Purpose: We investigated the clinical performance of (1 → 3)-β-D-glucan (BG), as an early marker of invasive fungal infections (IFI), in different clinical settings.
Methods: BG serum levels were assessed by Fungitell (Associates of Cape Cod, Inc), in parallel with galactomannan (GM) when requested by clinicians. By a prospective monocentric study, 270 episodes at risk or with suspect of IFI were enrolled, namely 58 proven-probable invasive aspergillosis (IA), 27 proven invasive candidiasis (IC), 11 possible IC, 16 P.jirovecii pneumonia (PJP), 4 episodes of other IFI and 154 non-IFI controls.
Results: We found that (a) the BG overall sensitivity, specificity, positive predictive value and negative predictive value (NPV) were 87.9, 80.5, 76.7 and 89.9 %, respectively; (b) the highest sensitivity was found in the IC groups, followed by PJP, IA and other IFI groups; (c) an association was observed between BG kinetics and patients outcome; (d) in the IA episodes, the combination of BG or GM vs GM alone increased sensitivity from 60.0 to 83.3 % in the haematological patients; (e) false-positive BG results were related to Gram-negative infections or infusion of polyclonal IgM-enriched immunoglobulins, where high levels of BG were indeed detected.
Conclusion: Besides strengthening its overall good clinical performance, we provide evidence that serum BG correlates with clinical outcome and that, once used in combination with GM, BG allows to enhance IFI diagnosis rate. The high sensitivity and NPV, observed in the Intensive Care Unit setting, open to BG validation as a marker for assessment of antifungal treatment.
Keywords: (1 → 3)-β-D-glucan; Galactomannan; Invasive fungal infections (IFI); Kinetics; Performance.