Abstract
The optimal diagnostic approach to pneumonia provides a rapid microbiologic identification of pulmonary pathogens by the least invasive means. The technique of sputum induction has been useful in the evaluation of patients with Pneumocystis carinii or mycobacterial pneumonia. It is not known whether induced sputum samples are preferable for the detection of pathogens other than P. carinii or mycobacteria. Microbiologic yields were evaluated from identically processed induced and conventional sputum samples collected from 509 consecutive patients. No statistically significant differences were found between the microbiologic yields of induced and spontaneous sputum samples. Bacterial pathogens were isolated in 19.6% of induced and 23.5% of routine specimens. Mycobacteria were cultured from 13.1% of routine and 9.4% of induced specimens. Non-Candida albicans fungi grew from 24% of routine and 20% of induced specimens. The process of sputum induction with aerosolized hypertonic saline did not alter either the purulence or the bacterial quantitation of Gram-stained sputum specimens. Sputum induction has been useful for the cytologic diagnosis of malignancy, for the diagnosis of pneumocystosis and tuberculosis, and in patients unable to spontaneously produce sputum samples. The use of induced sputum samples for the diagnosis of other infections may not be necessary when routine sputum specimens are available.