Volume 166 Issue 10

Safety issues in herbal medicine: implications for the health professions Anna K Drew and Stephen P Myers The use of herbal medicines in Australia is widespread. A number of factors make assessment of adverse effects associated with these products more complex than for pharmaceuticals. Problems have resulted from contamination with heavy metals and adulteration with prescription drugs in overseas herbal products. A classification is proposed for adverse effects associated with herbal medicines, and medical practitioners are encouraged to include use of these preparations in a patient's drug history and in reports of suspected adverse drug reactions. It may be necessary to develop a separate database to promote adverse drug reaction reporting for herbal medicine and the wider field of complementary and alternative medicine. (MJA 1997; 166: 538-541) For editorial comment see Shenfield et al. Introduction - Regulation of herbal medicines in Australia - Safety of herbal medicines - Proposed classification of adverse effects of herbal medicines - Adverse drug reaction (ADR) reporting - Conclusion - References - Authors' details - ©MJA1997 Introduction The World Health Organization estimates that 65%-80% of the world's population use traditional medicine as their primary form of health care.1 The use of herbal medicine, the dominant form of medical treatment in developing countries, has been increasing in developed countries in recent years.2 Assessment of the safety and efficacy of these medicines is an important issue for the health professions. We focus here on the safety of these preparations; the issue of their efficacy is not addressed. A classification of potential adverse effects associated with these preparations is proposed, and we encourage the reporting of any adverse drug reactions (ADRs). Herbal medicine, in which plants (dried or in extract form) are used as therapeutic substances, is one of a number of practices encompassed by the term "complementary and alternative medicine" (CAM). Recent studies have highlighted the extent to which CAM is used in Australia. A 1993 survey of 3004 South Australians by MacLennan et al.3 found that, in the previous year, 48.5% had used at least one form of CAM preparation and 20.3% of all respondents had visited at least one alternative practitioner. Herbal medicine accounted for approximately 26% of CAM use in this survey. Estimates of the national cost of both CAM preparations and practitioner visits were about one billion dollars when extrapolated to the Australian population. Results for the use of CAM were similar in a survey of 325 patients attending a Sydney teaching hospital emergency department in 1994,4 and only 35.5% of users had informed their medical practitioner about any use of CAM. Of the women who had borne children, 12 (14.5%) had taken one to 18 herbal preparations during pregnancy, and eight of 34 (23.5%) patients under 16 had been given between one and eight herbal preparations. Evidence suggests that CAM preparations or therapies are used for conditions such as cancer, high blood pressure and allergies, as well as for general wellbeing.3,5,6 Regulation of herbal medicines in Australia In Australia, products for human medicinal use must be placed on the Register of Therapeutic Goods [Therapeutic Goods Act 1989 (Cwlth)] in one of two categories -- "listed" or registered. Formulations can be listed for a small fee if they contain substances regarded by the Therapeutic Goods Administration (TGA) as being of low public health concern and comply with the Therapeutic Goods Advertising Code. This restricts wording of claims to "assist" rather than "treat" and limits indications to minor self-limiting conditions. The products have to be manufactured by a TGA-licensed manufacturer following a recognised code of Good Manufacturing Practice. Labelling requirements are the same as those for registered products. Efficacy data have to be held by the manufacturer/distributor of such products and can be called on at any time by the TGA or the Australian Competition and Consumer (formerly, the Trade Practices) Commission. About 4500 plant-based products are listed; these are given an "AUST L" number, indicating their listing on the register and that they can be sold legally in Australia. Registered products, which bear an "AUST R" number, contain herbs that are either restricted by the federal Standards for the Uniform Scheduling of Drugs and Poisons, those for which efficacy claims are more substantial, or those which are specified by the TGA as being of some health concern. For registration, which is more costly, appropriate documentation outlining clinical trial work must be submitted to the Traditional Medicines Evaluation Committee (established in 1991 -- soon to be replaced by the Complementary Medicines Evaluation Committee) which advises the TGA. Fewer than five CAM products have been evaluated in this way. Although Australia has more regulatory controls than many other countries for CAM preparations, including herbals,7 most of these preparations are not exposed to the premarketing evaluation process that prescription and scheduled proprietary medicines undergo. Few CAM preparations can be patented, so they are not subject to the financial incentive that drives the pharmaceutical market. Safety of herbal medicines Although it is widely perceived that "natural" products are safe, the evidence suggests that CAM use is not without risk. Of 90 patients with rheumatoid arthritis, 82% had tried more than one form of alternative medicine or therapy, including dietary modification, and 31% of these patients had experienced at least one adverse effect.8 Of 1701 consecutive patients admitted to the Prince of Wales Hospital, Hong Kong, three (0.2%) had had adverse effects attributed to traditional Chinese medicines and 75 (4.4%) to "Western" medications.9 A review of 5563 enquiries received by the National Poisons Unit, London, showed that 77.7% involved vitamin preparations and 19.3%, herbal extracts, royal jelly, hormonal products and other natural products. Exposure was linked to adverse effects in 49 (0.9%) of these cases.10 In ascertaining whether a substance is associated with an adverse effect, the medical literature may be of limited help -- there may be no previous report of such an event, as was the case for fatal anaphylaxis which occurred in an 11-year-old child with asthma after her third exposure to royal jelly.11 Prior to this event, contact dermatitis had been documented with royal jelly (which contains proteins, carbohydrates, amino acids, vitamins, lipids and fatty acids), but the allergen had not been identified.12 At the time of the child's death, the Adverse Drug Reactions Advisory Committee (ADRAC) of the Commonwealth Department of Health and Family Services had three reports of adverse reactions to royal jelly on file: one of anaphylaxis and two of bronchospasm (Dr Ian Boyd, Adverse Drug Reactions Advisory Committee, Canberra; data, 1972-May 1993; personal communication). Raised awareness of this problem resulted in the TGA advising manufacturers to label royal jelly products to warn of their potential to cause severe allergic reactions in people who suffer from asthma or allergies.13 ADRAC have now received a total of 18 reports of allergic reactions to royal jelly, including two fatalities (Dr Ian Boyd, Adverse Drug Reactions Advisory Committee, Canberra; data, 1972-February 1997; personal communication). Proposed classification of adverse effects of herbal medicines Adverse effects of herbal medications may be intrinsic or extrinsic (Box 1). The patient's age, genetic constitution, nutritional state, concomitant diseases and concurrent medication may affect the risk and severity of adverse events, as can consumption of large amounts or a wide variety of herbal preparations, or long-term use.15,16 Intrinsic effects Intrinsic effects are those of the herb itself and are characterised, as for pharmaceuticals, as type A (predictable, dose-dependent) and type B (unpredictable, idiosyncratic) reactions.17 Yohimbine, an alkaloid found in Pausinystalia yohimbe bark that has α2 -adrenoceptor antagonist activity, is taken for male impotence, and can cause hypertension and anxiety in a predictable, dose-related manner (type A reaction); it has also been associated with the serious idiosyncratic reactions of bronchospasm and increased mucus production when taken in normal doses by a patient with severe allergic dermatitis (type B).18,19 Type A reactions with herbal preparations also include effects with deliberate overdose or accidental poisoning and interactions with pharmaceuticals. Extrinsic effects Extrinsic effects are not related to the herb itself, but to a problem in commercial manufacture or extemporaneous compounding. Potential failures to adhere to a code of Good Manufacturing Practice, while not specific to herbal medicine, can occur, particularly in developing countries where such a code is not in place. This makes it more difficult for medical practitioners and other health professionals to assess the adverse effects of herbal preparations compared with pharmaceuticals. Misidentification: It is difficult to track and identify adverse effects of herbal ingredients, as the plants can be named in four different ways -- the common English name, the transliterated name, the latinised pharmaceutical name, and the scientific name.20 It is essential that plants are referred to by their binomial Latin names for genus and species; misidentification can occur when other names are used. For example, the scientific name of the Chinese herb that is variously transliterated as "dong quai", "dong guai", "danggui" and "tang kuei" is Angelica polymorpha (formerly sinensis). The common English name "angelica" and the latinised name "Radix Angelica" could refer either to this species, which is used in Australia, or to the European species Angelica archangelica, depending on the country of origin. Misidentification can result in erroneous associations being made, with potential clinical implications. Plant material can be misidentified at the time of the manufacturer's bulk purchase or when wild plants are picked. Lack of standardisation: The therapeutic/toxic components of plants vary depending on the part of the plant used, stage of ripeness, geographic area where the plant is grown, and storage conditions. Therefore, batch-to-batch reproducibility of plant material should be assessed in the production of marketed products, but, in practice, product variation in herbal medicines can be significant. The content of ginsenoside, the glycosylated steroid to which most of the biological activity of ginseng (Panax ginseng) has been ascribed, was examined in 50 commercial brands of ginseng sold in 11 countries.21 In 44 of these products, the concentration of ginsenoside ranged from 1.9% to 9% w/w; six products contained no ginsenoside, and one of these six contained large amounts of ephedrine (for which a Swedish athlete was accused of doping). Contamination: During growth and storage, crude plant material can become contaminated by pesticide residues, microorganisms, aflatoxins, radioactive substances and heavy metals;22 lead, cadmium, mercury, arsenic and thallium have been reported as contaminants of some overseas herbal preparations.23-25 In a case series of five patients in the United Kingdom with lead poisoning from Asian traditional remedies, the preparations implicated contained 6%-60% w/w lead by weight.26 The Australian Code of Good Manufacturing Practice specifies detection of microorganisms and leaves estimation of other contaminants (not specified in internationally recognised pharmacopoeial standards) to the discretion of manufacturers.14 Substitution: A report of nine cases of rapidly progressive interstitial nephritis in young women taking a Belgian slimming treatment27 led to the discovery that Aristolochia fangchi, containing the nephrotoxic component aristolochic acid, had been introduced in place of Stephania tetrandra.20 Eighty cases have now been identified and more than half of these patients developed terminal renal failure.28,29 Adulteration: The intentional use of pharmaceutical adulterants has been reported. Cases of acute interstitial nephritis, reversible renal failure, loss of blood pressure control and peptic ulceration have been reported with a product called "Tung Shueh" pills, taken for arthritic complaints.30-32 The product contained mefenamic acid and diazepam, neither of which was included on the label. Adulterants can also be added by unethical herbalists compounding preparations for individual patients. In a recent Victorian court case, a Chinese herbalist was prosecuted for adding a steroid cream to a herbal preparation, which produced severe facial erythema in a patient. 33 Incorrect preparation/dosage: The processing of crude plant material carried out by a manufacturer, CAM practitioner or the patient is a major determinant of the pharmacological activity of the finished product. A Western Australian patient had a heart attack when he failed to follow a herbalist's instructions to boil aconite (a restricted plant in Australia) in three pints of water for one hour and take the decanted liquid; the patient increased the dose and shortened the boiling time. 34 Boiling changes the alkaloid composition, rapidly reducing the plant's toxicity, 35 and can substantially reduce microorganism contamination. 36 Another point to consider is that the activity of crude plant material may differ from that of the purified constituents, as some constituents may modify the toxicity of others. 35 Inappropriate labelling/advertising: In early 1996, a direct-mailing campaign to individuals who had purchased exercise bicycles included information on seaweed (Fucus vesiculosus) patches for weight loss. Seaweed, or kelp, contains iodine, and it was claimed that the patches would reverse hypothyroidism by releasing iodine into the body, speeding up the body's metabolism, resulting in weight loss. This claim was unproven. Hyperthyroidism has been reported in people who take kelp products orally,36 and if iodine were to be absorbed transdermally it could lead to hyperthyroidism in susceptible individuals. The TGA became aware of the product promptly and secured a promise that no further supplies would be imported, but keeping abreast of potentially unsafe products is a mammoth task. Adverse drug reaction (ADR) reporting ADR reporting is as essential for CAM products as it is for pharmaceuticals in providing postmarketing surveillance. In Australia, reporting of adverse effects of any medication, whether alternative or conventional, is usually undertaken by a medical practitioner, pharmacist or dentist, who completes and forwards a "blue card" to ADRAC. Although the person reporting need not assess the association between the medication and the adverse effect, this process enables trends to be spotted. ADRAC has received 154 reports relating to CAM in 25 years (Dr Ian Boyd, Adverse Drug Reactions Advisory Committee, Canberra; data, 1972-February 1997; personal communication). Given the widespread use of CAM, this low number of reports suggests that CAM has either a low risk of adverse effects or that such effects are significantly under-reported. Although limited evidence suggests that CAM products may be associated with a lower risk than conventional medicines, 9 under-reporting is likely, as: ADRAC does not actively encourage the reporting of adverse effects by practitioners and consumers of alternative medicine; CAM use is not routinely included in patients' drug histories or in reports of adverse effects; The public perception that "natural" products are safe biases against an association being made between CAM products and adverse effects. Conclusion The incidence of adverse effects of CAM products requires further study, and more education about CAM is needed. Medical practitioners should be encouraged to routinely ask for information about CAM use when they take a drug history and to include CAM products in ADR reports (Box 2, below). It is also important to promote an avenue for alternative practitioners and consumers to report adverse effects to CAM products, as a large proportion of alternative medicines are sold through health food outlets, supermarkets and by direct marketing (including via the Internet). Development of a separate or parallel database could fulfil this purpose. The Government response to recommendations arising from the recent TGA review accepts the need to extend the coverage of ADRAC to complementary and alternative medicines. 37,38 As there is increasing pressure to regulate CAM products to pharmaceutical industry standards of quality and safety, 7 the challenge for the Government and the CAM industry is to provide a level of postmarketing surveillance at least equivalent to that in place for pharmaceuticals. References Anonymous. Program Profile: international liaison brings global vision to OAM. Complementary and Alternative Medicine at the NIH 1996; 3: 3 . British Medical Association. Complementary medicine. New approaches to good practice. Oxford: Oxford University Press, 1993: 9-36. MacLennan AH, Wilson DH, Taylor AW. Prevalence and cost of alternative medicine in Australia. Lancet 1996; 347: 569-572. Kristofferson SS, Atkin PA, Shenfield GM. Uptake of alternative medicine [letter]. Lancet 1996; 347: 972. Sawyer MG, Gannoni AF, Toogood IR, et al. The use of alternative therapies by children with cancer. Med J Aust 1994; 160: 320-322. Eisenberg DM, Ronald CK, Foster C, et al. Unconventional medicine in the United States; prevalence, costs, and patterns of use. New Engl J Med 1993; 328: 246-252. De Smet PAGM. Should herbal medicine-like products be licensed as medicines [editorial]? BMJ 1995; 310: 1023-1024. Kestin M, Miller L, Littlejohn G, Wahlqvist M. The use of unproven remedies for rheumatoid arthritis in Australia. Med J Aust 1985; 143: 516-518. Chan TYK, Chan AYW, Critchley JAJH. Hospital admissions due to adverse reactions to Chinese herbal medicines. J Trop Med Hyg 1992; 95: 296-298. Perharic L, Shaw D, Colbridge M, et al. Toxicological problems resulting from exposure to traditional remedies and food supplements. Drug Saf 1994; 11: 284-294. Bullock RJ, Rohan A, Straatmans, J-A. Fatal royal jelly-induced asthma [letter]. Med J Aust 1994; 160: 44. Takahashi M, Matsuo I, Ohkido M. Contact dermatitis due to honeybee royal jelly. Contact Dermatitis 1983; 9: 452-455. Anonymous. Royal jelly: warning label required. TGA News 1994; 16: 4. Bensoussan A, Myers SP. Towards a safer choice. The practice of traditional Chinese medicine in Australia. Sydney: Faculty of Health, University of Western Sydney (Macarthur), 1996: 54. De Smet PAGM. Health risks of herbal remedies. Drug Saf 1995; 13: 81-93. Huxtable RJ. The harmful potential of herbal and other plant products. Drug Saf 1990; 5 (Suppl 1): S126-S136. Rawlins MD, Thompson JW. Pathogenesis of adverse drug reactions. In: Davies DM, editor. Textbook of adverse drug reactions. Oxford: Oxford University Press, 1977: 44. Landis E, Shore E. Yohimbine-induced bronchospasm. Chest 1989; 96: 1424. De Smet PAGM, Smeets OSNM. Potential risk of health food products containing yohimbe extracts. BMJ 1994; 309: 958. But P P-H. Need the correct identification of herbs in herbal poisoning [letter]. Lancet 1993; 341: 637. Cui J, Garle M, Eneroth P, Bjorkhem I. What do commercial ginseng preparations contain? [letter] Lancet 1994; 344: 134. Bisset NG, editor. Herbal drugs and phytopharmaceuticals. Stuttgart: Medpharm Scientific Publishers, 1994. Schaumburg HH, Berger A. Alopecia and sensory polyneuropathy from thallium in a Chinese herbal medication [letter]. JAMA 1992; 268: 3430-3431. Kew C, Morris C, Aihie A, et al. Arsenic and mercury intoxication due to Indian ethnic remedies. BMJ 1993; 306: 506-507. Wu M-S, Hong J-J, Lin J-L, et al. Multiple tubular dysfunction induced by mixed Chinese herbal medicines containing cadmium. Nephrol Dial Transplant 1996; 11: 867-870. Bayly GR, Braithwaite RA, Sheehan TMT, et al. Lead poisoning from Asian traditional remedies in the West Midlands -- report of a series of five cases. Hum Exp Toxicol 1995; 14: 24-28. Vanherweghem J-L, Depierreux M, Tielemans C, et al. Rapidly progressive interstitial renal fibrosis in young women: association with slimming regimen including Chinese herbs. Lancet 1993; 341: 387-391. Vanhaelen M, Vanhaelen-Fastre R, But P, Vanherweghem J-L. Identification of aristolochic acid in Chinese herbs [letter]. Lancet 1994; 343: 174. van Ypersele de Strihou C, Vanherweghem JL. The tragic paradigm of Chinese herbs nephropathy [editorial]. Nephrol Dial Transplant 1995; 10: 157-160. Diamond JR, Pallone TL. Acute interstitial nephritis following use of tung shueh pills. Am J Kidney Dis 1994; 24: 219-221. Abt AB, Oh JY, Huntington RA, Burkhart KK. Chinese herbal medicine induced acute renal failure. Arch Intern Med 1995; 155: 211-212. Gertner E, Marshall PS, Dean Filandrinos, et al. Complications resulting from the use of Chinese herbal medications containing undeclared prescription drugs. Arthritis Rheum 1995; 38: 614-617. Adams D. Herbal tea made woman's face "stop light", court told. The Age (Melbourne ) 12 December 1995; 7. Dean A. Herbalist's patient had heart attack. The Sydney Morning Herald 21 July 1994; 4. De Smet PAGM, Tognoni G. Drugs used in non-orthodox medicine. In: Dukes MNG, editor. Meyler's side effects of drugs. 12th edition. Amsterdam: Elsevier, 1992: 1209-1232. De Smet PAGM. Toxicological outlook on the quality assurance of herbal remedies. In: De Smet PAGM, Keller K, Hansel R, editors. Adverse effects of herbal drugs, I. Heidelberg: Springer-Verlag, 1992: 1-72. Therapeutic Goods Administration. Government response to recommendations arising from the Therapeutic Goods Administration Review. Canberra: AGPS, April 1997: 37-38. KPMG Management Consulting. Review of Therapeutic Goods Administration on behalf of the Department of Health and Family Services. Canberra: AGPS, January, 1997: 126-127. (Received 11 September 1996; accepted 18 February 1997) Authors' details Hunter Drug Information Service, Department of Clinical Toxicology and Pharmacology, Newcastle Mater Misericordiae Hospital, Newcastle, NSW. Anna K Drew, BPharm(Hons), Director. School of Natural and Complementary Medicine, Southern Cross University, Lismore, NSW. Stephen P Myers, BMed, ND, Head. Medical Doctoral Student, Discipline of Clinical Pharmacology, Faculty of Medicine and Health Sciences, University of Newcastle. No reprints will be available. Correspondence: Anna Drew, Director: Hunter Drug Information Service, Locked Bag 7, Hunter Region Mail Centre, NSW 2310. E-mail: oudanATcc.newcastle.edu.au Make a comment - - To top of article - ©MJA 1997 <URL: http://www.mja.com.au/> © 1997 Medical Journal of Australia. We appreciate your comments.

Anna K Drew · Stephen P Myers

The St Marys fragmentation grenade explosion Antony Nocera The accidental explosion of a fragmentation grenade in a munitions factory at St Marys injured four workers, two critically. The prompt response by ambulances and physician-staffed helicopter emergency medical service prevented deaths, but the incident suggests lessons for the future handling of urban explosions. (MJA 1997; 166: 545-548) For editorial comment see Fulde Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au/>". Introduction - Patient records - What makes a grenade - Discussion - Acknowledgement - References - Authors' details Make a comment - Register to be notified of new articles by e-mail - Current contents list - ©MJA1997 Introduction On 14 November 1995 an F1 fragmentation hand grenade exploded in a grenade testing facility in a munitions factory on a 1600 hectare site at St Marys, NSW. Seven employees were in the room. The grenade exploded at 8:36 am on a work bench and injured four workers around the bench (Figure 1). None were wearing body armour and there were no blast shields in the work area. Figure 1: Immediate area of the explosion. Patient C was sitting in the chair in the foreground, with Patient A in the chair immediately behind, while Patient B was standing behind the two. Patient D was sitting on the other side of the bench, opposite Patient C. The pallet to the left carried 1000 live grenades and was 1.85 m from the grenade burst point. Forty-eight grenades were damaged in the explosion and represented an explosive hazard during recovery operations. Photograph courtesy of the New South Wales Police. The first ambulances arrived at 8:41, 8:44 and 8:47. The four injured workers were extricated from the factory building to an adjacent courtyard by other employees and ambulance personnel. NRMA CareFlight, a Bell 412 HP helicopter emergency medical service (HEMS) with a physician on board, was placed on standby at 8:45 and dispatched at 8:50, landing at the site at 9:06. Two patients were transported by air to a trauma centre with full cardiothoracic facilities about 24 km (10 minutes' flying time) from the factory. Access to this centre by road would have encountered peak hour traffic flowing into Sydney via the major western arterial road routes. Two patients were transported by road to a local trauma centre (10 minutes' travelling time) which has limited cardiothoracic surgical facilities. Patient records Patient A Patient A was a 44-year-old woman sitting 0.25 m from the grenade burst point. On arrival of the HEMS, she had a Glasgow coma score of 12-13 and no recordable blood pressure. Only carotid pulses were palpable after an infusion of 1.5 L of polygeline by ambulance personnel; an additional 500 mL polygeline and 500 mL normal saline were infused. The patient's conscious state began to fluctuate, and a rapid sequence induction and endotracheal intubation were performed before transport by helicopter. The patient received a further two units of O negative blood during transport. On arrival at hospital, Patient A had a heart rate of 100/min, systolic blood pressure, 118 mmHg; haemoglobin level, 79 g/L (normal range, 115-165 g/L); pH, 7.29 (normal range, 7.35-7.45); PO2, 518 mmHg (normal range, 75-100 mmHg); PCO2, 30 mmHg (normal range, 35-45 mmHg); HCO3, 14 mmol/L (normal range, 22-26 mmol/L); and serum lactate, 3.04 mmol/L (normal range, 0.63-2.44 mmol/L). She was taken to theatre for thoracotomy and laparotomy, enucleation of left orbit, amputation of right ring and little fingers, debridement of right forearm and thigh, with internal fixation of her right forearm fractures. By the end of her first theatre session she had received (in addition to her prehospital fluids) 6.5 L of polygeline, 4 L of crystalloid, 24 units of packed red blood cells, 8 units of frozen plasma and 6 units of platelets. Patient A's injuries were: Multiple puncture wounds to the face, neck, chest and abdomen Singed hair and eyebrows Grenade fragments in the right maxillary antrum and right side of the nasopharynx Multiple puncture wounds to the left eye Bilateral perforated tympanic membranes Multiple fractured teeth Subtotal traumatic mastectomies (Figure 2a) Bilateral haemothoraces, with contusions of the right middle lobe and both lower lobes of the lungs Perforation of the right ventricle and apex of the heart, with a 250 mL pericardial haematoma Perforations of the stomach, left lobe of the liver, and splenic flexure of the colon Haematomas of the head of the pancreas, lesser omentum and left perinephric region Comminuted compound fracture of the right radius and ulna, with major soft tissue disruption to the dorsal surface of the right forearm (Figure 2b) Subtotal amputation of the right ring and little fingers A large skin and muscle defect to the anterior aspect of the right thigh (Figure 2c). After her initial surgery, Patient A required mechanically assisted ventilation for 11 days and underwent five additional operative procedures during her initial 39 days in hospital. In the next year she underwent another five operative procedures, with a further five reconstructive procedures scheduled for the following year. Patient B Patient B was a 45-year-old woman standing behind and to the left of Patient A, about 0.6 m from the grenade burst point. On arrival of the HEMS, she was in severe respiratory distress, with a heart rate of 105/min, systolic blood pressure of 60 mmHg, and Glasgow coma score of 10. Patient B was treated by ambulance personnel with 1.7 L of polygeline and transported to hospital by road. On arrival at hospital, Patient B had a heart rate of 88/min; systolic blood pressure, 140 mmHg; Glasgow coma score, 15; haemoglobin level, 83 g/L; pH, 7.30; PO2, 331 mmHg; PCO2, 36.6 mmHg; HCO3, 17.8 mmol/L; and base excess, - 7.2 (normal range, - 3 to 3). Patient B was taken to theatre for laparotomy and cholecystectomy. Her injuries were: Multiple puncture wounds to the face, neck, thorax and abdomen Forehead laceration and fractured nose Grenade fragments in the ethmoid and sphenoid sinuses, plus a fragment in the occipital lobe of the brain. Contusion of the lower lobe of the left lung Multiple perforations of the liver and gallbladder Lacerated right kidney and retroperitoneal haematoma Fragments adjacent to the abdominal aorta and the right L3 nerve root Delayed third-nerve palsy. She was discharged from intensive care after 1.5 days and discharged home after 18 days. She underwent six subsequent day procedures to repair soft tissue injuries and extract fragments. Patient C Patient C was a 54-year-old woman sitting to the right of patient A, 1.3 m from the burst point. On arrival of the HEMS, she had a heart rate of 80/min, systolic blood pressure of 130 mmHg and Glasgow coma score of 14. She was treated with 1 L of Hartmann's solution and 250 mL of polygeline and transported by air. On arrival at hospital, Patient C's heart rate was 83/min; systolic blood pressure, 151 mmHg; Glasgow coma score, 14; and haemoglobin level, 125 g/L. Her injuries were: Multiple small puncture wounds to the left upper limb and left side of the thorax A chin laceration and left periorbital haematoma; there were no perforations of the tympanic membranes A closed fracture of the left ulna A small left retinal tear and preretinal haematoma A fragment in the medial rectus of the left orbit. Patient C was discharged home after two days. Patient D Patient D was a 50-year-old woman sitting diagonally opposite Patient A about 1.6 m from the grenade burst point. On arrival of the HEMS, she had a heart rate of 105/min, systolic blood pressure of 70 mmHg and Glasgow coma score of 11. She was treated with 1 L of polygeline and transported by road. On arrival at hospital, Patient D had a heart rate of 90/min; systolic blood pressure, 150 mmHg; Glasgow coma score, 15; and haemoglobin level, 108 g/L. Her injuries were: Multiple fragment wounds to the right side of the face, neck, thorax and upper limb, including an intracapsular fragment in the right shoulder joint and a 4 cm scalp laceration Bilateral perforations of the tympanic membranes Right pneumothorax requiring drainage with an intercostal catheter, plus pulmonary contusions of the right mid and lower zones. She was discharged after seven days and subsequently underwent two outpatient procedures to extract fragments. Discussion The injuries of the four accident victims demonstrate the capability of military weapons to inflict serious physical injury -- but the potential psychological impact of these weapons extends well beyond their injury radius. The grenade exploded on a bench top, which directed most of the blast and fragments onto the upper torso of the victims, but it also blew a hole in the bench top, producing the wound to the right thigh of patient A. Any bomb or blast scene should always be considered dangerous until declared safe from the risk of secondary explosion by appropriate technical personnel. Specialist bomb disposal personnel were not part of the initial response to the St Marys incident, which contributed to the delayed recognition of a potential secondary explosion hazard from 48 damaged grenades at the site. The size and shape of modern military weapons often belie their wounding potential. Explosive devices may contain fuse mechanisms that detonate in response to stimuli other than that of direct impact. The bodies of victims may conceivably hold unexploded devices that pose a threat to rescue, medical and mortuary personnel.11,12 Major injuries and deaths occurred during the Gulf War among medical and service personnel handling souvenired battlefield ordnance.13 Injury from military weapons is uncommon in Australia. Army ammunition technical officers can provide expert advice on the composition and function of a particular device. This information is vital in the assessment of individuals wearing body armour, or who have been potentially exposed to either radiolucent fragments or items of ordnance which have become airborne.11,12 Early on-site specialist technical advice is critical to safe operations at the scene of an urban explosion. Helicopter emergency medical services have been used in the primary and secondary transport of blast victims.14,15 In this incident the HEMS rapidly provided a doctor on-site to assist in trauma triage. This allowed patients to be referred immediately for specialist surgical care, while dispersing casualties over a wider network of receiving hospitals (thus Patient A and Patient B were sent to different hospitals so that both could receive immediate surgical treatment; this would not have occurred under ambulance protocols). In addition, on-site medical care went beyond the scope of paramedic protocols16 (e.g., in performing a relaxant-assisted intubation and giving a blood transfusion). The NSW Ambulance Service Disaster Plan (December 1995) precludes the immediate dispatch of physician-staffed helicopter emergency medical services to potential multicasualty incidents. This may introduce avoidable delays in delivering seriously injured patients to definitive surgical care, especially in incidents geographically removed from designated trauma centres. Acknowledgement I thank Mr Mike Etzel for generously providing some of the background material. References Owen J. Infantry weapons of the world 1979. London: Brassey's 1979. Hogg IV. The illustrated encyclopaedia of ammunition. Sydney: Colporteur Press, 1985. Gander TJ, editor. Jane's infantry weapons. 22nd ed. 1996-97. London; Jane's Information Group, 1995. Rawlins JSP. Physical and pathophysiological effects of blast. Injury 1978; 9: 313-320. Huller T, Yaacov B. Blast injuries of the chest and abdomen. Arch Surg 1970; 100: 24-30. Cooper GJ, Maynard RL, Cross NL, Hill JF. Casualties from terrorist bombings. J Trauma 1983; 23: 955-967. Bellamy RF. The medical effects of conventional weapons. World J Surg 1992; 16: 888-892. Mellor SG. The relationship of blast loading to death and injury from explosion. World J Surg 1992; 16: 893-898. Handbook on weaponry. 2nd ed. DŸsseldorf: Rheinmetall GmbH, 1992. Courtney-Green PR. Ammunition for the land battle. London: Brassey's 1991. Clark MA. A fatal wound from an unusual military projectile: potential dangers of live military ordnance to the autopsy pathologist. J Forensic Sci 1987; 32: 793-797. Spencer JD. Accidental death by light anti-tank weapon: a dangerous autopsy? J Forensic Sci 1979; 24: 479-482. Thomson JD, Lisecki EJ. Injuries and deaths from collecting war souvenirs in Operation Desert Storm. Mil Med 1993; 158: 505-507. Brown MG, Marshall SG. The Enniskillen bomb: A disaster plan. BMJ 1988; 297: 1113-1116. Merriman M. Emergency medical response teams react swiftly to the Philips Plant tragedy. Occup Health Saf 1990; 59: 32-37. NSW Ambulance Service protocols, procedures and pharmacology. Sydney: NSW Ambulance Service, 1993. (Received 17 Jul 1996, accepted 18 Feb 1997) Authors' details NRMA CareFlight, Westmead, NSW. Antony Nocera, MB BS, Emergency Medicine Registrar. No reprints will be available. Correspondence: Dr Antony Nocera, NRMA CareFlight, PO Box 159, Westmead, NSW 2145. E-mail: tonynoceATozemail.com.au <URL: http://www.mja.com.au/> © 1997 Medical Journal of Australia.

Anthony Nocera