Finding the Balance: Ruling Out Endocarditis in Enterococcal Bacteremia

Enterococci, especially Enterococcus faecalis, are among the most common causes of infective endocarditis (IE) and are associated with high mortality rates.^1,2 Therefore, an accurate diagnosis of IE is an important component of treating patients with enterococcal bacteremia. In clinical practice, echocardiography is typically the first diagnostic tool used when IE is suspected. However, transthoracic echocardiography has limited sensitivity, whereas transesophageal echocardiography is more invasive and carries procedural risks.³

Certain patient-related factors, including structural heart disease or implanted cardiac devices, raise greater concern for IE. The routine use of echocardiography in patients may result in unnecessary testing, increased health care resource utilization, and prolonged hospital admissions.4 To address these challenges, clinical prediction scores such as NOVA and DENOVA have been developed to stratify IE risk in patients with enterococcal bacteremia. The adapted NOVA score incorporates 4 weighted clinical components: the number of positive blood cultures, an unknown source of bacteremia, preexisting valvular disease, and the presence of a cardiac murmur. The DENOVA score builds on this system by adding 2 criteria: prolonged symptom duration of at least 7 days and evidence of embolic events. Despite the scores’ potential, evidence supporting their use remains limited.^4,5

The main objective of these scores is to minimize missed IE cases and improve hospital resource utilization. Zimmermann and colleagues conducted a multicenter retrospective cohort study at 2 tertiary care hospitals in Switzerland, evaluating adults with enterococcal bacteremia who underwent evaluation for IE across multiple clinical cohorts. Adult patients with at least 1 positive blood culture for Enterococcus species were included; cases with insufficient clinical follow-up were excluded. IE was defined using the 2023 Duke International Society of Cardiovascular Infectious Diseases criteria, and patient cases were additionally reviewed by multidisciplinary endocarditis teams. The authors compared 3 clinical prediction scores to assess their ability to identify patients at low risk for IE and to guide echocardiography use. The primary outcome was the negative likelihood ratio (NLR), with a focus on rule-out performance.4 In this study, the adapted NOVA score was used for all episodes of enterococcal bacteremia to identify patients at higher risk of IE, whereas the DENOVA score was primarily used in cases of E faecalis bacteremia. Patients meeting predefined score thresholds were classified as high risk and considered candidates for echocardiography. The authors also evaluated an additional modified score, DENOVi, designed for use across enterococcal species and that incorporates the presence of cardiac implantable electronic devices into the criteria.⁴

The study included 827 adult patients with enterococcal bacteremia, who were predominantly older and male. Notably, intravenous (IV) drug use was uncommon in this cohort, suggesting that findings are most applicable to patients with a non-IV drug use history with enterococcal bacteremia. Most bacteremia episodes were caused by E faecalis (58%) rather than E faecium (39%). Echocardiographic evaluation was performed in just over half of all episodes, and IE was ultimately diagnosed in 21% of cases.⁴ All 3 prediction scores demonstrated high sensitivity for detecting IE (approximately 89%-99%), though meaningful differences in specificity and imaging burden were observed. Adapted NOVA and DENOVi scores demonstrated the lowest NLRs (both 0.04) with negative predictive values greater than or equal to 99%, while classifying 76% and 42% of patients as high risk requiring echocardiographic imaging, respectively. In contrast, DENOVA classified 26% as high risk but had a higher NLR (0.10), showing reduced rule-out safety. Overall, DENOVi provides the most favorable balance, maintaining NOVAequivalent rule-out performance (NLR, 0.04) while having substantially fewer echocardiograms. In contrast, DENOVA, despite a lower imaging burden, sacrifices rule-out safety by misclassifying a greater proportion of IE cases as low risk. Findings were consistent for both reference standards used for IE diagnosis, supporting the robustness of the results.⁴

The authors concluded that DENOVi offers a favorable balance between diagnostic safety and resource utilization and may help safely reduce echocardiography in selected patients with enterococcal bacteremia.

This study provides an important and contemporary contribution to the literature by validating existing prediction scores for IE in enterococcal bacteremia and introducing a modified, more pragmatic tool (DENOVi) that better reflects clinical practice. Its major strengths include a large multicenter cohort, use of multidisciplinary endocarditis team adjudication alongside the updated 2023 Duke-ISCVID criteria, and a deliberate focus on rule-out safety, which is a priority when decisions involve withholding echocardiography in a high-morbidity disease.⁴ By demonstrating that DENOVi achieves a very low NLR comparable to adapted NOVA while greatly decreasing unnecessary imaging, the study furthers prior work that either overestimated risk (NOVA) or prioritized specificity at the expense of missed cases (DENOVA).

The main limitations of the study include its retrospective design, restriction to 2 Swiss tertiary centers, incomplete imaging in a substantial minority of patients, and the lack of external validation. In practice, DENOVi may be a helpful adjunct to clinical judgment when deciding whether to defer echocardiography in low-risk patients. Importantly, this strategy does not appear to meaningfully increase the rate of missed IE. However, widespread adoption would be premature; multicenter validation in diverse health care settings and integration into standardized practice should be performed before DENOVi can be confidently incorporated into routine practice guidelines.

References

1.Baddour LM, Wilson WR, Bayer AS, et al; American Heart Association Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease of the Council on Cardiovascular Disease in the Young, Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and Stroke Council. Infective endocarditis in adults: diagnosis, antimicrobial therapy, and management of complications: a scientific statement for healthcare professionals from the American Heart Association. Circulation. 2015;132(15):1435-1486. doi:10.1161/CIR.0000000000000296

2.Seby R, Kim C, Khreis M, Khreis K. Enterococcus faecalis-induced infective endocarditis: an unusual source of infection and a rare clinical presentation. J Int Med Res. 2022;50(7):3000605221112019. doi:10.1177/03000605221112019

3.Petersen JK, Østergaard L, Fosbøl EL. Role of echocardiography in the diagnosis and clinical management of infective endocarditis. Indian J Thorac Cardiovasc Surg. 2024;40(suppl 1):16-28. doi:10.1007/s12055-023-01668-4

4.Zimmermann V, Fourré N, Ledergerber B, et al. Prediction scores identifying patients at high risk of endocarditis in enterococcal bacteremia. Open Forum Infect Dis. 2025;13(1):ofaf796. doi:10.1093/ofid/ofaf796

5.Danneels P, Chabrun F, Grandière-Pérez L, Touré A, Dubée V. Endocarditis risk stratification with scores: what about reproducibility? the case of NOVA and DENOVA scores for Enterococcus faecalis bacteremia. Braz J Infect Dis. 2026;30(1):104605. doi:10.1016/j.bjid.2025.104605