Comparison of ventricular volume and mass measurements from B- and C-scan images with the use of real-time 3-dimensional echocardiography: Studies in an in vitro model

Background: Real-time 3-dimensional (3D) echocardiography avoids geometric assumptions in volume analysis and permits immediate visualization in any plane without the need for cardiac or respiratory gating or computation time. This study compared the accuracy of volume and mass assessments between standard long-axis (B-scan) and short-axis (C-scan) views in a simplified but quantifiable left ventricular phantom. Methods and Results: The model comprised an inner balloon within an outer balloon separated by ultrasonographic gel. First, to mimic different chamber volumes, 12 volumes (40 to 180 mL) of water within the inner balloon were scanned with a real-time 3D system. Second, 10 volumes (80 to 170 mL) of gel were inserted between the balloons to mimic varying cardiac mass, and the gel volume space (mass) was calculated by subtracting the inner from the outer balloon volume. 'Chamber' and 'mass' measurements for both B and C scans correlated closely with the actual values (r = 0.99). However, chamber volumes from C scans were consistently less than B-scan values (mean difference from reference for C scans: -5.2 ± 1.2 mL, P < .0001; for the 2 orthogonal B scans: 0.03 ± 1.4 mL and -0.9 ± 1.5 mL, respectively, P = NS). Similarly, for gel volume measurements, B-scan results were closer to actual mass volumes (mean difference 0.3 ± 2.5 and 1.7 ± 2.9 mL) than those of C scans, which tended to underestimate (-4.5 ± 2.5 mL, P < .0001). Conclusion: Our study suggests that real-time 3D echocardiography should provide an accurate means of determining chamber volumes and cardiac mass. However, measurements performed from B-scan views may be closer to the actual values than those from C-scan views, presumably since they are less highly influenced by distortions related to lateral resolution.