Abstract 1961: Multi-Modality Comparison of Volumetric Quantification of Right Ventricular Volumes Using a Novel Semi-Automated Border Detection Software

Abstract

Background. Right ventricular volumes (RVV) and ejection fraction (RVEF) measurements are challenging due to the complex shape of the right ventricle (RV). Current methods are tedious and have limited ability to take into account the base of the RV and right ventricular outflow tract. Novel semi-automated border detection software, RV analysis (RVA) was recently developed (TomTec) to overcome these limitations. Our goal was to determine whether there are inter-modality differences between real-time 3D echocardiographic (RT3DE), cardiac computed tomography (CCT) and cardiac magnetic resonance (CMR) data using the same RVA. Methods. We acquired transthoracic RT3DE images of the RV (Philips iE33 w/X4 probe) from the apical window using a full-volume mode in 12 patients referred for CMR imaging (1.5T Siemens scanner) and CCT (Toshiba) on the same day. CMR and CCT images were obtained in a stack of short-axis slices from RV base to apex, four-chamber and coronal views, and reformatted to obtain Cartesian volume data. All three types of images were analyzed to obtain end-systolic and end-diastolic volumes (ESV, EDV) and EF using RV analysis software (RVA). RT3DE and CCT were compared to CMR values as the reference standard, using linear regression and Bland-Altman analyses. Results. Analysis of RT3DE, CCT and CMR data were feasible in all patients. RT3DE and CCT RVV and RVEF correlated well with CMR (Table). CCT overestimated RVV whereas RT3DE underestimated RVV. Both CCT and RT3DE slightly underestimated RVEF compared to CMR. Conclusion. This new technique for volumetric RV surface detection applied to RT3DE and CCT data provides accurate RVV and RVEF measurements in agreement with CMR.