Disease-stratified incidence, risk factors and outcomes of galactomannan positive invasive aspergillosis

T Mercier1,2, K Lagrou1,3, J Maertens1,4

Author address: 

1Department of Microbiology and Immunology, KU Leuven, Leuven, Belgium 2Department of Internal Medicine, University Hospitals Leuven, Leuven, Belgium 3Department of Laboratory Medicine, University Hospitals Leuven, Leuven, Belgium 4Department of Haematology, University Hospitals Leuven, Leuven, Belgium


Introduction: The epidemiology of invasive aspergillosis (IA) has been changing over the last decades. Since the incidence of IA is dependent on the underlying disease and the associated risk and host factors, the local epidemiology of these diseases can greatly influence the epidemiology of IA. Most previous epidemiological studies consisted of either IA registry studies, which collect all cases of IA in a certain region, but do not provide any data on the epidemiology of the underlying diseases. Other epidemiological studies are based on the incidence of IA in a certain population (e.g. transplant recipient registries), making direct comparisons between risk groups difficult. We therefore performed a hospital wide study in a tertiary care center, stratifying the incidence of IA by underlying disease.

Methods: We retrospectively reviewed digital patient records of all 6181 patients between 2011 and 2016 who had at least one galactomannan (GM) test performed. Records of 1364 patients with either a single broncho-alveolar lavage GM index ≥ 0.5 or at least two consecutive serum GM indices ≥ 0.5 were examined in more detail and were classified according to the 2008 EORTC-MSG criteria. We recorded 48 different possible risk factors described previously in different studies, including type of antifungal prophylaxis, renal and hepatic function, blood cell counts, and cardiac, neurological and pulmonary disorders. Outcome at week 6 and 12 was classified according to the 2008 EORTC-MSG criteria. Risk factors were compared between the hematological and solid organ transplantation (SOT) groups using the Chi squared test for discrete variables or Wilcoxon’s rank-sum test for continuous variables.

Results: We identified 47 cases of proven IA and 154 cases of probable IA. The total number of proven or probable IA cases, incidence of IA per total number of new patients with the underlying disease, and week 6 and week 12 all-cause mortality were as follows (in descending order of incidence): in acute myeloid leukemia 49, 121.0/1000/year, 18.4%, 36.7%; in allogeneic stem cell transplantation 36, 94.7/1000/year, 36.1%, 50%; in pancreatic transplantation 1, 52.6/1000/year, 0%, 0%; in lung transplantation 20, 50.5/1000/year, 20.0%, 20.0%; in heart transplantation 5, 36/1000/year, 0.0%, 20.0%; in autologous stem cell transplantation 6, 31.6/1000/year, 16.7%, 33.3%; in any hematological malignancy 61, 26.3/1000/year, 47.5%, 59.0%; in kidney transplantation 17, 20.8/1000/year, 35.3%, 41.2%; and in liver transplantation 3, 6.9/1000/year, 33.3%, 33.3%. All-cause mortality was similar between hematological patients and SOT recipients at 6 weeks (35.5% vs 27.8%, p = 0.520) and at 12 weeks (49.1% vs 33.3%, p = 0.145). Hematological patients were significantly more likely to have had probable IA (87.3% vs 63.9%, p = 0.004), to have received echinocandins as first line therapy (23.6% vs 5.6%, p = 0.032), and to have had their therapy switched to another class of antifungals after first-line therapy (50.0% vs 16.7%, p < 0.001).

Conclusion: Patient characteristics and incidence rates of invasive aspergillosis differ significantly between different patient populations, while outcomes appear to be similar.



Full conference title: 

The 8th Advances Against Aspergillus, Lisbon Conference Center, Lisbon, Portugal
    • AAA 8th (2018)