Comparison of two-dimensional and real-time three-dimensional transesophageal echocardiography in the assessment of aortic valve area

Abstract

Background: The accuracy of two-dimensional transesophageal echocardiography (2D-TEE) for the measurement of aortic valve area (AVA) in patients with aortic stenosis (AS) depends upon the cross-section selected for imaging. Real-time three-dimensional transesophageal echocardiography (3D-TEE) may overcome this limitation of 2D-TEE. The goal of this study was to compare 3D-TEE with 2D-TEE for the measurement of AVA.

Methods and results: Twenty-five patients with AS underwent TEE. In 2D-TEE, the aortic valve image was obtained at the orifice level in the short-axis view, and AVA was measured by planimetry of the acquired images (2D-AVA). In 3D-TEE, 3D data containing the entire aortic valve were obtained. Then, a short-axis cross-section containing the smallest orifice in mid-systole was cut from the 3D data during image postprocessing, and the AVA was measured by planimetry (3D-AVA). The 3D-AVA was significantly smaller than the 2D-AVA (0.79±0.35cm(2) vs. 0.93±0.40cm(2), p<0.0001), but there was a strong correlation between 3D-AVA and 2D-AVA (R=0.94). Although the frame rate was lower in 3D-TEE than in 2D-TEE (17±6Hz vs. 58±16Hz), the 3D-AVA determined at each frame during systole showed that the difference between 3D-AVA and 2D-AVA was not explained by the lower frame rate. The time required for image acquisition of the aortic valve was shorter with 3D-TEE than with 2D-TEE (p=0.0005).

Conclusions: The geometric AVA is smaller with 3D-TEE than with 2D-TEE, and the difference is not due to the lower frame rate of 3D-TEE. The improved accuracy of 3D-TEE along with reduced image acquisition time indicates that 3D-TEE is superior to 2D-TEE for the assessment of AVA.