Letters

Graham D Hocking Child Psychiatrist, 25 High St South, Kew, VIC 3101. ghockinATiprimus.com.au To the Editor: Catto-Smith gives a very good account of the medical management of constipation and soiling in children,1 but fails to mention psychological, interpersonal and social factors in the main part of his article. He does mention “behavioural abnormalities” towards the end, in the section “When to refer”. I think it is generally accepted among paediatricians and child psychiatrists that the problem of constipation and soiling, or encopresis, often has multiple determinants and varied psychological effects on the child and the family. Certainly, the older the child is, the more likely these effects will be present.2 If this condition is to be managed in general practice over a period of “6–12 months”, general practitioners need to be aware of these factors so they can be addressed. Twelve months is a long time in the life of a 5 year old, and in that time pathological patterns can become well established and hard to shift. Most children over 5 years with soiling have developed secondary psychological problems as a result of the soiling.3 At this stage, assessment by a child psychiatrist will often reveal that the child has developed a pathological fantasy world around what they believe is happening inside them. The physical management of constipation and soiling is an essential part of the management no matter what the aetiology, but addressing the psychological interpersonal and social factors is equally important. If these factors are obvious to the GP and are not responding to intervention, the family should be referred to a child psychiatrist. With children over 5 years, the secondary effects have almost always become significant, and I believe that all these families should be referred for assessment. Catto-Smith quotes a 30%–50% relapse rate,1 and “there is evidence that they do not improve on reaching puberty”. This is not my experience with families that have the benefit of a multidisciplinary approach to the disorder.

Graham D Hocking

Anthony G Catto-Smith Director, Gastroenterology and Clinical Nutrition, Royal Children's Hospital, Flemington Road, Parkville, VIC 3052. tony.cattosmithATrch.org.au In reply: Hocking emphasises secondary behavioural and emotional effects that occur in some children with longstanding faecal soiling. Fortunately, there is good evidence that these tend to resolve with effective multimodal treatment of the constipation.1 The relatively high long-term relapse rate of soiling among children who have been treated in tertiary centres has only recently been recognised,2 but the psychological features of this relapsing group are not well defined. My review was directed toward general practitioners and was as much as possible evidence-based. I am unaware of any good quality evidence to support Hocking’s assertion of the benefits of automatic referral to a child psychiatrist of all children over the age of 5 years with ongoing faecal soiling. Given the beliefs of both myself and Hocking, that constipation and soiling are likely to have multiple determinants and varying psychological effects, it would seem to be appropriate to triage “problem” patients through a general paediatrician, with referral for psychological assistance if deemed appropriate. This is best summed up in my article in the section that Hocking mentions, “When to refer”.3

Anthony G Catto-Smith

Preeti Sachdev,* Simon Chapman† * Postgraduate Honours Student, † Professor, School of Public Health, University of Sydney, NSW 2006. simonchapmanAThealth.usyd.edu.au To the Editor: Smokeless tobacco products (with the possible exception of Swedish “snus”1) are carcinogenic.2 They cause oral cancer, sometimes rapidly (within 7 years of use).3 A pinch of smokeless tobacco held in the mouth for 30 minutes delivers as much nicotine as 3–4 cigarettes.4 Sachet of smokeless tobacco In Australia, the sale of smokeless tobacco was first banned in 1986, in South Australia.5 Thereafter, all states enacted legislation outlawing its sale, and, in 1991, an amendment to the federal Trade Practices Act 1974 banned the manufacture, importation and commercial supply of the products.6 Nevertheless, permits to import smokeless tobacco for personal use were issued on application, and, in March 2002, an amendment to the federal Customs (Prohibited Imports) Regulations 1956 allowed importation without a permit of amounts up to 1.5 kg for personal use.7 Between September 2000 and March 2002 (when permits were still required for all importations), 2270 permits were issued, while between March 2002 and December 2004, a further 88 permits were issued for amounts exceeding 1.5 kg (Mr Tim Pulford, Australian Competition and Consumer Commission, Canberra [which administers permits], personal communication). Following observations of smokeless tobacco being sold in south Asian shops in Sydney, New South Wales, we sought to assess its availability. We selected 14 Sydney suburbs with large populations of residents from south Asia (defined as the Indian subcontinent), and surveyed all south Asian mixed businesses in the shopping precincts of these suburbs in March 2005. If no smokeless tobacco products were displayed, the person serving was asked if they had any “paan masala or guthka” (Hindi expressions for smokeless tobacco) for sale. Fifty of the 53 shops surveyed (94%) sold smokeless tobacco: 31 (62%) of these kept it under the counter, 14 (28%) on display behind the counter, and five (10%) on shelves accessible to consumers. No shopkeeper advised that sale of the products was illegal. The prevalence of “under the counter” storage suggests widespread awareness that it is illegal to sell the products. The federal Customs (Prohibited Imports) Regulations do not restrict the number of times a person may import up to 1.5 kg of smokeless tobacco for personal use without a permit. A typical sachet of guthka (Box) weighs 4.6 g, meaning that around 320 sachets could be legally imported for personal use. It would be entirely legal for a shopkeeper and each family member to import up to 1.5 kg on a daily basis if it was intended for personal use. Diversion of this into retail trade appears easy. The ease with which we were able to obtain smokeless tobacco suggests that the law prohibiting sale is not being enforced. The New South Wales Public Health Act 1991 empowers officers, such as environmental health officers, to investigate breaches of the Act. These officers should undertake surveillance of the readily identifiable shops in the manner that we did, confiscate the products being sold and warn that future sales will result in prosecution.

Preeti Sachdev · Simon Chapman

Chris E Rissel Clinical Associate Professor, School of Public Health, University of Sydney, Level 9, King George V Building, Missenden Road, Camperdown, NSW 2050. crissATemail.cs.nsw.gov.au To the Editor: The editorial by Sharman about exercise and air pollution1 makes the point that cars contribute substantially to air pollution, and air pollution is known to have adverse health effects. Therefore, Sharman posits that exercise, which is unequivocally good for human health, is best done away from sources of air pollution. This “common-sense” maxim to avoid air pollution when exercising is superficially reasonable as far as it goes, but is a very weak response to the health and social problems generated by motor vehicles or the need for increased levels of physical activity in the population. With only half the Australian population achieving adequate levels of physical activity,2 recommendations to patients to be more physically active are essential. To simultaneously promote exercise and then put a health warning on this physical activity effectively undermines the recommendation. Part of the difficulty in judging the actual risks from air pollution and benefits of physical activity is that the science of pollutant exposure is not well understood at the individual level. It is not currently possible to say that exercising in a particular environment will have a net negative effect. Thinking laterally, perhaps physical activity even boosts the immune response in a way that helps the body resist adverse effects of air pollution? Perhaps only under more extreme conditions would outdoor activities need to be curtailed. One body of relevant research that Sharman did not consider is the research on pollutant exposure by travel mode, which clearly indicates that car drivers and passengers have pollutant exposures at least twice that of pedestrians walking on the same street.3,4 The longer people sit in cars, the greater their exposure to air pollutants, not to mention the increased risk of obesity.5 Therefore, a highly sensible approach is to recommend to patients that they avoid travelling in cars, particularly if the patient is sensitive to air pollutants or if traffic is congested. The most obvious common-sense solution to reduce air pollution and increase individual and population levels of physical activity is to recommend to patients that they replace short car trips with walking or cycling. As little as two 15-minute active transport trips per day can achieve recommended levels of physical activity to maintain health. It is possible to change travel behaviour, and this is a far better recommendation for all patients than telling them not to exercise near traffic.

Chris E Rissel

Louis A du Plessis Former Senior Lecturer, School of Applied Science, Riverina–Murray Institute of Higher Education, Wagga Wagga, NSW (retired). eldupeATbigpond.com To the Editor: Health professionals in the Sydney Greater Metropolitan Region have been found to be less aware of air pollution and its health effects than are patients susceptible to such effects.1 Therefore Sharman’s brief review of the harmful effects of air pollution is welcome.2 In offering advice on how to minimise the exacerbation of pollution-induced harm by exercise, Sharman concentrates on the spatial distribution and temporal variation of traffic. The advice is sound for the metropolitan population, but incomplete for non-metropolitan residents exposed to smoke from burning biomass. Australia’s National Environment Protection Measure defines limits for inhalable particulate matter suspended in ambient air (PM10). The NSW Department of Environment and Conservation operates a network of monitoring stations to measure PM10 and other pollutants, and posts the results daily on its Internet site <www.epa.nsw.gov.au/index.htm>. The National Environment Protection Measure requires environmental authorities to work towards reducing the number of days on which PM10 exceeds the daily limit to no more than 5 days per year. This goal is far from being realised in some places. One such place is Wagga Wagga, NSW, where monitoring of PM10 started on 11 April 2001. From that date up to 6 July 2005, the city experienced 118 days on which PM10 exceeded the limit, as well as many days of pollution near but below the limit. In the same period, monitoring stations in the Sydney Greater Metropolitan Region registered between 11 and 33 days of excess PM10. Of the 118 days of above-limit PM10 in Wagga Wagga, 30 occurred in the period from 30 October 2002 to 26 January 2003, when there were severe bushfires in south-eastern Australia. Most of the rest had a cause that is very evident in the surrounding countryside in autumn — the burning of paddocks to prepare them for sowing. Rural residents’ health is worse than urban residents’ health for many reasons, but biomass burning is not widely recognised as one of them. The Australian Medical Association’s Rural Reference Group, which is being convened to improve rural health,3 may wish to add environmental health to its agenda. There are two steps that the Group could take to lessen the effects of smoke from burning biomass. The first is to acquaint rural doctors with information such as that presented by Sharman. The second is to persuade the Department of Environment and Conservation and NSW Health to issue rural health warnings based on the continuous, real-time output of PM10 monitors in regional centres.

Louis A du Plessis

James E Sharman Postdoctoral Research Fellow in Cardiovascular Physiology, Princess Alexandra Hospital, Ipswich Road, Woolloongabba, Brisbane, QLD 4102. jsharmanATsoms.uq.edu.au In reply: As emphasised in my editorial, regular aerobic exercise is to be encouraged, as it is of undeniable benefit to health.1 The crux of the intended message was for people to undertake an exercise program, but not near busy roads. There was no suggestion to curtail outdoor activities. Although the effect of traffic pollution on an individual is not well understood, there are many hundreds of scientific papers consistently finding that whole automotive pollution, or components thereof, damage biological tissue and promote disease.2 The World Health Organization recognises urban air pollution as a major risk to human health, with exposure to particulate matter alone accounting for an estimated 800 000 deaths a year globally.3 Would it be ethical to confine this information to the annals of scientific literature, or should some attempt be made to inform those who may be unknowingly and unnecessarily exposing themselves to veritable risk? People should have access to all the available information so that they can make an informed decision on where to exercise. Not only is it common sense, but it is entirely reasonable to suggest that people would be better off avoiding exercise alongside roadways congested with traffic. Quite separate to the question of exercising beside busy roads, but equally important from a health perspective, is the issue of persistently high ambient levels of particulate air pollution in certain regions. The air quality problem encountered in Wagga Wagga is exacerbated by geographical and climatic factors that encourage entrapment of air pollution, owing to a temperature inversion layer that is particularly apparent during winter. Other cities, such as Launceston, Tasmania, suffer the same fate and, in both cases, smoke from domestic wood-heaters is thought to be the biggest contributor to poor air quality. What are people to be told regarding exercise in these regions? It may be reasonably argued that habitual exercise in such environments would be detrimental to health, but it would be a very poor public health outcome if people were advised to stop exercising. In the affected areas mentioned, community education programs have been in existence for years, but these appear to be of limited value, as daily air pollution limits are regularly exceeded,4 as highlighted by du Plessis. Although unpopular, the answer lies in stricter regulations to clean the air, such as banning wood-heaters and tighter monitoring of rural burning.

James E Sharman

Robert D Goldney Professor of Psychiatry, University of Adelaide, The Adelaide Clinic, 33 Park Terrace, Gilberton, SA 5081. Robert.goldneyATadelaide.edu.au To the Editor: Two recent items in the Journal might potentially lead to misinterpretation of the evidence on managing depression in young people. The first was the book review entitled Darker side of “wonder drugs” by Jureidini1 in which there was no disclosure that the author of the review is president of Healthy Skepticism, a body which has been quite strident in its opposition to antidepressant therapy. The second was the unattributed comment in the editorial by Rey and Dudley describing “parents who believe their children killed themselves because they were taking SSRIs [selective serotonin reuptake inhibitors] . . .”,2 which may imply subtly that this has occurred frequently. In a review of the United Kingdom General Practice Research Database of more than three million people,3 there were no suicides among the 6976 aged 10–19 years who had been prescribed one of two SSRIs or two tricyclic antidepressants; however, 15 people in that age group who had not received an antidepressant drug died by suicide. Furthermore, in a review of 14 857 suicides in Sweden, of the 52 involving people under 15 years, no SSRIs were detected, and in the 15–19-years age group, those taking SSRIs had a lower relative risk of commiting suicide than those taking other antidepressants.4 Clinicians with responsibility for children and adolescents can be reassured by these data, and also by the fact that the American Food and Drug Administration “black box” warning (their most potent warning) about antidepressants has recently been modified.5 Furthermore, the American Academy of Child and Adolescent Psychiatry and the American Psychiatric Association have provided a new resource about the use of medication in treating childhood and adolescent depression,6 which has been endorsed by over a dozen United States organisations comprising a “national coalition of concerned parents, providers, and professional associations”. This should allay questions that have rightly been raised, but that have been answered in favour of the judicious use of antidepressants, along with other therapeutic measures for children and adolescents with severe depression. In view of the strong association between child and adolescent mood disorders and suicide,7 the above research findings and the recommendations of respected professional bodies raise the issue of potential legal action for not at least trialling antidepressant medication in young people with severe depression if non-pharmacological measures are ineffective.

Robert D Goldney

Peter R Mansfield,* Melissa K Raven,† Jon N Jureidini‡ * Research Fellow, University of Adelaide, SA; † Lecturer, Flinders University, Adelaide, SA; ‡ Head, Department of Psychological Medicine Women's and Children's Hospital, Adelaide, SA. peter.mansfieldATadelaide.edu.au To the Editor: Rey and Dudley cite clinical experience as the basis of their recommendation of selective serotonin reuptake inhibitors (SSRIs) — chiefly fluoxetine — for youth with severe depression plus severe impairment or failure of non-drug therapy.1 They do not discuss the evidence on efficacy because they claim that it is “ambiguous enough for scholars to be divided”. It is true that industry-funded scholars are continuing to suggest that SSRIs (chiefly fluoxetine) provide a worthwhile benefit.2 However, the evidence is unambiguous. The four published comparisons of fluoxetine versus placebo for children and adolescents have all been negative on their pre-specified primary endpoints.3,4 A tiny average benefit is likely, but the magnitude of this benefit is unlikely to exceed the magnitude of less frequent but more severe harms. Furthermore, the common clinical impression of worthwhile benefit is to be expected given the large average improvements seen in placebo groups. Rey and Dudley speculate that psychosocial treatments may be less effective with uncooperative teenagers.1 However, that group may also be at higher risk of the dangers of intermittent use of, and overdosing with, antidepressant drugs. Rey and Dudley cite Timimi’s critique of the concept of childhood depression5 as supporting “treating depression primarily as a moral or social problem”. However, Timimi did not even allude to depression as a moral problem, and advocated a multi-perspective approach that normalises emotional responses to adverse life experiences and includes interventions addressing biological factors, such as diet, exercise, and cognitive abilities. Rey and Dudley use a related straw-man argument in their final sentence when they suggest that the only alternatives to SSRIs are tricyclic antidepressants, victim blaming, and non-treatment. Rey and Dudley deny being influenced by the gifts and funding that they have received from drug companies. There is compelling evidence that gifts and funding are effective, on average, for influencing beliefs, especially among people who have an illusion of invulnerability.6 We are not aware of any way that any individual can know that he or she has not been influenced.

Peter R Mansfield · Melissa K Raven · Jon N Jureidini

Joseph M Rey,* Michael J Dudley† * Professor, Psychological Medicine, University of Sydney, PO Box 142, North Ryde, NSW 1670. † Senior Lecturer in Psychiatry, University of New South Wales, Randwick, NSW. jreyATmail.usyd.edu.au In reply: The data available are inconclusive, but suggest that treatment with selective serotonin reuptake inhibitors (SSRIs) may increase the short-term (less than 14 weeks) risk of suicidal thoughts or self-harm in children and adolescents slightly, by about 2%. However, SSRI treatment may actually decrease the number of completed suicides,1 as Goldney also highlights. To show whether SSRIs influence the risk of completed suicide, a rare event, requires a randomised trial including up to two million individuals.2 This will not happen. Hence, clinicians must rely on accumulated data from experimental, epidemiological, and observational studies. Disagreements about interpretation will doubtless continue. In response to Mansfield and colleagues, we personally know of media reports influencing some practitioners to revert to using tricyclic antidepressants, and child psychiatrists to avoid treating depressed adolescents. We do not shrink from our interpretation of the implications of Timimi’s reconceptualisation of “depression” as “unhappiness”. Regardless of how childhood depression is classified or named, we remain concerned that the impetus for clinicians to diagnose and treat it not be lost. Its social correlates include stigma and racism, which often involve seeing mental health problems as moral failures of character. Our view is that fluoxetine shows a favourable harm–benefit profile in moderate to severe depression. According to the Treatment for Adolescents with Depression study,3 which was not funded by drug companies, four children need to be treated with fluoxetine for one to show much or very much improvement attributable to medication. This compares with having to treat 21 children for one to display a widely defined harm-related event. The numbers improve further when fluoxetine is combined with cognitve behavioural therapy (3 and 50, respectively). Pending new studies, clinicians would be unwise to ignore these data when treating serious depression in young people, a recurring illness that produces much suffering, physical and psychosocial disability, and suicide (odds ratio estimates ranging from 11.0 to 27.0).4 Our opinions are consistent with those of the recently released joint clinical guidance by the colleges of psychiatrists, general practitioners, and physicians.5 Mansfield and colleagues suggest that our editorial’s content might have been influenced by drug company gifts. We provided the educated readers of the Journal with information to judge this for themselves.

Joseph M Rey · Michael J Dudley

Duncan Topliss Chairman, Adverse Drug Reactions Advisory Committee, Therapeutic Goods Administration, Department of Health and Ageing, Canberra, ACT 2601. adracAThealth.gov.au Comment: Three essentially independent reviews of the use of selective serotonin reuptake inhibitor (SSRI) antidepressants in children and adolescents have been undertaken in Australia in the past 9 months.1-3 The review by the Adverse Drug Reactions Advisory Committee1 had input from representatives of the Royal Australian and New Zealand College of Psychiatrists and the Division of Paediatric and Child Health, Royal Australasian College of Physicians (RACP). All three reviews noted the paucity of information to support the efficacy of these and other antidepressants in children and adolescents, and the frequent observation of increased suicidal thoughts and self-harm in clinical trials. The colleges’ review2 and the National Prescribing Service Rational Assessment of Drugs And Research (RADAR) review3 support the ADRAC advice that: Any use of SSRIs in children and adolescents with MDD [major depressive disorder] and other psychiatric conditions should be undertaken only within the context of comprehensive management of the patient. Management should include careful monitoring for the emergence of suicidal ideation and behaviour which may particularly develop early in therapy, or if therapy is interrupted or irregular because of poor compliance. Cognitive behaviour therapy, if it is available, may enhance the outcome in MDD. An SSRI should be chosen for a child or adolescent with MDD or other psychiatric condition only after taking into account the recent evaluations of clinical trial data and the Australian product information. Prescribers should be aware that the marketers of fluvoxamine and sertraline (indicated for obsessive compulsive disorder) advise against their use in children and adolescents with MDD, and the marketers of citalopram, escitalopram, paroxetine, venlafaxine and fluoxetine warn or caution against their use in patients aged less than 18 years for any indication. It is important to note that children and adolescents who are being treated for MDD with an SSRI should not have their medication ceased abruptly.

Duncan Topliss

Gordon R W Davies Psychiatrist, 33 Smith Street, Wollongong, NSW 2500. alienistATihug.com.au To the Editor: The recent editorial by Andrews1 quotes a report suggesting that the integration projects funded by the Australian Government produced substantial benefits to patient care at no extra cost. Working in an area in which one of these projects was funded, I suggest that such an inference is unwarranted, particularly as many patients report increased difficulty in accessing public mental health services. The Illawarra, being geographically circumscribed and with a relatively small medical population, has always had a high degree of interaction between services, although it is true to say that these have somewhat declined in recent years with larger bureaucracies and increased privacy concerns. In my early days in the area, there was a monthly meeting involving police, Youth and Community Services, the Housing Commission, school counsellors, and hospital and community social and mental health workers to coordinate the management of problem families. Sadly, this no longer occurs. Unfortunately, the major effect of the integration project was simply to add a management structure to the prior interaction, and not to significantly increase it. It seemed that the core issue was control and not service provision. Useful coordination projects, such as some commonality of core records, never seemed to happen, and with the passing of the project, things have, in fact, been worse, as fundamental community services such as the crisis team and chronic care components have been cut. I have always been a strong supporter of a more integrated approach to care, but one that does not grow primarily from the workers involved in day-to-day clinical care and that addresses their needs is unlikely to be lasting and successful.

Gordon R W Davies

Andrew S Kemp Professor, Allergy Immunology and Infectious Diseases, The Children’s Hospital at Westmead, Locked Bag 4001,Westmead, NSW 2145. andrewk5ATchw.edu.au To the Editor: Publicity such as that on the recently televised “Sunday” show (Channel 9) entitled “When food can be fatal” (http://sunday.ninemsn.com.au/sunday/cover_stories/transcript_1770.asp), which contained statements that “30 in every 1000 [3%] children in Australia are at risk of a severe allergic reaction [anaphylaxis] to a food”, and a reference to a “tsunami of children” with serious allergies, provokes understandable concern and anxiety. Some perspective on this issue is required. To determine the risk, it is essential to study a population-based cohort. Allergies to peanuts or tree-nuts are the most common cause of severe childhood food anaphylaxis and death.1 What is the risk for Australian children of peanut-induced anaphylaxis that is likely to require adrenaline? Of a population-based cohort of 456 Tasmanian children aged 7–8 years, none reacted to a peanut skin-prick test.2 In the Australian Childhood Asthma Prevention Study (CAPS),3 a high-risk cohort, 4.9% of 3 year olds were prick-test positive to peanut (unpublished data) using a liberal cut-off of ≥ 2 mm (for clinical testing the usual cut-off is ≥ 3 mm). Perhaps the most helpful information comes from a population-based study of 13 971 preschool children in the United Kingdom who were followed from birth to 6 years of age. Forty-nine (0.35%) children had an allergic reaction to peanut, of whom only two (0.014%) had what was described as anaphylaxis.4 Thirty-six of the children underwent formal peanut challenge, 23 reacted and three had reactions for which adrenaline was given. Combining these three with the previous two gives a severe reaction rate requiring adrenaline of 0.036%. This suggests that, of the 49 children in the UK study who had an allergic reaction to peanut, only 10% were at risk of a severe reaction requiring adrenaline. Only a third to a half of children with a positive peanut skin test will react if exposed.5 Applying these considerations to Australian children indicates that the proportion at risk of a severe peanut reaction is only 0.25% (4.9% × 1/2 × 5/49) even in a high-risk cohort such as the CAPS. This would be substantially lower in a population-based cohort. For the cohort of 7–8-year-old Tasmanian children referred to above, the risk would be much less than 0.2%, considering none of 500 children was prick-test positive to peanut allergen. There has been a substantial increase in childhood food allergy in recent decades;5 however, sensationalist statements and inaccurate figures are unlikely to be helpful in developing appropriate responses. The Australasian Society of Clinical Immunology and Allergy recently published guidelines for the prevention of food anaphylactic reactions,6 and has other useful information for patients and medical practitioners on its website (http://www.allergy.org.au/).

Andrew S Kemp

Grant A Mackenzie Paediatrician and Postgraduate Student, Ear Health and Education Unit, Menzies School of Health Research, PO Box 41096, Darwin, NT 0811. grantmacATmenzies.edu.au To the Editor: Funding of universal varicella zoster vaccine (VZV) at ages 18 months and 10–13 years was recently announced in Australia. Health professionals should be aware of a number of related issues. Varicella vaccination was recommended in the United States from 1996 for all children aged 12–18 months, with catch-up vaccination to age 13 years. US surveillance shows: Decreased varicella mortality in all age groups except those aged ≥ 50 years (average varicella deaths per year: 145 in 1990–1994 versus 66 in 1999–2001).1 Decreased varicella cases in all age groups, with a non-significant reduction in hospitalisations (average annual hospitalisations in three surveillance regions: 40 before vaccination versus 14 after vaccination).2 US data on herpes zoster have not yet been published. There are concerns that, in the longer term, universal varicella vaccination may increase the incidence of adult varicella and herpes zoster, similar to the effect of pertussis vaccination on adult pertussis. Modelling in the United Kingdom predicted that universal infant vaccination would initially reduce varicella, but would result in increases in herpes zoster 5–10 years later and adult varicella 20–40 years later.3 In contrast, modelling of adolescent vaccination predicted a small decrease in varicella, but no increase in later adult varicella.3 Varicella is generally perceived as a mild illness, while vaccination is largely valued for preventing serious, life-threatening conditions. Anecdotal reports of low levels of private purchase of VZV in Australia suggest it may not be a priority for some families. With 36% of general practitioners concerned about unknown side effects of VZV,4 and public concern about vaccine adverse events in the face of low disease rates, the level of acceptance of universal varicella vaccination by providers and consumers is uncertain. Alternatives to universal varicella vaccination were a high-risk strategy (vaccination of children with chronic illness and family members of high-risk individuals) or waiting until US disease patterns were established. These were real options as: A high-risk strategy may prevent up to 45% of paediatric hospitalisations.5 Hospitalisation and herpes zoster contribute more to health costs than treatment in the community or acute varicella.5 Natural infection, at the cost of disease, immunises most of the population. Any increase in adult varicella and herpes zoster caused by varicella vaccination may be alleviated by booster doses, but the added cost, difficulty in reaching the target population, and potential impact on community confidence in vaccination may be significant problems. The universal varicella vaccination program will test providers’ and consumers’ acceptance of vaccination against what is perceived as a mild illness.

Grant A Mackenzie

Kristine Macartney,* Peter McIntyre† * Senior Research Fellow, † Director, National Centre for Immunisation Research, The Children’s Hospital at Westmead, Locked Bag 4001, Westmead, NSW 2145. petermATchw.edu.au Comment: Recently published data have added considerably to the evidence about the impact of varicella vaccination programs, and address many of the concerns raised by Mackenzie. First, in the United States where universal varicella vaccination was recommended a decade ago, recent data show that — despite much slower implementation than is expected in Australia — varicella-related disease has declined by up to 90%, and hospitalisation rates and deaths from varicella by more than two-thirds, due to herd immunity.1,2 Second, data have now been published on the incidence of herpes zoster in areas of sentinel surveillance in the US, showing no change in age-specific rates to 2002.3 Along with the success of a recent trial of high-dose varicella vaccine in reducing herpes zoster in older adults,4 these data add to confidence that any increase in herpes zoster — as predicted in some models — will be detected and effectively combated by vaccinating people aged over 60 years. A substantial allowance for surveillance of both varicella and herpes zoster was included in the 2005 federal budget, to accompany the introduction of universal varicella vaccination in Australia. Mackenzie is correct that varicella is perceived by some as a mild illness, but it is important for general practitioners to emphasise to patients that this is incorrect.5 Each year in Australia, varicella causes around seven to eight deaths and more than 1500 hospitalisations,6 many associated with serious complications, such as invasive bacterial infection, pneumonia, and encephalitis. Although complications are more likely in adults and immunocompromised patients, 42% of hospitalisations are in children aged 0–4 years,6 most of whom are otherwise healthy.7 Patients can also be reassured about the safety of varicella vaccines, as clinical trials now date back 30 years, and more than 40 million doses of vaccine have been distributed in the US. Mackenzie suggests alternatives to universal childhood varicella vaccination, such as vaccination of “high risk” patients and their families, or of adolescents alone. However, these programs would not prevent morbidity among otherwise healthy young children and older age groups, as they would be insufficient to generate herd immunity. Moreover, age-based vaccination strategies have been shown to be easier to implement than more targeted programs. In the absence of a publicly funded universal program, the private market could sustain modest varicella vaccination rates of around 40%–50% in Australia.8 This would increase the number of adolescents and adults susceptible to varicella, because of reduced exposure to the virus and lack of vaccination; these groups also experience greater morbidity with infection than children. A universal program vaccinating young children and adolescents against varicella offers the best current option to reduce morbidity and mortality from this disease in Australia. Ongoing surveillance of varicella and herpes zoster in Australia and elsewhere will reveal whether there is a need for further interventions, such as a second dose of varicella vaccine in children and high-dose varicella vaccine to prevent herpes zoster in older adults.

Kristine Macartney · Peter McIntyre

Adam P Morton Physician, Department of Medicine, Mater Adult Hospital, Raymond Terrace, South Brisbane, QLD 4101. AmortonATmater.org.au To the Editor: The excellent report by Harding et al in the Diagnostic Dilemmas article in the Journal highlighted medications and conditions that may cause false positive results of biochemical tests for phaeochromocytoma.1 Another group of patients, those with obstructive sleep apnoea (OSA), may have raised urine noradrenaline levels in the absence of a phaeochromocytoma. Of about 170 patients seen at a hypertension screening service at the Mater Adult Hospital between 1998 and 2000, six had elevated levels of urine noradrenaline and normetadrenaline up to twice the upper limit of the normal range on repeated testing. Five were obese and were proven to have significant OSA. All required at least three antihypertensive drugs for reasonable control of their blood pressure, and had normal suppression of catecholamines with clonidine. A recent report described a series of five patients with OSA presenting as pseudophaeochromocytoma who had consistently elevated levels of noradrenaline on measurement of 24-hour urinary catecholamine levels; normetadrenaline levels were not measured.2 Noradrenaline levels became normal in all five patients after treatment with continuous positive airway pressure, and blood pressure levels improved significantly. Excess urinary noradrenaline, rather than being adrenal in origin, was thought most likely to be due to increased neuronal release of noradrenaline from small arteries and arterioles as a result of sympathetic nerve activity and synaptic overflow. In conclusion, OSA is an important reversible cause of elevated urine noradrenaline and normetadrenaline levels in patients with resistant hypertension.

Adam P Morton

Stan B Sidhu In reply: We thank Morton for his letter which highlights another group of patients in whom raised urinary noradrenaline levels exist in the absence of a phaeochromocytoma. His experience and our group of patients1 should serve as a note of caution when making the diagnosis of phaeochromocytoma. A combination of positive results of biochemical tests, along with results of anatomical and functional imaging, should serve to minimise false positive diagnoses.

Stan B Sidhu

Raymond W Cook Retired anaesthetist, PO Box 6135, O’Connor, ACT 2602. dcookATactonline.com.au To the Editor: A great deal has been said and written in recent years about inadequate numbers in the medical workforce. As a result, many scientific and political articles about the need to increase the number of medical students and how to train them for the workforce have been written.1,2 Yet there has been little discussion of how those already trained and in the workforce should be retained, or of the rate of attrition of those in the workforce. After 30 years as an anaesthetist, I can recall only one positive change in my conditions of work — the introduction of exhaust gas scavenging. All other changes have been negative: longer hours, greater stress (from multiple factors, such as increased complexity, day surgery and admission on day of surgery) and higher public expectations. My motives in suggesting the need for such research are purely selfish — having recently retired, and enjoying the lack of stress, I wish to be sure there is an adequate workforce in my old age.

Raymond W Cook

Rebecca J Guy,* David E Leslie,† Kleete Simpson,‡ Beth Hatch§ Jennie Leydon,¶ Margaret E Hellard,** Heath A Kelly†† * Epidemiologist, ** Head, Centre for Epidemiology and Population Health Research, Macfarlane Burnet Institute for Medical Research and Public Health, GPO Box 2284, Melbourne, VIC 3001; † Microbiologist, ¶ Senior Serologist, †† Head, Victorian Infectious Diseases Reference Laboratory, Melbourne, VIC; ‡ Surveillance Manager, § Partner Notification Officer, Blood Borne Viruses and Sexually Transmissible Infections Program, Department of Human Services, Melbourne, VIC. Rebecca. GuyATburnet.edu.au To the Editor: In Victoria, notifications of infectious syphilis infection (primary, secondary and early latent [< 2 years’ duration]), reported by the Department of Human Services, have increased more than fivefold in the past decade, from 16 in 1995 to 85 in 2004 (Box). An increase in notifications has also been observed in Sydney.1 Whereas previously in Victoria, very few infectious syphilis notifications were reported among men who have sex with men (1 of 16 cases in 1995), in 2004, 63 of a total of 85 cases (74%) were in this group (68% of these were acquired in Victoria). The Victorian Infectious Diseases Reference Laboratory (VIDRL) conducts testing for sexually transmitted infections (STI) and HIV for three Melbourne sexual health clinics with a high proportion of patients who are men who have sex with men. In 2004, 62 male patients tested positive for infectious syphilis, and 40% of these were HIV-positive. This is similar to the situation in Sydney, where in 2003, 54% of infectious syphilis cases were reported among HIV-positive men who have sex with men.1 Syphilis outbreaks among men who have sex with men have been reported elsewhere in recent years. There was an outbreak of syphilis in this group in Greater Manchester; between 1999 and 2002, and 37% of cases were in HIV-positive men.2 In this population, syphilis infection was associated with unprotected oral sex with high numbers of partners, seeking sexual partners at venues (darkrooms, cruising areas and saunas) and use of drugs (GHB [gamma hydroxybutyrate] and poppers [amyl nitrate]).3 A San Fransisco study in 2000, performed in response to a syphilis outbreak among men who have sex with men, reported that meeting sexual partners through use of the Internet was a factor significantly associated with syphilis infection.4 It is likely that some or all of the factors reported in these outbreaks overseas are contributing to the sustained increased in infectious syphilis notifications in Victoria, but it is important to have local data to ensure interventions are targeted appropriately and cost effectively. In Victoria, responses to the increase in syphilis notifications have already included an alert to general practitioners to encourage men who have sex with men to have syphilis testing and individual counselling, and syphilis testing of men who have sex with men at a popular sex-on-premises venue over a 4-week period. Depending on further studies in this population in Victoria, other responses could include enhancing outreach at Internet chat rooms, intensive counselling of HIV-positive men who have sex with men, and education interventions such as peer-led community-based strategies for countering unsafe sex and substance-use behaviours. Finally, it is vital that interventions are multidisciplinary, collaborative and evidence-based. Infectious syphilis notifications by year, Victoria Data from the Notifiable Infectious Diseases Surveillance System Database, Communicable Diseases Section, Victorian Department of Human Services.

Rebecca J Guy · David E Leslie · Kleete Simpson · Beth Hatch · Jennie Leydon · Margaret E Hellard · Heath A Kelly

To the Editor: Lymphogranuloma venereum (LGV) is an uncommon sexually transmitted infection caused by Chlamydia trachomatis serovars L1–3. LGV is not endemic in Australia, and rare Australian cases of LGV have been seen in patients who have either acquired the infection while travelling overseas in an endemic area, or have had local contact with an imported case. Currently, there is an outbreak of LGV in western Europe (in particular, The Netherlands) and the United States.1-4 A case of LGV in an Australian man with no history of overseas travel was managed recently. A 42-year-old bisexual man with previously treated early syphilis and hepatitis C infection presented to a Melbourne hospital in August 2004 complaining of 3 months of tender right inguinal lymphadenopathy. An excisional biopsy showed the formation of necrotising granuloma indicative of LGV. He had no history of penile ulceration, urethritis or proctitis. The surgical wound healed normally. The patient gave a history of attending sex-on-venue premises (“gay saunas”) and “beats”. He reported having oral sex with men, and recently having non-insertive sex involving masturbation with an unknown casual male contact who was apparently an overseas visitor. The patient had a female sexual partner with whom he had irregular, unprotected vaginal intercourse. The diagnosis of LGV was confirmed by polymerase chain reaction (PCR), which detected C. trachomatis, identified as serovar L2 by nucleotide sequencing, from the excised lymph gland. IgG and IgA antibodies to C. trachomatis were demonstrated by enzyme-immunoassay. Tests for other active sexually transmitted infections were negative. The patient was treated with doxycycline (100 mg twice daily for 3 weeks). His asymptomatic female partner was also treated. LGV is endemic in developing countries in our region, but occurs only sporadically in industrialised countries. The first stage of disease consists of a papule or ulcer that may occur on the penis, urethra or cervix. Proctocolitis may also be present, mimicking inflammatory bowel disease. Regional lymphadenopathy develops in the secondary stage of disease when there may be systemic symptoms. Fistula formation at these sites can be prevented by early recognition and treatment. Late, severe genital ulceration is rarely seen. Confirmation of a diagnosis of LGV requires showing C. trachomatis serovars L1–3 by serological tests or PCR on genitourinary specimens. Lymph node resection is not favoured because of the possibility of sinus formation. Prolonged treatment with doxycycline or roxithromycin for 3 weeks is required for affected patients. Asymptomatic contacts are treated with doxycycline for 1 week or a single dose of azithromycin. This case of locally acquired LGV highlights the features of this progressive disease that may now be recognised more frequently in Australian men who have sex with European or North American men. Histological section of the lymph node showing the thickened node capsule and necrotising granuloma Image courtesy of Dr Malcolm Buchanan, Department of Anatomical Pathology, Royal Melbourne Hospital.

Damon P Eisen

To the Editor: A 47-year-old woman presented with a 1-week history of spontaneous prominent and painful bruising and haematomata, 5–6 weeks after commencing meloxicam 15 mg daily for osteoarthritis and plantar fasciitis. The bruises varied in size from 2 cm × 2 cm to 3 cm × 4 cm. She had a past history of acne rosacea, for which she was taking clonidine 100 μg daily, and reflux oesophagitis, for which she was taking pantoprazole 40 mg daily. Meloxicam, being the only new drug taken by the patient, was suspected as the causative agent and was therefore discontinued. Activated partial thromboplastin time (APTT) at presentation was 58 s (reference range, 22 s –38 s). A week after stopping meloxicam, it had fallen to 37 s. All other haematological parameters, including platelet count and international normalised ratio, were normal, as were renal and liver function. New bruises and haematomata stopped appearing 2 days after the patient stopped taking meloxicam. She continued taking clonidine and pantoprazole. Meloxicam is a selective nonsteroidal anti-inflammatory drug (NSAID) (a cyclo-oxygenase 2 [COX-2] inhibitor). It was chosen for this patient in view of her reflux oesophagitis and because she had already been unresponsive clinically to one of the other COX-2 inhibitors. Unlike non-selective NSAIDs, meloxicam has been shown to have negligible effect on platelet function as measured by bleeding time.1-3 Rinder et al4 reported no prolongation of APTT or prothrombin time after 8 days of regular administration of meloxicam at 7.5 mg, 15 mg or 30 mg. In spite of these contrary findings, I believe the prolongation of APTT and spontaneous bruising and haematomata in this patient were directly attributable to meloxicam, as the bruises disappeared 2–3 days after stopping the drug (with no other changes in the patient’s existing medication) and a repeat APTT a week after cessation of the drug was normal.

Anil M Kurien

Diane A Longstreet,* Deanne L Heath,† Robert Vink‡ * Dietitian, † Research Scientist, Townsville Aboriginal and Islander Health Services, 57–59 Gorden Street, Garbutt, QLD 4814; ‡ Head, Department of Pathology, University of Adelaide, SA. dlongstreetATtaihs.net.au To the Editor: Diabetes in Indigenous Australians occurs at a younger age and at almost four times the rate in non-Indigenous Australians. The age-adjusted prevalence of diabetes among Indigenous people is 16% in remote areas and 9% in non-remote areas, with the actual prevalence estimated to be between 20% and 25%, and possibly higher than 30% in some remote areas.1 The cause for this disparity in diabetes incidence is multifactorial, and recent evidence suggests that nutrition — particularly magnesium intake — may play a role. Although central obesity remains a major risk factor, magnesium deficit has been posited to be an underlying common mechanism for the insulin resistance found in type 2 diabetes, as well as in metabolic syndrome, hypertension, and impaired glucose tolerance.2 The clinical correlations between low magnesium and diabetes have been well documented,3 with serum magnesium deficits being reported in 25%–39% of diabetic outpatients in the United States and Switzerland, and up to 73% of diabetic outpatients in Mexico. With magnesium deficits being observed in diabetes, studies examining the effects of magnesium-rich foods on diabetes risk become relevant. The Nurses’ Health Study and the Health Professionals’ Follow-up Study, which included 85 060 women (18 years follow-up) and 42 872 men (12 years follow-up), demonstrated that, after adjusting for confounding variables, a magnesium-rich diet reduced the relative risk of developing diabetes by 34% in women and 33% in men.4 A similar inverse correlation between magnesium intake and diabetes risk was shown in the Iowa Women’s Health Study with a cohort of 35 988 older women,5 and in the Honolulu Heart Program and the Women’s Health Study with cohorts of 8006 men and 39 345 women, respectively.6,7 Despite this growing body of evidence supporting the involvement of magnesium in diabetes, consideration of magnesium status has not been integrated into Australian medical care for diabetes, and more specifically, for Indigenous Australians. It is known that the traditional diet of hunter-gathers such as Indigenous Australians was much more nutrient- and magnesium-rich than the current estimated Australian intake.8 Nonetheless, there remains a lack of information about current magnesium status, including dietary intake, in Indigenous Australians. It is possible that dietary magnesium intake may be too low to maintain normal serum magnesium homoeostasis, and that this might contribute to the development of type 2 diabetes. Further research into this issue may provide this information.

Diane A Longstreet · Deanne L Heath · Robert Vink

To the Editor: The clinical update on venous thromboembolism by Lee and colleagues advises that “Ventilation perfusion (V/Q) isotope scanning reliably establishes the diagnosis of PE [pulmonary embolism] if the V/Q features suggest a high probability of PE . . .”.1 Although this is probably true for patients with intermediate or high pretest probability, a discordant result (low pretest probability and high probability V/Q) should be regarded with suspicion. From the original PIOPED data, high probability V/Q was predictive of angiographically confirmed PE in 80% of patients,2 which drops to 56% by Bayesian analysis if the pretest probability is low. False positive results may be due to previous PE or unrelated parenchymal lung disease. There is significant potential morbidity associated with a false positive result for PE, both from the acute anticoagulation and for future presentations with PE-type symptoms, where PE will be accorded a higher probability because of the previous documented diagnosis. As Lee and colleagues also state, D-dimer testing must be combined with an estimate of pretest probability to be useful. They advocate excluding PE on the basis of low pretest probability and negative D-dimer result. However, a negative D-dimer result (rapid enzyme-linked immunosorbent assay [ELISA] type) may be used to exclude PE in intermediate as well as low probability patients.3 This is dependent on the type of assay available as well as the local PE prevalence, and local guidelines should therefore be developed.

Matthew J Bragg

To the Editor: I read with interest the article by Lee et al regarding the investigation and treatment of pulmonary embolism (PE).1 The investigation of patients presenting with PE as a diagnostic possibility is of great interest to emergency physicians, and such presentations are a daily occurrence in emergency departments around the country. Unfortunately, only a small amount of text is devoted to describing the relative merits of ventilation perfusion (V/Q) scanning and computed tomography pulmonary angiography (CTPA), and no guidance is provided as to which is the test of choice when both are available. The British Thoracic Society has recommended CTPA as the lung imaging modality of first choice for patients presenting with non-massive PE.2 There is a large and increasing body of evidence that CTPA provides superior specificity to V/Q scanning in the detection of PE. CTPA also provides the opportunity of establishing diagnoses other than PE and, in addition, a negative multi-slice CTPA is of sufficient sensitivity to enable the withholding of anticoagulation.3 It is also my experience that CTPA is easier to obtain out of hours, compared with V/Q scanning. The authors state that V/Q scanning “reliably establishes the diagnosis of PE if the V/Q scan features suggest a high probability of PE . . .”.1 Unfortunately, this statement is incorrect. It is essential that V/Q scan results be interpreted in the light of the patient’s clinical probability for PE. In the PIOPED study, only 56% of patients with high probability V/Q scan reports had pulmonary embolism if the pretest probability was low.4 No mention is made of the special situation of pregnant women presenting with pleuritic pain, or which lung imaging test is considered “safest” for both mother and baby. Although the risks of PE are generally agreed to be increased in pregnancy, it is my experience that pregnant women are extremely reluctant to undergo any form of diagnostic investigation that exposes the fetus to radiation. The Wells criteria have been validated for the assessment of PE in emergency department patients only, and provide a means for clinicians with little experience to make an accurate assessment of an individual patient’s clinical probability of PE.5 Once initiated, clinical assessment of the patient with possible PE is straightforward. The key question facing emergency physicians is this: is there a group of patients that have such low probability for PE that no investigation at all is required?

Paul M Bailey

John W Eikelboom,* Graeme J Hankey,† Wai Khoon Ho,‡ Cindy H Lee§ * Haematologist, Thrombosis Service, McMaster University, HHS General Divison, 237 Barton Street East, Hamilton, ON L8L2X2, Canada; † Neurologist, ‡ Fellow in Haematology, § Senior Registrar in Haematology, Royal Perth Hospital, Perth, WA. eikelbj@mcmaster.ca In reply: Pulmonary embolism (PE) remains a complex diagnosis despite the availability of validated prediction models and D-dimer testing to direct the need for diagnostic imaging. We agree with Bailey that the ability to exclude the diagnosis of PE on clinical grounds in patients with a low pretest probability is highly desirable. Unfortunately, clinical features lack sensitivity and specificity for the diagnosis of PE, and clinical prediction models, laboratory investigations, and diagnostic imaging are likely to remain an integral part of the clinical work-up. As suggested by Bragg, it may be possible to simplify the diagnostic approach by using a highly sensitive D-dimer assay, and simplified pretest probability models have been proposed. However, this may come at a cost of reduced specificity,1 which leads to unnecessary diagnostic imaging studies and thus limits the clinical utility of these approaches. Further improvements in the diagnostic approach to PE are clearly needed. There are emerging data demonstrating the accuracy of computed tomography pulmonary angiography (CTPA) for the diagnosis of PE. However, CTPA has limitations (a large contrast load, high radiation dose, and lack of sensitivity of first generation scanners for small thrombi2), some of which are evident in the recently published validation study referred to by Bailey:3 25% of screened patients with suspected PE were not eligible for this study because of renal impairment, a contraindication to CT, or other reasons. The diagnostic algorithm that we provided in our review suggests that either ventilation perfusion (V/Q) scanning or CTPA can be used for patients with suspected PE who require diagnostic imaging,2 with the choice determined by patient factors and availability. The diagnosis of PE during pregnancy is challenging because of concerns about radiation exposure and uncertainty about whether CTPA or V/Q delivers more radiation to the fetus.4 Furthermore, clinical decision rules have not been validated in pregnancy. However, recommendations from experts and professional bodies suggest that V/Q scanning can be used in combination with compression ultrasound to establish or exclude the diagnosis of PE during pregnancy in most cases with minimal fetal radiation exposure.5,6 The comments by Bailey and Bragg concerning the interpretation of high probability V/Q scan results highlight the pitfalls of performing diagnostic imaging without considering the patient’s pretest probability of PE. Although a high probability V/Q scan is diagnostic in patients with a moderate or high pretest probability of PE (prevalence of disease ≥ 90%), the prevalence of disease is only about 50% in those with a low pretest probability.7,8 Therefore, V/Q scanning should not be performed in patients with a low pretest probability unless the D-dimer test is positive. In this situation the algorithm for moderate or high pretest probability should be followed,2 and a high probability scan reliably establishes the diagnosis.

John W Eikelboom · Graeme J Hankey · Wai Khoon Ho · Cindy H Lee

Richard F O’Reilly,* Ian A Burgess,† Bernard Zicat‡ * Physician, † Radiologist, ‡ Orthopaedic Surgeon, Mater Misericordiae Hospital, Rocklands Road, North Sydney, NSW 2060. roreillyATbigpond.net.au To the Editor: In a recent editorial, Gallus estimates the cost of doing ultrasonography in all patients after unilateral hip or knee replacement, with further testing in the 9% or 26% of patients, respectively, found to have deep vein thrombosis (DVT), to be about $200 000 per 1000 patients.1 We agree. He then states, “Many would argue that extended prophylaxis is likely to be the simplest, cheapest and perhaps safest solution”. However, prophylaxis is also expensive. Subcutaneous enoxaparin 40 mg administered daily for 30 days costs $170, or $170 000 per 1000 patients.2 In our study, we found DVTs in 1086 of 5999 patients (18.1%) before discharge,3 so that extended prophylaxis would involve 81.9% of patients receiving prophylactic doses of anticoagulants, with the risk of unwanted bleeding, despite the absence of DVT on ultrasound at Day 7 postoperatively. In addition, if an ultrasound scan was not done before discharge, the 18.1% of patients with a DVT would receive only prophylactic (not therapeutic) doses of anticoagulant for their DVT. We plan a further study to check the prevalence of post-discharge DVT by repeating ultrasonography at 90 days postoperatively in patients without DVT on ultrasound at Day 7. We suspect the prevalence is lower than suggested in the literature, as the data on late presentation of DVTs have been obtained by retrospective study of the number of patients re-admitted to hospital with DVT. Finally, on the question of whether performing ultrasonography on all patients has clinical benefit, we concur with Gallus when he writes that “Logic suggests it should . . .”.

Richard F O’Reilly · Ian A Burgess · Bernard Zicat