Use of 3-Dimensional Color Doppler Echocardiography to Measure Stroke Volume in Human Beings: Comparison with Thermodilution

Background: The availability of accurate noninvasive measurements of cardiac output (CO) would be useful in assessing disease severity and the effects of therapeutic interventions in many different clinical settings. Current noninvasive methods are limited by their dependence on geometric assumptions. We tested the feasibility of a new technique for CO measurements based on 3-dimensional color Doppler echocardiographic (3D-CD) imaging. Objective: We sought to compare the accuracy of CO determination in human beings as measured by 3D-CD and conventional 2-dimensional echocardiography (2DE) using thermodilution as the gold standard for comparison. Methods: Simultaneous 3D-CD, 2DE, and thermodilution data were acquired in 47 patients postcardiac transplantation with good acoustic windows who required routine hemodynamic evaluation with a pulmonary artery catheter. Data were stored on compact disc and analyzed offline using custom software. Echocardiographic data were compared against thermodilution using linear regression and Bland-Altman analysis. Results: Correlation coefficients for 3D-CD and 2DE of the left ventricular outflow tract were r = 0.94 and r = 0.78, respectively. Correlation coefficients for 3D-CD and 2DE of the mitral valve were r = 0.93 and r = 0.75, respectively. Compared with 2DE, 3D-CD demonstrated a smaller bias and narrower limits of agreement in the left ventricular outflow tract (-1.84 ± 16.8 vs -8.6 ± 36.2 mL) and mitral valve inflow (-0.2 ± 15.6 vs 10.0 ± 26 mL). Conclusion: The 3D-CD determination of CO is feasible and accurate. Compared with previous noninvasive modalities, 3D-CD has the advantages of independence of geometric assumptions and ease of image acquisition and analysis.