Introduction Mediastinal staging in lung cancer is a core function of EBUS-TBNA. There has been an explosion of EBUS -TBNA services across the UK over recent years. However, quality standards and adherence to such standards are not widely known. The aim of this study was to describe the current practices of four independent EBUS centres serving a large UK cancer network.

Materials and Methods In 2012, the number of centres providing EBUS-TBNA in this Network increased from one to four. This prompted the development of an EBUS sub-group and service specification that mandates the collection of pre-defined data for all EBUS procedures. Analysis of this prospectively maintained database was undertaken for this report.

Results 741 lung cancer patients underwent EBUS-TBNA in the study period. 56.4% (418/741) were for nodal staging, with the remaining performed for pathological confirmation of lung cancer. The proportion of staging procedures performed at each centre varied significantly (range 4.8% - 80.3%, p < 0.0001). In those patients undergoing EBUS for mediastinal staging, the average number of lymph stations sampled per procedure varied from 1.3 to 1.9 across the four centres and the proportion of patients with 3 or more stations sampled varied from 0% to 26.6%. Furthermore, there was variability in the rate of inadequate sampling (2.4–7.6%), sensitivity (69.2–90.5%), negative predictive value (50–87.5%) and rate of surgical sampling of negative nodes (6.3–35.0%) across the EBUS centres.

Discussion There is variability in practice in all parameters of EBUS practice examined in our Network, in the absence of agreed standards and protocols for mediastinal staging. Such protocols and standards have now been agreed and implemented across our Network and the EBUS sub-group is committed to ongoing data collection and publication to drive quality outcomes in this pivotal lung cancer service.