Bungarus multicinctus multicinctus Snakebite in Taiwan

Introduction

Six venomous land snakes are considered medically important in Taiwan, including Bungarus multicinctus multicinctus (B. multicinctus) and Naja atra in family Elapidae and Trimeresurus stejnegeri, Protobothrops mucrosquamatus, Deinagkistrodon acutus, and Daboia siamensis in family Viperidae.^1,2 Bungarus multicinctus, also known as the many-banded krait or umbrella snake, is the only krait species in Taiwan. Its inhabitated area extends throughout south China, Myanmar, Laos, north Vietnam, and Thailand.³ Bungarus multicinctus bites are rare and account for only 7.5% (range: 0–17.1%) of all snakebites in Taiwan.¹ However, the bite of B. multicinctus was ranked as second in lethality only to that of D. acutus.¹ The case-fatality rate is reported as 7–50%, with death occurring 6–30 hours post-bite if no antivenom therapy was provided.^4–7 Although the Taiwan government produces specific antivenoms against the six major venomous snakes, published recommendations on managing B. multicinctus bites are scarce.^6,8 We conducted a retrospective case series study to better understand the clinical outcome and risk factors associated with respiratory failure after B. multicinctus envenomation at two medical centers, Taichung Veterans General Hospital (VGH-TC) in central Taiwan and Taipei Veterans General Hospital (VGH-TP) in northern Taiwan in an attempt to provide better management guidelines and foster further studies on B. multicinctus snakebites.

Materials and Methods

Results

Discussion

We carried out a preliminary analysis of the predictors of respiratory failure based on patients' clinical characteristics. General weakness, ptosis, dysarthria, and dilated pupils were significantly associated with the development of respiratory failure. However, after careful review of the medical records, we found that the abovementioned manifestations were usually recorded during or shortly after the onset of respiratory failure that occurred quite fast (< 6–7 hours) post-bite. Therefore, their predictive role could not be accurately determined in such a retrospective study. As a result, all suspected cases of envenomation should be closely monitored during the first few hours, and sufficient doses of antivenom should be administered promptly if signs and symptoms of B. multicinctus envenomation appear.⁹ Notably, seven of our patients who developed respiratory failure received no antivenom or only a small fraction (0–10 vial) of the eventual total dose prior to intubation. Unfamiliarity with managing rare snakebites and limited timely access to antivenom may delay or impede timely antivenom administration.

In Taiwan, only envenomation by B. multicinctus, N. atra, and D. siamensis is known to cause neurologic effects. With D. siamensis, the principal clinical effects are hematologic disturbances and renal failure in addition to wound swelling and pain.¹⁰ Although presynaptic neurotoxic phospholipase A₂ (PLA₂) is present in the venom of D. siamensis, in human cases only transient drowsiness, dizziness, and faintness, but not neuromuscular block were infrequently observed.^10,11 In N. atra snake envenomation, the principal effect is wound necrosis or infection, with a few cases manifesting transient and mild weakness or ptosis in the presence of wound swelling and pain.^2,12,13 On the other hand, B. multicinctus causes negligible wound reaction despite the presence of overt neuromuscular blockade and descending paralysis.^7,9 Although assay of B. multicinctus venom was not performed in this study because of its unavailability in all health-care facilities in Taiwan, physicians could easily differentiate B. multicinctus envenomation from other venomous snakebites based on clinical examination. In addition, the venom yield of B. multicinctus is only one-fifth to that of Bungarus caeruleus (4.4 mg versus 22.7 mg), which may make its venom detection in serum samples more difficult.^2,9,14 In this study, all cases were carefully examined by physical examination, serial wound inspection, and a relevant history; in addition, the signs and symptoms were similar between definite, suspected, and clinical case groups (data not shown). Given that only single krait species (Bungarus sp.) was distributed to Taiwan, clinical diagnosis of B. multicinctus envenomation is very likely to be convincible.

Other differential diagnosis of respiratory paralysis with/or resembling descending paralysis may include botulism, myasthenia gravis, and Miller–Fisher syndrome.^15,16 All 12 patients with respiratory paralysis in this study were carefully assessed for the abovementioned disorders. All of them were bitten when they caught the snake and were free from neuromuscular symptom before the snakebite. Only one case consumed alcohol every day and the blood alcohol level before performing endotracheal intubation was 171 mg/dL. There was also no cluster occurrence of paralysis among the study subjects, which made food-borne botulism less likely.¹⁶ Moreover, all patients recovered rapidly after the administration of specific bivalent antivenom for B. multicinctus and N. atra, a finding that should have excluded the diagnosis of botulism.

Bungarus multicinctus venom contains several neurotoxins and enzymes, including muscle or neuronal nicotinic acetylcholine receptor (AChR) and muscarinic receptor antagonists. α-Bungarotoxin, first isolated from B. multicinctus in Taiwan, binds to the mammalian muscle nicotinic AChR with high affinity, causing a rapid curare-like neuromuscular block, whereas β-bungarotoxin with PLA₂ activity acts presynaptically in the neuromuscular junction, causing damage to nerve endings and further inhibiting impulse conduction.^17,18 κ-Bungarotoxin blocks postsynaptic nicotinic AChRs in autonomic ganglia,¹⁹ whereas the novel γ-bungarotoxin, which has less characterized clinical effects, inhibits muscarinic and nicotinic AChRs and platelet aggregation.²⁰ The weight percentages of α-, β-, and κ-bungarotoxin in B. multicinctus venom are 13–61%,^21,22 22–26.4% (sum of β1–5),^22,23 and 0.1%,¹⁹ respectively. The median lethal doses (LD₅₀) of the crude venom and α-, β-, and γ-bungarotoxin in mice were 0.16, 0.3, 0.089 (s.c),²² and 0.15 (i.v.) mg/kg,²⁴ respectively. These are rough estimates because venom from individual snakes may not include all the toxin isoforms. Other enzymes or peptides of unevaluated clinical significance are present in the venom as well.²⁵

The hallmark of B. multicinctus envenomation is neuromuscular blockade caused by bungarotoxins.^9,22 Twenty-seven percent of our patients developed respiratory failure within 1.5–6.5 hours post-bite. It is generally accepted that presynaptic neurotoxin (e.g., β-bungarotoxin) is the primary cause of muscular paralysis.^9,26 Rapid fixation and irreversible damage to the nerve terminals by β-bungarotoxin may account for the resistance to antivenom therapy when administration is delayed.^22,26 Clinical experience suggested that unless antivenom is administered within 4 hours after a snakebite causing presynaptic neurotoxicity, most patients continued to deteriorate and required endotracheal intubation with mechanical ventilation.²⁷ In our study, patients developed respiratory failure in spite of antivenom administration within 4 hours post-bite, and in a few patients, the preintubation doses were significantly higher than the currently recommended dose of 2–4 vials.² Our observation suggests that the recommended doses of B. multicinctus and N. atra antivenom are not sufficient to prevent respiratory failure, perhaps because of its intrinsic pharmacokinetic properties compared with the far more rapid distribution of toxins.^28,29 On the other hand, our study is consistent with previous reports that early antivenom administration was associated with accelerated recovery which might have been related to various proportions of β-bungarotoxin in the individual snakes.^23,30,31 Nevertheless, the contribution of α-bungarotoxin in neuromuscular paralysis is unassessed.

The second most frequent distinct symptom of Bungarus snakebite is pain remote from the bite site of undetermined mechanism, which has been described in cases of Bungarus spp. envenomations including abdominal pain due to bites from Bungarus cf. sindanus,³² B. caeruleus,³³ and Bungarus lividus³⁴; generalized burning sensation due to bites from B. lividus³⁴; or muscle pain due to bites from B. caeruleus,³⁵ Bungarus candidus,³⁶ Bungarus niger,³⁷ and B. multicinctus.⁷ A significant proportion of our patients developed severe general pain. Patients described the pain as similar to myalgia after heavy exercise but much more intense, especially in the head and neck and trunk in addition to the limbs. It is known that curare-like muscle relaxation (e.g., α-bungarotoxin or nondepolarizing muscle relaxants) is not associated with muscle pain, whereas depolarizing relaxation induced by succinylcholine can lead to muscle pain in the throat, shoulder, and abdomen.^38–40 The biochemical mechanisms that have been associated with succinylcholine-induced myalgia included increased myoplasmic calcium, activation of cellular PLA₂ and resultant myocyte phospholipid degradation, and release of free fatty acids.⁴¹ β-Bungarotoxin, which has PLA₂ activity, might degrade the postsynaptic muscle membrane as well as the presynaptic nerve terminals.⁴² In fact, the CK level was slightly, but significantly higher in patients with general pain, a phenomenon that is also observed in succinylcholine-induced myalgia.⁴¹ Moreover, the painful area occurred in anatomic regions did not correspond to the distribution of peripheral nerves. It is possible that general pain developed after B. multicinctus envenomation is related to the enzymatic effect of β-bungarotoxin on muscles. In our observation, patients usually did not experience pain relief after nonsteroidal anti-inflammatory drugs (NSAIDs, e.g., ketoprofen or ketorolac) administration. Instead, the pain was ameliorated by tramadol treatment. Further prospective evaluation is warranted to quantify the nature of pain and see whether better pain management can be achieved with antivenom, NSAIDs, or tramadol/other opioids.^39,40

Autonomic effects have been described after B. multicinctus^4,6,7 and other Bungarus spp. envenomations,^34,43 including dilated pupils, nausea or vomiting, urine retention, changes in blood pressure or heart rate, sweating, and hyperglycemia. Prolonged decreases in parasympathetic functions (i.e., mydriasis, tachycardia, hypertension, constipation, or difficulty in urinating) lasting for months to years, have been reported in patients with B. candidus envenomation who were not given antivenom.⁴⁴ Two patients in our study had prolonged constipation or GI upset, and two others experienced abnormal taste sensations. β-Bungarotoxin may also have an inhibitory effect on the parasympathetic nerve terminals at the neuroeffector junction in mammals.⁴⁵ By contrast, α-bungarotoxin did not affect ganglionic nicotinic neurotransmission in mammals.⁴⁶ Although in in vitro studies, the minor component κ-bungarotoxin (0.1%) potently blocked postsynaptic nicotinic AChRs of various species, including avian ciliary ganglia (parasympathetic system), avian and rat sympathetic ganglia, cultured bovine chromaffin cells, and even invertebrate central nervous system cells,^46,47 given the difference in weight percentages of β- and κ-bungarotoxin in the venom, these effects are more likely to be induced by β-bungarotoxin.¹⁹ The mechanism of abnormal taste remains unknown but might also involve the acetylcholine transmission pathway.⁴⁸ As the effect on autonomic and taste functions was not vigorously evaluated in this study, the clinical significance of these effects may have been overlooked.

Recently, a new syndrome of life-threatening hyponatremia was identified in cases of B. multicinctus and B. candidus envenomation in Vietnam.^49,50 Approximately 42–50% of patients who did not receive antivenom developed significant hyponatremia (< 130 mEq/L) 2–3 days post-bite.^7,30,49 In Taiwan, life-threatening hyponatremia was not observed after B. multicinctus envenomation, which might be due to geographical variation in the venom of B. multicinctus. Nevertheless, early administration of specific antivenom might have prevented serious hyponatremia, as only three cases in this study had developed hyponatremia and all spontaneously recovered.

Four types of antivenom, ammonium sulfate–precipitated and lyophilized formulations containing F(ab′)₂, are available in Taiwan. They are bivalent B. multicinctus and N. atra antivenom, bivalent T. stejnegeri and P. mucrosquamatus antivenom, and monovalent D. acutus and D. siamensis antivenom.² One vial of bivalent specific antivenom for B. multicinctus and N. atra roughly neutralizes 2 mg venom of B. multicinctus.² Because the mean quantity of B. multicinctus venom extracted by milking is 4.4 mg, the Taiwan Poison Control Center has thus recommended 2–4 vials of antivenom to be administered in an envenomed case.² The abovementioned recommendation, however, was not validated in any prior study. Because F(ab′)₂ fragment has much slower distribution kinetics than venom, a higher dose may have been required to achieve effective tissue accumulation after envenomation.²⁹

In this study, not all treatments adhered to the protocol recommended by the Taiwan Poison Control Center.² Therefore, it is difficult to evaluate the effectiveness of different dosing regimens of antivenom. However, it appears that the dosage administered in respiratory failure cases is higher than that in patients without respiratory failure. Hung and others recommended that first and second doses totaling 5–10 vials of antivenom given 6–8 hours apart be used to treat a B. multicinctus bite.³⁰ Silva and others reported using 20 vials of antivenom once to treat a patient bitten by B. caeruleus.^9,14,51 Based on the limited evidence, we suggest that an initial dose of four vials of antivenom be administered in all envenomation cases, and a subsequent four vials be considered if the patient's condition is deteriorating.⁵² Further evaluation is needed to determine the optimal dosing regimen for management of B. multicinctus bites.

An antivenom skin test is recommended by the manufacturer before administration of the antivenom.² In our study, four patients had a positive skin test, and skin rashes occurred in another four patients (11.4%) who had a negative skin test. The reported incidence of early adverse reactions after the administration of ammonium sulfate–precipitated IgG or F(ab′)₂ antivenom ranges from 10–87%.⁵³ It has been proposed that routine premedication with low-dose epinephrine would decreased severe allergic reaction, in which the incidence of such reaction was high.⁵⁴ In Taiwan, only 0.7–3% of patients receiving bivalent T. stejnegeri and P. mucrosquamatus antivenom were reported to have a skin allergy.^55,56 Moreover, anaphylaxis was only reported in this study despite the fact that nearly 1,000 cases received antivenom in Taiwan each year.¹ Because of the overall low incidence of allergic reactions after antivenom administration, routine premedication with epinephrine or antihistamines and corticosteroids is not recommended in Taiwan.⁵⁷ Moreover, all patients received antivenom regardless of the result of antivenom skin test in this study. Given that the administration of a small test dose of antivenom to identify patients who may develop acute adverse reactions to the antivenom is both insensitive and nonspecific,⁵⁴ the antivenom skin test should better be omitted.⁵²

Conclusions

Of 44 cases of B. multicinctus snakebite, nine (20.5%) were asymptomatic, 12 patients (27.3%) experienced respiratory failure, and 12 (27.3%) had transient but debilitating severe general pain. The toxic effects of venom began between 0.5–6 hours post-bite and respiratory failure occurred within 1.5–6.5 hours post-bite. The neuromuscular, pain, or autonomic effects may have been caused, at least partly, by β-bungarotoxin. Because patients envenomed by B. multicinctus may develop respiratory failure within a short time, all suspicious envenomed cases should be closely monitored in the first few hours. We suggest that four vials of antivenom be administered in all envenomed cases and a subsequent four vials be considered if the patient's condition is deteriorating. Further prospective study is warranted to characterize the nature of pain and to identify the optimal antivenom dosing regimen in the management of B. multicinctus envenomation.

Disclaimer: The authors alone are responsible for the content and writing of this paper.