Assessment of left ventricular volumes, ejection fraction and regional wall motion with retrospective electrocardiogram triggered 320-detector computed tomography: a comparison with 2D-echocardiography

Abstract

Left ventricular (LV) volumes, ejection fraction (LVEF) and regional wall motion (LVRWM) have important treatment and prognostic implications in patients with coronary artery disease. We sought to determine the accuracy of 320-row multidetector computed tomography (MDCT) for the assessment of LV volumes, LVEF and LVRWM, using 2D-echocardiography as the reference standard. We evaluated 50 consecutive patients (mean age 60 ± 14 years, 66% male) who underwent 320-detector MDCT (dose-modulated retrospective electrocardiogram-triggering) and 2D-echocardiography within 14 days for investigation of known or suspected coronary artery disease. Two blinded readers measured LV volumes on MDCT and visually assessed LVRWM with a 3-point scale using a 17-segment model. A separate experienced echocardiologist, blinded to MDCT findings, assessed LVRWM on 2D-echocardiograms and determined LV volumes and LVEF using Simpson's biplane method. 2D-echocardiography served as the reference standard. Mean LVEF was 59 ± 9% (range 26-75%) on 2D-echocardiography and 60 ± 9% (range 27-76%) on MDCT. Using linear regression analysis, MDCT agreed very well with 2D-echocardiography for assessment of LVEDV (r(2) = 0.88; P < 0.001), LVESV (r(2) = 0.95; P < 0.001) and LVEF (r(2) = 0.90; P < 0.001). Mean differences (±standard deviation) of 14 ± 13 ml, 5 ± 7 ml and 1 ± 3% were observed between MDCT and 2D-echocardiography for LVEDV, LVESV and LVEF, respectively. On 2D-echocardiography, 81/850 (9.5%) segments had abnormal LVRWM. Agreement for assessment of LVRWM between 2D-echocardiography and MDCT was excellent (96%, k = 0.76). Accurate assessment of LV volumes, LVEF and LVRWM is feasible with 320-detector MDCT, with MDCT demonstrating slightly larger LV volumes than 2D-echocardiography.