Sampling Utility of the Convex Probe Endobronchial Ultrasound Visible Intrapulmonary Lesion.

Abstract

The value of endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) in the mediastinal staging of lung cancer has been well established. However, data regarding its utility in the diagnosis of intrapulmonary lesions has been sparse. This study assesses the sampling utility of convex probe EBUS-visible intrapulmonary lesions not visualized by the white-light bronchoscopy. A retrospective analysis of EBUS-TBNA of EBUS-visible intrapulmonary lesions was performed between January 2010 and March 2015. Patients with visible endobronchial lesions by white-light bronchoscopy were excluded from analysis. Among 108 procedures, the diagnostic yield of EBUS-TBNA for EBUS-visible intrapulmonary lesions was 87%. Following diagnoses were established: lung cancer (73/67.6%), lung metastases (10/9.2%), infection (5/4.6%), lymphoma (1/<1%), sarcoma/spindle cell sarcoma or neoplasm (3/2.8%), unspecified malignancy (1<1%), and hamartoma (1/<1%). EBUS-TBNA was nondiagnostic in 14 (13%); among these, 9 turned out to have benign disease based on additional bronchoscopy samples or other testing and/or follow-up imaging. Five were ultimately diagnosed with a malignant condition: lymphoma (1), epithelioid hemangioendothelioma (1), and non-small cell lung cancer (3). The sensitivity and the negative predicted value of EBUS-TBNA for differentiating malignancy from benign disease was 94.7% and 75%, respectively, while the accuracy for diagnosing the neoplastic disease was 95.3%. There was one major bleeding requiring bronchial artery embolization and 1 pneumothorax requiring chest tube drainage. EBUS-TBNA is safe and effective in the diagnosis of EBUS-visible intrapulmonary lesions. It should be considered as the diagnostic test of choice in patients with these lesions undergoing EBUS-TBNA for the staging of suspected lung cancer.