Endocrinology

In reply: Seeman and colleagues agree that most patients with minimal trauma fractures do not have osteoporosis. The figures are clear: only 13% of patients with a peripheral fracture have a hip bone mineral density (BMD) T-score less than or equal to − 2.5 and only 25% have a score less than or equal to − 1.5. The corresponding figures for vertebral fractures are 25% and 38%, respectively.1 We do not argue that the − 2.5 T-score threshold for defining osteoporosis is sacrosanct, but simply that some bone density threshold be defined for subsidised therapy, for which virtually all the supporting evidence is based on treatment of patients with established osteoporosis. Osteopenia is an artificial concept with an arbitrary definition, but we agree that the T-score threshold for subsidised therapy need not be as low in those with prevalent adult fracture as in those without — perhaps − 1.5, which is the threshold recently adopted for patients receiving corticosteroid therapy. We disagree about the predictive power of bone densitometry; it is comparable to that of blood cholesterol level for heart attacks and blood pressure for stroke.2 It therefore makes sense to measure BMD in all women at menopause and all men at age 60 years to identify those with osteoporosis before they sustain fractures, as well as those with normal but negative T-scores, who have a fracture risk twice that of those with positive T-scores.3 Those with proven osteoporosis could receive subsidised therapy, and those with low normal values could be advised on lifestyle measures, such as calcium supplementation (which significantly delays or prevents bone loss in postmenopausal women).4 People with positive T-scores can be reassured. To suggest that no trials have demonstrated the antifracture efficacy of nutritional measures is to argue against three large meta-analyses showing significant prevention of fractures with vitamin D and calcium supplementation.5-7 The additional cost of confirming low bone density before providing subsidised therapy in fracture cases is likely to be more than offset by the savings from reduced inappropriate therapy; bone densitometry costs about $80 per test, but bisphosphonate therapy costs about $50 a month for each patient. The extra cost of bone densitometry for every woman at menopause and every man at age 60 years could be $20 million a year, but even with subsidised therapy for those without fracture but proven osteoporosis (with a T-score less than or equal to − 2.5, for instance), the cost is also likely to be more than offset in the long term by reducing the enormous cost of osteoporotic fractures ($8 billion annually8). We find it hard to understand why any of our colleagues would not support proposals that would transfer treatment from those who do not need it to those who do.

B E Christopher Nordin · Michael Horowitz

To the Editor: The ASPREE (ASPirin in Reducing Events in the Elderly) study may provide useful data on the benefits and risks of aspirin therapy in patients aged ≥ 70 years, as described by Woods and colleagues.1 However, the decision to allow general practitioner co-investigators to “help decide whether the patient is a suitable candidate for the placebo-controlled trial” introduces a source of selection bias that may limit the generalisability of the results. Without pre-specified objective selection criteria, it is likely that primary-prevention patients assessed by GP co-investigators as being at high vascular risk will be excluded because the GPs believe they should be taking antiplatelet agents. Similarly, those at low risk may be thought inappropriate participants because the risks of random allocation to this therapy might outweigh the perceived benefits, as has been shown in previous meta-analyses.2,3 ASPREE may end up with a disproportionate number of intermediate-risk patients. In the case of diabetes, a recent observational study from our group highlighted patients with diabetes and retinopathy and those taking a sulfonylurea as being at increased risk of complicated peptic ulcer disease.4 By contrast, we did not find that aspirin use, positive serological results for Helicobacter pylori, or the interaction of these two factors predicted complicated peptic ulcer disease. If GP co-investigators were aware of these findings, they might also influence the screening and recruitment of patients with diabetes to ASPREE. According to the trial registration details (ISRCTN83772183), patients with diabetes were eligible for recruitment to ASPREE from late February 2009, even though the trial started 6 years ago.5 Given this delayed eligibility, the fact that a substantial proportion of patients with diabetes older than 70 years will already have vascular disease, and the expected total sample size of 19 000,1 the trial might include fewer than 1000 patients with diabetes and thus have insufficient statistical power to assess the risks and benefits of aspirin for primary prevention in this important subgroup. We question why subjective assessment forms part of patient selection for a potentially important study such as ASPREE, and also what steps the investigators are taking to determine whether the sample they recruit is representative. In addition, details of planned statistical analyses involving diabetic participants in this non-superiority trial would be reassuring.

Timothy M E Davis · Brett A Sillars · Wendy A Davis

In reply: Recruitment to clinical trials through general practice is representative of the population, as a high proportion of all Australians regularly attend their general practitioners.1 GP co-investigators are appropriate to decide whether their patients are suitable for the ASPREE (ASPirin in Reducing Events in the Elderly) study because their assessment includes objective inclusion and exclusion criteria that must be satisfied before enrolment in the study (clinical trial registration number ISRCTN83772183),2 as well as patient-specific potential risks with using aspirin, and known medical factors likely to influence patient survival during the trial. These include the risk of complicated peptic ulcer disease in patients with diabetes treated with a sulfonylurea.3 GP co-investigators support participation in ASPREE by eligible patients because of aspirin’s therapeutic equipoise for primary prevention in older patients4 and in those with diabetes.5 Because of age alone, ASPREE participants will be at least at intermediate risk of cardiovascular disease and also at increased risk of bleeding. Determining the aspirin balance underpins the importance of collecting more data in older people, who have been under-represented in previous primary prevention trials. ASPREE is a superiority trial with pre-specified subgroup analyses, including for the subgroup with diabetes.2 The study is powered to address the primary question reliably in the total cohort rather than subgroups. To date, fewer than 500 participants have been randomly allocated, with recruitment slowed subject to National Institutes of Health funding deliberations. Recruitment will be reinvigorated in late 2009, and will continue to include people with diabetes.

Robyn L Woods · Mark R Nelson · Andrew M Tonkin · Christopher M Reid

To the Editor: In Australia, patients with type 1 diabetes and private health insurance are eligible for rebates on the purchase price of insulin pumps if deemed necessary for treatment. In contrast, hospital-funded or donated pumps are often used by non-insured patients. Hospitals may provide pumps to certain patients for various reasons — for example, to pregnant women (to improve their glycaemic control), to patients who want to try the pump to determine their preference or their ability to use it, or to patients waiting for private health insurance cover to be activated. Patient selection is important, as insulin pumps are cost-effective only if they reduce levels of glycated haemoglobin (HbA1c) and the frequency of hypoglycaemia1 — although quality of life may also be an important benefit. We conducted a study to compare outcomes for patients with public hospital-funded pumps (Group A) with outcomes for those with private health insurance-funded pumps (Group B). All pump starts between June 2000 and January 2008 at Fremantle Hospital and Rockingham General Hospital in Western Australia were assessed. HbA1c levels before and 6 months after pump initiation were recorded. Diabetes-related hospital admissions over a 1-year period before and a 1-year period after pump commencement were recorded using hospital software (TOPAS KEA! 340, version 5.106) that tracked admissions to all hospitals within the greater metropolitan area of Perth. Patients were excluded from our study if they had type 2 diabetes; had commenced pump use at a different hospital; had moved during the study period to a region not captured on the database; or had used a pump for less than a year (this last exclusion criterion was to ensure that admission rates for the subsequent 12 months were representative of the influence of pump therapy). We identified 109 patients (32 in Group A, 77 in Group B). There were no significant differences between the two groups in age, diabetes duration, initial HbA1c levels (9.2% v 8.7%; P = 0.29) or sex, although the proportion of females was higher in both groups (65.6% and 70.1%, respectively). Patients in Group A had more hospital admissions than those in Group B before and after commencement of pump therapy (0.7 v 0.2 admissions/year before [P = 0.02]; 0.7 v 0.2 admissions/year after [P = 0.04]). After commencing pump therapy, HbA1c levels fell significantly in Group B patients (8.7% v 8.0%; P < 0.005) but not in Group A patients (9.2% v 8.9%; P = 0.17). The mean interval between pump initiation and follow-up HbA1c readings was similar in both groups (10.3 months [Group A] v 10.8 months [Group B]; P = 0.70). There was no significant difference in diabetes-related admissions before and after commencement of pump therapy in either group.

Ken Y Thong · P Gerry Fegan · Bu B Yeap

Objective: To provide an estimate of the morbidity and mortality resulting from abdominal overweight and obesity in the Australian population.Design and setting: Prospective, national, population-based study (the Australian Diabetes, Obesity and Lifestyle [AusDiab] study).Participants: 6072 men and women aged ≥ 25 years at study entry between May 1999 and December 2000, and aged ≤ 75 years, not pregnant and for whom there were waist circumference data at the follow-up survey between June 2004 and December 2005.Main outcome measures: Incident health outcomes (type 2 diabetes, hypertension, dyslipidaemia, the metabolic syndrome and cardiovascular diseases) at 5 years and mortality at 8 years. Comparison of outcome measures between those classified as abdominally overweight or obese and those with a normal waist circumference at baseline, and across quintiles of waist circumference, and (for mortality only) waist-to-hip ratio.Results: Abdominal obesity was associated with odds ratios of between 2 and 5 for incident type 2 diabetes, dyslipidaemia, hypertension and the metabolic syndrome. The risk of myocardial infarction among obese participants was similarly increased in men (hazard ratio [HR], 2.75; 95% CI, 1.08–7.03), but not women (HR, 1.43; 95% CI, 0.37–5.50). Abdominal obesity-related population attributable fractions for these outcomes ranged from 13% to 47%, and were highest for type 2 diabetes. No significant associations were observed between all-cause mortality and increasing quintiles of abdominal obesity.Conclusions: Our findings confirm that abdominal obesity confers a considerably heightened risk for type 2 diabetes, the metabolic syndrome (as well as its components) and cardiovascular disease, and they provide important information that enables a more precise estimate of the burden of disease attributable to obesity in Australia.

Adrian J Cameron MPH · David W Dunstan PhD · Neville Owen PhD · Paul Z Zimmet MD, PhD · Elizabeth L M Barr MPH · Andrew M Tonkin MD · Dianna J Magliano PhD · Shirley G Murray GradDipPractMan · Timothy A Welborn PhD · Jonathan E Shaw MD

To the Editor: We concur wholeheartedly with Gill and colleagues1 in support of recognising obesity as a public health issue, and we dispute claims that the current problem of obesity is being exaggerated. Gill and colleagues point out that obesity trends have climbed over decades, and state that 6%–8% of Australian school children are affected.1 While this is a substantial burden of over a quarter of a million children, we also consider that restricting definitions of obese to arbitrary cutoff points may underestimate the problem, given that the entire distribution of childhood weight is increasing, not just the extreme group classified as obese. Adiposity is related to cardiovascular outcomes such as myocardial infarction and stroke in a (curvi)linear fashion. Defining obesity by arbitrary cutoff points is vulnerable to differences between sexes, ethnicity and age, and limits our understanding of obesity-related diseases. It is well known that cardiovascular risk factors cluster, particularly the adiposity-driven components of the so-called metabolic syndrome. In the Western Australian Pregnancy Cohort (Raine) Study, we have used cluster analysis to identify a group of children at risk of future cardiovascular disease with features of the metabolic syndrome.2 The differences in characteristics are shown in the Box. The “high risk” and “low risk” cluster groups differ widely in terms of not only body mass index, the most widely used measure of obesity, but also waist circumference (a measure of central adiposity), insulin resistance, blood pressure, and levels of triglycerides, high-density lipoprotein cholesterol, total cholesterol (data not shown) and low-density lipoprotein cholesterol (data not shown). Not only the conventionally used 95% confidence intervals, but also the 99% confidence intervals do not overlap for any of these intermediate cardiovascular risk factors. We found that 29% of children were in the high-risk cluster at the age of 14 years2 and a similar analysis suggested that even at age 8 years, 25% of children were at increased risk of future obesity, cardiovascular disease and diabetes.3 C-reactive protein (CRP) level is known to be associated with future cardiovascular diseases in adults,4 and with an adverse metabolic profile in children.5 The “high risk” children had significantly higher CRP levels at the age of 14 years than their low-risk counterparts. Certainly, the magnitude of this problem, affecting up to a third of our youth, needs to be addressed by government and health-planning bodies. We suggest our approach of cluster analysis will help identify earlier those children at substantially increased risk of cardiovascular and other adiposity-related disorders in Australia. Features of the cluster groups with respect to components of the metabolic syndrome, showing 99% CIs* BMI = body mass index. HOMA = homeostatic model assessment (for quantifying insulin resistance). SBP = systolic blood pressure. HDL = high-density lipoprotein cholesterol. * From Huang et al.2 Reprinted with permission from the American Diabetes Association.

Rae-Chi Huang · Fiona J Stanley · Lawrence J Beilin

To the Editor: There is a substantial volume of evidence from a range of national and state-based surveys illustrating increases in the rates of obesity and overweight among Australian children over the past two decades,1 concurring with trends observed in most developed countries.2 The recent article by Gill and colleagues highlighted questions that have been raised publicly regarding the extent and impact of levels of obesity and overweight among Australian children, including whether trends have been exaggerated.1 To examine these issues using the latest data available, we present data from the three most recent national surveys in which weight and height of Australian children were measured: the Australian Health and Fitness Survey (1985),3 the National Nutrition Survey (1995),4 and the Australian National Children’s Nutrition and Physical Activity Survey (2007).5 We examined overweight and obesity levels among young Australians from comparable age groups at three time points over more than 20 years, using the same internationally accepted definitions of childhood overweight and obesity. For 1985 and 1995 data, we used the figures reported by Magarey et al in 2001,6 which compared results from the 1985 and 1995 surveys using new standard international definitions to classify overweight and obesity among Australian children and adolescents.7 We calculated body mass index for the 2007 Australian National Children’s Nutrition and Physical Activity Survey using the raw data file obtained through the Australian Social Science Data Archive,8 categorising children as overweight or obese based on the same international definitions used by Magarey et al.6 We based our calculations on the age group common to each of the three surveys: 7–15-year-olds. As shown in the Box, the prevalence of overweight and obesity in boys aged 7–15 years has risen from 11.0% (95% CI, 10.99%–11.01%) in 1985 to 20.0% (95% CI, 19.97%–20.03%) in 1995 and 23.7% (95% CI, 23.68%–23.72%) in 2007. In 7–15-year-old girls, the prevalence of overweight and obesity has increased from 12.2% (95% CI, 12.19%–12.21%) in 1985 to 21.5% (95% CI, 21.47%–21.53%) in 1995 and 25.8% (95% CI, 25.78%–25.82%) in 2007. While data from additional time points are required to map national trends more comprehensively, our analysis clearly indicates an upward trend in overweight and obesity levels in both boys and girls aged 7–15 years between 1985, 1995 and 2007. This trend is cause for alarm, given the widely recognised body of evidence on the significant short-term and long-term consequences of childhood obesity.9 Prevalence of overweight and obesity in Australian children aged 7–15 years, 1985–2007 * Data weighted for age, sex and region with the weighting variable in the raw data file obtained from the Australian Social Science Data Archive.8

Lyn M Roberts · Tessa R Letcher · Alexandra A Gason · Tim Lobstein

To the Editor: Davis and colleagues stated that the Framingham and United Kingdom Prospective Diabetes Study (UKPDS) cardiovascular risk equations are not suitable for predicting risk in an Australian population with type 2 diabetes.1 If confirmed, this would be extremely disappointing. However, before accepting this conclusion the following important considerations should be noted. Davis noted that the Fremantle Diabetes Study (FDS) patient group differed significantly from the UKPDS baseline group (eg, 38% of the FDS patients were aged outside the validated age range of the risk engine [25–65 years] and were assessed by non-validated extrapolation). Similarly, it cannot be assumed that the FDS group is representative of patients in general practice and hospital diabetes clinics around Australia. Moreover, it would be interesting to know how well the engine performs in FDS patients in the age group in which it was validated (ie, patients diagnosed with diabetes at age 25–65 years). It is likely that the low rate of cardiovascular events in the FDS (4.8% with at least one myocardial infarction, and 2.9% with at least one stroke)1 affects the accuracy of the results obtained with the UKPDS risk engine. The Framingham risk score has already been found to vary considerably in accuracy between populations, with predicted-to-observed ratios ranging from underprediction of 0.43 to overprediction of 2.87.2 Further, the UKPDS risk engine recently overestimated the risk of cardiovascular disease events in a UK general practice population.3 In purely pragmatic terms, most patients with type 2 diabetes aged over 50 years are at “high risk” for cardiovascular events (cardiovascular risk of more than 20% over 10 years),4 and the UKPDS risk engine is unlikely to influence prescribing practice significantly. However, we have found the engine to be a useful educational tool for explaining risk to patients. Even if the UKPDS risk engine is not optimally calibrated, the FDS analysis revealed that the coronary heart disease risk equation had modest discrimination (area under the receiver operating characteristic curve [AUC], 0.68), and the stroke risk equation had good discrimination (AUC ≥ 0.86),1 identifying those at highest risk. We believe that, rather than being irrelevant in Australians, the UKPDS risk engine continues to identify those at highest risk for cardiovascular events, operates well within its validated age group, and provides a motivational tool for encouraging changes in patient behaviour. Until a large dataset is pooled from various Australian studies, we believe the UKPDS risk engine should not be discarded.

Roland W McCallum · John R Burgess · Timothy M Greenaway

In reply: We thank McCallum and colleagues for their comments. In relation to their specific points: The Fremantle Diabetes Study (FDS) cohort is representative and drawn from a typical Australian urban centre.1 The 488 cardiovascular disease-free FDS participants with type 2 diabetes who were aged 25–65 years at both diagnosis and study entry had 22 coronary heart disease (CHD) events compared with 72 predicted, with a similar area under the receiver operating characteristic curve (AUC) to that for all 791 patients who were included in the analysis2 (0.66 v 0.68). Calibration indicated significant discrepancies between predicted and actual outcomes (P ≤ 0.02), and positive predictive values were low (≤ 3.5%). Therefore, restricting our patient sample to a “UKPDS” cohort did not alter our conclusions. We agree that the low observed CHD event rate in the FDS compared with that predicted by the UKPDS risk engine undermines its validity in Australians with type 2 diabetes. There was a similarly low CHD event rate in the FIELD study, which included many Australasians.3 Contemporary diabetes care clearly differs from that during the Framingham Study and UKPDS. Although the study cited by McCallum and colleagues, in a UK general practice population, is not strictly comparable to our study, it also found that the UKPDS cardiovascular disease risk engine performed only moderately (AUC, 0.72).4 Accurate risk prediction should be a basis for cost-effective care. We have developed an FDS risk calculator which should improve clinical management for Australians with diabetes.5

Wendy A Davis · Stephen Colagiuri · Timothy M E Davis

One of the dilemmas in delaying sex-assignment surgery is the increased risk of gonadal malignancy Few areas of medicine are as controversial as the management of disorders of sex development (DSD). The use of the term DSD to describe patients born with ambiguous genitalia has undergone major change from older terms with negative connotations, such as “intersex”, “testicular feminisation” and “hermaphroditism”.1 Meanwhile, international debate continues about the ethics of performing genital surgery on affected infants and children. In fact, the debate has been raging for more than a decade between the medical profession and patient advocacy groups in Western countries, and has been documented by anthropologist Katrina Karkazis in a recent book.2 A long-term outcome study of 50 patients aged 18–32 years who had been treated in Melbourne when they were children showed that mental and physical health outcomes were as good for most of the DSD patients as for those in two control groups; however, there was a small minority of patients whose gender identity as adults was a source of such profound discomfort that they felt compelled to undergo treatment to change it.3 Clearly, this is unsatisfactory, and management practices have been reviewed internationally by clinicians looking for ways of minimising the risk of making such mistakes about gender assignment. The main problem relates to feminising genitoplasty (Box), which involves the removal of phallic erectile tissues and skin that cannot be replaced. This type of operation is considered appropriate for 46,XX girls with congenital adrenal hyperplasia (Box), who rarely identify as male when they are adults if they are treated with appropriate hormones to maintain androgen suppression from soon after birth and throughout childhood.4 However, feminising genitoplasty is much more of a problem in patients with a Y chromosome. For example, in one study of 14 adult patients with genetically confirmed partial androgen insensitivity who were treated at Johns Hopkins University in the United States as children, 25% experienced gender dysphoria (Box) as adults, and a small number wanted to undergo sex change surgery.5 Although policy changes are still being discussed, it seems likely that fewer and fewer XY patients with frankly ambiguous genitalia due to DSD will have feminising genitoplasty and be raised female. The option to assign a gender but postpone surgery until the child is able to give consent has been strongly advocated in some quarters,6 but has not gained much traction because of concerns that children might suffer psychological harm if left with ambiguous genitalia. In 2008, clinicians from Melbourne’s Royal Children’s Hospital, recognised for their expertise in the management of DSD, were required to meet representatives of the Victorian state Justice Department. They were asked to respond to a proposal — advanced by an advisory committee representing the interests of the gay, lesbian, bisexual, transsexual and intersex communities — that doctors wanting to perform surgery to treat ambiguous genitalia in children too young to consent on their own behalf should have to seek approval from the Family Court of Australia on a case-by-case basis. Also in 2008, the Australian Human Rights Commission decided to initiate a public inquiry into the same question, and circulated a draft discussion paper called Genital surgery for babies born intersex to health professionals for comment. Thus, in Australia as elsewhere, the arm wrestle between medical professionals and patient advocacy groups continues. What has largely been missing from the debate is recognition of the fact that surgery forms a necessary part of the risk management strategy for preventing gonadal malignancy. In any DSD associated with a Y chromosome, there is an increased risk of germ cell cancer,7 especially when the testes are intra-abdominal (the risk of seminoma in partial androgen insensitivity is 50% for an intra-abdominal testis) or when there is gonadal dysgenesis. In this issue of the Journal, a salutary case report by Parker and colleagues8 reminds us of the need to be mindful of this risk, and also to take a long-term view of risk. If the intra-abdominal gonad in the patient described had been removed at the initial surgery, he would never have needed to fear this tumour. It had not been removed because, by today’s standards, he had been inadequately investigated in the past, and therefore the intersex condition was not recognised. The trend for surgeons to recommend male-sex rearing for greater numbers of children with DSD could also mean greater reluctance to remove testes that pose a significant risk of cancer on the grounds that physiologically useful hormone secretion might be retained. It is therefore imperative that a risk management strategy be prepared for each patient. This would mandate: educating parents and patients about risk; removing all intra-abdominal gonads that cannot be brought down into the scrotum; regular clinical and ultrasound surveillance of scrotal gonads with removal of any that contain suspicious lumps; biopsy of testes after the onset of puberty, looking for early signs of malignant change; and effective communication between paediatric and adult care-providers at the time of transition. It is also important for all children identified as having DSD to be referred to a centre of excellence where they will be seen by paediatric endocrinologists, surgeons and other health care professionals with expertise in the field and who recognise the importance of a multidisciplinary team approach.9 Case conferences about patients diagnosed as having a DSD in adult life would be enhanced if paediatric specialists in DSD were asked to comment. Of equally great importance is the need for an accurate aetiological diagnosis wherever possible. At the moment, about 40% of patients with 46,XY forms of DSD are left without a precise diagnosis.10 The application of microarray (gene chip) technology,11 which is available in Australia, is an exciting and promising step forward in identifying genetic mutations. In this technique, samples of very large numbers of genes are arranged in a regular pattern on a solid surface or membrane, which is then incubated with DNA from a patient. Alterations in known (and even unknown) genes are rapidly detected by studying patterns of matches and mismatches. The current challenge for researchers is to develop new tools, such as microarray technology, that will lead to gene discovery and to better methods of screening patients for mutations in all the known genes. Glossary of terms relating to disorders of sex development DSD: Disorders of sex development, previously known as intersex. Congenital conditions in which development of the chromosomal, gonadal or anatomical sex is atypical. Feminising genitoplasty: Surgery carried out to give genitalia that were originally ambiguous a more female appearance. Usually involves clitoral reduction (removal of erectile tissue) and surgery to create a vaginal opening separate from the urethra. Congenital adrenal hyperplasia: A genetic disorder caused by a deficiency of the enzyme 21-hydroxylase in the adrenal cortex, and the commonest adrenal disorder of childhood. Cause of virilisation in an affected female fetus. Partial androgen insensitivity: An X-linked genetic disorder causing ambiguous genitalia in 46,XY individuals. Caused by a lack of androgen receptors in androgen target tissues, such as genital skin. Gender dysphoria: Mental distress caused by unhappiness with one’s own sex and the desire to be identified as the opposite sex.

Garry L Warne MB BS, FRACP · Jacqueline K Hewitt MB BS

To the Editor: The management of pregnant women with pre-existing diabetes is often challenging for clinicians. A 38-year-old woman presented with an unplanned pregnancy at 8 weeks’ gestation. She had a 4-year history of type 2 diabetes treated with metformin, which she stopped taking on confirmation of her pregnancy. Her glycated haemoglobin (HbA1c) level was 9.0% at her first antenatal visit, and therapy with pre-meal insulin aspart and twice daily isophane was commenced. Her insulin requirement rapidly escalated, rising to 400 units per day by the end of the first trimester. Metformin therapy was reintroduced in the second trimester. Despite strategies including splitting her insulin doses and trialling different insulin regimens, control of her diabetes remained poor. At 30 weeks’ gestation, U-500 insulin became available, and that allowed rapid titration of insulin doses (Box). Her HbA1c level improved to 6.4% in the third trimester. At 35 weeks, her daily insulin requirement had reached 1455 units and her membranes ruptured prematurely. During labour, an intravenous insulin infusion rate of 90 units per hour was needed to achieve normoglycaemia. At birth, the neonate weighed 2005 g (11th percentile) with no evidence of congenital abnormalities, but she suffered transient hypoglycaemia on Day 1. Postpartum, the mother’s daily insulin requirement decreased to 28 units per day. In pregnant women with pre-existing type 2 diabetes, the insulin requirement increases substantially in the second half of pregnancy.1 In the past 12 months, among 25 pregnant women with type 2 diabetes who attended our antenatal clinic, the median insulin dose at the end of their pregnancies was 123 units per day, with 6 women requiring doses exceeding 200 units per day. At daily doses above 200 units, the therapeutic response to further increments in the insulin dose is attenuated.2 Current insulin preparations in Australia (100 units/mL) can be problematic for these patients as it is difficult to administer large volumes of insulin subcutaneously. U-500 is a preparation of regular insulin at a concentration of 500 units/mL. It is invaluable for patients with extreme insulin resistance, as a smaller volume is needed for injections. Its application during pregnancy has been previously reported.3-5 U-500 insulin is not readily available in Australia, and the preparation has to be administered by syringe. Hence, the volume of insulin must be drawn up accurately, as insulin syringes in Australia are designed for conventional lower-concentration insulin preparations (100 units/mL). This case highlights one of the many difficulties in managing pregnant women with pre-existing diabetes. The use of U-500 may be considered for women on extremely large doses of insulin and whose diabetes is still suboptimally controlled. The safety aspects of U-500 during pregnancy will need further examination. Insulin requirements during pregnancy of a 38-year-old woman with diabetes

Vincent W Wong · Alexia V Pape

To the Editor: We believe the current indications for subsidised treatment of osteoporosis specified in the Pharmaceutical Benefits Schedule (PBS) encourage over-prescribing on the one hand, yet, on the other, deny many patients with osteoporosis the treatment they need. For patients under 70 years of age, the PBS indication for specific treatment, such as bisphosphonate therapy, is “established osteoporosis with minimal trauma fracture”. Thus, perhaps surprisingly, treatment is indicated for secondary prevention only. More remarkable is that patients do not need to have osteoporosis to receive the benefit: repeated enquiries to Medicare (the most recent on 3 March 2009) have confirmed that prior measurement of bone mineral density (BMD), the only practical way to diagnose osteoporosis, is not required. Yet it is well known that only 20% of women with peripheral fractures from non-major trauma actually have osteoporosis, whatever BMD T-score is used diagnostically.1 Accordingly, the current policy simultaneously denies specific treatment to patients with osteoporosis who have not yet sustained a fracture while subsidising treatment to patients with fractures who are unlikely to have osteoporosis. There is little gain from bisphosphonate therapy in women who have normal BMD and no vertebral fracture,2 but good evidence that such therapy is effective if BMD is low.3 To remedy these anomalies, we believe that bone densitometry should be more readily available — not deferred until people have fractures or reach the age of 70 years. We support a bone density measurement for all women at the menopause (and perhaps all men at age 60 years). This would identify those with osteoporosis at high risk of fracture as well as those in the low-normal range who are at high risk of developing osteoporosis.4 The first group could be offered specific therapy to prevent fractures and the second group could be advised on preventive lifestyle measures such as calcium and vitamin D supplementation and appropriate exercise. We estimate that the cost, even if there was full acceptance, would be only about $20 million a year compared with the current $8 billion yearly cost of osteoporotic fractures.5 Our recommended policy revision could pay for itself many times over, even if there were only a 10% reduction in fracture rate, not to mention improvements to be gained in the quality of patients’ lives. Early recognition of low bone density and early diagnosis of osteoporosis has the long-term potential to transform the current depressing osteoporosis picture. We do not discount the importance of minimal trauma fracture, but believe that more weight should be given to vertebral fractures and less to peripheral fractures, as the former are much more likely to be osteoporotic, much more liable to recur,6,7 and much more responsive to specific therapies.3

B E Christopher Nordin · Michael Horowitz · Barry E Chatterton

Fixing sex. Intersex, medical authority, and lived experience. Katrina Karkazis. New York: Duke University Press, 2008 (xiii + 364 pp). ISBN 978 0 8223 4318 9. Currently, there is an intense ethical debate about genital surgery for infants born with ambiguous genitalia. The controversy rose to a new level of intensity in Australia in 2008 with the involvement of the Australian Human Rights Commission and the Victorian Government Department of Justice. Doctors in Europe and North America are facing the same dilemmas. This new book, possibly the best contribution to the debate yet published, is very welcome, not only because it is timely but because it is deeply thoughtful, thoroughly researched and very respectful of all points of view. The author, Katrina Karkazis, PhD, MPH, is a Senior Research Scholar with the Center for Biomedical Ethics at Stanford University in the United States. In addressing the historical basis for current understanding of sex and gender, Karkazis discusses the contribution to the understanding of sex development made by John Money (a psychologist at Johns Hopkins University) in depth, in a way that is refreshingly generous. She traces the development of what became the traditional treatment model, the scepticism that emerged, and the origins of Internet-based patient advocacy groups in the mid 1990s. Her exploration of what is posted on discussion boards is balanced by her careful study of what scientific long-term outcome studies have, and have not, delivered. She has also conducted hundreds of interviews with doctors, parents and adult patients. Her book concludes with the following:

Garry L Warne

Objectives: To conduct initial analyses and examine ways in which depression and anxiety are associated with outcomes after participation in the Healthy Living Course (HLC), an early-intervention diabetes prevention program for adults with prediabetes.Design: Randomised controlled study using pre-intervention and postintervention measures to examine relationships between depression, anxiety and diabetes-related program outcomes.Participants and setting: 185 adults from urban and rural Victoria with prediabetes who had completed the HLC program and for whom postintervention measure data were available. Data were collected between 15 June 2006 and 15 June 2008.Main outcome measures: Baseline and postintervention scores on mood (anxiety, depression), biochemical (fasting plasma glucose, oral glucose tolerance), anthropometric (body mass index [BMI], waist circumference), cognitive (self-efficacy, diabetes knowledge) and behavioural (healthy eating, physical activity) measures; correlations between these measures.Results: The intervention alleviated depression, and improved eating patterns and scores on cognitive, anthropometric and biochemical measures. Cultural group and sex did not influence most results. Baseline mood was not associated with anthropometric or biochemical outcomes; however, more positive baseline mood factors were associated with activity changes, and with greater subsequent activity rates, self-efficacy and diabetes knowledge. In turn, baseline self-efficacy was associated with postintervention healthy eating. Changes towards healthier eating correlated with anthropometric and biochemical changes, while baseline cognitive measures were also associated with physiological outcomes. As expected, reductions in BMI and waist circumference were related to biochemical changes.Conclusion: Our findings highlight the importance of assessing mood factors in prediabetes, and the need to develop theoretical models of change mechanisms for mood in health outcomes.

Michael Kyrios BA, MPsych, PhD · Susan M Moore BSc(Hons), MEd, PhD · Naomi Hackworth BSc, BA(Hons), DPsych(HealthPsych) · Simone A Buzwell BA(Hons), PhD · Naomi Crafti BBSc(Hons), DPsych(Counselling) · Christine Critchley BA(Hons), PhD · Elizabeth Hardie BA(Hons), PhD

This position statement was prepared by the Working Group of the Australian and New Zealand Bone and Mineral Society and Osteoporosis Australia. The final statement was endorsed by the Endocrine Society of Australia. Currently, the balance of evidence remains in favour of fracture prevention from combined calcium and vitamin D supplementation in elderly men and women. Adequate vitamin D status is essential for active calcium absorption in the gut and for bone development and remodelling. In adults with a baseline calcium intake of 500–900 mg/day, increasing or supplementing this intake by a further 500–1000 mg/day has a beneficial effect on bone mineral density. Calcium intake significantly above the recommended level is unlikely to achieve additional benefit for bone health.

Kerrie M Sanders GradDipDiet, MHumNutr, PhD · Caryl A Nowson DipNutrDiet, PhD · Mark A Kotowicz MB BS, FRACP · Kathryn Briffa BAppSc(Physio), PhD · Amanda Devine GradDipDiet, PhD · Ian R Reid MB BS, FRACP

Objective: To assess the prevalence of metabolic syndrome and its association with sociodemographic, clinical and lifestyle variables among Australian patients with a variety of psychiatric disorders.Design and setting: Cross-sectional study of patients attending a public mental health service in Western Australia between July 2005 and September 2006.Participants: Patients who were aged 18–65 years; diagnosed with schizophrenia, schizoaffective disorder, bipolar disorder, major depressive disorder with psychotic symptoms, drug-induced psychosis or borderline personality disorder; and currently taking at least one antipsychotic drug for a minimum of 2 weeks.Main outcome measures: Prevalence of metabolic syndrome diagnosed with International Diabetes Federation criteria; fasting blood glucose and lipid levels; sociodemographic and lifestyle characteristics.Results: Of 219 patients invited to participate, 203 agreed and had complete data. Prevalence of metabolic syndrome was 54% overall, and highest among patients with bipolar disorder or schizoaffective disorder (both 67%), followed by schizophrenia (51%). Sociodemographic variables, including age and ethnic background, were not significantly associated with metabolic syndrome, but a strong association was seen with mean body mass index. Other cardiovascular risk factors, such as smoking and substance misuse, were common among participants.Conclusions: Prevalence of metabolic syndrome in this population was almost double that in the general Australian population, and patients with schizophrenia had a prevalence among the highest in the developed world. Prevalence was also high in patients with a variety of other psychiatric disorders.

Alexander P John MB BS, MD, FRANZCP · Radhakrishnan Koloth MB BS, DPM · Milan Dragovic PhD · Stephen C B Lim PhD

Objective: To assess the performance of the Framingham and United Kingdom Prospective Diabetes Study (UKPDS) cardiovascular risk equations in Australian patients with type 2 diabetes who were initially free of cardiovascular disease (CVD).Design and setting: The Fremantle Diabetes Study (FDS), a community-based longitudinal observational study; data for the period 1993–2006 were used.Patients: Of the 815 FDS participants with type 2 diabetes who were initially CVD-free, 791 (97%) were eligible for assessment using the UKPDS equations, and 697 (86%) using the Framingham equation.Main outcome measures: CVD endpoints during 5 years of follow-up. For the UKPDS equations, these were fatal myocardial infarction (MI) or sudden death (fatal coronary heart disease [CHD]); hospitalisation for/with or death from MI or sudden death (all CHD); fatal stroke; and all stroke. For the Framingham equation, they were all MI, sudden death or angina pectoris (CHD).Results: During follow-up to first CVD event, death or 5 years, there were 38 MIs (11 fatal) and 23 strokes (13 fatal) in the UKPDS-assessable cohort of FDS participants. The UKPDS risk equations for all CHD, fatal CHD, and all stroke overestimated the number of events by 6.5, 2.8 and 1.8 times, respectively. The risk equation for fatal stroke underestimated the number of events by 38%. The UKPDS CHD risk equations showed modest discrimination and poor calibration, while the stroke risk equations showed good discrimination and calibration. The Framingham equation predicted 28% fewer CHD events than occurred (93 v 130), and discrimination and calibration were poor.Conclusions: While the UKPDS stroke risk equations performed relatively well, the UKPDS and Framingham CHD risk equations are not suitable for predicting risk in Australians with type 2 diabetes.

Wendy A Davis MPH, PhD · Stephen Colagiuri FRACP · Timothy M E Davis DPhil, FRACP

To the Editor: Gilbert and colleagues1 report thyroid function test results in a large number of pregnant women in Western Australia during the first trimester. While assessment of thyroid status is increasingly important in pregnancy, they do not present a strong enough argument for their reference ranges to be adopted. Their controls consisted of only 100 blood donors, and it is not clear whether these were age-matched with patients. Differences between pregnant and non-pregnant thyroid hormone ranges were too small to justify use of separate ranges. We assume from the article that the controls were not screened for thyroid antibodies. Prevalence of thyroid autoimmunity is high in women of reproductive age, whether or not they are pregnant.2 Serum thyrotropin (TSH) concentration is reduced in up to 20% of women during their first trimester, often with modestly increased thyroid hormones. The thyroid-stimulatory effect of human chorionic gonadotropin may help ensure adequate thyroxine delivery to the fetus. It is surely more important for clinicians to understand this than to have reference ranges that conceal normal physiological changes. Gilbert et al do not state whether patients with multiple or assisted-conception pregnancies were included — both are more likely to have abnormal thyroid test results.2 Their detection limit for TSH and the lower limit of normal differed by only 0.01 mU/L — they could therefore not reliably distinguish low TSH from suppressed TSH. They screened only 61% of pregnant women in WA. It is inconceivable that there was not a selection bias, as current guidelines3 advocate only screening high-risk groups such as those with a history of thyroid disease or previous poor obstetric outcome. Ethnic differences in TSH levels have been reported. However, data from the United States National Health and Nutrition Examination Survey (NHANES) suggest that TSH levels in Hispanics are no different to those of white people,4 contrary to what is suggested by Gilbert et al.1 Increased miscarriage risk may relate to autoimmunity itself, rather than altered thyroid function. The study by Negro et al5 is, to date, the only one showing a decrease in miscarriages when thyroxine is given to thyroid antibody-positive women. However, the TSH level before thyroxine was given was comfortably within the normal range reported by Gilbert et al. Publications in this complex area are only informative if they tell us something about thyroid physiology or about diagnosis and management of thyroid disorders. While laboratories must validate their reference ranges, it is unlikely that those reported by Gilbert et al could be generalised to the ethnically diverse and geographically dispersed Australian population. Also, as described,1 patients would have to be screened routinely for thyroid antibodies to ensure that the quoted ranges were applicable.

Richard L Kennedy · Usman H Malabu · David Porter

Objective: To explore use of bone densitometry in Australia and to identify any sex and geographic differences, as a marker of osteoporosis diagnosis and care.Design and setting: Analysis of claims data from Medicare Australia in patients aged over 45 years during the period 2001–2005.Main outcome measures: Age-standardised rates of bone densitometry use, by sex and by metropolitan, rural or remote classification.Results: Bone densitometry use increased by 26% over the 5 years. Rates were lower for rural and remote populations, with people in capital cities about three times as likely to undergo the investigation as those in remote areas. The sex ratio for the rate of bone densitometry use (women to men) decreased from more than 6 : 1 in 2001 to 4 : 1 in 2005.Conclusion: Although the sex ratio for osteoporotic fracture is close to 2 : 1 (women to men), the sex ratio for testing is much higher, suggesting underuse of bone densitometry in men. Sex and rural inequities in use of the investigation need to be addressed as part of a national approach to reducing minimal trauma fracture.

Dan P Ewald FRACGP, MAppEpid, FAFPHM · John A Eisman FRACP, PhD, AO · Ben D Ewald BMed, MClinEpid, PhD · Tania M Winzenberg FRACGP, MMedSci(ClinEpid), PhD · Markus J Seibel MD, PhD, FRACP · Peter R Ebeling MB BS, MD, FRACP · Leon A Flicker MB BS, FRACP, PhD · Peter T Nash MB BS(Hons), FRACP

To the Editor: Type B insulin resistance is an uncommon syndrome characterised by abnormal glucose homeostasis (hypo- and/or hyperglycaemia), the presence of insulin receptor (IRec) autoantibodies, and intact IRec structure. It occurs mostly in African Americans, often with coexisting autoimmune disease.1 We report a case of type B insulin resistance with atypical features. A 44-year-old white male with a 22-year history of poorly controlled diabetes was referred to us with severe Graves ophthalmopathy. His home blood glucose levels ranged from 3.0 to 25.0 mmol/L (reference range [RR], 3.5–5.5 mmol/L) and he required over 800 units of insulin daily. His ophthalmic condition had been diagnosed 18 months earlier and had been observed until its recent deterioration into diplopia. Examination confirmed severe bilateral Graves ophthalmopathy, with 6/6 visual acuity bilaterally, and euthyroidism. Thyroid function tests showed the following levels: thyrotropin-stimulating antibody, 69 U/mL (RR, < 10 U/mL); thyroid-stimulating hormone, 2.3 mU/L (RR, 0.4–4.0 mU/L); and free tetraiodothyronine, 18.5 pmol/L (RR, 10.2–24.5 pmol/L). Routine biochemical and immunological studies were normal. Orbital magnetic resonance imaging showed marked ocular muscle hypertrophy and adipose tissue congestion, consistent with Graves ophthalmopathy. After 3 months of combination immunosuppressant therapy, the patient showed minimal improvement. In view of his diabetes, he was commenced on intravenous immunoglobulin rather than a glucocorticoid. Within 24 hours, he developed marked hypoglycaemia, with a glucose level of 2.1 mmol/L. Despite ceasing insulin treatment, the patient’s home blood glucose levels hovered between 4 and 6 mmol/L for the next 7–10 days. This continued until Day 14, when small doses of insulin were required, escalating to about 800 U/day before the next course of intravenous immunoglobulin. A clinical diagnosis of type B insulin resistance was made. The clinical picture suggested the presence of dual stimulating and blocking autoantibodies in the presence of structurally intact IRecs. At maximal insulin resistance, the IRec concentration is thought to be normal, but probably with reduced affinity. The blocking antibodies are believed to be polyclonal.2 The pathophysiology of the hypoglycaemic phase is unknown, but the antibodies would need to behave as IRec stimulators, possibly via partial agonism and increased IRec numbers.3 The mechanism of the selectivity of intravenous immunoglobulin therapy and its preferential removal of the inhibitory autoantibodies is yet to be unravelled. Proposed mechanisms include suppression of the inhibitory antibody activity and modulation of the immune system in favour of IRec-stimulating (hypoglycaemia-inducing) autoantibodies. Treatments include immunosuppression, plasmapheresis and rituximab therapy.4,5 As far as we are aware, the incidental but favourable response to intravenous immunoglobulin described here has not been previously observed. Our report highlights the atypical characteristics of this fascinating syndrome, including male sex, European ethnicity and a novel treatment modality.

Huy A Tran · Glenn E Reeves

To the Editor: A 50-year-old man presented with gynaecomastia and galactorrhoea, reporting diminished libido and energy over 12 months. Previous medical and psychiatric histories were unremarkable. The patient had a tender increase of the right breast tissue. His testes appeared normal. He had markedly elevated prolactin levels (410 μg/L; reference range [RR], < 15 μg/L) and decreased testosterone levels (5.6 nmol/L; RR, 10–33 nmol/L); results of other biochemical tests were unremarkable. Pituitary magnetic resonance imaging (MRI) showed a microadenoma. Cabergoline 0.5 mg twice weekly was commenced. One year later, the patient had normal prolactin (8 μg/L) and testosterone (14 nmol/L) levels. His libido and sexual function had improved — he claimed his “mates are envious”. MRI demonstrated no changes to the tumour. He was lost to follow-up. Five years after his last review, the patient re-presented with his estranged wife, who was concerned about changes to his behaviour after starting cabergoline. He had engaged in excessive casino and horse-racing gambling, resulting in financial losses (> $100 000), and excessive libido had led to hypersexual activities and divorce proceedings. His prolactin levels were normal (10 μg/L), but testosterone levels were low (8 nmol/L). Cabergoline was ceased. On review 3 months later, the patient’s change in behaviour was dramatic. All gambling and hypersexuality issues had ceased, and divorce proceedings were on hold. His prolactin levels had increased (78 μg/L); testosterone levels were unchanged (8 nmol/L). No changes were seen on MRI. Pathological gambling has been reported in patients with Parkinson’s disease who take dopamine agonists — particularly pramipexole but also cabergoline (4.5% of published cases).1 Most were also prescribed levodopa.1 A minority had concomitant hypersexuality.1 The prevalence of pathological gambling in patients with Parkinson’s disease has been estimated at 6.1%, compared with 0.25% in age- and sex-matched controls.2 There has been one published case report of pathological gambling (but not hypersexuality) following use of a dopamine agonist (cabergoline 0.25 mg weekly) for prolactinoma.3 However, the dose of cabergoline normally used in Parkinson’s disease is higher (0.5–6 mg/day).4 Normalising prolactin levels usually leads to increased libido and vitality, but not pathological gambling and hypersexuality. Our patient had not engaged in these activities before commencing cabergoline, and there was no personal or family history of psychiatric illness. Moreover, his testosterone concentrations during treatment ranged from low to low–normal, never high. His Naranjo score was 6, indicating a “probable” adverse drug reaction.5 No reduction in tumour size was seen, raising the question of a partial non-functioning pituitary adenoma. Cabergoline-induced pathological gambling and hypersexuality are probably under-reported, and physicians should consider screening for these in patients treated with dopamine agonists.

Henrik Falhammar · Jennifer Y Yarker

A 62-year-old woman presented with left frontotemporal pain, scalp tenderness and raised levels of inflammatory markers. Temporal arteritis was considered likely, and symptoms resolved with prednisone therapy. This delayed diagnostic bone biopsy until a soft tissue abscess formed, and Pott's puffy tumour associated with Prevotella osteomyelitis of the frontal bone was diagnosed. This case highlights the value of early histopathological examination, and is a reminder of a condition seen frequently in the pre-antibiotic era. Clinical recordA 62-year-old woman was referred with a 6-month history of spontaneous left frontotemporal scalp pain. Initially, the pain was associated with a soft swelling over the scalp and resolved within 6 weeks. It recurred 4 months later, without associated swelling, localised to the left frontal region and became progressively more severe, disturbing the patient’s sleep. She remained systemically well and had no history of fever, sinusitis, dental infection or diabetes mellitus. Investigation by the patient’s general practitioner revealed raised erythrocyte sedimentation rate (130 mm/h; reference range [RR], < 20 mm/h), C-reactive protein (66 mg/L; RR, < 3 mg/L) and alkaline phosphatase (121 U/L; RR, 25–100 U/L). There was no paraprotein in the serum or urine. A plain radiograph revealed apparent osteolysis (Box, A) and a bone scan revealed intense uptake of technetium-99m-methylenediphosphonate in the skull (Box, B). A computed tomography (CT) scan confirmed diffuse lucency in the left frontal bone, without evidence of underlying sinusitis. These radiological and scintigraphic features suggested osteoporosis circumscripta cranii (the lytic phase of Paget’s disease of the skull), although the raised inflammatory markers were inconsistent with this diagnosis. A specialist opinion was sought. At review, the patient was afebrile and results of physical examination were normal except for marked tenderness over the left temporal region, raising the possibility of giant cell arteritis. Urgent temporal artery biopsy was arranged while the patient began oral prednisone therapy (80 mg per day). Her condition improved dramatically — the pain was alleviated and levels of inflammatory markers decreased — but no histological evidence of vasculitis in the temporal artery was found. A presumptive diagnosis of biopsy-negative giant cell arteritis was made, and oral corticosteroid therapy was continued with a tapering regimen. Prophylactic trimethoprim–sulfamethoxazole (320 mg/1600 mg twice per week) was administered to prevent Pneumocystis infection; concurrently, bisphosphonate therapy was begun (40 mg alendronate per day) for suspected osteoporosis circumscripta cranii. The occurrence of two separate pathological processes was considered improbable, and a skull biopsy for definitive diagnosis was discussed with the patient. However, she was reluctant to undergo further invasive testing because of the rapid symptomatic improvement. She was monitored clinically and biochemically at regular intervals while the prednisone dose was reduced over the following 6 months. Throughout this period, she remained pain-free and systemically well, with normal levels of inflammatory markers. Six months after initiation of corticosteroid therapy, while the patient was taking 10 mg of prednisone per day, she developed a tender fluctuant swelling over the left frontal region of the scalp. The swelling enlarged rapidly, and a CT scan revealed a 6 × 11 × 4.5 cm collection overlying the left frontal bone (Box, C). Neurosurgical consultation and exploration of the swelling were arranged. During exploratory surgery, copious malodorous pus was drained and a biopsy sample of the underlying skull, which appeared discoloured, was taken. A pure growth of Prevotella sp., an anaerobic gram-negative rod-shaped bacterium, was isolated on culture of the pus. The organism was identified as either Prevotella melaninogenica or Prevotella oralis. Histopathological examination of the biopsy sample revealed infiltration of the marrow spaces by neutrophils (Box, D), and the presence of plasma cells with admixed fibrosis (Box, E), consistent with concurrent acute and chronic osteomyelitis. The diagnosis was revised to Prevotella osteomyelitis of the frontal bone with associated soft tissue abscess (Pott’s puffy tumour). Prednisone and alendronate were withdrawn, and the patient was treated with intravenous antibiotics for 6 weeks. As the organism was penicillin-resistant, clindamycin (600 mg four times per day) was administered for 2 weeks, and this was followed by ertapenem (1 g per day) as outpatient therapy. The intravenous antibiotics were combined with oral metronidazole (400 mg three times per day) and were followed by treatment with oral clindamycin (300 mg four times per day) for another 6 weeks. The patient’s condition responded rapidly to medical management and she remained clinically well with normal levels of inflammatory markers at 12-month follow-up. A subsequent CT scan of the calvaria showed resolution of the lytic changes without bony sequestrum. DiscussionAnaerobic organisms predominate in head and neck infections, occurring in a mixed growth in more than 90% of dental, oral and neck space infections.1 Clinically relevant anaerobes include the gram-negative rods Bacteroides, Prevotella, Porphyromonas, Fusobacterium and Bilophila, and gram-positive rods (eg, Clostridium, Actinomyces and Propionibacterium) and cocci (eg, Peptostreptococcus).1 Despite frequently being found in mixed bacterial populations, Prevotella sp. was isolated as a pure growth in our patient. Although trimethoprim–sulfamethoxazole therapy (used as Pneumocystis prophylaxis in our patient) is generally thought to lack useful activity against anaerobes,1 it may have suppressed the infection. Osteomyelitis of the skull is uncommon; recognised clinical syndromes include frontal bone osteomyelitis secondary to frontal sinusitis, and skull base osteomyelitis associated with malignant otitis externa. Skull osteomyelitis may also result from direct inoculation during surgery or occur in association with retrograde septic thrombophlebitis.2 Haematogenous seeding of bacteria in the skull is rare. Although our patient had no clinically or radiographically recognised sinusitis or history of dental instrumentation or infection, it is likely that the organism originated in the oral cavity or frontal sinus. The duration of the history suggests that chronic osteomyelitis may have accounted for our patient’s initial symptoms, but, if present, Paget’s disease of the skull may have predisposed to bacterial seeding. Paget’s disease is a reported risk factor for osteomyelitis of the jaw.2 Pott’s puffy tumour is a complication of frontal osteomyelitis that was first described by Sir Percival Pott in 1760.3 It appears as a circumscribed swelling on the forehead, representing a subperiosteal abscess that forms when infection breaks through the frontal bone.2 Epidural abscess, subdural empyema, brain abscess and cortical vein thromboses have been described in association with Pott’s puffy tumour. It was predominantly seen in children and adolescents before the widespread availability of antibiotics.2,4 In our patient, the administration of corticosteroids masked the underlying anaerobic infection until dose tapering permitted an abscess to form. Our patient illustrates a diagnostic challenge. It seemed unlikely that two unrelated diagnoses would account for a single presentation, but the prompt clinical response to corticosteroids led to a delay in diagnostic bone biopsy. The final diagnosis of frontal bone osteomyelitis with subperiosteal abscess is a reminder of a condition seen commonly in the pre-antibiotic era. Radiography and computed tomography (CT) scans, and bone biopsy specimens, of a patient with Pott’s puffy tumour associated with Prevotella osteomyelitis of the frontal bone A: Plain radiograph of the skull before corticosteroid therapy, showing patchy osteolytic areas in the frontal cranium. B: Delayed skull spot views on a bone scan before corticosteroid therapy, showing accumulation of technetium-99m-methylenediphosphonate. C: CT scan of brain (using intravenous iopamidol contrast medium) after 6 months of corticosteroid therapy showing a large collection over the left frontal bone. D: Bone biopsy specimen taken after 6 months of corticosteroid therapy, showing infiltration of the bone marrow spaces by neutrophils and scattered osteoclastic giant cells adjacent to the bony trabeculae, indicating active inflammation (haematoxylin and eosin stain; original magnification, × 400). E: Bone biopsy specimen taken after 6 months of corticosteroid therapy, showing plasma cells with admixed fibrosis, indicating chronic inflammation (haematoxylin and eosin stain; original magnification, × 400).

Lyndal J Tacon MB BS · Jonathon F Parkinson MB BS · Bernard J Hudson FRACP, FRCPA · Janice M Brewer MB BS, FRCPA · Nicholas S Little MB BS, FRACS · Roderick J Clifton-Bligh FRACP, PhD

To the Editor: I read with interest the letter by Lim and colleagues on anorexia nervosa and senna misuse.1 I have seen abnormal whole body bone scans in patients with severe eating disorders of exactly the same pattern (except for the avid bilateral apical lung and gastric uptake) as the case described. However, I disagree with the interpretation of the bone scan. There was increased periarticular tracer uptake involving long bones. The pattern was not that of hypertrophic osteoarthropathy (HOA). The pattern in HOA is linear tracer uptake by the periosteum, particularly along the distal ends of long bones.2 The scan in the case reported did not show uptake of this pattern, despite radiological evidence showing periosteal reaction and new bone formation of the tibia and fibula at the ankle. The pattern exhibited in this patient was more consistent with metabolic bone disease (increased tracer uptake by the ends of long bones periarticularly, the axial skeleton, calvaria, mandible, sternum and “beading” of costochondral junctions, with faint, or absent, renal uptake),3 although not all of these features were present in this case. Metastatic calcification of the gastric wall (not mentioned by the authors) and upper lobes of the lung was present in this patient. Metastatic calcification of the lungs can be diffuse4 or localised (most commonly) to the upper lobes, as in this case.5 With regard to the bone mineral density results in this case, the authors state that the lumbar and femoral neck T scores were elevated (1.2 and 1.3, respectively). The normal range of the T scores is ± 1.0 standard deviation of young adult normal values.6 Elevated bone mineral density measurements are generally not of pathological significance and are therefore clinically not relevant. In my experience they are usually decreased, and are often osteoporotic, in patients with severe eating disorders.

Andrew F McLaughlin

To the Editor: In their recent “Clinical Update” on adrenal insufficiency, Jung and Inder1 state: In patients with adrenal insufficiency who are fasting before procedures, glucocorticoid therapy must be continued, by parenteral routes if necessary. A recent case report has highlighted the adverse consequences of omitting oral steroid therapy in a patient who was fasting before a surgical procedure. The patient developed hypotension and acute renal failure. The patient described in the case report2 was admitted with septic arthritis. His usual cortisone dose of 12.5 mg had been omitted that evening, and his morning dose of 25 mg was not given the next day until after he had returned from the operating theatre. Over the next 3 days, he became overtly septic, and returned to theatre for another knee washout. Cortisone was not given during this time. When he developed acute renal failure on Day 5, dehydration and gentamicin toxicity were listed as possible causes. The article by Jung and Inder does not make it clear that the case report contains nothing of relevance to the management of patients with adrenal insufficiency who are fasting for routine surgical procedures, as this man’s hypotension and acute renal failure actually developed over 5 days in the context of sepsis, dehydration and possible gentamicin toxicity, in addition to prolonged withholding of cortisone and two operations.

Ian J Woodforth

To the Editor: The recent article by Jung and Inder1 provides sensible advice for the safe management of adults with adrenal insufficiency (AI) during illness and surgery, without risking adrenal crisis or excessive steroid dosing. However, the authors make no reference to paediatric practice and no guidelines have been provided for the body-size-related steroid doses required in paediatric patients with AI, either for routine steroid replacement or during illness and surgery. It is important that doctors be aware that the doses recommended by Jung and Inder are not suitable for children with AI. In keeping with recent studies of daily cortisol production, daily hydrocortisone replacement doses of 6–8 mg/m2/day are now recommended for children with secondary AI (eg, due to adrenocorticotropic hormone deficiency), provided the patient has no hypoglycaemia or symptoms of cortisol deficiency.2 In children with primary AI, higher hydrocortisone doses are often necessary (up to 10–15 mg/m2/day) — for example, to minimise adrenal androgen secretion in children with congenital adrenal hyperplasia.3 During minor illness (as defined in Box 3 of Jung and Inder’s article1), a child’s usual daily oral dose of glucocorticoid should be doubled or tripled until recovery.3,4 However, for children with secondary AI who are on the lower doses of daily hydrocortisone (about 6–8 mg/m2/day), these multiples may not constitute adequate doses during stress. In these patients, per-m2 dosing is more accurate (ie, 30–40 mg/m2/day for minor illnesses). During moderate-to-severe illness, for patients who are vomiting, those who have experienced trauma and those undergoing anaesthesia and surgery, the following doses of intravenous hydrocortisone are recommended: For children aged < 3 years: 25 mg initial dose then 25–30 mg/day; For children aged 3–12 years: 50 mg initial dose then 50–60 mg/day; and For adolescents and adults: 100 mg initial dose then 100 mg/day. These doses are in keeping with national4 and international3 recommendations, and equate to doses of 60–100 mg/m2/day of hydrocortisone. The more accurate per-m2 dosing should be used for children who are not within the normal weight range for their age. These recommendations for children are extrapolated from adult studies and also based on expert consensus. Attention to the specific body-size dose adjustments required in paediatric prescribing can provide safe levels of steroid cover while avoiding exposure to excessive steroid doses.

Ann M Maguire · Maria E Craig · Christopher T Cowell