Search

Background: The 2023 Duke-ISCVID (International Society of Cardiovascular Infectious Diseases) classification is a new diagnostic tool for infective endocarditis, updating the 2000 modified Duke and the 2015 European Society for Cardiology (ESC) classifications. In comparison, its sensitivity is higher; however, its specificity remains to be evaluated and compared to that of the 2 other classifications in endocarditis suspected patients., Methods: We retrospectively collected the characteristics of patients hospitalized in Bichat University's Hospital, Paris, in 2021, who had been evaluated for clinical suspicion of endocarditis, have had at least a transthoracic echocardiography, 2 pairs of blood cultures, 3-month follow-up and in whom endocarditis diagnosis was finally rejected. All patients were classified by 2000 modified Duke, 2015 ESC and 2023 Duke-ISCVID, as though the endocarditis diagnosis had not been rejected., Results: In total, 130 patients' charts were analyzed. Mean age was 62 years, 84 (64.6%) were male, 39 (30.0%) had prosthetic cardiac valve or valve repair, 21 (16.2%) cardiac implanted electronic device, and 30 (23.1%) other cardiac conditions. Overall, 5, 2, and 5 patients were falsely classified as definite endocarditis with the 2000 modified Duke, 2015 ESC, and 2023 Duke-ISCVID classifications, respectively. The corresponding specificities were 96.2% (95% confidence interval [CI] [90.8%, 98.6%]), 98.5% (95% CI [93.9%, 99.7%]), and 96.2% (95% CI [90.8%, 98.6%]). The rates of possible endocarditis were of 38%, 35%, and 35% in the 3 classifications, respectively., Conclusions: The 2023 Duke-ISCVID classification is highly specific for ruling out the diagnosis of definite infective endocarditis in patients who had been evaluated for IE., Competing Interests: Potential conflicts of interest. F. R. reports grants or contracts paid to institution by Fédération Française de Cardiologie and GE Healthcare. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed., (© The Author(s) 2024. Published by Oxford University Press on behalf of Infectious Diseases Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.)