ABSTRACT
Detection of serum galactomannan (GM) and (1,3)-ß-d-glucan (BG) is considered useful for non-culture diagnosis of invasive pulmonary aspergillosis (IPA) in neutropenic patients. Only few studies evaluated these seromarkers in non-neutropenic patients suspected of having IPA. The aim of this study was to evaluate both tests together with the Aspergillus fumigatus-specific serum IgG and IgA (IgAG) test for serological IPA diagnosis in non-neutropenic patients. Sera from 87 patients suspected of having IPA were retrospectively analysed. Patients were categorised into groups of proven IPA (n = 10), putative IPA (n = 31) and non-IPA colonisation (n = 46). When the GM, BG and IgAG assays were used for patients included in the study, the sensitivity/specificity/positive predictive value (PPV)/negative predictive value (NPV) were 48.8%/91.3%/83.3%/66.7%, 82.9%/73.9%/73.9%/82.9% and 75.6%/95.7%/93.9%/81.5%, respectively. Thus, the highest specificity and PPV were confirmed for the IgAG assay. Improvements in the sensitivity and NPV were achieved by "at least one positive" analysis with the GM and BG assays, with the sensitivity/specificity/PPV/NPV values being 85.0%/69.6%/71.4%/84.2%. Nevertheless, the highest sensitivity and NPV were achieved by the "at least one positive" analysis combining the GM, BG and IgAG tests (97.6% and 96.8%, respectively). The involvement of the IgAG assay could improve IPA diagnosis in non-neutropenic patients by increasing the sensitivity and NPV when combined with the GM or BG assays. Furthermore, improvement was achieved by combining the GM, BG and IgAG assays using the "at least one positive test" strategy, especially if doubt exists.
