Transtracheal puncture enables two samples of bronchial secretions to be taken-the product of transtracheal aspiration and simultaneously expectorated sputum (obtained in 71% of the cases)-for the purpose of testing for Mycobacterium tuberculosis in cases of suspected pulmonary tuberculosis. Two groups of patients were studied: Group I: 100 patients who were poor expectorators and who all underwent transtracheal puncture; Group II (Control): 100 patients who expectorated well or who had been given gastric lavages immediately on admission. Laboratroy analyses revealed M. tuberculosis in at least one of the samples obtained from each of the 200 patients. The authors compare the efficiency of the methods used within each group and between the two groups. Samples obtained by transtracheal aspiration and simultaneous expectoration (75% of positive results) more often contained M. tuberculosis than the other Group I samples (64% of total positive results; 64% of positive results for spontaneous sputa, 65% for gastric fluids), and as often as the Group II samples (76% of total positive results), particularly the expectoration samples (78% of positive results). Simultaneously expectorated sputum more frequently contained M. tuberculosis (82% of positive results) than transtracheal aspiration (69% of positive results). Transtracheal puncture and/or simultaneous expectoration were the only examinations revealing M. tuberculosis in 34 patients in Group I. Non-specific bacteriological findings are not relevant. However, the authors point out that this technique is not always innocuous (although no serious complications were observed in this series), and that transtracheal puncture must always be carried out by physicians trained in the technique.

No abstract