Diagnostic yield of transbronchial histology needle aspiration in patients with mediastinal lymph node enlargement.

摘要

BACKGROUND: Transbronchial needle aspiration (TBNA) is a safe, minimally invasive technique to assess the mediastinal spread of lung cancer. Excellent results have been published by experts. However, little information is available about the diagnostic yield of TBNA with the histology needle in a non-expert center. OBJECTIVES: The aim of this study is to assess the diagnostic yield of histology TBNA in the workup of suspected lung cancer. METHODS: In a non-university teaching hospital, TBNA data from patients diagnosed with lung cancer between June 1998 and July 2000 were analyzed retrospectively. TBNA had been performed by six different bronchoscopists in patients eligible for surgery with accessible N2 and N3 lymph nodes on computed tomography of the chest during the workup of an undefined mass. Cytology and histology specimens were obtained with the same 19-gauge needle. TBNA results were considered to be diagnostic if cytologic or histologic examination revealed a malignant lesion or non-malignant lymphoid cells. However, TBNA outcome was called non-diagnostic if no representative cells were obtained. RESULTS: From a group of 264 consecutive lung cancer patients, 106 (40%) patients were eligible for TBNA. In 79%, TBNA was diagnostic in cytology and/or histology specimens. Malignancy was demonstrated in 59% (63/106). In only 32/106 patients (30%), a histologic core of tissue could be sampled. In 87.5% of these patients (28/32), TBNA was diagnostic. For cytology only, this number was slightly lower (75%, 56/74). In 12 cases, diagnostic TBNA was verified by mediastinoscopy: these diagnoses were concordant. The sensitivity is 65% if all non-confirmed cases are considered false negative. Ten mediastinoscopies were avoided because TBNA demonstrated contralateral N2 (= N3) disease. The routine use of TBNA during bronchoscopy in suspected N2 disease is a cost-effective procedure, as the total additional costs of TBNA (9,540 EUR) were lower than the costs of 10 avoided mediastinoscopies (15,500 EUR). No complications were observed. CONCLUSION: The diagnostic yield of TBNA relied mainly on cytology specimens, despite the use of a histology needle. Representative histology specimens could only be obtained in 28/106 patients (26%). Since TBNA was performed in a general hospital by different bronchoscopists, this procedure is useful in the workup of lung cancer patients with enlarged lymph nodes.