Background: Diagnosis of infective endocarditis (IE) often is challenging, and mortality is high in such patients. Our goal was to characterize common diagnostic tools to enable a rapid and accurate diagnosis and to correlate these tools with mortality outcomes. Methods: Because of the possibility of including perioperative diagnostics, only surgically treated patients with suspected left-sided IE were included in this retrospective, monocentric study. A clinical committee confirmed the diagnosis of IE. Results: 201 consecutive patients (age 64 ± 13 years, 74% male) were finally diagnosed with IE, and 14 patients turned out IE-negative. Preoperative tests with the highest sensitivity for IE were positive blood cultures (89.0%) and transesophageal echocardiography (87.5%). In receiver operating characteristics, vegetation size revealed high predictive power for IE (AUC 0.800, p < 0.001) with an optimal cut-off value of 11.5 mm. Systemic embolism was associated with mortality, and N-terminal prohormone of B-type natriuretic peptide (NT-proBNP) had predictive power for mortality. Conclusion: If diagnostic standard tools remain inconclusive, we suggest employing novel cut-off values to increase diagnostic accuracy and accelerate diagnosis. Patients with embolism or elevated NT-proBNP deserve a closer follow-up.
Keywords: NT-proBNP; ROC analysis; infective endocarditis; mortality; sensitivity; specificity; valve histopathology.