Performance characteristics and predictors of yield from transbronchial needle aspiration in the diagnosis of peripheral pulmonary lesions

Background and objective: Transbronchial needle aspiration (TBNA) is useful for diagnosing peripheral pulmonary lesions (PPL). However, TBNA is largely underused and the variables that may be related to its diagnostic usefulness have not been specifically studied. The aim of the present study was to evaluate the performance characteristics and predictors of yield from TBNA of PPL, and to compare the performance characteristics of different bronchoscopic sampling methods. Methods: Consecutive patients with PPL were prospectively enrolled, and during the same examination, TBNA, transbronchial lung biopsy (TBLB) and bronchial washing (BW) were performed. Results: Two hundred and eighteen PPL in 218 patients were sampled. TBNA was more sensitive (65%) than either TBLB (45%, P <0.001) or BW (22%, P <0.001). TBNA was the only diagnostic procedure in 42/196 patients (21%) with malignant lesions, and was more likely to be the only diagnostic procedure for lesions lacking (23/85 patients, 27%) than for lesions with the bronchus sign (19/111 patients, 17%). In multivariate analysis, a lesion size >2 cm, malignancy and location in the middle lobe were independent predictors of a positive TBNA result. Conclusions: TBNA is the single best contributor to the success of bronchoscopy in the diagnosis of PPLs, and should be routinely used especially in the presence of lesions lacking the bronchus sign. Lesion size of > 2 cm, location in the middle lobe, and malignant nature are strong predictors of a positive TBNA result. In a large series of patients with peripheral pulmonary lesions, transbronchial needle aspiration was the single best method associated with successful diagnosis. Preference should be given to this method, especially for lesions lacking the bronchus sign on CT.