| Abstract |
Background
Candida infections occur very frequently in patients hospitalized in intensive care units and especially in those receiving long-term mechanical ventilation, broad-spectrum antimicrobial drugs or suffering from diseases that cause immunosuppression. The increased mortality and multi-level morbidity that they cause, impose targeted scientific research on the risk factors that lead to such an infection as well as the ways to avoid and treat them in a timely manner.
Purpose
The purpose of this specific study is to compare invasive fungal infections in patients hospitalized in intensive care units with invasive bacterial infections in matched patients, but also with patients with a similar duration of hospitalization who did not develop an invasive infection. Individual elements that will be studied are the prognosis and duration of hospitalization, the mortality and disability rate, as well as predictors of outcome and the need for timely administration of empirical antimicrobial drugs
Methodology
This is a retrospective study of patients with mycemia, bacteremia and some with blood culture without microbe growth, who were hospitalized in the ICU of the General Hospital of Chania in the period 01/01/2018-12/12/2023. As mycemia we defined the growth of fungal strains from at least one blood culture of the patient during his hospitalization, while as bacteremia the growth of bacterial strains. We excluded patients who were hospitalized for less than 5 days and those who were transferred to other hospitals for continued hospitalization.
The data were entered into an EXCEL 365 spreadsheet for Windows and processed qualitatively and quantitatively with the corresponding statistical models. The statistical analysis was done with the statistical program IBM SPSS Statistics 29.0. Results
The study included 75 patients, of whom 15 had fungemia, 30 had bacteremia, and 30 did not have confirmed systemic infection. Previous colonization with fungi did not differ among patients with fungemia (N=3, 20%), those with bacteremia (N=6, 20%), and those without hematogenous infection (N=3, 16%) (Fisher exact test, p=0.512). However, the number of colonization foci differed, with fungal infection being more frequent when there were more than 2 foci (Fisher exact test p=0.011). The calculation of the Candida score differed among the study groups, with the group exhibiting fungemia having the highest scores (ANOVA p<0.001).
The presence or absence of hematogenous infection was associated with co-infection with viral pathogens, such as Covid-19 (N=7, 9.3%), HIV (N=2, 2.7%), CMV (N=1, 0.4%), VZV (N=3, 4%), with the majority of patients with fungemia having co-infection with Covid-19 (Fisher exact test p=0.010). In multiple logistic regression analysis, independent factors associated with the development of fungemia compared to all other patients were the administration of steroids/immunosuppressive agents [exp(b) 0.073 CI 95% 0.014-0.392, p=0.002], Candida score [exp(b) 0.153 CI 95% 0.035-0.665, p=0.012], and total parenteral nutrition [exp(b) 0.033 CI 95% 0.003-0.360, p=0.005]. Mortality within the ICU varied among groups, with the lowest mortality observed in the fungemia group (40%).
Conclusions
In the present study, it appears that invasive fungal infections have a better prognosis than invasive bacterial infections in critically ill patients admitted to the Intensive Care Unit (ICU). Total parenteral nutrition, Candida score>2.5, and systematic administration of steroids/immunosuppressive drugs seem to be independently associated with an increased likelihood of developing invasive fungal infections. Finally, the number of white blood cells and neutrophils upon admission to the ICU, as well as the systematic administration of steroids/immunosuppressive drugs, appear to be independently associated with the mortality of individual patients.
|
Collections
