Should we be using 3D transesophageal echocardiography instead of 2D for the assessment of aortic annular diameter in transcatheter aortic valve implantation?

Objectives: Transcatheter aortic valve implantation is a relatively new procedure aimed at patients deemed too high risk for conventional aortic valve replacement. Accurate imaging of the aortic annulus is necessary for prosthetic valve sizing and we aimed to see if there was a difference between 2D and 3D transesophageal echo (TEE) regarding this measurement. Methods: 24 consecutive patients who underwent successful valve deployment were included (18 transapical and 6 transfemoral). The measurements obtained were a 2D anteroposterior (AP) diameter (as currently used), 3D AP and 3D maximal (max) diameter. The 3D max was obtained using a multiplane reconstruction viewer to obtain the maximal annular diameter. Results: All measurements were possible in all patients with both modalities. Correlation coefficients for interobserver and intraobserver variability were both 0.86 for 2D, and 0.85 and 0.87 respectively for 3D. The maximum time to obtain 3D measurements was 90 seconds. Mean diameter for 2D AP was 20.9mm compared to 22.1 for 3D (p=0.001) and 22.9mm for 3D max (p=0.0001). When we looked at patients who developed paraprosthetic regurgitation (mild or more), the mean discrepancy between 2D and 3D max measurements was 4mm for those with regurgitation compared to 1.5mm for those without (difference 2.5mm p=0.047). Fig. 1: Comparison of the mean discrepancy between 2D and 3D Conclusion: AAD measurements by 2D and 3D TEE are reliable and reproducible. However, 2D appears to underestimate when compared to 3D. This underestimation may be implicated in paraprosthetic leaks and may be important for valve sizing in the future.