Predictors of mortality of Enterococcus faecium bloodstream infections: results from a 5-year retrospective study at Padua University Hospital

Background: Enterococcus faecium bloodstream infections (EF BSIs) cause significant morbidity and mortality in healthcare settings. We herein report a cohort of EF BSIs aiming at identifying predictors of 30-day in-hospital mortality. Methods: Retrospective cohort including hospitalized patients with EF BSIs from 2019-2023. We collected data about demographics, clinical and microbiological information, laboratory findings, treatments and deaths. Cox proportional hazards models were used to estimate hazard ratios (HRs) with 95% confidence interval (CI) for 30-day in-hospital mortality, to quantify the prognostic significance of risk factors after multivariable adjustment. A backward variable selection process in the Cox regression model was implemented to identify risk factors with statistically significant association with mortality. Results: 604 patients were included. The overall 30-day in-hospital mortality rate was 25.8%. Significant predictors of mortality included presence of septic shock, Pitt score values at least of 2, liver cirrhosis, while early source control and infectious diseases consultation were associated with a reduction in mortality rates. No statistically significant differences were observed in terms of mortality between vancomycin susceptible and vancomycin resistant BSIs. Conclusions: EF BSIs mortality was influenced by host- and disease-specific factors, including its severity. Vancomycin resistance seemed to have not an impact on mortality. Early source control and ID consultation played a critical role in improving survival, Future research should focus on prospective validation of these predictors and the development of tools and scores to early identify high-risk populations, optimizing clinical management and patient's outcomes.