Accurate assessment of stenosis severity is crucial for the management of patients with aortic stenosis (AS), and transthoracic echocardiography (TTE) is the primary modality for this purpose. In recent guidelines, the criteria proposed to identify severe stenosis are peak aortic jet velocity ≥ 4.0 m/sec, mean transvalvular gradient ≥ 40 mm Hg, aortic valve area (AVA) < 1.0 cm 2 . However, cardiologists are often confronted with situations in which there are discrepancies among the different echocardiographic parameters of AS severity, and resolving these discrepancies may be challenging.
Discrepancies in the Grading of AS Severity: When AVA and Gradient Do Not Fit
When a patient has a high velocity or gradient across the aortic valve, the stenosis is severe. In such patients, it is not required to measure AVA unless there is significant coexisting aortic regurgitation. Indeed, moderate or severe aortic regurgitation increases the volume of flow across the aortic valve in systole and may thus result in a high gradient even if the stenosis is only moderate. Another pitfall that may result in overestimation of AS severity is to mistake the velocity signal originating from mitral regurgitation for aortic flow. Aortic valve replacement is recommended (class I) in patients with high-gradient severe AS if they are symptomatic or have left ventricular (LV) systolic dysfunction (LV ejection fraction < 50%). If the mean gradient is <40 mm Hg and AVA is >1.0 cm 2 , then stenosis is moderate and there is no indication for surgery.
The most challenging situation is when AVA is <1.0 cm 2 (consistent with severe AS) but the gradient is low (consistent with nonsevere AS). This “low-gradient AS” entity (i.e., low gradient with small AVA) raises uncertainty about the actual severity of the stenosis and thus about therapeutic management if the patient is symptomatic. Furthermore, this entity is highly heterogeneous and may include (1) patients with errors in the measurement of AVA or gradient, (2) patients with small body size, (3) patients with low LV outflow, and (4) patients with normal flow but AVA-gradient discordance due to inconsistencies in the guidelines criteria for severe AS. It is important to distinguish these different causes of low-gradient AS because they have markedly different clinical implications. Patients with low-flow, low-gradient AS may greatly benefit from valve replacement, whereas those with measurement errors or small body size would not.
Low-gradient AS may be related to small body size. Small AVA in a small patient may in fact correspond to moderate AS and thus a low gradient. In such cases, it is useful to calculate AVA indexed to body surface area; indexed AVA > 0.6 cm 2 /m 2 would rule out severe AS. Low-gradient AS is often due to the presence of low LV outflow, which is defined in the guidelines as a stroke volume index < 35 mL/m 2 . This low-flow, low-gradient entity may occur in patients with reduced LV ejection fractions (i.e., classical low flow) but also in those with preserved LV ejection fractions (i.e., paradoxical low flow). All parameters of stenosis severity are flow dependent, but the gradient is the most affected in this regard given that it is a squared function of flow. Hence, in presence of low flow, the gradient may be “pseudonormalized” and thus underestimate stenosis severity. On the other hand, the flow rate may not be sufficient to completely open a valve that is only moderately stenotic, and consequently, AVA may be “pseudosevere” and overestimate stenosis severity. Low-gradient AS may also occur in the context of normal flow, and in this case, the discordance between AVA (small) and gradient (low) may be explained by the fact that, from a fluid mechanics standpoint, the AVA cut point (1.0 cm 2 ) proposed in the guidelines to define severe stenosis corresponds to a mean gradient of 30 to 35 mm Hg, which is lower than the guideline cut point of 40 mm Hg. As a consequence, some patients with normal flow may have discordant AS grading depending on whether one uses AVA or gradient to classify stenosis severity. These patients generally have better outcomes than those with low-flow, low-gradient AS, and they benefit less from aortic valve replacement.
A frequent cause of low-gradient AS is an error in the echocardiographic measurement of AVA and/or gradient. If the continuous-wave Doppler beam is not well aligned with the aortic flow jet, this may result in an underestimation of the gradient. It is thus important to perform a multiple-window interrogation with continuous-wave Doppler that includes not only the apical windows but also the right parasternal and suprasternal windows to obtain the maximal transvalvular velocity and gradient. The measure of AVA is even more challenging and more prone to technical errors, particularly in elderly patients with calcified aortic valves. AVA is determined by the method of the continuity equation, whereby the numerator is the stroke volume calculated from the LV outflow tract (LVOT) diameter and the integral of flow velocity over systole, while the denominator is the time-velocity integral of flow through the stenotic aortic valve. Among these three parameters used for the calculation of AVA, LVOT diameter is the most important potential source of error. Given that LVOT diameter is squared in the continuity equation, a small error in LVOT diameter measurement may result in an important error in the calculation of AVA. In practice, LVOT diameter is measured by two-dimensional (2D) TTE at peak systole, although the LVOT is a dynamic three-dimensional (3D) structure. The calculation of stroke volume is based on the assumption that the LVOT cross-sectional area is circular, which is not necessarily true in all patients.
Implications of LVOT Remodeling and Dynamics in the Assessment of AS
In the elegant study reported by Mehrotra et al . in this issue of JASE , the authors compared the geometry and dynamics of the LVOT assessed by 2D TTE and 3D transesophageal echocardiography (TEE) in patients with AS compared with healthy control subjects. There were several main findings. First, at end-diastole, LVOT geometry is similar in patients with AS and control subjects. In both cases, LVOT cross-sectional area was indeed elliptical, with an ellipticity index close to 1.3. Second, at peak systole, the LVOT cross-section became more circular in patients with AS and control subjects. However, the LVOT was less distensible in patients than in control subjects, and as a consequence, systolic “circularization” was less important in patients with AS than in control subjects. Compared with end-diastolic measures, the eccentricity index decreased at peak systole, but the changes were blunted in patients with AS. Therefore, the eccentricity index at peak systole was significantly larger in patients with AS (1.18) than in control subjects (1.08) ( P < .001). Third, these differences in the distensibility and dynamic changes in LVOT geometry during the cardiac cycle were likely related to the remodeling of the LVOT in patients with AS. The LVOT posterior wall was indeed thicker in patients with AS than in control subjects. Furthermore, although this was not measured in this study, patients with AS generally have calcified LVOTs and aortic roots, which may reduce their distensibility during systole.
In the continuity equation, the cross-sectional area is calculated with the use of the diameter measured by 2D TTE in the parasternal long-axis view by assuming that the LVOT is circular. However, this method is inaccurate if the LVOT cross-section is elliptical. In the study of Mehrotra et al ., the minor- and major-axis diameters of the LVOT ellipse measured by 3D TEE were similar in patients with AS versus control subjects at end-diastole. However, because of the differences in LVOT distensibility, the minor-axis diameter became significantly smaller in patients than in control subjects at peak systole. The anteroposterior diameter of the LVOT measured by 2D TTE in the parasternal long-axis view ( Figure 1 ) generally corresponds to the smaller diameter. Hence, use of the 2D transthoracic echocardiographic measure of LVOT diameter may result in an underestimation of LVOT cross-sectional area and thus result in a smaller stroke volume and AVA. An underestimation of AVA may, in turn, lead to the false conclusion that the patient has low-flow, low-gradient severe AS, whereas he or she in fact has moderate AS.