Background

Pneumothorax remains a common reason for medical admission to hospital with a UK annual incidence of 16.7 cases per 100 000 for men and 5.8 cases per 100 000 for women.1 This equates to around 8000 admissions a year for pneumothorax accounting for 50 000 bed days given an average length of stay of just under 1 week.2 Despite its frequency and the existence of established guidelines,3 4 there is still wide variation in pneumothorax management, which is likely to be driven by the paucity of good quality evidence in this field.

Pneumothorax has traditionally been divided into primary spontaneous pneumothorax (PSP), with no known underlying lung disease and secondary spontaneous pneumothorax (SSP), with existing lung disease. This is likely to be too simplistic an approach. Smoking cessation has been shown to significantly reduce the risk of recurrence in PSP and this together with the presence of decreased lung density/early emphysematous change seen on CT imaging in the lung apexes of current smokers presenting with PSP suggests many cases of PSP having underlying lung abnormalities.5 6 As current guidelines triage treatment according to whether they are primary or secondary pneumothorax, a better definition may lead to better algorithms and patient treatments.3

So why is the existing evidence so poor? Perhaps because it is so difficult to recruit to good quality randomised controlled trials (RCTs) in this area. The reasons for …