The diagnostic yield and safety of ultrasound-assisted transthoracic biopsy of mediastinal masses

Background: Ultrasound (US)-assisted transthoracic biopsy offers a less invasive alternative to surgical biopsy in the setting of mediastinal masses.

Objectives: The aim of this 1-year prospective study was to assess the diagnostic yield and safety of a novel single-session sequential approach of US-assisted transthoracic fine-needle aspirations (TTFNA) with rapid on-site evaluation (ROSE) followed by cutting needle biopsies (CNB) performed by physicians on patients with anterosuperior mediastinal masses.

Methods: US-assisted TTFNA with ROSE was performed on 45 consecutive patients (49.5 ± 27.7 years, 24 males), immediately followed by CNB where a provisional diagnosis of epithelial carcinoma or tuberculosis could not be established, provided a safety range could be assured.

Results: TTFNA alone was deemed adequate by means of ROSE in 27 (60%) patients. CNB could be performed in 17 of the remaining 18. The on-site diagnosis corresponded to the final diagnosis in 26/45 (57.8%). An accurate cytological diagnosis was made in 33 (73.3%), and was more likely to be diagnostic in epithelial carcinoma and tuberculosis (28/30) than all other pathologies (5/15, p < 0.001). CNB yielded a diagnosis in 15/17 (88.2%). Overall, 42/45 patients were diagnosed by the single-session approach (93.3%). The final diagnoses included 41 neoplasms, with small cell lung cancer (n = 13) the commonest diagnosis. We observed no pneumothorax or major haemorrhage.

Conclusions: A single-session sequential approach of US-assisted TTFNA with ROSE followed by CNB, where indicated, has a high diagnostic yield for anterosuperior mediastinal masses, is safe and offers an alternative to surgical biopsy.