Volume 182 Issue 5

Desperate people seek desperate remedies The treatment alternatives available to Australians with endstage kidney failure are dialysis, transplantation or no active treatment. The last of these options allows kidney failure to progress spontaneously to uraemia and death. Over the past decade the number of Australians on dialysis has grown by 6% per annum, adding an additional $25 million yearly to healthcare expenditure.1 This growth is caused by both increasing numbers of people entering dialysis programs and a low rate of transplantation because of a shortage of donor kidneys. Kidney availability in Australia remains low and, if anything, is worsening, with only 6.8% of those on dialysis receiving transplants in 2002, compared with 11.7% a decade earlier.1 It is remarkable that Australia has been so slow to fully examine and take up systems that appear to work Dialysis is the only initial treatment option for most patients with endstage kidney failure. Transplantation without prior dialysis is increasingly popular, but requires a live donor available at the right time; currently, only 3% of patients undergoing transplantation have not been on dialysis beforehand. Of those on dialysis, only 23% overall and 39% of those aged under 65 years are on the waiting list for a deceased-donor transplant.2 Those not on the list are either not interested in undergoing transplantation, have medical barriers to safe transplantation, or are deemed too frail to tolerate the procedure and subsequent immunosuppression. Both patients and healthcare professionals believe that transplantation, when feasible, is the preferred therapeutic option. The scientific justification for this belief appears well founded. In particular, there is strong evidence that patients who receive transplants have a significant survival advantage. The annual mortality rate of an age-matched population maintained by transplantation is reduced about 80% beyond the first year compared with those remaining on dialysis on the waiting list.3 The major difference is an up to 30-fold increase in relative risk of cardiovascular events and death experienced by those on long-term dialysis.4 In most Australian states, the average wait for a kidney from a deceased donor is about 4 years, and some patients wait much longer. The prospect of an extended wait on dialysis, as well as the possibility that a suitable kidney may never become available, drives some patients to consider more drastic options. One pathway that is illegal in Australia, but open to those able to afford it, is to travel overseas to purchase a kidney transplant. It is not known how often Australians are choosing this option. The report by Kennedy et al in this issue of the Journal (page 224) describes the outcomes for 16 Sydney-based patients who travelled overseas for kidney transplantation over the past 14 years. The risk of going down this path is evident, with an increased risk of serious infection being a major hazard. The annual rate of deceased-donor kidney transplants in Australia for 2004 was a low 11 donors per million population.5 In 2003, the rate in Australia was 9.0 per million population, compared with 33.8 in Spain, 23.9 in Austria, 24.8 in Belgium, 18.3 in France and 22.1 in the United States. 6 Thus, the rate of organ donation in this country is low compared with other developed countries, and remains so despite the publicity campaign promoting organ donation following the untimely death following a brain injury of Australian cricket icon David Hookes. One response to the shortage in deceased-donor organs has been an increase in live kidney donation, and the proportion of live donations in 2003 was 40% of total transplants. The source of live kidney donors, previously restricted to close blood relatives, has broadened in recent years to allow unrelated and poorly matched emotionally connected donors. In the past 12 months, there have been several kidney transplants from altruistic strangers donating to the pool of waiting dialysis patients (so-called non-directed donations). We can now add overseas commercial sources as another contributor to live kidney donation for Australian residents. The real reasons for Australia’s poor performance in deceased-donor organ procurement have not been fully established. Clearly, there is no lack of public support, which has exceeded 80% in repeated surveys over many years.7 One outstanding observation that has received little prominence and no systematic study is the high and internationally competitive organ donor rate achieved in South Australia. Over the last decade, South Australia has consistently doubled the rate in all other Australian states.5 A similar variation in performance is seen in the teaching hospitals in capital cities, with some having double the rate of others. This marked variation in the donation rates between states and hospitals points to the probability that the barriers to increased organ donation are within the hospital system. The situation in Australia appears ripe for a collaborative approach, such as one reported from the United States that sought to “identify, learn, adapt, replicate and celebrate ‘breakthrough’ practices associated with higher donation rates”.8 It is remarkable that Australia has been so slow to fully examine and take up systems that appear to work in some regions or hospitals. Positive moves are being made. A special working group of the Australian Health Ministers’ Council has recently made 11 recommendations for change in the arrangements and process for organ donation. The most fundamental recommendation is for intensive care staff to routinely interrogate the Australian Organ Donor Registry to ascertain the recorded intent of all suitable patients with severe brain injury after the first set of brain death tests. Relatives will then be informed of the intent recorded on the Registry and be asked only if they are aware of any change. Importantly, in this approach, the family will not need to be asked for consent. If Australia’s organ donation rate could match that of its best-performing hospitals and states, the embarrassing situation driving dialysis patients to take the risks involved with travelling overseas for kidney transplantation would not exist. Much remains to be accomplished, but there are grounds for optimism in believing Australia’s deceased-donor organ donation rate could double if the barriers existing in the hospital system could be removed.

Timothy Mathew FRACP · Randall Faull PhD, FRACP · Paul Snelling FRACP

If Indigenous Australians are at high risk of H. pylori infection, the associated risks of peptic ulceration and gastric cancer may make screening and treatment necessary Interest in asymptomatic Helicobacter pylori infection seems largely to have waned in much of the developed world.1-3 It is almost universally accepted that H. pylori causes chronic gastritis and peptic ulcer disease,1-3 and the evidence that the infection also causes gastric adenocarcinoma is virtually unassailable.1-4 In high-risk populations, at least, prevention of gastric cancer by H. pylori eradication is theoretically feasible, although convincing clinical-trial evidence is lacking.2-5 However, perhaps because the incidence of gastric cancer is steadily declining in developed nations (and because H. pylori infection seems to be much less often acquired these days), interest in its prevention remains remarkably low key.1,2 Australian Aboriginals and Torres Strait Islanders are much more likely to suffer ill health and die at a younger age than non-Indigenous Australians.6 However, one health scourge that the Indigenous population seemed to have missed, despite their lower socioeconomic status, was H. pylori infection and its complications.7 For example, seroprevalence studies suggested H. pylori infection was very rare.7 The report by Windsor and colleagues in this issue of the Journal (page 210) is at variance with these earlier findings,8 and strongly suggests that H. pylori infection is as common in Australia’s Indigenous population as it is in developing countries in South America, Africa and parts of Asia.3 If confirmed, these findings may have major health implications; indeed, the high prevalence identified in children8 implies a public health risk which will continue long into the future unless this situation is rectified. Notably, the findings of Windsor et al are limited to a small and possibly unique Indigenous group, and may not be more widely generalisable — more data are needed. However, the fact that over half the population in the remote rural communities studied were tested gives some confidence that H. pylori is remarkably widespread in this community (where 91% were found to be infected by means of a “gold standard” test). On the other hand, in urban Perth, referral bias can not be ruled out as accounting for the high H. pylori prevalence (60%), although the burden of infection remains strikingly high — if bias is the sole explanation, it would have to be particularly pervasive. What are the potential implications of these findings? Gastric adenocarcinoma remains a major health problem in the developing world.2 In the developed world, the consequences of H. pylori infection have arguably been underappreciated. In the United Kingdom, it has been calculated on the basis of relatively conservative assumptions that by age 85, and allowing for competing causes of mortality, the cumulative risk of dying from gastric cancer arising because of H. pylori infection is a staggering 1 in 51 for men and 1 in 96 for women.9 Data from the Northern Territory suggest that the rate of mortality from gastric cancer among Indigenous people has not increased.10 However, gastric cancer may become a much greater burden for the Indigenous population if, as is hoped, life expectancy improves from the current unacceptable median age at death of just 51 years.6 Older studies have suggested that peptic ulcer is uncommon in the Indigenous population,8 but, if Indigenous people undergo fewer endoscopies, this impression might be driven purely by referral bias. It has been calculated that, in the UK population, 1 in 154 men and 1 in 173 women die from peptic ulcer caused by H. pylori.9 The ulcer rates may, however, be truly low in Indigenous Australians, as undernutrition and lower socioeconomic status are associated with lower gastric acid secretion. This allows H. pylori to spread throughout the stomach (causing pan-gastritis and increasing the risk of cancer if people live long enough), but protects against ulcer disease.2,3 Circulatory disease, particularly ischaemic heart disease, is a major cause of mortality in the Indigenous population.6,10 Is H. pylori an important and modifiable risk factor for ischaemic heart disease? Unfortunately, the evidence for this is mixed and, at best, equivocal.11 Overall, widespread eradication of H. pylori to try to reduce ischaemic heart disease in this population is not currently justifiable. Other health problems in the Indigenous population just might be attributable to H. pylori, at least in part. Indigenous Australians have been reported to be more likely to give birth to infants who are small for gestational age,12 and H. pylori infection has been associated with growth restriction in young children.13 A Sydney study also found (among 448 pregnant women attending for routine examinations in the third trimester) that intrauterine growth restriction was twice as common in H. pylori-seropositive women than in seronegative women.14 This association was not explained by smoking or maternal height, although these were also independent risk factors for intrauterine growth restriction.14 While the data clearly need confirmation, the general health risks of a high carriage rate of H. pylori in Indigenous people may have been underestimated and may be another antenatal risk factor. Might H. pylori offer protection from some other diseases, and, if so, might attempts at its widespread eradication lead to more ill health in the Indigenous population? Such concerns have been raised,15 but seem overexaggerated and unlikely. In particular, recent data suggest that H. pylori eradication in peptic ulcer disease neither provokes reflux oesophagitis nor aggravates symptoms of gastroesophageal reflux.16 We need large, population-based intervention trials to define any risks, but whether it is wise to wait for such results before actively intervening remains debatable. H. pylori may be very common in Indigenous Australians, particularly among those living in remote parts of Australia. Future research needs to focus on the prevalence of H. pylori (and strains of different virulence) and the burden of dyspepsia in these populations. Although the exact route of transmission of H. pylori remains controversial, close contact within families is probably at least one key risk factor, and faecal–oral spread is feasible.2,3 Thus, strong consideration needs to be given to trying to reduce the acquisition of the infection in Indigenous children by reducing family crowding, and providing adequate sanitation and clean water. Doctors need to be aware that there may be a high risk of H. pylori infection in subpopulations within Australia. They should consider testing and treating Indigenous patients who present with dyspepsia (or non-ulcer dyspepsia), have a history of peptic ulcer complications, or have a family history of gastric cancer.2,3,17 However, to reduce the disease burden of H. pylori in the next 50 years, population screening and treatment may be the only solution. Assuming the data of Windsor et al are generalisable, a randomised controlled trial to test the health benefits (and risks) of population-based screening and antibiotic treatment in Indigenous Australians should be a government research-funding priority.

Nicholas J Talley MD, PhD, FRACP

Vaccination programs can act as a paradigm for effective health programs in Indigenous people The inferior health status of Indigenous Australians has been extensively reported1 and is linked inextricably to their ongoing social and economic disadvantage.1,2 Measures to decrease this health inequity are best focused on prevention and primary care as early in life as possible.2 Infectious diseases, although accounting for only a minority of the excess disease burden in Indigenous Australians,1,3 are particularly promoted by features of their disadvantage, such as overcrowding, poor infrastructure for health and hygiene, and poor nutrition. They are also more readily and rapidly targeted for intervention than chronic diseases.3 Vaccine-preventable diseases are particularly amenable to rapid and cost-effective prevention,4 and targeted vaccination programs have been shown to reduce health disparities.5 However, delivery of these programs, like other healthcare interventions, depends on culturally appropriate health services.2 The achievements and opportunities for vaccination programs to reduce morbidity and mortality among Indigenous Australians were highlighted in a recent report from the National Centre for Immunisation Research and Surveillance of Vaccine Preventable Diseases (NCIRS). This provided the first national assessment of the burden of vaccine-preventable diseases and vaccine coverage in Indigenous Australians.6 It showed that universal vaccination programs with highly effective vaccines (eg, measles, mumps, rubella, poliomyelitis, diphtheria and tetanus vaccines) have achieved excellent disease control in Indigenous people, similar to that in non-Indigenous people. Findings are similar in other countries with comparable Indigenous populations.4 In contrast, when vaccination programs are more limited (eg, hepatitis A, hepatitis B, influenza and pneumococcal disease), hospitalisation and death rates continue to be higher for Indigenous Australians than for non-Indigenous Australians.6 The report also showed gaps in vaccination coverage for both children and adults. Children identified as Indigenous had comparable coverage to other children at age 2 years, but lower coverage at age 12 months, suggesting greater delay in receipt of vaccines. Among adults aged 50–64 years (for whom influenza and pneumococcal polysaccharide vaccine have been funded since 1999 for Indigenous people alone), vaccination coverage was higher for Indigenous people than for non-Indigenous people (47% v 26% for influenza vaccine and 20% v 3% for pneumococcal vaccine), but still suboptimal. For both vaccines, Indigenous adults had higher coverage in remote areas than in non-remote areas (75% v 45% and 48% v 19% for influenza and pneumococcal vaccines, respectively). These findings highlight two key issues: the potential for expanded vaccination programs; and the need to improve delivery of current programs, especially in non-remote areas. The value of expanded vaccination programs is compellingly demonstrated by data on hepatitis A and influenza. The recognition of high infection rates and three deaths from hepatitis A in far north Queensland in the 1990s7 led to a regional vaccination program targeting Indigenous children aged under 5 years from 1999. This virtually eradicated hepatitis A in both the Indigenous and non-Indigenous population.8 Similar results after vaccination of high-incidence populations have been documented in the United States and Israel.9 Yet, the NCIRS report documented hepatitis A as an issue for Indigenous Australians outside north Queensland, with national hospitalisation rates 57 times higher in Indigenous children aged under 5 years than in non-Indigenous children the same age.6 This highlights the potential of broader national application of the strategy, which is currently being examined by the Australian Technical Advisory Group on Immunisation. In addition, deaths attributed to influenza and pneumonia are significantly higher in Indigenous Australians.6 Among age groups either not targeted for influenza vaccination (0–4 years), or where only those with risk factors such as chronic cardiac or pulmonary disease are targeted (25–49 years), the respective mortality rates are 17 times and 28 times higher for Indigenous Australians than for non-Indigenous Australians.6 In contrast, among Indigenous adults aged 50 years and over, in whom universal influenza and pneumococcal vaccination is funded, the differential mortality rate is much less (three times higher). This suggests that the vaccination program has had an impact and justifies examining more closely the expansion of universal vaccination of Indigenous people to younger age groups. As to the delivery of current programs, several presentations at the Public Health Association of Australia’s national immunisation conference in 200410 were pertinent. The conjugate pneumococcal vaccine program (introduced in 2001) provided another example of the success of appropriately targeted vaccination programs, reducing for the first time, in 2003, the national incidence of invasive pneumococcal disease in Indigenous children under 2 years of age to the levels in non-Indigenous children.11 This is similar to the impact of such a program in the United States.4 Other conference presentations highlighted the problems of vaccination delivery to Indigenous Australians in general practice. Although general practice is crucial for vaccine delivery in non-remote areas, only 27% of respondents to a national survey of GPs had a routine system for ascertaining whether their patients are Indigenous.12 Yet, Aboriginal and Torres Strait Islander people are happy to be asked about Indigenous status, provided it is done respectfully, and the reasons are explained.13 Improving GPs’ identification of Indigenous status could help increase vaccination rates of Indigenous Australians in general practice, which are lower than those of non-Indigenous Australians,14 and is vital for targeted vaccination. Improving identification is particularly important in non-remote areas, where vaccination coverage of Indigenous people is consistently lower than in remote areas,6 and where the characteristics of successful programs — accurate data on vaccination status, active promotion and outreach services, and effective collaboration across community, health authorities and providers15 — are less likely to be found. Vaccination is a safe, well accepted intervention which offers an opportunity to develop solutions in healthcare delivery that can be applied more generally. Although vaccination programs targeted only to Indigenous Australians will have less impact than universal programs, particularly in urban areas where Indigenous status may not be identified, the higher burden of disease in Indigenous people will continue to make targeting attractive. However, to translate the success shown by regional targeted programs8 to the national level requires substantial investment to improve both data collection (eg, ascertainment of Indigenous status in general practice and completeness of the Australian Childhood Immunisation Register) and service delivery (through culturally appropriate and accessible services). Provision of resources for active program monitoring and coordination equivalent to that found in successful programs in more remote areas is needed.

Peter B McIntyre PhD, FRACP · Robert I Menzies MPH

Objectives: To determine and compare the prevalence of Helicobacter pylori in an urban and a remote rural Western Australian Indigenous community.Design: Cross-sectional study of Helicobacter pylori status determined by urea breath tests between mid-January 2003 and the end of June 2004.Participants: 520 self-selected fasting participants, comprising 270 members of the Martu community at Jigalong, Punmu and Parnngurr in the East Pilbara region (129 men, 141 women; age range, 2–90 years) and 250 people from the Perth Indigenous community (96 men, 154 women; age range, 3–75 years.Results: The overall prevalence of H. pylori was 76%, but the prevalence in the remote rural community was 91%, compared with 60% in the urban community. The odds of having H. pylori were six times greater for rural than for urban participants (odds ratio [OR], 6.34; 95% CI, 3.89–10.33). Further, the overall odds of H. pylori infection in males (rural and urban combined) were greater than for females (OR, 1.61; 95% CI, 1.02–2.54). In both communities, the prevalence of infection remained relatively constant after the age of 10.Conclusions: The prevalence of H. pylori in the two Indigenous communities was two to three times higher than that in the non-Indigenous Australian population and higher than that shown in previous studies in Indigenous Australians.

Helen M Windsor PhD · Susan D Morrow BA · Barry J Marshall FRACP · Emmanuel A Abioye-Kuteyi MB BS · Jacqueline M Leber RN · Max K Bulsara MSc

Objective: To evaluate the effect of a behaviour modification program, taught to parents in a single visit to a trained nurse, in improving sleep performance in newborn infants.Design: Randomised controlled trial.Setting and participants: 268 families with normal newborn infants in the community, recruited between October 1996 and March 1997 from birth notices published in a South Australian daily newspaper.Intervention: A 45-minute consultation with a nurse 2–3 weeks after the birth, including a tutorial discussion on normal sleep patterns in newborn infants, supported by retained written material and, for infants with weight gain < 30 g daily, referral to their usual postnatal care provider.Main outcome measures: Hours of daytime sleep (0600–1800), night sleep (1800–0600) and total sleep per 24 h; and number of daily records with total sleep ≥ 15 h per 24 h, assessed by 7-day sleep diary at ages 6 and 12 weeks.Results: 268 families returned at least one sleep diary (137/171 intervention, 131/175 control), recording 3273 days. Two intervention infants were referred for low weight gain. Total sleep time was 15 h or more per 24 h on 62% of recorded days in the intervention group, compared with 36% in the control group (P < 0.001). At 6 weeks of age, intervention infants slept a mean 1.3 h per day more than control infants (95% CI, 0.95–1.65), comprising a mean 0.5 h more night sleep (95% CI, 0.32–0.69) and 0.8 h more daytime sleep (95% CI, 0.56–1.07). At 12 weeks, intervention infants slept a mean 1.2 h per day more (95% CI, 0.94–2.14), comprising 0.64 h more night sleep (95% CI, 0.19–0.89) and 0.58 h more daytime sleep (95% CI, 0.39–1.03). There was no significant difference in crying time between the groups.Conclusions: A single consultation supported by written material in the first 3 weeks of a child’s life improves sleep performance at 6 weeks of age. This improvement is maintained at 3 months.

Brian G Symon MD · John E Marley MD · A James Martin MD · Emily R Norman MB BS

Objective: To study the impact of various risk factors on survival time in a cohort of elderly Australians.Design, setting and participants: A longitudinal, prospective cohort study conducted in Dubbo, NSW. Participants were men and women aged 60 years or over living in the community, first assessed in 1988–1989 and followed for 15 years.Main outcome measures: Mortality rates; risk factors; survival times.Results: There were 668 deaths in 1233 men (54%) and 625 deaths in 1572 women (40%). Coronary heart disease was the major cause of death, rates being higher in men than women until age group 80+ years; stroke death rates were similar in both sexes; cancer and respiratory death rates were higher in men than women across all ages. In a proportional hazards model, the independent predictors of mortality were cigarette smoking, diabetes, very high blood pressure (BP), impaired peak expiratory flow (PEF), physical disability, and zero intake of alcohol. Over 15 years, the average reductions in survival time associated with various risk factors, in men and women respectively, were smoking, 22 and 15 months; diabetes, 18 and 18 months; very high BP, 16 and 9 months; impaired PEF, 14 and 17 months; physical disability, 16 and 12 months; zero alcohol intake, 9 and 5 months. Combinations of selected risk factors were associated with a multiplier effect.Conclusion: The reduction in survival time in elderly citizens demonstrated in the presence of smoking, diabetes and hypertension highlights a potential benefit to healthy ageing to be gained from prevention and intervention.

Leon A Simons MD, FRACP · Judith Simons MACS · John McCallum DPhil · Yechiel Friedlander PhD

Lack of donors has led to a worldwide increase in commercial kidney transplantation programs where recipients acquire kidneys either from executed prisoners or live non-related donors. Commercial transplantation is prohibited by legislation in Australia. Our centres have had 16 patients who have travelled overseas to receive a commercial kidney transplant; five have subsequently died. As has been found previously, patients who received commercial transplants were more likely to develop infections such as HIV, hepatitis B virus, cytomegalovirus and fungal infections. Previous reports have found that patient and graft survival were comparable to local results, whereas we found that patient and graft survival were worse than transplantation within Australia. Patients considering the option of overseas commercial donation should be advised that heightened risks to life and graft survival exist.

Sean E Kennedy FRACP · Yvonne Shen FRACP · John A Charlesworth MD, FRACP · James D Mackie FRACP · John D Mahony FRACP · John J P Kelly MD, FRACP · Bruce A Pussell PhD, FRACP

Although it is commonly assumed that the quality of medical school education in Australia is uniformly high, there is no national process for assessing its outcomes. There is substantial variability in the content of medical school curricula, and the process of curriculum change is becoming more challenging because of intense competition for time and space in the course. A national exit examination could provide a uniform standard of assessment for all medical school graduates in Australia, as well as foreign graduates applying to work in Australia. Such an examination could assess medical school outcomes, monitor the effects of curriculum change, and provide a benchmark for new medical schools that would help medical curricula evolve to better meet society’s needs.

Bogda Koczwara FRACP, MBioethics · Martin H N Tattersall MD, FRACP · Martin R Stockler FRACP, MSc · Michael B Barton FRANZCR · Brendon J Coventry PhD, FRACS · Ian N Olver MD, PhD, FRACP · Joanna M Dewar FRACP · Jeremy L Millar FRANZCR, FAChPM · Max A Schwarz FRACP, FACP, FAChPM · Darren L Starmer BN, RN · David R Turner PhD

We recently diagnosed rickettsial spotted fever in four patients from the south-eastern coastal region of South Australia near Adelaide, an area not known to be endemic for this infection. All infections were acquired within the geographic range of Aponomma hydrosauri, the tick vector of Rickettsia honei. Infection by R. honei was confirmed in two patients. This extension of the known geographic range of R. honei infection may be explained, in part, by alterations in host–parasite ecology. Two spotted fever group (SFG) rickettsia species endemic to Australia are known human pathogens — Rickettsia australis and Rickettsia honei.1 R. australis, the agent of Queensland tick typhus (QTT), is transmitted by ticks of the genus Ixodes, principally I. holocyclus,2,3 and has been associated with infections along the eastern seaboard from tropical north Queensland to Wilsons Promontory, in Victoria. The hosts for these tick vectors are mammals, including native rats and bandicoots.2 Confirmed cases of R. honei infection, or Flinders Island spotted fever, have previously been described only on Flinders Island, in Bass Strait,4 although a rickettsial spotted fever illness, possibly caused by R. honei, has also recently been reported from the east coast of Tasmania.5 Recently published findings indicate that the parasitic tick Aponomma hydrosauri, which has a variety of reptile species as its hosts, may be the principal vector of R. honei.6 We describe four patients from coastal South Australia (SA) near Adelaide who presented with an illness characterised by fever, severe malaise, diffuse maculopapular rash, and laboratory findings indicating acute SFG rickettsial infection. Clinical recordsPatient 1In October 2001, a 65-year-old woman from the south coast of SA (Box 1) was referred after 5 days of high fever, rigors, prostration, headache and generalised muscle pain. On Day 5, a non-pruritic rash appeared on the trunk and subsequently spread to the extremities, including the palms and soles. Two days before the onset of fever, while working in her garden, she had experienced sudden pain in her right groin, which she attributed to an ant bite; no tick or other arthropod was seen. She developed a mild purple skin discoloration in her right groin, but no eschar. She had not travelled recently or been bushwalking and had had no contact with animals, but her house bordered on bushland. On admission she appeared unwell, with tachycardia and temperature over 39°C. There was a diffuse maculopapular rash (Box 2), but no eschar, and crackles were audible in the left lung base. Chest x-ray showed bilateral reticular infiltrates that were most pronounced in the left mid and lower zones. Results of laboratory tests, including rickettsial serological testing, culture and polymerase chain reaction (PCR), are shown in Box 3. There was no initial response to empiric therapy with flucloxacillin, gentamicin and azithromycin. Three days after admission, she was reviewed by an infectious diseases physician and her antibiotic therapy was changed to 100 mg doxycycline twice daily, which resulted in rapid improvement of her condition, with resolution of fever and near-complete disappearance of rash within 48 hours. Doxycycline therapy was continued for 7 days. A week after discharge, she was well and there were no clinical or laboratory abnormalities. Patient 2In December 2002, a 74-year-old man from the south coast of SA (Box 1) was referred after 5 days of high fever, rigors, malaise, and severe prostration with mild confusion. He had a generalised, non-pruritic maculopapular rash, predominantly on the trunk but also on the extremities, including the palms and soles (Box 2). There was no history of recent travel or animal or arthropod exposure, but his house was on the edge of bushland. He had noticed numerous lizards in the vicinity of his house, but had not had any direct contact. On examination, he appeared unwell, with a temperature of 40°C and diffuse rash, but no other abnormal physical findings. Empiric therapy with flucloxacillin, ceftriaxone and gentamicin was started, but doxycycline (100 mg twice daily) was substituted 24 hours later after review by an infectious diseases physician. Within 48 hours, the fever resolved and he was discharged home. On review 6 weeks later, he was well, with no abnormal clinical or laboratory findings. Patient 3In March 2003, a 44-year-old woman from a southern Adelaide suburb (Box 1) presented after 8 days of lethargy, fever, rigors, headache, arthralgia and muscle pains; 4 days after the initial symptoms appeared, a maculopapular rash appeared over her trunk and limbs. She gave a history of probable insect bite 4 or 5 days before symptom onset. She had a temperature of 39°C and a widespread macular, blanching eruption over her trunk and limbs, including the palms and soles. A number of blood cultures, as well as serological tests for cytomegalovirus, Epstein–Barr virus, parvovirus B19, Ross River and Barmah Forest viruses, and also leptospirosis, were negative. On Day 4 of admission (12 days after onset of symptoms), she started a 7-day course of doxycycline (100 mg twice daily), and the fever resolved within 24 hours. At follow-up 3 weeks later, there was complete resolution of rash and normalisation of all haematological and biochemical parameters. Patient 4In May 2003, a 58-year-old man from a southern Adelaide suburb (Box 1) presented with 7 days of high fever, followed 2 days later by a macular rash on the trunk and limbs, which resolved within 72 hours. On the night of admission he awoke with dyspnoea, but there was no cough or sputum. There was no history of tick bite or recent direct animal exposure. On examination, his temperature was 38.4°C and there were bilateral basal chest crackles. There was no rash or eschar. Chest x-ray showed bilateral basal interstitial shadowing. After initial treatment with intravenous benzylpenicillin and erythromycin, his antibiotic therapy was changed to 100 mg oral doxycycline twice daily as treatment for presumed atypical pneumonia. Acute and 2-week convalescent serological tests showed no evidence of recent infection with Chlamydophila species, legionella, Mycoplasma pneumoniae, Q fever, parvovirus B19, Epstein–Barr virus, cytomegalovirus or Rickettsia spp. PCR testing of a throat swab for M. pneumoniae and Chlamydophila spp. was negative. The fever resolved within 48 hours of starting doxycycline therapy, and he was discharged home. The patient subsequently moved interstate and was lost to follow-up. Laboratory methodsMethods used for rickettsial serological testing, culture and PCR were all established and optimised at the Australian Rickettsial Reference Laboratory, Geelong Hospital, Victoria, and are described in detail elsewhere (N Unsworth, Postgraduate Student, Australian Rickettsial Reference Laboratory, Geelong Hospital, Geelong, VIC, personal communication). Serological tests were performed using antigen from SFG species R. honei, R. australis, R. akari, R. conorii, R. sibirica and R. rickettsii in an indirect microimmunofluorescence assay (IFA). IFA against the typhus-group rickettsiae R. typhi and R. prowazekii was also performed and was negative in each case. Rickettsial culture of buffy coat cells from peripheral blood and homogenised skin biopsy tissue was performed in Vero (African green monkey kidney)-cell monolayers. PCR with primers designed to amplify conserved sequences of the SFG rickettsial 17 kDa gene was performed on DNA extracted from buffy coat cells and skin tissue, and from any culture isolate. Rickettsial species was assigned on the basis of DNA sequence of the amplified portion of the 17 kDa gene. DiscussionWe describe four cases of rickettsial spotted fever in the south-eastern coastal region of South Australia, where this infection has not previously been recognised. In Patients 1 and 2, there was clearcut serological evidence of acute SFG rickettsial infection, with at least a fourfold rise in antibody titre. The causal pathogen was identified in Patient 2 as R. honei. Patient 3 had a compatible illness and history of antecedent arthropod bite, and developed low level SFG rickettsial antibodies 33 days after symptom onset. Patient 4 had positive rickettsial blood culture, from which R. honei DNA was identified. In this patient there were no detectable rickettsial SFG antibodies in serum taken during the acute phase of the illness or in Week 2 of convalescence. The serological response to some SFG rickettsial infections may be delayed or even absent, especially in patients treated with tetracyclines.7 After inoculation and dissemination through the bloodstream, SFG rickettsiae proliferate within vascular endothelial cells, resulting in vasculitis and (usually) a characteristic rash. The four patients described presented with fever, maculopapular rash, severe malaise, muscle pains, and, in two cases, respiratory symptoms and signs, together with pulmonary infiltrates. In the original case series of Flinders Island spotted fever, common symptoms included fever, headache, myalgia, arthralgia, cough, and maculopapular rash.4 All our patients required hospital admission, but it is likely that mild or asymptomatic infection occurs in others and remains undiagnosed.8 In all cases, there were significant abnormal laboratory findings, including deranged liver function, activated neutrophils on blood film, and markedly elevated C-reactive protein levels (Box 3). White blood cell counts were either low or normal, and one patient had significant thrombocytopenia. Antibiotic therapy with tetracyclines reduces the duration and severity of illness in most reported series of SFG rickettsial infection, including Flinders Island spotted fever.4,9 Symptoms and abnormalities in laboratory findings resolved rapidly for all our patients after doxycycline therapy was initiated; three of the four had had no discernible response to previous antimicrobial regimens, including, in 2 cases, macrolide drugs, which have been proposed elsewhere as a suitable alternative treatment for rickettsial SFG infections if tetracyclines are contraindicated.10 Of the four rickettsial species implicated in human infections in Australia, R. australis and R. honei cause spotted fever, which, until now, has been reported only along the eastern seaboard. Uniquely for arthropod-transmitted infections of humans, the vector of Flinders Island spotted fever has recently been shown to be Aponomma hydrosauri, a parasitic tick found on reptiles.6 This tick is found in the coastal geographic range of its host species in Tasmania, Flinders Island in Bass Strait, as well as much of southern mainland Australia, including south-eastern South Australia,11 where the cases we describe occurred. Although two of our patients gave histories suggesting arthropod bites a few days before onset of symptoms, there was no unequivocal tick exposure in any patient. This accords with the original case series of Flinders Island spotted fever, in which only four of 26 patients gave a history of tick bite, and reinforces the importance of not discounting possible tick-borne infection based on apparent lack of exposure. The interval of 2–5 days between possible tick exposure and onset of fever in two of our patients is consistent with reported incubation periods in individual cases of Flinders Island spotted fever where a tick has been isolated,4 whereas the incubation period of R. australis infection is probably somewhat longer.1 That most of our cases occurred in warmer months is also consistent with the observed seasonality of Flinders Island spotted fever,4 and with the predicted activity of A. hydrosauri. These cases define a previously unknown focus of rickettsial spotted fever in regions of recent population growth near Adelaide, South Australia. Environmental changes bringing humans into unusual propinquity with native fauna and their arthropod ectoparasites could increase both the geographic range and number of rickettsial species causing disease. Other ecological disturbances, including climate change and complex alterations in predator–prey relationships that increase populations of biting arthropods, may also lead to increases in the incidence of arthropod-borne infection.12,13 On the other hand, our recognition of Patient 1 as a case of SFG rickettsial infection clearly resulted in heightened local awareness of this diagnosis and increased active case finding. It is possible that locally acquired cases of rickettsial infection have been presenting to clinicians for many years without being recognised. Increased awareness and improved diagnostic capability, particularly molecular-amplification and DNA-sequencing technologies, are also likely to increase the number and species range of human rickettsial diagnoses.14 The microbiology, ecology and epidemiology of SFG rickettsial infections in South Australia require further study. Meanwhile, clinicians should be aware that rickettsial infections, particularly Flinders Island spotted fever, may occur outside previously described geographic ranges in Australia. Clinically compatible cases should be further investigated with appropriate serological and microbiological tests, and empiric doxycycline therapy should be considered to shorten the duration of illness. 1 Map of the South Australian coast, showing the locations south of Adelaide where the four patients with rickettsial spotted fever lived 2 Diffuse maculopapular rash associated with spotted fever group rickettsial infection Patient 1 Patient 2 3 Laboratory test results for four patients with rickettsial spotted fever Laboratory test (reference range) Patient 1 Patient 2 Patient 3 Patient 4 White cell count (4.0–11.0 × 109/L) 3.3 × 109/L 8.6 × 109/L 8.5 × 109/L 7.8 × 109/L Blood film Left shift and toxic changes in neutrophils Left shift and toxic changes in neutrophils Left shift and toxic changes in neutrophils Mild reactive lymphocytosis Platelet count (150–450 × 109/L) 74 × 109/L 189 × 109/L Normal 194 × 109/L Alanine aminotransferase (< 50 U/L) 311 54 305 120 C-reactive protein (< 6 mg/L) 262 149 243 148 Acute SFG antibody titre < 128 (Day 6)* < 128 (Day 6)* < 128 (Day 9)* < 128 (Day 7)* Convalescent SFG antibody titre 512 (Day 19)* > 1024 (Day 17)* 128 (Day 33)* < 128 (Day 18)* Rickettsia SFG culture of buffy coat cells and skin Buffy coat and skin negative (Day 9)† Buffy coat and skin negative (Day 8)† Buffy coat and skin negative (Day 12)† Buffy coat positive (Day 12);†no skin biopsy taken Rickettsia SFG 17 kDa gene PCR of DNA extract from tissue Buffy coat and skin negative Buffy coat negative; skin positive Buffy coat negative Buffy coat and skin negative 17 kDa gene sequence analysis nd Ricksettia honei (100% homology) nd Ricksettia honei (100% homology) All test samples were obtained on the day of admission to hospital unless otherwise specified: * day after symptom onset; †day of collection after symptom onset. All skin biopsies were obtained from the anterior abdominal wall. SFG = Spotted fever group. PCR = Polymerase chain reaction. nd = Not determined.

John R Dyer MB BS, FRACP · Lloyd Einsiedel BM BS, PhD, FRACP · Patricia E Ferguson BMed (Hons) · Andie S Lee MB BS · David L Gordon PhD, FRACP, FRCPA · Nathan B Unsworth BBiomedSc, BSc (Hons) · Stephen R Graves PhD, FRCPA

Clinical records Patient 1 In 2004, a 63-year-old woman presented to the emergency department with a 1-month history of nausea and intermittent vomiting. Three weeks before presentation she was empirically prescribed esomeprazole for management of dyspepsia. A week before presentation, she stopped taking this medication, as she suspected it was exacerbating the malaise, nausea and vomiting. On presentation, she was pale, with blood pressure of 160/60 mmHg and no signs of fluid overload. Dipstick urinalysis showed blood +, protein ++, and leukocytes. Serum creatinine level was 1878 μmol/L (reference range [RR], 60–125 μmol/L), and serum urea level, 42 mmol/L (RR, 2.5–6.5 mmol/L). A normochromic, normocytic anaemia was present (haemoglobin level, 87 g/L; RR, 130–180g/L), with no eosinophilia. Urine microscopy revealed isomorphic red blood cells, some white blood cells, but no casts. Urine culture showed no growth. On admission, the patient’s remaining medications (irbesartan [150 mg daily] and atorvastatin [40 mg at night]) were also withdrawn as a precautionary measure. As acute interstitial nephritis was suspected, she was treated with methylprednisolone (500 mg daily for 3 days) followed by oral prednisone (50 mg daily). Supportive haemodialysis was begun. On Day 4 of admission, a renal biopsy was performed; results were consistent with acute interstitial nephritis. Prednisolone therapy was continued for a total of 4 weeks on a tapering dose. Supportive dialysis was required for 4 days, by which time renal function had improved. However, at follow-up 8 months later, serum creatinine level remained abnormal at 187 μmol/L (Box 1). Patient 2 In the same year, a 63-year-old man presented to the emergency department with a 3-week history of nausea, vomiting, weight loss and oliguria. Five weeks before presentation, he was prescribed rabeprazole by his family physician as empirical treatment for nausea and dyspepsia. This was replaced by esomeprazole a week later, as symptoms had not resolved. The patient discontinued the latter after 2 weeks’ therapy (12 days before presentation), as he had still gained no relief of symptoms. Medications on presentation were irbesartan/hydrochlorothiazide (300/12.5 mg daily), amlodipine (10 mg daily) and rofecoxib (12.5 mg daily) as required. He had been taking these medications for the previous 3 years, but used rofecoxib infrequently. On presentation, the patient was pale, with no signs of fluid retention. Dipstick urinalysis showed blood ++, protein +, and trace leukocytes. Serum creatinine and urea levels were 1110 μmol/L and 31 mmol/L, respectively. Six months previously, serum creatinine level had been 109 μmol/L. The patient also had normochromic, normocytic anaemia (haemoglobin level, 90 g/L), with no eosinophilia. Urine microscopy revealed isomorphic red blood cells, granular casts, hyaline casts, some white blood cell casts, but no red blood cell casts. Urine culture showed no growth. A clinical diagnosis was made of acute interstitial nephritis induced by a proton-pump inhibitor (PPI). The patient was admitted to hospital and treated with prednisolone (60 mg daily). Renal biopsy 2 days later confirmed acute interstitial nephritis (Box 2). Renal function improved after admission, and serum creatinine level was 600 μmol/L on discharge 5 days after presentation. Subsequently, renal function declined slowly, and, 9 months after presentation, long-term peritoneal dialysis was begun. In these two cases of acute interstitial nephritis, esomeprazole was implicated as the likely causative agent (although Patient 2 was also briefly exposed to rabeprazole). In both cases, renal function improved after esomeprazole was withdrawn and corticosteroid treatment begun, although Patient 2 went on to require long-term dialysis. Omeprazole was first implicated as a cause of acute interstitial nephritis in 1992.1 Since then, 22 case reports of omeprazole-induced acute interstitial nephritis have been referenced in Medline. 2-7 Recently, pantoprazole and lansoprazole have also been implicated as causes of acute interstitial nephritis, 3,8 along with rabeprazole (unpublished data from our centre). Before the cases reported here, the Adverse Drug Reactions Advisory Committee (ADRAC) had been notified of four cases of acute interstitial nephritis induced by esomeprazole, along with six cases of acute renal failure and two of renal impairment. In October 2004, the manufacturer of esomeprazole, AstraZeneca, reported being aware of at least 15 cases worldwide of acute interstitial nephritis possibly induced by esomeprazole, and at least 200 cases worldwide induced by omeprazole (data on file, Astra-Zeneca). Two of these esomeprazole cases and 20 of the omeprazole cases were from Australia (it is unclear whether these two esomeprazole cases were included in those notified to ADRAC). MIMS online lists acute interstitial nephritis as a complication of omeprazole but not esomeprazole.9 A recent hospital series reported drug-induced acute interstitial nephritis as the cause of biopsy-proven acute renal failure in 8% of cases, with PPIs accounting for eight of the 14 cases. 2 The diagnosis of acute interstitial nephritis is most common when renal biopsy is performed for unexplained renal impairment, with a reported prevalence in this setting of 27%.10 PPI-induced acute interstitial nephritis poses a particularly difficult diagnostic challenge, as symptoms are non-specific and may mimic the original indications for which the PPI was prescribed. Most patients diagnosed with PPI-induced acute interstitial nephritis recover renal function, but some fail to recover fully and, in extreme cases, require long-term renal replacement therapy, 2 as in Patient 2. Little is known about the relationship between the risk of PPI-induced acute interstitial nephritis and duration and dosage of PPI therapy, delay in diagnosis, or other factors. In addition, it is not clear whether the prognosis differs between acute interstitial nephritis induced by PPIs and that induced by other drugs. Although both our patients were treated with corticosteroids, evidence for their use remains anecdotal and is not derived from randomised controlled trials. A recent report of the largest retrospective series published to date found no statistically significant difference in outcome, as determined by serum creatinine level, between patients who received corticosteroid therapy and those who did not at 1, 6 and 12 months after presentation.11 PPIs are now the third most commonly prescribed drug in Australia. Although acute interstitial nephritis is a rare complication, it is a potentially catastrophic cause of acute and chronic renal failure. All medical practitioners need to be aware of this potential class reaction. Early recognition may prevent the development of irreversible renal injury. Lessons from practice Acute interstitial nephritis is a rare but serious adverse effect of proton-pump inhibitor (PPI) therapy, which may potentially lead to chronic kidney failure. Acute interstitial nephritis has now been described in association with all PPIs currently marketed in Australia. The initial symptoms of PPI-induced acute interstitial nephritis are usually non-specific, often mimicking the symptoms of dyspepsia for which the PPI was initially prescribed. A high index of suspicion is needed, and serum creatinine level should be checked promptly if the diagnosis is suspected. 1 Serum creatinine level over time after presentation in Patient 1 Shaded area indicates the reference range, 60–125 μmol/L. 2 Renal biopsy specimen in Patient 2 Biopsy specimen from the renal cortex taken 2 days after presentation, showing an interstitial inflammatory infiltrate composed of lymphocytes and eosinophils (I), with mild acute tubular necrosis and a cellular cast (C). (Original magnification × 40; haematoxylin–eosin stain.)

Nimeshan Geevasinga BSc · Lukas Kairaitis FRACP · Gopala K Rangan FRACP · Patrick L Coleman MRCI, FRACP

Various claims have been made about the benefits of transdermal progesterone creams for relieving symptoms of menopause. Peer-reviewed articles have reported that the creams can raise plasma progesterone levels slightly, but have no effect on vasomotor, psychosexual or mood symptoms, bone metabolism or plasma lipid levels. Currently available progesterone creams can not be recommended for treatment of symptoms associated with menopause.

Barry G Wren MD, MHPEd, FRANZCOG

Constipation is common in children, with prevalences ranging from 0.3% up to 28%. In most children, constipation is functional (ie, without objective evidence of a pathological condition). Painful defecation has been proposed as the primary precipitant of functional faecal retention in early childhood. Faecal soiling is often secondary to constipation, and may occur during spontaneous relaxation of the sphincters precipitated by rectal distension. Management in general practice combines behavioural modification techniques with prolonged courses of laxatives. Treatment is usually successful, but may take up to 6–12 months. Significant numbers of children with initially good responses to therapy for constipation relapse in the long term. Long-term relapse is more frequent in children under 4 years at onset of symptoms and in whom there is a history of faecal soiling associated with constipation.

Anthony G Catto-Smith MD, FRACP

Bradley T Martin,* Pamela Palasanthiran,† Iain B Gosbell,‡ Thelma Barbagiannakos,§ Emma J Best,¶ Richard L Henry** * Respiratory Fellow (currently, Department of Respiratory Medicine, Children’s Hospital at Westmead, Locked Bag 4001, Westmead, NSW 2145), † Infectious Diseases Specialist, ¶ Infectious Diseases Fellow, Sydney Children’s Hospital, Sydney, NSW; ‡ Director, § Hospital Scientist, SWAPS Staphylococcal Reference Facility, South Western Area Pathology Service, Sydney, NSW; ** Head of School of Women’s and Children’s Health and Senior Associate Dean, Faculty of Medicine, University of New South Wales, Sydney, NSW. bradleymATchw.edu.au To the Editor: We report a case of severe pneumonia in a previously healthy 3-year-old girl of European background. Non-multiresistant methicillin-resistant Staphylococcus aureus (MRSA) was isolated from her sputum, and she was treated with intravenous vancomycin, followed by oral rifampicin and fusidic acid. After antibiotic therapy ceased, symptoms recrudesced, and computed tomography of the chest showed bronchiectasis. Sputum again grew MRSA (now also resistant to rifampicin, fusidic acid and erythromycin), as well as Pseudomonas aeruginosa. She was treated with intravenous vancomycin, ceftazidime and tobramycin. As the MRSA persisted, vancomycin was replaced with intravenous linezolid, followed by a course of oral linezolid and trimethoprim–sulfamethoxazole. More than a year after her original illness, she continues to have a productive cough and requires nebulised tobramycin to prevent exacerbations. The patient’s only risk factor was contact with her mother, who had an MRSA buttock abscess incised several months earlier. Phage typing and pulsed-field gel electrophoresis revealed that the mother’s and daughter’s isolates were identical (Box). They were found to belong to the Queensland strain of community-acquired MRSA (CA-MRSA) and to possess Panton–Valentine leukocidin, a virulence factor which is highly associated with necrotising pneumonitis and invasive primary skin infection. 1 CA-MRSA is a growing problem in Australia, and severe pneumonia due to this organism has recently been reported in adults. 2,3 The isolation of CA-MRSA before prolonged courses of antibiotics, its characterisation as a virulent strain, the response to appropriate treatment and the recrudescence of symptoms on cessation of therapy demonstrate that it was the causative organism in this case. The case is also noteworthy for the documented intrafamilial spread and the fact that the patient did not belong to the Pacific Islander community, in which CA-MRSA infections in south-western Sydney most commonly occur.4 Most S. aureus strains in the community are sensitive to flucloxacillin and dicloxacillin, so these remain the empirical treatments of choice, unless CA-MRSA is isolated or strongly suspected. CA-MRSA strains are non-multiresistant, and oral antibiotic options include clindamycin, trimethoprim–sulfamethoxazole or rifampicin and fusidic acid. 5 Intravenous vancomycin is commonly used in severe infections, but recent evidence suggests that linezolid, an oxazolidinone antibiotic with efficacy against multiply resistant bacteria, including MRSA, penicillin-resistant Streptococcus pneumoniae and vancomycin-resistant enterococci, may be more effective.6 Issues of cost, toxicity, local availability, potential development of resistance and sensitivity of isolates in vitro need to be considered before determining appropriate treatment. In conclusion, clinicians should be aware of the growing problem of CA-MRSA and the potentially devastating consequences of infection with this organism, even in otherwise healthy children. Pulsed-field gel electrophoresis (PFGE) of Staphylococcus aureus strains 1. Low-range PFGE marker. 2. Control S. aureus strain (NCTC 8325). 3. S. aureus isolate from patient. 4. S. aureus isolate from mother. 5. Queensland strain of community-acquired methicillin-resistant S. aureus (CA-MRSA).

Bradley T Martin · Pamela Palasanthiran · Iain B Gosbell · Thelma Barbagiannakos · Emma J Best · Richard L Henry

John R Gowardman,* Catherine Kelaher,† Joy Whiting,‡ Peter J Collignon§ * Intensive Care Physician (currently Launceston General Hospital, Launceston, TAS 7250), † Medical Student, ‡ Data Manager, § Director of Infectious Diseases and Microbiology, The Canberra Hospital, Canberra, ACT, and Professor, Canberra Clinical School, Australian National University, ACT, and University of Sydney, NSW. john.gowardmanATdhhs.tas.gov.au To the Editor: Bloodstream infections are frequent in healthcare settings and cause significant mortality and morbidity. 1,2 Most of these infections are caused by intravenous catheters, particularly central venous catheters (CVCs). Over 250 000 catheter-related bloodstream infections occur annually in the United States, 1 and over 3000 in Australia.2 Many CVCs are retained when no longer essential. For example, a recent one-day audit in a US teaching hospital found that 15% of CVCs (11/74) were “unjustified” most of these had been inserted in the intensive care unit but retained unecessarily after discharge from the unit.3 The risk of bloodstream infection is much higher with CVCs than with peripheral venous catheters (4.0 versus 0.2 per 1000 line-days).2,4 Such simple facts are often overlooked or inadequately emphasised in preventive programs, and CVCs may be retained for convenience. Our intensive care unit maintained an informal clinical practice of routinely removing CVCs when patients were discharged from the unit. However, an audit found that many CVCs were retained, often inappropriately, thus exposing patients to needless increased risk.5 A formal intervention policy aimed at improving CVC removal was implemented. This included a month of staff education, culminating in introduction of a formal written policy in March 2003. CVCs were to be removed when no longer clinically required or at discharge from the intensive care unit, unless the patient met predetermined retention criteria (ie, administration of vasoactive or venotoxic drugs [eg, dopamine or vancomycin] or parenteral nutrition solutions; poor peripheral venous access [after two attempts] with ongoing need for intravenous therapy; or transfer to another intensive care or coronary care unit). We undertook a prospective observational study of all patients with CVCs in the intensive care unit of our hospital in the period March to August 2003. Patients were grouped according to whether the CVC was removed per policy before or at discharge from the intensive care unit; whether it was retained per policy at discharge from the unit; or whether it was retained in breach of policy. All patients were followed up for 7 days after CVC removal. Those who died within this time were excluded from the analysis. We studied a total of 305 CVCs in 272 patients (Box). We observed: high compliance with the written policy (91%), significantly lower CVC in-situ times when policy was followed (5.1 v 8.1 days), low CVC reinsertion rates (7%), no difference in incidence of bloodstream infections between the groups. This study demonstrates that a formal policy directed at early CVC removal is effective in lowering CVC in-situ times without incurring clinical cost to the patients (eg, excessive CVC reinsertion rates). Policy breaches were infrequent (8% of all CVCs), but, when they occurred, CVC retention appeared unnecessary, and CVC in-situ times were significantly prolonged. The risk of sepsis with CVCs may be substantially lowered by policy-driven removal of CVCs, without compromising patient care. Comparison of patient characteristics and CVC outcomes when removal policy was followed versus when it was breached Policy followed Policy breached P (policy followed v breached) CVC removed CVC retained* Total CVC retained Patient characteristics Number of patients 176 71 247 25 Age (years) (SD) 60.2 (17.9) 64.9 (15.7) 61.7 (17.5) 69.0 (15.3) 0.02 ICU length of stay (days) (SD) 4.7 (8.2) 3.3 (5.2) 4.3 (7.6) 2.1 (2.0) 0.06 APACHE II score (SD) 14.7 (6.9) 14.8 (7.0) 14.6 (17.5) 14.2 (5.1) 0.31 Ventilation time (h) (SD) 51 (89) 46 (123) 51 (103) 31 (55) 0.19 CVC outcomes No. of CVCs (% of all CVCs) 202 (66%) 77 (25%) 279 (91%) 26 (8%) nt In-situ time Hours (SD) 97 (115) 202 (186) 124 (148) 197 (136) 0.009 Days 4.0 8.4 5.1 8.1 Tips cultured (% of CVCs) 136 (67%) 51 (66%) 187 (67%) 19 (73%) nt Tips infected (% of CVCs) 20 (9%) 11 (14%) 31 (11%) 4 (15%) 0.51 Catheter-related bloodstream infections Total no. 2 (1%) 0 2 (1%) 1 (4%) nt Per 1000 CVC days 2.5 0 1.4 6.0 0.33 CVC reinsertions (% of CVCs) 15 (7%) 4 (5%) 19 (7%) 0 0.38 Mean no. of ports idle (per day) na 1.5 na 1.6 nt Peripheral catheters Total no. 260 36 296 50 nt Mean no. per patient 1.5 0.5 1.2 0.5 nt Mean in-situ time (h) 63 66 64 81 nt CVC = central venous catheter. nt = not tested. na = not applicable. * Reasons for appropriate CVC retention were drug administration (32%), poor peripheral access (34%), transfer to another high dependency unit (25%) and total parenteral nutrition (9%).

John R Gowardman · Catherine Kelaher · Joy Whiting · Peter J Collignon

Vanessa M McDonald,* Peter G Gibson† * Clinical Nurse Consultant, † Professor, Department of Respiratory and Sleep Medicine, John Hunter Hospital, Locked Bag 1, Hunter Region Mail Centre, Newcastle, NSW 2310. Peter. GibsonAThunter.health.nsw.gov.au To the Editor: The delivery of asthma drugs via inhalation offers the best balance between efficacy and safety. However, poor inhalation technique limits the efficacy of this approach. In recent years, there has been a progressive increase in the types of inhalation devices used in asthma management. We questioned whether this would lead to “inhaler-device polypharmacy”, a situation in which an individual used multiple types of inhalation device to deliver his or her asthma medications. We conducted a novel investigation of this issue in 2004. We examined the computerised records of adults with asthma who had been enrolled in a standardised, evidence-based asthma management and education program1 between 2000 and 2004. We noted the number and type of inhaler devices used, as well as competence with each device (a trained asthma educator had observed and scored inhalation technique). We defined “inhaler-device polypharmacy” as the use of two or more different types of inhalation device. The devices assessed in the education program included a pressurised metered-dose inhaler (with and without a spacer), turbuhaler, accuhaler, aeroliser, autohaler, and handihaler. Nebuliser use was not included in the evaluation. We assessed a total of 511 patients: 278 (107 male; mean age, 37 years) between 1 January 2000 and 1 January 2002 (Period 1), and 233 patients (55 male; mean age, 40 years) between 2 January 2002 and 1 January 2004 (Period 2). Period 1 patients were distinct from Period 2 patients in that the latter began their treatment after the release of combination asthma therapy in a single inhaler, when polypharmacy may have been expected to diminish. Inhaler-device polypharmacy was present in 203 (73%; 95% CI, 68%–78%) patients during Period 1 and 164 (70%; 64%–75%) in Period 2 (P = 0.3) (Box). In Period 1, inhalation technique was inadequate with at least one device in 58 (29%) patients using inhaler polypharmacy and in 19 (25%) using only one device (Box). In Period 2, inhalation technique was inadequate with at least one device in 85 (52%) patients using inhaler polypharmacy and in 25 (36%) using only one device. In both Period 1 and Period 2 patients, inadequate inhaler technique with at least one device increased with the number of devices used (P values 0.02 and 0.05, respectively) (Box). We conclude that inhaler-device polypharmacy is a common problem among adults with asthma. Inadequate inhalation-device technique is also common, especially among patients using three or more delivery devices. Inhaler-device polypharmacy could lead to poor asthma control through inadequate delivery of medication. Patients with poor asthma control should be evaluated for their asthma management skills, including competency in using inhaler devices. These skills should be optimised before a new drug and/or device is added to their treatment regimen. We see no justification for the use of more than two inhalation delivery devices in asthma management. Number of asthma patients using single or multiple inhalation devices and proportion of those patients with inadequate technique, over two time periods One device Two devices Three or more devices Period 1* (n = 278) 75 (27%) 150 (54%) 53 (19%) Period 2† (n = 233) 69 (30%) 129 (55%) 35 (15%) Patients with inadequate inhalation-device technique Period 1* 19/75 (25%) 35/150 (23%) 23/53 (43%) Period 2† 25/69 (36%) 64/129 (50%) 21/35 (60%) * 1 Jan 2000–1 Jan 2002. † 2 Jan 2002–1 Jan 2004.

Vanessa M McDonald · Peter G Gibson

Raoul A Walsh,* Judith Lumley† * Senior Research Academic, Centre for Health Research and Psycho-oncology, The Cancer Council NSW/University of Newcastle, Locked Bag 10, Wallsend, NSW 2287. † Director, Mother and Child Health Research, La Trobe University, Melbourne, VIC. Raoul. WalshATnewcastle.edu.au To the Editor: Problems with interpreting odds ratios reported in meta-analyses of smoking-cessation interventions have recently been highlighted.1 Ford and Dobson2 have erred in a different way when applying the findings of the Cochrane review on smoking cessation interventions in pregnancy3 to calculate the public health benefits of delivering such interventions to all pregnant women in Australia. When all methodologically acceptable randomised controlled trials were considered, the Cochrane review did find the prevalence of smoking at end-of-pregnancy was 6% lower in intervention than control groups.3 However, this does not equate to a 6% reduction in the population prevalence of smoking among pregnant women, as Ford and Dobson assume. A mean between-group difference reported in a meta-analysis is not equivalent to a difference of exactly the same magnitude in a population prevalence of a risk factor unless 100% of the population exhibit that risk factor. Clearly, as Ford and Dobson have reported, this is not the case with smoking in pregnancy, where they correctly note that about 20% of pregnant women report current smoking at their first antenatal visit.2 Therefore, smoking-cessation interventions would not reduce the prevalence of smoking by 6% from 20% to 14%. The expected reduction can be calculated as follows: expected reduction in prevalence of smoking in pregnant women = current smoking prevalence in pregnant women (20%) × between-group difference in smoking prevalence (0.06) = 1.2%. This calculation rests on two assumptions: namely, that all pregnant women in Australia currently receive usual smoking-cessation care equivalent to that of control group conditions in the Cochrane review3 and that, in the short term, antenatal care can be transformed to the point where all future pregnant women receive smoking-cessation care equivalent to that received by those in intervention groups in the Cochrane review. Therefore, it is obvious that the expected smoking prevalence of 18.8% (20% minus 1.2%) is considerably higher than the 14% calculated by Ford and Dobson.2 Unfortunately, this means the rates of reduced infant deaths, hospital separations and costs to the healthcare system estimated by Ford and Dobson have also been overstated. In summary, the gains to be expected by clinical interventions with pregnant smokers are modest. Furthermore, past evaluations of media campaigns directed specifically at pregnant women have not shown significant positive effects.4 This reinforces the importance of tobacco-control strategies which target the whole population in addition to those which target pregnant women.5

Raoul A Walsh · Judith Lumley

Jessica H Ford,* Annette J Dobson† * Research Assistant, † Professor of Biostatistics, School of Population Health, University of Queensland, Herston Road, Herston, Brisbane, QLD 4006. J. FordATsph.uq.edu.au In reply: We thank Walsh and Lumley for correcting the error in our letter. The 6% reduction in smoking during pregnancy referred to an absolute difference in prevalence of continued smoking in late pregnancy among women who smoked early in pregnancy, from 91% in the control groups to 85% in the treatment groups.1 We incorrectly hypothesised a reduction from 20% to 14% in prevalence of any smoking during pregnancy. In fact, there was a decline in smoking during pregnancy, from 22% in 1994 to 17% in 2001 in New South Wales.2 These figures illustrate well the final point that Walsh and Lumley make: whole-of-population approaches to smoking reduction can yield much greater benefits (a 5% reduction in 7 years in NSW) than high-risk approaches (from our data, the 1.2% calculated by Walsh and Lumley). Our estimates of the adverse effects of smoking in pregnancy are, at present, correct. Although we unfortunately overstated the possible reductions resulting from interventions targeted only at pregnant women, such reductions are plausible for whole-of-population approaches.3

Jessica H Ford · Annette J Dobson

Mark Mackay,* Peter Millard† * PhD student, Department of Psychology, Level 4, Hughes Building, University of Adelaide, Adelaide, SA 5005, and Principal Project Officer, SA Department of Health; † Emeritus Professor of Geriatric Medicine, St George’s Hospital Medical School, and Visiting Professor — Health and Social Care Modelling Group, University of Westminster, London. mmbmATiprimus.com.au To the Editor: Gray, Yeo and Duckett used the wrong basis of measure for their analysis of bed trends.1 Bed use per thousand of the population masks the trends in total bed-days or separations and does not address the issue of supply. These issues have important ramifications for policy decision-making. Using the same sources of data,2,3 we compared bed-related statistics and population changes for the periods 1993–94 and 2001–02. What should be of most interest to planners is that the number of multi-day bed-days only declined marginally (from 14 434 to 14 231; −1.4%), despite the significant increase (from 1698 to 3343; + 96.8%) in same-day activity. Although multi-day separations and bed-days did decline (separations, − 4.2%; bed-days, − 14.9%) for those aged 65–74 years, for those aged 75 years or over bed-days and separations increased significantly (separations, + 41.6%; bed-days, +27.7%). Furthermore, same-day activity increased significantly for those aged 65 or more years. Furthermore, the authors failed to highlight the implications of changes in the relative age mix of activity. For those aged 75 years or more, the increase in proportion of separations (+ 5.8 percentage points) and bed-days (+ 1.8 percentage points) was greater than the increase in this proportion of the population (+ 1.1 percentage points). For the 65–74-years age group, the proportion of same-day hospital activity increased (+ 1.3 percentage points), unlike the reduction in that proportion of the population (− 0.2 percentage points). Moreover, the question of whether an ageing population has resulted in the need for more beds can not be answered without considering the supply of beds. From our experience, the growth in same-day activity has been achieved, at least in part, by substituting same-day beds for inpatient beds. The need for increased same-day beds has been considerable. Statistics relating to same-day beds do not appear to be reported for Australia as a whole. However, the increasing implied bed occupancy (including same-day) shown in the Box supports this conclusion. We surmise that the reduction in supply of multi-day beds combined with a marginally altered demand for multi-day beds has led to increasing numbers of bed crises. Given that relative growth in same-day activity can be attributed to people aged 65 years or over, and that the number of multi-day bed-days for those aged 75 years or more has increased, it appears that the ageing of the population, combined with the manner in which the substitution of beds has occurred, has contributed to increasing bed crises. Changes in implied bed occupancy Financial year Change from 1998–99 to 2001–02 1998–99 1999–00 2000–01 2001–02 Total bed-day utilisation (000s) 22 323 22 597 22 467 23 218 + 4.0% Total available bed-days (000s) 28 868 28 540 28 675 28 787 − 0.3% Implied occupancy 77% 79% 78% 81% + 4.3%

Mark Mackay · Peter Millard

Len C Gray,* Stephen J Duckett† * Professor in Geriatric Medicine, University of Queensland, Princess Alexandra Hospital, Ipswich Road, Woolloongabba, QLD 4102; † Professor of Health Policy, Latrobe University, Melbourne, VIC lgrayATsoms.uq.edu.au In reply: Mackay and Millard have raised some important issues in relation to our analysis. Our article was developed to encourage wider reflection and their response is thus welcomed. The primary criticism levelled by Mackay and Millard was that we underplayed the importance of supply of beds in our interpretation of the trends. We agree that bed supply is an important driver of utilisation patterns. We acknowledged this, in part, in the discussion as a possible explanation for rising separation and declining bed-utilisation rates in the older patient population. We are also sympathetic to the hypothesis that there may be a process of substitution of same-day separations for multi-day separations. However, data relating to bed availability are not readily available, and thus could not be included in our study. Our article was designed to highlight different trends between age groups, which have not previously been reported. Now that these trends have been identified, further research and analysis is required to fully explain them, with a view to supporting an intelligent strategy to prepare for future population ageing.

Len C Gray · Stephen J Duckett

Paddy A Dewan Head of Paediatric Surgery, Sunshine Hospital, PO Box 152, Parkville, VIC 3052. Paddy.dewanATwh.org.au To the Editor: Three recent articles in the MJA about complaints by patients attending hospital, 1 attitudes of hospital staff toward incident reporting2 and whistleblowing3 show that complaints are common, that cultural change is needed to allow staff to understand that a complaint from a patient or staff member should be viewed as an opportunity for change, and that quality assurance sometimes relies on whistleblowers but does not always appreciate their efforts. Faunce and Bolsin report on three whistleblower incidents,3 but fail to mention one at the Royal Children’s Hospital, Melbourne, in which I was involved. Attempts to highlight deficiencies in delivery of paediatric surgical care and concerns about the response to adverse events were managed with threatening tactics (of dismissal) by the division of surgery. This was followed by a hospital board investigation that, in my opinion, had neither the skill mix nor the terms of reference to adequately investigate the quality of care or the bullying. The subsequent investigation by the Department of Human Services involved narrowly focused terms of reference and failed to consider outcomes in some circumstances, thereby facilitating the “shooting of the messenger”. Current legislation does not effectively allow for dealing with threatening behaviour in the workplace, particularly when the refusal to look at complaints and adverse events in a productive manner goes well beyond the confines of the hospital involved. The Community Advisory Committee parliamentary enquiry was held in camera, with evidence being kept from the public. Worksafe legislation on bullying does not deal well with the complex situations that arise in the healthcare industry.4 The Colleges and other professional bodies, such as the AMA and the Medical Boards, need to take a proactive rather than a reactive role if further whistleblower incidents are to be avoided. I concur with the statement of Faunce and Bolsin that: “Even after substantiation of their allegations, the whistleblowers . . . received little respect and support from their institutions or professions”.3 From personal experience, I am very aware of the lack of support that stems from an ethos wary of public criticism, and the reactive bullying to which the whistleblower is often subjected. Until the culture of healthcare focuses on quality and caring, whistleblower sagas will continue to occur.

Paddy A Dewan

Luke R Bereznicki,* Shane L Jackson,† Gregory M Peterson‡ * PhD Candidate, † Research Fellow, ‡ Professor of Pharmacy, Unit for Medication Outcomes Research and Education, University of Tasmania, Private Bag 26, Hobart, Tasmania 7001. lberezniATutas.edu.au To the Editor: If a new drug such as ximelagatran is to be considered as a replacement for warfarin in preventing the thromboembolic complications associated with atrial fibrillation (AF), drug cost becomes an important issue. Brighton’s recent article in the Journal, 1 while comprehensive, does not discuss the cost-effectiveness of ximelagatran treatment. Ximelagatran was approved in several European countries for the prevention of venous thromboembolism associated with orthopaedic surgery. The cost of the drug for this indication (24 mg given twice daily) is 4.5 euros (A$7.7) per day.2 This represents the best available estimate of the cost of using ximelagatran for AF, although the dose is higher in AF (36 mg twice daily), and there are limitations in applying the drug cost in one country to another country. Routine monitoring of the antithrombotic effect of ximelagatran (ie, international normalised ratio [INR] testing) was not conducted in clinical trials. While this is potentially advantageous, frequent testing of alanine aminotransferase (ALT) levels is recommended at baseline and monthly for the first 6 months of therapy, every second month for the remainder of the first year, and every third month thereafter, for safety reasons.3 This is because some patients taking ximelagatran will develop elevated ALT levels (about 6.1% of patients to greater than threefold normal, and 3.4% to greater than fivefold normal) when ximelagatran therapy is commenced.3 The costs of INR and ALT tests are very similar (about $25 and $22, respectively). Although INR monitoring may be more frequent with warfarin than ALT testing with ximelagatran, the cost difference associated with therapeutic monitoring would remain far less than the likely cost of ximelagatran. We estimate the cost associated with treating 1000 patients with AF with ximelagatran instead of warfarin for 1 year, taking into account drug costs, monitoring costs and the slight difference in major bleeding rates, to be about $2.4 million (Box). A United States Food and Drug Administration advisory committee has recently raised concerns about the safety of ximelagatran (after episodes of severe liver damage), and has recommended that it not be granted any indication for use without further safety data. In particular, ALT monitoring did not prevent 3 deaths attributable to ximelagatran-associated hepatocellular necrosis.3 In light of the recent withdrawal of rofecoxib (Vioxx; Merck Sharp & Dohme), warfarin carries the intangible benefits of a long and proven track record. It certainly requires careful management and ongoing monitoring, but healthcare resources might be better spent on improving the use of warfarin rather than paying substantially increased costs for a drug with similar efficacy and an uncertain safety profile. Estimated costs of treating 1000 patients with atrial fibrillation (AF) with ximelagatran or warfarin for the first year of therapy. Ximelagatran* Warfarin† Total cost $2 803 821.00 $106 800.00 Cost per patient $2 803.82 $106.80 Monitoring Test (frequency/year) ALT (10)‡ INR (20)§ Total cost $217 000.00 $507 000.00 Cost per patient $217.00 $507.00 Major bleeding¶ Annual incidence** 1.6% 2.2% No. of expected events 16 22 Total cost $38 730.00 $53 253.00 Ischaemic stroke†† Annual incidence‡‡ 1.6% 1.6% No. of expected events 16 16 Total cost $101 936.00 $101 936.00 Overall cost Total $3 161 487.00 $768 989.00 Per patient $3 161.49 $768.99 Cost difference compared with warfarin Total $2 392 498.00 — Per patient $2 392.50 — ALT = Alanine aminotransferase. INR = International normalised ratio. * Cost of giving ximelagatran (24 mg twice-daily) to prevent venous thromboembolism post-surgery (German data; the dose for prevention of thromboembolism in AF is 36 mg twice-daily).3† Cost of warfarin taken from the Australian Pharmaceutical Benefits Scheme, December 2004. ‡ Monitoring cost derived from the cost of conducting ALT testing (Medicare Benefits Schedule, December 2004) according to the manufacturer’s directions (tests at baseline, monthly for the first 6 months, 2-monthly for remainder of the first year). § Cost derived from Medicare Benefits Schedule (December 2004) based on a frequency of 20 tests per annum. ¶ The cost of a hospital admission caused by a major bleed was $2420.60 in Australia for the years 2002–2003 (estimate based on 2002–2003 public hospital data).4 Intracranial haemorrhage may be associated with significant ongoing costs, but a smaller proportion of major bleeding incidents. ** No significant difference between warfarin and ximelagatran in either SPORTIF III5 or V;6 statistically significant when data from both trials were combined at P < 0.05. †† Cost ($6371) taken from the NEMESIS study7 and covers acute admission to an Australian hospital with ischaemic stroke only; this is an underestimate of the ongoing costs associated with ischaemic stroke. ‡‡ No significant difference between warfarin and ximelagatran in SPORTIF III5 and V;6 no significant difference when data from both trials were combined at P < 0.05.

Luke R Bereznicki · Shane L Jackson · Gregory M Peterson

Donald B Reid Co-convenor, Colleague of First Contact PO Box 896, Bridgetown, WA 6255 dbreidAThotlinks.net.au To the Editor: Thank you for your in-depth look at some of the concerns in doctors’ health in the October 2004 issue of the MJA. 1 I believe it would have been useful to include in the issue some practical information for doctors wanting to seek help, either for themselves or for a colleague. The attached table of contact phone numbers does not cover every state and territory, but it is a starting point. The services offered are confidential, and can be anonymous if desired. In Western Australia, the contact can include access to a list of doctors willing and able to be GPs for their colleagues. Further information is available on the Doctors’ Health Advisory Service website <www.doctorshealth.org.au>. Editor’s note: The AMA website also has a very useful “Doctors’ Health Database” at <www.ama.com.au/web.nsf/tag/doctors-health-database>. Australasian Association of Doctors’ Health Advisory Services (DHASs): helplines for doctors Victorian Doctors’ Health Programme (VIC) (03) 9495 6011 DHAS (SA) (08) 8273 4111 Colleague of First Contact (WA) (08) 9321 3098 DHAS (NSW) (02) 9437 6552 DHAS (New Zealand) (04) 471 2654 DHAS (QLD) (07) 3833 4352 DHAS (TAS) (03) 6223 2047 (in hours) (03) 6235 4165 (after hours)

Donald B Reid

Max Kamien Emeritus Professor, Senior Research Fellow, Discipline of General Practice, University of Western Australia, Perth, WA mkamienATcyllene.uwa.edu.au To the Editor: Over the past two years, the Journal has pointed out the health hazards and the lack of logic in the Health Insurance Commission’s closure of its previously cost-effective and successful “Doctor Shopping Hotline”.1-4 But it has taken the death of a 25-year-old “prescription shopper” in Cairns, his crusading mother, a scathing report by the Queensland Coroner5 and public exposure of these problems by Mark Bannerman on ABC TV (The 7.30 Report, 22 Dec 2004) for discernible action to occur. On that program, the Federal Minister of Health and Ageing promised that a “Prescription Shopper Line” would be up and running by the end of January 2005, and indeed it was activated on 31 January. This leaves two outstanding issues. The first is for the Health Insurance Commission to engage in an open exercise of mutual education by clearly reviewing its process of thinking in closing the previously successful Doctor Shopping Hotline and its lack of urgency in reinstituting its proposed better and broader successor.6 The second, and more important, issue is in understanding the underlying factors and thought processes of those doctors whom prescription shoppers describe as an “easy touch”. 7,8

Max Kamien

A 49-year-old man presented with medial knee pain. Arthroscopy of the knee revealed meniscal fraying and early articular cartilage degeneration (Figure). It was thought that a twisting injury may have had a hand in the matter.

John C Tuffley MB BS, FRACS

Double standards in medical research in developing countries. Ruth Macklin. Cambridge: Cambridge University Press, 2004 (viii + 280 pp). ISBN 0 521 54170 0 Inequity, when looked at steadily, is appalling. Here we see it, like a Dementor straight out of Harry Potter, staring us down and threatening our sanity in the middle of nice, clean drug trials. Increasingly, pharmaceutical companies conduct clinical trials of new drugs in developing countries because, without adequate mechanisms for the protection of human subjects and the interference of busybody ethics committees, they can obtain rapid answers and get on with the business of profiteering. Clinical trials of new medications conducted in developing countries first provoked comment in 1997, when a study of relatively low dose AZT, given late in pregnancy to HIV-positive mothers about to give birth, was compared with treatment with a placebo. By then we knew that high-dose AZT (expensive) given for a longer period during pregnancy had positive consequences for the newborn child. The idea of not treating the controls caused uproar. Why worry? Firstly, the cost of the new treatment under test is often unaffordable for the country in which the trial is being conducted. When the trial ends, questions arise about continued use of the drug in that country. Secondly, trials using placebo controls bother some observers. They feel that the control subjects should be offered best care as provided in the country from whence the drug trial emanates. This is difficult when standard care is well beyond the reach of the country where the trial is running. What a mess! Thirdly, others worry when a trial is conducted, say, in Botswana of a drug of limited relevance there but highly relevant in Germany, where the costs of the trial would be much higher. Rip off? Ruth Macklin is a New York-based bioethicist with a legal background, which makes her a formidable commentator. She has served on many prestigious international bioethics agencies and inside the NIH. She writes compellingly, lucidly, and in deep detail. She examines the three questions mentioned above, and explores the interplay of science with concerns for human rights, justice, equity, and development. This is not a book for anyone in search of a quick fix or a slick answer. It is, however, chock-full of up-to-the-minute, sophisticated, sound and helpful ethical reasoning and reflection in an area that challenges our humanitarian values head-on. Stephen R LeederProfessor of Public Health and Community Medicine University of Sydney, Sydney, NSW

Stephen R Leeder

Fast facts: Endometriosis. Second edition. Botos Rizk and Hossam Abdalla. Oxford: Health Press, 2003 (131 pp). ISBN 1 903734 20 7. What a joy to be asked to review a book on one of the most topical and baffling disorders in gynaecology and to find it presented clearly and succinctly by highly regarded experts in the fields of gynaecology and infertility. Rizk, from the USA, and Abdalla, from the UK, pertinently quote Sir William Osler in their overview: "He who knows endometriosis knows gynaecology." Despite much research and many recent advances in management, endometriosis remains a major threat to the physical and psychological health of millions of women worldwide, especially in developed countries. (In underdeveloped countries it is rare, because women are either pregnant or breastfeeding for most of their reproductive lives, thus seldom menstruating.) This small book is a practical overview of this mysterious disorder, which, the authors estimate, affects over 5 million women in the USA and may be found in up to 25% of women attending infertility clinics in both countries. The "implant" theory (retrograde menstruation) is still considered a likely cause, but other theories are presented. The authors make an impassioned plea for earlier recognition and diagnosis of this complaint. Despite greater public awareness, medical awareness is still sadly lacking, they say, and endometriosis becomes "psychologised" (meaning, I assume, that the woman is often told "It's all in your mind"). Sadly, this is true, as has been my own experience as a gynaecologist in practice for over 40 years. This book would be of great value to general practitioners as well as specialists. I think that lay people would also find it easy to read and informative. The diagrams are excellent and the highlighted "key points" in each chapter are clear and precise. The references are also comprehensive without being exhaustive. This book is a little goldmine and is very good value for money. I will certainly treasure my review copy. Margaret SmithGynaecologist, West Leederville, WA

Margaret Smith

More than lipid-lowering US investigators have suggested that lowering serum C-reactive protein (CRP) levels (with statins or other anti-inflammatory agents) could reduce vascular risk in patients with acute coronary syndromes. The PROVE IT-TIMI 22 Investigators studied serum CRP and LDL cholesterol levels in 3745 patients receiving statin therapy (80 mg atorvastatin or 40 mg pravastatin, daily) after an acute myocardial infarction or for unstable angina. Irrespective of the serum LDL cholesterol level achieved, patients with low (< 2mg/L) serum CRP levels after statin therapy had better clinical outcomes than those with higher levels. N Engl J Med 2005; 352: 20-28 Worse than landmines In Afghanistan, many more children are now injured while tampering or playing with unexploded ordnance than by contact with landmines, say US authors. They reported that of 6114 recorded munitions-related injuries occurring in Afghanistan in 1997-2002, only 13% were associated with military activity. Further, over this period, the proportion of injuries caused by unexploded ordnance rose from 37% to 57%, with 42% occurring in children. The authors say policy debates must go beyond landmines to address the effects of widespread deployment of other explosives, including cluster bombs, which may fail to explode on impact and persist for years. They wonder whether such munitions, which are more visible than landmines, can be designed to be less attractive to children. BMJ 2005; 330: 127-128 Global pressure rising More than a quarter of the world’s adult population — that is, 972 million, or nearly a billion, people — had hypertension in the year 2000, say US and UK authors. Their estimate of 26.4% was based on data from the largest and most recent studies conducted around the world, including Australia. They also calculated that by 2025 the number of people in the world with hypertension will have risen to more than 1.5 billion. Lancet 2005; 365: 217-223 Long shift? Drivers beware Driving home after working for more than 24 hours poses a safety hazard for interns, according to the Harvard Work Hours, Health, and Safety Group. The group conducted a US nationwide, prospective, web-based survey of 2737 interns who filled out a total of 17 003 monthly reports about their work hours and events that occurred while driving home. After an extended work shift of 24 or more hours, interns were more than twice as likely to experience a motor vehicle crash and more than five times more likely to have a near-miss incident. In months when interns worked five or more extended shifts, they were more likely to fall asleep while driving or stopped in traffic. N Engl J Med 2005; 352: 125-134 Managing anticoagulation Patient self-management of oral anticoagulation therapy at home can be as effective and even safer than clinical management, according to Spanish researchers. They randomised 737 patients seen at their hospital for oral anticoagulant treatment to receive either continuing clinic care or self-management. Self-management involved a short training course, followed by weekly use of a portable coagulometer and self-adjustment of the treatment dose of anticoagulant. After an average follow-up period of nearly one year, the researchers found that clinic-based treatment and self-management were comparable in terms of time spent in the therapeutic range of INR and percentages of in-range INR values. However, the self-management group experienced fewer complications, including severe haemorrhage and thromboembolism. Ann Intern Med 2005; 142: 1-10 Multiple sclerosis and May In the northern hemisphere, multiple sclerosis is more common in people born in the month of May. It is also less common in people born in November. So says an international collaborative study group that examined pooled data from more than 42 000 people with multiple sclerosis from Canada, Great Britain, Denmark and Sweden. They found that 9.1% more people with multiple sclerosis were born in May and 8.5% fewer people in the month of November. The researchers said this abrupt change in risk by month suggested a threshold effect for both increased and decreased risk. They wondered whether sun exposure and maternal serum levels of vitamin D during pregnancy might be relevant factors. BMJ 2005; 330: 120-123 Dr Ann Gregory, MJA