Endobronchial ultrasound-guided transbronchial needle aspiration in routine care - plenty of benign results and follow-up tests.

BACKGROUND: Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is a widespread technique for tissue sampling from hilar and mediastinal lymph nodes (LN). The diagnostic yield of this method is reported to be very high even outside clinical trials. We aimed to assess the diagnostic accuracy of EBUS-TBNA after its implementation at a University hospital.
METHODS: We analysed the first 100 consecutive patients who underwent an EBUS-TBNA procedure at our institution with respect to indication and cytological results. Sensitivity and negative predictive value (NPV) were calculated on the basis of histological confirmation or follow-up.
RESULTS: From 03/2007 to 03/2008, EBUS-TBNA of 218 LNs was performed on the basis of chest computed tomography. The primary indication for EBUS-TBNA was lymphadenopathy of unknown cause with (44%) or without (37%) pulmonary nodule(s). Only 19% of patients had known cancer and underwent the procedure for (re-)staging. In 73% of patients a non-diagnostic cytology was reported. A diagnosis could be established in only 27% by EBUS-TBNA including four patients with sarcoidosis. Sensitivity and NPV were low with 61.4% and 76.7%, respectively. Diagnostic yield increased over time and was better in cancer patients than in patients with incidental lymphadenopathy.
CONCLUSION: Although EBUS-TBNA is reported to have a very high diagnostic yield in selected patients, the predominant finding in routine care, depending on the patient population, can be a non-diagnostic cytology result with the need for surgical procedures or follow-up studies. This should be considered in the approach to patients with mediastinal or hilar lymphadenopathy.