The aim of this study was to compare classification of pneumothoraces into size groups for treatment using the British Thoracic Society [BTS], American College of Chest Physicians [ACCP] and Belgian Society of Pulmonology [BSP] guidelines and the range of pneumothorax sizes in each group calculated using the volumetrically-derived Collins' method.

This was a retrospective cohort study. Participants were patients with primary spontaneous pneumothorax [PSP] attending emergency departments of two teaching hospitals between 1996 and 2005. Participants were identified from a pre-existing database. Data collected included demographics, side of PSP and interpleural distances for size classification based on BTS, ACCP, BSP and Collins' method requirements measured on inspiratory X-rays. The outcome of interest was comparison of classification into size groups according to each guideline. Analysis was by descriptive statistics, overall agreement and Kappa analysis for agreement between guideline pairs.

Forty-nine episodes [44 patients] were studied. Median age was 22 years; 66% of patients were male. Median PSP size [Collins' method; inspiratory films] was 24%; range 5–100%. Based on inspiratory films, the BTS guideline classified 10% of PSP as large compared with 47% by the BSP guideline and 49% by the ACCP guideline. The three guidelines agreed in their classification in only 47% of cases.

Size classification of PSP based on available treatment guideline definitions shows poor agreement. This goes some way to explain management variation between regions and limits comparability of reported outcomes. There is a strong case for international agreement in size classification/estimation in order to facilitate high-quality studies into optimal management approaches.