Evaluation of Vegetation Size and Its Relationship With Embolism in Infective Endocarditis

Abstract

Background— Two-dimensional (2D) echocardiography studies have shown that the maximum length of vegetation (MLV) ≥10 mm is a predictor of embolic events (EEs) in patients with infective endocarditis. However, 2D measurements probably underestimate the vegetation dimensions. In this study, we evaluated the feasibility of real-time 3-dimensional transesophageal echocardiography (RT3DTEE) in determining MLV and its accuracy in identifying the risk for EEs compared with 2D transesophageal echocardiography (2DTEE). Methods and Results— We analyzed 60 patients with vegetations. RT3DTEE measurement of MLV was obtained with Advanced QLAB Quantification Software by cropping the 3D volume with the appropriate 2D plane to obtain the largest value. The standard 2DTEE images were also evaluated to determine the MLV. Major EEs were registered from medical records, and a logistic regression analysis was performed to determine the association between MLV and EEs. The RT3DTEE MLV was larger than the 2DTEE value with a mean difference of 3.2 mm (95% confidence interval, 2.1–4.2 mm). The best cut-off value for prediction of EEs was MLV ≥20 mm with RT3DTEE and MLV ≥16 mm with 2DTEE. The positive predictive value increased from 59.1% to 65.2% when RT3DTEE was used. The accuracy of classification of patients with EEs increased from 65% to 70% with this new technique. Conclusions— RT3DTEE is a feasible technique for the analysis of vegetation morphology and size that may overcome the shortcoming of 2DTEE, leading to a better prediction of the embolism risk in patients with infective endocarditis.