OBJECTIVES: Our objective was to determine the contribution of endobronchial ultrasound in the diagnostic yields of acid-fast bacillus smear, nucleic acid amplification tests, and culture in bronchoalveolar lavage fluid for pulmonary tuberculosis. METHODS: During a 1-year interval, 99 patients who had initial sputum-negative acid-fast bacillus smears or no sputum but were later proven to have a positive culture for Mycobacterium tuberculosis in their sputum or bronchoalveolar lavage fluid were retrospectively studied. Among them, 56 patients underwent bronchoscopy with endobronchial ultrasound (EBUS group) and 43 patients received conventional bronchoscopy for bronchoalveolar lavage (non-EBUS group). RESULTS: The diagnostic yields of the nucleic acid amplification tests (89.3%, 50/56; P = .006), acid-fast bacillus smear (30.4%, 17/56; P = .013), and M tuberculosis culture in bronchoalveolar lavage fluid (67.9%, 38/56; P = .041) were significantly higher in the EBUS group of patients. The results of those who underwent conventional bronchoscopy were 65.1% (28/43), 9.3% (4/43), and 46.5% (20/43), respectively. Combining bronchoalveolar lavage fluid smear and nucleic acid amplification tests, we made a rapid diagnosis of pulmonary tuberculosis in 51 (91.1%) of the 56 EBUS patients and 29 (67.4%; P = .004) of the 43 non-EBUS patients. CONCLUSIONS: The introduction of endobronchial ultrasound increases the diagnostic yield of the nucleic acid amplification tests, acid-fast bacillus smear, and M tuberculosis culture from bronchioalveolar lavage fluid in patients with pulmonary tuberculosis who have negative sputum smear or no sputum production. Copyright 2010. Published by Mosby, Inc.
OBJECTIVES: Our objective was to determine the contribution of endobronchial ultrasound in the diagnostic yields of acid-fast bacillus smear, nucleic acid amplification tests, and culture in bronchoalveolar lavage fluid for pulmonary tuberculosis. METHODS: During a 1-year interval, 99 patients who had initial sputum-negative acid-fast bacillus smears or no sputum but were later proven to have a positive culture for Mycobacterium tuberculosis in their sputum or bronchoalveolar lavage fluid were retrospectively studied. Among them, 56 patients underwent bronchoscopy with endobronchial ultrasound (EBUS group) and 43 patients received conventional bronchoscopy for bronchoalveolar lavage (non-EBUS group). RESULTS: The diagnostic yields of the nucleic acid amplification tests (89.3%, 50/56; P = .006), acid-fast bacillus smear (30.4%, 17/56; P = .013), and M tuberculosis culture in bronchoalveolar lavage fluid (67.9%, 38/56; P = .041) were significantly higher in the EBUS group of patients. The results of those who underwent conventional bronchoscopy were 65.1% (28/43), 9.3% (4/43), and 46.5% (20/43), respectively. Combining bronchoalveolar lavage fluid smear and nucleic acid amplification tests, we made a rapid diagnosis of pulmonary tuberculosis in 51 (91.1%) of the 56 EBUSpatients and 29 (67.4%; P = .004) of the 43 non-EBUSpatients. CONCLUSIONS: The introduction of endobronchial ultrasound increases the diagnostic yield of the nucleic acid amplification tests, acid-fast bacillus smear, and M tuberculosis culture from bronchioalveolar lavage fluid in patients with pulmonary tuberculosis who have negative sputum smear or no sputum production. Copyright 2010. Published by Mosby, Inc.