Intended for healthcare professionals


Simple aspiration versus intercostal tube drainage for spontaneous pneumothorax in patients with normal lungs

BMJ 1994; 309 doi: (Published 19 November 1994) Cite this as: BMJ 1994;309:1338
  1. J Harvey,
  2. R J Prescott
  1. Lung Function Unit, Southmead Hospital, Bristol BS10 5NB
  2. Department of Public Health Sciences, Edinburgh University Medical School, Edinburgh EH8 9AG.
  1. Correspondence to: Dr Harvey.
  • Accepted 19 August 1994

Two clinical trials in the 1960s led to opposing recommendations for initial management of spontaneous pneumothorax,1,2 and the debate has continued since. Recently, simple aspiration has gained favour as a more acceptable procedure for patients and doctors.3,4 We conducted a randomised comparison of simple aspiration with intercostal drainage to assess acceptability and outcome at one year.

Subjects, methods, and results

Patients who presented with spontaneous pneumothorax and whom the admitting team thought required a drainage procedure were randomly allocated to either simple aspiration or intercostal drainage. Patients with signs of a tension pneumothorax or with lung disease other than previous pneumothorax were excluded.

Simple aspiration was undertaken by inserting a 16-18 gauge catheter under local anaesthetic and aspirating air through a three way tap with the exit tube under water. The procedure was continued until no more air could be aspirated, the patient became uncomfortable, or a maximum of three litres had been removed. Intercostal drainage was managed according to the participating physician's usual practice. No sclerosing drugs were allowed with either technique.

Patients completed symptom score charts indicating the degree of pain experienced during drainage and throughout their hospital admission. Patients were reviewed at one and 12 months, and recurrence of pneumothorax or referral for thoracic surgery was recorded. The size of pneumothorax in presenting radiographs was graded (see table).

Characteristics and clinical details of patients having aspiration or intercostal drainage for pneumothorax. Values are means (SD) unless stated otherwise

View this table:

Both groups were similar with respect to age, sex, height, weight, smoking history, lung function, and history of pneumothorax (table). There were no significant differences between the two groups in either the size or side of pneumothorax. No difference was reported in pain experienced while undergoing drainage, but those treated by intercostal drainage experienced significantly more pain during their hospital admission and spent an average of two days longer in hospital (table).

In all 28 out of 35 aspirations were successful, although five patients required two aspirations. The remaining seven patients were subsequently treated by intercostal drainage. None of them had a recurrence or required pleurectomy within one year. The amount of air aspirated during the first aspiration was significantly different in successful and unsuccessful aspirations (successful 1.59 (SD 0.72) v unsuccessful 2.52 (0.91) 1; P<0.01). A logistic regression model found no significant associations with failed aspiration. In particular, there was no association with a history of pneumothorax or initial radiographic appearances. No significant differences were found in the recurrence rate at one year, but more patients who had intercostal drainage required pleurectomy than those patients who had had aspiration (P=0.02).


Simple aspiration is a simple and safe procedure and should be the initial treatment of choice for patients with normal lungs who present with a spontaneous pneumothorax, irrespective of its size. This study has shown that aspiration is less painful than intercostal drainage, leads to a shorter admission, and reduces the need for pleurectomy with no increase in recurrence rate at one year.

As a result of this study, and after consultation with over 150 British respiratory physicians and thoracic surgeons, the standards of care committee of the British Thoracic Society has published guidelines covering the management of spontaneous pneumothorax.5

The British Thoracic Society's research committee would like to thank the following respiratory physicians for taking part in this study. I A Campbell, Llandough Hospital, Penarth; II Coutts, Treliske Hospital, Truro; A D Ferguson, Royal Devon and Exeter Hospital, Exeter; H R Gribbin, Middlesbrough General Hospital, Middlesbrough; B D W Harrison, West Norwich Hospital, Norwich; R N Harrison, North Tees General Hospital, Stockton on Tees; J E Harvey, Southmead Hospital, Bristol, (coordinator); C M B Higgs, Royal United Hospital, Bath; A W Matthews, Queen Alexandra Hospital, Portsmouth; C R McGavin, Derriford Hospital, Plymouth; C P Mustchin, Cumberland Infirmary, Carlisle; E Neville, St Mary's Hospital, Portsmouth; M R Partridge, Whipps Cross Hospital, London; J A Siddorn, Royal Devon and Exeter Hospital, Exeter; J A M Turner, Royal Victoria Hospital, Bournemouth; R J White, Frenchay Hospital, Bristol. Drs Ferguson, Harvey, and McGavin were members of the pneumothorax subcommittee.

We thank Mrs Elsie Miller for help during the study, Miss Vivian Wilson for preparing the results and typing the manuscripts, and Jian-Hua Mao for statistical analysis.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.