Galactomannan detection in bronchoalveolar lavage fluid corrected by urea dilution for the diagnosis of invasive pulmonary aspergillosis among nonneutropenic patients.

Abstract

To investigate the diagnostic performance of galactomannan (GM) detection in bronchoalveolar lavage fluid (BALF) corrected by urea dilution and modification of the AspICU clinical algorithm. GM detection in serum and BALF samples was performed in nonneutropenic patients on the day of clinically suspected invasive pulmonary aspergillosis (IPA) between January 2016 and June 2018, and urea was measured in the plasma and BALF. The BALF GM concentration was corrected by urea dilution, and receiver operating characteristic (ROC) curves were generated to determine the optimal cut-off value. A total of 184 patients who were suspected of IPA, were enrolled in this prospective study together with 30 patients with lung cancer as a control group. Seventy-eight patients were diagnosed with IPA, including 37 who were verified by pathology. The urea plasma-to-urea BALF ratio in the IPA group [4.18 (IQR, 3.52-4.91)] was greater than that in the non-IPA group [3.42 (IQR, 3.12-3.76), P<0.001]. The ROC curve showed that defining the cut-off value as 2.94 optical density index (ODI) for the corrected BALF GM resulted in a sensitivity and specificity of 85.91% and 94.07%, respectively, and was more accurate than the use of the uncorrected values (P<0.05). The corrected BALF GM was valuable for diagnosing IPA in nonneutropenic patients. The modified AspICU clinical algorithm based on this measurement represents a reliable diagnostic instrument in clinical settings.