Transbronchial fine-needle aspiration. Reliability and limitations.

Transbronchial needle aspiration is a new method of sampling pulmonary, paratracheal, and mediastinal masses. A total of 116 aspirations of the lung performed during a two-and-a-half-year period were reviewed. The results were compared with those of bronchial wash, bronchial brush, sputum, transbronchial forceps biopsy, and histologic material subsequently obtained during surgery or at autopsy. Clinical information was also reviewed. Transbronchial needle aspiration was performed on 104 patients, yielding 116 specimens. With the use of a combination of aspiration, wash, brush, sputum, and forceps biopsy, malignancy was detected in 48 patients. Open biopsy, mediastinoscopy, and transcutaneous needle aspiration increased the number of malignant neoplasms detected to 72. The tumors were detected in 67% of these 72 cases by a combination of all bronchoscopic methods. Transbronchial aspiration correctly identified 36% of malignant cases compared with 35% identified by wash, 39% by brush, 25% by sputum, and 59% by forceps biopsy. The overall accuracy of the techniques was 56% for aspiration, 48% for wash, 56% for brush, 35% for sputum, and 71% for forceps biopsy. Malignant cases were associated with cellular transbronchial aspirates, good preservation of cell detail, and a high degree of cytologic/histologic correlation regarding cell type. Transbronchial fine-needle aspiration has the potential for improving the diagnostic ability of bronchoscopy. However, the high incidence of false negative diagnoses is the main limitation for the technique, and the responsible factors are discussed.