Predictors of infective endocarditis in enterococcal bloodstream infections: a contemporary 17-year cohort study
V Fritz, M Gruebler, L Antonitsch, R Westhofen, O Assadian, V Braun, D UrbanAbstract
Background
Enterococcus spp. bloodstream infections (eBSI), particularly those caused by E. faecalis, are considered to carry a substantial risk of infective endocarditis (IE). Current guidelines provide limited guidance on risk stratification and indications for cardiac imaging. Clinical predictors of IE in patients with eBSI are not well defined, and routine cardiac imaging is often considered standard of care. However, contemporary cohort data on risk factors remain limited.
Purpose
This study aims to identify risk factors for IE in patients with eBSI.
Methods
We conducted a retrospective, single-centre study at an academic tertiary care hospital, including adult patients with at least one positive blood culture for Enterococcus spp. between 2007 and 2023. Clinical and microbiological data were obtained from the microbiology registry and electronic medical records. Death certificates were obtained from the civil registry. Collected variables included the Charlson Comorbidity Index, presence and type of implanted cardiac devices, presence of prosthetic valves, 60-day and one-year mortality, Enterococcus species, and antimicrobial resistance profiles. IE was classified according to the modified Duke criteria. Predictors of IE were analysed using multivariable logistic regression in SPSS (version 26.0; IBM, Chicago, USA).
Results
A total of 521 eBSI episodes in 471 patients (35% female) with a mean age of 68.2 ± 12.6 years were included. According to the modified Duke criteria, 26 episodes (4.2%) were classified as definite IE, 38 (6.2%) as possible IE, and 516 (83.6%) as rejected IE. In multivariable analysis, the Charlson Comorbidity Index, malignancy, age, and ampicillin resistance were not independently associated with definite or possible IE. Prior prosthetic valve replacement emerged as the only strong predictor of definite and possible IE.
Conclusions
In this contemporary cohort, IE was uncommon among patients with enterococcal BSI. Commonly assessed clinical risk factors, including age, sex, and immunosuppression, were not associated with definite or possible IE. Prior valve replacement was identified as an independent predictor. Contemporary studies report variable prevalences of IE in enterococcal BSI. Our findings may question the need for routine comprehensive cardiac imaging in all patients with enterococcal BSI. Further multi-centre studies are required to validate these results and guide risk-adapted diagnostic strategies.