Author's reply

EDITOR,—When the British Thoracic Society's research committee began the study in 1985 some doctors declined to take part because they thought that all pneumothoraces should be treated by simple aspiration. An equal number declined to take part because they thought that all pneumothoraces required intercostal tube drainage. It was because of these uncertainties that a well designed, randomly selected study was undertaken. Rather than “reinventing the wheel,”1 we were testing which of the two commonly used wheels was the right one to use.

Since our aim was to test current clinical practice, patients with signs of tension pneumothorax were specifically excluded from the trial. Most doctors do not now measure intrapleural pressures and would not diagnose tension pneumothorax simply on the basis of complete collapse. Whatever the numbers might indicate, the distribution of patients between the two randomly selected arms showed no greater variation than would be expected by chance, and the interpretation of our findings is not affected by the greater number of patients with complete collapse in the intercostal drainage group. Aspiration was successful in six of 10 patients with complete collapse, so that even in this group an initial attempt at aspiration would not be contraindicated. We do not agree, therefore, that routine thoracoscopic surgery is indicated as the initial treatment for all patients with spontaneous pneumothorax who present with complete collapse, as Anthony P C Yim proposes, although it may in future be more commonly used for recurrent episodes.

Whether intercostal drainage is successful depends on how success is defined. Two of the 38 patients treated by intercostal drainage required pleurectomy during the same admission and a further five within the next year because of recurrent pneumothorax, whereas none of the patients in whom aspiration either failed or was successful required pleurectomy during the subsequent year. The size of pneumothorax is not, therefore, an “alternative explanation for the higher rate of pleurectomy in the group treated by intercostal drainage.”2

We accept that minimal intervention thoracoscopic surgery is now rapidly replacing pleurectomy, but it was not in widespread use during the years of this study. We also agree that “small rim” pneumothoraces would not usually require any drainage procedure. Only four of the 73 patients in our study had small rim pneumothoraces, and they were judged by the doctors to merit a drainage procedure.

As we mentioned, the results of this study have been widely discussed with respiratory physicians and thoracic surgeons and form the basis of our guidelines.3 Given the successful outcome measures, a less painful and less traumatic procedure that requires a shorter stay in hospital seems to be the best initial management for all spontaneous pneumothoraces in patients with normal lungs.