P392Prediction of percutaneous aortic valve size by transthoracic and transesophageal two-dimensional echocardiography

    loading  Checking for direct PDF access through Ovid


Measurement of the aortic annulus is critical for appropriate patient selection and successful Transcatheter Aortic Valve Implantation (TAVI). Aim of the study was to compare Transesophageal (TEE) and Transthoracic (TTE) echocardiographic methods of annulus measurements and to evaluate their potential clinical impact on TAVI.Methods: 44 consecutive patients (27 females, mean age 82 years) with severe aortic stenosis referred to our centre for TAVI underwent a comprehensive TTE and TEE using the GE-Vivid 7 ultrasound system. Annulus diameter was measured at the insertion of the leaflets in mid-systole from the para-sternal long-axis view (TTE) or from the 130° long-axis view (TEE). TAVI was performed using Edwards-Sapien prosthesis (23 or 26 mm size), or Corevalve prosthesis (26 or 29 mm size).Statistics: A paired t test and Pearson correlations were used for the comparisons of annulus measurement. To assess for error and bias, the Bland-Altman analysis method was used. A contingency analysis (expressed by the kappa value) was performed to assess the agreement between the different imaging methods with regard to the choice of prosthesis' size.Results: Aortic annulus was significantly larger when measured with TEE, 21.6±2.6 mm vs 20.7±2.3 (p 0.001). Correlation between TTE and TEE was good (r 0.75; p <0.001). The Bland-Altman analysis showed a trend for overestimation using TEE (mean difference 0,9 mm; IC 95% -2.5 to 4.4); a difference of more than 1 mm was observed in 13 patients. We evaluated the impact of the measurement of the annulus diameter using TTE or TEE on choice of the prosthesis size: agreement between TTE and TTE was only k=0.29. The choice would have been totally different in 6 patients (13%); in 7 patients (16%) the choice would have been the same with a particular model of prosthesis but different with the other model. Implantation was performed in 33 patients (75%), 24 with Edwards and 9 with Corevalve. The size of the implanted valve did not fit with the one expected on the basis of TEE and TTE measurements in 7 patients (21%) and 11 patients (33%) respectively. Implantation was successful in all but 3 patients with grade 3/4 aortic regurgitation at the 3-months follow-up: 1 with a valve smaller than expected; 2 with a valve size according with the expected.Conclusions: Measurements of the aortic annulus using TTE and TEE were not identical, with poor agreement regarding the choice of the prosthesis size. The use of TEE as reference was associated with good clinical results

    loading  Loading Related Articles