Evaluation of Different Versions of the Duke Criteria for the Diagnosis of Infective Endocarditis Among Patients With Enterococcal Bacteremia; a Multicenter Study.
Enterococci are a common cause of infective endocarditis (IE). This study aimed to assess the diagnostic performance of the 2015 and 2023 Duke versions of the European Society of Cardiology (ESC) Duke criteria, as well as the 2023 Duke International Society of Cardiovascular Infectious Diseases (ISCVID) clinical criteria, for identifying IE among patients with enterococcal bacteremia.
We included adult retrospective patients with enterococcal bacteremia from 3 independent cohorts across 2 Swiss university hospitals between 2015 and 2024. An interdisciplinary Endocarditis Team classified each case as either IE or not IE. Each episode was then classified as definite, possible, or rejected IE according to the 2015 Duke-ESC, 2023 Duke-ESC, and 2023 Duke-ISCVID clinical criteria. Patients with IE (reference standard) classified as definite IE by the Duke criteria were considered true positives, while those without IE classified as rejected IE were considered true negatives.
Among 827 episodes with enterococcal bacteremia, IE was diagnosed in 173 (21%) episodes. The sensitivity of the 2015 Duke-ESC, 2023 Duke-ISCVID, and 2023 Duke-ESC clinical criteria for diagnosing IE was 67% (95% CI, 59%-74%), 79% (95% CI, 72%-85%), and 74% (95% CI, 67%-80%), respectively. Specificity was 86% (95% CI, 83%-89%) for the 2015 Duke-ESC criteria, 55% (95% CI, 51%-59%) for the 2023 Duke-ISCVID criteria, and 69% (95% CI, 65%-72%) for the 2023 Duke-ESC criteria.
Among the evaluated Duke criteria versions, the 2023 Duke-ISCVID criteria demonstrated the highest sensitivity for diagnosing IE in patients with enterococcal bacteremia. However, this was at the expense of specificity.
We included adult retrospective patients with enterococcal bacteremia from 3 independent cohorts across 2 Swiss university hospitals between 2015 and 2024. An interdisciplinary Endocarditis Team classified each case as either IE or not IE. Each episode was then classified as definite, possible, or rejected IE according to the 2015 Duke-ESC, 2023 Duke-ESC, and 2023 Duke-ISCVID clinical criteria. Patients with IE (reference standard) classified as definite IE by the Duke criteria were considered true positives, while those without IE classified as rejected IE were considered true negatives.
Among 827 episodes with enterococcal bacteremia, IE was diagnosed in 173 (21%) episodes. The sensitivity of the 2015 Duke-ESC, 2023 Duke-ISCVID, and 2023 Duke-ESC clinical criteria for diagnosing IE was 67% (95% CI, 59%-74%), 79% (95% CI, 72%-85%), and 74% (95% CI, 67%-80%), respectively. Specificity was 86% (95% CI, 83%-89%) for the 2015 Duke-ESC criteria, 55% (95% CI, 51%-59%) for the 2023 Duke-ISCVID criteria, and 69% (95% CI, 65%-72%) for the 2023 Duke-ESC criteria.
Among the evaluated Duke criteria versions, the 2023 Duke-ISCVID criteria demonstrated the highest sensitivity for diagnosing IE in patients with enterococcal bacteremia. However, this was at the expense of specificity.
Authors
Zimmermann Zimmermann, Fourré Fourré, Ledergerber Ledergerber, Epprecht Epprecht, Siedentop Siedentop, Monney Monney, Tzimas Tzimas, Frank Frank, Senn Senn, Ianculescu Ianculescu, Niclauss Niclauss, Kirsch Kirsch, Van Hemelrijck Van Hemelrijck, Dzemali Dzemali, Guery Guery, Hasse Hasse, Papadimitriou-Olivgeris Papadimitriou-Olivgeris
View on Pubmed