Comparison of two techniques of sputum induction in the diagnosis of pulmonary tuberculosis.

SETTING: Specialised tuberculosis out-patient clinic covering a population of 350 885. OBJECTIVES: To compare two techniques, ultrasonic nebuliser (UN) and Venturi-type face mask nebuliser (VFMN), for obtaining sputum in patients suspected of having active pulmonary tuberculosis (PTB) and who cannot expectorate, and to analyse the usefulness and cost of sputum induction with UN in a specialised medical practice. DESIGN: Prospective study from October 1997 until March 2000. RESULTS: Of 94 subjects who completed all tests, at least one sample of sputum was obtained from 89 (95.6%). UN provided adequate samples in 86 (93.4%) and VFMN in 66 (71%), a significant difference (P < 0.001). Among 43 patients diagnosed with active PTB, the diagnosis was confirmed in 39 with the samples obtained by sputum induction. Direct acid-fast bacilli smear was positive in 8 (19%). Direct costs were low and the technique can be performed on the day the patient attends the outpatient clinic. CONCLUSION: Sputum induction with an ultrasonic nebuliser was well-tolerated, of low cost, allows adequate samples to be obtained in the majority of patients who cannot produce sputum, and was more effective than the Venturi-type face mask.