Performance of the Bronchoalveolar Lavage Fluid Aspergillus Galactomannan Lateral Flow Assay With Cube Reader for Diagnosis of Invasive Pulmonary Aspergillosis: A Multicenter Cohort Study

Abstract

Abstract Background The Aspergillus Galactomannan Lateral Flow Assay (LFA) is a rapid test for the diagnosis of invasive aspergillosis (IA) that has been almost exclusively evaluated in patients with hematologic malignancies. An automated digital cube reader that allows for quantification of results has recently been added to the test kits. Methods We performed a retrospective multicenter study on bronchoalveolar lavage fluid (BALF) samples obtained from 296 patients with various underlying diseases (65% without underlying hematological malignancy) who had BALF galactomannan (GM) ordered between 2013 and 2019 at the University of California, San Diego, the Medical University of Graz, Austria, and the Mannheim University Hospital, Germany. Results Cases were classified as proven (n = 2), probable (n = 56), putative (n = 30), possible (n = 45), and no IA (n = 162). The LFA showed an area under the curve (AUC) of 0.865 (95% confidence interval [CI] .815–.916) for differentiating proven/probable or putative IA versus no IA, with a sensitivity of 74% and a specificity of 83% at an optical density index cutoff of 1.5. After exclusion of GM as mycological criterion for case classification, diagnostic performance of the LFA was highly similar to GM testing (AUC 0.892 vs 0.893, respectively). LFA performance was consistent across different patient cohorts and centers. Conclusions In this multicenter study the LFA assay from BALF demonstrated good diagnostic performance for IA that was consistent across patient cohorts and locations. The LFA may serve a role as a rapid test that may replace conventional GM testing in settings where GM results are not rapidly available.