VAPAGuide - Biomedical database - Terrestrial snakes, Pit vipers - Calloselasma rhodostoma

Clinic

Calloselasma rhodostoma

Studies

Malaysia

Reid et al. 1963a: 250 verified Calloselasma rhodostoma bites; identification: criteria not specified.

Classification:

Local envenoming:

None (no local swelling) 48/250.
Negligible (maximum extent of local swelling ≤1 cm difference in circumference between the bitten and healthy extremity) (24/250).
Mild (maximum extent of local swelling >1 to <4 cm difference in circumference between the bitten and healthy extremity, no necrosis) (57/250).
Moderate (maximum extent of local swelling >4 cm difference in circumference between the bitten and healthy extremity, no necrosis) (94/250).
Necrosis 27/250.

Systemic envenoming (haemostatic defect) 97/250:

None (no haemostatic defect) 56/250.
Mild: haemostasis with impaired clot formation (Reid et al. 1963b) 28/250.
Moderate (incoagulable blood) 32/250.
Severe (haemorrhagic syndrome) 37/250.

Thailand

Warrell et al. 1986: 46 Calloselasma rhodostoma bites; identification: morphological or immunological with ELISA (Ho et al. 1986a). All patients had incoagulable blood as a sign of systemic envenoming [see below, paragraph "Treatment (specific)"].

Case reports

Thailand

Brown and Brown 1987: 1 Calloselasma rhodostoma bite; identification: ELISA.

Brown and Brown 1988: 1 Calloselasma rhodostoma bite; identification: initially incorrectly identified as a Vipera russelli bite and treated with monospecific Russell's viper antivenom, QSMI. Identification of the snake revised and successful treatment with Malayan pitviper antivenom, QSMI.

Signs & symptoms

Autopharmacological effects

Gastrointestinal, pulmonary and cutaneous signs and symptoms that could be caused by autopharmacological processes were rarely or never observed (Reid et al. 1963a).

Local effects

Local pain is correlated with the degree of severity of the envenoming.

No swelling or only minor local swelling practically excludes a clinically relevant injection of venom. Swelling began within minutes after the bite, maximum after 24–48 h (in all patients ≥75% of the maximum extent of the local swelling was reached after 12 h).

Very good correlation of the difference in circumference between the bitten and healthy extremity and the extent of the swelling with the amount of venom injected and the degree of systemic envenoming:

Difference in circumference and systemic envenoming: difference in circumference at the initial investigation of the patient in cm < the number of hours since the bite, or maximum difference in circumference <3 cm: exclusion of severe systemic envenoming. In contrast, a third of the patients with a difference in circumference of 4–8 cm and all of those with a difference in circumference of ≥9 cm had severe systemic envenoming.
Extent of the swelling and systemic envenoming: 32/33 patients with swelling that extended beyond the level of the knee proximally and 8/12 who had swelling that extended beyond the level of the elbow proximally developed severe systemic envenoming. In contrast, 86 patients in whom the swelling was confined to the area below the knee or elbow, respectively, did not develop severe systemic envenoming.
Extravasation, blistering (haemorrhagin activity) and systemic envenoming: extensive blistering is a reliable sign of injection of a large amount of venom: 5/6 patients with this clinical sign had severe systemic envenoming, 1/6 moderate envenoming. Extravasation of erythrocytes and plasma leads to discolouration of the skin, but in dark skin this is barely discernible.

Local necroses 27/250, but these were more common with bites on the toes and fingers. Bacterial infections practically never occur if there is no necrosis (Reid et al. 1963a).

Haemostatic effects

Systemic bleeding 37/250, by definition all patients with severe systemic envenoming.

Haemoptysis 29/37, at the earliest 30 min after the bite, for a maximum of 4 days (this clinical sign probably overestimated, as in a large proportion of these patients, the blood actually originated exclusively from the oral cavity). Haemoptysis observed most notably in patients with pulmonary cavities (tuberculosis) (Warrell 1993, pers. comm.).

Discoid ecchymosis 21/37, at the earliest 2 ¾ h after the bite, for a maximum of 3 days.

Gingival bleeding 17/37, at the earliest 2 ¾ h after the bite, maximum duration 3 days.

Persistent oozing of blood from the site of the bite 11/37.

Haematemesis 3/37, at the earliest 2 h after the bite, for a maximum of 1 day. Macroscopic haematuria 1/37, 24 h after the bite, for 3 days. Intracerebral haemorrhage 1/37, 36 h after the bite, patient died. Shock 8/37; if the shock occurred late (>24 h), the most probable cause was hypovolaemia (fluid and erythrocyte loss into the region of the swelling). If the state of shock occurred before the bitten extremity was noticeably swollen, then other mechanisms must be assumed to have caused it. The patients with shock responded extremely well to antivenom treatment (see below) (Reid et al. 1963a).

Neurological effects

Neurological symptoms and clinical signs due to a direct effect of toxin were not observed in >1,000 cases (Reid et al. 1963a).

Renal effects

Rare and, if present, secondary (arterial hypotension) (Reid et al. 1963a).

Other signs & symptoms

Fever 12/250, of whom 5/12 had severe and 7/12 moderately severe systemic envenoming (Reid et al. 1963a).

Morbidity

The local necrotising effect of the venom is a common cause of morbidity. Gangrene can lead to the loss of toes, fingers or whole extremities; chronic infections (osteomyelitis) can occur (Warrell et al. 1986).

Local necroses occur chiefly with bites on the fingers and toes and if tourniquets are used.

If there is no necrosis, healing starts quickly. In these cases the time it takes for the swelling to subside is practically identical to the time it takes for complete recovery of the bitten extremity: these time periods range from an average of 3.6 days (1–10 days) for swelling to subside and complete recovery in patients with negligible local envenoming and up to 22.3 days (8–42 days) in patients with severe systemic envenoming. If necroses occur, the time to complete recovery is drastically increased to an average of 67.6 days (13–322 days) (Reid et al. 1963a). Infections complicate the course of recovery in patients with necroses.

Of 46 patients treated with TRC, GPO or Twyford antivenom, a 5-year-old girl developed local necrosis to such an extent that residual deformity and loss of function were expected. In 1 patient a necrotic toe had to be amputated. One patient suffered peroneal paralysis; fasciotomy was indicated but the patient refused the operation. 3 patients developed necroses of ≤9 cm² (Warrell et al. 1986).

Intracerebral haemorrhages with permanent neurological deficits are very rare.

Overall, incoagulability of the blood does not appear to be associated with a high risk of systemic bleeding (excluding blood loss into the local swelling), at least not in hospitalised patients. However, this is not necessarily the case for patients who engage in physical labour despite incoagulability of the blood, especially if they are exposed to trauma while working (Warrell et al. 1986).

Case fatality rate

Mortality, which is chiefly caused by haemorrhages and secondary infections, is low (Warrell et al. 1986).

On the whole, incoagulability of the blood does not appear to be associated with a high risk of fatal systemic bleeding, at least not in hospitalised patients. However, this is not necessarily the case for patients who engage in physical labour despite incoagulability of the blood, especially if they are exposed to trauma while working (Warrell et al. 1986).

Before specific antivenom became available, the mortality rate in hospitalised patients was around 1% (Reid et al. 1963a). In the study of Reid et al. (1963a), of a total of 291 patients with verified Calloselasma rhodostoma bites, only 2 patients died, and their deaths could only be indirectly attributed to the snakebites. One patient died of tetanus and one from a combination of an anaphylactic reaction to the antivenom, an intracerebral haemorrhage and severe pre-existing anaemia.

In 23 fatalities due to Calloselasma rhodostoma bites recorded in northern Malaysia between 1955 and 1960, the average time between the bite and death was 64.6 h (5–240 h), the median time 32 h (Reid et al. 1963a).

Of 46 patients treated with TRC, GPO or Twyford antivenom, none died (Warrell et al. 1986).

According to a study of fatal snakebites in rural areas of Thailand, 13/46 were caused by Calloselasma rhodostoma (Looareesuwan et al. 1988).

Laboratory and physical investigations

1. Haemostasis
Studies

Hutton et al., in press: 10 Calloselasma rhodostoma bites; identification: criteria not specified.

Reid et al. 1963b: 29 patients with systemic envenoming due to verified Calloselasma rhodostoma bites:

moderate systemic envenoming 8/29,
severe systemic envenoming 21/29.

Haemostatic parameters investigated: clotting time according to Lee and White, bleeding time according to Ivy, tourniquet test, clot quality test, prothrombin time, thrombin time, fibrinogen, platelets.

Warrell et al. 1986: for a description of the study, see below, "Treatment (specific)".

Type of haemostatic defect

Defibrin(ogen)ation due to a fibrinogen-coagulating ("thrombin-like") enzyme (Arvin). It differs from thrombin in that it only splits the fibrinopeptide A from the Aα-chain and not the fibrinopeptide B from the Bβ-chain. It is not inhibited by ATIII (?) or the ATIII-heparin complex, does not activate factor XIII and does not induce platelet aggregation (Stocker 1990). An activator of factor X might also play a role (Denson 1969).

Thrombopaenia: the speed with which platelets increase again after antivenom treatment implies that sequestered platelets are re-entering the circulation, rather than there being a fresh release of platelets from the bone marrow. According to clinical criteria, the activation, aggregation and deposition of platelet complexes do not play a role (Hutton et al., in press).

Haemostatic parameters

Overview haemostasis

(FSP)

aPTT

FSP

VIII

XIII

ATIII

tPA

α₂AP

Essential

bed-side

tests

Tests for full clinical assessment

Tests for research purposes

H	haemorhagic effects
+	definite evidence in human envenoming
CT	full blood clotting test
(FSP)	FSP rapid test
Tc	platlets
PT	prothrombin time
aPTT	partial thromboplastin time

TT	thrombin time
I	fibrinogen
FSP	fibrinogen split products
D	D-dimer
II, V, VII, X, XIII
	clotting factors
PC	protein C
ATIII	antithrombin III

PI	plasminogen
tPA	tissue plasmin activator
α₂AP	α₂-antiplasmin

In this overview, the deviations from normal are recorded for those haemostasis para- meters only, for which good evidence is documented in the literature.

A	The time until the incoagulability of the blood became evident on laboratory tests (clotting time) varied greatly. In patients who still had coagulable blood at the initial investigation, the blood became incoagulable after a period of between several hours and 2–3 days. The latency period until the incoagulability of the blood becomes detectable with the clotting time test can be explained by a state of equilibrium between the consumption and production of fibrinogen, a state which is maintained for variable periods of time (Ho et al. 1986b). That is why patients need to be hospitalised for a sufficient period of time after a Calloselasma rhodostoma bite, and coagulation needs to be investigated regularly. Clotting time should be assessed at least twice a day. If antivenom treatment is indicated and administered, clotting time is assessed every 6 h in order to confirm the efficacy of the antivenom treatment or to determine whether a further dose of antivenom is indicated. It must be noted, however, that there is a delay in the time it takes for the clotting time to return to normal compared to the disappearance of the venom from the bloodstream. This can be explained by the increase in fibrinogen, the speed of which may vary. As fibrinogen is an acute-phase protein, the speed with which it increases depends, among other factors, on the number of concurrent inflammatory reactions present that can accelerate this process (Hutton et al. 1990). The clotting time test is thus misleading if it is performed too early, i.e. antivenom may be administered although there is no longer any venom in the circulation. Other haemostatic tests are more sensitive and better suited to determining the status of the dynamic process of the haemostatic defect at any given moment (fibrinogen, FSP, see below), but they also do not solve the problem completely and are much more complicated. This problem will only be solved when reliable and fast ELISA tests for detection of venom antigen levels in the serum are available (Ho et al. 1986b). Incoagulability of the blood can re-occur after a long latency period (see below). This is probably due to continued absorption of venom from a depot in the region of the bite or saturation of extravascular binding sites. Recurrent venom antigenaemia was detected with the aid of ELISA tests, both in the presence and absence of circulating antivenom. In order to detect and treat relapses, patients should be kept in hospital for at least 5 days after initial treatment, and coagulation should continue to be investigated twice daily (Warrell et al. 1986).
B	The coagulation disorder is the prominent sign of systemic envenoming due to Calloselasma rhodostoma. Incoagulable blood was observed as early as 30 min after the bite. In 43 patients who did not receive antivenom, the average duration of incoagulability of the blood in 8 patients with severe envenoming was 8.0 days (5–11 days), and in 24 patients with moderate envenoming it was 6.6 days (1–15 days) (Reid et al. 1963a). In the absence of specific treatment, the coagulation defect can persist for more than 3 weeks (Reid et al. 1963a).
C	Correlation between the degree of severity of the haemostatic defect and other systemic signs of envenoming: none of the patients with a mild (haemostasis with impaired clot formation) or moderate haemostatic defect (incoagulable blood) had additional signs of systemic envenoming (Reid et al. 1963a); in other words, there are no clinical signs or symptoms that initially point to a haemostatic defect, as long as it does not have a complicated course (haemorrhage). Patients whose only symptom is incoagulable blood appear remarkably well, and the risk of systemic bleeding (excluding blood loss into the area of local swelling) appears to be small (Reid et al. 1963a, Reid and Chan 1968). This seems to apply at least to hospitalised patients, but not necessarily to patients who engage in physical labour despite incoagulability of the blood, especially if they are exposed to trauma while working (Warrell et al. 1986).
D	Haemorrhagic activity: haemorrhagins appear to be active primarily locally, i.e. in the region of the swelling.
E	Clotting time: 29/29 (study inclusion criterion) (Reid et al. 1963b). It is possible that a patient's blood may first become incoagulable as late as up to 72 h after the bite; in contrast, circulating antivenom (see ELISA) and increased serum FSP may be detectable significantly earlier. In groups of patients with local signs of envenoming, increased clotting time was present in 69/202 (Reid et al. 1963a) and 57/147 (Viravan et al. 1992).
F	Platelets: marked thrombopaenia 27/29; 10,000–99,000/mm³ (Reid et al. 1963b). Determination of platelet count at the time of maximum defibrin(ogen)ation directly before administration of antivenom: <150,000/μl (7/10), of whom 4/7 <50,000/μl. The patients with platelets <50,000/μl all had systemic bleeding, but only 1/6 patients with a platelet count >50,000/μl. If the platelet count was low prior to antivenom treatment, it increased rapidly, i.e. within 3 h, after treatment was commenced. The speed of the platelet increase implies that sequestered platelets re-entered the circulation, rather than there having been a fresh release of platelets from the bone marrow. According to clinical criteria, the activation, aggregation and deposition of platelet complexes do not play a role. The platelet count should be taken into consideration when monitoring the course of the envenoming and assessing indications for antivenom treatment (Hutton et al., in press).
G	PT: increased 28/29 (Reid et al. 1963b).
H	TT: increased in all patients investigated (Reid et al. 1963b).
I	Fibrinogen: in all patients investigated 10–160 mg/100 ml, mean level 46.9 mg/100 ml (normal levels 140–420 mg/100 ml) (Reid et al. 1963b). Slow increase in fibrinogen after administration of antivenom (12–24 h) (Hutton et al. 1990).
J	FSP: on average 238 μg/ml (35–750 μg/ml; n = 15) (Warrell et al. 1986). On average 550 μg/ml (10–750 μg/ml; n = 15) (Warrell et al. 1986). On average 200 μg/ml (36–750 μg/ml; n = 16) (Warrell et al. 1986). Increased FSP can be detectable early on, even if a patient's blood only becomes incoagulable as late as up to 72 h after the bite. This discrepancy may possibly be explained by a prolonged state of equilibrium between the synthesis and consumption of fibrinogen. As FSPs are eliminated from the bloodstream very rapidly, they are a good indicator for persistent or acute coagulation defects (Ho et al. 1986b).

2. Haemoglobin

32 patients were investigated, of whom 25 had severe and 7 moderate systemic envenoming. Hb decrease <1 g/100 ml: 8/32; 1.0–1.9 g/100 ml: 9/32; 2.0–2.9 g/100 ml: 7/32; 3.0–8.3 g/100 ml: 8/32 (Reid et al. 1963a).

The cause of significant anaemia was generally loss of erythrocytes into the bitten extremity (Reid et al. 1963a, Warrell et al. 1986).

3. Leucocytes

In systemic envenoming on average considerably >10,000/μl (Warrell et al. 1986).

4. ELISA

- Determination of the serum antivenom concentration.

- Determination of the venom concentration (venom antigen level).

The serum venom antigen level at the initial investigation of the patient is correlated with the incidence of spontaneous systemic bleeding, the incoagulability of the blood and the plasma fibrinogen and FSP concentrations, as well as with the extent of local swelling and the occurrence of necroses (Ho et al. 1986b). If reliable and fast test systems become commercially available in the future, it will be possible to monitor the serum venom antigen level and to assess the efficacy of antivenom treatment or indications for further doses of antivenom; see also Haemostatic parameter A, above (Ho et al. 1986b, Tan et al. 1992). In this way unnecessary antivenom treatment can be avoided. Adverse reactions to antivenom can thus be reduced, and antivenom can be spared.

First aid

The efficiency of tourniquets as a first aid measure was investigated in 6 patients who had applied them. In none of these patients was there convincing evidence of a delay in the absorption of venom into the blood circulation (Ho et al. 1986b).

Treatment (symptomatic)

Blood transfusion: efficacy: improvement in patients in shock or with anaemia, no effect on the duration of the coagulation defect (Reid et al. 1963b).
Fibrinogen: efficacy: temporary improvement of some haemostatic parameters, no lasting effect (Reid et al. 1963b).
Prednisone: efficacy: none, neither with regard to local nor systemic effects of the venom (Reid et al. 1963b).
A patient with anterior tibial compartment syndrome had a pressure of 63 mmHg in this compartment; fasciotomy was indicated, but the patient refused the operation. The consequence was a severe peroneal paralysis (Warrell et al. 1986).
ATIII (in addition to antivenom treatment) as yet only successful in animal experiments (Pukrittayakamee et al. 1990): this treatment approach is based on the assumption that the currently available antivenoms successfully neutralise the fibrinogen-activating components ("thrombin-like", Arvin), but not the additionally present activators of thrombin and factor X. The thrombin and factor Xa generated lead to consumption of ATIII.
Anti-infectious treatment: according to an investigation of Calloselasma rhodostoma in Thailand, there is a wide spectrum of principally gram-negative faecal rod-shaped bacteria in the oral cavity and to a lesser degree also in the venom. A combination of benzylpenicillin and gentamicin is recommended, preferably prophylactically; also prophylactic tetanus injection (Theakston et al. 1990a).

Treatment (specific)

Antivenoms

Malayan pitviper antivenom (TRC), Queen Saovabha Memorial Institute, Thai Red Cross Society, Bangkok, Thailand (unrefined, equine, freeze-dried, monospecific antivenom, has been produced since 1939; clinically evaluated in Malaysia by Reid et al. 1963b,c).

Anti-Malayan-pitviper venom serum (GPO), Thai Government Pharmaceutical Organization, Bangkok, Thailand (refined, equine, freeze-dried, monospecific antivenom).

Malayan pitviper antivenom, Twyford Pharmaceutical GmbH, Ludwigshafen, Deutschland (refined, caprine, liquid, monospecific antivenom).

Studies

Efficacy

With regard to local swelling: TRC does not appear to be effective (Reid et al. 1963c).
There was no significant difference with regard to the extent of local swelling and the speed with which it subsided between patients treated with TRC, GPO or Twyford antivenom (Warrell et al. 1986).
With regard to the formation of necroses: there was no significant difference with regard to the incidence of necroses between patients treated with TRC, GPO or Twyford antivenom, and there was no conclusive evidence for any of the antivenoms of activity against the necrotic effect of the venom (Warrell et al. 1986). However, it must be taken into account that as a rule, necroses cannot be prevented if more than an hour has elapsed between the bite and administration of antivenom (Warrell et al. 1975, Warrell et al. 1976b).
With regard to symptoms of shock (isolated observations): TRC is extremely effective (Reid et al. 1963b,c).
With regard to the haemostatic effect:
- Clinically: gingival bleeding ceased in all patients within 1 h and did not recur, regardless of whether they were treated with TRC, GPO or Twyford antivenom (Warrell et al. 1986).
- Coagulation defect:
  - TRC: if this antivenom is administered i.v., coagulability of the blood returns to normal within 9.0 h on average (2–18 h). If too little antivenom is administered, coagulability of the blood is not completely restored or the blood becomes incoagulable again (see below, Dose). Antivenom is just as effective if it is first administered days after the bite (Reid et al. 1963b,c).
    Initial restoration of coagulability within 4 h (1–10 h) after the initial dose (5 vials) (11/15). Late recurrence of incoagulability after 20–130 h (all 3 antivenoms), after the coagulation defect had initially been corrected (4/15) (Warrell et al. 1986).
  - GPO: initial restoration of coagulability within 4 h (1–16 h) after the initial dose (5 vials) (13/15). Late recurrence of incoagulability after 20–130 h (all 3 antivenoms), after the coagulation defect had initially been corrected (2/15) (Warrell et al. 1986).
  - Twyford: initial restoration of coagulability within 5.5 h (1–6 h) after the initial dose (5 vials) (16/16); in 2 of these patients the blood was incoagulable again after 1–3 days and repeat administration of antivenom was necessary. Late recurrence of incoagulability after 20–130 h (all 3 antivenoms), after the coagulation defect had initially been corrected (2/16) (Warrell et al. 1986).
- Thrombopaenia
  - TRC does not have a significant effect on the increase in platelets (Reid et al. 1963b).
    If the platelet count was low prior to antivenom treatment, it increased rapidly, i.e. within 3 h, after treatment was commenced (Hutton et al., in press).

Dose

TRC: 50–100 ml achieved lasting normalisation of blood coagulability (22/23), smaller doses appear to be ineffective (Reid et al. 1963c).

1–2 doses (of 5 vials each) restored blood coagulability permanently (12/15), 1 patient needed 4 doses and 2 patients were classified as treatment failures (definition: recurrence of incoagulability after 4 doses of TRC antivenom = 20 vials = 200 ml). In both cases coagulability of the blood was quickly restored with a single dose of Twyford antivenom (Warrell et al. 1986).

GPO: 1–2 doses (of 5 vials each) restored blood coagulability permanently (15/15) (Warrell et al. 1986).

Twyford: 1–2 doses (of 5 vials each) restored blood coagulability permanently (16/16) (Warrell et al. 1986).

Pharmacokinetics

TRC, GPO and Twyford antivenoms show very similar pharmacokinetic properties despite the different methods of production. The decrease in the serum concentrations after i.v. administration is biphasic: the initial quick phase is attributed to the creation of venom-antivenom complexes, while the final phase most probably corresponds to clearance of antivenom by the reticuloendothelial system. The recurrence of venom antigenaemia and incoagulability of the blood does not correlate with the elimination half-life of the respective antivenom. In contrast, incoagulability of the blood recurs when the serum antivenom level falls below a certain level (TRC <approx. 5 μg/ml, GPO <approx. 4 μg/ml, Twyford <approx. 1 μg/ml) (Ho et al. 1990).

Adverse reactions

TRC: antivenom reactions (antivenom i.m.) 10/33: moderately severe anaphylaxis 2/33, late serum reactions: severe 3/33, moderately severe 4/33, mild 1/33 (Reid et al. 1963c).

Acute anaphylactic reactions (antivenom i.v.) 13/15: severe 2/15 (Warrell et al. 1986).

Pyrogenic reactions 8/15 (Warrell et al. 1986).

Late serum reactions 2/15 (data not reliable, as the follow-up period was not sufficiently long) (Warrell et al. 1986).

GPO: acute anaphylactic reactions (antivenom i.v.) 6/15: severe 1/15. Significant difference between TRC and GPO (Warrell et al. 1986).
Pyrogenic reactions 1/15 (Warrell et al. 1986).

Late serum reactions 1/15 (data not reliable, as the follow-up period was not sufficiently long) (Warrell et al. 1986).

Twyford: acute anaphylactic reactions (antivenom i.v.) 8/16. Difference between TRC and Twyford not significant (Warrell et al. 1986).

Pyrogenic reactions 0/16 (Warrell et al. 1986).

Late serum reactions 1/16 (data not reliable, as the follow-up period was not sufficiently long) (Warrell et al. 1986).

Recommendations for antivenom treatment
1. Choice of antivenom

GPO or Twyford antivenom for systemic envenoming due to Calloselasma rhodostoma. The advantage of fast and comfortable handling with the liquid Twyford antivenom must be weighed against the greater stability at higher outdoor temperatures of the freeze-dried GPO antivenom (Warrell et al. 1986).

The efficacy of the TRC antivenom with regard to the coagulation defect is unsatisfactory, both clinically and in mouse experiments; the risk of anaphylactic and pyrogenic reactions is unacceptably high (Warrell et al. 1986).

2. Recommended initial dose

5 vials (Warrell et al. 1986).

Identification at the initial investigation of patients who will require more than 1 dose of antivenom (= 5 vials) for permanent restoration of blood coagulability: this is to be expected in patients with epistaxis, evident local swelling, marked venom antigenaemia prior to treatment, marked leucocytosis and low haematocrit (Warrell et al. 1986). A larger initial dose should be given to patients, especially children, with clinical signs of severe envenoming (shock, gastrointestinal bleeding, massive local swelling) (Warrell et al. 1986).

3. Treatment failure

This is defined as recurrent or persistent incoagulability of the blood after administration of a total dose of 20 vials of antivenom. A different specific antivenom should then be used (see above, "Dose: TRC"), and the species identification should be reviewed (Brown and Brown 1988).