Abstract

Objective: The aims of this study were to compare the estimated size of primary spontaneous pneumothorax (PSP) calculated on inspiratory and expiratory radiographs using the volumetrically derived Collins method and to determine whether radiograph type influences size classification for treatment according to published guidelines.

Method: This retrospective cohort study included patients treated for PSP in the emergency departments of two metropolitan teaching hospitals. Data collected included patient demographics and interpleural distances required to calculate pneumothorax size by the Collins method and to classify PSP according to guidelines. The outcomes of interest were the difference in size estimate between radiograph types and agreement in size classification for treatment according to guidelines. Analysis is by bias-plot analysis, kappa analysis and descriptive statistics.

Results: A total of 49 pneumothoraces (44 patients) were studied. Median age was 22 years; 66% were men. Median PSP size on inspiratory radiographs was 24% (IQR 14% to 31%, range 5% to 100%). The average size difference between expiratory and inspiratory films was 9%, with size on expiratory radiographs being larger. The 95% limits of agreement were wide (−5% to 23%). For each guideline, size estimation on expiratory rather than inspiratory radiographs would have suggested a change in treatment for an additional seven patients (14%, 95% CI 7% to 27%).

Conclusions: On average, PSP size calculated on expiratory radiographs is 9% higher than that calculated on matched inspiratory radiographs. Applying current management guidelines, the size difference between inspiratory and expiratory x-rays may alter initial treatment recommendation for some patients.