Transbronchial needle aspiration improves the diagnostic yield of bronchoscopy in sarcoidosis

Rocco Trisolini, Luigi Lazzari Agli, Alessandra Cancellieri, Venerino Poletti, Piero Candoli, Daniela Paioli, Marco Alifano, Carmine Tinelli, Marco Patelli

Research output: Contribution to journalArticlepeer-review


Background and aim: Transbronchial needle aspiration (TBNA) is a minimally invasive bronchoscopic procedure that allows sampling of hilar and mediastinal lymph nodes in close contact with the airways. We undertook this study to assess the value of TBNA in the diagnosis of sarcoidosis manifesting with intrathoracic lymphadenopathies (stages I and II), and to compare its yield with that of transbronchial lung biopsy (TBLB). Methods: The results of bronchoscopy with combined TBNA and TBLB in 32 patients with stage I or II sarcoidosis were retrospectively analyzed. Results: Sensitivity was 65.6% for TBNA (stage I, 82.3 %; stage II, 46.6%), and 62.5% for TBLB (stage I, 52.9%; stage II, 73.3%). The combination of the two methods was associated with the highest diagnostic yield (93.7% overall sensitivity), and allowed significantly better results over both TBNA alone (93.7% vs 65.6%; p = 0.011) and TBLB alone (93.7% vs 62.5%; p = 0.005). Conclusions: The results of our study suggest that a diagnostic approach combining TBNA and TBLB is safe and effective in the setting of stage I and II sarcoidosis. It also confirmed the value of TBNA, with excellent diagnostic yields especially in stage I of the disease.

Original languageEnglish
Pages (from-to)147-151
Number of pages5
JournalSarcoidosis Vasculitis and Diffuse Lung Diseases
Issue number2
Publication statusPublished - Jun 2004


  • Bronchoscopy
  • Lymphadenopathy
  • Sarcoidosis
  • Transbronchial biopsy
  • Transbronchial needle aspiration

ASJC Scopus subject areas

  • Pulmonary and Respiratory Medicine


Dive into the research topics of 'Transbronchial needle aspiration improves the diagnostic yield of bronchoscopy in sarcoidosis'. Together they form a unique fingerprint.

Cite this