Protected Bronchoalveolar Lavage for the Diagnosis of HIV-Associated Pneumonia

Abstract

Abstract: The aim of protected bronchoalveolar lavage (PBAL) is to improve the diagnostic yield in bacterial pneumonia by combining the high sensitivity of bronchoalveolar lavage (BAL) with the specificity of the protected specimen brush (PSB). We evaluated PBAL and PSB in patients infected with human immunodeficiency virus (HIV) known to have a high incidence of bacterial pneumonia and colonization of the tracheohronchial tree. Sixty-three consecutive episodes with symptoms suggestive of pneumonia were investigated prospeetively. A comprehensive microbiologic workup was performed, including quantitative cultures for bacterial agents. Each episode was classified as “pneumonia,” “control,” or “non-infectious infiltrates.” Fifty-one episodes were identified as pneumonia. A definite microbial etiology could be established in 34/51 (67% cases, including a bacterial etiology in 15/51 (29%). The sensitivity for definite and probable bacterial pneumonia was 13/22 (59%) for PSB and 15/22 (68%) for PBAL (p value not significant). Specificity was 8/8 (100%), but there were 4 false-positive results in cultures of PSB and in cultures of PBAL effluent in patients with infiltrates on chest radio-graphs. PBAL effluent had a significantly higher mean bacterial index and specific bacterial index than PSB (5.4 ± 4.9 vs. 3.3 ± 3.2, p < 0.01, and 5.6 ± 0.8 vs. 3.8 ± 2.1. p < 0.001). The yield for Pneumocystis carinii was 13/15 (87%). PBAL, as compared to PSB, had a similar diagnostic yield for bacterial pneumonia. Both techniques were highly specific, but the specificity may be lower in patients with abnormal chest radiographs. PBAL most accurately reflected the bacterial load of the area investigated, and also conserved the yield of BAL reported for P. carinii. PBAL is a rational technique for the diagnosis of pneumonia in patients with HIV.