Fiberoptic bronchoscopy in the diagnosis of pulmonary tuberculosis.

We reviewed the records of 80 patients with culture-positive pulmonary tuberculosis. Forty (50%) had their diagnoses established by sputum smears and cultures alone, 20 (25%) by brush/wash specimens from fiberoptic bronchoscopy alone, 18 (22.5%) from both sources, and two (2.5%) by gastric smears. The average age of the patients was 71.3 years, and only 20% had symptoms typical of pulmonary tuberculosis. Our data reveal that findings in smears from fiberoptic bronchoscopy were of comparable sensitivity (80%) to those of sputum (72.5%); in 25/80 (31.25%), diagnosis was made exclusively by fiberoptic bronchoscopy. Such a high sensitivity of fiberoptic bronchoscopy in the diagnosis of pulmonary tuberculosis indicates a threefold rise (P = .018) in the number of diagnoses made in this manner compared to our figures from 1983 (10% of diagnoses made by fiberoptic bronchoscopy). Forty-three of the 80 patients (53.8%) had either a negative sputum smear or no sputum available. Thirty of the 43 patients (69.8%) had diagnostic bronchoscopy, which provided an immediate diagnosis (smear positivity) in 18 patients (60%). Transbronchial biopsy was most useful in excluding associated malignancy. Fiberoptic bronchoscopy is playing an increasingly significant role in the diagnosis of pulmonary tuberculosis. Further studies are essential to evaluate cost effectiveness, specificity of fiberoptic bronchoscopy, and the influence of the procedure on morbidity and mortality in pulmonary tuberculosis.