Document Type


Publication Date



Maine Medical Center, Medical Education, Maine Medical Center Research Institute, Critical Care

MeSH Headings

Bronchoscopy, Endoscopic Ultrasound-Guided Fine Needle Aspiration


Background: Endobronchial ultrasound (EBUS) guided transbronchial needle aspiration (TBNA) of mediastinal lymph nodes is the procedure of choice for the diagnosis and staging of lung cancer. EBUS TBNA has very high positive predictive value and obviates the need for more invasive procedures. In recent years, treatments that target specific oncogenic mutations have become the standard of care in advanced stage non-small cell lung cancer (NSCLC), especially in adenocarcinomas, and sample sufficiency for molecular testing has become more important. We evaluated sample sufficiency, and other related procedural variables over time during EBUS procedures at MMC.

Methods: Retrospective chart review of all patients undergoing EBUS TBNA at MMC from July 2013 through June 2017. All EBUS cases were identified from a bronchoscopy database, and procedure variables including number of nodes sampled, number of passes at each node, needle gauge, real time cytology, pre-procedure PET, type of airway used (endotracheal or laryngeal mask airway), specific proceduralist, and PGY of pulmonary fellow involved, were extracted. Procedures from the database were then matched to pathology results in EPIC, which included sample sufficiency for molecular testing. Trends in procedural variables were described, including molecular sufficiency data over three sequential 18-month periods.

Results: There were 313 EBUS procedures conducted, diagnosing 194 malignancies, of which 87 were adenocarcinoma. Of these adenocarcinomas, 43 samples were sufficient for molecular testing. Over sequential periods, overall volume of procedures increased, and sample sufficiency rose, from 48%, to 59%, and then to 69%. During the first two periods 21G needles were used in 90% of the procedures, while during the last period 19G needles were used in 55% of the procedures. During the second and third time periods there was one additional node sampled on average, and the average number of total passes per procedure rose from 6 to 9. Of note, over time there was also a trend toward the use of endotracheal airways and away from laryngeal mask airway. No major differences were noted in the other variables examined.

Conclusion: Yields for molecular testing in adenocarcinoma increased from 48 to 69 percent over the course of 48 months. During the same time period there were an increased number of nodes sampled, increased total passes per procedure, and transition to a larger gauge biopsy needle (22G to 19G). These procedural changes may explain some of the increase in yields, though we are unable to exclude other changes in practice which may have also affected yields. Future directions could include prospective collection of data including node size, correlation with real time cytology results, and whether other biopsy types were obtained during the procedure, all of which may also be predictive of specimen adequacy for molecular markers. Protocolization of EBUS procedures may also be helpful to further assess sample sufficiency for molecular testing.


2020 Costas T. Lambrew Research Retreat