Echocardiography is inferior to computed tomography in predicting balloon expandable transcutaneous implantation valve size in routine clinical setting-single centre study

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background Transcatheter aortic valve replacement (TAVR) became standard of care for selected patients with severe aortic stenosis. Computed tomography (CT) and 2D/3D echocardiography (ECHO) are used for aortic annulus sizing. As a result of increasing number of TAVR more imaging specialists participate in pre-procedural TAVR assessment. Purpose The aim of this study was to compare accuracy of ECHO aortic annulus measurements vs CT based parameters for Edward Sapiens TAVR in such environment. Methods Data of 145 consecutive patients with TAVR (Sapiens XT or Sapiens 3S) were analysed retrospectively. One radiologist and six echocardiographers trained in aortic annulus size measurements participated in pre-procedural aortic anulus assessment. Aortic annulus area and area derived diameter were measured/calculated in each patient from 3D ECHO data set acquired from mid-oesophageal view and from CT data set, using multiplanar reconstruction. 2D ECHO aortic annulus size was measured from mid-oesophageal 120º view in early systole. Results 139 (96%) patients had favourable TAVR result (max. mild aortic regurgitation, single valve implanted). 3D ECHO aortic anulus area and area derived diameter were smaller than corresponding CT parameters (464 ± 99 vs 479 ± 88 mm2, p < 0.001 and 24.2 ± 2.7 mm vs 25.0 ± 5.5, p = 0.002, respectively) and differed between echocardiographers (p < 0.001). 2D ECHO anulus measurement were smaller in comparison to both CT and 3D ECHO area derived diameter (22.6 ± 2.9 vs 25.0 ± 5.5mm, p = 0,013 and 22.6 ± 2.9 vs 24.2± 2.7mm, p < 0.001, respectively). In implanted valves with favourable procedure result, concordance of CT and 3D ECHO aortic annulus area with manufacturer recommended ranges was 79.4% vs 61% (p= 0.001) and for area derived diameter 80.1% vs 61.7% (p = 0.001). Using 3D ECHO measurements 33% of the patients would have hypothetically received inappropriate valve size. Conclusion 3D ECHO aortic annulus measurements are generally smaller then CT ones. If 3D ECHO based parameters only were used for TAVR (Edward Sapiens) size selection, it would have resulted in underestimation of the valve size in up to 1/3 of the patients. CT aortic annulus size assessment should be preferred method over 3D ECHO in daily practice. Abstract Figure 1 Abstract Table 1