Quantitative assessment of aortic stenosis by three-dimensional anyplane and three-dimensional volume-rendered echocardiography

Aortic stenosis is a challenge for three-dimensional (3-D) echocardiographic image resolution. This is the first study evaluating both 3-D anyplane and 3-D volume-rendered echocardiography in the quantification of aortic stenosis. In 31 patients, 3-D echocardiography was performed using a multiplane transesophageal probe. Within the acquired volume dataset, five parallel cross sections were generated through the aortic valve. Subsequently, volume-rendered images of the five cross sections were reconstructed. The smallest orifice areas of both series were compared with the results obtained by two-dimensional (2-D) transesophageal planimetry and those calculated by Doppler continuity equation. No significant differences were found between Doppler (0.76 +/- 0.18 cm(2)), 2-D echocardiography (0.78 +/- 0.24 cm(2)), and 3-D anyplane echocardiography (0.72 +/- 0.29 cm(2)). The orifice area measured smaller (0.54 =/- 0.31 cm(2), P < 0.001) by 3-D volume-rendered echocardiography. Bland-Altmann analysis indicated that for 3-D anyplane echocardiography, the mean difference from Doppler and 2-D echocardiography was - 0.04 +/- 0.24 cm(2) and - 0.06 +/- 0.23 cm(2), respectively. For 3-D volume-rendered echocardiography, the mean difference was -0.23 +/- 0.24 cm(2) and - 0.25 +/- 0.26 cm(2), respectively. In the subgroup with good resolution in the 3-D dataset, close limits of agreement were obtained between 3-D echocardiography and each of the reference methods, while the subgroup with poor resolution showed wide limits of agreement. In conclusion, planimetry of the stenotic aortic orifice by 3-D volume-rendered echocardiography is feasible but tends to underestimate the orifice area. Three-dimensional anyplane echocardiography shows better agreement with the reference methods. Accuracy is influenced strongly by the structural resolution of the stenotic orifice in the 3-D dataset.