Risk of immediate effects from F(ab)2 bivalent antivenin in Taiwan

Transcription

1 Wilderness and Environmental Medicine, 11, (2000) ORIGINAL RESEARCH Risk of immediate effects from F(ab)2 bivalent antivenin in Taiwan JIH-CHANG CHEN, MD; MICHAEL J. BULLARD, MD, PRCP; TE-FA CHID, MD; CHIP-JIN NG, MD; SHIDMN-JEN LIAW, MD Department ofemergency Medicine, Chang Gung Memorial Hospital, Kweishan Hsiang, Taoyuan Hsien, Taiwan, ROC (Drs Chen, Chiu, NG, and Liaw), and the Division of Emergency Medicine, University of Alberta, Edmonton, Alberta, Canada (Dr Bullard). Objective.-To evaluate the incidence of immediate adverse effects from equine fragment antigen binding F(ab)2 bivalent antivenin produced by the National Institute of Preventive Medicine (NIPM) in Taiwan, Methods.-A retrospective chart review of patients presenting to a 600-bed general hospital over a 3-year period with snakebite who were treated with NIPM antivenin. Results.-A total of 130 snakebite victims presented to the emergency department over the study period, and 159 vials of antivenin were given. One hundred two patients (78.5%; 95% CI: 70, 85) received only hemorrhagic bivalent antivenin, 2 (1,5%; 95% CI: 0, 5) received only neurotoxic bivalent antivenin, and the remaining 26 (20.0%; 95% CI: 13, 28) received both kinds of bivalent antivenin. Three received a second vial of hemorrhagic antivenin because of progression of symptoms. Fortytwo patients (32.3%; 95% CI: 24, 41) had positive skin tests, but following pretreatment with diphenhydramine and hydrocortisone, only 1 patient developed a skin rash thought to be related to antivenin. No patient developed an anaphylactic reaction. Conclusions.-The use of NIPM F(ab)2 antivenin in snakebite victims in Taiwan has a very low risk of acute adverse reactions. Key words: snakebite, F(ab)2 antivenin, anaphylaxis, Taiwan Introduction There are 6 major venomous snakes in Taiwan: 4 from the Crotalidae family and 2 from the Elapidae family. The Crotalidae family includes the Taiwan habu (Trimeresurus mucrosquamatus), green habu (Trimeresurus stejnegeri), hundred-pace snake (Deinagkistrodon acutus), and Russell's viper (Doboia russelli jormosensis). These all produce toxic venom that produces pronounced hemorrhagic effects. The Elapidae family includes the banded krait (Bungarus multicinctus) and Taiwan cobra (Naja atra), both of which produce a neurotoxic venom. The median reported snakebite mortality rate in Taiwan from 1965 to 1971 was 5.4%.1 The breakdown was 19.2% for the banded krait, 17.6% for the hundred-pace Presented at 7th International Conference on Emergency Medicine, Vancouver, British Columbia, CaIi:lda, March 25 to 29, Corresponding author: Jih-Chang Chen, Chang Gung Memorial Hospital, 5 Fu-Hsing St, Kwei-Shang, Tao-Yuan 333, Taiwan, ROC ( a @ms3.hinet.net). snake, and 1.7% for the Taiwan habu. 1 In response to public fear of snakebites, in 1970 the National Institute of Preventive Medicine (NIPM) developed 3 kinds of equine fragment antigen binding (Fab) antivenin: a bivalent hemorrhagic antivenin for the Taiwan habu and green habu, a bivalent neurotoxic antivenin for the banded krait and Taiwan cobra, and a monovalent antivenin for the hundred-pace snake. Antivenin is available through emergency departments (ED), and the Department of Health (DOH) recommends that an appropriate antivenin be offered to all victims of poisonous snakebite. 2 In accordance with this policy, 200 to 300 vials of NIPM hundred-pace snake antivenin, 3000 vials of bivalent hemorrhagic antivenin, and 2000 vials of bivalent neurotoxic antivenin are administered each year in Taiwan. The recent reported snakebite mortality in Taiwan is quite low, with an average of only 7.2 deaths per year from 1992 to On the basis of the antivenin used, we estimated that about 3000 poisonous snakebites

2 164 would occur per year and that the calculated mortality rate would be 0.24%. The Taiwan habu and green habu produce the lowest mortality rates among venomous snakebites in Southeast Asia (0.12%).4 Despite this, patients continue to request antivenin. Furthermore, the DOH has not softened its recommendations, and doctors are reluctant to withhold antivenin for fear of reprisal should there be an adverse outcome. Although antivenin is clearly beneficial in severe snakebites, a number of studies have suggested that antivenin should not be used routinely because of the expense and risk of adverse reactions. 5,6 The objective of this study was limited to determining the incidence of immediate adverse reactions to NIPM antivenin. Materials and methods SETIING AND PATIENTS The study involved emergency visits to a 600-bed general hospital located in northern Taiwan between May 1991 and March Snakebite victims who received at least 1 vial of antivenin were identified through computerized records. Charts were located through the medical records database by searching for patients with a discharge diagnosis of snakebite. These charts were referenced with pharmacy records for all patients receiving antivenin. TREATMENT REGIMEN In accordance with DOH recommendations, the ED policy during the study period was to ask snakebite victims to attempt snake identification from a picture chart to determine which kind of NIPM antivenin to use. Since the hundred-pace snake and the Russell's viper are rarely seen in this region, if the snake could not be identified, only the bivalent antivenins (hemorrhagic antivenin for the Taiwan habu and green habu, and neurotoxic antivenin for banded krait and Taiwan cobra) were used, The bivalent antivenin, which contains less than 0.2 grams of protein per milliliter, was produced from horse serum and supplied as 1 vial of crystallized powder and one 10-mL vial of distilled water. Patients received the antivenin as early as possible. All were skin tested by use of 1: 100 diluted antivenin solution. A positive skin test was represented by immediate swelling and redness > 5 cm in circumference within 30 minutes of the injection. If negative, 1 vial of antivenin in 10 ml of distilled water was diluted in 100 ml of isotonic saline and given as an intravenous (IV) infusion over 30 minutes. If the skin test was positive, the patient was pretreated with diphenhydramine (Benadryl) 50 mg IV and hydro- Chen et at cortisone (Solucortef) 100 mg IV, followed by IV infusion of the antivenin over 60 minutes. Only 1 vial of NIPM antivenin is recommended as an initial dose, with more given if symptoms progress. If the emergency physician decided to give both kinds of bivalent antivenin, 2 skin tests were done, and the second antivenin was given immediately after the first one. If the symptoms did not progress during an observation period in the ED, no further antivenin was given, and disposition of the patient (discharged or admitted for further care) was at the discretion of the emergency physician. DATA COLLECTION, VALIDITY, AND ANALYSIS Data reviewed included type of snake, patient age and sex, severity of envenomation on the basis of a grading scale as modified by McCollogh and Gennaro 7 (Table 1), response to skin test, amount of antivenin used, and adverse reaction to antivenin. The majority of charts (120 of 130) were reviewed by 2 blinded data extractors (Chen and Chin) to ensure interrater reliability. Kappa values were calculated. Data were analyzed using the SPSS-PC statistical software program (SPSS Inc, Chicago, IL). Categorical values are reported as counts and percentages. Continuous variables are reported as means and standard deviations. Odds ratios with 95% confidence intervals are also reported. Results PATIENTS Over the 3-year study period, 130 cases of venomous snakebite were treated in the ED. Patient demographic data revealed a male predominance (91:39), with an age range of 3 to 80 years (mean 44.1 ± 20.1 years). Patients were treated with antivenin within 3.1 ± 5.6 ( ) hours after being bitten. There were 3 patients suffering from grade 0, 99 patients from grade 1, 21 from grade 2, and 7 from grade 3 envenomations. SNAKE IDENTIFICATION The green habu (68 patients) and Taiwan habu (31 patients) accounted for 76.1 % (95% CI: 68, 85) of the bites. In the remaining 31 cases (23.8%; 95% CI: 17, 32), the snake was not positively identified. However, 28 of the wounds were swollen, erythematous, and painful, suggesting hemorrhagic venom, indicative of one of the habus.

3 Antivenin in Taiwan 165 Table 1. Envenomation grades* Grade o Degree None Minimal Moderate Severe Very severe * Modified by McCollogh and Gennaro 7 USE OF ANTIVENIN Findings Fang marks Pain, 1-5 inches of edema, and erythema Pain, 6-12 inches of edema, erythema, and occasional systemic symptoms Pain, > 12 inches of edema, erythema, systemic symptoms, and coagulation defects Rapid edema, erythema, ipsilateral ecchymoses, blebs, and coagulation defects A total of 159 vials of bivalent antivenin were administered during this study. Initially 102 patients (78.5%; 95% CI: 70, 85) received only bivalent hemorrhagic antivenin, 2 (1.5%; 95% CI: 0, 5) received only bivalent neurotoxic antivenin, and the remaining 26 (20%; 95% CI: 13, 28) received both (Table 2). Three patients initially receiving hemorrhagic antivenin received a second vial because of progression of their symptoms. ADVERSE REACTIONS Forty-two patients (32.3%; 95% CI: 24,41) had positive skin tests. Of the 104 patients receiving a single bivalent antivenin, 31 (29.8%; 95% CI: 21, 40) had a positive skin test. Of the 26 patients receiving both kinds of antivenin, 6 had positive skin tests to both, 4 to hemorrhagic antivenin only, and I to neurotoxic antivenin only. No patient developed anaphylaxis. Three patients developed skin rash during treatment in the ED, but only one seemed related to antivenin use. One appeared related to codeine sensitivity, and the other was likely a contact sensitivity to a dressing material. Twenty-nine patients returned to an outpatient clinic in the hospital. One of these had symptoms consistent with serum sickness. DATA RELIABILITY Charts were evaluated by 2 independent reviewers who achieved an almost perfect kappa (k) agreement regarding snake identification (simple agreement [SA] = 95.4%, k = 0.87) and skin test results (SA = 93%, k = 0.84). Simple agreement remained high in identifying Table 2. Type of antivenin received and the results of skin testing Antivenin No. of No. of No. ofpositive used patients vials skin tests H* t 31 (H) N 2 2 o(n) Hand N (H and N), 4 (H), 1 (N) Total * H indicates hemorrhagic bivalent antivenin; N, neurotoxic bivalent antivenin. t Three patients received a second vial of antivenin. those patients who developed postantivenin rashes, but the kappa was only fair (SA = 95.6%, k = 0.39). This was because only 3 patients developed rashes, and initially I reviewer did not try to identify the cause. Envenomation grading achieved an SA of 73.3% and a k of Almost all differences were a single grade-level difference, and the final adjudication was the maximal grade reached during the ED stay. Discussion Most antivenins are currently made from horse serum. Adverse reactions include urticaria, anaphylaxis, anaphylactoid reaction, and serum sickness. Hypersensitivity is dependent both on the origin and purity of the antivenin, with the former responsible for anaphylactic reactions and the latter for anaphylactoid reactions and serum sickness. 8 Antivenin dose directly contributes to the risk of serum sickness via an increased risk of immune complex formation and deposition. Human anaphylaxis is the result of a foreign antigen's (eg, horse serum) reacting with a native antigen's (generally immunoglobulin E) bond to tissue mast cells or circulating basophils. 5 An anaphylactoid reaction is due to the anticomplementary activity of an infused protein and produces symptoms similar to anaphylaxis. 8 Serum sickness presents with malaise, fever, arthralgias, lymphadenopathy, and swollen joints due to type 3 (complex-mediated) hypersensitivity, which occurs within 30 (usually 5-20) days after injection of the antivenin. 9,lo The clinical symptoms and signs of serum sickness typically resolve within 1 to 2 weeks. II In a study by Jurkovich et al 5 from Oregon, 26 patients received 507 vials of antivenin. Immediate hypersensitivity developed in 6 patients (23%), cutaneous manifestations occurred in 3, and anaphylaxis occurred in another 3; 50% of patients developed serum sickness.

4 166 White and Weber l2 from Texas reported on 23 patients receiving a total of 133 vials of antivenin. Thirteen (56%) had adverse reactions, and 9 developed severe anaphylaxis with shock. Both of the studies used Wyeth polyvalent antivenin (equine origin), which is the only available commercial antivenin against pit vipers in the United States. The primary cause of side effects elicited by commercial antivenins is their impurities. 13 Wyeth antivenin is prepared with venom from 4 different species of Crotalidae. Purification using ammonium sulfate fractionation of the hyperimmune serum to decrease immunogenicity results in a fraction of the neutralizing antibody's being lost. Although the potency diminishes significantly, amounts of equine proteins remain. I I To decrease the risk of allergic side effects, higher purity Fab fragment equine antivenins were developed in the late 1960s in Europe to treat European viper (Vipera berus) bites, with a reported 10% incidence of adverse effects. 14 In Nigeria, 4.3% of patients showed minor immediate adverse events that may have been related to a polyvalent antivenin composed of highly purified F(ab)2. 15 In Sri Lanka, 34% of patients had early mild reactions after treatment with a monospecific ovine Fab fragment antivenin. 16 In Taiwan, the NIPM antivenin is prepared using horse serum. Since 1970, pepsin has been used to digest Fab and Fc. This results in F(ab)2 fragments, which are then purified using ammonium sulfate. Fab immunotherapy offers a higher specificity and affinity, rapid and extensive tissue distribution, low antigenicity, and rapid clearance. 13 These characteristics may explain the low rate of adverse reactions in our series and other studies. 14, 17 Although skin testing is an unreliable predictor ofthose patients who will develop an allergic reaction, I 8-20 it is still recommended unless a patient is hemodynamically unstable and requires the antivenin regardless of test result. Low-dose subcutaneous adrenaline's preventing acute adverse reactions due to antivenin was reported recently.21 Pretreatment with steroids and antihistamines has been suggested for patients with positive skin tests. 10, 14,22 But a randomized, double-blind, placebocontrolled study revealed that promethazine (HI antihistamine) could not produce prophylaxis against early anaphylactic reactions to antivenin for Bothrops snakebites. 23 The pretreatment policy in our positive-skin-testing patients may reduce our immediate adverse reaction rate. In our negative-skin-testing patients who did not receive pretreatment, none developed immediate adverse reactions. In the study by Karlson-Stiber and Persson l4, all their patients received cortisone in varying doses before antivenin administration. Ifthey had performed skin tests, the majority of their patients may not have needed cortisone therapy. Chen et a1 This study catalogues the prevalence of adverse reactions to antivenin in a region where snakebites are relatively common and where a standardized approach to treatment exists. In keeping with DOH recommendations, all patients were treated with antivenin, even though 78.5% were grade 0 or 1. Moreover, 42 patients (32%) had positive skin tests, suggesting some risks to patients exposed to the antivenin. Patients were treated within 3.1 hours on average, and all had good outcomes with no episodes of anaphylaxis, even among those with positive skin tests. Although this review provides no evidence to support the value of antivenin for all snakebites, it does appear that the risk of anaphylactic and anaphylactoid reactions is extremely low for NIPM antivenin. This study has several potential limitations. As the study was retrospective, cases may have been missed and some relevant data not recorded on the charts. The computer system, however, allows for patient identification by diagnosis, and the pharmacy records of patients receiving antivenin allows cross-referencing. No formal follow-up was arranged, and as such, only 29 patients returned to an outpatient clinic in the hospital. Among these returnees, a single patient who received both antivenins after a positive skin test presented with symptoms consistent with serum sickness. Serum sickness prevalence among the remaining 101 patients cannot be determined. Because this is the largest hospital in the community, however, it is unlikely that patients would seek out alternative hospitals for care of potential complications. Because only 3 patients received more than 1 vial of the same antivenin, the risk of serum sickness should have been minimized. Given that the risk of mortality among this group of patients (bitten by habu) if they had received no antivenin would have been negligible, a careful prospective study is needed to determine in which patients it might be reasonable to withhold antivenin and avoid exposing them to costly, potentially allergenic therapy. Conclusion The liberal, early use of NIPM F(ab)2 bivalent antivenin for habus, as recommended by the Taiwan DOH, appears to be associated with a low incidence of adverse reactions. Although the potential for allergies exists on the basis of skin test results, antivenin induced anaphylaxis was not observed in patients pretreated with diphenhydramine and hydrocortisone. Acknowledgment The authors wish to thank Dr Brian Rowe for his critical review of this manuscript and assistance in calculating the kappa values.

5 Antivenin in Taiwan References 1. Sawai Y, Koba K, Okonogi T, et al. An epidemiological study of snakebites in the Southeast Asia. Jpn J Exp Med. 1972;42: Antivenin of Tr. muscrosquamatus and Tr. gramineus lyophilized [package insert]. Taiwan, Republic of China: National Institute of Preventive Medicine, Department of Health; Department of Health. Annual Statistics. Taiwan, Republic of China: Department of Health; Snakebites in the tropics. Lancet. 1972;785: Jurkovich GJ, Luterman A, McCullar K, Ramenofsky ML, Curreri pw. Complications of Crotalidae antivenin therapy. J Trauma. 1988;28: Burch JM, Agarwal R, Mattox KL, Feliciano DV, Jordan GL Jr. The treatment of crotalid envenomation without antivenin. J Trauma. 1988;28: McCollogh NC, Gennaro IF. Evaluation of venomous snakebite in southern United States from parallel clinical and laboratory investigations. J Fla Med Assoc. 1963;49: Sutherland SK. Serum reactions. An analysis of commercial antivenoms and the possible role of anticomplementary activity in de novo reactions to antivenoms and antitoxins. Med J Aust. 1977;1: Thomas A, SmithTA II, Figge HL. Treatment ofsnakebite poisoning. Am J Hosp Pharm. 1991;48: Forks TP. Evaluation and treatment of poisonous snakebites. Am Fam Physician. 1994;50: , Horotz RS, Dart RC. Antivenins and immunobiologicals: immunotherapeutics of envenomation. In: Auerbach PS, ed. Wilderness Medicine: Management ofwilderness and Environmental Emergencies. 3rd ed. St Louis, MO: Mosby Year-Book; 1995: White RR, Weber RA. Poisonous snakebite in central Texas. Ann Surg. 1991;213: Sullivan JB Jr. Past, present, and future immunotherapy of snake venom poisoning. Ann Emerg Med. 1987;16: Karlson-Stiber C, Persson H. Antivenom treatment in Vipera berus envenoming-report of 30 cases. J Intern Med. 1994;235: Meyer Wp, Habib AG, Onayade AA, et al. First clinical experiences with a new ovine Fab Echis ocellatus snake bite antivenom in Nigeria: randomized comparative trial with Institute Pasteur Serum (Ipser) Africa antivenom. Am J Trop Med Hyg. 1997;56: Ariaratnam CA, Meyer Wp, Perera G, et al. A new monospecific ovine Fab fragment antivenom for treatment of envenoming by the Sri Lankan Russell's viper (Daboia Russelii Russelii): a preliminary dose-finding and pharmacokinetic study. Am J Trap Med Hyg. 1999;61: Otero-Patino R, Cardoso JL, Higashi HG, et al. A randomized, blinded, comparative trial of one pepsin-digested and two whole IgG antivenoms for Bothrops snake bites in Uraba, Colombia. Am J Trop Med Hyg. 1998;58: Thachil RT, Tony JC, Jude E, Ross C, Vincent ND, Sridhar CB. Antisnakevenom: an over-used medication. Trap Doct. 1992;22: Malasit P, Warrell DA, Chanthavanich P, et al. Prediction, prevention and mechanism of early (anaphylactic) antivenom reactions in victims of snakebites. Br Med J (Clin Res Ed). 1986;292: Wingert WA, Wainschel 1. Diagnosis and management of envenomation by poisonous snakebites. South Med J. 1975;68: Premawardhena Ap, de Silva CE, Fonseka MM, Gunatilake SB, de Silva HJ. Low-dose subcutaneous adrenaline to prevent acute adverse reactions to antivenom serum in people bitten by snakes: randomised, placebo controlled trial. BMJ. 1999;318: Kurecki BA III, Brownlee HJ Jr. Venomous snakebites in the United States. J Fam Pract. 1987;25: Fan HW, Marcopito LF, Cardoso JL, et al. Sequential randomised and double blind trial of promethazine prophylaxis against early anaphylactic reactions to antivenom for bothrops snake bites. BMJ. 1999;318: