thrombocytopenia has been recorded for 7 to 14 days after envenomation.[12,13] Indeed, one case [3] has been described of recurrent thrombocytopenia after the use of antivenin (Crotalidae) polyvalent after envenomation by the Blacktail rattlesnake (Crotalus molossus).

There are two fundamental requirements for a coagulation defect to occur after envenomation. First, a venom component that affects coagulation must be present in the blood. Crotalid venoms contain several components that affect platelets, the coagulation factors themselves, or both.[14] The second requirement is that the dose of the component is sufficient to cause an abnormality. An insufficient dose will cause only mild coagulation abnormalities or perhaps no apparent abnormality at all; it is not unusual to find normal PT and platelet counts in a patient with elevated fibrin split products -- a finding that would be missed if the test were not done. Effective treatment occurs when antivenom is administered and binds the venom component, thereby decreasing the concentration of the free form of that antigen to zero. Antivenin (Crotalidae) polyvalent has been shown to reverse most cases of coagulation abnormality induced by crotalid snakes.[15,16]

This model predicts that a coagulation abnormality could recur if the concentration of antivenom dropped below a critical level needed to bind all of the venom components. If the antivenom was excreted but venom continued to be absorbed slowly from the bite site, the continuing absorption might produce venom levels after the antivenom had been eliminated, allowing the coagulation defect to recur. Since the half-life of antivenin (Crotalidae) polyvalent is 61 to 194 hours, recurrence would not be expected for 2 to 3 days.[17] If a small dose of antivenom were used, it seems possible that recurrence could develop sooner.

This theory may explain the severe hypofibrinogenemia detected in case 1 (Table 3). The coagulation abnormality was initially mild, and the fibrinogen concentration was normal. The patient was administered 10 vials of antivenom and had a moderate allergic reaction consisting of chills and hives. No further antivenom was given, and allergic symptoms resolved with antihistamine and steroid treatment. The serum fibrinogen decreased to 60 mg/dL at 26 hours, then further decreased to 29 mg/dL at 38 hours. It is also possible that two different venom components were responsible for this effect, one that was absorbed initially and interfered with the PT and a second that was absorbed after the first and degraded fibrinogen.

The primary limitation of our study is its retrospective design. Several types of bias occur in retrospective studies. This makes it impossible to determine the incidence, frequency, or prevalence of recurrence from our data. In addition, a retrospective study by definition does not collect data in a systematic manner. It would have been helpful in our study to have more frequent determination of laboratory values. It is likely that we missed the peak coagulation abnormalities or missed cases altogether due to this problem.

Another limitation is that our data base primarily included patients from the southwestern United States. Of the six cases from outside the Southwest that included recurrence analysis, one case from the Southeast showed signs of coagulopathy recurrence. Lack of recurrent cases from the Northeast and Northwest are likely due to a small sample size.

Recurrence of coagulation abnormalities after envenomation by North American crotalid snakes occurred in patients administered crotalid antivenom. Most of these episodes were not life-threatening. However, at least two patients had coagulation abnormalities that could lead to spontaneous bleeding or hemorrhagic complications of injury such as motor vehicle trauma. It may be prudent to test for return of coagulation abnormalities in patients with crotalid snake envenomation. However, further research is needed to determine the appropriate clinical implications of this phenomenon.

References

  1. Bieber AL: Metal and nonprotein constituents in snake venoms. Handbook of Experimental Pharmacology. No. 52, Snake Venoms. Lee CY (ed). New York, Springer-Verlag, 1979, pp 295-306

  2. Nicole M, Boquet A: Elements de microbiologie generale et d'immunologie. Paris, Doin, 1925, p 268

  3. Hardy DL, Jeter M, Corrigan JJ Jr: Envenomation by the northern blacktail rattlesnake (Crotalus molossus molossus): report of two cases and the in vitro effect of the venom on fibrinolysis and platelet aggregation. Toxicon 1982; 20:487-493

  4. Bush SP, Jansen PW: Severe rattlesnake envenomation with anaphylaxis and rhabdomyolysis. Ann Emerg Med 1995; 25:845-848

  5. Russell FE: Medical problems of snakebite. Snake Venom Poisoning. Great Neck, NY, Scholium International, 1983, pp 306-343

  6. Budzynski AZ, Pandya BV, Rubin RN, et al: Fibrinogenolytic afibrinogenemia after envenomation by western diamondback rattlesnake (Crotalus atrox). Blood 1984; 63:1-14

  7. Dart RC, Seifert SA, Carroll L, et al: Affinity-purified, mixed monospecific crotalid antivenom ovine Fab for the treatment of crotalid venom poisoning. Ann Emerg Med 1997; 30:33-39

  8. Boyer-Hassen LV, Seifert SA, Clark F, et al: Recurrent and persistent coagulopathy following pit viper envenomation. Arch Intern Med 1999; 159:706-710

  9. Roberts JR, Otten EJ: Snakes and other reptiles. Goldfrank's Toxicologic Emergencies. Goldfrank LR, Flomenbaum NE, Lewin NA, et al (eds). Stamford, Appleton & Lange, 6th Ed, 1998, pp 1603-1619

  10. Otten EJ: Venomous animal injuries. Emergency Medicine Concepts and Clinical Practice. Rosen P, Barkin R (eds). St. Louis, CV Mosby, 4th Ed, 1998, pp 924-932

  11. Wallace JF: Disorders caused by venoms, bites, and stings. Harrison's Principles of Internal Medicine. Wilson JD, Braunwald E, Isselbacher KJ, et al (eds). New York, McGraw-Hill, 12th Ed, 1991, pp 2187-2189

  12. Hasiba U, Rosenbach LM, Rockwell D, et al: DiC-like syndrome after envenomation by the snake, Crotalus horridus horridus. N Engl J Med 1975; 292:505-507

  13. Tallon RW, Koch KL, Barnes SG, et al: Envenomation coagulopathy from snake bites (Letter). N Engl J Med 1981; 305:1347-1348

  14. Markland FS: Snake venom fibrinogenolytic and fibrinolytic enzymes -- an updated inventory -- on behalf of the registry of exogenous hemostatic factors of the scientific and standardization committee of the international society on thrombosis and haemostasis. Thromb Haemost 1998; 79:668- 674

  15. Burgess JL, Dart RC: Snake venom coagulopathy: use and abuse of blood products in the treatment of pit viper envenomation. Ann Emerg Med 1991; 20:795-801

  16. Riffer E, Curry SC, Gerkin R: Successful treatment with antivenin of marked thrombocytopenia without significant coagulopathy following rattlesnake bites. Ann Emerg Med 1987; 16:1297-1299

  17. Ownby CL, Reisbeck SL, Allen R: Levels of therapeutic antivenin and venom in a human snakebite victim. South Med J 1996; 89:803-806