Evaluation of the Protected Brush Catheter and Bronchoalveolar Lavage in the Diagnosis of Nosocomial Pneumonia

We assess the sensitivity and specificity of the protected brush catheter (PBC) and bronchoalveolar lavage (BAL) in diagnosing nosocomial pneumonia in nonimmunocompromised critically ill patients. Computerized bibliographic literature searches of MEDLINE were performed, and the reference list of each article selected was reviewed. Of 496 citations, there were 19 articles (describing 18 studies) that proved relevant. Study quality was assessed, and descriptive information concerning study populations, interventions, and clinically relevant outcome measurements was extracted. The sensitivity and specificity of PBC were high (pooled estimates, 89.9 and 94.5%, respectively). Criteria for a positive BAL have varied between studies, and sensitivity ranged from 53.3 to 100%, whereas specificity was 98.6%. Most studies did not report whether antibiotics were withheld on the basis of negative test results. In those that did, the incidence of adverse outcomes consequent on withholding antibiotics was low. BAL and PBC, combined with the use of quantitative cultures, appear to increase accuracy in diagnosing pneumonia. The strength of inference is hampered, however, by the absence of a "gold standard" for the diagnosis of pneumonia. Moreover, the generalizability of these findings is limited by the fact that there are so few methodologically sound studies from so few centers. A randomized trial of PBC is needed.