Sputum and bronchial secretion samples are equally useful as bronchoalveolar lavage samples for the diagnosis of invasive pulmonary aspergillosis in selected patients.

Abstract

In the absence of histopathology studies of lung biopsies, the bronchoalveolar lavage (BAL) sample is preferred for the diagnosis of invasive pulmonary aspergillosis. Isolation of Aspergillus fumigatus from sputum and bronchial secretion samples are commonly interpreted as colonization or laboratory contamination, particularly in nonneutropenic patients. We studied if sputum/bronchial secretions and BAL samples are equally useful for the diagnosis of invasive pulmonary aspergillosis. We retrospectively selected 14 patients with proven (n = 1) or probable (n = 13) invasive pulmonary aspergillosis from whose samples A. fumigatus had been simultaneously isolated in BAL and sputum/bronchial secretions between 2006 and 2012. The isolates were identified by sequencing the β-tubulin gene and genotyped using the STRAf assay. Matches between BAL and sputum/bronchial secretions were observed in patients with identical genotypes in BAL and sputum/bronchial secretions. All patients had clinically suspected pneumonia, before the diagnosis of invasive pulmonary aspergillosis. The sample from which A. fumigatus was initially isolated was collected as a result of the presence of fever (50%), abnormal radiological findings (100%), and/or pneumonia that did not respond to antibiotics (36%). The underlying conditions varied, although the most common predisposing factors were hematological malignancies (21.5%) and COPD (43%). In 13 of the 14 patients (93%), we found matching genotypes in the BAL and the sputum/bronchial secretion samples. Genotyping showed that samples of sputum or bronchial secretions were equally useful as samples of BAL for the diagnosis of invasive pulmonary aspergillosis.