Objective: The goal of this study was to determine the optimal number of aspirations per lymph node (LN) station during endobronchial ultrasound (EBUS)-guided transbronchial needle aspiration (TBNA) for maximum diagnostic yield in mediastinal staging of non-small cell lung cancer (NSCLC) in the absence of rapid on-site cytopathologic examination.
Methods: EBUS-TBNA was performed in potentially operable NSCLC patients with mediastinal LNs accessible by EBUS-TBNA (5 to 20 mm). Every target LN station was punctured four times.
Results: We performed EBUS-TBNA in 163 mediastinal LN stations in 102 NSCLC patients. EBUS-TBNA confirmed malignancy in 41 LN stations in 30 patients. Two malignant LN stations were missed in two patients. The sensitivity, specificity, positive predictive value, negative predictive value (NPV), and accuracy of EBUS-TBNA in predicting mediastinal metastasis were 93.8%, 100%, 100%, 96.9%, and 97.9%, respectively. Sample adequacy was 90.1% for one aspiration, and it reached 100% for three aspirations. The sensitivity for differentiating malignant from benign LN stations was 69.8%, 83.7%, 95.3%, and 95.3% for one, two, three, and four aspirations, respectively. The NPV was 86.5%, 92.2%, 97.6%, and 97.6% for one, two, three, and four aspirations, respectively. Maximum diagnostic values were achieved in three aspirations. When at least one tissue core was obtained by the first or second aspiration, the sensitivity and NPV of the first two aspirations were 91.9% and 96.0%, respectively.
Conclusions: Optimal results can be obtained in three aspirations per LN station in EBUS-TBNA for mediastinal staging of potentially operable NSCLC. When at least one tissue core specimen is obtained by the first or second aspiration, two aspirations per LN station can be acceptable.