Preliminary experience with endobronchial ultrasound-guided transbronchial needle aspiration for sampling of mediastinal lymph nodes

Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) has become established as the first-line procedure for nodal staging in lung cancer patients. We report our preliminary experience with this technique after its introduction in our Unit in October 2012. EBUS-TBNA was performed using an ultrasonic endoscope (Olympus BF-UC180F) and either 21- or 22-Gauge needles. Cytological samples were obtained from mediastinal lymph nodes enlarged at chest CT-scan and/or with an increased FDG uptake at PET-scan. Up to now we collected 244 nodal samples from 196 patients with suspected lung cancer. The final diagnosis was 174 lymph nodes positive for cancer, 48 benign lesions and 22 nodal samples without a definitive diagnosis and thus considered as lost in follow-up (LFU). EBUS-TBNA was positive for metastatic involvement in 144 and negative in 100 lymph nodes. The latter included 48 true negatives, 30 false negatives and 22 LFU. The sensitivity of EBUS-TBNA for lung cancer was 83% (144 out of 174). However, including among the false negatives the 22 LFU, sensitivity decreased to 73%. The overall diagnostic accuracy of EBUS-TBNA was 79% (144 samples positive for cancer plus 48 true negatives out of a total of 244 nodal samples). The most frequent nodal stations examined and the respective diagnostic accuracy were: station 7 (95 EBUS-TBNAs, 76%); station 4R (72 EBUS-TBNAs, 85%) and station 10R (33 EBUS-TBNAs, 79%). These results show that the diagnostic yield of EBUS-TBNA is high right away from the beginning of its use. Accordingly, it can be considered a reliable tool immediately after its introduction in an operative unit