Medizinische Universität Graz - Research portal
Selected Publication:
Duettmann, W.
Novel and established biomarkers in invasive fungal infections in patients with hematological malignancies.
Doktoratsstudium der Medizinischen Wissenschaft; Humanmedizin; [ Dissertation ] Graz Medical University; 2020. pp. 121
[OPEN ACCESS]
FullText
- Authors Med Uni Graz:
- Advisor:
- Hönigl Martin
- Krause Robert
- Rabensteiner Jasmin
- Altmetrics:
- Abstract:
- Background: Invasive fungal infections (IFI) caused by Aspergillus and Candida are difficult to diagnose. They are associated with high mortality rates especially in patients with hematological malignancies. Newly developed biomarkers or diagnostic bundles may help to establish early diagnosis, especially when used in varying specimen. Methods: This prospective and observational trial was performed at the Division of Hematology at the Medical University of Graz from July 2012 to May 2013. Patients at risk for IFI were included. Serum and urine samples were collected twice per week. Bronchoalveolar lavage (BAL) fluids were collected if ordered by the doctor in charge. Galactomannan (GM) antigen assay, mannan antigen (MAG) test, mannan antibody (MAB) test and 1,3-β-D-glucan (BDG) test were performed in all serum samples. GM assay was also assessed in urine samples. BAL fluid samples were tested for BDG assay, GM assay, and the novel Aspergillus Lateral Flow Device (LFD) test. Medical imaging and mycological laboratory work-up were performed if required and only ordered by the doctors in charge. We collected medical information of patients by chart review and compared it to the test results. According to Aspergillus, the participants were classified regarding the modified EORTC/MSG criteria. Results: Eleven cases were classified as probable invasive pulmonary infection (IPA) and 22 as possible IPA. Three participants developed a candidemia. Regarding the GM assay a cutoff of 0.5ODI was used, except for urine samples, were we used a cutoff of 0.1ODI. It showed in serum samples the sensitivity, specificity, PPV and NPV of 63.6%, 90.1%, 36.8% and 96.5%. In BAL fluid samples the sensitivity, specificity, PPV and NPV was 94.7%, 100%, 100% and 98.4%. In urine samples, it showed sensitivity, specificity, PPV and NPV of 50%, 61.5%, 13% and 91.4%. Regarding the BDG assay a cutoff of 80pg/ml was used. In serum it had sensitivity, specificity, PPV and NPV of 54.5%, 35.1%, 7.5% and 88.9% in the Aspergillus subgroup and 100%, 38.6%, 3.7% and 100% in the Candida subgroup. In BAL fluids the BDG assay showed a sensitivity, specificity, PPV, and NPV of 58.3%, 85.7%, 77.8% and 70.6%. According to the combined evaluation of MAG assay (cutoff >125pg/ml) and MAB assay (cutoff >10AU/ml) sensitivity, specificity, PPV and NPV of 0%, 96.9%, 0% and 98.4% were found. Due to invalid test charges of the Aspergillus LFD test, results were not evaluable. Conclusions: Clinical and radiological findings are still the basis of diagnosis, but a pre-emptive initiation of antifungal therapy seems feasible with help of biomarkers. GM assay in urine samples or MAG assay test results in serum samples seem to be useful to rule out IA or candidiases. BDG assay in BAL fluid samples was useful to recognize colonization with Candida but was not feasible to diagnose IPA. The GM assay in serum and BAL fluid samples confirms its value and the BDG assay is sensitive to diagnose candidemia.