Bronchoscopic diagnosis of pulmonary infections. Comparison of protected-specimen brush and cytology brush with lung aspirates.

In a recent study the use of a new plugged double-lumen protected-specimen brush with the flexible fiberoptic bronchoscope was advocated to isolate pathogens in lower respiratory tract infections while avoiding upper respiratory tract contamination. To compare the efficacy of this brush and a standard single-lumen cytology brush in identifying the etiologic agent in lower respiratory tract infections, we studied 18 patients with lung infections. Transthoracic lung aspiration was done in all but two patients in an attempt to identify the specific etiologic agent. In these two cases, cultures of specimens of blood or postmortem lung tissue yielded the causative organism. In 12 patients anaerobic or aerobic bacteria (or both) were identified, whereas one patient had a mixed bacterial and fungal infection. Using the cytology brush and the protected-specimen brush we identified at least one pathogen in 10 of 12 and 10 of 13 cases, whereas both brushes missed one or more causative organisms in 8 of 12 and 8 of 13 cases, respectively. Nonetiologic organisms were found in 8 of 12 cases by the cytology brush and 8 of 13 cases by the protected-specimen brush. Quantitative culture techniques improved the specificity of the brush results in infections where aerobes predominated. Our data show that bronchoscopic cultures of lower respiratory tract infections do not consistently recover the causative agent and are frequently subject to contamination by nonetiologic organisms. There was no difference between the brushes in avoiding contamination.