Endosonography-guided fine-needle aspiration in the diagnosis of sarcoidosis: a randomized study

INTRODUCTION There are no widely accepted standards for the diagnosis of sarcoidosis. OBJECTIVES The aim of this study was to assess the relative diagnostic yield of endobronchial ultrasound fine-needle aspiration (EBUS-FNA) and endoscopic ultrasound fine-needle aspiration (EUS-FNA), and to compare them with standard diagnostic techniques such as endobronchial biopsy (EBB), transbronchial lung biopsy (TBLB), transbronchial needle aspiration (TBNA), and mediastinoscopy. PATIENTS AND METHODS This was a prospective randomized study including consecutive patients with clinical diagnosis of stage I or II sarcoidosis. EBB, TBLB, and TBNA were performed at baseline in all patients. Subsequently, patients were randomized to group A (EBUS-FNA) or group B (EUS-FNA). Next, a crossover control test was performed: all patients with negative results in group A underwent EUS-FNA and all patients with negative results in group B underwent EBUS-FNA. If sarcoidosis was not confirmed, mediastinoscopy was performed. RESULTS We enrolled 106 patients, of whom 100 were available for the final analysis. The overall sensitivity and accuracy of standard endoscopic methods were 64% each. When analyzing each of the standard endoscopic methods separately, the diagnosis was confirmed with EBB in 12 patients (12%), with TBLB in 42 patients (42%), and with TBNA in 44 patients (44%). The sensitivity and accuracy of each endosonographic technique were significantly higher than those of EBB+TBLB+TBNA (P = 0.0112 vs P = 0.0134). CONCLUSIONS The sensitivity and accuracy of EBUS-FNA and EUS-FNA are significantly higher than those of standard endoscopic methods. Moreover, the sensitivity and accuracy of EUS-FNA tend to be higher than those of EBUS-FNA.