Diagnosis of culture-negative endocarditis: The role of the Duke criteria and the impact of transesophageal echocardiography

Abstract

Background The Duke criteria have been shown to be more sensitive than the von Reyn criteria in the diagnosis of culture-positive endocarditis but to date have not been fully validated for culture-negative endocarditis (CNE). The aim of this study was (1) to compare the diagnostic accuracy of the Duke criteria versus clinical judgment and the von Reyn criteria in CNE and (2) to assess the diagnostic impact of transesophageal echocardiography (TEE) on the Duke criteria in CNE. Methods The study group consisted of 49 patients with suspected CNE in whom the presence (n = 32) or absence (n = 17) of endocarditis was confirmed by surgery, autopsy, or both. All patients underwent transthoracic echocardiography (TTE) and TEE. They were classified into a Duke category initially with TTE data only, and then the Duke categories were reevaluated with the additional TEE data. Results The Duke criteria demonstrated a significantly higher sensitivity (72%) than the von Reyn criteria (28%; P =.0008) and a higher specificity (100%) than clinical judgment (76%; P =.02). No major differences were noted between sensitivities of the Duke criteria and clinical judgement. TEE significantly augmented the capacity to diagnose CNE by Duke criteria versus TTE (P <.05). Conclusions The Duke criteria are of high diagnostic validity for the conduction of clinical studies on CNE. They have the potential to affect clinical decision-making, based on the higher specificity versus clinical judgment. TEE appears to be crucial for the diagnosis of CNE when the Duke criteria are applied. The diagnostic differentiation between CNE, sclerotic valve degeneration, and nonbacterial thrombotic endocarditis remains a challenge. (Am Heart J 2001;142:146-52.)