RESUMO
BACKGROUND AND AIM: The aim of this retrospective study was to assess endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA), a minimally invasive technique, to determine its diagnostic accuracy and morbidity in the etiological diagnoses of mediastinal lymph nodes. METHODS: A total of 84 patients underwent EUS-FNA biopsy of the mediastinal lymph nodes, and were classified as either malignant disease, inflammatory disease or incidental diagnosis, according to the suspected clinical condition. To evaluate the diagnostic accuracy of EUS-FNA in each group, a comparison of the cytological results obtained with 19- and 22-gauge needles was performed. RESULTS: All 84 procedures were carried out between January 2004 and June 2008. Six patients were excluded because of non-contributory results. On analyzing the results of EUS-FNA in the malignant group (n=41), the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were 93, 100, 100 and 85%, respectively. In the inflammatory (n=20) and incidental diagnosis (n=17) groups, the sensitivity, specificity, PPV and NPV were 85, 100, 100 and 75%, and 80, 100, 100 and 77%, respectively. There was no significant difference between the FNA results with the 19-gauge (20/78 patients) versus 22-gauge (58/78 patients) needles (P=0.55). Also, no complications were reported either during the procedure or after the 6-month follow-up. CONCLUSION: EUS-FNA is a safe and accurate diagnostic procedure for the study of mediastinal lymph nodes not only for malignancy, but also for inflammatory diseases and incidental diagnoses. In addition, there is no difference in diagnostic accuracy whether FNA is carried out with a 19-gauge or 22-gauge needle.