Comparison of the 2015 and 2023 European Society of Cardiology versions of the Duke criteria among patients with suspected infective endocarditis.

Abstract

Diagnosing infective endocarditis (IE) poses a significant challenge. This study aimed to compare the diagnostic accuracy of the 2015 and 2023 Duke clinical criteria introduced by the European Society of Cardiology (ESC) in a cohort of patients suspected of having IE. Conducted retrospectively at two Swiss University Hospitals between 2014-2023, the study involved patients with suspected IE. Each hospitals' Endocarditis Team categorized case as either IE or not IE. The performance of each iteration of the Duke-ESC clinical criteria was assessed based on the agreement between definite IE and the diagnoses made by the Endocarditis Team. Among the 3127 episodes with suspected IE, 1177 (38%) were confirmed to have IE. Using the 2015 Duke-ESC clinical criteria, 707 (23%) episodes were deemed definite IE, with 696 (98%) receiving a final IE diagnosis. With the 2023 Duke-ESC clinical criteria, 855 (27%) episodes were classified as definite IE, of which 813 (95%) were confirmed as IE. The 2015 and 2023 Duke-ESC clinical criteria categorized 1039 (33%) and 1034 (33%) episodes, respectively, as possible IE. Sensitivity for the 2015 Duke-ESC and the 2023 Duke-ESC clinical criteria was calculated at 59% (95% CI: 56-62%), and 69% (66-72%), respectively, with specificity at 99% (99-100%), and 98% (97-98%), respectively. The 2023 ESC criteria demonstrated significant improvements in sensitivity compared to the 2015 version, although one-third of episodes were classified as possible IE by both versions.