A chest X-ray (CXR) is recommended after bronchoscopies with an increased risk of pneumothorax (PTX). However, concerns regarding radiation exposure, expenses and staff requirements exist. A lung ultrasound (LUS) is a promising alternative for the detection of PTX, though data are scarce. This study aims to investigate the diagnostic yield of LUS compared to CXR, to exclude PTX after bronchoscopies with increased risk. This retrospective single-centre study included transbronchial forceps biopsies, transbronchial lung cryobiopsies and endobronchial valve treatments. Post-interventional PTX screening consisted of immediate LUS and CXR within two hours. In total, 271 patients were included. Early PTX incidence was 3.3%. Sensitivity, specificity, and the positive and negative predictive values of LUS were 67.7% (95% CI 29.93-92.51%), 99.2% (95% CI 97.27-99.91%), 75.0% (95% CI 41.16-92.79%) and 98.9% (95% CI 97.18-99.54%), respectively. PTX detection by LUS enabled the immediate placement of two pleural drains along with the bronchoscopy. With CXR, three false-positives and one false-negative were observed; the latter evolved into a tension-PTX. LUS correctly diagnosed these cases. Despite low sensitivity, LUS enables early diagnosis of PTX, thus preventing treatment delays. We recommend immediate LUS, in addition to LUS or CXR after two to four hours and monitoring for signs and symptoms. Prospective studies with higher sample sizes are needed.
Keywords: chest X-ray; diagnostic yield; interventional bronchoscopy; lung ultrasound; pneumothorax.