Discrepancies between cardiovascular magnetic resonance and Doppler echocardiography in the measurement of transvalvular gradient in aortic stenosis: the effect of flow vorticity

Abstract

BackgroundValve effective orifice area EOA and transvalvular mean pressure gradient (MPG) are the most frequently used parameters to assess aortic stenosis (AS) severity. However, MPG measured by cardiovascular magnetic resonance (CMR) may differ from the one measured by transthoracic Doppler-echocardiography (TTE). The objectives of this study were: 1) to identify the factors responsible for the MPG measurement discrepancies by CMR versus TTE in AS patients; 2) to investigate the effect of flow vorticity on AS severity assessment by CMR; and 3) to evaluate two models reconciling MPG discrepancies between CMR/TTE measurements.MethodsEight healthy subjects and 60 patients with AS underwent TTE and CMR. Strouhal number (St), energy loss (EL), and vorticity were computed from CMR. Two correction models were evaluated: 1) based on the Gorlin equation (MPGCMR-Gorlin); 2) based on a multivariate regression model (MPGCMR-Predicted).ResultsMPGCMR underestimated MPGTTE (bias = −6.5 mmHg, limits of agreement from −18.3 to 5.2 mmHg). On multivariate regression analysis, St (p = 0.002), EL (p = 0.001), and mean systolic vorticity (p < 0.001) were independently associated with larger MPG discrepancies between CMR and TTE. MPGCMR-Gorlin and MPGTTE correlation and agreement were r = 0.7; bias = −2.8 mmHg, limits of agreement from −18.4 to 12.9 mmHg. MPGCMR-Predicted model showed better correlation and agreement with MPGTTE (r = 0.82; bias = 0.5 mmHg, limits of agreement from −9.1 to 10.2 mmHg) than measured MPGCMR and MPGCMR-Gorlin.ConclusionFlow vorticity is one of the main factors responsible for MPG discrepancies between CMR and TTE.

Semiautomated Analysis of Aortic Stenosis Parameters on Velocity-Encoded Phase-Contrast MR Images in Patients with Severe Aortic Stenosis: A Comparison with Echocardiography

“…The VENC PC sequence is an established and non-invasive tool used to quantify the velocity and blood flow based on the accumulated phase of moving protons [14,15]. Previous studies have manually assessed AVA on cine sequences and used in-house or customized software to assess hemodynamic parameters from a VENC PC sequence in valvular stenosis [3,8,[16][17][18]. We only used VENC PC sequences and commercial software for easy access in daily clinical practice.…”

“…The tendency to underestimate was apparent in severe AS due to intravoxel dephasing signal loss related in part to acceleration, alignment of the image plane with the jet, flow turbulence, partial volume effect, background noise, and phase wrap [3,5,9,[19][20][21][22]. Caruthers et al [14] reported that the VENC PC sequence underestimated velocity-time integral in more than half of patients with an AVA less than 0.8 mm 2 .…”

“…Transthoracic echocardiography (TTE) is the standard tool used to evaluate AS severity [3]. However, TTE is limited in some patients by poor acoustic windows and can be affected by the experience of the sonographer, particularly in assessing PG and AVA [4].…”

“…Cardiac catheterization is the gold standard for hemodynamic assessment, but it is invasive and requires exposure to ionizing radiation [5]. Cardiovascular magnetic resonance (CMR) has emerged as an alternative non-invasive modality to evaluate AS severity without exposure to ionizing radiation [3,5]. Accurately measuring velocity requires velocityencoded (VENC) phase-contrast (PC) pulse sequences based on the accumulated phases of moving protons, but this technique tends to underestimate the actual velocity in severe AS [4,5].…”

Valve, Ventricle, and Vessel

“…1 In fact, flow vorticity has been determined as one of the main factors responsible for discrepancies in terms of aortic transvalvular pressure gradients between CMR and echocardiography. 9 In addition to vortical structures, 10 4D-flow magnetic resonance imaging allows quantifying the eccentricity of the flow, 11 which can be calculated as the normalized flow displacement. 1 The loss of mechanical energy caused by the perturbations of the aortic blood flow as a result of vorticity and eccentricity might lead to a significant increase of the left ventricular afterload, 12 hence offering a possible explanation for the observed association between flow displacement and left ventricular hypertrophy in the study from von Knobelsdorff-Brenkenhoff et al 1 Indeed, the energy loss index, 13 which is dependent on the size relationship between the stenotic aortic valve orifice and the aorta, 12 provides additional prognostic value in the evaluation of patients with asymptomatic aortic stenosis.…”