Objective: Transthoracic sonography (TS) has evolved as an important imaging technique for diagnosing pleural and pulmonary conditions. However, the value of TS in either excluding or diagnosing pneumothorax is still under debate. This study was conducted to examine whether TS could replace chest radiography for the diagnosis of post-interventional pneumothorax and hydropneumothorax.
Methods: 53 patients (21 females, 32 males; median age 64 years, range 37-94 years), 35 of whom underwent transbronchial biopsy (TBB) and 18 patients who had an ultrasound-guided chest tube placement (U-GCTP) were enrolled in the study. TS was performed three hours after either TBB or removal of a chest tube, followed by postero-anterior chest radiograph (CRX). If any discrepancy between TS, the clinical presentation and the CRX became apparent, either a lateral CRX or a computed tomography (CT) of the thorax was performed. TS was assessed according to the presence of the following criteria: (1) "gliding sign" of the pleural line, (2) comet tail artifacts, (3) reverberation artifacts, (4) air/fluid mirror, (5) hyperechoic reflectors within the pleural effusion and (6) "lung point".
Results: In four out of the 53 patients (7.5%) a post-interventional pneumothorax or hydropneumothorax occurred. One out of the 35 patients (2.9%) developed a pneumothorax after TBB, requiring chest tube placement. Three patients (16.7%) developed a hydropneumothorax due to U-GCTP which was detected by sonography but was missed by postero-anterior CRX in one patient. The sensitivity, specificity and accuracy of TS were 100% in excluding post-interventional pneumothorax/hydropneumothorax.
Conclusion: TS is a cost-effective and safe bed-side-method, allowing for an immediate exclusion or diagnosis of post-interventional pneumothorax/hydropneumothorax in patients who have undergone TBB or U-GCTP. Thus, these preliminary results suggest that CXR may only be required in patients with pneumothorax diagnosed by TS in order to assess its extension or to exclude any discrepancy between the TS-result and the clinical presentation.