Real-time three-dimensional echocardiographic left ventricular systolic assessment: head-to-head comparison with cardiac computed tomography

Introduction

There is a paucity of information concerning side-by-side comparison of real-time 3D echocardiography (RT3DE) and cardiac computed tomography (CCT) ventricular systolic performance assessment. We sought to compare those techniques with different temporal and spatial resolution, regarding left ventricle (LV) systolic function and volumes.

Methods

We studied 67 consecutive patients (37 males, 55 ± 11 years) by RT3DE and by 64-slice CCT. We analysed by both techniques the LVEF, LVEDV and LVESV, and by RT3DE the LV dyssynchrony percentage indexes (DI%) (6, 12, 16 segment model). RT3DE and CCT data were compared by coefficients of determination (r: Pearson), Bland-Altman test and linear regression, 95% CI.

Results

RT3DE data: LVEF ranged from 30 to 78.6 (63.1 ± 7.33)%; LVEDV ranged from 44.1 to 210 (104.9 ± 29.7) ml; LVESV from 11.4 to 149 (38.9 ± 19.3) ml; 6S DI% ranged from 0.25 to 29.7 (1.92 ± 4.67)%; 12S DI% ranged from 0.29 to 26.78 (2.10 ± 5.10)%; 16S DI% ranged from 0.29 to 27.81 (2.57 ± 4.37)%. CCT data: LVEF ranged from 28 to 86 (66 ± 8.4)%; LVEDV ranged from 51 to 212 (110.3 ± 31.2) ml; LVESV from 7 to 152 (38.2 ± 19.2) ml. Correlations relative to RT3DE and CCT were: LVEF (r = 0.74, P < 0.0001, 95% CI = 0.6169 to 0.8379); LVEDV (r = 0.8213, P < 0.0001, 95% CI = 0.7229 to 0.8870); LVESV (r = 0.9096, P < 0.0001, 95% CI = 0.8929 to 0.9627). RT3DE (x) LVEF was compared with CCT (y) LVEF as: y = 19.7862 + 0.6525 x, R² = 0.5586, P < 0.0001, coefficient slope = 0.6525; RT3DE (x) LVEDV was compared with CCT (y) LVEDV as: y = 15.7057 + 0.7823 x, R² = 0.6745, coefficient slope = 0.7823, P < 0.0001; and RT3DE (x) LVESV was compared with CCT (y) LVESV as: y = 2.7997 + 0.9439 x, R² = 0.8828, coefficient slope = 0.9439, P < 0.0001.

Conclusion

In this series, adequate correlation was observed between real-time 3D echocardiography and cardiac computed tomography regarding ventricular systolic function and geometry assessment.