Background: The diagnostic accuracy of endobronchial ultrasound-guided transbronchial needle aspiration for the diagnosis of isolated mediastinal lymphadenopathy in patients with previous malignancy is not well defined. Positron emission tomography scanning has been proven to be a significant advance, but false-positive results are common. The purpose of this prospective and controlled study was to assess the yield of endobronchial ultrasound-guided transbronchial needle aspiration to reveal mediastinal lymph node metastases in patients with previous malignancy and possible mediastinal involvement on computed tomography and positron emission tomography.
Methods: Seventy-three lymph nodes were tested by transbronchial needle aspiration on 48 consecutive patients, each patient underwent to mediastinoscopy or thoracoscopy immediately after needle aspiration for histological confirmation. A cytological sampling adequate for diagnosis was obtained in 45 patients (93.7%); the three cases of inadequate sampling resulted as negative for cancer.
Results: The endobronchial ultrasound guided transbronchial needle aspiration gives a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 97.4%, 100%, 100%, 87.5%, and 97.7% respectively. Disease prevalence was 84.4%. All the endoscopic procedures were well tolerated and no immediate complications were recorded.
Conclusions: Transbronchial needle aspiration under endobronchial ultrasound guidance is a valuable technique for cytological diagnosis of isolated mediastinal lymphadenopathy in patients with history of malignancy. Tissue sampling by invasive surgical procedures (mediastinoscopy or thoracoscopy) remains mandatory in case of inadequate or negative transbronchial needle aspiration cytology.