ABSTRACT
PURPOSES: Enterococcal BSI is associated with significant morbidity and mortality, with fatality rates of approximately 20-30%. There are microbiological and clinical differences between E. faecalis and E. faecium infections. The aim of this study was to investigate differences in predisposing factors for E. faecalis and E. faecium BSI and to explore prognostic factors. METHODS: This study was a post-hoc analysis of PROBAC, a Spanish prospective, multicenter, cohort in 2016-2017. Patients with E. faecalis or E. faecium BSI were eligible. Independent predictors for BSI development in polymicrobial and monomicrobial BSI and in-hospital mortality in the monomicrobial group were identified by logistic regression. RESULTS: A total of 431 patients were included. Independent factors associated with E. faecium BSI were previous use of penicillins (aOR 1.99 (95% CI 1.20-3.32)) or carbapenems (2.35 (1.12-4.93)), hospital-acquired BSI (2.58 (1.61-4.12)), and biliary tract source (3.36 (1.84-6.13)), while congestive heart failure (0.51 (0.27-0.97)), cerebrovascular disease (0.45 (0.21-0.98)), and urinary tract source (0.49 (0.26-0.92)) were associated with E. faecalis BSI. Independent prognostic factors for in-hospital mortality in E. faecalis BSI were Charlson Comorbidity Index (1.27 (1.08-1.51)), SOFA score (1.47 (1.24-1.73)), age (1.06 (1.02-1.10)), and urinary/biliary source (0.29 (0.09-0.90)). For E. faecium BSI, only SOFA score (1.34 (1.14-1.58) was associated with in-hospital mortality. CONCLUSIONS: The factors associated with E. faecium and E. faecalis BSI are different. These variables may be helpful in the suspicion of one or other species for empiric therapeutic decisions and provide valuable information on prognosis.
Subject(s)
Bacteremia , Enterococcus faecalis , Enterococcus faecium , Gram-Positive Bacterial Infections , Humans , Enterococcus faecalis/isolation & purification , Male , Gram-Positive Bacterial Infections/microbiology , Gram-Positive Bacterial Infections/mortality , Gram-Positive Bacterial Infections/epidemiology , Female , Prospective Studies , Enterococcus faecium/isolation & purification , Enterococcus faecium/drug effects , Aged , Middle Aged , Bacteremia/microbiology , Bacteremia/epidemiology , Bacteremia/mortality , Risk Factors , Aged, 80 and over , Hospital Mortality , Anti-Bacterial Agents/therapeutic use , Prognosis , Spain/epidemiology , Cross Infection/microbiology , Cross Infection/epidemiology , Cross Infection/mortalityABSTRACT
Biliary-tract bloodstream infections (BT-BSI) caused by Enterococcus faecalis and E. faecium are associated with inappropriate empirical treatment and worse outcomes compared to other etiologies. The objective of this study was to investigate the risk factors for enterococcal BT-BSI. Patients with BT-BSI from the PROBAC cohort, including consecutive patients with BSI in 26 Spanish hospitals between October 2016 and March 2017, were selected; episodes caused by E. faecalis or E. faecium and other causes were compared. Independent predictors for enterococci were identified by logistic regression, and a predictive score was developed. Eight hundred fifty episodes of BT-BSI were included; 73 (8.5%) were due to target Enterococcus spp. (48 [66%] were E. faecium and 25 [34%] E. faecalis). By multivariate analysis, the variables independently associated with Enterococcus spp. were (OR; 95% confidence interval): cholangiocarcinoma (4.48;1.32 to 15.25), hospital acquisition (3.58;2.11 to 6.07), use of carbapenems in the previous month (3.35;1.45 to 7.78), biliary prosthesis (2.19;1.24 to 3.90), and moderate or severe chronic kidney disease (1.55;1.07 to 2.26). The AUC of the model was 0.74 [95% CI0.67 to 0.80]. A score was developed, with 7, 6, 5, 4, and 2 points for these variables, respectively, with a negative predictive value of 95% for a score ≤ 6. A model, including cholangiocarcinoma, biliary prosthesis, hospital acquisition, previous carbapenems, and chronic kidney disease showed moderate prediction ability for enterococcal BT-BSI. Although the score will need to be validated, this information may be useful for deciding empirical therapy in biliary tract infections when bacteremia is suspected. IMPORTANCE Biliary tract infections are frequent, and a significant cause of morbidity and mortality. Bacteremia is common in these infections, particularly in the elderly and patients with cancer. Inappropriate empirical treatment has been associated with increased risk of mortality in bacteremic cholangitis, and the probability of receiving inactive empirical treatment is higher in episodes caused by enterococci. This is because many of the antimicrobial agents recommended in guidelines for biliary tract infections lack activity against these organisms. To the best of our knowledge, this is the first study analyzing the predictive factors for enterococcal BT-BSI and deriving a predictive score.