Candida species in the lower respiratory tract

In critically ill, intubated and mechanically ventilated patients Candida spp. are frequently isolated from respiratory tract secretions such as endotracheal aspirates and BAL. Invasive pulmonary Candida infection however is considered to be very rare. In spite of this, preemptive antifungal therapy based on isolation of Candida spp. from the respiratory tract is often initiated in critically ill patients. The disadvantages of this approach include increased selective pressure for the development of antimicrobial resistance, potential risks of adverse drug reactions and high treatment costs. On the other hand, recent studies demonstrated that the presence of Candida spp. in the lower respiratory tract was associated with worse clinical outcome, was independently associated with increased hospital mortality, and that immediate administration of appropriate antifungal therapy increased favorable outcome for patients with invasive fungal infections. The diagnosis of Candida pneumonia is currently based on histological demonstration of the invasive yeast in lung tissue associated with inflammation. However, there is lack of clarity regarding the significance of Candida spp. in lung tissue accompanied by histologic changes of pneumonia but without the histologic demonstration of invading yeasts into lung tissue. Additionally, lung biopsies are very dangerous in intubated and mechanically ventilated patients due to imminent major side effects of this procedure and thus cannot be used in clinical management of patients with suspected pulmonary Candida infection. Current microbiological methods are also inappropriate for the differentiation of pulmonary Candida colonization or infection. Hence, novel diagnostic approaches assessing the significance of Candida in the lower respiratory tract have to be established either based on pathogen-related or on host-related factors. The aim of this study is to identify biological markers to differentiate between pulmonary Candida spp. colonization and infection of the lower respiratory tract in critically ill, non-neutropenic, intubated and mechanically ventilated patients. For this purpose, pathogen related factors such as Candida prevalence, Candida quantity, invasive properties, phospholipases, secreted aspartyl proteinases, chromosome length polyporphism, transcriptional profiles, DFG16, SAP1-3, and SAP5 mRNA as well as human cellular and serum markers such as Dectin-1, CARD 9, Dectin polymorphism, TLR-2, TLR-4, TNF-a , IL-2, IL-10, IL-12, procalcitonin and (1?3) ß-D-Glucan, the indigenous pulmonary bacterial flora and underlying risk factors will be investigated in 4 different patient groups in this project. This should prospectively lead to a more targeted antifungal therapy and to a better outcome as well as to a reduction of unnecessary antifungal treatments and to a reduction of treatment costs in the future.

In critically ill, intubated and mechanically ventilated patients Candida spp. are frequently isolated from respiratory tract secretions such as endotracheal aspirates and bronchoalveolar lavage. Invasive pulmonary Candida infection however is considered to be very rare. In spite of this, preemptive (based on clinical suspicion of infection) antifungal therapy after isolation of Candida spp. from the respiratory tract is often initiated in critically ill patients. The disadvantages of this approach include increased selective pressure for the development of antimicrobial resistance, potential risks of adverse drug reactions and high treatment costs. On the other hand, recent studies demonstrated that the presence of Candida spp. in the lower respiratory tract was associated with worse clinical outcome and that immediate administration of appropriate antifungal therapy increased favorable outcome for patients with invasive fungal infections. Up to initiation of this project no clinical or laboratory based decision rule allowed clinically applicable discrimination of Candida colonized and infected patients. The aim of this study was to identify biological markers to differentiate between pulmonary Candida spp. colonization and infection of the lower respiratory tract in critically ill patients. For this purpose various biomarkers were identified an tested in different patient groups. We developed an automated test for detection fungal antigens (1,3 beta D Glucan Test) which has meanwhile been introduced into clinical practice. Furthermore, we identified molecules (interleukin 17A, kynurenine) to differentiate between Candida colonized and infected patients. We also investigated the bacterial and fungal lung microbiota using next generation sequencing to demonstrate shifts in colonizing microorganisms as risk factors for invasive fungal infections. We found that patients at intensive care units have rapid (within 2-3 days) shifts in fungal lung microbiota to Candida dominated profiles. This shift was not due to antibiotic therapy. We did not found invasive candidiasis originating from fungal lung colonization. In summary, we developed an automated test to differentiate between Candida colonized and infected patients. We further identified new biomarker for differentiation between Candida colonized and infected patients which have to be confirmed in future studies. Finally we demonstrated rapid shifts in bacterial and fungal microbiota which have not been identified before. Project data should allow more targeted antifungal therapy in patients treated in ICUs with risk for invasive candidiasis.