Bedside bronchoalveolar lavage for the diagnosis of Pneumocystis carinii pneumonia in patients with the acquired immunodeficiency syndrome.

We studied the utility of bronchoalveolar lavage (BAL) as the primary diagnostic procedure in 46 separate episodes of suspected Pneumocystis carinii pneumonia (PCP) in 42 patients with AIDS. 35 procedures were performed at the bedside. A separate group of 40 historical controls with AIDS and suspected PCP had transbronchial biopsy (TBB) as the primary procedure. At least 100 cc of saline in 50 cc aliquots was used for BAL. Specimens were processed using rapid silver methenamine, Papanicolau, and Ziehl Neelson stains with appropriate cultures. There were 29 positives and 17 true negatives for PCP with BAL, confirmed by biopsy in 11, and using clinical criteria in 6. Accuracy, sensitivity, and predictive value of a negative result were all 100%. There was no significant difference in yield between TBB and BAL (p greater than 0.10). BAL was useful to diagnose other opportunistic organisms including CMV, atypical mycobacteria, and fungi. The frequency of isolates in the TBB group were comparable. TBB was complicated by 4 episodes of major bleeding. The data indicate that bedside BAL is effective, and safer than TBB in patients with AIDS and suspected PCP. We advocate BAL as the primary diagnostic procedure which can be performed at the bedside in patients with suspected PCP and AIDS. The high accuracy and predictive value of a negative BAL for PCP suggests confirmatory procedures may not be necessary in many cases.