True Negative Predictive Value of Preoperative Staging Endosonography by Combined Endobronchial Ultrasound and Endoscopic Ultrasound in Non-Small Cell Lung Cancer

Abstract

SESSION TITLE: Interventional Pulmonary Procedures SESSION TYPE: Original Investigation Slide PRESENTED ON: Wednesday, November 1, 2017 at 02:45 PM - 04:15 PM PURPOSE: Complete endosonographic mediastinal staging modalities have become a new standard in the pre-operative staging of non-small cell lung cancer in some centers. This study was performed to understand the value of mediastinal lymph node (LN) sampling and dissection at time of pulmonary resection in the setting of a negative pre-operative staging endosonography by combined EBUS & EUS. METHODS: This study consisted of a retrospective, single-institution, tertiary care referral center review of a prospectively maintained database between January 2009 and December 2014. Two hundred and thirty-two consecutive patients with non-small cell lung cancer with negative results on lymph node staging by combined endobronchial ultrasound (EBUS) and endoscopic ultrasound (EUS) fine needle aspiration (FNA) that subsequently underwent pulmonary resection or attempt at pulmonary resection were included. The endosonographic results were compared with the final pathological staging utilizing final pathology results of lymph nodes sampled during surgery. Chi-Square test was used for statistical analysis. A p-value of <0.05 was considered statistically significant. RESULTS: Twenty-five out of 232 patients with negative results on LN staging by combined EBUS / EUS - FNA had metastases in LNs sampled during pulmonary resection. 10 patients had N1 disease and 15 patients had N2 disease. The negative predictive value (NPV) of EBUS/EUS was 89.2% (p<0.01). Twenty-one out of these 25 patients who had surgical LN sampling at resection did not have the same station sampled pre-operatively by EBUS/ EUS. Therefore a direct comparison at surgical resection could not be made on these 21 patients. Eleven out of these 21 patients had Station 5 and Station 6 involvement and these stations are routinely difficult to sample using endoscopic modalities. If these 21 patients are omitted for comparison, only 4 out of 232 patients had false negative results by EBUS / EUS - FNA giving a true negative predictive value of 98.3% (p<0.01). 3 out of these 4 patients had N2 disease (Station 4L, 5 and 7) and 1 patient had N1 (Station 11R) disease detected at Pulmonary resection. CONCLUSIONS: Staging endosonography by combined EBUS & EUS in potentially resectable patients is associated with a high NPV. If pre-operative results of biopsy of a particular LN station by EBUS/EUS-FNA are negative, then sampling of the same station at surgery can potentially be omitted. LN stations not sampled by endosonography should be sampled at surgery. CLINICAL IMPLICATIONS: Can potentially change the practice and the way lymph node station sampling and dissection is done during pulmonary resection. This may reduce the operative time and avoid unnecessary dissection of the lymph nodes already biopsied at pre operative staging procedure. DISCLOSURE: The following authors have nothing to disclose: Pravachan Hegde, Vicky Thiffault, Vipul Jain, Akshatha Gowda, Pasquale Ferraro, Moishe Liberman No Product/Research Disclosure Information