Abstract 16763: Can Three-Dimensional Echocardiography Replace Two-Dimensional Biplane Simpson’s Method? A Comparison of Fully Automated Three-Dimensional Transthoracic Echocardiography and Cardiac Magnetic Resonance Imaging for Quantification of Left Ventricular Volume and Ejection Fraction
Introduction: Three-dimensional transthoracic echocardiography (3-D TTE) using manual detection of the endocardial surface has been compared with Cardiac magnetic resonance (CMR) for quantification of left ventricular (LV) volumes and ejection fraction (EF). The objective of this study was to compare LV volumes and EF by 3-D TTE using a fully automated algorithm to volumes obtained using CMR.
Methods: CMR studies were clinically indicated while 3-D TTE studies were performed for research purposes only (X-5, Philips). LV volumes by 3-D TTE were measured by two blinded echocardiographers using fully automated endocardial tracking software (Q-lab, Philips), and by manual contour tracing of 2-D images. CMR was performed with a 1.5T scanner (Siemens), and volumes were measured using automated offline software (MedVoxel).
Results: Between October 2012 and May 2013, 75 patients underwent clinically indicated CMR followed by 3-D TTE performed within 90 minutes. The initial 35 studies were performed to optimize 3-D image acquisition and consistency of measurement technique; 40 subsequent studies were used for this analysis.
Mean LV end diastolic volume (EDV) by 3-D TTE was 102.9 mL (± 34.8) and 100.1 mL (± 33.5) by CMR (p=0.66). Mean LV end systolic volume (ESV) by 3-D TTE was 34.6 mL (± 19.7) and 41.2 mL (± 21.0) by CMR (p=0.31). Mean EF was 63.4% (± 11.1) and 60.5% (± 9.9) (p=0.14) by CMR.
When automated tracing of LV endocardium by 3-D TTE was compared to CMR the correlation for EDV, ESV and EF was r=0.28 (p=0.08), r=0.42 (p=0.007) and r=0.31 (p=0.05) respectively. Comparison of manual LV myocardial tracing by 2-D TTE to CMR showed the correlation for EDV, ESV and EF to be r=0.64 (p=0.005), r=0.81 (p=<0.001) and r=0.72 (p<0.001) respectively. Exclusion of 3-D TTE studies with poor image quality, depressed LV EF (<40%), or segmental wall motion abnormalities did not improve the correlation of 3-D TTE with measurements made using CMR.
Conclusions: In this series of prospectively enrolled, unselected patients, LV volume by automated tracking of LV myocardium by 3-D TTE correlated poorly with automated measurement by CMR. Improved correlation was seen using manual tracking of biplane 2-D images. Manual image manipulation may be necessary to maximize clinical usefulness.
- © 2013 by American Heart Association, Inc.