Letters

James Tibballs,* Sharon Kinney† * Associate Director, Intensive Care Unit, Royal Children’s Hospital, Flemington Road, Parkville, Melbourne, VIC 3052; † Lecturer, School of Nursing, University of Melbourne, Melbourne, VIC. james.tibballsATrch.org.au To the Editor: The introduction of a medical emergency team (MET) at the Austin Hospital significantly reduced cardiac arrest and deaths, and reduced time spent by survivors of cardiac arrest in the intensive care unit (ICU) and in hospital.1 We note that the evaluation of the MET was preceded by a 12-month period of education and a 2-month “run-in” period before the effects of the MET were analysed. We would be interested to know the incidence of death and cardiac arrest, and the duration of ICU and hospital admission in survivors of cardiac arrest during these two periods. Another MET service also claimed substantial benefits in patient outcomes,2 but was criticised on the basis that the results may have been due to better education of ward staff in recognising the antecedent signs of cardiac arrest and/or the creation of more “do-not-resuscitate” orders.3 While the latter criticism cannot be levelled at the study by Bellomo et al, no attempt is made to separate out the effects of the lengthy education period and the operation of the MET. This is a pity, because it would have been a relatively simple matter to do so without detracting from the obvious benefit of the service. Why was a 2-month “run-in” period between education and operation of the MET allowed before analysis of results? Was this a post-hoc decision or were there foreseeable difficulties during the introduction of the MET? Did patient outcomes change during these two periods compared with the period before the MET?

James Tibballs · Sharon Kinney

Rinaldo Bellomo Director, Department of Intensive Care, Austin and Repatriation Hospital, Studley Road, Heidelberg, VIC 3084. Rinaldo. BELLOMOATarmc.org.au In reply: Tibballs and Kinney raise important questions about our trial of a medical emergency team (MET).1 My colleagues and I are, in fact, currently studying these issues. Preliminary (not fully double-checked) data show that during the education period there were 69 cardiac arrests — an average of 23 cardiac arrests per 4-month block. This is a clear reduction from the 63 recorded during the 4-month control period and similar to the 22 cardiac arrests reported during the 4-month MET period. These 69 cardiac arrests led to a total of 227 intensive care unit bed-days or a 75 bed-day average for each 4-month block, about 50% of the number recorded during the control period, but more than twice as many as during the MET period. These patients remained in hospital for a total of 986 hospital bed-days or 328 days per 4-month block, close to an 80% reduction compared with our control period, and twice as many as the number achieved during the MET period. Unfortunately, although we are pursuing mortality data, we won’t be able to provide them for another 6 months because of a changeover in the computer system at our hospital. Tibballs and Kinney are invited to contact me directly by mid-2004. The 2-month run-in period was chosen prospectively, as we expected that the uptake of the MET (a major change of culture) might be slow and require time and encouragement. We were wrong: the MET was taken up with zest and enthusiasm. The histogram (Box 4) in our article1 shows no cardiac arrests during the run-in period, not because they were not recorded, but because there were literally none for 2 months in a row! Obviously, there were also no post-cardiac-arrest bed-days. Again, mortality figures for this period should be available by mid-2004. As we stated in the Discussion of our article, the educational program associated with the MET may indeed have been partly responsible for the findings. We were careful at all times to say that introducing the MET was effective, not the MET per se (see Conclusion).1 We are not aware of any prospective studies testing the effectiveness of introducing a hospital-wide education program aimed at increasing awareness of the significance of physiological instability. Our findings support a powerfully beneficial role of education, but only represent a post-hoc analysis and require validation in other settings and institutions. The role of education was prospectively and separately investigated in the recently completed multicentre cluster-randomisation MERIT study (Medical Early Response Intervention and Therapy). Its results should be available in the second half of this year.

Rinaldo Bellomo

Timothy H J Florin Director of Gastroenterology, Department of Medicine, University of Queensland, Mater Health Services’ Adult Hospital, South Brisbane, QLD 4101. t.florinATuq.edu.au To the Editor: In their article on adoption of celecoxib and rofecoxib by Australian general practitioners, Kerr et al noted that “the increase in COX-2 [cyclooxygenase-2] prescribing coincided with a period of energetic marketing to the medical profession, which promoted the message that the new C2SNs [COX-2-selective non-steroidal anti-inflammatory drugs] were ‘safer’ than traditional NSAIDs [non-steroidal anti-inflammatory drugs].”1 The implication is that the decision of Australian GPs to prescribe these new drugs may have been less than independent or rationally informed. The reason for prescribing C2SNs is that, like traditional NSAIDs, they relieve arthritic pain and so promote mobility, although, unlike traditional NSAIDs, they do not inhibit cyclooxygenase-1. While the power of advertising is undeniable, the simple message about C2SNs is that there is an approximate 50% reduction in clinically significant gastrointestinal (GI) complications compared with traditional NSAIDs.2 There are over a dozen articles to support the better GI side-effect profile of C2SNs. Most data support a non-cumulative, reversible, but constant, risk of peptic and other, more distal GI bleeds, or perforation, with the coefficient of risk being significantly greater for NSAIDs.3 Although the data from the CLASS study did suggest that the higher GI morbidity of NSAIDs seemed to diminish with time,4 that of celecoxib remained at a constantly lower rate.5 This is clinically important for all our patients, and especially for our ageing population with their comorbid conditions and polypharmacy. I suggest that it is for this single reason that many doctors have been quick to take up C2SNs for their patients. The better GI safety enfranchised patients who previously could not take NSAIDs safely, and could explain why the overall anti-inflammatory market increased by 20%.1 However, no one suggests that the C2SNs are free of non-GI side-effects. To the best of my knowledge, COX-2-specific NSAIDs have not been promoted as being free of non-GI side-effects or better than COX-1-specific NSAIDs in this regard. While agreeing with the last sentence of Dowden’s editorial that “new is not always better”,3 the opposite — that “new is sometimes better” — is also true. Thus, we persuaded the accountants in our hospital, who rightly participate in the determination of which drugs are available on its formulary, to accept one of the COX-2 drugs because of its better GI complication profile. While this will not reduce its pharmacy budget, it is anticipated to reduce overall hospital costs in this area,5 which should allow a reapportionment of its budget to other areas of need. Of course, the hospital is watching carefully for any unforeseen “serious adverse effects which sometimes only emerge after marketing”.3

Timothy H J Florin

Stephen J Kerr,* Andrea Mant,† Fiona E Horn,‡ Kevin McGeechan,§ Geoffrey P Sayer¶ * Decision Support Coordinator, National Prescribing Service, PO Box 1147, Strawberry Hills, NSW 2012; † Area Advisor, Quality Use of Medicines, South East Health, Sydney Hospital, Sydney, NSW; ‡ Research Analyst, § Senior Research Analyst, ¶ General Manager — Research, Health Communication Network, St Leonards, NSW. skerrATnps.org.au In reply: The advantage of the COX-2-selective NSAIDs (C2SNs) is the reduction in clinically significant gastrointestinal complications compared with conventional NSAIDs, but, as Florin agrees, other toxicities, including the risk of renal failure and heart failure, are similar for C2SNs and the older drugs.1 We speculated that doctors may have been more aware of the differences between the new and the conventional anti-inflammatories rather than the similarities: between 4.7% and 7.9% of patients in our study cohorts were treated with a combination of drugs which placed the patient at risk of renal complications. Florin points out that there is an approximate 50% reduction in clinically significant gastrointestinal (GI) complications with C2SNs compared with conventional NSAIDs. If, in a population, the annual incidence of serious GI complications with NSAID use is around 1.4%,2 then the absolute risk reduction is 0.7%. This means that about 140 patients would need to be treated with a C2SN for one year to prevent one serious GI complication. Messages conveyed in this way may be more pertinent to clinical decision making than a statement about relative risk reduction. Florin also notes the problems with elderly patients who often take multiple medications, and are probably at increased risk of upper-GI events with NSAIDs. Our data demonstrated very high prescribing rates in patients who were not elderly. Over 20% of patients in our cohorts were aged less than 50 years, and over 50% were aged less than 65 years. Furthermore, between 34.5% and 61.3% had no pain medication prescribed in the 12 months before the first C2SN prescription, suggesting that C2SNs may have been used as a first-line pain medication in these patients. Quality use of medicines advocates prescribing which is safe, judicious, effective and cost-effective. Recent pharmacoeconomic studies suggest the cost-effectiveness may only be realised when prescribing of C2SNs is confined to patients who are at high risk of GI complications.3,4

Stephen J Kerr · Andrea Mant · Fiona E Horn · Kevin McGeechan · Geoffrey P Sayer

John S Dowden Editor in Chief, Australian Prescriber, Suite 3/2 Phipps Close, Deakin, ACT 2600. jdowdenATnps.org.au In reply: The general practitioners’ decision to prescribe COX-2 inhibitors was rational, but the information underpinning their decision was less than independent. A big reduction in short-term relative risk can be persuasive, even if the absolute benefit is small. General practitioners deal with whole patients, so they consider the overall risks of treatment, and not just one adverse effect. While COX-2 inhibitors may have gastrointestinal advantages, they may have cardiovascular disadvantages. Treatments for chronic conditions should be based on long-term data. The observation that most of the ulcer complications in the second half of the CLASS study were in patients taking celecoxib is therefore important.1 Undoubtedly, some patients who could not take non-selective non-steroidal anti-inflammatory drugs (NSAIDs) were able to tolerate COX-2 inhibitors. However, Kerr et al found that up to 61% of patients given a COX-2 inhibitor had not previously been prescribed any analgesia.2 It seems unlikely that so many people suddenly required analgesia that only a COX-2 inhibitor could provide. The Pfizer-funded study by MacDonald et al shows that UK general practitioners tended to prescribe COX-2 inhibitors for patients at risk of gastrointestinal haemorrhage.3 This follows the advice of the National Institute of Clinical Excellence (NICE). However, NICE also recommended against the routine use of COX-2 inhibitors.4 A review by the Canadian Co-ordinating Office for Health Technology Assessment has also concluded that COX-2 inhibitors may have no significant safety advantage over diclofenac.5 Solomon et al conclude that the cost of adverse effects of NSAIDs in low-risk elderly patients is modest.6 However, there is no comparison with COX-2 inhibitors, so we do not know if they reduce this cost. Hospital accountants may be interested to know that researchers at the Mayo Clinic concluded that, in terms of averting gastrointestinal events, the most cost-effective analgesic is paracetamol.7

John S Dowden

Alan Rodger Director, and Professor of Clinical Oncology, Beatson Oncology Centre, Western Infirmary, Dumbarton Road, Glasgow, Scotland G11 6NT, UK. alan.rodgerATnorthglasgow.scot.nhs.uk To the Editor: In their amusing Postcard from the UK, Jamrozik, Heller and Weller1 paint a picture familiar to most of us in the National Health Service (NHS), but, as with most art, some licence has been permitted. Comparing the size of the UK NHS workforce with that of the Chinese army is unfair: there are, as they know, four different health services in the UK (England, Scotland, Wales and Northern Ireland), which function separately and distinctly. The NHS Jamrozik et al describe is much more the English model. In Scotland, targets for cancer care have been set, although they will not be applied for over a year yet. In the meantime, NHS Scotland has awarded the three Scottish regional cancer networks an extra £25 million (A$62.5 million) to improve and enhance agreed cancer services to assist them achieve the agreed targets on waiting times, etc. Another example of the difference between the Scottish and English NHS can be found in the detail of the new consultant contract. Scotland is offering its consultants a sabbatical and England does not. I am reliably informed by one of the negotiators that the seed of that idea was sown by me when I was recounting enthusiastically some of the better experiences of working in Australia’s own, if complex, national health service. Finally, those of us who laboured during the 1990s at the coal face of Victoria’s health service will not be unfamiliar with targets. Infringement of targets set for the upper levels of the waiting list (no one in Category 3 could ever expect treatment, so no target was set), and for 12-hour waits in the emergency department, carried huge financial penalties for the institution. The good old NHS is far from unique in its fondness for targets.

Alan Rodger

John B Dixon Senior Research Fellow, Monash University Department of Surgery, Alfred Hospital, Commercial Road, Melbourne, VIC 3181. john.dixonATmed.monash.edu.au To the Editor: I am concerned about a possible misrepresentation in the Journal. A recent Practice Essentials article included a case report describing the management of type 2 diabetes.1 The case involved a 51-year-old man with a 6-year history of type 2 diabetes and body mass index (BMI) of 32 kg/m2. The case report included a photograph, but this was not of a man with a BMI of 32 kg/m2. Ten consecutive professional staff working at a large Melbourne centre for bariatric (obesity) surgery were asked to estimate the man’s BMI. Estimates varied from 50 to 65 kg/m2. None accepted that his BMI was 32 kg/m2. All estimates placed this gentleman in the obesity class III category (BMI > 40 kg/m2). A man with a true BMI of 32 kg/m2 is shown in the Box. If indeed the man depicted previously has class III obesity, he is likely to suffer significant obstructive sleep apnoea,2 physical disability and poor quality of life, in addition to the metabolic syndrome and type 2 diabetes. The case report focused on the management of an unacceptable level of glycosylated haemoglobin (HbA1c) of 8.9%. The management may, arguably, be appropriate if his BMI were 32 kg/m2, although treatment options may promote weight gain. Unfortunately, the options and discussion were inadequate for the man depicted. This ill, disabled man deserves better assessment and care. His obesity-related conditions should be fully explored. The only management likely to have a significant effect on conditions related to class III obesity (diabetes, metabolic syndrome, sleep apnoea, physical disability and poor quality of life) is significant weight loss. The only treatment that reliably provides significant sustained weight loss for those with class III obesity is surgery. Obesity surgery therefore should be an essential inclusion in his treatment options. The beneficial effects of modern obesity surgery on type 2 diabetes, the metabolic syndrome, sleep and quality of life are well documented.3,4 I believe the inclusion of this illustration is a serious misrepresentation. Editor's note: We agree with Dr Nixon's remarks, and have replaced the inappropriate illustration in the article with the image he has kindly supplied. Patient with a body mass index of 32 kg/m2 (height, 172 cm; weight, 96 kg)

John B Dixon

Sanjaya N Senanayake,* Mark J Ferson,† Susan J Botham,‡ Roslyn T Belinfante§ * Infectious Diseases Physician, † Director and Medical Officer of Health, ‡ Infectious Diseases Team Coordinator, § Environmental Health Officer, South Eastern Sydney Public Health Unit, Locked Bag 88, Randwick, NSW 2031. djayaATbigpond.com To the Editor: Keeping fish as pets, as with other animals, carries the risk of zoonotic infections. A 2000 review of bacterial zoonoses that can be acquired topically from fish commented on infection with Aeromonas hydrophila, Edwardsiella tarda, Erysipelothrix rhusiopathiae, Mycobacterium marinum, Streptococcus iniae, Vibrio vulnificus and Vibrio damsela.1 Our public health unit was recently notified of a case of Salmonella infection acquired from a fish tank. In October 2003, a 14-month-old boy was admitted to hospital with a 2-day history of fever, vomiting and diarrhoea. Culture of three stool specimens revealed a Salmonella isolate. This was identified by serotyping (at the Institute of Medical and Veterinary Science, Adelaide, South Australia) and phage typing (at the Microbiological Diagnostic Unit, Melbourne, Victoria) as Salmonella enterica serotype Paratyphi B var Java phage type Dundee. The child had no recent history of overseas travel or overseas visitors. Other family members were well, and their stool specimens were negative for Salmonella spp. However, the family kept a tropical fish tank, and the child’s parents reported that he would place his hands in the water to help feed the fish. Culture of water from the tank revealed a Salmonella isolate identical to that in the stool specimens. We believe that the boy became infected after touching the water while feeding the fish. The fish tank contained red-eyed tetras, bala sharks, silver dollars and angel fish, which had been purchased from three local aquariums several years previously. None appeared sick at the time of the child’s illness. It is important for clinicians to recognise that Salmonella infections are not always foodborne in origin. Salmonellosis is a well-known zoonosis that can be found in a variety of pets, including cats, dogs, birds, rodents and even reptiles.2 In fact, an estimated 90% of all reptiles shed Salmonella spp. in their faeces.3 Salmonella spp. have been isolated from tropical fish aquariums previously; unusual Salmonella serotypes were found in eight of 100 tropical aquariums sampled in Wales.4 Fish can excrete Salmonella without appearing unwell.5 Our patient was a 14-month-old child. In a Canadian outbreak of S. enterica serotype Paratyphi B linked to aquariums, five of seven cases were also in children aged under 10 years.5 This may reflect a combination of immature immunity and behaviour — young children may not wash their hands properly (or at all) before eating or touching their mouths. This case highlights the importance of good handwashing at all times after contact with an aquarium, regardless of the appearance of the fish. It may also be worthwhile recommending close supervision of children under 5 years of age around aquariums.

Sanjaya N Senanayake · Mark J Ferson · Susan J Botham · Roslyn T Belinfante

Lyndon E Llewellyn,* Cedric E Robillot,† Andrew P Negri‡ * Principal Research Scientist, Bioactive Molecule Discovery; † Senior Research Specialist, Bioinnovation; ‡ Senior Research Scientist, Bioinnovation, Australian Institute of Marine Science, PMB 3, Townsville, QLD 4810. L. LlewellynATaims.gov.au To the Editor: Seafood is not only consumed as food, but also as dietary supplements and complementary medicines. Examples are capsules of freeze-dried oysters and mussels, or freeze-dried extract of shellfish meat, sold as reputed antihypertensives, cardioprotectants, and anti-inflammatories, among other medical claims. However, oysters and mussels can become dangerously toxic after they ingest poisonous microscopic algae. If these molluscs are sold as food in Australia, they are subject to the Food Standards Code,1 under which their sale is prohibited if biotoxin levels per kilogram of wet shellfish meat exceed 800 g of paralytic shellfish poisons, 200 mouse units of neurotoxic shellfish poisons, 200 g of diarrhoetic shellfish poisons, or 20 mg of amnesic shellfish poisons. Shellfish capsules can be simply manufactured by milling dried meat and encapsulating the powder, a process unlikely to degrade shellfish biotoxins, which are stable to heat, pressure and freeze-drying. 2,3 Such capsules may then become subject to regulation by the Australian Therapeutic Goods Administration (TGA), which distinguishes therapeutics from food, on the basis of whether there is a “tradition of use as a food in the form presented”, especially if there is an associated health claim. Such complementary medicines can be either “registered” or “listed”. Registered medicines require extensive safety, quality and efficacy data. Listed medicines are considered to pose a lower risk than registered medicines, and regulations allow product sponsors to “self-assess” products. Listing is a route commonly taken for complementary medicines. A pertinent example is the TGA listing of therapeutic goods containing dried green-lipped mussel (Perna canaliculus).4 Where manufacturers of shellfish capsules have undertaken the responsibility of ensuring product safety, they invariably adopt existing biotoxin testing protocols developed for food safety. However, as the allowable biotoxin level is based on wet weight, and the dry weight of bivalve shellfish is only 10%–15% of the wet weight,5 safety limits for shellfish meat as food are incorrect by an order of magnitude, and potentially more for capsules containing extracts of shellfish meat. While important for acute exposure to these toxins, this may be even more relevant in chronic exposure. Okadaic acid, the cause of diarrhoetic shellfish poisoning, is a tumour promoter,6 and epidemiological studies suggest that rates of cancer have increased in regions with regular dietary exposure to low levels of this toxin.7 Capsules are available that contain 500 mg of dried shellfish meat, which may equate to 5 g of wet shellfish meat.5 Unlike a shellfish meal, which may be considered equivalent to a single acute exposure, recommended doses for shellfish capsules can be as many as five capsules a day for many weeks, if not months, therefore magnifying the risk of chronic exposure. It is known that different classes of biotoxins can co-occur in shellfish, adding to the potential hazard outlined here.8 Further complications arise because some shellfish capsules include other natural extracts (such as ginseng) or pharmaceutical formulations that might affect toxin uptake. While this situation needs to be subjected to risk assessment, testing products in accordance with an inappropriate standard can make them seem safe when they might not be. This is especially so for products which are usually self-prescribed, and where patients can exceed recommended doses in the belief that more is better. For products such as shellfish capsules that straddle the food/therapeutic divide, it is better for manufacturers to test the final consumer product and not the raw supply.

Lyndon E Llewellyn · Cedric E Robillot · Andrew P Negri

Shanthi Kannan,* Christine H Rowland,† Gregory I Hockings,† Peta M Tauchmann,‡ Edwin A Blackwell§ * Endocrine Advanced Trainee, † Endocrinologist, ‡ Nurse Educator, § Haematologist, Greenslopes Private Hospital, Brisbane, QLD. Correspondence: Dr Christine Rowland, Greenslopes Private Hospital, Newdegate Street, Greenslopes, Brisbane, QLD 4120. c.rowlandATmailbox.uq.edu.au To the Editor: Intravenous immunoglobulin preparations containing maltose (eg, Intragam P [CSL Limited]) can interfere with the readings of blood glucose monitors that use test strips with glucose dehydrogenase, such as Advantage (Roche).1 Glucose dehydrogenase is an enzyme of the Pyrroloquinolinequinone class which reacts with the disaccharide isomer maltose present in Intragam P at concentrations of 10 g/100 mL,2 resulting in falsely elevated blood glucose level results. Blood glucose monitors, such as Precision (Medisense products), Medisense (Medisense products)3 or Accutrend (Roche), that use the glucose oxidase system, do not react with maltose and can be safely used for patients receiving Intragam P. We have had two patients in whom capillary blood glucose levels were over-estimated. One was a 64-year-old woman with type 2 diabetes who was being treated with haemodialysis, and who received Intragam P for immune-mediated thrombocytopenic purpura. Before being admitted, her diabetes was reasonably controlled by 10 units of protophane twice daily. She received prednisone (100 mg/day) from Day 1 to Day 13 and Intragam P intravenously each day from Day 2 to Day 6 and from Day 14 to Day 18. While the patient was receiving Intragam P, the Advantage monitor gave persistently higher readings than both the Precision monitor and plasma glucose level measurements. The following paired readings were obtained: Advantage capillary glucose reading of 9.3 mmol/L while the plasma glucose level was 2.3 mmol/L; and Advantage capillary glucose reading of 24.4 mmol/L while the plasma glucose level was 10.4 mmol/L. During the second course of Intragam P we noted that readings with the Precision monitor were equivalent to the measured plasma glucose level. This patient developed hypoglycaemia because her insulin doses were increased on the basis of falsely elevated capillary glucose readings as measured on Advantage blood glucose strips. A Precision monitor was used for this patient until after the haemodialysis treatment which followed the final dose of Intragam P. After this time Advantage blood glucose readings approached those obtained with the Precision monitor. A second patient — a 35-year-old woman who was not known to have diabetes, but was having her glucose levels monitored while undergoing total parenteral nutrition — was given Intragam P for idiopathic thrombocytopenic purpura. While receiving Intragam P, capillary glucose levels by the Advantage monitor were elevated by 13–20 mmol/L compared with concurrent (and normal) plasma glucose level measurements. It has previously been reported that icodextrin, used in some peritoneal dialysis fluids, can have similar effects, as it is hydrolysed to oligosaccharides, including maltose, maltotriose and maltotetraose.4,5 The reaction with maltose has important clinical implications. Although it is stated in the product information for Advantage test strips that glucose readings may be affected by maltose levels over 16 mg/dL (0.89 mmol/L), clinicians may not be aware that Intragam P is in a maltose solution. For patients receiving Intragam P, the Advantage monitor is not suitable, and the Precision, Medisense or Accutrend blood glucose monitors, which use glucose oxidase which does not react with maltose, should be used instead. We have reported our experience to the Australian Red Cross Blood Service as an adverse event.

Shanthi Kannan · Christine H Rowland · Gregory I Hockings · Peta M Tauchmann · Edwin A Blackwell

Anna Peeters,* Jason Ting,† Mark R Nelson,‡ John J McNeil§ * Research Fellow, Department of Epidemiology and Preventive Medicine, Monash University, and Department of Public Health, Erasmus Medical Centre, PO Box 1738, Rotterdam, 3000DR, The Netherlands; † Intern, Geelong Hospital, Geelong, VIC; ‡ NHMRC Research Fellow; § Head, Department of Epidemiology and Preventive Medicine, Monash University, Prahran, VIC. a.peetersATerasmusmc.nl To the Editor: Coronary heart disease (CHD) risk prediction for primary prevention now focuses on multifactorial risk,1-3 and various risk calculation tools exist.1-3 The need for such tools depends on the degree to which risk status can be estimated by healthcare professionals. We performed a study to assess general practitioners’ intuitive calculation of multifactorial CHD risk for patients likely to be considered for lipid-lowering therapy. In 1999 we posted a survey to a random sample of 400 GPs in Victoria, and received back 155 completed surveys (39% response rate). The respondent population was demographically similar to the Australian GPs and trainees who billed Medicare in 1998–99. GPs were asked to estimate 5-year absolute risk of CHD for four vignettes (two primary prevention and two secondary prevention), each based on the average characteristics of a published lipid-lowering clinical trial cohort4-7 (vignette descriptions are available from the authors). For the primary prevention vignettes, GPs were also requested to estimate the risk relative to other Australians of the same age and sex. Estimated risks were compared with reported risks to derive risk ratios. GPs accurately estimated the relative CHD risk for the two vignettes of patients with no prior CHD, with mean risk ratios of 0.84 (95% CI, 0.77–0.91) and 1.23 (95% CI, 0.97–1.49). However, they overestimated absolute risk, with risk ratios ranging between 2.79 (95% CI, 2.61–2.96) and 6.43 (95% CI, 5.59–7.27). The proportion of GPs estimating within 10% of the actual risk was 9% for a 62-year-old man with prior CHD and average cholesterol level, 13% for a 60-year-old woman with prior CHD and high cholesterol level, 17% for a 55-year-old man with no prior CHD and high cholesterol level, and 43% for a 58-year-old man with no prior CHD and average cholesterol level. Although the two primary prevention vignettes had 5-year absolute CHD risks below 10%, most GPs’ estimates (93% and 71%, respectively) were greater than 10%. In conclusion, GPs in Victoria have a good understanding of a patient’s relative risk of CHD, but they consistently overestimate absolute risk. The problem with absolute risk estimation may not be the sophisticated multifactorial calculations required, but rather a general overestimation of risk within the population, at least for middle age. The GPs’ estimates of absolute risk for the four vignettes were correctly ranked, suggesting that they recognised the degree to which a risk was lower or higher, but were unfamiliar with the scale. This overestimation led most GPs to categorise patients such as those from the AFCAPS/TexCAPS trial, with only a 3.3% 5-year risk of CHD,6 as having a risk of greater than 10%. GPs would therefore incorrectly consider these patients appropriate for lipid-lowering therapy according to national and international guidelines. 1-3 Education may improve understanding and accuracy of risk communication for CHD in middle-aged patients, but tools for accurate assessment of coronary risk are needed in routine clinical practice.

Anna Peeters · Jason Ting · Mark R Nelson · John J McNeil

Keri Lockwood,* Ian D Cameron,† Susan E Kurrle‡ * Coordinator, Hip Protector Studies Unit; ‡ Director, Rehabilitation and Aged Care Service, Hornsby Ku-ring-gai Hospital, Hornsby, NSW; † Head, Rehabilitation Studies Unit, University of Sydney, PO Box 6, Ryde, NSW 1680. iancATmail.usyd.edu.au To the Editor: Hip protectors may be an effective means of preventing hip fracture in residents of aged-care facilities, if there is adequate compliance with their use.1 Research studies conducted in the Northern Sydney and Illawarra health regions2,3 have found that, if a hip protector is worn at the time of a fall, the chance of hip fracture is reduced by about 80%. We wished to determine the extent to which hip protector use had become a routine part of healthcare in residential aged-care facilities generally, and whether there was a greater uptake of their use in areas where the research had been conducted. A brief questionnaire about hip protector use was sent to all residential aged-care facilities in the Northern and South Eastern Sydney health regions, and the region covered by the New South Wales Southern Area Health Service. The response rate was 60% in each region, and responses are summarised in the Box. Only 30 residential aged-care facilities participated in the previous study, and the results indicate that use is now much more widespread. Another contrast is that, in the research study, hip protectors were provided free of charge, whereas most current users pay for hip protectors. Overall, the survey indicates that 61% of aged-care facilities across three regions of NSW are either using hip protectors or have used them in the past. Hip protector use was similar in Northern Sydney and Southern NSW; however, comparing Northern Sydney (where previous research had been conducted) with the other two regions combined showed a greater prevalence related to conducting previous research. Issues such as cost, laundering of the hip protectors, and comfort are seen as major barriers to their use, but many aged-care facilities continue to use them for selected residents at high risk of hip fracture. It is our impression that the more residents within a facility who wear hip protectors, the better the adherence to hip protector use and the management of these issues. It is also evident that hip protector research encourages use of hip protectors. We are unsure of the mechanism involved, but believe that knowledge and enthusiasm for use of hip protectors is spread by word of mouth between health professionals, staff working in nursing homes and hostels and, in some cases, through relatives of the residents of residential aged-care facilities. Hip protector use in residential aged-care facilities in three health service regions of New South Wales Region (no. of residential aged-care facilities) Any hip protector use* Northern Sydney (99) 74% South Eastern Sydney (104) 43% Southern NSW (35) 75% * χ2 = 7.5, df = 1, P = 0.006 for comparison of frequency of use in Northern Sydney with use in the other two area health services combined.

Keri Lockwood · Ian D Cameron · Susan E Kurrle

Timothy J McCulloch Anaesthetist, Royal Prince Alfred Hospital, Missenden Road, Camperdown, NSW 2050. tmccullATusyd.edu.au To the Editor: Chilov and colleagues have presented an updated set of guidelines for management of hip fracture, which included the statement “regional anaesthesia is recommended for most patients”.1 The evidence for this recommendation was graded as Level I (National Health and Medical Research Council) and was supported by a single reference, namely a systematic review from the Cochrane Database by Parker et al.2 Parker et al performed a meta-analysis of the published trials examining the effect of regional versus general anaesthesia on a variety of outcomes after surgery for hip fracture. A possible difference in 1-month mortality was found in favour of regional anaesthesia, but this difference was borderline using one statistical model (relative risk, 0.7; 95% CI, 0.5–1.0) and non-significant using another model. There was no significant difference in mortality at 3 months or 1 year, and no significant difference in a variety of other outcomes. Appropriately, the authors concluded that “both regional and general anaesthesia produce comparable results and therefore anaesthetists should choose which technique is most appropriate for each individual patient”.2 One of the many difficulties in interpreting meta-analyses of regional anaesthesia is that most of the published trials were performed some decades ago. For example, one study that contributed a large proportion of the data within the Cochrane meta-analysis was conducted between 1980 and 1982, and patients were explicitly excluded if they were receiving low-dose anticoagulation therapy.3 The relevance of such trials to patients receiving general anaesthesia today is highly questionable, given the improvements in general anaesthetic drugs and techniques and the importance now placed on routine thromboprophylaxis. There is a wide range of opinion within the specialty of anaesthesia regarding the place of major regional blockade, with little outcome-based evidence to support any particular advantage of these techniques. Although medical practitioners can benefit greatly from the efforts of reviewers to develop guidelines based on the best available evidence, care must be taken to ensure that recommendations do not go beyond what is supported by available data. Particular care needs to be taken when recommendations are made for areas of practice outside the reviewers’ expertise. The authors of these guidelines might consider withdrawing their recommendation regarding choice of anaesthesia.

Timothy J McCulloch

Michael N Chilov,* Ian D Cameron,† Lynette M March‡ * Intern, Concord Hospital [corresponding author], 50 Mi Mi Street, Oatley, NSW 2223; † Chair, Rehabilitation Medicine, University of Sydney; and Director, Aged Care and Rehabilitation Services, Northern Sydney Area Health Services, Rehabilitation Studies Unit, Ryde, NSW; ‡ Senior Staff Specialist in Rheumatology and Clinical Epidemiology, Department of Rheumatology, Royal North Shore Hospital, St Leonards, NSW. mchilovATgmp.usyd.edu.au In reply: We thank McCulloch for his comments regarding the use of regional anaesthesia in the surgical management of hip fracture. He makes the point that surgical and anaesthetic techniques have improved and implies that the advantage seen for regional anaesthesia in published studies may no longer be present. Given that controversy still exists, we would recommend that further randomised controlled trials be conducted. However, for the following reasons, we stand by our recommendation that the available evidence supports the use of regional anaesthesia for most patients with this condition. Our current recommendation is unchanged from the earlier version of the guideline (published in the Journal in 1999),1 and is also consistent with at least one other published guideline.2 A number of the concerns raised by McCulloch were addressed in the response to a letter by another correspondent after the publication of the original guidelines.3 While we acknowledge that the review by Parker et al4 only found the reduction in mortality at 1 month to be of borderline significance, when our review team reassessed the original articles using the Cochrane Collaboration protocol we reached a summary odds ratio for mortality of 0.68 (95% CI, 0.49–0.96). With time and further studies we expect that this estimate of effect will become more precise as the power of the meta-analysis is increased. This view is supported by a systematic review of all randomised studies comparing regional anaesthesia with general anaesthesia across surgical specialties. The study of Rodgers et al found a statistically significant reduction in mortality (odds ratio, 0.70; 95% CI, 0.54–0.90) when regional anaesthesia was compared with general anaesthesia.5 This overall point estimate is very similar to that of Parker et al in their meta-analysis of patients with hip fracture. Although lack of power meant that statistical significance did not exist within individual surgical specialties, there was, in fact, little difference in the effect across surgical groups, with no significant heterogeneity between studies. Serious complications of regional anaesthesia (eg, spinal haematoma) are extremely rare, as shown in the recent PEP study in Australia and New Zealand that reported no cases in 4603 patients undergoing regional blockade.6 This should be compared with the number needed to treat with regional anaesthesia to prevent one death of 38, according to the data of Parker et al.4 There is no doubt that our recommendation needs to be considered in the context of individual patient characteristics and, while the recommendation may not apply to all patients with hip fracture, we feel that the available evidence supports the use of regional anaesthesia.

Michael N Chilov · Ian D Cameron · Lynette M March

Alan Rodger Medical Director, and Professor of Radiology Oncology, Beatson Oncology Centre, Western Infirmary, Dumbarton Road, Glasgow, G11 6NT, UK. alan.rodgerATnorthglasgow.scot.nhs.uk To the Editor: I strongly support the editorial comments and recommendations of McNeil et al1 on public funding of clinical trials. Having worked in the Australian healthcare system for 11 years, and having returned to a changed National Health Service in Scotland a few months ago, I can vouch for the benefits that accrue from adequate funding for clinical trials. While McNeil and colleagues focus on large-scale trials, their comments apply equally to smaller trials. Certainly, in oncology, several trials organisations in Australia have struggled for years to continue conducting trials in spite of inadequate government funding. The ANZ Breast Cancer Trials Group and the Trans Tasman Radiation Oncology Group are but two organisations with which I am familiar. In addition to precarious funding, I believe the consequences in the past 2 years of upheaval in the insurance industry have placed all such groups on an uncertain and untenable footing. Clinical trials must be ethical, scientific and well managed. Clinicians entering patients into trials need support from essential data managers and clinical nurse specialists. Governments encourage and, in fact, demand evidence-based medicine. The only effective way to produce the evidence is to conduct clinical trials. That costs money. In Victoria, cancer trial management was supported by about $800 000 per annum in grants from the Cancer Council Victoria. Those funds provided start-up assistance to institutions new to clinical trials and supported the others that could not rely on pharmaceutical company largesse. However, only part of that money was state government funded and then only for rural and regional centres or for breast cancer. The diagnosis-related-group-based casemix funding of Victorian hospitals included a notional element for research. That sop was lost in budget deficits. In Scotland, where health matters are totally devolved to the Scottish Parliament, the latter’s Scottish Executive Health Department has very recently enhanced funding for cancer care. This includes £500 000 (A$1.25 million) annually to support clinical research in cancer. Each of the three cancer networks has a guaranteed share of that sum to resource clinical trials in all the associated health boards and cancer units. This is in addition to the excellent clinical trials units in the main cancer centres, often funded by the charity Cancer Research UK and industry. There is also a national system of considering and approving clinical research in cancer. Such approval places obligations on health boards to support such trials. Lastly, accreditation of cancer centres can depend on clinical trial participation. The evidence that this (still imperfect) system has an effect is seen in trial entry at our oncology centre, where 11% of patients are already entered into trials. The new funding should see that increase. Government needs to put its money where its mouth is: evidence needs resourcing. The alternative is to rely on charity or on industry (whose eye is more often on marketing than science).

Alan Rodger

Paul Walker Paediatric Otolaryngologist, John Hunter Children’s Hospital, PO Box 293, New Lambton, NSW 2305; and Conjoint Associate Professor, Disciplines of Surgery and Paediatrics, University of Newcastle. walkerpATtpgi.com.au To the Editor: The study by Paradise et al on tympanostomy tubes for persistent otitis media,1 which was expertly reviewed by Morris and Leach in the Journal recently,2 has since been updated.3 The findings of both studies from Pittsburgh should only be applied with caution in Australia. Indications for inserting ventilating tubes (VTs) can be divided into three: bilateral hearing loss of more than 25–30 dB continuously for 3 months after failed non-operative management; structural damage to the tympanic membrane (TM) which may lead to irreversible hearing loss or cholesteatoma; and a miscellany which includes under-lying sensorineural hearing loss or learning difficulties or similar conditions with deterioration associated with bilateral middle ear effusion (MEE), and recurrent middle ear infections with use of VTs as an alternative to antibiotic prophylaxis, among others. The conclusion of the more recent article by Paradise et al3 — that there was no difference in expressive or receptive speech or cognition between children in whom VTs were inserted early or late — is valuable, but may not readily be extended to Australian practice. Although 6350 children were enrolled, only 397 were actually randomly allocated into the early or late treatment groups. Thus, the numbers are not large. Of more concern is that only 18% of those analysed had bilateral continuous MEE (40 in the early and 32 in the late treatment group), with the remaining 82% having unilateral (continuous or discontinuous) or bilateral discontinuous MEE. Only the 18% with bilateral continuous MEE would ordinarily be candidates for VTs in Australia, as Paradise et al underline the fact that intermittent and/or unilateral MEE is not associated with speech and language difficulties in the absence of other handicaps to learning. An abnormal hearing test result was identified by the study as a 15 dB loss. This could well fall in the normal range for the Australian Hearing Service for children wearing headphones, and a minimum threshold of 25-30 dB should typically be required for considering VTs in Australia. As Morris and Leach2 pointed out for the earlier study,1 the later study also excludes children “not otherwise healthy”,3 and the results cannot be generalised to such children, or to those with moderate rather than mild hearing loss. Pointing to studies such as that of Paradise et al can be very helpful in reassuring parents who want VTs for their child with unilateral or intermittent hearing loss that not having VTs does not place the child’s speech and language development at risk.

Paul Walker

Stacey L Emmett,* Adam J O’Brien,† Joseph E Ibrahim† * Research Officer, † Consultant Physician, Clinical Liaison Service, Victorian Institute of Forensic Medicine and the State Coroner’s Office, 57-83 Kavanagh St, Southbank, VIC 3006. staceyeATvifm.org To the Editor: Root-cause analysis is an established, retrospective, structured investigative technique1 that was first introduced into wide clinical practice in public hospitals in Victoria in 2001.2 It is usually reserved for investigating infrequent and significant adverse events and explores the nature and causes of organisational systems failures.1 It is important for all healthcare professionals to understand this technique, as it is used increasingly by healthcare organisations. Cooper and Blamey’s Lesson from Practice described the outcome of an investigation using root-cause analysis of an occupational exposure to HIV from a needlestick injury.3 We commend the analyses that identified multiple failures of the system for reporting and responding to occupational exposures to hazardous material. The practice changes that ensued at Southern Health demonstrate the value of root-cause analysis. However, more information and analysis is required about the mistaken use of the stored serum samples. As the authors explain, the initial information was that the source patient tested negative for HIV antibodies. Some time later, it was discovered that the specimen tested was not from the source patient but from a patient of the same surname in the same ward. This caused a 3-day delay between the initial test and the Infectious Diseases Unit being notified that the source patient had twice tested positive for HIV antibodies. We contend this is an important and common systems failure that usually makes headlines of the form “Wrong site, wrong procedure, wrong person surgery”. The Joint Commission on Accreditation of Healthcare Organisations developed a universal protocol with the intention of highlighting the causative systems failures and minimising the frequency of recurrences of wrong-site surgery.4 The information given by Cooper and Blamey does not clearly explain why (ie, the root cause) the incorrect specimen was tested initially. The pathology department’s review identified the presence of unacceptable “informal norms” in the practice of blood collection and labelling. The suggested remedy that “all serum should be collected with strict adherence to blood collection and labelling protocols” is unlikely to prevent a recurrence. Exhortation to do better rarely solves the underlying problem. It is therefore important to understand why health professionals violate procedures and protocols.5 The experiences of Cooper and Blamey demonstrate that some of the limitations and benefits of root-cause analysis depend on the depth of the investigation. The early and unquestioning acceptance that strict adherence to an existing protocol will prevent another “wrong person” error is not convincing. This contrasts with the well-conducted inquiry and subsequent management of occupational exposure to needlestick injuries.

Stacey L Emmett · Adam J O’Brien · Joseph E Ibrahim

Elizabeth E Cooper,* Stephen L Blamey† * Sterilisation and Infection Control Coordinator, Southern Health Infection Control and Epidemiology, Southern Health, Locked Bag 29, Clayton, VIC 3168; † Head, Department of GastrointestinaI Surgery, Monash Medical Centre, Melbourne, VIC. elizabeth.cooperATsouthernhealth.org.au In reply: Emmett and colleagues request more information and analysis about the mistaken use of stored serum samples. The pathology staff member correctly labelled the specimen of the patient being bled but did not follow the protocol in identifying that the patient was the same as on the request slip. It had not been highlighted that there were two patients with the same surname (but different first names) in the ward, and blood was collected from one patient with a request slip labelled for another. At specimen reception, the protocols were again not followed, as the staff did not check that the minimum identifiers on the specimen label and request form matched. It is recognised that violations of procedures are not root causes and are not directly manageable. The cause of the procedural violation must be managed.1 The collection and labelling of blood protocols were reviewed after this incident and found to be appropriate. The root-cause analysis identified that the protocols were not followed and that unacceptable “informal norms” had become practice in the collection and labelling of specimens. Staff training was examined and revised to ensure that staff were aware of the content of the protocols and that they followed them accordingly. All staff members were counselled about the importance of following correct procedures and the consequences of not doing so. Up to 1000 specimens are received at specimen reception each weekday. New “front end processing” technology is to be introduced at the end of 2003. This electronically scans the specimen and request slip to ensure details match. In the interim, in recognition that mislabelling will occur, all specimens relating to occupational exposures are collected at the time of the incident. Previously available results and serum stored in the laboratory are not relied on.

Elizabeth E Cooper · Stephen L Blamey

Nicola Magnavita Researcher, Italian Study Group on Hazardous Workers and Institute of Occupational Medicine, Catholic University School of Medicine, Largo Gemelli 8, Rome, 00168, Italy. magnavitaATrm.unicatt.it To the Editor: The article by Charles and colleagues1 is an interesting contribution to the development of Australian protocols for healthcare workers infected with hepatitis C virus (HCV). To date, most European countries have no national policy for HCV-infected healthcare workers, and existing guidelines are advisory in nature and poorly enforced. A panel of European and American experts recently failed to reach consensus on management of HCV-infected healthcare workers who perform exposure-prone procedures, and concluded that screening for HCV infection and restricting infected healthcare workers is not justified, based on current published data.2 Today, the effectiveness of guidelines relies solely on self-assessment of HCV status from healthcare workers. However, collaboration of healthcare workers might be problematic if management criteria are not defined, and workers’ rights are not guaranteed. Issues such as practice restriction, disclosure of serological status to patients, privacy and discrimination need to be resolved. Given the risk of HCV transmission from healthcare workers to patients is not clear, the burden of uncertainty rests entirely with healthcare workers. Because of the fear of discrimination, needlestick injuries may be under-reported, and infected workers may not seek diagnosis and treatment because they have greater legal protection if they can honestly say that they did not know their serological status.3 Moreover, the largely asymptomatic nature of HCV infection may leave healthcare workers unaware of their infective status. The results of Charles and colleagues suggest up to tenfold underreporting of occupational injuries with blood exposure in Australian healthcare workers.1 With this number of unreported exposures, there may be two or three new cases of HCV infection in healthcare workers in metropolitan hospitals in Melbourne each year — a figure similar to the prevalence of occupational HCV infection from notified injuries. Paradoxically, the prevalence of HCV infection in healthcare workers and the transmission risk for patients cannot be assessed without compulsory testing of healthcare workers, but without risk assessment there is no reason for this compulsory testing. Overcoming this Catch-22 with well-targeted epidemiological studies may help create broad consensus about policies for HCV-infected workers.

Nicola Magnavita

Patrick G P Charles,* M Lindsay Grayson,† Peter W Angus,‡ Joseph J Sasadeusz§ * Registrar, Department of Infectious Diseases, † Director, Infectious Diseases and Clinical Epidemiology, ‡ Director, Gastroenterology and Hepatology, Austin and Repatriation Medical Centre, Studley Road, Heidelberg, VIC 3084; § Director, and Infectious Diseases Physician, Victorian Infectious Diseases Service, Royal Melbourne Hospital, Melbourne, VIC. patrick.charlesATmh.org.au In reply: Magnavita correctly points out the need for a national policy on the management of healthcare workers who are either infected with or occupationally exposed to hepatitis C virus (HCV). Compulsory testing of healthcare workers is neither practical, logical nor fair in the current Australian healthcare environment. Firstly, most healthcare institutions do not currently have adequate staff health systems in place to ensure that staff are appropriately vaccinated against readily preventable diseases, such as hepatitis B, measles and varicella, let alone to test all staff for a disease such as hepatitis C, for which there is no vaccine. Secondly, awareness about important issues, such as healthcare worker transmission-risk assessment is embryonic (at best) in most institutions. The risk of HCV transmission from infected healthcare workers to their patients is generally considered to be extremely low, but probably depends on a number of factors, including the nature of the patient’s procedure and the healthcare worker’s injury and level of viraemia at the time. Simplistic legal opinions about such matters rarely help. Finally, we agree that the rights of healthcare workers are often neglected in this era of litigation-driven medicine. If these rights are not considered, and infected or exposed healthcare workers are simply excluded from all types of work without any appropriate risk assessment or compensation, then compliance with any form of postinjury testing is unlikely. However, rather than ignoring this important workplace issue, as we believe many Australian institutions currently do, a logical assessment of potential transmission risk is possible that is both fair to the patient and the infected worker. Stratification of healthcare workers according to their level of HCV viraemia and whether they are involved in exposure-prone procedures is a logical start — this we have attempted in our proposed guidelines.1 Some healthcare workers may avoid being tested so that they can have the protection of not knowing their serological status.2 However, recent studies suggesting the efficacy of early treatment of acute HCV infection3 mean that it will actually be in healthcare workers’ interest to know if they have recently acquired HCV infection — as long as they are treated in a manner that protects their health and workplace rights, while also protecting the rights of their patients.

Patrick G P Charles · M Lindsay Grayson · Peter W Angus · Joseph J Sasadeusz

Caroline M de Costa Obstetrician and Gynaecologist, Cairns Base Hospital, Cairns, QLD 4870. carolinedecAThotkey.net.au To the Editor: As a “grand multip” myself, I turned with interest to Humphrey’s recently published study of grand multiparity and pregnancy risk at Cairns Base Hospital.1 However, I cannot support his conclusions. Throughout the period of the study, most grand multiparous women giving birth at this hospital were actively managed in the third stage of labour with a regimen designed to prevent postpartum haemorrhage (intravenous ergometrine/oxytocin). Women of lesser parity were also given preventive therapy, but generally in lower doses and less consistently. This is a major confounding factor not addressed in Humphrey’s study. Clearly, an unknown, but probably significant, number of haemorrhages were prevented by this treatment. Given that 9.2% of grand multiparous women still had a postpartum haemorrhage, any prospective randomised controlled trial that allowed some of these women not to have active management of the third stage would be unethical. Numerous other studies have shown an association between grand multiparity and postpartum haemorrhage. 2-4 These include the study of Babinszki and colleagues, which limited study subjects to upper-class, private patients and thereby eliminated many confounding variables.4 Humphrey’s study does not report the severity of the postpartum haemorrhages that did occur; even a small number of life-threatening haemorrhages could justify continuing very active preventive measures in grand multiparous women. The incidences of anaemia and previous postpartum haemorrhage, not recorded in the study, would also be of interest; both are independent reasons to actively manage the third stage in women of any parity, and there are good reasons to believe that both may be more common in grand multiparous women. Humphrey also states that grand multiparous women did not have higher perinatal mortality rates or poorer maternal outcomes than women of lower parity. However, whenever possible throughout the period of the study, grand multiparous women were identified on booking into antenatal care and treated by senior obstetricians. Many had conditions such as diabetes, hypertension, anaemia and heart disease managed carefully to ensure as good a pregnancy outcome as possible. This focused obstetric care could well have counteracted any natural tendency of grand multiparous women towards poorer outcomes, and thus it is not possible to draw any conclusions about perinatal results from the data provided. What is clear from these data is that grand multiparous women in far north Queensland are often economically and socially disadvantaged compared with women of lower parity. This is a common finding in almost all studies of grand multiparity. 2,3,5 We should remember that these women are taking home a new baby to conditions that may already be quite compromised. They deserve the best obstetric care we can offer them, and we should be very cautious when reviewing protocols that we do not increase risks to these women or their babies.

Caroline M de Costa

Michael D Humphrey Director, Obstetrics and Gynaecology, Women's and Children's Health Service, King Edward Memorial Hospital, PO Box 134, Subiaco, WA 6904. Michael. HumphreyAThealth.wa.gov.au In reply: I thank de Costa for her interest in the debate about excessive medicalisation of normal childbirth in women with significant parity. The obstetric protocol manual of Cairns Base Hospital did not, at any time, discriminate in the details of management of the third stage of labour between grand multiparous and non-grand multiparous women. The statistical analysis in my report shows that, once confounding factors are accounted for, grand multiparous women who labour spontaneously are twice as likely as their less parous counterparts to have a spontaneous vaginal birth, with no statistically greater risk of a postpartum haemorrhage requiring transfusion.1 The purpose of multivariate analysis is to remove, as far as possible, the influences of the confounding factors that de Costa’s appraisal relies on. The recommendation from my study is not that grand multiparous women be ignored, but that, if labour occurs spontaneously at the end of an uncomplicated pregnancy, they be treated no differently to their less parous sisters in terms of venous cannulation and blood cross-matching. I am on record as strongly recommending sensible, routine oxytocin-based management of the third stage of labour in all pregnancies.2 In the end, the question is whether or not evidence wins out over an individual’s historically influenced clinical beliefs.

Michael D Humphrey

Neil H Cradick General Practitioner, PO Box 1127, Buderim, QLD 4556. cradwhitAToptusnet.com.au To the Editor: On opening the MJA Focus document “Prevention of cardiovascular disease: an evidence-based clinical aid”,1 I expected to find useful and contemporary guidelines for general practice. However, I was surprised to read two of the recommendations — for the use of antihypertensive and antiplatelet drugs in “low risk” patients (those without risk-associated clinical conditions or end-organ damage). The document recommends drug treatment only if systolic blood pressure is > 180 mmHg or diastolic blood pressure is > 100 mmHg in people under 60 years, or systolic blood pressure is > 160 mmHg in people over 60 years. While Fulcher et al1 volunteered that this was at odds with clinical practice, they supported the recommendations by stating that they were in accordance with published guidelines. The data for these conservative hypertension parameters were published nearly 10 years ago,2 or derived from textbooks,3 and are at odds with the 1999 WHO/ISH guidelines,4 and even more at odds with the excellent Joint National Committee (JNC 7) report.5 The latter publication recommends, after lifestyle recommendations, drug treatment for a blood pressure of 140–159/90–99 mmHg for those without end-organ damage. Furthermore, the focus document recommends primary prevention with aspirin in those with a calculated annual cardiovascular event risk > 3%. Hayden et al6 suggest antiplatelet treatment should be offered to those with a 5-year cardiovascular event risk > 3%, which equates to a > 0.6% annual risk. The benefit to harm ratio needs to be explained to the individual, and therapy should only be initiated once blood pressure is controlled. As an interested general practitioner and user of evidence-based guidelines, I usually check the funding of publications, and, with the heavy emphasis on the use of ACE inhibitors (in particular ramipril) and statins (which are produced by Aventis Pharma), it is difficult to rely on the evidence as presented. I would hope that independent bodies such as the National Prescribing Service or Australian Prescriber could take the pharmaceutical lead and produce desktop references with a more unbiased opinion on the latest collection of evidence that is shaking us up in primary care medicine. I would again refer readers to the excellent hypertension guidelines mentioned above4,5 (in particular the JNC 7 reference card available on the Internet at www.nhlbi.nih.gov/guidelines/hypertension/jnc7card.htm), and caution them to remain wary of easy-reference desktop items funded by pharmaceutical companies.

Neil H Cradick

Gregory R Fulcher,* John V Amerena,† Greg W Conner‡ * Director, Department of Diabetes, Endocrinology and Metabolic Medicine, Level 3, Main Block, Royal North Shore Hospital, St Leonards, NSW 2065; † Cardiologist, Department of Clinical and BioMedical Sciences, University of Melbourne, Melbourne, VIC; ‡ Vascular Physician, Cardiovascular Diagnostic Services, Sydney, NSW. gfulcherATmed.usyd.edu.au In reply: We thank Cradick for his critical review of our publication.1 His interesting points illustrate some of the reasons we were keen to publish this work. Firstly, the average practitioner is confused about which guidelines (both national and international) to follow. Cradick refers to the JNC 7 report and WHO/ISH guidelines (the committee is familiar with these), but which to follow? The committee decided a priori that Australian national guidelines, if they existed, would be referenced in preference to overseas ones. We mentioned that current Australian hypertension guidelines should be revised, and that clinical practice was probably at odds with these recommendations. Secondly, guidelines quickly become dated. Cradick cites the recommendation of Hayden et al that aspirin should be introduced if the annual risk of a cardiovascular event is > 0.6%.2 The latest Australian guidelines3 recommending a > 1% annual risk as the treatment threshold were published just after we received Cradick’s letter. The Hayden article is thus (for some patients) at odds with current local recommendations. Consistent with our approach, we will include the Australian recommendations in the first update of our document (July 2004). Cradick’s comments imply bias in the presentation of the data; yet he fails to substantiate this. Neither the clinical trials quoted (nearly all of which are funded by the pharmaceutical industry) nor the conclusions or inferences drawn have been challenged. He refers to the “heavy” emphasis on the use of ACE inhibitors and statins, implying that this is inappropriate. We would argue that, given the strength of the evidence, this emphasis is appropriate, and reflects current specialist practice in tertiary centres. Careful reading of the National Prescribing Service publications will show that “heavy” reference is made to the 4S,4 LIPID,5 CARE,6 and WOSCOPS7 studies (for example, NPS News 20 February 20028), as well as to the prescribing information for Lipitor (Pfizer), Pravachol (Bristol-Myers Squibb) and Zocor (Merck Sharp & Dohme). We must caution against “throwing the baby out with the bathwater”. The HPS, CARE, LIPID, WOSCOPS, HOPE, PROGRESS, 4S, CURE, and CREDO studies represent substantial and widely acclaimed clinical trials that have had a positive impact on clinical care. They have all been funded by the pharmaceutical industry. Finally, about 130 GPs in clinical practice had some input in compiling and formatting this document. The document was tested before publication in over 30 000 patients, and was peer reviewed by several clinicians. It is a pity that Cradick was not a participant in any of these processes.

Gregory R Fulcher · John V Amerena · Greg W Conner

Michael Thomsen,* Luis Vitetta,† Mathias Schmidt,‡ Avni Sali§ * Research Associate, † Director of Research, § Head, Graduate School of Integrative Medicine, Swinburne University, 9 Frederick Street, Hawthorne, VIC 3122; ‡ Head of Toxicology, Society of Nutritional Medicine and Dietetics, Harsewinkel, Germany. lvitettaATmedicine.swin.edu.au To the Editor: Gow and colleagues recently attributed fulminant hepatic failure in an Australian patient to an over-the-counter herbal product containing kava (Piper methysticum) that is extensively used in the community.1 We are not convinced that kava caused this patient’s liver failure. The hepatotoxicity could have been due to other unidentified contaminants of the herbal preparation or to chromium (from the mineral supplements also taken by the patient). We tested a sample of the product Kava 1800 Plus (Eagle Pharmaceuticals, Batch No. 10711) by thin layer chromatography (TLC) and high pressure liquid chromatography (HPLC) methods, and assessed it against defined standards. Kavalactones were identified and quantified by normal-phase HPLC. Kava was positively identified, with a content of about 47 mg of kavalactones per tablet (the product label specifies 60 mg per tablet). However, flavonoids from Scutellaria lateriflora, analysed by reverse-phase HPLC, could not be found. Passiflora incarnata, analysed by TLC, was also not found, with none of the typical flavonoid bands being detectable. The absence of two ingredients listed on the product label, P. incarnata and S. lateriflora, raises a new concern that there may be batch variations, as well as the possibility of unknown adulterants or contaminants being present. The possibilities of contaminants and batch variation were not excluded in the case report by Gow and colleagues.1 The Therapeutic Goods Administration established that Kava 1800 Plus did not contain common germander (Teucrium chaemaedrys), a known adulterant for S. lateriflora. However, Teucrium species are known to have hepatotoxic effects and there are about 100 species, any one of which could have been a contaminant. Gow et al refer to 68 international case reports of suspected hepatotoxicity with the use of kava.1 An independent analysis of these 68 cases concluded that only two were probable kava-associated hepatotoxicities.2 An incidence calculation from these case reports indicates that hepatotoxicity from kava occurs in 0.008 cases per million daily doses,2 which represents an extremely low risk of adverse reactions associated with kava. Recently, a study on the potential hepatotoxicity of kava found that the aqueous extract of kava does not affect results of liver function tests in rats. Extracts were administered in daily dosages of 200 or 500 mg of active kavalactones per kilogram of bodyweight for 2 or 4 weeks. Sera were assayed for four enzymes that are markers of liver toxicity, and liver homogenates were assayed for malondialdehyde formation, which indicates changes in lipid peroxidation. Kava did not elevate malondialdehyde or the enzymes alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase or lactate dehydrogenase. In fact, in certain instances, these enzymes were significantly reduced, suggesting a lack of toxic effect of kava on the liver.3 TLC analyses have shown that there is no qualitative difference between aqueous, acetone or ethanol extracts of kava.4 We propose, given that patients can consume other over-the-counter products such as chromium (shown to have hepatotoxicity at high concentrations),5 that all products consumed by patients must be evaluated before causality can be attributed to a particular herb. Reports of case records associated with hepatic failure due to kava require proper verification and documentation of all the findings.

Michael Thomsen · Luis Vitetta · Mathias Schmidt · Avni Sali