Estimation of Stroke Volume and Aortic Valve Area in Patients with Aortic Stenosis: A Comparison of Echocardiography versus Cardiovascular Magnetic Resonance

Background: In aortic stenosis, accurate measurement of left ventricular stroke volume (SV) is essential for the calculation of aortic valve area (AVA) and the assessment of flow status. Current American Society of Echocardiography and European Association of Cardiovascular Imaging guidelines suggest that measurements of left ventricular outflow tract diameter (LVOTd) at different levels (at the annulus vs 5 or 10 mm below) yield similar measures of SV and AVA. The aim of this study was to assess the effect of the location of LVOTd measurement on the accuracy of SV and AVA measured on transthoracic echocardiography (TTE) compared with cardiovascular magnetic resonance (CMR). Methods: One hundred six patients with aortic stenosis underwent both TTE and CMR. SV was estimated on TTE using the continuity equation with LVOTd measurements at four locations: at the annulus and 2, 5, and 10 mm below annulus. SV was also determined on CMR using phase contrast acquired in the aorta (SVCMR-PC), and a hybrid AVA(CMR-PC) was calculated by dividing SVCMR-PC by the transthoracic echocardiographic Doppler aortic velocity-time integral. Comparison between methods was made using Bland-Altman analysis. Results: Compared with the referentmethod of phase-contrast CMR for the estimation of SVCMR-PC and AVA(CMR-PC) (SVCMR-PC 83616 mL, AVA(CMR-PC) 1.2760.35cm(2)), the best agreement was obtained by measuring LVOTd at the annulus or 2 mm below (P = NS), whereas measuring 5 and 10 mm below the annulus resulted in significant underestimation ofSVandAVAbyup to 15.9617.3mLand 0.2460.28cm(2), respectively (P<.01 for all). Accuracy for classification of low flow was best at the annulus (86%) and 2 mm below (82%), whereas measuring 5 and 10 mm below the annulus significantly underperformed (69% and 61%, respectively, P <.001). Conclusions: Measuring LVOTd at the annulus or very close to it provides the most accurate measures of SV and AVA, whereas measuring LVOTd 5 or 10 mm below significantly underestimates these parameters and leads to significant overestimation of the severity of aortic stenosis and prevalence of low-flow status.