Letters

To the Editor: Forder et al conveyed that trials without allocation concealment have the potential to mislead.1 However, it is not true in any meaningful sense that “Without exception, allocation concealment is achievable in all randomised clinical trials. In contrast, it is not always possible to blind people to study treatments received.” Rather, “Masking may be defined as either the process (researchers not revealing treatment codes until the database is locked) or the result (complete ignorance of all trial participants as to which patients received which treatments). A masking claim indicates only the former . . . If masking is possible only some of the time, then clearly reference is being made to the result, and not the process. To be fair, then, one would have to ask if the result of allocation concealment is always possible . . . only the process of allocation concealment, but not its result, can be ensured.”2 Forder et al also state that certain methods (including sealed envelopes) are considered to be adequate concealment methods. Sadly, this is true, but only if the emphasis is on the word “considered”, because sealed envelopes are both imperfect at preventing direct observation of future allocations and useless at preventing the prediction of future allocations, even without direct observation. Because the extent of prediction depends on the specific restrictions used on the randomisation,3 allocation concealment is not even a binary phenomenon, and so to truly assess allocation concealment in a given trial, one must ask how much prediction is possible in that trial. Allocation concealment is perfect if no observation or prediction is possible, and only partially effective if some prediction is possible. Many trials use randomised blocks, and smaller block sizes tend to allow for substantial prediction.3-5 So, while methods aimed only at preventing the direct observations of future allocations may be considered to be adequate, it is clear that in reality they are not. That the authors failed to use this opportunity to set the record straight indicates their implicit agreement with the incorrect statement that methods aimed only at preventing the direct observations of future allocations are not only considered adequate, but actually are adequate. Pretending that allocation concealment is binary, and hence that it suffices to use methods aimed only at preventing the direct observations of future allocations, represents ignorance that may be bliss, but certainly is not harmless.

Vance W Berger

To the Editor: In their article on controlled trials, Forder et al1 described the trial by Karlowski et al on vitamin C and the common cold2 as an example of how patients’ or investigators’ preconceptions about the value of the treatment may affect a trial’s results. However, their presentation of this trial is misleading in two respects. Firstly, the Karlowski et al trial was reanalysed and the “placebo-effect explanation” of the original authors was shown to be erroneous.3 For example, their subgroup analysis of “blinded” and “non-blinded” participants excluded 42% of all episodes of colds, even though the subgroups were presented as complementary; numerous further problems are detailed elsewhere.3 Thus, the trial by Karlowski and colleagues cannot be seen as an example of the placebo effect in action. The concept of large and omnipresent placebo effects can be traced back to an early article by Beecher, who chose “15 illustrative studies” covering such conditions as, “severe postoperative wound pain, cough, headache, seasickness, etc.”4 Beecher calculated that the “average placebo-effect” was 35.2% (SE, ± 2.2%). However, these studies did not use a control group. The comparison was “before–after”, which is affected by the regression to the mean phenomenon as most of these conditions are self-limiting. Thus, Beecher’s studies did not measure the “effect” of placebo. A recent meta-analysis of 114 trials comparing a placebo group with a no-treatment group found no evidence of placebo effect on binary outcomes, and only a rather small effect on pain, thus disproving Beecher’s notion of great and universal placebo-effects.5 This empirical evidence was disregarded by Forder and colleagues. Although there are reasons to use placebo whenever practicable, the bias caused by the absence of a placebo control should not be exaggerated, and the “placebo effect” should also not be misused to support investigators’ own preconceptions.3 Secondly, the trial by Karlowski et al was focused on the effect of vitamin C on the common cold,2 and thus the “placebo effect explanation” in this particularly influential trial is crucial to the biological question. A recent meta-analysis of 55 placebo-controlled trials found that regular vitamin C supplementation had no effect on the incidence of colds in the general population (relative risk [RR], 0.98; 95% CI, 0.95–1.00), but reduced the incidence of colds in people exposed to substantial physical or cold stress (RR, 0.50; 95% CI, 0.38–0.66).6 Also, regular vitamin C intake reduced the duration of colds in adults by 8% (95% CI, 3%–13%) and in children by 13.5% (95% CI, 5%–21%). Although further studies are needed to evaluate the practical significance of these findings, it is evident that the interpretation by Karlowski and colleagues that the effect of vitamin C on the common cold may be explained by the break in the double blind2 is false and should not be reiterated.

Harri Hemilä

Peta M Forder,* Val J Gebski,† Anthony C Keech‡ * Statistician, † Principal Research Fellow, ‡ Deputy Director, NHMRC Clinical Trials Centre, University of Sydney, Locked Bag 77, Camperdown, NSW 1450. enquiryATctc.usyd.edu.au In reply: Allocation concealment refers to ignorance of future treatment assignment before randomisation whereas masking or blinding is most commonly used to refer to the concealment of treatment assignment after randomisation.1 There are two criteria for successful concealment of allocation: (i) physical concealment of the process of random assignment to treatment, and (ii) concealment of any pattern of consecutive assignments. Successful concealment of the process must prevent unauthorised access to randomisation lists, envelopes or algorithms; the best way is to use a centralised or remote service for randomisation, whereby an independent party other than the clinician or investigator accesses a secure sequence list or a secure computer system to generate the next allocation.2,3 Successful concealment of the pattern of random assignments prevents investigators from predicting a future treatment assignment on the basis of pattern recognition of allocations to date. Identifying a pattern of previous allocations can occur in open-label trials, in which all parties are aware of allocated treatments after randomisation, or if the blinding of patients and investigators has been compromised. The likely success of concealing the allocation process can reasonably be judged by its description in most trial reports (usually found in the Methods section). However, it is usually more difficult to assess the likelihood that investigators could have predicted future allocations. Unsuccessful concealment of treatment assignment after randomisation (masking or blinding) should be detailed in the trial report. In circumstances where the blinding has been substantially compromised, exploring the results of treatment separately among participants who were unblinded and those who remained blinded, should be considered, although these are no longer randomised comparisons. In the study by Karlowski et al,4 the placebo did not match the active treatment in taste, which alerted the investigators to the likely occurrence of significant unblinding within the study. To their credit, the investigators sought to quantify the extent of unblinding by means of a questionnaire at study close-out, and reported their findings by results of these responses. The particular grouping of responses, however, has been the subject of some discussion,5,6 and while the interpretation of a possible placebo effect has been challenged, it has not necessarily been disproven. (The absence of a placebo effect could be proven only if information concerning perceived benefits of vitamin C related more to cold frequency than cold symptoms. Biologically, it is far more plausible for a placebo effect to result in fewer cold symptoms reported than fewer colds reported.) This trial highlights the impact of compromised blinding in the reporting of trial results, emphasising the importance of maintaining adequate blinding for reliable and unbiased trial results. Good quality reporting of trials, in accordance with the CONSORT statement,7 includes describing the processes in enough detail to assure readers that any pattern of randomisation is not predictable. Authors should report issues relating to allocation concealment, blinding (where appropriate) and randomisation sufficiently to convey the message that these essential trial principles were successfully achieved.8

Peta M Forder · Val J Gebski · Anthony C Keech

H Patrick McNeil,* Michael C Grimm† * Associate Dean (Medical Education), South Western Sydney Clinical School, † Associate Professor of Medicine, St George Clinical School, Faculty of Medicine, University of NSW, Sydney, NSW 2052. p.mcneilATunsw.edu.au To the Editor: We read with interest the article by Koczwara and colleagues proposing a national exit examination for all Australian medical school graduates.1 It is refreshing to see interest in educational outcomes, a distinctly different trend from earlier reforms that shifted curricular focus from content to the learning process, exemplified by problem-based learning (PBL). Although the early process-focused programs were based on sound pedagogy current at their time, their educational outcomes have been relatively disappointing, with marginal or no demonstrable improvements in knowledge structures, clinical skills, or generic capabilities such as self-direction.2 Rather than an indictment of PBL, the results may reflect what was missing in those programs: explicit focus on educational outcomes, alignment of assessments with outcomes, and attention to the learning environment. There is widespread agreement on the outcomes desired by medical schools. They include teamwork, effective communication, critical evaluation and reflective practice, as well as more traditional outcomes.3 Unfortunately, assessment methods have been slow to match curricular reforms, as these outcomes require new approaches, such as group and assignment work, peer assessment and portfolio examination, which are only now emerging in Australia.4 A national exit examination for Australian graduates is unlikely to adequately measure this range of outcomes. While Koczwara and colleagues recognise that a national examination “might need to include a clinical component” and “would necessarily entail the explicit statement of professional values and expectations”, they support a multiple-choice question examination, suggesting such performance “can correlate well with clinical skills and future performance in multiple disciplines”.1 While this might “complement rather than replace” other medical school assessments, the message sent by its failure to address personal and professional attributes would be invidious. In recognition of the limitations of multiple-choice questions, national examinations in North America now include a clinical component.5 This has major resource implications and, like all high-stakes assessments, uses relatively reliable, but much less valid measures — standardised or simulated clinical encounters. This is at odds with current initiatives in medical schools, which are moving to clinical assessments with higher face validity, such as the mini-CEX (mini-clinical examination exercise).6 It would be near impossible to adequately measure generic outcomes, such as teamwork, communication and reflection, in a single national examination. Koczwara et al recognise that insufficient attention has been paid to ensuring that achievement of educational outcomes is embedded in reform of medical curricula. Their solution is overly simple for a highly complex set of issues.

H Patrick McNeil · Michael C Grimm

Christopher Lawson-Smith Surgeon, and Surgical Examiner, Australian Medical Council, 1/10 McCourt Street, Leederville, WA 6007. lawsmithATbigpond.net.au To the Editor: Foreign medical graduates sitting for the Australian Medical Council (AMC) examinations are expected to achieve a standard comparable to that of Australian medical students. If we do not measure the level of knowledge and problem-solving ability nationally, there is no reasonable basis for presuming that the AMC examination is fair. A uniform examination-based assessment should be passed by all potential medical practitioners before registration in Australia. I would be in favour of a national exit examination.1

Christopher Lawson-Smith

To the Editor: The recent article by Koczwara and colleagues proposing a national exit examination for medical students1 prompted me to recall a 1970 trial of a national examination in surgery.2 Seven of the then eight medical schools participated. Interstate differences were wide for some questions; separate analyses of the 15 teaching hospitals showed variation to be even wider within a university than between universities. Do local differences still undermine the validity of a national examination? It is still uncertain what is actually tested by questions on paper. Context-free, standardised questions and answers assume clinical teaching and practice are standardised. However, clinical teachers writing examination items know well that many colleagues choose the “wrong” answer. Consensus may be imposed on those who differ. Teachers then forget their disparity, but expect candidates to choose only one “true” answer! Clinical performance is interactive, multifaceted, situation-specific and value-laden. Complex judgement and decision-making cannot be measured by ticking predetermined boxes. Clinical experts develop personal subsets of specific evidence, and seek different data for diagnosis and management. But separate, context-free tasks, as in an objective structured clinical examination (OSCE), naively assume they do not.3 OSCE even standardises scoring; examiners become recorders rather than assessors. Reductionist standardisation reflects a pseudoscientific attempt to apply objectivity, consistency and precision to complex human interactions around incomplete evidence and uncertainty, approximations, judgements, trade-offs and locally-determined decisions.4 Internal consistency of measuring instruments does not confer external validity in real world clinical practice. The inexorable growth of medical knowledge and technological opportunities continuously expands “what every doctor should know”. Medical learning today embraces a mix of science-based, problem-based and work-based learning experiences, with community-based experiences5 increasingly included. In their recent article, Koczwara and colleagues identified gaps in oncology education,2 a field ranging from molecular processes to euthanasia. Is oncology managed and taught consistently across different medical schools and hospitals across Australia? Which facets would you test in a national exit examination?6 Clinical performance today includes patient/person management, case management, health system management and self-management. Clinicians can judge student performance consistently. However, formal clinical examinations lack the range of cases and open-ended time that allow examiners to observe all the patient-care skills espoused by today’s curricula.7 Assessment of performance in case management and procedural skills within hospital practice can be conducted simpy by paired examiners.8

Ken Cox

In reply: We appreciate the insightful responses to our proposal.1 Lawson-Smith alludes to one of the most significant justifications for a national examination — fairness. One cannot expect foreign medical graduates to attain a standard comparable with that of Australian graduates if we do not measure this standard. We propose that we owe fairness not only to foreign graduates coming to Australia, but also to Australian medical students who have a right to confidently expect an education that will lead to similar knowledge, skills and attitudes, irrespective of which university they choose. And finally, we owe fairness to society, which would also expect the same standards of graduates irrespective of where they come from. Unless we consider what are acceptable standards, we operate within an environment where standards of outcome differ from place to place and as we do not measure outcomes uniformly, we do not know how they differ nor have a system to address potential deficiencies. McNeil and Grimm point out that medical education has focused less on outcomes and more on process, and raise concerns that assessment methods lag in the sophistication necessary to assess outcomes. We wonder whether the reason for this lack of sophistication lies in the relative lack of interest in this field, and also in the lack of agreement on what constitutes acceptable outcomes. The process of outcome assessment is indeed complex, and the first step is national consensus on appropriate outcomes to be uniformly achieved. McNeil and Grimm also point out that some of the desirable outcomes, such as teamwork, effective communication, critical evaluation and reflective practice, may be harder to test than medical knowledge. While this is certainly the case, reliable assessment methods do exist, such as the Moral Judgment Interview2 and Rest’s Defining Issues test.3 The mini-CEX (mini-clinical examination exercise) that McNeil and Grimm refer to, allows testing of judgement, professionalism, communication, organisation and efficiency.4 Cox questions whether a written examination can test the complex judgement and decision-making process that is better tested in the clinical setting by experienced clinicians. We propose that the national examination is not meant to replace clinician-based assessments and ongoing learning and feedback. Furthermore, a national examination does not need to be conducted only in the written form. Specialty exams already conducted nationally incorporate a clinical element and are conducted in multiple locations. The main objective of a national examination is to ensure the comparison of outcomes against agreed acceptable national standards. This objective should not impose specific limitations on the structure of the examination. Cox warns of the risk of “reductionist standardisation” and asks whether oncology is taught consistently across different medical schools in Australia. We acknowledge that uncertainty is ever present in medical decision-making, but remain hopeful that there exist core knowledge, skills and attitudes that patients can expect and that can provide a foundation for national standards. Oncology is not taught consistently across different medical schools in Australia today. While its mode of delivery may differ, we propose that its outcomes should not. And we hope that agreement on national outcome standards and a national process of assessment of these outcomes will be a first step in achieving that objective.

Bogda Koczwara

To the Editor: GP-Psych Support is a national advice service on mental health management that links general practitioners (GPs) with psychiatrists by phone, fax or email within 24 hours. The service is federally funded through the Better Outcomes in Mental Health Care Initiative, and began operating in March 2004. Over the first 6 months of operation of the phone/fax arm, there were 726 case discussions between GPs and psychiatrists. A third of the GPs were rurally located, and 17% used the service twice or more. Most GPs (94%) accessed the service through the 1800 freecall number, rather than by fax. Three-quarters identified no other suitable, accessible source of urgent psychiatric advice. Two-thirds of cases involved female patients, and 80% were aged 19–65 years. In only 14% of cases was another mental health service provider involved (a psychiatrist in 2%). The primary diagnoses are shown in the Box. The most common topic discussed was medication (77%), with less demand for discussions of general management principles (12%) or diagnosis (7%). However, cases were discussed broadly, and referral to another mental-health care provider was recommended in a third of cases, including to a psychiatrist in 8% of cases. An evaluation survey with 450 randomly selected GP service users received 129 replies. Although the survey response was low (29%), the feedback was very positive, with over 99% of respondents indicating that they would consider using the service again. Over 95% of respondents reported satisfaction with the service in terms of ease of use, helpfulness of advice and ease of interaction with the psychiatrist. Compared with other sources of mental-health care advice, over 85% rated GP-Psych Support as better in regard to accessibility, reliability and the appropriateness of the advice given. Over 70% stated that contact with the service had increased their knowledge about the management of mental disorders and their confidence in managing mental health problems, and had improved the quality of care they provided to their patients. Fifty-three per cent of GPs reported an increased willingness to manage more complex mental health problems. From these data, it appears that psychiatrist telephone support has been well received and adds significantly to the resources available to GPs working in mental health care. The 10 most common primary diagnoses among 726 patients discussed in GP-Psych Support Diagnosis Percentage of patients Depression 45.9% Mixed anxiety and depression 12.6% Chronic psychotic disorder 8.8% Anxiety disorder 8.6% Acute psychotic episode 5.2% Bipolar disorder 5.1% Organic or cognitive disorder 3.6% Drug or alcohol use disorder 2.5% Somatoform disorder 1.5% Eating disorder 1.2%

Stephanie E Bradstock FRANZCP · Andrew J Wilson FRANZCP · Matthew J Cullen FRANZCP · Katie L Barwell RN

Ashwin Swaminathan,* Iain B Gosbell,† Nicholas A Zwar,‡Mark W Douglas§ * Infectious Diseases Registrar, † Director and Associate Professor, § Infectious Diseases Physician, Department of Microbiology and Infectious Diseases, Liverpool Hospital, South Western Area Pathology Service, Locked Bag 7090, Liverpool, NSW 1871; ‡ Director and Professor of General Practice, Sydney South West Area Health Service General Practice Unit, Fairfield Hospital, Sydney, NSW. Iain. GosbellATswsahs.nsw.gov.au To the Editor: Infestation with the sandflea Tunga penetrans, or “chigoe flea”, is rarely encountered in Australia, but has been noted in children recently arrived from Central–East Africa. There have been only two previous Australian case reports of this parasitic infection, both in adult travellers returning from Africa.1,2 Several families who had been living in crowded refugee camps in Tanzania underwent routine screening for infection within 2 weeks of arrival in Australia. Four of 14 children examined had cutaneous lesions on their feet — mainly on the toes, nail beds and interdigital spaces (Box). These lesions were papular, less than 1 cm in diameter, pale yellow with dark centres, and were variably painful and/or itchy. Chronic, adjacent skin and nail bed changes were evident, as were small, loosely attached seed-like objects. Papules could be lifted with a sterile needle, leaving a small, non-bleeding cavity. Tunga penetrans, with numerous attached eggs, was identified by microscopy. Tungiasis occurs when an impregnated female sandflea burrows into the unprotected skin of a warm-blooded host. There is a predilection for the feet, although the perineum, buttocks and arms may also be infected.3 The head of the sandflea breaches the upper dermis to feed on blood vessels, while the abdomen traverses the epidermis, with its posterior components (anus, genital opening and respiratory spiracle) reaching the surface, forming a papule. Over several weeks, the flea releases hundreds of eggs before dying. After hatching, the larvae thrive in dust, soil and sand; they are found on beaches and in animal stockyards of tropical countries.3,4 Infection of pigs and other livestock, the usual host reservoirs, has led to significant problems in the livestock industry.1,4 Apart from pruritis and pain caused by local inflammation, morbidity results from ulceration and secondary bacterial infection, including tetanus and gas gangrene.1,2,4 Fleas can be removed using a sterile needle and forceps, and secondary complications prevented with tetanus prophylaxis, and antibiotics as appropriate. Successful outcomes with antiparasitic agents, such as ivermectin and thiabendazole, have also been recently reported.5 Tungiasis is indigenous to Latin America and the Caribbean, but was introduced to Africa, where it is now endemic, and to parts of South Asia.4 Given the current influx of African refugees into Australia, including the tropical north, the obvious concern is whether Tunga penetrans could establish itself here. We encourage medical practitioners dealing with newly arrived African refugees to examine for tungiasis in their screening evaluation. If, as suspected, this condition is prevalent, national infection control guidelines aimed at preventing establishment of the disease in Australia may be needed. Lesions caused by Tunga penetrans, the “chigoe” sandflea A characteristic Tunga penetrans lesion (thick arrow), with pale-yellow papule and dark centre, and a less obvious lesion (thin arrow) with surrounding chronic skin changes and multiple, loosely attached eggs.

Ashwin Swaminathan · Iain B Gosbell · Nicholas A Zwar · Mark W Douglas

Raymond F Raper,* Elizabeth Fugaccia,† Yahya Shehabi‡ * Board Member, † Member, ‡ Chairman, NSW Regional Committee, Joint Faculty of Intensive Care Medicine, 117 Alexander Street, Crows Nest, NSW 2065. yshehabiATozemail.com.au To the Editor: We are writing in response to the pejorative, unhelpful and somewhat misleading editorial “The shortage of kidneys for transplantation in Australia”.1 There are many possible reasons for lower organ donation rates in Australia. Several Australian initiatives have led the way in reducing the incidence of severe, traumatic brain injuries. These include the compulsory wearing of seat belts and helmets, random breath testing, and a zero blood alcohol limit for inexperienced drivers. Intensive care medicine is better structured and organised in Australia and New Zealand than in most of the countries cited by Mathew et al, with higher organ donation rates. Intensive care outcomes in Australia are world-leading. So the donor rate may be lower because patient outcomes are better. A comprehensive chart audit of donor potential in New South Wales carried out under the supervision of the Organ Donation Network NSW/ACT, identified very few missed donors (T Wills, Manager, Organ Donation Network NSW/ACT, personal communication), and a Victorian audit has suggested that the donor pool may be much lower in Australia than previously estimated.2 Similarly, organ donation rates will appropriately vary considerably among hospitals. To improve outcomes, critically ill patients are transported to centres with specific experience and expertise, resulting in a preponderance of potential donors in hospitals with trauma and neurosurgical services compared with hospitals lacking these. The intensive care community supports these life-saving initiatives, notwithstanding the effect they may have on organ donation potential. In fact, the principal “barrier” to organ donation in Australia appears to be the consent rate. From 2000 to 2004, 44% of families declined organ donation when faced with an actual rather than a hypothetical request.3 The intensive care community represents the interests of critically ill patients and their families. We will continue our best endeavours to improve both the survival and quality of life of patients suffering devastating brain injuries (our performance standard). When all brain function ceases, despite our best efforts, we will continue to facilitate organ donation, in discussion with the family, and in consideration of the patient’s known or projected wish. The intensive care community has led the way in developing organ donation-related practice guidelines4 and in related education. Australian Donor Awareness Program — Training (ADAPT) workshops are now a compulsory component of Fellowship of the Joint Faculty of Intensive Care Medicine training. The editorial implication of poor performance and lack of commitment is inaccurate and offensive.

Raymond F Raper · Elizabeth Fugaccia · Yahya Shehabi

Timothy H Mathew,* Randall J Faull,† Paul L Snelling‡ * Medical Director, Kidney Health Australia, GPO Box 9993, Adelaide, SA 5001; † Nephrologist, Royal Adelaide Hospital, Adelaide, SA; ‡ Nephrologist, Royal Prince Alfred Hospital, Camperdown, NSW. tim.mathewATkidney.org.au In reply: We regret that Raper et al have misinterpreted our editorial on the shortage of donor kidneys in Australia.1 We are particularly concerned and indeed mystified by their last statement, where they state that we implied that “poor performance and . . . lack of commitment” were to blame. This was in no manner our message. Rather, we sought to emphasise that all possibilities to optimise local donation rates should be explored, so desperate patients seeking grafts from potentially dangerous overseas sources need not expose themselves to serious potential morbidity and mortality. We consider that our intensive care colleagues perform superbly under the most difficult of circumstances when managing potential organ donation. We understand they are often unsupported with managing potential donations while they must at the same time deal with the grieving family, and the immediate demands of treating other seriously ill patients. We simply suggest that problems within the system (for example, differences between states in the number of intensive care beds per head of population or in the provision of specifically funded donor coordinators) that might hinder increasing organ donation should be carefully examined. We believe it is unhelpful to suggest that the South Australian experience should simply remain unexplained. In the article by Opdam and Silvester (cited by Raper et al), of 112 potential donors, 46 were considered medically suitable unrealised potential donors, and their estimated maximal potential donor rate was 30 per million,2 remarkably similar to the rates seen in Spain (the country with the highest organ donor procurement rate) and South Australia. We agree with the conclusions of Opdam et al that “an increase in the organ donation rate may be possible through increasing consent and the identification and support of potential donors”. As they also state, this would require substantial changes in clinical practice, with resource and ethical complications We did not mean to offend our intensivist colleagues. We merely suggest that we all need to assess the systems in which donation occurs and attempt to improve donation rates, for the sake of the many desperate people awaiting organ transplants in this country.

Timothy H Mathew · Randall J Faull · Paul L Snelling

To the Editor: The article by Senanayake and Ferson1 on detention for tuberculosis included two case histories. In the case of “Patient 2”, we believe the details published were misleading and unnecessarily disclosed potentially identifying information. This case was presented in sufficient detail to allow identification of “Patient 2” by including date of admission, personal details, city of admission and details regarding his past history of alcohol addiction and attempts to self-discharge. All this information was not necessary and did not take into account all the relevant medical complicating factors. We were surprised that, as the primary treating team involved in this patient’s care over the inpatient stay of 3 months and the following 6-month outpatient follow-up, we were not informed of the authors’ plans for publication, nor requested to comment on the facts of the report. We were also surprised to read details of the article in The Sydney Morning Herald on the day of the article’s publication in the Journal.2 An individual’s right to privacy is a fundamental human right. It is unfortunate that the careful consideration that had been given to his detention was not extended to publishing his case details. As Senanayake and Ferson point out, “Patient 2” recognised that his human rights were being “infringed” because he was being held in a “jail cell”, which was a temporarily modified isolation room in a public hospital under 24-hour guard. In law, information provided to a medical practitioner by a patient becomes subject to a statutory duty to protect the patient’s privacy and a common-law duty of confidence is owed by the treating medical practitioner to the patient. The NHMRC Guidelines approved under Section 95A of the Privacy Act 1998 indicate that, when a patient history is published, an important principle is that a patient may not be identified or held up to ridicule.3 Furthermore, public access to medical journals on the Internet has allowed increased availability of such reports to the general public and increases the chance of family members and others identifying individuals. The principle of protecting patient privacy has previously been respected by the Journal. An article published in 1994 reported a 1979 outbreak of tuberculosis in medical students who attended an autopsy of an immunosuppressed patient with unsuspected active tuberculosis.4 A report of the incident was not published contemporaneously, mainly to protect the privacy of the students involved. As the report by Senanayake and Ferson suggests, our patient was socially disadvantaged, and several aspects of his behaviour were probably attributable to a Jarisch–Herxheimer reaction in conjunction with acute alcohol withdrawal. Although we recognise the need to serve the public interest in health service management activities, this must be balanced against the requirement for patient privacy regardless of social class. We suggest that the publisher has an obligation to ensure that patient consent is obtained, and that the primary treating team has been involved in the review of case details so that misleading and potentially identifying information is not released inappropriately.

Medical Registrar · Respiratory Physician

Ruth M Armstrong,* Martin B Van Der Weyden† * Deputy Editor, † Editor, The Medical Journal of Australia, Locked Bag 3030, Strawberry Hills, NSW 2012. medjaustATampco.com.au In reply: We concur with the principle that patient privacy should be protected in case reports. As recommended by the International Committee of Medical Journal Editors,1 measures in place at the Journal include asking authors to obtain patient permission for publication where possible, and directing authors to remove potentially identifying patient information. The article in question was not a case report. Case details were given to exemplify the circumstances that might lead to detention of a patient for public health reasons in New South Wales, and dates and some details were retained to illustrate the temporal flow of the story. We agree that this may have made the patient identifiable, if not to the general public, to himself or his family. We regret this editorial lapse, and have reworded the patient details in the electronic version of the article.2 Although chastened by our anonymous colleagues’ observations, we are somewhat puzzled as to why, given their concern about their patient’s privacy, they are determined to draw further attention to the exposing details. We assume that, in this case (as in the cases of detaining patients for treatment of tuberculosis), public interest prevails. We also agree that case reports of detailed clinical histories require the input of the primary treating team, but the question of authorship should be determined by the involved parties, not the Journal. Be that as it may, as the article was a public health report (and thus not meant to be a detailed clinical exposition), we do not believe that the input of the treating team was required.

Ruth M Armstrong · Martin B Van Der Weyden

Jim Stockigt Senior Endocrinologist, Alfred Hospital, Melbourne; and Professor of Medicine, Monash University, VIC 3004. jrsATnetspace.net.au To the Editor: In May 2004, Sigma, the sole Australian supplier of l-thyroxine sodium, instructed pharmacists that thyroxine tablets should be stored refrigerated, both in pharmacies and after dispensing. Thyroxine bottles now carry explicit labels: “keep refrigerated” or “refrigerate at all times”. This instruction seems to have been accepted by health professionals, but patient-support groups immediately questioned the refrigeration directive. In response, Sigma conceded that thyroxine tablets can be stored at room temperature (< 25°C) for up to 4 weeks, with refrigeration still the preferred option. There are major unresolved issues about the potency, stability and bioavailability of various thyroxine preparations that are marketed competitively in the United States.1 With a single supplier in Australia, we can avoid between-preparation variations, provided that stability and consistency are maintained. The instruction to refrigerate thyroxine tablets seems to be uniquely Australian. None of my co-authors of the website <www.thyroidmanager.org>2 is aware of a refrigeration directive in any other country. The local instruction seems to have followed interaction between the Therapeutic Goods Administration and the manufacturer, so that unopened bottles could be marketed with a longer shelf life. Is the rest of the world missing out on something important? Is there something peculiar about the Australian formulation that makes it unstable at room temperature? Could this directive be without firm basis, or even dangerous? There is currently no evidence on whether thyroxine in already-opened, unsealed bottles is more or less stable at 4°C than at room temperature, but the need to keep the tablets dry has been widely emphasised.3 Consider the condensation that will occur during 200 daily openings of a refrigerated glass bottle, whatever it contains. If damp tablets lose potency, this would lead to apparent under-treatment. In the months since refrigerated storage was recommended in Australia, preliminary observations suggest that apparent under-dosage (ie, unexpected rises in serum TSH) may indeed occur in previously compliant patients (personal observation). If dosage were increased, the adjustment could result in over-treatment after a change to a fresh preparation. Thyroxine has a narrow therapeutic window, and excessive dosage can have serious effects, especially if there is associated cardiac ischaemia. While refrigeration of sealed bottles of thyroxine might extend the shelf life, the instruction to refrigerate unsealed bottles seems ill-advised. When an existing formulation is modified, it is generally the obligation of a manufacturer to demonstrate safety. The stability of tablets in sealed bottles and those in current use are quite separate issues. To establish how tablets in current use are influenced by refrigeration, it is necessary to measure the thyroxine content of remaining tablets from bottles of 200, opened and used daily for up to 6 months. Without such data, it is preferable to instruct patients not to store currently used bottles of thyroxine at refrigerator temperature.

Jim Stockigt

Ovais Siddiqui Regulatory and Medical Manager, Sigma Pharmaceuticals, Locked Bag 268, South Croydon, VIC 3136. ovais.siddiquiATsignet.com.au In reply: Sigma Australia acquired Oroxine (thyroxine sodium) from the original manufacturer in 1999, and launched Eutroxsig, an identical product, in 2002. During 2002–03, as a result of advances in analytical technology for some pharmaceutical products, product specifications, including shelf life and storage conditions, were updated, so that the product’s quality, safety and efficacy could be maximised or maintained throughout the claimed shelf life. For Oroxine and Eutroxsig, the new stability data showed a loss of up to 10% of thyroxine sodium in the first 6 months when stored below 25°C, with some plateauing thereafter. As an interim measure, Sigma, in agreement with the Therapeutic Goods Administration (TGA), decided to immediately reduce the shelf life from 24 to 12 months (“store below 25°C”) and set the lower release to 98.0% (up from 92.5%), while investigating reasons behind the loss in potency. The manufacturing process was confirmed to consistently yield tablets with very reproducible chemical and physical attributes in accordance with the release criteria. During manufacturing, however, about 2% of the thyroxine sodium is lost, with a corresponding similar increase in degradants. To limit the degradants responsible for the reduction in potency of thyroxine at room temperature, it was agreed with the TGA that thyroxine should be stored at 2°– 8°C (“Refrigerate. Do not freeze”), based on good stability data generated at this temperature. Consumer Medicine Information (CMI) and Product Information (PI) were updated in May/June 2004 to reflect this change. The new stability studies support the storage of Oroxine and Eutroxsig in the refrigerator; however, repeated in-use handling may result in an increase in condensation and microbial contamination. This may lead to changes in the physical characteristics of these products, including the growth of mould. There may be a further increase in condensation if the lid is not tightly closed. One possible solution is for patients to place up to 4 weeks’ supply of tablets in a spare, previously used, Oroxine or Eutroxsig amber-coloured bottle and store out of the fridge (below 25°C) for current use, while keeping the remaining stock in the fridge. Sigma is looking at options to improve the packaging so that the above problems are minimised or eliminated. Oroxine and Eutroxsig, manufactured by Sigma, are sold in Australia only. Sigma does not have access to formulation details, stability results and justification for the storage conditions used in other countries; therefore, we are unable to comment on such issues. As an Australian company, we are obliged to follow the regulatory guidelines of the Therapeutic Goods Act 1989 (Cwlth). Sigma recommends that the label instructions regarding storage conditions after opening be strictly followed to maximise the quality, safety and efficacy of the product.

Ovais Siddiqui

Christopher M MacIsaac,* Mark A Page,† Beverley-Ann Biggs,‡ Kumar Visvanathan§ * Associate Intensivist, † Registrar, ‡ Associate Professor, The Royal Melbourne Hospital, Grattan Street, Melbourne, VIC 3050; § Senior Research Fellow, Murdoch Children’s Research Institute, Melbourne, VIC. Christopher.macisaacATmh.org.au To the Editor: We report a recent case of toxic shock syndrome associated with menstruation which illustrates that this syndrome still occurs, even when tampons are used appropriately. A potential diagnostic test for the syndrome is also discussed. An 18-year-old woman presented with a 1-day history of fever, chills and severe back pain, with no other focal symptoms. On examination, she was febrile with a blood pressure of 75/40 mmHg, and had begun vomiting. She was treated empirically with intravenous ceftriaxone and flucloxacillin and resuscitated with intravenous fluids. Over several hours, the back pain resolved, and a widespread erythrodermic rash developed, centred mainly on the trunk. Further questioning revealed that the patient had removed a tampon shortly before presentation, as she had just ceased menstruating. Renal ultrasound examination, chest x-ray and blood cultures were non-diagnostic. She was treated with intravenous antibiotics for 4 days and discharged home with a further 10-day course of oral amoxycillin and clavulanic acid. At outpatient follow-up 3 weeks after admission, she reported desquamation of the skin of her palms and soles. Toxic shock syndrome was first described in 1978,1 and a strong association with Staphylococcus aureus, menstruation and tampon use was established in 1980.2 Toxic shock syndrome toxin-1 (TSST-1), a protein secreted by S. aureus, was the first of many toxins associated with the syndrome to be identified. The term “superantigen” was adopted to describe the ability of these toxins to cause a remarkable expansion of T lymphocytes displaying specific β chain variable regions of the T-cell antigen receptor. Superantigens bypass normal antigen presentation and can stimulate over 20% of all T cells, whereas a conventional antigen stimulates only in the order of 1 in 10 000 T cells. The signature feature of superantigen activity is the expansion of lymphocyte populations bearing the particular Vβ chains that bind the superantigen. In the case of TSST-1, this is Vβ2.3 Our patient consented to blood being sampled to investigate the Vβ profile of her T cells at follow-up. This investigation was part of a broader study on superantigens in sepsis that was approved by the Ethics Committee of the Royal Melbourne Hospital. The blood was stained with monoclonal antibodies against 24 Vβ families4 and analysed by flow cytometry. This showed a massive expansion of Vβ2 cells, which accounted for 28% of all CD4 lymphocytes (Box). Currently, there is no diagnostic test for toxic shock syndrome. Toxin production from cultured organisms can be established in vitro by some laboratories, but does not confirm toxin production in vivo. Detection of a “skewed” Vβ repertoire is a potential diagnostic test. Clearly, the sensitivity and specificity of the assay would need to be established before general application. To date, we have found skewed Vβ T-cell profiles in six independent cases of toxic shock syndrome. This patient had used tampons appropriately, including replacing tampons at least every 4 hours and not using them overnight, but nevertheless developed a life-threatening disease. The incidence of toxic shock syndrome peaked in the United States in 1980 and has since fallen substantially, as a result of factors including changed tampon absorbency. However, the incidence may be now increasing.5 Our case serves to remind us all to be vigilant for toxic shock syndrome in association with menstruation, and to consider the diagnosis in all patients with severe sepsis. Vβ profile of the T-cell antigen receptor of CD4 lymphocytes in a patient with toxic shock syndrome Vβ profile of CD4 cells from a patient 21 days after onset of toxic shock syndrome compared with the mean profile from 11 adult intensive-care patients with no evidence of infection. Note the massive expansion of cells carrying Vβ 2, for which toxic shock syndrome toxin-1 has known affinity.

Christopher M MacIsaac · Mark A Page · Beverley-Ann Biggs · Kumar Visvanathan

Patrick M Schlievert Professor, Microbiology, University of Minnesota, 420 Delaware Street SE, Minneapolis, Minnesota 55455, USA. patsATlenti.med.umn.edu Comment: As noted by MacIsaac et al above, my colleagues and I recently reported an increase in the incidence of staphylococcal toxic shock syndrome (TSS) in Minneapolis–St Paul in the United States, from 0.8 per 100 000 (in January 2000) to 3.4 per 100 000 (in December 2003).1 We noted that physicians across the United States were reporting TSS cases in increasing frequency. There are two major categories of staphylococcal TSS, menstrual and non-menstrual.2,3 Menstrual TSS is defined as occurring during menstruation or within the 2 days preceding its onset or the 2 days following its cessation; the illness is primarily, but not exclusively, associated with tampon use. Menstrual TSS is nearly always caused by the superantigen exotoxin, TSS toxin-1 (TSST-1).4 Superantigens significantly overactivate the human immune system to release cytokines that cause the clinical features of TSS (interleukin-1β [endogenous pyrogen]; tumor necrosis factor-α and β [capillary leak]; and interferon-γ and interleukin-2 [rash]).5 Non-menstrual TSS may occur in anyone, young or old, male or female, and today commonly follows superinfection of the upper respiratory tract after viral infection. Non-menstrual TSS is caused by TSST-1 (50%) or by staphylococcal enterotoxin B or C (together nearly 50%). The important question is what accounts for the fourfold rise in TSS that was reported in our 2004 study? We proposed several hypotheses. First, the increase in incidence partly results from the emergence of three strains of methicillin-resistant Staphylococcus aureus (MRSA), at least two of which are emerging worldwide. These strains are termed (by Centers for Disease Control [CDC] nomenclature) USA 1100 (TSST-1 positive), USA 400 (SEB/SEC, Panton–Valentine leukocidin [PVL] positive), and USA 300 (positive for an unknown superantigen as well as PVL). In our studies, USA 1100 strains currently comprise 20% of submitted isolates, compared with none before the year 2000. These isolates may produce 10 to 100 times more TSST-1 in vitro than their methicillin-sensitive S. aureus counterparts matched by pulsed-field gel electrophoresis profile. Thus, these organisms rapidly produce high levels of TSST-1, leading to TSS even when lower-absorbency tampons are used. In addition, the USA 400 and USA 300 strains are also emerging and are associated with increases in non-menstrual TSS. These latter isolates also produce more superantigens than their methicillin-susceptible counterparts. Secondly, in our 2004 study, physicians who submitted cultures to our laboratory defined cases of TSS based on patient presentation and the presence of an S. aureus strain producing one of the three causative exotoxins. Our TSS definition is likely to be broader than the strict CDC definition. Finally, we also noted that it is possible that women are beginning to menstruate and to use tampons at earlier ages. In addition, teenagers are bombarded with media advice that TSS is no longer a problem; failure to recognise the illness may lead to it becoming more severe before presentation. These lifestyle and awareness changes, combined with the emergence of high-toxin-producing strains and the expanded definition of TSS, may account for the observed increase in TSS. The increase does not appear to be caused by changes in tampon composition or absorbency.

Patrick M Schlievert

Andrew Firestone Psychiatrist; and Honorary Senior Lecturer, Monash University, Clayton, VIC 3168. afireATtpg.com.au To the Editor: Emerging evidence suggests that the so-called second-generation antipsychotics (SGAs), especially olanzapine and clozapine, can cause abnormal weight gain and increase the risk of diabetes mellitus.1-3 In Australia, there are calls for a prospective multicentre trial to compare the rates of weight gain and diabetes between SGAs.4,5 The Australian data presented here underline the pressing need for such a study. Data were examined for the 10-year period January 1994 to December 2003 for: Total prescriptions dispensed by the Pharmaceutical Benefits Scheme (PBS) and the Repatriation Pharmaceutical Benefits Scheme for 12 antipsychotic drugs; and Reactions reported in the same period to the Adverse Drug Reactions Advisory Committee (ADRAC) for each of these drugs, involving excessive weight gain or obesity, and diabetes mellitus or hyperglycaemic reactions. Reports were included in the survey only when it was considered that no other drug could be responsible. As clozapine is dispensed and recorded differently from other SGAs in Australia, complete data on numbers of prescriptions dispensed were not available. However, total Australian expenditure was available for each tablet strength of clozapine for the full 10-year study period, along with number of prescriptions dispensed and costs for the private hospital sector for the 4 years July 2000 to June 2004. Therefore, I calculated the average script cost for each tablet strength, and extrapolated the script numbers for the 10-year period, as shown in Box 1. Box 2 shows the “report rate” for each SGA for the side effects of weight gain or obesity, and diabetes or hyperglycaemia. The report rate for side effects was greater for clozapine than for any other SGA. Unfortunately, the true situation may be still worse. Clozapine is usually prescribed a month at a time, while the other drugs are prescribed for up to 6 months. Adjusting for this would widen the gap further. Moreover, as ADRAC promotes reporting for new drugs, the report rates for the five drugs introduced during the study period are probably inflated. Clozapine is not one of them. The limitations of ADRAC data are well known.4 Nevertheless, these are currently our best Australian data and strongly suggest that SGAs, of which risperidone has the most favourable profile, cause weight gain and diabetes much more often than the older antipsychotic agents. These data accord with previously published studies1 and support the US advice to avoid olanzapine and clozapine if possible. Recent PBS approval in Australia for use of olanzapine in bipolar disorder further underlines the urgent need for a prospective multicentre study to compare weight gain and glucose metabolism in patients taking antipsychotic drugs. Meanwhile, I suggest that: Patients who have abnormal weight gain with an SGA might be treated with chlorpromazine, trifluoperazine or haloperidol. PBS regulation of clozapine might be amended, to discourage its prescription until after failure of a “first-generation” as well as a second-generation antipsychotic drug. 1 Estimation of the total number of clozapine scripts in Australia Tablet strength (mg) Private hospitals data (Jul 2000–Jun 2004) Total clozapine used (Jul 1994–Jun 2004) Total prescriptions Cost ($) Average cost/ prescription ($) Cost ($) Estimated total prescriptions 25 3 650 230 929 63.27 9 636 814 152 313 50 25 1 443 57.72 85 980 1 490 100 21 057 6 632 499 314.98 153 276 771 486 624 200 69 24 392 353.51 463 131 1 310 Total 24 801 6 889 263 – 163 462 696 641 737 2 Report rates for side effects of antipsychotic drugs No. of years* No. of prescriptions dispensed No. of ADRAC reports Report rate (per million prescriptions dispensed) Weight gain† Diabetes‡ Weight gain† Diabetes‡ Chlorpromazine 10 950 221 0 0 0 0 Fluphenazine 10 327 126 0 0 0 0 Trifluoperazine 10 937 605 1 0 1.07 0 Pericyazine 10 657 514 0 0 0 0 Thioridazine 10 1 983 915 1 3 0.50 1.51 Haloperidol 10 1 499 254 2 0 1.33 0 Flupenthixol 9 121 132 0 0 0 0 Zuclopenthixol 8 93 839 0 1 0 10.66 Olanzapine 6 2 786 334 47 19 16.87 6.82 Quetiapine 4 271 957 1 4 3.68 14.70 Risperidone 9 1 298 156 6 2 4.62 1.53 Clozapine 10 641 737§ 41 61 63.89 95.05 ADRAC = Adverse Drug Reactions Advisory Committee. * Number of years with data available (as some drugs were introduced only after the start of the 10-year period). † Weight gain or obesity. ‡ Diabetes mellitus or hyperglycaemia. § Estimated number (see Box 1).

Andrew Firestone

To the Editor: The interesting case report by Currie and colleagues describes a variable cutaneous eruption of uncertain aetiology in a cluster of methadone-dependent patients.1 The rash was described as including pruritic, exanthematous, purpuric and eventually desquamative components, and typically as involving the trunk and extremities, particularly palms and soles. Secondary syphilis classically presents in a similar fashion, but no mention was made as to whether this had been excluded by serological testing. Indeed, the histology of the rash (perivascular inflammation, including plasma cell infiltrate, progressing to endarteritis) is similar to that seen in skin biopsies from methadone patients with secondary syphilis. However, an allergic or toxic cause appears to be implicated, in view of previous, well documented reports of hallucinogenic or other drug-related vasculitis published by ourselves2 and others.3-5

Vernon J Heazlewood

To the Editor: As a Victorian always on the lookout for something new, I read with interest the report by Currie and colleagues of a syndromic rash in patients attending methadone clinics in New South Wales.1 From the title I expected to read about a rash occurring as part of a syndrome, yet no group of concurrent symptoms was described. In fact, there was a long list with each patient of negative findings. I also had trouble deciding whether the four patients described indeed had the same rash. While the “lumpers” among us may consider it pedantic to split “rash” into more than one category, some doctors make an occupation of it quite successfully. For example, Patient 1 had petechiae and purpura, but no erythema and no involvement of the palms and soles. No photo, but nevertheless a nice description of vasculitis — common among intravenous drug users. Patient 2 had, from the look of the photo, a toxic erythema that resolved with desquamation of the palms and soles. No petechiae or purpura. Therefore, must be a different rash to Patient 1. Patient 3 is described as having “ a prominent purpuric rash involving both lower limbs”. However, the photo shows a macular erythema with some associated purpura that looks almost certainly to be an incidental manifestation of dependency. Difficult to say from a photo, as touch is so important in the diagnosis of true purpura. Of course, a 2 mm punch biopsy of the skin could resolve this almost instantly. Again, it is not clear whether this rash is similar to that seen in either Patient 1 or Patient 2. Patient 4 is described as having a red and itchy rash (erythematous and pruritic), but, from the photograph, we can clearly see that the rash is urticarial. This raises the possibility of urticaria, or urticarial vasculitis, or even erythema multiforme. Again, a skin biopsy would be very useful. The severe palmar peeling almost seems incongruous, but it does give me faith that buried in this report there might actually be a new desquamating rash associated with methadone use. In summary, I am still not clear whether the four patients described had the same rash, but I concur with the authors that several of these patients might warrant specialist assessment. Let’s hope they get it.

Rodney D Sinclair

In reply: The purpose of our report1 was to alert the wider medical community to the recent outbreak of a “syndrome” (“a group of symptoms and signs, which, when considered together, are known or presumed to characterise a disease or lesion”2) that included the development of various forms of rash in patients taking methadone syrup. Our report included four cases illustrating the different types of rash encountered to date. From October 2004, over 400 cases were reported from methadone clinics in New South Wales, although very few new cases have been reported since February 2005, presumably reflecting the success of preventive measures instituted by the NSW Health department. To date, the cause of this methadone-associated syndrome has not been elucidated. Skin biopsies of rash lesions have been performed in a number of our patients. All have shown chronic perivascular inflammation, with most demonstrating hyperkeratosis. A small number of patients have had a true leukocytoclastic vasculitis. As Heazlewood has commented, both secondary syphilis and illicit drugs such as amphetamines and cocaine have been reported to cause vasculitic rashes. However, none of the more than 50 patients in whom we have performed syphilis serological testing has had positive results, and few of our affected methadone patients have had urine drug-test results positive for amphetamine or cocaine use. We therefore believe that the syndrome we have described remains specific to the patients’ current use of methadone syrup. We are unaware of a rash that is “an incidental manifestation of dependency”, as suggested by Sinclair, but we would assure him that specialists from a wide variety of fields, including dermatology, immunology, immunopathology, infectious diseases, addiction medicine and epidemiology, have all been involved in the assessment and treatment of patients with this syndrome, and in the wider investigation of its pathogenesis.

Jon N Currie · Lisa Snell · Elizabeth M Benson

Huy A Tran Director and Associate Professor, Department of Clinical Chemistry, University of Newcastle, John Hunter Hospital, Locked Bag No. 1, Hunter Region Mail Centre, New Lambton Heights, NSW 2310. huy.tranAThnehealth.nsw.gov.au To the Editor: The interesting article by Gordon et al1 warrants further discussion with regard to the hypophosphataemia, “inappropriately” high level of serum 25‑hydroxyvitamin D, biochemical diagnosis of pancreatitis and management of hypercalcaemia. Fibroblast growth factor-23 (FGF-23) is a recently described 254‑amino-acid peptide that has been shown to have significant phosphaturic effect. It probably plays a major role in phosphate metabolism and homeostasis by rising after an oral phosphate load and falling after dietary phosphate restriction. In the patient discussed by Gordon et al, elevated FGF-23 level may, hypothetically, have been a major contributing factor to the low serum phosphate level. Although the understanding of this factor is still in its infancy, measuring the serum level of FGF-23 in the patient might have shed more light on the role of FGF-23 in phosphate homeostasis. However, FGF-23 levels do not correlate directly with serum phosphate levels, suggesting that FGF-23 exercises control via renal tubular cells, regulation of calcitriol levels or intestinal phosphate absorption. FGF-23 levels are also markedly elevated in chronic renal failure, partly in response to the chronic hyperphosphataemia and partly because of reduced renal clearance.2 Its action is independent of the traditional and better understood regulators of phosphate level, including parathyroid hormone and parathyroid hormone-related protein. The triad of high vitamin D level, hypercalcaemia and hypophosphataemia points strongly to a diagnosis of vitamin D intoxication, despite a patient history to the contrary. An alternative explanation is an inaccurate vitamin D assay from the supporting laboratory. This issue, which has been highlighted recently, has therapeutic relevance in monitoring vitamin D replacement therapy.3 In supporting the diagnosis of pancreatitis, serum lipase level remains the best biochemical test and is more specific than amylase level.4 The practice of dual amylase and lipase ordering in the investigation of such conditions is excessive, confusing and costly to the community and should be discouraged. The indication for bisphosphonate treatment in milk-alkali syndrome remains unclear and contradicts the underlying pathogenesis, which is believed to be that of excessive calcium ingestion. In the patient in question, excessive calcium ingestion overwhelmed the calcium homeostatic mechanism, resulting in severe hypercalcaemia. In such a milieu, osteoclasts would be heavily suppressed and inhibited, and thus the use of a bisphosphonate, whose major action is also by osteoclastic suppression, would be of little value other than in precipitating hypocalcaemia.5 Thus, expectant management as outlined by the authors would be sufficient to achieve normocalcaemia. As bisphosphonates are not without adverse effects,6 they should only be used after a clear diagnosis of hypercalcaemia has been made.

Huy A Tran

Adam P Morton Physician, Mater Hospital, Raymond Terrace, Brisbane, QLD 4101. AmortonATmater.org.au To the Editor: I read with interest the Lessons from Practice article on milk-alkali syndrome during pregnancy.1 I would like to offer the following comments. The patient’s alkalosis was in fact more impressive than presented, as the authors used the reference range for serum bicarbonate in non-pregnant patients. During pregnancy, serum bicarbonate levels typically fall by about 4 mmol/L to compensate for the respiratory alkalosis caused by elevated progesterone levels stimulating respiratory drive. Given the patient’s life-threatening calcium level on presentation, I am interested to know whether calcitonin treatment or even dialysis was considered while waiting for the pamidronate to take effect. An important aspect that the authors did not discuss in relation to this case is the reassuring data on the safety of both proton-pump inhibitors and H2-receptor antagonists in pregnancy. While there is more experience with the latter, two recent studies found no evidence of teratogenicity in almost 900 cases of exposure to proton-pump inhibitors in the first trimester.2,3 Clinicians should feel comfortable about prescribing these medications in pregnancy. After reporting a similar case,4 I wrote to Walco, the manufacturers of Quick-Eze, who subsequently changed their product labelling to include a warning about the number of tablets that could be safely taken each day. Disappointingly, they did not include a warning about ingestion during pregnancy, as I suggested.

Adam P Morton

Michelle V Gordon,* P Shane Hamblin,† Lawrence P McMahon‡ * Registrar, † Head, Department of Endocrinology, Western Hospital, Private Bag, Gordon Street, Footscray, VIC 3011; ‡ Head, Department of Obstetric Medicine, Sunshine Hospital, St Albans, VIC. hamblin1ATbigpond.net.au In reply: Our case appears to be consistent with typical milk-alkali syndrome. While measuring fibroblast growth factor-23 (FGF-23) level might have been of hypothetical interest, it is unlikely that it would have influenced management. Vitamin D intoxication was considered once the 25-hydroxyvitamin D results became available, and we closely questioned our patient in relation to this possibility. She was insistent that she had not taken any vitamin D supplements. It is possible either that the patient did not wish to admit to taking vitamin D or that the assay was misleading, as suggested by Tran. We agree that bisphosphonate therapy should not be advocated when the diagnosis of milk-alkali syndrome is clear. In this case, however, the patient was drowsy and very ill; the full history relating to antacid ingestion was not obtained until after the bisphosphonate therapy had been given. With regard to Morton’s comments, the hypercalcaemia settled promptly, so fortunately calcitonin treatment and other measures did not need to be considered. Drug safety in pregnancy is a difficult issue, as the effects of fetal or neonatal damage may carry lifelong implications, and even relatively rare associations need to be considered with care. In addition, many pregnant women are uncomfortable about taking prescription medications during pregnancy, even though their doctors may have a more relaxed view. Currently, over-the-counter antacids are classed as category A drugs for pregnancy, whereas H2-receptor antagonists and proton-pump inhibitors are category B1 and B3, respectively. Cimetidine has been associated rarely with neonatal hepatic abnormalities,1 and it is still too early to state with confidence that proton-pump inhibitors are “safe”, despite promising initial analyses. Ironically, the potential dangers of over-the-counter calcium-containing antacids, as demonstrated in this case report and others, are not currently adequately acknowledged. We have written to the manufacturers of Rennie tablets requesting a package label warning advising consumers not to exceed six tablets a day.

Michelle V Gordon · P Shane Hamblin · Lawrence P McMahon

Tulsi Menon Resident Medical Officer, Department of Orthopaedic Surgery, Royal Perth Hospital, 19 Morgan Road, Redcliffe, WA 6104 kaltulAToptusnet.com.au To the Editor: I report an infection with Dracunculus medinensis diagnosed incidentally on x-ray. The patient was a Sudanese immigrant, who had sustained a left knee joint effusion after falling. During management at Royal Perth Hospital, a calcified lesion with a serpentine appearance was seen on x-ray within soft tissues near the left knee joint. Its characteristic appearance, combined with the patient’s background, led to a diagnosis of D. medinensis (known as guinea-worm), a nematode parasite which causes dracunculiasis. The patient’s knee effusion was managed conservatively and improved within a few weeks. The effusion was secondary to the fall and believed to be unrelated to the calcified D. medinensis. Transmission of dracunculiasis is through consumption of contaminated water. The guinea-worm larvae mature and migrate towards the skin surface over 1 year (during which the patient remains asymptomatic), with over 90% of the worms appearing from the lower limbs. When in contact with water, the exposed guinea-worm releases larvae, and the lifecycle is completed when people ingest contaminated water.1 If the guinea-worms die before maturation, they usually calcify. The Global Dracunculiasis Eradication Campaign was established in 1981 with the aim of eliminating dracunculiasis through water sanitation efforts.1 With no vaccine or treatment, prevention is the only method of eliminating dracunculiasis. Since the campaign was established, the number of people affected by dracunculiasis has decreased by 98%. Currently, Sudan alone accounts for 73% of cases.1 Our patient migrated to Australia as a refugee from Sudan during the civil war. Her village in Sudan had only one source of water used for daily activities, including drinking. The patient knew many people with dracunculiasis, but did not know she had been infected. When calcified guinea-worms are discovered during routine radiological examination, they usually do not need treatment. Many people are not aware they have been infected. Muller reported 89% of patients with calcified guinea-worms were asymptomatic.2 No known previous case of a radiologically diagnosed calcified guinea-worm has been reported in Australia. The consequences of war and famine, with a resultant increase in refugees and immigrants from affected nations, is likely to increase the number of incidental calcified guinea-worms found in non-endemic countries. This disease may have a significant impact if affected immigrants arrive during the incubation period (when asymptomatic) and the parasite emerges from the skin after immigration (rather than calcifying), similar to the patient described by Spring.3 Thus, it is important for health personnel to be aware of dracunculiasis, including its radiological manifestations. Lateral view of left knee joint and distal left thigh A calcified lesion, representing Dracunculus medinensis within the soft tissues, is visible posterior and lateral to the distal femur near the knee joint. The calcification has a serpentine appearance: the proximal part coiled in appearance, the middle having a string-like linear appearance, and the distal part having dense curvilinear opacity.

Tulsi Menon