The first consensus definitions for invasive fungal diseases (IFD) were published in 2002. Advances in diagnostic tests and a clear need for improvement in certain areas led to a revision of these definitions in 2008. However, growing data on Aspergillus galactomannan (GM) thresholds and the introduction of new polymerase chain reaction-based diagnostic tests resulted in a further update by EORTC and Mycoses Study Group Education and Research Consortium (MSGERC) in 2020. Compared to the 2008 version, the 2020 EORTC/MSGERC criteria have stricter definitions, especially regarding GM levels, which should lead to improved specificity. Thus, our study aimed to evaluate diagnostic changes, based on GM levels, resulting from these new definitions and ascertain the impact of the new classification on mortality rates.

Patients hospitalized in a single tertiary care center with hematologic malignancies and undergoing bronchoscopy for suspected IPA between April 2004 and December 2019 were included in this retrospective study.

The study population consisted of 327 patients with 31 patients (nine patients with proven IPA and 22 patients with no IPA) excluded from the study. 194 patients were classified as probable IPA cases according to 2008 EORTC/MSG criteria. However, 53 (27.3%) of these patients were re-classified as possible IPA according to 2020 EORTC/MSGERC criteria, due to novel galactomannan cut-off levels. Compared to re-classified possible IPA patients, those remaining in the probable IPA category experienced a higher incidence of septic shock (34.0% vs 16.9%, p=0.02), and required more non-invasive (12.0% vs 0.0%, p=0.004) and invasive (44.6 vs 24.5%, p=0.01) mechanical ventilation. There was a higher in-hospital mortality rate in probable IPA patients than in the re-classified possible IPA group (42.5% vs 22.6%, p=0.01). Patients reassigned to possible IPA had similar underlying diseases, radiological features and prognosis to patients already classified as possible IPA. Independent risk factors for mortality were classification as probable IPA according to 2020 EORTC/MSGERC criteria, lack of remission from hematologic malignancy, and number of nodules in Thorax CT.

The use of 2020 EORTC/MSGERC criteria resulted in a 27.3% significant reduction in probable IPA diagnoses and created a more homogeneous category of patients with respect to treatment response, prognosis and mortality. Therefore, 2020 EORTC/MSGERC criteria afford more reliable mortality prediction than 2008 EORTC/MSG criteria.