Abstract
BACKGROUND: The role of endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) validated with video-assisted mediastinoscopic lymphadenectomy (VAMLA) for mediastinal restaging of patients with non-small cell lung cancer (NSCLC) after induction therapy has never been described. OBJECTIVE: To report on our experience in this clinical setting. DESIGN: Retrospective analysis of a prospectively built database. METHODS: Patients with stage IIIA (N2) NSCLC who underwent EBUS-TBNA for mediastinal restaging after induction therapy were included. The sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and diagnostic accuracy of EBUS-TBNA and VAMLA for mediastinal restaging were calculated. The number of patients needed to undergo confirmatory VAMLA (NNT) after a negative EBUS-TBNA for mediastinal restaging to avoid a case of pathologic (p) N2 disease after resection was also calculated. RESULTS: Forty-six patients underwent EBUS-TBNA which was positive in 12 patients and negative in 34. Patients with a negative EBUS-TBNA underwent VAMLA which was positive in seven cases. Of the other 27 patients with a negative VAMLA, 26 underwent resection that did not show N2 disease. The sensitivity, specificity, NPV, PPV, and diagnostic accuracy of EBUS-TBNA for restaging were 63.1%, 100%, 79.4%, 100%, and 84.7%, respectively. The sensitivity, specificity, NPV, PPV, and diagnostic accuracy of confirmatory VAMLA after EBUS-TBNA was 100%. The NNT confirmatory VAMLA after a negative EBUS-TBNA to avoid a case of pN2 disease at resection was five patients. CONCLUSION: EBUS-TBNA must remain as the first-choice test for invasive mediastinal restaging. However, the results of our study in terms of sensitivity and NPV, even considering the small size of our population, suggest that negative results of EBUS-TBNA should be interpreted with caution and surgical exploration of the mediastinum (specially VAMLA, if available) should be considered in these patients.