Prediction Scores Identifying Patients at High Risk of Endocarditis in Enterococcal Bacteremia

Area of Science:

• Infectious Diseases
• Cardiology
• Clinical Prediction Models

Background:

• Enterococcal bacteremia poses a risk for infective endocarditis (IE).
• Clinical prediction scores like NOVA and DENOVA aim to identify low-risk patients to avoid unnecessary imaging.
• Evaluating existing scores and developing improved tools is crucial for efficient patient management.

Purpose of the Study:

• To evaluate the performance of the NOVA and DENOVA scores in predicting IE in patients with enterococcal bacteremia.
• To introduce and assess a novel modified tool, DENOVi, for IE risk stratification.
• To determine a pragmatic and objective approach for IE risk assessment in this patient population.

Main Methods:

• Retrospective study of adult patients with enterococcal bacteremia at two Swiss tertiary centers (2015-2024).
• IE diagnosis adjudicated by multidisciplinary Endocarditis Teams using 2023 Duke-ISCD criteria.
• Stratification of patients using adapted NOVA (≥4), DENOVA (≥3), and DENOVi (≥2) scores, with DENOVi incorporating intracardiac electronic devices.

Main Results:

• Out of 827 episodes, 172 (21%) had IE. The adapted NOVA, DENOVA, and DENOVi scores identified 76%, 26%, and 42% as high risk, respectively.
• Negative Likelihood Ratios (NLRs) for IE were 0.04 (95% CI, .01-.15) for NOVA, 0.10 (0.06-0.16) for DENOVA, and 0.04 (0.02-0.10) for DENOVi.
• DENOVA and DENOVi scores would have reduced the need for echocardiograms to 26% and 42% of episodes, compared to 76% with adapted NOVA.

Conclusions:

• Both adapted NOVA and DENOVi scores reliably ruled out IE in low-risk patients.
• The DENOVi score offers a balanced approach, optimizing diagnostic safety and resource utilization.
• DENOVi is a pragmatic and objective tool for IE risk stratification in enterococcal bacteremia, requiring prospective validation.