Endobronchial ultrasound-guided transbronchial needle aspiration in the diagnosis of non-lymph node thoracic lesions

AIMS: Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) has shown excellent diagnostic capabilities for mediastinal and hilar lymphadenopathy. However, its value in thoracic non-lymph node lesions is less clear. This study was designed to assess the value of EBUS-TBNA in distinguishing malignant from benign thoracic non-lymph node lesions. METHODS: From October 2009 to August 2011, 552 patients underwent EBUS-TBNA under local anesthesia and with conscious sedation. We retrospectively reviewed 81 of these patients who had tracheobronchial wall-adjacent intrapulmonary or isolated mediastinal non-lymph node lesions. On-site cytological evaluation was not used. Immunohistochemistry (IHC) was performed to distinguish the origin or type of malignancy when necessary. RESULTS: EBUS-TBNA was performed in 68 tracheobronchial wall-adjacent intrapulmonary and 13 isolated mediastinal non-lymph node lesions. Of the 81 patients, 77 (95.1%, 60 malignancies and 17 benignancies) were diagnosed through EBUS-TBNA, including 57 primary lung cancers, 2 mediastinal tumors, 1 pulmonary metastatic adenocarcinoma, 7 inflammation, 5 tuberculosis, 3 mediastinal cysts, 1 esophageal schwannoma, and 1 focal fibrosis. There were four false-negative cases (4.9%). Of the 60 malignancies, there were 9 (15.0%) which originally had no definite histologic origin or type. Thus, IHC was performed, with 7 (77.8%) being subsequently confirmed. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of EBUS-TBNA in distinguishing malignant from benign lesions were 93.4% (60/64), 100% (17/17), 100% (60/60), 81.0% (17/21), and 95.1% (77/81), respectively. CONCLUSION: EBUS-TBNA is a safe procedure with a high sensitivity for distinguishing malignant from benign thoracic non-lymph node lesions within the reach of EBUS-TBNA, with IHC usually providing a more definitive diagnosis.