PURPOSE: Early diagnosis and specific therapy of opportunistic infections is the cornerstone of successful treatment in immunocompromised patients with fever and pulmonary infiltrates. Flexible bronchoscopy (FB) plays an important role in identifying the cause of pulmonary infiltrates in this patient population. The purpose of this study was to determine the diagnostic yield of FB and to evaluate the impact of bronchoscopic methods on therapeutic decisions and outcome in non-HIV immunocompromised patients with pulmonary infiltrates at our institution.
METHODS: Seventy-two consecutive episodes of pulmonary infiltrates in neutropenic hosts were prospectively evaluated during a 9 month period. The non-invasive diagnostic methods included serological tests, blood and urine antigen detection, and blood, sputum and tracheobronchial aspirate (TBAS) cultures. Bronchoscopic techniques included fibrobronchial aspirate (FBAS), protected specimen brush (PSB), bronchoalveolar lavage (BAL) and transbronchial biopsy (TBB).
RESULTS: A diagnosis was obtained in 55 (76%) of the 72 patients. The etiology of the pulmonary infiltrates was infectious in 43 (60%) and non-infectious in 12 (16%); 17 (24%) remained undiagnosed. Non-invasive techniques led to the diagnosis of pulmonary infiltrates in 14 patients (19 %). Bronchoscopic techniques led to the diagnosis of pulmonary infiltrates in 36 patients (50%): FBAS 27/72 (37.5%), BAL 28/72 (39%), and PSB 11/72 (15%), TBB 30/72 (42%). The combined diagnostic yield of BAL and TBB (70%; 95% CI, 57 to 80%) was higher than that of BAL alone (p < 0.01).The results obtained with the different bronchoscopic techniques led to a change in antibiotic treatment in 29 cases (40%).
CONCLUSION: Bronchoscopic procedures are useful techniques for the diagnosis of pulmonary infiltrates in immunocompromised patients, yielding a definitive answer in as many as 76% of cases and are an important tool in adjusting medical management.
CLINICAL IMPLICATIONS: TBB should be performed whenever possible in an attempt to optimize the diagnostic yield.
DISCLOSURE: Alexander Panda, No Financial Disclosure Information; No Product/Research Disclosure Information