Journal of J Accid Emerg Med: first published as 10.1136/emj.11.4.234 on 1 December 1994. Downloaded from Accident and Radiographic features in 1016 adults admitted to Emergency with acute asthma Medicine 1994 hospital 11, 234-237 C.M. PICKUP, P.A. NEE & P.E. RANDALL

North Manchester General Hospital, Delaunays Road, Crumpsall, Manchester

that are likely to influence treatment in otherwise SUMMARY uncomplicated cases admitted with dyspnoea and A retrospective review was carried out of the radio- wheeze.3'4 graphic features of 1016 adults admitted to hospital It is clearly undesirable to irradiate a patient with acute asthma over a 4-year time period. unnecessarily and to expend financial and time The radiographic features were classified into resources on an investigation with a poor return in five groups: (I) normal, 536 patients (52.9%); (11) detecting clinically significant lesions. To date there features compatible with obstructive lung disease, has been no large scale United Kingdom based 323 patients (31.8%); (III) complications of asthma survey of the radiographic findings in patients including infection, segmental or greater atelectasis, hospitalized with acute asthma. one case of pneumomediastinum and one case of pneumothorax, 83 patients (8.2%); (IV) unimportant SUBJECTS AND METHODS incidental findings, six cases (0.6%); and (V) impor- tant incidental findings including tuberulosis, heart We carried out a retrospective review of the radio- failure, and bronchial neoplasm, 68 cases (6.7%). graphic features of all patients discharged from an We conclude that in this large series of patients inner-city district general hospital with a diagnosis presenting with asthma symptoms severe enough compatible with acute asthma (International Class- to merit admission there is an incidence of clinically ification of Disease Codes 493.0, 493.1 and 493.9) significant radiographic abnormalities of appro- between the 1 April, 1989 and 31 March, 1993. The ximately 15%. Admission chest radiography is names and district numbers of all these patients therefore indicated in adults who are hospitalized were accessed by means of a computerized search http://emj.bmj.com/ with acute asthma. of the Patient Administration System. These data Key words: adult in-patients, asthma, asthma were then input into the Kodak Radiological Infor- radiography, chest radiography mation System and the radiologist's report was obtained for the admission chest radiograph relating to each episode.

INTRODUCTION on October 1, 2021 by guest. Protected copyright. Radiographs were ordered by junior doctors in The majority of patients admitted to hospital with the specialties of accident and emergency (A&E) or symptoms of acute asthma undergo a chest radio- general medicine. Widely publicized guidelines on graph on admission. Routine admission chest the management of acute severe asthma in adults5 radiography is aimed at identifying the important and the utility of radiographs6 were available for complications of asthma such as pneumothorax, most of the study period. pneumomediastinum and atelectasis as well as The radiographs were reported by a radiological identifying treatable associated conditions such as staff of six consultants, three senior registrars and Correspondence: pneumonia and heart failure. It is also necessary to six registrars. P.A. Nee, Consultant rule out pulmonary tuberculosis before initiating in Accident and steroid therapy. RESULTS Emergency Previous studies have reported a substantial yield Medicine, Whiston on the admission chest In this 4-year time period 1218 patients were dis- Hospital, Prescot, of abnormalities radiographs Merseyside L35 of asthmatic patients.1' 2 However, others have charged (or died) with a diagnosis of asthma. 5DR, UK suggested that it is unusual to find abnormalities Admission radiographs were performed in 1016 C. Pickup et al. patients (384 males). Children are not admitted to (IV) Unimportant incidental findings. There were J Accid Emerg Med: first published as 10.1136/emj.11.4.234 on 1 December 1994. Downloaded from this hospital, the age range of all patients was six patients (0.6%) in this group. Radio- 16-94 years. Some patients were admitted more graphic features were two cases each of than once and some had more than one feature on fibrosis and old fractured ribs and one case their radiograph. For the purposes of this study each of pulmonary hypoplasia and previous multiple admissions were regarded as separate pneumonectomy. episodes. (V) Important incidental features. Sixty-eight Radiological findings were classified into the patients (6.7%) had radiographic findings following five groups. consistent with an alternative diagnosis for (I) Normal. Five hundred and thirty six radio- their presenting symptoms. Radiographic graphs (52.9%) revealed no abnormal signs. findings included evidence of heart failure (II) Features compatible with a diagnosis of (cardiomegaly, pulmonary oedema or pleural uncomplicated obstructive lung disease. There effusion), pulmonary tuberulosis and broncho- were 323 cases (31.8%) in this group. Patients genic carcinoma. were included in this group if the radiograph The incidence of the various radiological features contained features such as hyperinflation, are summarized in Table 1. increased lung markings or a combination of Groups (Ill) and (V) were regarded as including the two. the radiological features that would have a direct (111) Important complications. There were 83 bearing on the clinical management of the patients. patients (8.2%) in this group including 68 with When added together they account for 151 radio- radiographic evidence of pulmonary infection graphs, approximately 15% of the total. There was or consolidation and 13 with atelectasis of no significant difference in age or sex distribution segmental or greater extent. There was just between these groups and the rest of the population. one case of pneumothorax, a 43-year-old female, and one case of pneumomediastinum, a 17-year-old male.

Table 1. Findings in 1016 Row Mean admission chest radiographs in Column (%) age adults admitted to hospital with Group Finding Incidence (%) Males (range) acute asthma

I Normal 536 52.7 35 45.5 (16-87) http://emj.bmj.com/ (a) Hyperaeration (HA) 184 (b) Increased lung markings/subsegmental atelectasis (ILM/SSA) 57 (c) HA + ILM/SSA 82

Totals 323 31.8 39 61.1 on October 1, 2021 by guest. Protected copyright. (a) Pulmonary infection/ (16-84) consolidation 68 (b) Segmental atelectasis 13 (c) Pneumothorax 1 (d) Pneumomediastinum 1 Totals 83 8.2 41 43.9 IV (a) Pulmonary fibrosis 2 (21-80) (b) Hypoplasia 1 (c) Pneumonectomy 1 (d) Old fractured ribs 2 Totals 6 0.6 50 42.5 V (a) Cardiac failure 57 (27-61) (b) Pulmonary tuberculosis 10 (c) Carcinoma 1 Totals 68 6.7 34 63.6 (22-94) 235 examination of radiological findings in adult patients J Accid Emerg Med: first published as 10.1136/emj.11.4.234 on 1 December 1994. Downloaded from Asthma DISCUSSION radiography admitted to hospital with exacerbations of asthmatic It is routine practice to obtain a radiograph of the symptoms. The diagnosis of asthma was ascribed chest in patients admitted to hospital with severe to these patients by the senior physician in charge obstructive lung disease including asthma. Chest of the case. Many patients would therefore have radiographs comprise approximately 33% of all acquired a clinical coding compatible with a diag- exposures in UK practice.6 Most of these radio- nosis of asthma when there was an element of graphs are normal or show features consistent with reversible bronchospasm, regardless of the presence uncomplicated chronic obstructive lung disease.7'8 of chronic obstructive lung disease. The occurrence The radiological features most frequently observed of abnormalities likely to lead to a change in the are hyperinflation, increased lung markings, peri- clinical management of the patient the percentage bronchial thickening, and subsegmental atelectasis. incidence is 14.8%, which relates closely to the Some authorities believe that routine chest radio- findings of others. graphy is indicated to rule out important compli- We have not attempted to correlate clinical cations like pneumonia and pneumothorax and findings with radiological features. Others have associated conditions which may mimic acute made this correlation in smaller studies on the basis bronchospasm such as heart failure and, especially of the history,4 the physical signs and the results of in children, inhaled foreign body.1' 2 It is also essential special tests.11 They have shown that application of to exclude potential complications of assisted exclusion criteria can reduce admission radiographic ventilation especially pneumothorax9 and any exposures by around 50%. Others, however, have contra-indications to the commencement of steroids pointed to the difficulty in diagnosing the presence therapy such as evidence of pulmonary tuberculosis. of pneumonia on a clinical basis.14. The use of The frequency of radiological abnormalities on antibiotics is significantly affected by the demon- routine chest radiographs in adult asthmatics has stration of focal pulmonary opacities even in the been shown to be as low as 2.2%.10 It has therefore absence of physical signs of pneumonia.12 been suggested that many patients could be spared Pneumothorax is a dangerous complication of the time, the expense and the radiation of multiple asthma which is frequently clinically silent until the exposures which rarely contribute to clinical man- patient suffers a catastrophic deterioration.15 The agement decisions.11 incidence of pneumothorax or pneumomediastinum A selective approach is recommended for patients in the present study (0.2%) is similar to the 0.15% presenting to the A&E department whereby radio- incidence reported by Burke.16 There was a rela- graphs are ordered depending upon clinical criteria tively high frequency of other conditions likely to http://emj.bmj.com/ such as acute deterioration,6 failure to respond affect management in the present series. Atelectasis to therapy12 or a history of immunosuppression.4 for example, at the segmental or greater level par- Guidelines produced by the British Thoracic Society ticularly influences the physiotherapy regime. Heart and others recommend chest radiography to failure will determine the prescription of diuretic exclude pneumothorax in patients with severe therapy and a chest film compatible with pulmonary acute asthma.5 tuberculosis will prompt a search for activity of the on October 1, 2021 by guest. Protected copyright. Patients admitted to in-patient beds represent the disease as well as a cautious approach to the more severe end of the spectrum and it is usual for prescription of oral steroids. them to have a chest radiograph performed prior to Our study may be criticized on the grounds that it admission. is a retrospective review of patients by discharge There have been few studies of the frequency of diagnosis of asthma. It is theoretically possible, radiographic abnormalities in adult patients admitted therefore, that some patients may have been coded to hospital with exacerbations of obstructive lung under another diagnosis. For example an asthmatic disease. One group looked at 242 admission chest presenting with wheeze and dyspnoea complicated radiographs11 and another a total of 125 radio- by pulmonary infection may be coded as pneumonia. graphs.4 The only UK based study that we know of This is unlikely to affect our findings seriously examined the radiographic findings in 135 adults because of the large number of episodes we cap- and children seen in an A&E department.13 The tured using the present methodology. In general range of significant radiographic abnormalities in most asthmatic patients presenting to A&E depart- these three studies was 10.4-14.8%. ments do receive codes consistent with a primary 236 The present study represents the largest ever diagnosis of asthma. In one review of 6000 emerg- C. Pickup et al. ency department visits no example of the occurrence 2. Eggleston P.A., Ward B.W., Pierson W.E. & Bierman J Accid Emerg Med: first published as 10.1136/emj.11.4.234 on 1 December 1994. Downloaded from of inappropriate discharge diagnosis was found.11 C.W. (1974) Radiographic abnormalities in acute We are unable to report which particular radio- asthma in children. Pediatrics 54, 442-449. graphic technique was used to obtain the chest 3. Findley L.J. & Sahn S.A. (1981) The value of chest films in all cases. However, practical experience Roentgenograms in acute asthma. Chest 80, teaches us that the majority of these films will be 535-536. 4. Aronson S., Gennis P., Kelly D., Landis R. & Gallagher single view anteroposterior films taken portably J. (1989) The value of routine admission chest radio- within the resuscitation area. This introduces another graphs in adult asthmatics. Annals of Emergency potential bias in our study. In a previous study the Medicine 18(11), 1206-1208. yield of major abnormalities was 34% when adult 5. British Thoracic Society and others (1990) Guidelines asthma patients received a standard erect postero- for the management of asthma in adults. British anterior and left lateral film.12 If anything then, our Medical Journal 301, 797-800. finding of major abnormalities in 14.8% is an under- 6. Royal College of Radiologists (1989) Making the representation. Best Use ofa Departmentof Radiology. Royal College In conclusion, this large scale review of radio- of Radiologists, London. graphic findings on the admission radiographs of 7. Royle H. (1952) X-ray appearances in asthma. British adult patients admitted to hospital with asthma Medical Journal 1, 577-580. has revealed an incidence of major abnormalities 8. Brooks L.J., Cloutier M.M. & Afshani E. (1982) Sign- ificance of roentgenographic abnormalities in children likely to affect treatment of 14.8% of patients. hospitalized for asthma. Chest 82(3), 315-318. Complications of asthma, such as segmental or 9. Rebuck A.S. (1970) Radiological aspects of severe greater atelectasis and important associated condi- asthma. Austral Radiology 14, 264-268. tions such as pneumonia and heart-failure occurred 10. Zieverink S.E., Harper P.A., Holden R.W., Klatte E.C. relatively frequently. Pneumothorax and pneumo- & Brittain H. (1982) Emergency room radiography of mediastinum, even in patients with symptoms asthma: An efficacy study. Radiology 145, 27-29. severe enough to cause their admission to hospital, 11. Sherman S., Skoney J.A. & Ravikrishnan K.P. (1989) occur rarely. Archives of Internal Medicine 149(11), 2493-2496. 12. White C.S., Cole R.P., Lubetsky H.W. & Austin J.H.M. ACKNOWLEDGEMENTS (1991) Acute asthma. Admission chest radiography in hospitalized adult patients. Chest 100, 14-16. The authors are grateful to Mrs Gill Dewey and her 13. Dalton A.M. (1991) A review of radiological abnor- staff in the Medical Audit Department and to the malities in 135 patients presenting with acute asthma. Medical Records Staff at the North Manchester Archives of Emergency Medicine 8, 36-40. General Hospital. We are also grateful to Mrs Olive 14. Heckerling P.S. (1986) The need for chest Roent http://emj.bmj.com/ Tracey for typing the manuscript. genograms in adults with acute respiratory illness. Clinical predictors. Archives of Internal Medicine REFERENCES 146(7), 1321-1324. 15. Swain D.G. (1984) Pneumothorax and acute asthma 1. Peterham I.S., Kerr I.H. & Collins J.V. (1981) Value of (letter). British Medical Journal 289, 109. chest radiographs in severe acute asthma. Clinical 16. Burke G. (1979) Pneumothorax complicating acute on October 1, 2021 by guest. Protected copyright. Radiology 32, 281-282. asthma. South African Medical Journal 55, 508-51 0.

237