Bronchoalveolar lavage in immunocompromised patients. Clinical and functional consequences.

Fiberoptic bronchoscopy and bronchoalveolar lavage are major tools in the diagnosis of acute pneumonia in immunocompromised patients. We conducted a prospective study to assess the morbidity associated with this procedure in 14 patients with AIDS and 16 patients with drug-induced immunosuppression. No patient had a PaO2 lower than 70 mm Hg with additional oxygen. Clinical data, chest roentgenogram, pulmonary function test, forced vital capacity, forced expiratory volume in one second, and arterial blood gases were recorded before and after bronchoscopy. Arterial oxygen saturation was monitored during the procedure, and initial, lowest, and final saturation values were noted. The patients were separated into three groups on the basis of chest roentgenographic findings. No procedure-induced pneumonia or need for tracheal intubation occurred. Minor clinical symptoms induced by the lavage in seven patients resolved spontaneously. By contrast, mean SaO2 decreased markedly during the procedure from 94 +/- 3 to 87 +/- 5 percent (p less than 0.0001) and returned to only 89 +/- 5 percent at the end of the procedure. Lowest SaO2 during the procedure and final SaO2 correlated poorly with initial SaO2 but correlated well with initial FVC and FEV1 (p less than 0.01). The PFT values were lower following bronchoscopy. O2 desaturation was more pronounced in patients with severe roentgenographic abnormalities. No significant differences were found between the three groups of patients, or between the AIDS and DII patients in terms of changes in PFT values. We conclude that in immunocompromised patients, bronchoscopy with BAL induces severe arterial oxygen desaturation which is correlated with initial PFT and chest roentgenographic findings, and most of these abnormalities are transient and do not lead to major complications.