Antivenoms a Review of Current Status and Future Developments

BioDrugs 1997 May; 7 (5): 366-375 BIOPHARMACEUTICALS 1173-8804/97/0005-0366/$05.00/0

Antivenoms A Review of Current Status and Future Developments

R. David G. Theakston 1 and Damon C. Smith2 1 World Health Organization Collaborating Centre for the Control of Antivenoms, Liverpool School of Tropical Medicine, Liverpool, England 2 Medical College of Sf Bartholomew's Hospital, London, England

Contents Summary ...... 366 1. Standard Methods of Antivenom Production 367 2. Storage of Antivenoms . . . . 368 3. Labelling of Antivenoms ...... 368 4. Different Types of Antivenoms ...... 369 5. Recent Developments in Immunotherapy 369 5.1 Choice of Host Animal . 369 5.2 Affinity Purification ...... 370 5.3 Use of Fab Fragments ...... 370 5.4 Future Developments in Immunotherapy 370 6. Antivenom Reactions ...... 371 7. Assessment of Antivenom Quality 372 7.1 Conventional Methods 372 7.2 Clinical Trials. 372 8. Conclusions ...... 373

Summary The first advance on crude filtered immune serum or plasma for treating sys temically envenomed bite victims was the use of IgG (predominantly equine) precipitated by salt treatment. This was followed by the development of F(ab')z anti venoms, prepared using pepsin digestion to remove both the highly reactive and nonspecific Fc part of the IgG molecule and also other non-antibody proteins from the material. Within normal limitations, both mono specific (raised against the venom of a single species) and poly specific (raised against a mixture of venoms of different species) F(ab')z anti venoms have proved to be generally effective. Many prepa rations are poor for a wide range of reasons often not related to the method of preparation. The main problem associated with them is the high reaction rate in patients caused by complement-mediated anaphylactic reactions. The most significant recent development in immunotherapy has been the pro duction of ovine Fab anti venoms prepared by replacing pepsin digestion with papain digestion. Theoretically, these possess advantages over F(ab')2 anti venoms, such as a greater volume of distribution and more rapid kinetics. A major disadvantage of Fab anti venoms is, however, their shorter clearance time, which Recent and Future Developments of Antivenoms 367

may result in inadequate blood anti venom concentrations for neutral ising venom entering the system late from a venom depot at the bite site. Affinity purification of both Fab and F(ab')2 anti venoms also results in a major increase in specific activity when compared with the original serum source, but this adds greatly to the cost of production, rendering its use prohibitive in the rural tropics where bites and stings are a major problem. The use of sheep as opposed to horses for immunisation also results in a cheaper product, due to the lower cost of animal maintenance; theoretically, sheep preparations should also be safer, causing fewer sensitivity reactions, due to the nature of ovine as opposed to equine protein. Additionally, it is a great advantage in the rural tropics to have a Iyophilised, as opposed to a liquid, anti venom because of the longer shelf-life of the former at ambient tropical temperatures. However, Iyophilisation does add considerably to the cost of production, involves additional testing and has also in the past been reported to result in some decrease in neutral ising potency. There may well be a case for the development of a combined ovine Fab/F(ab')2 anti venom which combines the advantages of each product. Alternatives to con ventional anti venoms should also be explored in the future.

1. Standard Methods of curs.!4,5] Unfortunately, horses do not tolerate such Antivenom Production adjuvants well,f6] Other additives, such as benton ite,!7] aluminium hydroxide gel and sodium algin Because of biochemical and immunological dif atel8] also have a similar overall effect. ferences between snake and other animal venoms Although it is remarkable that even today a few from different areas of the world, anti venoms are producers supply crude filtered horse serum alone produced commercially by about 70 different lab with all its additional attendant problems,!2] most oratories in 37 countries. I 1,2) Antivenoms are usu antivenoms usually comprise an ammonium sul ally made by immunising large animals with an phate-precipitated, pepsin-digested, concentrated individual venom or a range of venoms from large F(ab'h fraction of the IgG molecule (fig. I) sepa venom pools, which take account of seasonal, ge rated from the serum or plasma of hyperimmunised ographical, genetic and sex-related variation in horses.l7,9.11] The heat-labile fraction of the IgG venoms within the same species. Ideally, an anti molecule, the Fc fragment, produces no direct anti venom should be capable of neutralising all the venom effect and is responsible for some of the toxic components in a venom, and to achieve this early reactions induced by inadequately purified it is generally considered best to use unmodified anti venoms. This fragment is therefore removed by venom for immunisation, rather than to rely on de pepsin digestion; the ammonium sulphate used for toxification techniques which may result in thc rc precipitation of the IgG is eliminated by dialysis duction or complete loss of important antigens. 13 ) and lipids are removed by aluminium hydroxide With highly toxic venoms, however, such as those gel agglutination. Filtration is carried out by mo of some elapids and sea snakes, the use of formol lecular filtration using gels such as Sephadex®. The toxoids (prepared under acid conditions, as alkalin final product is subsequently tested for pyrogenic ity can lead to loss of both toxicity and antigenic activity (usually using rabbits), bacterial sterility, ity) was, in the past, considered unavoidable. Ven immunological specificity using electrophoresis or oms used in the preparation of anti venoms are best immunodiffusion, and potency before the final pro presented combined with an adjuvant, such as cedure of ampouling under sterile conditions to Freund's Complete, which results in a decrease in produce either a lyophilised preparation or a toxici ty because onl y gradual release of venom oc- straightforward fluid anti venom.

F (ab'h (divalent) Fab (monovalent) World Health Organization Collaborating Centre for the Control of Antivenoms (WHO CCCA) maintained ampoules of I batch (number 2070(6) of Behringwerke North and West Africa anti venom L___l whose expiry date was June 1980 at 22 to 24°C for over 15 years with no loss of neutralising potency; other batches of the same anti venom deteriorate within the expiry period. Even at ambient temper P",i" ",,"" ,jO,' P"",;;; ..,,,, ,," atures, most liquid anti venoms maintain clitrity and I activity up to and over the expiry date printed on Fe the label. Facilities for storage of anti venoms are inadequate in many developing countries, both Fig. 1. Schematic representation of the IgG molecule showing F(ab')2' Fab and Fc fragments. Sites of papain and pepsin di during transport and at the final destination (hospi gestion are also indicated (arrows). tal or dispensary). In the rural tropics, it is possibly preferable to maintain such anti venoms at ambient temperatures rather than to store them in a refrig Antivenoms for use against the venoms of ani erator with fluctuating electricity supply and con mals other than snakes (e.g. medically important sequent fluctuating temperature. r I] scorpions, spiders, ticks, jellyfish and fish) are also Although lyophilised preparations are more ex produced by a few companies throughout the world pensive to produce than liquid anti venoms, they are 121 using the same basic system as described above. generally more satisfactory due to the longer shelf There are some exceptions to this, one being an life at ambient tropical temperatures in areas where anti venom produced by CSL Ltd. (Australia) cold storage facilities are unreliable or unavailable. against the venom of the funnel web spider (Atrax robustus). This is a protein A-purified intact rabbit IgG.[2 1 It has been reported that over 100 patients 3. Labelling of Antivenoms have received this anti venom with no recorded ad verse reactions.112! Simple labelling requirements have been laid down by the WHO Expert Committee on Biologi cal Standardisation.IIS ] The label should include in 2. Storage of Antivenoms formation on: • potency estimation using a standardised recog- Although there are few published data on the nised method (not yet satisfactory) stability of anti venoms, it is well established that • the nature of the preparation liquid antivenoms stored at different temperatures suffer different rates of activity loss.II.13,14J As a • the method of reconstitution general rule, the higher the ambient temperature, • restrictions, if any, on its use in a particular the more rapid the loss of neutralising potency; it country or area has been concluded that the potency of an anti • the identity of each reference venom against venom held in the liquid state is maintained for at which the potency is expressed least I year when stored at 30°C,113] although it has • a list of snake species for which cross-protection also been reported that antivenom stored at 37°C may be expected.ll ] did lose potency at the rate of25% per year and that The instructions contained in the package leanet opacity precedes loss of activity)13! There are ex should be clear and printed in letters large enough ceptions to this rule: liquid anti venoms may be to be read under restricted light conditions, and the highly variable both from batch to batch and from ampoule should be screen-printed as paper labels manufacturer to manufacturer. For example, the are susceptible to deterioration in humid condi-

tions)I) Other details of leat1et requirements are 5. Recent Developments given by the WHO)I) in Immunotherapy

4. Different Types of Antivenoms 5.1 Choice of Host Animal

If a single venom is used for immunisation, the Although in the past the horse has been the an antiserum is termed monospecific (monovalent); if imal of choice for anti venom production because more than one venom is used the antiserum is poly its use permits the collection of large volumes of specific (polyvalent). It was often assumed that blood, more recently the use of more immunolog polyspecific antisera should be produced by im ically suitable, practical and cheaper animals for munising animals with a mixture of venoms rather venom immunisation has been investigated. than by mixing monospecific anti venoms, as the Goats,f l?-19) rabbits,120) dogsf2 ] and even hens121.22) latter method was thought to result in dilution of have been used. Avian venom antibodies (IgY) each component)?] This, however, may not be true have been purified from the yolks of eggs from in all cases, as a competitive effect or antigen hens immunised with venom coupled to an affinity induced suppression may occur which could result resin,l22) An advantage of Ig Y is that it does not fix in an alteration or decrease in the response against mammalian complement,122) thus reducing the risk one or more of the many antigens involved.l l6 ) It is of adverse effects. currently thought that the mixing of monospecific One of the main problems with horses is the antisera, each produced in separate animals, max high cost of maintenance; a second, and probably imises the number of available antibody-producing more important, problem is that, in order to ensure cells and, in theory, should produce a wider range that sufficient anti venom is given, excessively of specific antibodies. This can obviously result in large amounts of horse serum proteins are often a dilution effect, but since many venom antigens given to the victim of a snake bite. This, in con are highly cross-reactive, thc int1uence of such ef junction with the fact that many people are sensi fects will be reduced. 161 tive to horse serum proteins, often results in pa Generally, clinicians prefer to use polyspecific tients exhibiting severe adverse effects typified by anti venoms, prepared by immunising the animal both early and delayed sensitivity reactions.l 231 against a range of venoms present within a geo This sensitivity is thought to be due, in part, to the graphical area, as this avoids the necessity for de well-documented124) immunogenicity of equine ciding which preparation to use for patient treat immunoglobulin containing a high proportion of ment; frequently the clinical signs of envenoming the IgG(T) subclass, which produces a much do not provide a clear indication as to which spe stronger immune response than that produced by cies is responsible for the bite. However, the ad either pig or sheep immunoglobulin.f25] vantages of using a monospecific preparation (pre Sheep are easy and cheap to maintain and sheep pared by immunising the animal against a single immunoglobulin is not associated with any of the venom) are that: (i) it is usually more effective, in immunogenicity of equine IgG(T). Although that a lower volume is required for treatment, thus horses are still the predominant host animal, it is reducing the problems of anaphylactoid reactions; not surprising that sheep are now being used more and (ii) it is useful in areas where a bite from a frequently for the preparation of immune sera.l 6.261 single species of snake causes readily identifiable Sheep, unlike horses, also respond well to venom clinical symptoms, for example failure of blood administered in an adjuvant preparation such as clotting following bites by the carpet viper, Echis Freund's Complete, and do not demonstrate the se ocellatus, in West Africa. vere local reactions manifested in horses)6)

5.2 Affinity Purification ing both pre- and postsynaptic neurotoxins when preincubated and injected into experimental mice, If the total immunoglobulin fraction is adminis clinically it does not work well against presynaptic tered, the high proportion of non-effective mole neurotoxins and therc is some doubt that it works cules simply increases the burden of foreign pro at all once the toxin is bound to its site at the neu tein administered to the envenomed patient and the romuscular junction. This was demonstrated in pa possible risk of adverse effects. Sheep hyper tients bitten by the Papuan taipan (Oxyuranus immune antisera typically contain 25 to 30 grams scutellatus canni) in Papua Ncw Guinea, in which of immunoglobulin per litre, of which only 25 to taipoxin, the main presynaptic neurotoxin, was not 33% is antigen-specific. 1261 Affinity chromatogra effectively neutralised by either the CSL poly phy purification techniques have been applied to specific or monospecific antivenoms. 1281 The neu anti venom research in recent years in order to over rotoxins are themselves distributed very rapidly come this, since they enable the isolation of anti through the tissues and the intravenous administra bodies specific only to the venom antigen. tion of Fab to the patient may prove more cffica The successful production of an ovine affinity cious in treating these patients. purified Fab-based anti venom combines all the ad Envenoming by many snake and some spidcr vantageous features mentioned. The treatment em species is oftcn associated with severe necrosis at ploys specific Fab fragments which have been the site of the bite, an effect as yet unalleviated by purified using agarose gels as a support matrix. Af commercially available antivcnoms. Thc more finity purification can result in a 4-fold increase in rapid effects of Fab may therefore be beneficial in specific activity when compared with the original this type of therapy where rapid treatment is vital. source serum, and a new anti venom prepared in It may also be feasiblc to administer Fab anti this way was shown to have a 9- to LO-fold supe venoms locally to alleviate severe local venom ef riority over current commercially available anti fects and to provide rapid anti venom treatment be venoms in protecting mice from the lethal effects fore hospitalisation. of the venom. 1271 The removal of the Fc region and the cleavage of F(ab'h into 2 Fab fragments by papain digestion 5.3 Use of Fab Fragments (fig. I) has the added beneficial effect of reducing the molecular weight of the therapeutic agent, thus A recent development in snake venom immuno improving its ability to diffuse through membrane therapy is the use of Fab fragments (fig. I) pro barriers. Also, since Fab fragments are only one duced by digesting JgG with the plant enzyme pa third the size of intact IgG (approximately 50kD pain. 161 Fab fragments have a number of advantages compared with 150kD), then theoretically less Fab over F(ab')z. They are not associated with type III based anti venom need be given. One problem with hypersensitivity since, with only a single antigen Fab, however, could be its more rapid c1carance via binding site, they do not form immune complexes the kidneys, thus preventing the maintenance of of sufficient size. They also have both greater vol adequate serum anti venom concentrations for neu umes of distribution and faster kinetics, both of tralising venom entering the system late from a which properties are appropriate to therapeutic venom depot at the site of the bite, or venom reen agents. tering the circulation from tissues. 1291 Rapid distribution is of particular importance in the treatment of snake envenoming, because of the 5.4 Future Developments in Immunotherapy low molecular weights of certain venom compo nents such as the neurotoxins which occur predom Possibilities for the future include the use of inantly in elapid venoms. It is well established that even smaller fragments of the IgG molecule (e.g. although F(ab'h anti venom is efficient in neutralis- Fv fragments),I:lOI the use of liposomal immunisa-

tion,16] and the use of biochemically isolated, re ment of early reactions,l34] It is possible that aggre combinant and modified venom toxins and mono gates of immunoglobulins or their fragments may clonal antibodies in anti venom production.l6•31 ] be either partly or wholly responsible for early re The latter may be useful in the case of venoms actions. Many patients receiving even so-called re containing I major toxin (e.g. the neurotoxins of fined F(ab')z antivenom develop either early (ana some elapid venoms), where an avid monoclonal phylactoid) or delayed serum sickness type antibody may be of use. However, the use of reactions, or both. One of the reasons for this is that monoclonal antibodies in neutralising viperine anti venoms do not always conform to their official venoms, which are multicomponent in nature, does description. For example, F(ab')z preparations are not at present appear to be practicaJ.f31 1 The use of frequently contaminated by aggregates, by Fc alternatives to conventional antivenoms, such as components and even by intact IgG molecules. synthetic peptide inhibitors directed against the ac Any measures, such as the use of more refined IgG tive sites of important venom toxins, and the use derived fragments (for example Fab or Fv) and im of certain plant extracts should also not be over proved purification procedures, that reduce the in looked. cidence of reactions should be encouraged, although there are indications that purification pro 6. Antivenom Reactions cedures do reduce the neutralising potency of anti venoms.[38] Although it has been stated in the past that, fol Results of sensitivity testing have proved that lowing purification of anti venoms by ammonium such tests have no predictive value for the occur sulphate precipitation and enzyme digestion, the rence of early reactions, including severe anaphy incidence of anaphylactic reactions is reduced to laxis, and they therefore appear to be a waste of about 10%,[321 more recent studies suggest that rates far in excess of this (30 to 84%) may be ob time. Reactions to homologous immunoglobulins tained.l33] Malasit et al,f34] stated that the reported resemble reactions to anti venom in that they are incidence of early, potentially life-threatening, re not predicted by the results of skin tests,l34] actions ranges from 3 to 54%, the incidence being Anaphylactoid reactions are overcome in the highest when larger doses of unrefined anti venom normal way by stopping anti venom administration are given; the incidence also appears to be related as soon as they occur, injecting 0.5ml of 1 : 1000 to the speed at which the anti venom is adminis adrenaline (epinephrine) subcutaneously and cau tered. 1351 tiously restarting the anti venom with antihistamine Sutherland[36 1 reported that most commercial (H I antagonist) cover. More recent studies per anti venoms are anticomplementary in vitro, and formed in Brazil have suggested that antihistamine his results led to the recommendation that anti given prophylactically does not decrease the level venom should be administered diluted and by slow of reactions.[ 39 1 However, in patients with severe intravenous infusion. However, more recently no systemic envenoming and a history of allergy, anti significant differences were found in the incidence venom should still be given with the added precau of early reactions when anti venom was adminis tions applied as described above. Although the use tered by slow 'push' injection or by slow intrave of prophylactic medications, such as subcutaneous nous infusion. 134] adrenaline administered before anti venom therapy Generally, complement activation[34.36.37] and commences, is not generally considered to be ac other mechanisms may be involved in addition to ceptable practice routinely, in Australia this proce immediate (type I) hypersensitivity to equine se dure is recommended.l401 In most parts of the rum, but the precise causes remain obscure. There world, pretreatment with adrenaline or antihista appears to be no apparent association between the mines is only recommended if the patient has a presence of immune complexes and the develop- history of allergy, has experienced previous anti-

venom reactions or has reacted earlier to other via the intradermal, subcutaneous or occasionally horse serum-based products. the intramuscular route. The currently approved WHO efficacy assay therefore comprises in vitro 7. Assessment of Antivenom Quality preincubation experiments (where venom and anti venom are mixed together), followed by an in vivo 7.1 Conventional Methods assessment phase of in vitro immunising activity)11 In the future, experimental assessment would be In 1979 a WHO coordination meeting was held more meaningful if the venom was administered in Zurich to examine the work in progress through subcutaneously and the anti venom, after a set time, out the world on the characterisation of medically by the intravenous route. The major disadvantage important venoms and the standardisation of anti of the latter system is, however, that because of the venom preparations.lll Following this meeting, the routes used for injection of venom and the time Liverpool School of Tropical Medicine was desig between venom injection and anti venom adminis nated as the WHO Collaborating Centre for the tration, the dose ranges are much wider, which Control of Antivenoms and given the responsibility means that many more animals have to be used in for organising the collection and characterisation efficacy assessment studies. of selected medically important (International Ref As different venoms possess a wide range of erence) venoms and the production and assessment different pharmacological properties, the anti of (International Standard) anti venoms against venom should ideally be titrated against each im these. A great deal of work and expense was in portant venom activity. Therefore, other simple an curred both by the CCCA and by the 13 interna cillary tests currentiy recommended by WHOIII are tional laboratories recruited to the study, which is those for the assessment of in vivo venom still in progressJI,411 The potency of any commer haemorrhagic, necrotising and defibrinogenating cial or laboratory antivenom should be determined activities and in vitro tests for venom coagulant and by reference to that of a stable, standard anti venom. procoagulant activities. I17,421 Further tests for eval These are currently held in the WHO laboratories uation of systemic neuromuscular paralytic and in Copenhagen. The fact that such reference and systemic myotoxic activity are under develop standard materials are available should greatly as ment.143 I Antivenoms are assessed by their ability sist collaboration among laboratories in different to neutralise these venom activities. Because ani countries. mal-based assays ofvenom/antivenom interactions Antivenom producers and independent asses are expensive, time-consuming and involve the use sors of anti venom potency, such as the WHO of many experimental animals, and because there CCCA, have to make some estimation of the neu often remains some uncertainty as to what activi tralising potency of products before clinical trials ties are being measured, it is important that further can be performed. The most important measure of efforts are also encouraged to develop other in vitro anti venom efficacy is its ability to save the life of biochemical and immunological1441 methods of as an experimental animal, and this is assessed by the say. EDso (median effective dose) test. Different doses of anti venom under test are mixed with, for exam 7.2 Clinical Trials ple, 5 intravenous LDsos (median lethal doses) of venom, incubated for 30 minutes, injected intrave The only truly meaningful method of assessing nously into mice of known weight, sex and strain, anti venoms is of course by randomised clinical and survi val after 24 hours is recorded) 17 ,421 The trial in human victims of envenoming. It is notable intravenous route of injection is normally used in that very few well-controlled randomised trials these assays; this is an artificial system, as when a have been performed,117.33,4S-48 1 and this reflects snake bites a human venom is normally introduced the logistic problems associated with organising

such trials in developing countries where snake fect of the anti venom on the resolution of local bite is a problem (for example, lack of basic facil effects such as swelling and necrosis should be as ities such as constant and stable electricity supply sessed, as should the adverse effects (for example 'or reliable water supply). early anaphylactic, pyrogenic and late serum sick It is obviously necessary to conduct compara ness type reactions). tive trials on a randomised, double-blind basis with The ideal standard dose of antivenom for treat a detailed protocol and clinical proforma available. ment is highly variable, as it depends on the type There are usually 2 main aims of such studies: and potency of the anti venom used and the severity (i) potency comparison of different antivenoms in of envenoming. In an ideal situation the amount of humans (potencies estimated from animal models antivenom provided in I vial or ampoule should be are frequently unreliable); and (ii) the calculation adequate for treating the most severe envenoming of the optimum initial dosage of an antivenom in encountered. relation to the defined severity of envenoming. Both involve accurate monitoring of antivenom 8. Conclusions treatment using both clinical[17,48] and labora tory[17,42] parameters. Methods of antivenom production have not al tered significantly since their conception almost a Ideally the species involved should be con century ago. One of the main reasons for this is firmed either by the taxonomic identification of the probably the conservatism that exists among anti snake or by positive identification of specific venom manufacturers, who have been unwilling in venom in the body fluids (blood, wound aspirate the past, possibly due to financial constraints, to or urine) of the victim by either enzyme or radio invest in new or improved existing methodology. immunoassay.[49-52] Patients who have already re The fact that antivenoms are sold to many devel ceived antivenom prior to admission, are very old oping countries who cannot afford high prices or very young, pregnant or already infirm should means that quality control probably suffers and the be excluded from the study. Informed consent by resulting product is therefore inefficient, with large patients or relatives for inclusion into the study volumes being required to reverse the symptoms should be obtained. of systemic envenoming. This in tum leads to a Methods used in such studies should include re high incidence of serum reactions. cording the history of the patient and results of Conversely, in many countries who produce physical examination on a standard proforma. This government-subsidised products of high neutralis should be divided into local signs (for example ing efficacy, excessively high amounts of anti fang marks, evidence of prehospital treatment such venom are frequently administered to patients, as incisions, extent of swelling, local bleeding, which again results in unacceptably high inci bruising, blistering, signs of infection and necro dences of early anaphylactic reactions. Patients sis) and systemic signs (for example evidence of may also be treated with anti venom even in the spontaneous systemic bleeding, shock, cardiac ar absence of signs of systemic envenoming, which rhythmias or neurotoxicity). obviously puts them unnecessarily at risk. Laboratory studies should include, among oth Another major problem is the frequent lack of ers, the assessment of the kinetics of venom and availability of antivenoms. For example, in Nigeria anti venom using immunoassay techniques[49,52,53] at present Pasteur and Behringwerke anti venoms to estimate the half-lives and clearances of both are virtually unobtainable unless the patients them venom and anti venom in the system, correction of selves, their families or their villages can afford to haemostatic and blood coagulation abnormalities pay prices massively inflated by the additional and reversal of neurological and myolytic effects, costs incurred in delivering the drugs to local hos depending on the type of snake involved. The ef- pitals and dispensaries in inaccessible regions;

many people are dying simply because they cannot 2. Theakston RDG. Warrell DA. Antivenoms: a list of hyperimmune sera currently available for the treatment of afford the anti venom or enough of it to have ade envenoming by bites and stings. Toxicon 1991; 29: 1419-70 quate therapeutic effect. 3. Latifi M. Commercial production of commercial anti-snakebite There is a real need to adapt modern technology serum (antivenin). In: Gans C. Gans KA. editors. Biology of the reptilia. Vo1.8. Physiology. London: Academic Press. to anti venom production in order to produce 1978:561 cheaper, more effective and less reactive therapeu 4. Christensen PA. South African snake venoms and anti venoms. Johannesburg: South African Institute for Medical Research. tic agents. There are indications that this may be 1955 happening, as demonstrated by the use of more 5. Sawai Y. Chinzei H. Kawamura Y. et al. Studies on the improve suitable animals for immunisation, the introduc ment of treatment of Habu (Trimeresurus j/avoviridis) bites. IX. Studies on the immunogenicity of the purified habu tion of affinity purification and the application of venom toxoid by alcohol precipitation. Jpn J Exp Med 1972; smaller JgG fragments such as Fab. A combined 42: 155-64 Fab/F(ab')z antivenom may also be useful, combin 6. Theakston RDG. Smith DC. Therapeutic antibodies to snake venoms. In: Landon J. Chard T. editors. Therapeutic antibod ing the long elimination half-life of the F(ab')z ies. London: Springer. 1995: 109-33 fragment with the improved volume of distribution 7. Christensen PA. The preparation and purification of anti venoms. Mem Inst Butantan 1966; 33: 245-50 of the Fab fragment. Other possibilities for the fu 8. Amies CR. The use of topically formed calcium alginate as a ture include research into methods of active im depot substance in active immunisation. J Pathol Bacteriol munisation of people in areas of high snake bite 1959; 77: 435-42 9. Christensen PA. Venom and antivenom potency estimation. incidence and mortality, and the study of possible Mems Inst Butantan 1966; 33: 305-26 alternatives to antivenom therapy. 10. Grasset E. Christensen PA. Enzyme purification of polyvalent antivenine. South and Equatorial African colubrine and viper The input of the WHO into this field is impor ine venoms. Trans R Soc Trop Med Hyg 1947; 41: 207-11 tant as, at present, there is no established inde II. Latifi M. Manhouri H. Antivenin production. Mem Inst pendent control of quality, although it should be Butantan 1966; 33: 893·7 12. Hartman LJ. Sutherland SK. Funnel-web spider (Atrax stressed that the development of such a system is robustus) anti venom in the treatment of human envenoma in progress.l431 Companies producing anti venoms tion. Med J Aust 1984; 141: 769-99 should be required to submit them for independent 13. Jerne NK. Perry WLM. The stability of biological standards. Bull World Health Organ 1956; 14: 167-82 assessment, preferably under the auspices of the 14. Christensen PA. The stability of refined anti venom. Toxicon WHO. For this, both the active and enthusiastic 1975; 13: 75-7 interest of the WHO in the programme as well as 15. World Health Organization. 23rd Report of the WHO Expert Committee on Biological Standardization. Geneva: World the provision of some financial support is vital. Health Organization, 1971. WHO Technical Report Series No. 463 16. Adler FL. Antibody formation after injection of heterologous Acknowledgements immunoglobulin. II. Competition of antigens. J Immunol 1957; 78: 201-10 We are grateful to all our colleagues in the Venom Re 17. Warrell DA, Looareesuwan S, Theakston RDG, et al. search Unit, Liverpool School of Tropical Medicine, and at Randomised comparative trial of three monospecific anti venoms for bites by the Malayan pit viper (Cal/oselasma Therapeutic Antibodies, London, without whom it would rllOdostoma) in southern Thailand: clinical and laboratory have been impossible to write this article. We would like to correlations. Am J Trop Med Hyg 1986; 35: 1235-47 thank Professor J. Landon, Department of Clinical Pathol 18. Russell FE, Timmerman WF. Meadows P. Clinical use of anti ogy, St. Bartholomew's Hospital, London, and Dr Gavin venin prepared from goat serum. Toxicon 1970; 8: 63-5 Laing and Ms Alison Richards, Venom Research Unit, Liv 19. Kochalty WF, Bowles-Ledford E, Daly JG, et al. Preparation of erpool School of Tropical Medicine, for advice and help. We coral snake anti venom from goat serum. Toxicon 1971; 9: are especially grateful to Professor D.A. Warrell, Centre for 297-8 Tropical Medicine, University of Oxford, for his continuing 20. Russell FE. Use of Crotalus monovalent anti venom from rabbit serum. Curr Ther Res 1961; 3: 438-40 support over many years of collaborative effor!. 21. Thalley BS. Carroll SB. Rattlesnake and scorpion anti venoms from the egg yolks of immunized hens. Biotechnology 1990; 8: 934-8 References 22. Carroll SB. Thalley BS, Theakston RDG. et al. A comparison I. World Health Organization. Progress in the characterization of of the purity and efficacy of affinity purified avian antivenoms venoms and standardization of anti venoms. WHO Offset Pub with commercial equine crotalid anti venoms. Toxicon 1992; lication No. 58. Geneva: World Health Organization. 1981 30: 1017-25

23. Russell FE, Sullivan JB, Egan NB, et al. Preparation of a new Proceedings of the First International Congress on Enveno anti venom by affinity chromatography. Am J Trop Med Hyg mations and their Treatments; 1995 Jun 7-9. Paris: Institut 1985; 34: 141-50 Pasteur. 1985: 207 24. Weir RC, Porter RR. Comparison of the structure of the immu 40. Tibballs J. Premedication for snake anti venom. Med J Aust noglobulins from horse serum. Biochem J 1966; 100: 53-8 1994; 160: 4-7 25. Cartledge C, Mclean C, Landon J. Production of polyclonal 41. Theakston RDG. Reid HA. Development of simple standard antibodies in ascitic tluid of mice: time and dose relation assay procedures for the characterization of snake venoms. ships. J Immunoassay 1992; 13: 339-53 Bull World Health Organ 1983; 61: 949-56 26. Smith DC, Reddi KR, Laing G, et al. An affinity purified ovine 42. Laing GD. Lee L, Smith DC. et al. Experimental assessment of anti venom for the treatment of Vipera benls envenoming. a new, low-cost anti venom for treatment of carpet viper Toxicon 1992; 30: 865-71 (Echis ocellatlls) envenoming. Toxicon 1995; 33: 3(l7-313 27. AI-Joufi A, Bailey GS, Reddi K, et al. Neutralisation of kinin 43. Theakston RDG. Characterization of venoms and stand releasing enzymes from viperid venoms by anti venom IgG ardization of anti venoms. In: Harris JB, editor. Natural toxins, fragments. Toxicon 1991; 29: 1509-12 animal, plant and microbial. Oxford: Clarendon Press, 1986: 28. Lalloo DA, Trevett AJ, Korinhona A, et al. Snake bites by the 287-303 Papuan taipan (OXYllrllIlIlS sClitel/allIs canlli): paralysis, he 44. Theakston RDG. Reid HA. Enzyme-linked immunosorbent as mostatic and electrocardiographic abnormalities, and effects say (ELISA) in assessing anti venom potency. Toxicon 1979; of anti venom. Am J Trop Med Hyg 1995; 52: 525-31 17: 511-5 29. Choumet V, Audebert F, Riviere G, et al. Toxicokinetics of en 45. Warrell DA. Davidson NMcD, Ormerod LD. et al. Bites by the venomations and anti venom labstractJ.ln: Proceedings of the saw-scaled or carpet viper (Echis carinatus): trial of two spe First International Congress on Envenomations and their cific antivenoms. BMJ 1974; 4: 437-40 Treatments; 1995 Jun 7-9. Paris: Institut Pasteur, 1995: 71 46. Warrell DA, Warrell MJ, EdgarW, et al. Comparison of Pasteur 30. Landon J, Chard T, Coxon RE. An introduction to the clinical and Behringwerke anti venoms in envenoming by the carpet applications of antibodies. In: Landon J, Chard T, editors. viper (Echis carinalus). BMJ 1980; I: 607-9 Therapeutic antibodies. London: Springer, 1995: 1-32 47. Jorge MT, Cardoso JLC. Castro SCB. et al. A randomised 31. Theakston RDG. New techniques in anti venom production and active immunisation against snake venoms. Trans R Soc Trop blinded comparison of two doses of anti venom in the treat Med Hyg 1989; 83: 433-5 ment of Bothrops envenoming in Sao Paulo, Brazil. Trans R Soc Trop Med Hyg 1995; 89: 111-4 32. Reid HA. Symptomatology, pathology and treatment of land snake bite in India and Southeast Asia. In: Biicherl W, Buck 48. Meyer WP. Habib HG. Onayade AA. et al. First clinical expe ley EE, Deulofeu V, editors. Venomous animals and their ven riences with a new ovine Fab Echis ocel/atlls snakebite anti oms. Vol. I. Venomous vertebrates. New York: Academic venom in Nigeria: randomised comparative trial with Pasteur Press, 1968: 611-42 Ipser Africa anti venom. Am J Trop Med Hyg. In press 33. Cardoso JLC, Fan HW, Fran~a FOS, et al. Randomized com 49. Theakston RDG, Lloyd-Jones JM. Reid HA. Micro-ELISA for parative trial of three anti venoms in the treatment of en detecting and assaying snake venom and venom antibody. venoming by lance-headed viper (Bo/hm!,s jarl/rl/Cll) snakes Lancet 1977; II: 639-41 in Sao Paulo, Brazil. Q J Med 1993; 86: 315-25 50. Coulter AR. Cox JC. Sutherland SK. et al. A new solid-phase 34. Malasit P, Warrell DA, Chanthavanich P, et al. Prediction, pre sandwich radioimmunoassay and its application to the detec vention and mechanism of early (anaphylactic) anti venom tion of snake venom. J Immunol Methods 1978; 23: 242-52 reactions in victims of snake bites. BMJ 1986; 292: 17-20 51. Coulter AR. Harris RD, Sutherland SK. Enzyme immunoassay 35. Antivenom therapy and reactions leditoriall. Lancet 1980; I: for the rapid identification of snake venom. Med J Aust 1980; 1009-10 I: 433-5 36. Sutherland SK. Serum reactions: an analysis of commercial 52. Ho M, Warrell MJ. Warrell DA. et al. A critical reappraisal of anti venoms and the possible role of anticomplementary ac the usc of enzyme-linked immunosorbent assays in the study tivity in de novo reactions to antivenoms and antitoxins. Med of snake bite. Toxicon 1986; 24: 211-21 J Aust 1977; I: 613-5 53. Theakston RDG. Fan HW. Warrell DA. et al. Use of enzyme 37. Pugh RNH, Theakston RDG. Antivenom reactions and comple immunoassays to compare the effect and assess dosage re ment depletion in snake bite. Ann Trop Med Parasitol 1987; gimens of three Brazilian Bothmps anti venoms. Am J Trop 81: 73-5 Med Hyg 1992; 47: 593-604 38. Morais JI', De Frietas MCW, Yamaguchi IK. et al. Snake anti venoms from hyperimmunised horses: comparison of the anti venom activity and biological properties of their whole IgG and F(ab'h fragments. Toxicon 1994; 32: 725-34 39. Fan HW. Mendes LP. Fran~a FOS. et al. Lack of efficacy of Correspondence and reprints: Dr R. David G. Theakston, antihistamine in preventing early reactions to horse anti Venom Research Unit, Liverpool School of Tropical Medi venom administration: a retrospective study labstract]. In: cine, Pembroke Place, Liverpool L3 5QA, England.