Methods: Two hundred twenty-eight patients (165 males, 63 females; mean age 58.1 years; range 22 to 84 years) who were performed EBUS-TBNA between April 2011 and January 2013 were retrospectively analyzed. Midazolam sedation was applied in all patients. The diagnostic verification of the samples collected from lymph nodes was carried out by surgical intervention or clinical and radiological follow-up.
Results: In a total of 228 EBUS-TBNA procedures, 1,447 aspirations (average 6.47 for each patient, range 1-12) were applied to 635 lymph nodes (average 2.78 for each patient, range 1-6). In 98.6% of the patients (225/228), the samples were histopathologically appropriate. The histopathological examination identified 88 reactive hyperplasias (39%), 82 neoplastic diseases (36%), 39 granulomatous diseases (17%) (tuberculosis compatible in four patients), and five patients with suspicion of malignant diseases (2%). The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy rates were 91.7%, 100%, 100%, 88.9%, and 95%, respectively. No mortality was observed associated with EBUS-TBNA procedure.
Conclusion: This study suggests that EBUS-TBNA is a reliable method with high diagnostic value in the evaluation of the mediastinal and hilar lymph nodes.