Endobronchial ultrasound-guided transbronchial needle aspiration for differentiating N0 versus N1 lung cancer.

BACKGROUND The aim of this study was to assess the value of endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) for differentiating cN0 versus cN1 non-small cell lung cancer. METHODS A retrospective review of EBUS-TBNA results in patients with potentially resectable clinical N0 or N1 non-small cell lung cancer based on computed tomography and positron emission tomography was performed. Systematic mediastinal and hilar lymph node sampling was performed by EBUS-TBNA. Lymph nodes larger than 5 mm in short axis or suspicious nodes were targeted. In the absence of N2 or N3 disease, patients underwent resection with lymph node dissection. RESULTS A total of 981 patients underwent EBUS-TBNA during the study period, of which 163 patients met the study criteria. There were 94 cN0 and 69 cN1 patients. A total of 453 lymph nodes (338 mediastinal and 115 N1 lymph nodes, average 2.8 nodes/patient) were sampled. Endobronchial ultrasound upstaged 9 (5.5%) patients to N2 disease, but was falsely negative in the mediastinum in 7 (4.3%) patients. In cN0 patients, EBUS confirmed N0 in 87 (53.4%) and upstaged in 7 (4.3%, N1 in 1, N2 in 6). In cN1 patients, EBUS confirmed N1 in 19 (11.7%), downstaged in 47 (28.8%), and upstaged in 3 (1.8%). The sensitivity, specificity, diagnostic accuracy, and negative predictive value of EBUS-TBNA to accurately differentiate between N0 and N1 disease was 76.2%, 100%, 96.6%, and 96.2%, respectively. The accuracy of mediastinal staging was 95.7%. CONCLUSIONS Endobronchial ultrasound-guided transbronchial needle aspiration can accurately access the hilar and interlobar lymph nodes in patients with potentially resectable lung cancer. Accurate assessment of cN0 versus cN1 by EBUS-TBNA may be used to guide induction therapy before surgery.