Respiratory disease

Objective: To audit requests for computed tomography (CT) examination of the chest emanating from general practitioners and assess the appropriateness and usefulness of these requests.Methods: We reviewed 50 consecutive requests for CT examination received by two private radiology practices in Cairns between August 2004 and March 2005. Clinical details were abstracted from request forms and clarified by telephone if necessary. A subjective assessment of the appropriateness of the investigation was made by the authors. The study was performed in a large regional centre.Main outcome measures: Indications for requesting a CT scan; appropriateness of CT scan for indication specified.Results: Fifteen patients had had recent normal chest x-rays, all of whom proved to have normal CTs; eight had not had a recent chest x-ray performed. The CT scan was considered appropriate in 16 cases (32%), but 10 of these patients required referral to specialists anyway. Thirty-four CT scans (68%) were felt to be inappropriate and, of these, 10 were subsequently referred to specialists. In only six cases did the CT scan resolve the GP’s clinical problem. In six cases the wrong type of CT scan was performed (five were conventional CT scans instead of high-resolution scans; one was a high-resolution instead of low-resolution scan).Conclusions: Many CT examinations of the chest requested by GPs could be avoided or replaced by simpler, cheaper tests with lower radiation exposure. Assuming a fatal cancer risk of 1 in 3000, the radiation exposure involved in unnecessary chest CT scans could be responsible for about 40 fatal cancers a year in Australia.

Graham Simpson MD, FRACP, FRCP · Garry S Hartrick MB BS

Objective: To assess the use of domiciliary oxygen therapy (DOT) in Tasmania and the impact of a specialist oxygen clinic on service provision.Design: Retrospective observational study.Participants and setting: Patients prescribed government-funded DOT in Tasmania between December 2002 and April 2004.Main outcome measures: Indications for DOT; usage, prescription, reassessment and costs of DOT; influence of a specialist-run oxygen clinic.Results: 490 patients were using DOT, an overall rate of 102 patients per 100 000 population (varying between regions from 95 to 116 per 100 000 population). Of 267 patients (54%) prescribed DOT during hospitalisation, only 72% met national guidelines for DOT at commencement. Chronic obstructive pulmonary disease (COPD) was the most common indication (48% of prescriptions). The median time to reassessment after prescription was 5.5 months. Median usage in patients with COPD was 18.3 hours per day. The average cost per patient was $1498 per year, but differed regionally in relation to costs of ambulatory supplies. The oxygen clinic in the north-west region substantially reduced oxygen prescriptions, improved compliance with guidelines, decreased time to first reassessment from 21 to 6.6 months, and produced major cost savings.Conclusions: Prescription of DOT was often not in keeping with national guidelines. Reassessment was poor, despite more than half the patients being prescribed DOT as an inpatient. A dedicated oxygen clinic resulted in more appropriate prescription, decreased time to reassessment and a reduction in costs.

Andrew Jones MB BS · Richard Wood-Baker DM, FRACP · E Haydn Walters MA, DM, FRACP

Cyril Evans was born on 27 April 1921 in Sydney and attended Fort Street Boys’ High School. He graduated in medicine with credit from the University of Sydney in 1943, despite having had to work during the course to support his mother and sister and pay his university fees. He obtained a Diploma in Tropical Medicine in 1946. After doing his residency at Royal Prince Alfred Hospital, Sydney, Cyril joined the Australian Army in 1945. When the War ended, he worked for several years as a missionary doctor in the Solomon Islands. Between 1950 and 1954, he completed specialty training in internal medicine in the United Kingdom, working first at Hammersmith Hospital, London, and later in Cornwall and Wales. He became a Member of the Royal College of Physicians (London) in 1953. Cyril spent the next 21 years specialising in chest diseases, particularly tuberculosis, first in North Carolina, USA (1954–1955), then as Deputy Director of Tuberculosis Services in Queensland (1956–1968). During this time, he was seconded to the World Health Organization for 2 years (1964–1965) to work at the Tuberculosis Chemotherapy Centre in Madras, India. From 1969 to 1973, he served as Director of Tuberculosis Services for South Australia and then as Commonwealth Director of Tuberculosis Services in Canberra (1974–1975). In 1975, Cyril was appointed Deputy Director-General of the Commonwealth Department of Health. Over the next few years, he became a Fellow of the Royal Australasian College of Physicians (1975), the Royal College of Physicians (London) (1978) and the Royal Australian College of Medical Administrators (1979). He was highly regarded by his colleagues — respected not only for his expertise in public health, but also for his patience and consideration towards others. He was made an Officer of the Order of the British Empire in 1978. After retiring from government service in 1982, Cyril spent a year as Adviser in Chronic Diseases at the Western Pacific Regional Office of the WHO in Manila, The Philippines, and then over 10 years as Medical Director of the Australian Kidney Foundation (1986–1997). Cyril was passionate about the welfare of his fellow human beings. He decided against continuing to work in the USA in 1956 because of the racism he saw in the hospital and the community. From the 1960s, he volunteered his time to various programs to help people stop smoking and, more recently, was a keen supporter of Canberra ASH (Action on Smoking and Health) Inc. He was also a supporter of the Medical Association for Prevention of War and a long-time member of the Board of the Richmond Fellowship, a charity providing mental health care and accommodation to adolescents with behavioural problems. Cyril and his wife, Beryl, also provided a home-away-from-home for scores of people, particularly international students in Canberra, for more than 30 years. Cyril died on 1 February 2007 after a period of failing health associated with Parkinson’s disease and Alzheimer’s disease. He is survived by Beryl and children Bronwyn, David, Susan and Annette.

David B Evans · David de Souza

Two people employed for several years in the spawning shed of a mushroom farm developed mushroom worker’s lung. The first patient presented in respiratory failure, with radiological features characteristic of hypersensitivity pneumonitis. The condition of the second patient was subacute on presentation, with a computed tomography (CT) scan showing ground-glass opacities. With absence from the workplace and no steroid therapy, the symptoms of both patients subsided and the results of lung function tests and CT scans improved markedly. Clinical recordTwo employees of the same mushroom farm presented to our hospital within a 5-month period. The farm is a large commercial producer of Agaricus bisporus mushrooms. Both workers were employed in the spawning shed, where mushroom compost is tipped onto a conveyor belt for mushroom spawn (sterilised grain inoculated with mushroom mycelia) distribution. The process is associated with increased levels of ambient organic dust.1 The principal means of minimising organic dust in the shed was local exhaust ventilation. Neither worker recalled receiving instructions about respiratory protection or the specific hazard of organic dust exposure during their employee-induction process. Case 1A 36-year-old man, who was a non-smoker, had been employed at the mushroom farm for 8 years, and had worked in the spawning shed for 3 years. He described a 4-month history of non-productive cough that was noticeably worse in the afternoons at work and improved on weekends. Two weeks before presenting, he developed daily chills, sweats, myalgia, chest tightness and exertional dyspnoea. His symptoms consistently commenced 5 hours after arriving at work and persisted into the evening at home. They abated sufficiently by morning and over the weekend to allow him to return to work. He had lost 7 kg in weight over 4 weeks. He presented to the emergency department with worsening dyspnoea after a day at work. He was pyretic (38.1°C) and had bilateral basal inspiratory crackles. Measurement of arterial blood gases confirmed hypoxaemia (Pao2, 57 mmHg; reference range [RR], 80–100 mmHg). Inflammatory markers were elevated: C-reactive protein level, 92 mg/L (RR, < 8 mg/L); and erythrocyte sedimentation rate, 17 mm/h (RR, 8–12 mm/h). Results of a full blood examination were within normal limits. Chest x-ray showed a diffuse bilateral reticulonodular infiltrate, and a high resolution computed tomography (HRCT) scan showed changes consistent with hypersensitivity pneumonitis (Box 1). Mushroom worker’s lung was diagnosed and the patient was admitted for observation and oxygen administration. Corticosteroid treatment was not administered. Respiratory function tests showed a borderline restrictive ventilatory defect, with moderately impaired transfer factor for carbon monoxide diffusion (TLco) (Box 2). During 4 days of observation, there was an improvement in oxygenation, his fever abated, and C-reactive protein levels fell to 15 mg/L. He was discharged home and advised not to return to work. Over the next 4 months, with ongoing avoidance of workplace exposure, ventilatory function, gas transfer and vital capacity improved significantly, and the abnormalities seen on HRCT scan partly resolved, with persistence of tiny centrilobular nodules (Box 1). Precipitin testing for antibodies to A. bisporus was not available. Precipitin testing for antibodies to Micropolyspora faeni (a fungus of compost, hay and grain), done at the 1-month review, gave negative results. Case 2A 40-year-old man, an ex-smoker, who had worked in the spawning shed for 6 years, had experienced 3 months of non-productive cough, fatigue, exertional dyspnoea and weight loss. His working day in the spawning shed started at 6 am, with onset of symptoms usually occurring around midday. He presented on two occasions to another hospital after a full day’s work complaining of dyspnoea, cough, chest tightness, myalgias and fever. At the first of these two presentations, resting Spo2 (oxygen saturation measured by pulse oximetry) was mildly reduced at 93%. Results of a full blood examination showed neutrophilia (9.4 × 109/L; RR, 2.0–8.0 × 109/L), and the C-reactive protein level was 13.5 mg/L. No abnormalities were seen on chest x-ray. He was diagnosed with a respiratory tract infection, prescribed antibiotics and discharged. His symptoms abated during a period of sick leave, but recurred 2 hours after returning to work in the spawning shed. On presenting to our outpatient clinic, 9 days after his last work exposure, he reported that his cough and fever had abated, but exertional dyspnoea persisted. Spirometry tests showed no abnormality, but TLco was impaired at 22.8 mL·min-1·mmHg-1 (58% predicted). Serum precipitin testing for M. faeni gave negative results. HRCT scan of the chest showed subsegmental air-trapping on expiratory scans and subtle patchy ground-glass opacities in both lower lobes (Box 3). After 4 months of avoiding further exposure, he noted no recurrence of symptoms, slow improvement in exercise tolerance, and improved gas transfer. DiscussionMushroom cultivation in Australia is a large agricultural industry, employing over 2500 people,2 yet mushroom worker’s lung has not previously been reported in the Australian medical literature, nor to an occupational lung disease notification scheme.3 It is likely that there is considerable under-recognition of this condition, as it is estimated that 5%–15% of those exposed to the causative antigens may develop hypersensitivity pneumonitis.4 In the largest cross-sectional study of workers at an Agaricus mushroom farm, 20% of those heavily exposed to organic dust reported experiencing symptoms consistent with mushroom worker’s lung.5 Unfortunately, with no thorough epidemiological studies, specifically cohort studies, it is not possible to estimate the true incidence of respiratory disease in mushroom farm workers. Several outbreaks of mushroom worker’s lung have been reported in the international literature since the 1950s. Workers with high exposure to organic dust from mushroom compost, such as spawners and compost handlers, are commonly affected and hence the more specific term “mushroom compost worker’s lung” is occasionally used.1,6 In this form of mushroom worker’s lung, M. faeni (currently known as Saccharopolyspora rectivirgula of the class Thermoactinomycetes), which is present in mushroom compost, is the most commonly implicated allergen.1,5,7 Organic dust from mushroom compost consists of a vast array of microorganisms and organic antigens; failure to demonstrate precipitins to M. faeni, as in the patients reported here, does not exclude the diagnosis of mushroom worker’s lung.1,4,8 Japanese mushroom farm workers have been reported to suffer an alarmingly high incidence of allergic respiratory disease.9,10 In a 3-year follow-up study by Tanaka et al, 40% of workers left the industry due to intolerable respiratory symptoms.10 Japanese mushroom varieties such as Hypsizygus marmoreus (Bunashimeji) are grown on wet wood dust rather than compost and release billions of spores 4–6 μm in diameter before being harvested.10,11 The inhaled mushroom spore (rather than Thermoactinomycetes) is the causative allergen in this setting, with the term “mushroom picker’s lung” used to more accurately describe the group of workers at risk of this form of mushroom worker’s lung.11 Although commercial cultivation of “exotic” mushrooms in Australia is small (1000 tonnes per year compared with 52 250 tonnes of Agaricus mushrooms), as demand for and cultivation of these mushrooms increases, employers must be aware of the significant hazard posed by these varieties when developing safe work practices.2 The most important component of identifying hypersensitivity pneumonitis is recognition of exposure to a causative antigen, reinforcing the importance of a thorough occupational history, and identification of workplace hazards (Box 4).7,12 A temporal relationship between the development of symptoms (cough, fever, chills, dyspnoea, chest tightness and malaise) 4–8 hours after the start of exposure, and an improvement during weekends or vacations, is quite indicative of this condition.4,8 Organic dust toxic syndrome, a form of inhalation fever, may be difficult to differentiate from acute hypersensitivity pneumonitis and is estimated to be 30–50 times more common.12 Organic dust toxic syndrome may result from a single heavy exposure to organic dust, and is self-limiting, with symptoms rarely exceeding 36 hours.8 Optimal management of hypersensitivity pneumonitis requires early recognition and complete avoidance of further exposure to the causative antigen;8,7,12 a change of occupation may be necessary. Although corticosteroid therapy has been shown to result in more rapid improvement in lung function and may be warranted in severely unwell patients, it has not been shown to improve long-term outcomes.9 Recurrence of acute hypersensitivity pneumonitis is more common in patients treated with steroids; this may be due to their improved sense of wellbeing and less stringent adherence to antigen avoidance.8,13 The natural history of hypersensitivity pneumonitis has been poorly described, primarily due to a lack of longitudinal studies.8 With repeated acute or chronic low-level exposure in farmer’s lung, permanent lung damage caused by pulmonary fibrosis and emphysema has been shown to occur, with associated chronic dyspnoea and permanent impairment.8,9 Even patients who remain asymptomatic may have long-term physiological sequelae.8 Australian occupational health and safety legislation describes in broad terms employers’ responsibilities to ensure every reasonable action is taken to preserve the health and safety of workers. Obligations to control hazardous non-organic substances, such as isocyanates and silica, are further described by subordinate Occupational Health and Safety (Hazardous Substances) Regulations (Vic) and the accompanying Hazardous Substances Code of Practice. Despite organic dust clearly having the potential to harm human health, the requirement to control organic dust falls outside the domain of hazardous substance legislation in Australia. Therefore, for their duty of care to be discharged, employers in the agricultural sector must demonstrate due diligence in their identification and control of all workplace hazards, including organic dust. The National Occupational Health and Safety Commission (now known as the Australian Safety and Compensation Council) has established limits for some organic dusts, such as cotton.14 However, organic dust in most agricultural settings is a complex and variable mixture of constituents, impairing the ability to set useful standards.8 Episodic high concentrations of dust exposure, rather than static ambient levels, may precipitate respiratory diseases, further increasing the difficulty of determining “safe” exposure standards. These factors impair our ability to advise employers how best to control this hazard. It is also difficult to determine what can reasonably be expected of employers as far as monitoring is concerned. Urgent research has been called for in this area by the American Thoracic Society.8 Employers in agricultural industries should demonstrate awareness of the hazard of organic dust, and aim to reduce exposure levels using the “as low as reasonably practicable” (ALARP) principle. Mushroom farm workers specifically should be educated about the risk of developing hypersensitivity pneumonitis and be advised of the symptoms and warning signs.5 1 Lung imaging — Patient 1 A: High resolution computed tomography (HRCT) scan of the chest of Patient 1 at presentation showing small, ill-defined centrilobular ground-glass nodules < 5 mm in diameter. Scans of the lower zones (not shown) revealed more confluent areas of ground-glass opacity, without discrete nodules. B: Repeat HRCT scan performed 1 month later (1 month without workplace exposure) showing significant improvement, but with persistence of tiny centrilobular nodules, particularly in the upper zones. 2 Respiratory function tests — Patient 1 Normal range Time since exposure (percentage of mean predicted value) Tests Presentation 1 month 4 months FEV1 (L) > 3.34 3.24 (76%) 4.50 (107%) 4.54 (108%) FVC (L) > 4.27 4.11 (78%) 5.17 (99%) 5.51 (105%) FEV1/FVC (%) > 72% 79% 87% 82% TLco (mL·min-1·mmHg-1) > 30.3 19.8 (53%) 26.8 (72%) 32.7 (88%) VA (L) > 5.8 5.3 (78%) 6.4 (95%) 6.9 (104%) FEV1 = forced expiratory volume in 1 second. FVC = forced vital capacity. TLco = transfer factor for carbon monoxide diffusion. VA = alveolar volume. 3 Lung imaging — Patient 2 High resolution computed tomography scan of Patient 2 at presentation, showing normal upper lobes (A) and patchy, centrilobular ground-glass opacities, with expiratory subsegmental air- trapping at the lung bases (B, C). This is a non-specific pattern, compatible with hypersensitivity pneumonitis. 4 Occupational causes of hypersensitivity pneumonitis — disease and source of exposure9,12 Farmer’s lung: mouldy hay, grain; compost Bagassosis: mouldy sugarcane Mushroom worker’s lung: mushroom compost, mushroom spores Ventilation pneumonitis: humidifier, air conditioner Machine operator’s lung: contaminated metal working fluids Humidifier lung: ultrasonic cool-mist humidifiers Floor finisher’s lung: mouldy wood floors Malt worker’s lung: mouldy malt dust (brewing) Compost lung: compost Tobacco worker’s lung: mouldy tobacco Sequoiosis: contaminated red-wood dust Wood worker’s lung: mouldy wood dust Wood trimmer’s disease: mouldy wood trimmings Wine grower’s lung: mouldy grapes Suberosis: mouldy cork dust Cheese worker/washer’s lung: cheese mould Salami worker’s lung: salami seasoning Saxophonist’s lung: mouldy saxophone reed Bird fancier/breeder/handler’s lung: pigeon, duck, chicken, turkey, parrot Furrier’s lung: cat hair, fur dust Laboratory worker’s lung: laboratory rat or gerbil urine Oyster shell lung: shell dust Tobacco grower’s lung: tobacco dust Coffee worker’s lung: coffee bean dust Tea grower’s/worker’s lung: tea leaves Streptomyces hypersensitivity pneumonitis: contaminated fertiliser Detergent worker’s disease: detergent

Ryan F Hoy MB BS · Jeffrey J Pretto BAppSc, GDBI, CRFS · David van Gelderen MB BCh, FRANZCR · Christine F McDonald MB BS, PhD, FRACP

Clinical record A 23-year-old woman presented to the emergency department with acute radicular lower back pain that became apparent when she was lifting books. She had normal blood pressure and 100% oxygen saturation breathing room air, but her heart rate was 110 beats/min and she had a temperature of 38.4°C. Inexplicably, she was discharged with a diagnosis of mechanical back pain. She presented again 2 days later with back pain, increasing shortness of breath, vomiting, myalgia, fever and sweating. She had also developed a dry cough and anterior pleuritic chest pain. There was an erythematous lesion on her left elbow. She and other family members had a history of recurrent furunculosis. When examined on admission, the patient was tachycardic (heart rate, 160 beats/min), hypotensive (blood pressure, 80/50 mmHg) and hyperpnoeic (respiratory rate, 32 breaths/min), with an oxygen saturation of 100% on a non-rebreather mask with oxygen flow at 15 L/min. She was febrile, with a temperature of 38.2°C, and had a furuncle on her left elbow. She had tenderness in the right upper quadrant; the spleen was not palpable. There was midline and left paraspinal tenderness over T8/9. There was no tampon in situ. Initial investigations showed a predominantly neutrophilic leukocytosis (18.4 × 109 cells/L [reference range (RR), 3.5–11 × 109 cells/L]); coagulopathy (prothrombin time, 20 s [RR, 9–14 s]; activated partial thromboplastin time, 40 s [RR, 25–38 s]); thrombocytopenia (platelet count, 59 × 109/L [RR, 140–400 × 109/L]); renal dysfunction (urea level, 14.2 mmol/L [RR, 3–8 mmol/L]; creatinine level, 172 μmol/L [RR, 50–100 μmol/L]); and an elevated serum troponin I level (1.4 μg/L [RR, < 0.2 μg/L]). A chest x-ray showed bilateral multilobar consolidation. Initial therapy included large-volume fluid resuscitation, a noradrenaline infusion, intravenous hydrocortisone and empirical intravenous antibiotics (3.1 g ticarcillin/clavulanate, 400 mg gentamicin and 500 mg azithromycin, within 50 minutes of arrival). Subsequently 2 g dicloxacillin was administered intravenously. The patient required intubation and mechanical ventilation 6 hours after admission, due to markedly deteriorating respiratory function. At this time, arterial blood gas results measured with the patient on 100% oxygen were: pH, 7.13 (RR, 7.35–7.45); partial pressure of carbon dioxide (Paco2), 52 mmHg (RR, 35–45 mmHg); Pao2, 237 mmHg (RR, 75–100 mmHg); base deficit, –12.3 mmol/L (RR, –3 to 3 mmol/L); and bicarbonate, 16 mmol/L (RR, 22–33 mmol/L). A computed tomography scan revealed multiple small areas of airspace opacification in a perivascular distribution, as well as bilateral extensive lower-lobe consolidation. Nine hours after admission, in view of worsening shock, drotrecogin alpha (activated protein C) and vasopressin were commenced. Despite a high-dose infusion of noradrenaline and adrenaline, the patient’s circulatory status continued to deteriorate. Staphylococcus was grown from initial blood cultures after 14 hours, and intravenous vancomycin 1000 mg was administered. Sixteen hours after admission, the patient had an episode of ventricular tachycardia, which reverted to sinus rhythm after a single precordial thump. However, ventricular tachycardia recurred and progressed to asystole. The patient died 17 hours after presentation, despite resuscitation. Subsequently, methicillin-resistant S. aureus (MRSA) was grown from blood cultures, endotracheal aspirates, and furuncle swabs and biopsies. The organism was sensitive to erythromycin, clindamycin, gentamicin, tetracycline, ciprofloxacin and vancomycin. Isolates were typed using a real-time polymerase chain reaction method based on single nucleotide polymorphisms (SNP) of the core genome and the presence or absence of variable genes, including the gene for Panton–Valentine leukocidin (pvl).1 All isolates had an SNP and variable gene profile characteristic of the Queensland clone (ST93-MRSA-IV) of community-associated MRSA (CA-MRSA), including the presence of pvl. Queensland clone CA-MRSA was also isolated from nose swabs subsequently collected from three family members, two of whom had suffered from recurrent furunculosis. Postmortem examination showed that the principal pathology lay in the lungs and myocardium. The lungs showed multiple foci of bronchopneumonia, many coalescing to form extensive areas of lobar pneumonia. However, the most striking feature seen on histology was involvement of the pulmonary vasculature by staphylococcal septicaemia. Staphylococci had invaded the walls of multiple blood vessels, producing a florid vasculitis with subsequent secondary thrombosis of the involved vessels (Figure). This process involved both large and small vessels to such an extent that a lethal degree of bilateral arterial thrombosis had developed. The larger thrombosed vessels were obvious at macroscopic examination of lung slices. Multiple small thrombi were seen on microscopy. The myocardium showed focal abscesses containing staphylococcal colonies. Adjacent myocardial fibres showed necrosis, which correlated with the patient’s raised troponin level. The other organs of the body were remarkably free of sepsis, the spleen was normal, and the spinal column showed no evidence of osteomyelitis. A furuncle on the left elbow was confirmed. Virulent strains of methicillin-resistant Staphylococcus aureus (MRSA) have recently emerged in community settings around the world (including many parts of Australia)2 and are causing community-acquired infection with increasing frequency.3 Most of the virulent strains carry the genes for producing Panton–Valentine leukocidin (PVL), a potent necrotising toxin. They most commonly cause primary skin and soft tissue infections such as furuncles and abscesses, but can also give rise to severe invasive conditions, including necrotising pneumonia.4 While uncommon, necrotising pneumonia is associated with a high mortality rate. In Australia, two major strains of PVL-positive, community-associated MRSA (CA-MRSA) are currently circulating: the Queensland (QLD) clone and the south-west Pacific (SWP) clone. Currently, these strains predominate among CA-MRSA in Queensland, New South Wales and the Australian Capital Territory, while in other states, PVL-negative strains are more common.2 Lessons from practice A history of recurrent furunculosis in a patient or in family members may precede severe Staphylococcus aureus sepsis, including necrotising pneumonia. Patients with recurrent infection due to S. aureus should be tested for persistent nasal carriage. Treatment aimed at eradication could be considered. The prevalence of virulent strains of community-associated methicillin-resistant S. aureus (CA-MRSA) is increasing in many parts of Australia. Knowledge of local prevalence would be valuable in guiding empirical treatment. In communities where CA-MRSA is prevalent, suspected severe sepsis due to S. aureus should be treated with a combination of vancomycin and one of dicloxacillin, flucloxacillin or cephalothin until culture and susceptibility results are available. Necrotising pneumonia due to PVL-positive S. aureus is often rapidly fatal, as in the case described here. A study by Gillet et al recorded a mortality rate of 37% within 48 hours of presentation.5 A significant association with preceding furunculosis was also noted. Most cases occurred in otherwise healthy children and young adults. A recently reported fatal case of CA-MRSA necrotising pneumonia in an Indigenous person was also in a previously healthy young adult.6 The patient in our case had a history of recurrent furunculosis and a furuncle on her elbow at presentation, both commonly caused by PVL-positive S. aureus. Two family members had also suffered from recurrent furunculosis. All isolates from the patient and from nose swabs of three family members belonged to the QLD clone. QLD and SWP clones are frequently sensitive to numerous non-β-lactam antimicrobials.2 Agents such as clindamycin and cotrimoxazole may be used to treat mild-to-moderate CA-MRSA infections such as furunculosis, depending on the organism’s susceptibility.7 Agents that act against protein synthesis (and therefore toxin production) have a theoretical advantage in the treatment of toxin-related infectious syndromes, but good clinical studies in this area are lacking. The use of clindamycin for treating invasive CA-MRSA infections is supported by one retrospective study in children.8 Linezolid, a new agent also active against protein synthesis, has been shown to be superior to vancomycin, but only in complicated skin and soft tissue infections.9 Use of one of these agents, perhaps in combination with established anti-staphylococcal antibiotics, is worthy of prospective study. The current national recommendation for treating suspected MRSA pneumonia is to administer vancomycin together with a β-lactam antibiotic (dicloxacillin, flucloxacillin or cephalothin) until susceptibility data are known.7 The severity of this case and rapidity of progression make it unlikely that more appropriate antibiotic therapy would have led to survival. Indeed, azithromycin, which was administered soon after admission, is active against erythromycin-sensitive strains of S. aureus. Nevertheless, early optimum antimicrobial treatment will give the best chance of survival. The possibility of MRSA pneumonia should be considered in the context of severe community-acquired pneumonia, particularly in children or young adults, and especially if there is evidence of preceding staphylococcal infection, such as folliculitis or furunculosis. Histological section of the patient’s lung at autopsy The section shows confluent staphylococcal bronchopneumonia with invasion of vessels in the lungs, producing a florid vasculitis (arrow). Secondary thrombosis is occluding the pulmonary vascular system. (Haematoxylin–eosin stain; original magnification × 100)

David C Risson MB BS, BVSc · Enda D O’Connor MB BCh, MRCP(Irl), FJFICM · Roger W Guard FRCPA · Jacqueline M Schooneveldt MAppSci, MASM, GCM · Graeme R Nimmo FRCPA, FASM, MPH

Objective: To examine evolving changes in asthma and chronic obstructive pulmonary disease (COPD) in South Australia and Australia as a whole from the perspective of hospital admissions, ventilatory support and mortality data.Design: Retrospective analyses, for the period 1993–2003, of hospital separations data from the Australian Institute of Health and Welfare and the Integrated South Australian Activity Collection, and mortality data from the Australian Bureau of Statistics and South Australian hospital morbidity collection.Main outcome measures: Hospital separations, ventilatory support episodes, mortality rates, burden-of-disease rankings.Results: Between 1993 and 2003, in SA and nationally, hospital separations for asthma declined but separations for COPD increased significantly. Falling mortality rates from asthma in both men and women, and from COPD in men, contrast with increasing rates of COPD-related hospitalisation and mortality in women.Conclusions: Hospital admissions and mortality associated with asthma have fallen. Admission rates for COPD are declining for men, but there is no indication that admission rates for women have reached a peak. There is a need for higher prioritisation of COPD, including policies to reduce smoking in women, and medical practice initiatives to support primary and secondary prevention, pulmonary rehabilitation and appropriate drug therapies.

David H Wilson PhD, MPH, BEd · Graeme Tucker BSc · Peter Frith MD · Sarah Appleton BA · Richard E Ruffin MD · Robert J Adams MD

A 55-year-old woman was admitted to hospital with a 1-year history of recurrent haemoptysis. A chest radiograph showed a well defined, rounded left hilar mass with surrounding air-space shadowing (Box, A). Contrast-enhanced computed tomography revealed a thrombosed aneurysm of the left main pulmonary artery that extended into its branches, with surrounding lung consolidation (Box, B). Echocardiography excluded any valvular lesions or shunt. Tests for tuberculosis and HIV were negative, and antinuclear antibody levels were not raised. The aneurysm was thought to be idiopathic. It was surgically resected, and the postoperative course was uneventful. A: Chest radiograph showing a left hilar mass. B: Contrast-enhanced computed tomography of the thorax showing an aneurysm of the left pulmonary artery extending distally into its branches with evidence of curvilinear thrombus (1), patent vessels (left pulmonary artery [2], right pulmonary artery [3], ascending aorta [4], superior vena cava [5] and descending aorta [6]), and consolidation of lung parenchyma adjoining the thrombus, with pleural thickening (7).

Zubair Ahmad MD, FCCP · Imrana Masood MB BS · Saurabh K Singh MB BS

Objective: To elucidate oxygen administration practices in the setting of acute exacerbations of chronic obstructive pulmonary disease (COPD) and compare these practices with those recommended in internationally accepted guidelines.Design: Retrospective audit.Participants and setting: 65 patients admitted to a Melbourne university teaching hospital via the emergency department (ED), identified through medical records by a discharge diagnosis (discharged between 1 June and 30 September 2005) of acute exacerbation of COPD (AECOPD). Those included had respiratory function test results consistent with British Thoracic Society guidelines for the diagnosis of COPD.Main outcome measures: Length of stay, need for high dependency unit (HDU) admission, use of non-invasive ventilation (NIV), and use of arterial blood gas (ABG) tests.Results: Our audit showed that 95% of patients defined as retaining carbon dioxide received oxygen at a flow rate greater than 2 L/min. This process began in the ambulance and continued in the ED, often without monitoring of ABG levels. Length of stay was significantly longer (P = 0.029); need for NIV on admission greater (P = 0.0124); and rate of admission to the HDU higher (P = 0.0124) in patients who achieved a partial pressure of arterial oxygen (Pao2) ≥ 74.5 mmHg compared with those with a Pao2 < 74.5 mmHg.Conclusions: The vast majority of patients with AECOPD presenting to our university teaching hospital receive oxygen therapy outside of internationally accepted guidelines, often without monitoring of ABG levels. The use of high-flow oxygen may contribute to an increased length of stay, more frequent admission to HDU and greater use of NIV among patients who achieve a higher Pao2.

Simon A Joosten MB BS, BMedSc · Mariko S Koh MB BS, MRCP · Xiaoning Bu MB BS · David Smallwood MB BS, FRACP, PhD · Louis B Irving MB BS, FRACGP, FRACP

Controlling oxygen delivery to limit oxygen saturation should reduce the incidence of hyperoxic hypercapnia The report by Joosten et al in this issue of the Journal (page 235)1 is a timely reminder of the importance of avoiding the induction of hyperoxic hypercapnia in patients with acute-on-chronic respiratory failure. The complication of acute hypercapnic respiratory failure precipitated by giving oxygen has long been recognised; most resident medical and nursing staff are aware of this problem. The natural intervention in patients presenting with acute-on-chronic respiratory failure is to relieve any hypoxia with supplemental oxygen, but this can be associated with carbon dioxide retention, narcosis, respiratory acidosis, and death. That the use of controlled oxygen flow rates could avoid this complication (and the need, in those days, for tracheostomy and invasive ventilation) was first recognised in the 1940s and 1950s.2 Although hyperoxic hypercapnia can now be managed with non-invasive ventilation, the article by Joosten et al reminds us that it still has adverse consequences for morbidity, length of stay and the use of hospital resources.1 Studies over the past 20–30 years have identified the characteristics of the patients most likely to have this problem, and have gone some way to identifying the mechanisms responsible. Chronic respiratory failure is the usual predisposing condition, and the most common cause of chronic respiratory failure is chronic obstructive pulmonary disease (COPD). Interestingly, hyperoxic hypercapnia is a phenomenon of acute exacerbations of COPD — giving oxygen to patients with stable hypercapnia rarely, if ever, causes clinically significant further hypercapnia.3 This may, of course, be related to the lower flow rates used for stable hypercapnia. Importantly, the degree of hypoxaemia at presentation is a better predictor of hyperoxic hypercapnia progressing to narcosis than is the initial degree of hypercapnia.4 Usual clinical teaching is that high concentrations of inspired oxygen remove the hypoxic drive to ventilation in susceptible hypoxaemic patients; the narcotic effect of the rising hypercapnia amplifies this effect, promoting further hypoventilation. However, a number of studies have cast doubt on this as the most important mechanism, at least up to the point of narcosis.5-7 The most comprehensive study of mechanisms, using the multiple inert gas elimination technique, suggests that relative hypoventilation is the defining event in those who retain carbon dioxide, but that worsening ventilation–perfusion mismatching and an accompanying increase in dead space ventilation contribute about 50% of the increase in carbon dioxide levels.8 This finding is clinically important because it identifies non-invasive ventilatory support as the appropriate intervention before narcosis progresses. The dictum “hypoxia kills quickly, hypercapnia slowly” engages the clinician when confronted with this situation. How can hyperoxic hypercapnia be avoided without exposing these patients to the more acute risk of inadequate oxygenation? Oxygen delivery controlled to an appropriate flow rate appears to be the answer, but there are no large-scale studies to indicate how the oxygen “dose” should be determined and monitored.9,10 In the absence of clinical trial evidence, it is reasonable to control oxygen flow rate to achieve an arterial oxygen saturation of 90%, but not above 93%–95%. This corresponds with an arterial oxygen tension of 60–70 mmHg at the start of the “flat part” of the oxyhaemoglobin dissociation curve, and ensures adequate arterial oxygen content and delivery in most circumstances. This is also consistent with the data of Joosten and colleagues, who found an arterial oxygen tension of less than 74.5 mmHg to be protective.1 The ready availability of continuously reading pulse oximeters makes the above recommendation a practical procedure, and its wide application in wards, emergency departments and, particularly, ambulances should substantially reduce the incidence of the hazardous and largely unnecessary complication of hyperoxic hypercapnia.

Iven H Young BSc, PhD, FRACP

Over a 3-year period, seven patients who were taking HMG-CoA reductase inhibitors (statins) presented to our respiratory service with interstitial pneumonitis. Clinical course varied, with the condition responding to prednisolone treatment and cessation of statins in three patients, and progressing slowly despite this management in another three, while one patient died of associated cardiac disease. While a causative role cannot be confirmed, clinicians should be aware of the possible association. Two patients who presented with interstitial lung disease to our hospital’s respiratory service in 2000 prompted us to research a possible association between this disease and therapy with statins (hydroxymethylglutaryl-coenzyme A [HMG-CoA] reductase inhibitors). A literature review revealed several isolated cases of pneumonitis in the setting of statin therapy.1-3 Thereafter, we prospectively recorded patients referred to our unit who had interstitial lung disease and were undergoing statin therapy, where no other clear cause of the pneumonitis was evident. All cases have been reported to the Adverse Drug Reactions Advisory Committee. Clinical recordsBetween January 2000 and December 2003, our service saw 58 new presentations of interstitial lung disease, including the two patients discussed above. Data on these patients were retrieved from an ambulatory care database of newly presenting patients kept prospectively by our service. Their diagnoses are shown in Box 1. Eight cases were thought to be drug-associated: five of these were potentially linked to statin therapy, two to nitrofurantoin and one to amiodarone. Another two patients were referred to our service during hospital admissions in other specialties not included in our database. Details of these patients were retained from consultation records. Clinical details of the seven patients taking statin therapy are shown in Box 2. Most patients presented with dyspnoea and non-specific examination findings consistent with interstitial lung disease, such as bilateral crepitations on chest auscultation. None had clubbing. Some patients had a background of smoking and mild chronic obstructive airways disease, while others had no specific risk factors. Statins potentially implicated were atorva-statin (10–40 mg daily), pravastatin (40 mg daily) and simvastatin (10–40 mg daily). No patients were taking other medications known to be implicated in interstitial lung disease. Pneumonitis was diagnosed based on clinical assessment, along with demonstration of interstitial infiltrates on high-resolution computed tomography and reduced transfer factor for carbon monoxide diffusion on lung function testing (Box 3). Management comprised prednisolone or other immune-modifying treatment and/or withdrawal of the statin. In three patients treated with both prednisolone and statin withdrawal, pneumonitis decreased (Patients 3, 5 and 7). In another three, the condition progressed slowly: one of these (Patient 1) did not take prednisolone, and another (Patient 2) initially continued statin therapy and experienced respiratory failure necessitating home oxygen therapy before it was ceased. He died 18 months later of respiratory failure. The third (Patient 4) experienced slow progression despite statin withdrawal and treatment with prednisolone and azathioprine. The remaining patient (Patient 6) was treated with prednisolone but continued statin therapy, and died from coexisting cardiac disease, probably exacerbated by interstitial lung disease. DiscussionStatins are the most often prescribed class of medication for treating hypercholesterolaemia. They act primarily by inhibiting HMG-CoA reductase, thereby inhibiting cholesterol biosynthesis and improving lipid profiles. However, recent research has revealed multiple immunomodulatory, vascular endothelial, antioxidant and other effects of statins.4 These so-called “pleiotrophic” effects have led to statins being studied in a host of unrelated clinical settings, including osteoporosis, multiple sclerosis and Alzheimer’s disease. While research is ongoing, it appears that statins have profound multisystem effects that extend well beyond lipid metabolism. Statins are, on the whole, well tolerated, with the most commonly reported adverse effects being gastrointestinal upset, headache, rash and a dose-dependent elevation in serum levels of liver transaminases. The best characterised rare, but potentially serious, adverse effects are myopathy and polyneuropathy. These adverse effects are probably dose-related and may occur more often in patients taking medications known to inhibit statin metabolism.5 There are also a few reports of lupus-like syndromes, polymyositis/dermatomyositis with lung involvement, and hypersensitivity pneumonitis associated with statin therapy.1-3,6-9 The timing of onset appears unpredictable, with many patients having been taking statin therapy for many months or years before symptoms develop. Clinical features varied in severity from mild dry cough and rash through to severe and progressive respiratory failure. Low-titre antinuclear antibody (ANA) positivity and a raised erythrocyte sedimentation rate were also described in some patients. There are only four reports of open lung biopsy in these cases, two showing hypersensitivity pneumonitis with granuloma formation, one showing diffuse alveolar damage and the other showing non-specific interstitial pneumonitis. Most — but not all — patients responded to drug cessation and therapy with prednisolone or other potent immunosuppressive agents.2,8,9 Our patients shared many features with these patients. However, we did not observe rash of a dermatomyositic, lupoid or urticarial type, or lupus or polymyositis-like syndromes in our patients. Instead, all presented with respiratory symptoms — shortness of breath with or without dry cough — generally of insidious onset. Two out of six patients tested were positive for ANA with no other clinical features of connective tissue disease. The findings on transbronchial biopsy, when performed, were non-specific. Radiological appearances varied between alveolitis and fibrosis, but none showed the characteristic subpleural basal honeycombing that is common in usual interstitial pneumonia. The occurrence of alveolar eosinophilia in two out of the four patients who underwent bronchoalveolar lavage, which has previously been described in drug-hypersensitivity pneumonitis, and the response in several patients to drug cessation and/or corticosteroid therapy also point to a potential drug-induced pneumonitis. One previous case of pneumonitis has been reported in the setting of statin therapy, which was confirmed by open lung biopsy, and where the findings closely resembled those in amphiphilic drug toxicity, such as that reported with amiodarone. The authors hypothesised that a toxic mechanism, possibly mediated by statin effects on lipid metabolism, led to the observed intralysosomal lamellar inclusions in pneumocytes and interstitial cells.9 Thus, while not fully characterised, there is a putative mechanism through which statins may cause interstitial pneumonitis. Detecting rare side effects of commonly prescribed medications has always been a challenging task for the clinician. It is even more difficult when symptoms present insidiously, months or years after medication has been commenced, and the disease process has multiple potential aetiologies, which are not fully characterised. Similarly, interstitial lung disease remains a challenging diagnostic area even for experts in the field. Failure to recognise an under-lying cause of the pneumonitis often leads to the diagnosis of usual interstitial pneumonia being accepted. With continuation of the causative agent, the expected clinical pattern is progressive deterioration leading to respiratory failure and death, which also closely resembles the expected course of usual interstitial pneumonia. We hope that our description of our patients and review of the possible role of statins in interstitial lung disease will raise awareness of the potential association between statin therapy and this uncommon and often fatal condition. The effects of withdrawal of statin therapy on development of interstitial lung disease in patients taking this class of medication require further study. 1 New presentations of interstitial lung disease, 2000–2003 Diagnosis Number of patients Sarcoidosis 26 Usual interstitial pneumonitis 10 Non-specific interstitial pneumonitis 9 Drug-associated 8 Hypersensitivity pneumonitis 2 Asbestosis 2 Connective tissue disease-associated 1 2 Clinical details of seven patients with interstitial lung disease receiving statin (HMG-CoA reductase inhibitor) therapy Patient Sex, age Presentation Comorbidities Medications at presentation Investigations* Treatment Outcome 1 F, 78 Progressive dyspnoea, dry cough Hypertension, type 2 diabetes, hyperlipidaemia, non-smoker Atorvastatin (10 mg, 1 year), fosinopril, sertraline HRCT: extensive bilateral fibrosis TLCO: 46% No lavage or biopsy ANA-positive (1/40) Colchicine, atorvastatin withdrawn Slow progression: lung function deteriorated but clinical condition stable, little dyspnoea on 3-year follow-up 2 M, 78 Progressive dyspnoea (3 weeks), no cough or fever Chronic obstructive airways disease, IHD, atrial fibrillation, stroke, hyperlipidaemia, depression, ex-smoker Pravastatin (40 mg, 10 years), aspirin, frusemide, isosorbide mononitrate, perindopril, sertraline HRCT: extensive mid and upper zone emphysema, coarse bilateral basal fibrosis TLCO: 14%, mildly obstructed flow volume loop No lavage No biopsy (as poor respiratory reserve) ANA-positive (1/640) Prednisolone (50 mg), pravastatin withdrawn Progression, discharged with home oxygen, died 18 months later (respiratory failure) 3 F, 74 Cough and fever (3 days) (consistent with pneumonia) on background of worsening exertional dyspnoea Chronic obstructive airways disease, IHD, congestive cardiac failure, hyperlipidaemia, ex-smoker Simvastatin (10 mg for 2 years, then 20 mg for 1 year), aspirin, thyroxine, frusemide, diltiazem, nicorandil, long-acting nitrates HRCT: diffuse ground glass left upper lobe and bilateral lower zone No TLCO Lavage: 78% macrophages, 12% neutrophils, 9% lymphocytes Biopsy: non-specific interstitial pneumonitis No ANA test Prednisolone (50 mg), simvastatin withdrawn Gradual reduction in infiltrative change on imaging, lung function stable at 9-month follow-up 4 M, 83 Progressive dyspnoea (6 months) IHD, CABG and aortic valve replacement, atrial flutter, hyperlipidaemia, ex-smoker Pravastatin (40 mg, 1 year), aspirin, digoxin, frusemide, ramipril, ranitidine HRCT: scattered bilateral fibrosis TLCO: 33% Lavage: 68% neutrophils, 26% macrophages, 3% lymphocytes, 1% eosinophils Biopsy: non-diagnostic ANA-negative Prednisolone (40 mg), azathioprine, pravastatin withdrawn Slow progression 5 F, 67 Chronic mild dyspnoea (9 months), dry cough (6 months) Hyperlipidaemia, atypical chest pain, non-smoker Simvastatin (40 mg, 5 years), aspirin HRCT: patchy bilateral upper and lower zone ground glass TLCO: 22% No lavage or biopsy Negative for ANA, ANCA, normal ACE Prednisolone (25 mg, reduced to 10 mg after 3 months), simvastatin withdrawn Marked improvement (TLCO: 51% after 1 month, then 65% after 1 year) 6 M, 68 Progressive dyspnoea and hypoxia IHD, CABG and aortic valve replacement (10 years before), gastro-oesophageal reflux disease, hyperlipidaemia, ex-smoker Simvastatin (10 mg, 2 years), atenolol, candesartan, esomeprazole, frusemide, warfarin HRCT: bilateral fibrosis TLCO: 46%, mild restrictive defectLavage: 19% eosinophils, 19% neutrophils Biopsy: mixed inflammatory and fibrotic change ANA-negative Prednisolone (37.5 mg), azathioprine, simvastatin continued Progressive cardiac failure, died 9 months after presentation 7 M, 64 Progressive dyspnoea, dry cough IHD, CABG and aortic valve replacement (11 years before), peptic ulcer disease, hyperlipidaemia, ex-smoker Atorvastatin (20 mg for 3 years, then 40 mg for 2 years), fosinopril HRCT: bilateral ground glass infiltrates, fibrosis, some traction bronchiectasis TLCO: 44%, mildly obstructed flow volume loop Lavage: 38% eosinophils, 15% lymphocytes, 5% neutrophils Biopsy: thickened alveolar walls with interstitial fibrosis, minimal inflammation ANA-negative Prednisolone (initially 50 mg, then 10 mg maintenance), atorvastatin withdrawn Slight initial improvement (TLCO: 52% at 2-month follow-up), then stable disease HMG-CoA = hydroxymethylglutaryl-coenzyme A. IHD = ischaemic heart disease. CABG = coronary artery bypass graft surgery. HRCT = high-resolution computed tomography. TLCO = transfer factor for carbon monoxide diffusion. ANA = antinuclear antibody. ANCA = antineutrophil cytoplasmic antibody. ACE = angiotensin-converting enzyme. * All patients underwent HRCT and TLCO measurement, while most underwent bronchoscopy with broncho-alveolar lavage and transbronchial biopsy. 3 Investigations in two patients with interstitial lung disease High-resolution computed tomography in Patient 7 at presentation showed bilateral interstitial infiltrates with an area of honeycombing. Transbronchial biopsy specimen from Patient 7 showed a lymphocytic interstitial infiltrate and intra-alveolar macrophages (haematoxylin and eosin stain; original magnification, × 100). Transbronchial biopsy specimen from Patient 6 showed thickened alveolar walls with a low grade inflammatory infiltrate (haematoxylin and eosin stain; original magnification, × 400).

Tim Walker MB BS(Hons) · Joe McCaffery MB BS · Chris Steinfort FRACP

To the Editor: A seminal 1997 article by Fine et al described the pneumonia severity score from the Pneumonia Patient Outcomes Research Team study and raised the role for Hospital in the Home (HIH): For the remaining patients in [risk] classes II and III for whom treatment at home with oral antimicrobial therapy is judged to be unsuitable, there are alternatives to traditional inpatient care. These include parenteral antimicrobial therapy at home or a short stay . . . in a hospital observation unit.1 A recent article in the Journal by Charles et al2 omitted a role for HIH in managing community-acquired pneumonia (CAP). Where their protocol mentions outpatient care, readers are led to interpret this as oral therapy only, managed by a general practitioner. Similarly, it is implied that inpatient therapy relates to traditional treatment in a hospital ward. No further clarification is given. This is a surprising omission, given that one of the authors has written extensively in support of HIH in the past.3 HIH administers hospital-level therapy (intravenous antibiotics, oximetry, rehydration, medical and nursing attendance, with 24-hour cover) to a clinical subgroup of CAP patients who can be defined and included within any protocol. Evidence suggests that HIH can offer effective and safe treatment of patients with acute CAP referred directly from hospital emergency departments after diagnosis.4-6 Many patients with pneumonia appreciate the option of well organised, acute, home-based care. An important and growing subgroup of patients living in residential nursing care facilities can also receive acute CAP treatment in facilities with HIH involvement.7 A significant proportion of patients receiving HIH care have failed oral therapy.4-7 Why the omission of HIH? Protocols are tools of influence to be tussled over. This sometimes conflicts with their general aim of organising science into process and progress. Fine and colleagues’ intent in investigating the use of pneumonia severity scores was to help address the question of where and how to treat acute pneumonia. One of the aims of developing scores was to broaden the treatment options, not to narrow them.

Michael Montalto

To the Editor: The recent editorial on community-acquired pneumonia (CAP) stated that “even in an era in which penicillin resistance appears to be increasing among some Streptococcus pneumoniae isolates, there have been no documented failures of high-dose penicillin in treating pneumococcal pneumonia or bacteraemia”.1 The medical literature suggests otherwise. Firstly, North American guidelines do not mention penicillin at all, and, in one analysis of 25 996 hospitalised patients who received monotherapy, mortality was about 50% higher with penicillin monotherapy than with monotherapy with ceftriaxone, another cephalosporin, a macrolide or a quinolone.2 More importantly, however, the same study found that the mortality rate in patients (even low-risk patients) treated with two antibiotics, one of which was a macrolide, was half the mortality rate of patients treated with one antibiotic. Dual therapies used were a macrolide agent in combination with ceftriaxone, another cephalosporin, a penicillin or a quinolone. Best outcomes were achieved with a ceftriaxone–macrolide combination. In a review of seven studies, Waterer3 found that patients with severe pneumococcal pneumonia or bacteraemic pneumococcal disease who were treated with two antibiotics had a significantly lower mortality rate than patients treated with a single antibiotic. This was despite the fact that the patients treated with a single antibiotic were not as ill initially as those treated with two antibiotics. Research by Waterer and colleagues4 showed that the benefit of taking two antibiotics was most apparent in the highest risk hospitalised patients, in whom mortality was five times higher in those receiving one antibiotic than in those receiving two. Thus, it is imperative that all patients with severe pneumococcal CAP be treated with two antibiotics, one of which should be a macrolide.

Patrick J Bradley

In reply: Whether Hospital in the Home (HIH) care is suitable for managing patients with community-acquired pneumonia (CAP) depends on what is considered an appropriate use of resources. Overall, we see relatively few indications for treating CAP patients with parenteral antibiotics via HIH, as, in our experience, most patients who do not need supplemental oxygen and are well enough to be treated at home are usually also well enough to be treated with oral antibiotics. If they are not well enough to take oral antibiotics, then admission to hospital as an inpatient is generally appropriate. The occasional exceptions to this are selected patients in nursing homes (where around-the-clock supervision is available if required) and some patients with CAP caused by pathogens like Pseudomonas or Acinetobacter who benefit from longer treatment courses and may not have the option of oral antibiotics. Furthermore, a report submitted to the Victorian Department of Human Services regarding HIH care of CAP patients at a number of Melbourne HIH units identified significantly worse outcomes at some centres, mainly related to inappropriate patient selection. Notable, but fortunately rare, cases included some patients with pulmonary embolism incorrectly diagnosed as CAP. Given that between 20% and 50% of patients given a diagnosis of CAP in the emergency department do not have pneumonia confirmed by a radiologist,1-3 the ability of busy emergency department doctors to select patients appropriately is definitely a concern. Thus, HIH treatment of CAP patients may be appropriate occasionally, but very careful patient selection is vital. In response to Bradley, the statement that there have not been any failures in treating pneumococcal CAP with penicillins refers to microbiological failures due to antibiotic resistance. Although several studies have suggested that combination therapy may reduce mortality from bacteraemic pneumoccocal infections, all of these have been retrospective observational studies. Thus, they lack the ability to control accurately for potential confounding variables such as disease severity, patient or family wishes, pre-morbid quality of life, or “not for resuscitation” status. Data are also lacking on co-infection with “atypical” pathogens such as Legionella. The immunomodulatory effects of macrolides, quinolones and tetracyclines on treatment response are also still being elucidated.4 We agree with the CAP treatment recommendations in the Australian antibiotic guidelines,5 which recommend dual therapy with a β-lactam antiobiotic plus either a macrolide or doxycycline for all patients who are not allergic to these drugs.

Patrick G P Charles · Paul D R Johnson · M Lindsay Grayson

United airway disease is characterised by inflammation of the respiratory tract, in which asthma and rhinitis are the upper and lower respiratory tract manifestations, respectively, of the same disease process. Irrespective of cause, the upper and lower respiratory tract manifestations are characterised by a systemic inflammatory response. Patients with rhinitis or asthma should always be assessed for coexistent disease in the reciprocal area. Treatment of upper airway disease can modify the severity of lower airway disease and vice versa. The potential for early treatment of allergic rhinitis to prevent progression to asthma merits further study.

Janet Rimmer MD, MB BS, FRACP · John W Ruhno MB BS, FRACP

Clinically useful CAP management guidelines are still elusive Community-acquired pneumonia (CAP) continues to generate controversy. Although CAP is common and generally mild, it can be life-threatening. For the treating clinician there are many questions. How much effort should be directed to establishing the aetiology, given that the responsible pathogen is infrequently diagnosed? Should the patient be managed in hospital or at home? Should one choose older, established antibiotics that work most of the time or broad-spectrum therapy that treats all imaginable pathogens but is probably unnecessary, has a less established safety record and is likely to contribute to increased cost of treatment and the emergence of resistance? To help clinicians with these questions, international guidelines for managing CAP have been published.1-5 However, their clinical usefulness in the Australian health care context is questionable, as they are not based on particularly robust evidence and there is marked disagreement between Europeans and North Americans on the correct approach. The North Americans recommend that extensive investigations should be avoided, many patients should be treated at home, and broad-spectrum antibiotics should be used.1-3 The British and European guidelines are less focused on treating patients at home, suggest the use of cheaper, narrow-spectrum agents, and do not recommend dual therapy to treat both “typical” (eg, pneumococcus) and “atypical” (eg, Mycoplasma, Chlamydophila or Legionella) pathogens, except in patients with more severe illness.4,5 In comparison, the Australian antibiotic guidelines6 steer a middle course, suggesting the use of the Pneumonia Severity Index (PSI)7 to guide the decision regarding the need for hospital admission but then also using the PSI as a tool to assist with empirical antibiotic selection. This latter feature is somewhat unique and is based on local (as yet, unpublished) data. Some authors have suggested that investigations for CAP aetiology are not cost-effective.8 However, these opinions are often based on studies in which sputum samples were of poor quality or were collected after antibiotics were commenced. In most cases, such investigations won’t affect choice of therapy if the doctor treats for both “typical” and “atypical” agents. However, not performing these tests will mean missing the occasional unusual cause of CAP (such as Staphylococcus aureus, Legionella, community-acquired methicillin-resistant S. aureus and gram-negative organisms such as Pseudomonas). Furthermore, for hospitalised patients, the cost of these investigations is minimal compared with the cost of inpatient stay. Neglecting these investigations could lead to inappropriate management of patients who are initially thought to have CAP but who turn out to have an illness such as urinary sepsis or endocarditis. For patients who are sufficiently ill to require admission to hospital, we recommend that at least blood cultures and sputum Gram stain and cultures be performed. Apart from clinical acumen, what other tools can be used to assess the severity of CAP in an individual patient and hence guide the decision on site of care? The two most commonly used CAP severity scoring systems are the PSI7 and CURB-65 (Confusion, elevated Urea, elevated Respiratory rate, low Blood pressure and age at least 65).9 The key purpose of the PSI is to identify CAP patients who could be safely managed at home. However, it is reasonably complex, requiring input of 20 features of patient demographics, premorbid illnesses, initial vital signs and investigation results to calculate the PSI score. In addition, the PSI gives high weighting to patient age and past history but lower weighting to potentially important clinical features such as hypoxia. Thus, young, previously well patients can be classed as having mild CAP (PSI classes I–III), despite being hypoxic and having clinically severe disease. Despite these criticisms, the PSI has been validated on over 40 000 patients and appears to be accurate for predicting 30-day mortality both in the United States and Australia.7,10 For this reason it has been recommended in the current Australian antibiotic guidelines, but its uptake by Australian doctors has been limited.11 CURB-65 has the advantage of being simpler and more focused on the severity of the episode of CAP rather than the patient’s past history. However, a disadvantage is that it appears less useful for determining who is safe to be treated at home.12,13 Neither the PSI nor CURB-65 appears particularly useful for predicting accurately whether an individual patient will require admission to an intensive care unit.14 A recent Australian study suggested a modified version of CURB-65 as being more accurate for this purpose, but this is yet to be validated.15 Given these limitations, clinicians should be mindful that features such as hypoxia, vomiting, poor social circumstances, unstable comorbid conditions and empyema often indicate the need for hospital admission regardless of the PSI or CURB-65 score. Australian recommendations for empirical therapy are much closer to those of the UK and European guidelines than the North American guidelines, promoting the use of cheaper, narrow-spectrum agents such as penicillin and doxycycline.6 It is notable that penicillin is not mentioned at all in the North American guidelines.1-3 The Australian guidelines are supported by the fact that, even in an era in which penicillin resistance appears to be increasing among some Streptococcus pneumoniae isolates, there have been no documented failures of high-dose penicillin in treating pneumococcal pneumonia or bacteraemia. In comparison, there are documented cases of treatment failure with fluoroquinolones, and the widespread use of these agents has been clearly associated with increased levels of resistance to quinolones in S. pneumoniae and other previously susceptible bacteria.16 Given the importance of these issues, a large Australian, multi-state study of CAP is currently under way, with results to be made available in the next 12 months, to better guide clinicians. In the meantime, clinicians may find our general approach to patients with CAP useful (Box). Approach to managing the patient with community-acquired pneumonia (CAP) confirmed by chest x-ray PSI = Pneumonia Severity Index. * Abnormal vital signs are respiratory rate ≥ 30 breaths/minute, heart rate ≥ 125 beats/minute, systolic blood pressure < 90 mmHg. Young patients are less likely to be tachypnoeic. † Hypoxia is defined as partial pressure of oxygen (Pao2) < 60 mmHg or oxygen saturation measured by pulse oximetry (Spo2) ≤ 90% (in patients aged ≤ 40 years, use Pao2 < 70 mmHg or Spo2 ≤ 93%). ‡ Approximate costs: sputum testing, $34; blood culture, $31; Legionella urinary antigen testing, $29.

Patrick G P Charles MB BS, FRACP · Paul D R Johnson MB BS, FRACP, PhD · M Lindsay Grayson FRACP, MD, FAFPHM

Evidence “beyond reasonable doubt” may never be achievable for low-volume drugs The recent publication of a randomised controlled trial (RCT) of inhaled cyclosporin in the New England Journal of Medicine represents another milestone in the evolution of lung transplantation (LTx) as a standard therapy in the management of severe lung and pulmonary vascular diseases.1,2 RCTs have been few and far between in lung transplantation, and this is the first in such a high-profile general journal. However, the big question is: how will we integrate the study results into the clinical practice in Australia? Lung transplantation is a relatively recent and very demanding form of solid organ transplantation. The first long-term survivor received a lung transplant in 1981,3 and, compared with kidney and liver transplantation, there have been proportionally fewer lung transplantation procedures performed annually in Australia (100 lung, 400 kidney and 170 liver cadaveric transplants in 20044). Nonetheless, the four Australian lung transplant programs have performed about 6% of all lung transplants worldwide.5 At present, 100 Australians are awaiting suitable donor lungs. The efficacy of the various lung transplantation procedures themselves is determined primarily by comparison with historical control data, and the results of small cases series. The clinical practice of lung transplantation has generally developed by inference from other types of solid organ transplantation, as well as clinical anecdote and deduction. Indeed, we can identify only a handful of RCTs performed in lung transplant recipients that directly guide therapeutic decision making. Typically, the conduct of RCTs requires substantial resources and patient numbers. In lung transplantation, unless the effect size is huge, this typically mandates multicentre, multinational trials.6,7 Australian lung transplantation centres have recognised the need for good quality trials, and have consistently taken lead or major roles in virtually every significant RCT of an immunosuppressant published over the last decade.6-8 Five-year survival after lung transplantation (50%) lags well behind that of other forms of solid organ transplantation (kidney, 85%; liver, 80%4), and we continue to strive to increase our evidence base with the ultimate objective of improving clinical outcomes. Therefore, it seems somewhat ironic that, despite core Australian involvement in clinical trials generating evidence of the utility of newer immunosuppressive agents (tacrolimus, sirolimus, mycophenolate mofetil, everolimus), this evidence has not been readily integrated into Australian clinical practice. The reimbursed immunosuppressive protocols employed by lung transplantation programs in this country (cyclosporin, azathioprine and prednisolone) reflect decade-old practices. By contrast, in the United States, United Kingdom and Europe, newer agents have been rapidly integrated into practice. In Australia, there appears to be a progressively widening gap between the point at which there are sufficient data to support the use of a drug in a clinical situation (clinical risk versus benefit), and the point at which funding can be achieved that will allow the drug to actually be administered to patients. We submit that central to the problem in Australia is the extremely complex system of health care funding. Gaining Therapeutic Goods Administration (TGA) registration of a drug allowing it to be prescribed for a series of indications is only the first (relatively small) step. After that, there is extreme complexity as to who (if anyone) will pay. Previously, public hospitals have provided drugs that were indicated on clinical grounds but not subsidised by the (federal) Pharmaceutical Benefits Scheme (PBS). With a shift to predominantly casemix funding, hospitals with cutting-edge clinical programs have been forced to limit access to non-PBS items to prevent institutional “fiscal meltdown” from these largely unsupported costs. Generally, equitable access to an expensive drug in Australia requires listing on the PBS on the advice of the Pharmaceutical Benefits Advisory Committee (PBAC). The process is rigorous, evidence-based, and includes detailed pharmacoeconomic evaluation with a benchmark of cost-effectiveness. Our concerns are that the same level of evidence is required for frequently used medications (such as statins) compared with low volume drugs (such as anti-rejection drugs for lung transplant recipients) and the cost-effectiveness benchmark is not stated specifically. It is quite appropriate for regulatory authorities, hospitals and individual transplant physicians to appreciate the financial implications of novel agents, and not just focus on the evidence of efficacy. It is also important for 21st century physicians to recognise the commercial reality of drug patent laws, generic alternatives and pharmaceutical company marketing strategies. However, these issues create particular difficulties in small clinical fields like that of lung transplantation, where there may be sufficient evidence to support the use of a therapy, despite the absence of the multiple randomised trials specific to lung transplantation. It becomes apparent that it may not be commercially viable for a pharmaceutical company to perform the necessary trials and successfully negotiate registration and reimbursement for such a small patient group as lung transplant recipients. These issues are again brought to our attention by the recent RCT in the New England Journal of Medicine.2,8 This small single-centre RCT in lung transplantation is welcomed with great interest, as it shows efficacy for a novel inhaled cyclosporin formulation in enhancing key outcomes, including chronic rejection-free and overall survival (Box).2 The study had an accompanying editorial noting several methodological issues and, consistent with conservative evidence-based medicine, the editorial suggests the lung transplantation community undertake multicentre trials to avoid being “doomed to recreate a series of anecdotal experiences”.8 We are concerned, however, that repeating the RCT will take years to complete. Indeed, given the particular complexities of this agent, where multiple entities hold patents and licensing rights to different components of the therapy, no trial may ever be undertaken. There is a push to market the treatment on the existing evidence, and a case could be made that not to facilitate access to this novel therapy for Australian lung transplant recipients may cause greater harm by preventing the use of what appears to be a safe agent that strikingly reduces the risk of chronic rejection and death. We face a dilemma as lung transplant physicians (which we believe will resonate with many other clinicians) that on the “balance of probabilities” (to use a legal analogy), the benefits to our patient are likely to be greater than the risks, so it would be better to use the drug than not. However, to have the drug funded (and thus, available to our patients in a practical sense) realistically requires evidence “beyond reasonable doubt”. For the many reasons alluded to, this standard may never be achievable for low-volume drugs. Australia has been very well served by a centralised drug reimbursement system like the PBS, but we also believe it disadvantages patients with uncommon diseases who need high-cost drugs. If health economics are to be the principle determinant of funding, we need a transparent, consistent benchmark of cost-effectiveness across the entire health system. In the meantime, we ask that the thresholds of support for pharmaceutical funding be based on clinical appropriateness from the best available evidence, otherwise evidence-based medicine is in danger of becoming an economic weapon rather than a key clinical tool. Key findings of the inhaled cyclosporin lung transplant study2 Design and primary outcome 58 patients were randomly allocated, early after transplant, to receive 300 mg inhaled cyclosporin or placebo for 2 years, in addition to standard therapy. The primary outcome was rate of histological acute rejection. Results Acute rejection rates were the same for the two groups (0.44 v 0.46 episodes per patient; P = 0.87) Nephrotoxicity and opportunistic infection rates were similar. Survival was improved in the inhaled cyclosporin group (relative risk of death, 0.2; P = 0.01) Chronic rejection-free survival was improved in the inhaled cyclosporin group (relative risk of chronic rejection, 0.38; P = 0.01)

Greg Snell MB BS · Tom Kotsimbos MD, FRACP · Trevor J Williams MB BS, FRACP

Objectives: To identify expressions used by patients to describe worsening asthma; to examine the relevance of the word “exacerbation” to patients’ experience; and to investigate whether their language is influenced by the severity of the episode and/or the target audience such as family members, friends and work colleagues.Design and setting: Qualitative study carried out from 1 January to 30 December 2004 among community volunteers to a research institute. Semistructured face-to-face interviews were used to elicit descriptions of episodes of worsening asthma, and further questioning was used to examine language used with family, employer and doctor.Participants: 25 people with asthma, aged 22–75 years.Main outcome measure: Themes identified by open coding about patient language for worsening asthma.Results: 12 participants were not familiar with “exacerbation” and only three would use it themselves. “Attack” was the only specific term spontaneously volunteered (20 participants), but it was used for anything from mild to life-threatening episodes. Patients often downplayed the severity of worsening asthma to their families. Different language was used with employers, sometimes to justify sick leave and sometimes because of fear of perceived discrimination. When communicating with clinicians about worsening asthma, patients used symptom descriptors rather than specific terms.Conclusion: There are important differences in the language patients and clinicians use to describe worsening asthma, and the word “exacerbation” has poor utility for communication with patients.

Stephen D Vincent MB BS, FRACP · Brett G Toelle DipAppSc(Nursing), BA(Psychology), PhD · Rosalie A Aroni PhD · Christine R Jenkins MB BS, FRACP, MD · Helen K Reddel MB BS, FRACP, PhD

The aetiology and management approach for cough in children differs greatly to that in adults, so the empirical approach commonly used in adults is unsuitable for children. Clinical evaluation of cough in children should include an assessment of environmental factors, particularly tobacco smoke, parental concerns and expectations. Most children with acute cough are likely to have an uncomplicated viral acute respiratory tract infection, but the possibility of a more serious problem, especially aspiration of foreign material, should always be considered. Isolated chronic cough in children is rarely asthma, and the term “cough variant asthma” should not be used. Over-the-counter and prescription medications are ineffective for the symptomatic relief of acute cough. Treatment for chronic cough should be based on aetiology. Because of the favourable natural history of cough, a “positive” response in medication trials should not be assumed to be due to the medication. Children should be reassessed within the expected timeframe of response to therapy.

Anne B Chang MPHTM, PhD, FRACP · Lou I Landau MD, FRACGP · Peter P Van Asperen MD, FRACP · Nicholas J Glasgow MD, FRAGP · Colin F Robertson MD, FRACP · Julie M Marchant MB BS · Craig M Mellis MD, MPH, FRACP

Long-acting β2 agonists are an effective and convenient treatment for chronic obstructive pulmonary disease (COPD), but do not significantly improve lung function. The long-acting anticholinergic tiotropium, which can be taken once daily, decreases exertional dyspnoea and increases endurance by reducing hyperinflation. The role in COPD of the combination of a long-acting β2 agonist and a glucocorticoid in a single inhaler remains unclear. The minimum duration of an effective pulmonary rehabilitation program that includes exercise training is 6 weeks. Long-term treatment with inhaled glucocorticoids may reduce the rate of decline in lung function, but the effect is small. Aminophylline should no longer be routinely used in acute exacerbations of COPD. Non-invasive positive pressure ventilation (NPPV) reduces mortality and hospital stay in patients with acute hypercapnic ventilatory failure; it is also an effective weaning strategy for patients who require intubation. Further studies are required to clarify the role of NPPV in the long-term management of stable COPD.

Michael J Abramson PhD, FRACP, FAFPHM · Alan J Crockett PSM, PhD, FANZSRS · Peter A Frith MD, FRACP · Christine F McDonald PhD, FRACP

Middle lobe syndrome — recurrent atelectasis and/or bronchiectasis involving the right middle lobe and/or lingula — has, up to now, not been reported as the pulmonary manifestation of primary Sjögren’s syndrome. We describe a patient in whom lymphocytic bronchiolitis in the atelectatic lobes was proved histologically from two separate transbronchial biopsies. The atelectasis responded well to glucocorticoid treatment, suggesting that the peribronchiolar lymphocytic infiltrates may have played an important role in the development of middle lobe syndrome in this patient. Clinical record A 53-year-old woman was admitted to our hospital in 2003 with symptoms, for the past 3 days, of shortness of breath, cough, and blood-tinged sputum. Primary Sjögren’s syndrome had been diagnosed in 2002, based on the presence of xerostomia, keratoconjunctivitis sicca, anti-Ro/La antibody, and rheumatoid factor, as well as a positive Schirmer’s test and the results of sialoscintigraphy. The patient had no prior history of smoking, fever, rhinitis or pharyngitis. Physical examination was unremarkable, except for a rapid respiratory rate (20 breaths/min) and coarse crackles heard during the early inspiratory phase in the middle lung field. A complete blood count, C-reactive protein levels, erythrocyte sedimentation rate, and the results of serum biochemistry were all within normal limits. A chest x-ray revealed a vague opacity in the right lower lung that obliterated the cardiac border in the posteroanterior view, and a wedge-shaped density in the cardiac area in the lateral projection (Box, A). Sputum studies for bacteria, mycobacteria and fungi, and serological tests for Mycoplasma pneumoniae and Legionella pneumophila all gave negative results. A computed tomography (CT) scan of the chest showed inhomogeneous opacities, as well as dilated airways crowded in the medial segment of the right middle lobe and inferior segment of the lingula, without mediastinal lymphadenopathy (Box, B). A fibrobronchoscopic examination to verify central bronchial patency showed no intraluminal obstruction in the affected bronchi. Cultures of the lavaged bronchial fluid were negative for tuberculosis and other microorganisms, and no malignant cells were found in aspirated specimens. A transbronchial biopsy from the atelectatic middle lobe revealed lymphocytic bronchiolitis with a moderate degree of mononuclear cell infiltration, predominantly by lymphocytes, in the terminal bronchiolar walls and adjacent interstitial areas, and no granuloma formation (Box, C). A diagnosis of middle lobe syndrome secondary to Sjögren’s syndrome was made and the patient was initially treated with methylprednisolone (125 mg/day for 4 days), followed by prednisolone (20 mg/day). The glucocorticoid dose was tapered off as the atelectatic lesions resolved. A repeat CT scan performed 1 year later showed that the atelectatic lesions had almost completely disappeared with this treatment (Box, D). Subsequently, the prednisolone dose was ceased. In 2005, atelectasis developed once again in the lingula. Malignancy and infection were excluded after extensive studies, including cytological examination, virus isolation, and bacterial and mycobacterial cultures from the endobronchial aspirates. Pulmonary tissue, obtained from the atelectatic lingula by biopsy, again showed evidence of lymphocytic bronchiolitis. With the use of medium-dose prednisolone (20 mg daily), the atelectasis gradually resolved. At the time of writing, there has been no further relapse in this patient. DiscussionMiddle lobe syndrome is a disorder of recurrent or fixed atelectasis involving the right middle lobe and/or lingula.1-3 It can result from either extraluminal or intraluminal bronchial obstruction, but also may develop in the presence of a patent lobar bronchus without identifiable obstruction.1-3 Inflammatory processes and defects in the bronchial anatomy and collateral ventilation have been designated as the non-obstructive causes of middle lobe syndrome.1-3 Middle lobe syndrome is commonly associated with many underlying lung conditions, such as neoplasms, asthma, broncholithiasis, microbial infections, granulomatous disorders, mucous plugging, or foreign body aspiration.1,2 It has, up to now, not been reported in patients with primary Sjögren’s syndrome, although similar radiological findings have been described in a patient with systemic lupus erythematosus with suspected secondary Sjögren’s syndrome.4 Primary Sjögren’s syndrome is a chronic inflammatory autoimmune disease characterised by lymphocytic infiltrations in the involved organs. This disorder predominantly affects the salivary and lacrimal glands, but can also involve internal organs resulting in many extraglandular complications.5 A wide spectrum of pulmonary manifestations has been described, including xerotrachea, obstructive airway disease, interstitial lung disease, and lymphoproliferative disease.6 With pulmonary involvement in primary Sjögren’s syndrome, the inflammation can be focal and predominantly cellular, with lymphocytes infiltrating the small airway walls.7,8 Local bronchial or bronchiolar lymphocyte infiltrates and inflammatory cell products may lead to impairment of tracheobronchial mucociliary clearance.8,9 In our patient, lymphocytic infiltrations in the terminal bronchioles were found in the atelectasis of the right middle lobe and lingula. The atelectatic and bronchiectatic lesions seen on the chest CT scan resolved after immunosuppressive drug treatment. Lymphocytic bronchiolitis with underlying bronchiectasis has also been described in patients with non-obstructive middle lobe syndrome. Our findings suggest that the peribronchiolar lymphocytic infiltrates may have been involved in the pathogenesis of the middle lobe syndrome in our patient. Lymphocytic bronchiolitis is common in primary Sjögren’s syndrome, but pulmonary atelectasis, as in our patient, is fairly rare.6,10,11 Therefore, mechanisms other than impaired mucociliary function and lymphocytic bronchiolitis may also have contributed to the development of this disorder. The right middle lobe bronchus is at higher risk of obstruction from inflammatory processes, as it has a relatively narrow diameter and an angular attachment to the intermediate bronchus. Compared with other areas of the lung, the middle lobe and lingula have a large ratio of pleural to non-pleural surface because of the presence of deep fissures, which are effective barriers to the collateral ventilation.,3 These anatomical features may thus predispose to the formation and persistence of atelectasis. Immunosuppressive drugs, either azathioprine alone or in combination with prednisolone, are effective in treating the pulmonary complications of primary Sjögren’s syndrome.7 In our patient, the lymphocytic bronchiolitis-associated atelectasis resolved after prednisolone therapy, but relapsed when the drug was ceased. Therefore, we think that immunosuppressive drugs, such as glucocorticoids, should be considered in this unusual pulmonary complication of primary Sjögren’s syndrome. Tapering of the dose must be gradual to avoid disease recurrence, and a long follow-up period is necessary. Box A A: Lateral chest radiograph showing a wedge-shaped dense area (arrows) with the apex pointing to the hilum. Box B B: Computed tomography scan of the lung showing atelectasis and bronchiectasis in the medial segment of the right middle lobe (arrow) and the inferior segment of the lingula (arrowhead). No pulmonary fibrosis or space-occupying lesion was observed elsewhere. Box C C: Lung tissue from a transbronchial biopsy revealing lymphocytic infiltrations in the terminal bronchiole (arrow) and adjacent interstitial area (arrowhead) (haematoxylin?eosin stain, original magnification, × 100). Box D D: Computed tomography (CT) scan performed 1 year after the first CT scan, showing almost complete resolution of the atelectatic lesions (arrow and arrowheads).

Horng-An Chen MD · Shinn-Liang Lai MD · Wei-Kang Kwang MD · Juhn-Cherng Liu MD · Chun-Hsiung Chen MD · De-Feng Huang MD

Objectives: To examine changes in asthma prevalence in the context of other population changes between 1990 and 2003, for specific age and sex groups.Design: Cross-sectional survey based on household interviews, repeated annually.Setting and participants: Representative samples of the South Australian population between 1990 and 2003 (around 3000 people per year).Main outcome measures: Current prevalence of doctor-diagnosed asthma and other health and demographic variables potentially associated with asthma, and asthma management.Results: Response rate was over 71%. Between 1990 and 2003, asthma prevalence increased significantly, doubling in females (from 7.3% in 1990 to 14.6% in 2003), with a smaller increase in males (from 7.8% to 9.4%). Asthma also increased in all age groups, but the largest relative increases occurred in people aged 55 years and older. Logistic regression analyses showed that obesity was a major predictive variable for every age group studied. The prevalence of asthma morbidity (waking at night and days lost from usual activities because of asthma) among those with asthma showed no significant changes between 1990 and 2003. Asthma action plans (introduced on a population basis in 1992) peaked in their distribution at 42% in 1994, and then declined to half that percentage in 2003. The increase in asthma prevalence occurred at the same time as increases in population prevalence of obesity (10.3% to 18.7%) and diabetes (3.1% to 6.9%), and decline in recent vigorous exercise (42.4% to 32.7%).Conclusions: The increase in asthma prevalence over a decade was large, but concentrated among specific sex and age groups. The increase accompanied population increases in obesity and diabetes and a decline in vigorous exercise.

David H Wilson PhD, MPH, BEd, AdvDipEd · Robert J Adams MD · Richard E Ruffin MD · Graeme Tucker BSc · Anne W Taylor MPH · Sarah Appleton BA

Objective: To evaluate the efficacy of supplementation with zinc and vitamin A in Indigenous children hospitalised with acute lower respiratory infection (ALRI).Design: Randomised controlled, 2-by-2 factorial trial of supplementation with zinc and vitamin A.Setting and participants: 187 Indigenous children aged < 11 years hospitalised with 215 ALRI episodes at Alice Springs Hospital (April 2001 to July 2002).Interventions: Vitamin A was administered on Days 1 and 5 of admission at a dose of 50 000 IU (infants under 12 months), or 100 000 IU; and zinc sulfate was administered daily for 5 days at a daily dose of 20 mg (infants under 12 months) or 40 mg.Main outcome measure: Time to clinical recovery from fever and tachypnoea, duration of hospitalisation, and readmission for ALRI within 120 days.Results: There was no clinical benefit of supplementation with vitamin A, zinc or the two combined, with no significant difference between zinc and no-zinc, vitamin A and no-vitamin A or zinc + vitamin A and placebo groups in time to resolution of fever or tachypnoea, or duration of hospitalisation. Instead, we found increased morbidity; children given zinc had increased risk of readmission for ALRI within 120 days (relative risk, 2.4; 95% CI, 1.003–6.1).Conclusion: This study does not support the use of vitamin A or zinc supplementation in the management of ALRI requiring hospitalisation in Indigenous children living in remote areas. Even in populations with high rates of ALRI and poor living conditions, vitamin A and zinc therapy may not be useful. The effect of supplementation may depend on the prevalence of deficiency of these micronutrients in the population.

Anne B Chang MPHTM, FRACP, PhD · Paul J Torzillo FRACP, FFICM · Peter M Stewart FRCPA · Naomi C Boyce BNurs · Andrew V White FRACP · Gavin R Wheaton FRACP · David M Purdie BSc(Hons), PhD · John Wakerman MB BS, MPH · Patricia C Valery MD, MPH, PhD

Allen C Cheng,* Andrew J Hughes,* Julian B Stella,† Eugene Athan* * Infectious Diseases Physician, Hospital in the Home Programme, Department of Infectious Diseases, † Emergency Department Physician, Geelong Hospital, Geelong, VIC 3220. allencATmenzies.edu.au To the Editor: We note with interest the studies published in the Journal by Richards et al and Ong et al.1,2 The authors conclude that treating pneumonia and pulmonary emboli in an ambulatory setting is safe for selected patients. However, this represents a large change in the conditions traditionally treated on this basis, from conditions that are associated with a very low mortality (such as cellulitis) to a subgroup of patients with potentially serious infections that are identified as being of low risk. We feel that safety is of prime importance in hospital-in-the-home programs because of limited or delayed access to acute medical care, and that both studies were underpowered to define this endpoint. Both studies incorrectly quote previous work that suggests that the groups they have identified have mortality rates of up to 5% (for pulmonary emboli) and up to 9.2% (for mild to moderate pneumonia). Published data suggest that the mortality of mild pneumonia (with CURB-65 scores ≤ 2) is in the range 1.7%–3%,3,4 and that mortality from treated sub-massive pulmonary emboli is in the range 1.0%–1.3% within the first week.5 These rates, although seemingly small, are still much higher than that associated with the treatment of soft tissue infections on ambulatory care programs. Recurrent pulmonary embolus, in particular, may be sudden and unexpected. Although admission to hospital may not necessarily prevent these deaths, the additional trauma of a death at home, particularly soon after transfer to ambulatory care, may carry a higher significance in the minds of patients, their families and the public than a death in hospital. We acknowledge that benefits for patients in ambulatory treatment programs need to be balanced against potential adverse outcomes. However, if these conditions are to be treated where access to medical attention may be delayed, it is imperative that informed consent be obtained from patients (including an awareness of the possibility of death), a mechanism be available for patients to summon urgent attention at any time, and patients and health care providers be aware that readmission to the hospital may be necessary in the event of clinical deterioration.

Allen C Cheng · Andrew J Hughes · Julian B Stella · Eugene Athan

Dee A Mangin (née Richards),* Les J Toop,† Michael J Epton,‡ Graham R B McGeoch,§ G Ian Town,¶ Simon M H Wynn-Thomas,** Robin D Dawson,†† Michael C Hlavac,‡‡ Anja M Werno,§§ Paul D Abernethy¶¶ * Senior Lecturer, † Head, †† Research Fellow, Department of Public Health and General Practice, ‡ Senior Lecturer, ¶ Dean, ‡‡ Research Fellow, Department of Medicine, Christchurch School of Medicine and Health Services, Otago University, PO Box 4345, Christchurch, New Zealand; § Director, Community Care, ** Medical Director, Extended Care @ Home, ¶¶ Manager, Health Services, Pegasus Health Independent Practitioners Association, Christchurch, NZ; §§ Community Patholgist, Canterbury Health Laboratories, Christchurch, NZ. derelie.manginATchmeds.ac.nz In reply: Thank you for the opportunity to reply to the letter from Cheng et al. The mortality figures we cited are correct.1 The cited article by Lim et al supports our statement that “Patients with a CURB-65 score of 0–2 have a low mortality (0.7%–9.2%)” (Table 4 shows mortality for CURB-65 score 0 is 0.7% and for score 2 is 9.2%).2 The 3% figure in the abstract is a summary measure obscuring the difference across the CURB 0–2 range — important information for anyone considering community management of community-acquired pneumonia where, we agree, safety is paramount. The rate of 1.7% cited by Cheng et al is for a modified CURB-65 score, which adds a further point, and thus is for the equivalent of CURB-65 scores of < 1.3 Cheng et al correctly observe our study was not powered to detect mortality differences. As explained in our discussion, mortality was not a primary outcome measure. With low mortality, large numbers are required to detect a statistically significant difference — the base rate of 3% in the validation study would require 10 602 patients in a randomised controlled trial to detect a 33% relative (1% absolute) increase in mortality. The study did provide for informed consent (including the possibility of readmission) and the ability to summon urgent attention. Careful patient selection, routine twice-daily nurse and daily doctor visits, along with a highly trained nurse available by telephone 24 hours a day who can dispatch a doctor or nurse immediately, provides a structure that should match hospital care. Careful patient monitoring will detect failure to respond as expected. It is important to treat in hospital those who will benefit, but not feasible to admit all with potential mortality risk (nor is there evidence of benefit). Hospitalisation also has risks. With this tool for predicting accurately who will suffer worse outcomes, it could be argued there has to be good evidence that better outcomes will result from continuing inpatient treatment of mild to moderate community-acquired pneumonia. These wider issues are worthy of debate. There is an assumption by some professionals and consumers that hospital-sanctioned death is more acceptable, that everything possible has been done, and that community-based death implies unsatisfactory management. As a counterpoint to this, there is a clear patient preference for treatment in the home where possible. Avian influenza may, of course, drastically redefine our expectations about locus of care and of death.

Les J Toop · Michael J Epton · Graham R B McGeoch · G Ian Town · Simon M H Wynn-Thomas · Robin D Dawson · Michael C Hlavac · Anja M Werno · Paul D Abernethy

Bin Soo Ong,* Margaret A Karr,† Daniel K Y Chan,‡ Anthony Frankel,§ Qing Shen¶ * Director, Department of Ambulatory Care, † Research Manager, ‡ Director, ¶ Research Assistant, Department of Aged Care and Rehabilitation, § Respiratory Physician, Bankstown-Lidcombe Hospital, Locked Mail Bag 1600, Bankstown, NSW 2200. bin.ongATswahs.nsw.gov.au In reply: We acknowledge the concerns of Cheng and colleagues regarding the safety of patients with pulmonary embolism (PE) treated in an ambulatory care setting. Caution is important as this is a relatively new area of treatment in ambulatory care compared with the management of deep venous thrombosis. The main objective of our study was to describe our experience in the management of PE in ambulatory care; it was not a randomised controlled study to conclusively define safety as such. As stated in our paper, there have been reports of the management of PE in the ambulatory care setting.1,2 We now know that more than 90% of patients with sub-massive PE will have a good response to treatment. The challenge is to accurately define this group. The mortality rate we quoted of less than 5% was derived from a review article on prognosis of patients with PE.3 This article quoted three studies on sub-massive PE, one of which was referenced by Cheng and colleagues in their letter.4 We note also that the specific study that was referenced4 included patients with cyanosis and shock; these patients would have been excluded by our selection criteria. We do not advocate management of all patients with sub-massive PE in the ambulatory care setting. It is also important to be conservative initially in the selection of these patients. There have been various studies examining prognostic indicators for PE, which we have referenced in our paper. There is evidence now that, for patients with specific prognostic indicators, the risk of death and adverse outcomes is significant and such patients should always be admitted. The practice of managing patients with sub-massive PE should only occur in ambulatory care units which are appropriately resourced, have strict admission criteria and well defined protocols and specialist medical input, consistent with the recommendation of the British Thoracic Society.5 In the meantime, further studies are required before this becomes standard practice in ambulatory care or hospital-in-the-home units.

Bin Soo Ong · Margaret A Karr · Daniel K Y Chan · Anthony Frankel · Qing Shen