Abstract
Background: Although endobronchial ultrasound (EBUS)-guided transbronchial biopsy (TBB) has been shown to increase the diagnostic yield over conventional bronchoscopic techniques, an important issue regarding the optimal number of biopsy specimens required has not been thoroughly investigated. Objectives: We sought to examine whether the number of biopsy specimens taken was associated with the diagnostic yield of EBUS-guided TBB and, if this was the case, to determine the optimal number of specimens required for the maximum diagnostic yield in peripheral pulmonary lesions. Methods: The medical records of patients undergoing EBUS-guided TBB for the diagnosis of peripheral pulmonary lesions from 2008 to 2010 were retrospectively reviewed. The association of clinical and radiological features, including the number of biopsy specimens, with the diagnostic yield was analysed. Results: A total of 384 patients were included for analysis. The overall diagnostic yield of EBUS-guided TBB was 73%, and the only factor influencing the diagnostic yield was the position of the probe. Patients in which the EBUS probe was placed within the lesions had a significantly higher yield (85%) than those in which the probe was adjacent to or outside the lesions (38%; p < 0.001). When the number of biopsy specimens was determined based on their adequacy, it was an insignificant factor in predicting the diagnostic yield. Conclusions: Probe position independently predicts the diagnostic yield of EBUS-guided TBB. In real-world practice, the optimal number of biopsy specimens should be decided on a case-by-case basis.