Primary spontaneous pneumothorax (PSP) occurs in apparently healthy young people with an incidence of 12.5 cases per 100 000 per year [1]. Attempts to develop standardised care guidelines for this condition have been severely hampered by a lack of high-quality clinical research into this condition. The American College of Chest Physicians (CHEST) concluded in 2001 that “insufficient data exist…to develop an evidence-based document” and so produced a consensus statement based on expert opinion [2]. Similarly, the British Thoracic Society (BTS) 2010 guidelines are based predominantly on nonanalytical studies and expert opinion [3]. In both documents, the size of the presenting PSP is used to determine initial treatment. A “small” PSP without respiratory compromise is thought not to require intervention, while a “large” PSP has typically been treated either by aspiration or intrapleural drainage. Implicit in these definitions is the belief that large pneumothoraces will not respond well to conservative management. Remarkably, no consensus regarding the definition of PSP severity exists, with CHEST and the BTS each using different arbitrary measurements of the presentation chest radiograph. When these measurements were compared directly to one another, they showed poor correlation [4]. This lack of a clinically useful radiological biomarker for pneumothoraces requiring intervention hinders the development of evidence-based care of this condition. We wished to determine whether the BTS definition of large pneumothorax (>2 cm at the hilum) or CHEST definition (>3 cm from apex to cupola) better predicts the requirement for intercostal chest drain (ICD) insertion.