| Abstract |
Backround: Invasive fungal infections (IFI) comprise a multisystemic disease mainly caused by strains of Candida spp. (Invasive Candidiasis- 89.4%)(1). They can affect any organ and cause a multitude of clinical manifestations, ranging from an episode of fever to the onset of septic shock. Therefore, the need for targeted and effective intervention remains crucial. In recent years, an increasing trend of fungal infections has also been recorded in the Intensive Care Units, where the majority of critically ill patients is being treated, in whom complex invasive therapeutic interventions often take place. In addition, difficulties in the surveillance of infections and the administration of appropriate initial therapy arise from several factors such as the increasing resistance to echinocandins, which are the first- line treatment for IFI, the changing epidemiological distribution towards non- albicans Candida strains and the outbreaks of emerging multidrug- resistant C. auris.
Objective: The main objective of this study is to compare critically ill patients hospitalized in the ICU who developed a fungal bloodstream infection with patients who developed bacteremia and those who did not develop any bloodstream infection during their ICU hospitalization. Secondary objectives are a) the risk factors for the occurrence of fungal and bacterial bloodstream infections, b) the discriminating ability of the risk factors in the early differential diagnosis of candidemia versus bacteremia, c) the epidemiology of the pathogens most frequently identified in blood cultures in the ICU, d) the study of the resistance pattern against antifungal and antimicrobial treatment and e) the outcome of the patients.
Methods: This is a single- center retrospective observational study of patients admitted to the Intensive Care Unit of Venizeleio General Hospital during the period 2018-2023. Information on demographic, clinical, laboratory and microbiological data of each patient were collected. Patients were divided into groups according to whether a fungus (Group A) or bacteria (Group B) or no pathogen (Group C) was isolated from blood cultures. For each patient in group A, a match- case was selected from Groups B and C based on age (± 5 years), gender, duration of hospitalization (± 15 days) at the time of positive blood cultures at a ratio of 1:2:2. Patients with community acquired pathogens and patients with a hospitalization of less than 5 days were excluded from the study.
Results: 52 patients with candidemia, 103 patients with bacteremia and 101 patients without bloodstream infection were included in the study. From the candidemia group, the majority of strains isolated was non albicans Candida (N=29), whereas N= 16 strains comprised of Candida albicans. Twenty two strains (N= 22, 42%) were resistant to at least one antifungal agent, while non- albicans Candida recorded greater resistance pattern against antifungals (p<0.001). High Candida score (p<0.001) and immunodeficiency/immunosuppression (p=0.040) proved to be independent risk factors favoring the development of candidemia over bacteremia. In addition, the increased Candida score index (p<0.001), the prolonged use of invasive mechanical ventilation (p= 0.021) as well as the number of broad- spectrum antibiotics administered (p<0.001) are independent risk factors for the occurrence of candidemia versus the nonoccurrence of bloodstream infection. Regarding the outcome of patients with candidemia, independent predictive factors of mortality were the prolonged duration of mechanical ventilation (p=0.005) and hospitalization (p=0.016) as well as the laboratory markers NLR (p= 0.034), CRP (p=0.007), CRP/Alb (p=0.017), PLTs (p<0.001). Regarding the outcome of all patients included in the study, the occurrence of candidemia or even bacteremia was not an independent predictor of mortality neither in the ICU nor in the hospital.
Conclusion: In the present study, risk factors for the development of candidemia in critically ill patients included the high Candida score, the presence of immunodeficiency/immunosuppression, the prolonged duration of mechanical ventilation and the number of administered classes of antibiotics. Independent predictors for death in these patients included the prolonged duration of mechanical ventilation and hospitalization in the ICU, as well as the laboratory biomarkers NLR, CRP, CRP/Alb, PLTs on the day of positive blood culture. Using these data and applying a stricter infection stewardship program, the aim of this research is to implement appropriate therapeutic algorithms to avoid excessive administration of antimicrobial agents.
|
Collections
