Introduction

According to recent recommendations on echocardiographic assessment of aortic valve stenosis (AS) direct measurement of AS peak jet velocity (VmaxAS), calculation of transvalvular mean gradient (ΔPmeanAS) from the velocities using the Bernoulli equation and calculation of the effective aortic valve area (AVAeff) by continuity equation are the appropriate primary key instruments for grading AS severity in all patients with AS [1,2,3,4,5]. The calculation of the AVAeff by maximum left ventricular outflow tract (LVOT) velocity (VmaxLVOT) and VmaxAS, the ratio between VmaxLVOT and VmaxAS and the planimetry of the anatomic or geometric aortic valve area (AVAgeom) are declared as only reasonable when additional information is needed in selected patients. Regarding the fact that grading of the AS severity is based on only one direct measurable parameter (AS jet velocity), which is reliable only in patients with normal left ventricular (LV) function and forward LV stroke volume (LVSV), and on calculations based on the simplified Bernoulli equation and on LVOT diameter assessment by transthoracic echocardiography (TTE), which usually results in an underestimation of the normally oval shaped LVOT cross sectional area (CSALVOT), it is obvious that no definite gold standard can certainly be declared for grading the AS severity [6,7,8,9]. However, it is unequivocal and reflected in current guidelines that physiological assessment using TTE does provide the current gold standard in clinical practice for AS detection and grading AS severity, in preference to invasive assessment of transvalvular pressure gradient or planimetry in TEE [1, 5]. Thus, conclusions of the exclusive evaluation of AS patients by Doppler echocardiography seem to be questionable due to the susceptibility to errors and consecutive shortcomings caused by methodological limitations, mathematical simplifications and inappropriate documentation [10, 11].

The present paper will address theoretical and practical issues of TTE documentation to satisfy the needs to analyze different AS scenarios due to various flow conditions and pressure gradients (Table 1). TTE measurements always interfere with the hemodynamic situation depending on factors like afterload, aortic annulus size, pressure recovery. The necessity to characterize different AS scenarios has led to the classification according to normal or low flow conditions as well as high or low gradients [1, 5, 12].

Table 1 Recommendations for a transparent standardized documentation in patients with aortic valve stenosis by transthoracic echocardiography: views and spectra, important pitfalls, and tips to avoid them
Full size table

Appropriate parameters for assessing the AS severity in all AS patients according to recent guidelines

Peak jet velocity (VmaxAS) and mean pressure gradient (ΔPmeanAS)

Antegrade systolic velocity across the narrowed aortic valve is measured using continuous-wave Doppler (CWD) ultrasound with a parallel intercept angle between the ultrasound beam and the direction of blood flow. Accurate recording of data involves the evaluation of all acoustic windows including the supra and right parasternal window for determining the highest velocities. VmaxAS is defined as the highest velocity signal obtained from any window after careful examination [8, 13]. Even when the imaging quality is poor, VmaxAS and ΔPmeanAS can successfully be determined by CWD in most patients. The outer edge of the spectral Doppler envelope is traced to provide the velocity–time integral (VTI) for both ΔPmeanAS and calculations with the continuity equation (Table 1).

In sinus rhythm averaging of VmaxAS and ΔPmeanAS in several beats seems to be crucial because the CWD spectrum is influenced by breathin manoeuvres and deviation of optimal Doppler angulation. In regular heart rhythm practical considerations favour measurements of the highest velocity signals to avoid errors. Averaging of at least five consecutive beats is recommended in patients with arrhythmia characterizing the hemodynamic sequelae of AS with respect to mean cardiac output. However, to characterize a representative hemodynamic situation normal LV filling with normal LV output can be analyzed by measuring VmaxAS and ΔPmeanAS in the second RR interval of two consecutive long RR intervals. Thus, representative sequences of beats have to be carefully selected during arrhythmias especially avoiding measurements after postextrasystolic beats. The shape of the CWD velocity curve is parabolic in valvular AS. In contrast, LVOT obstruction results in a late time-to-peak shape of the CWD velocity curve.

In general, VmaxAS and ΔPmeanAS can be rated as reliable during normal flow conditions, normal LVSV and normal LV function. Most recent studies on transvascular aortic valve replacement have included patients with a ΔPmeanAS > 40 mmHg, which seems to be a hemodynamically unquestionable diagnosis for severe AS [14,15,16]. However, AS severity is hardly possible to assess by VmaxAS and ΔPmeanAS alone, because increased transvalvular flow can also be observed in patients with moderate aortic stenosis and hyperdynamic LV function, e.g. in the presence of concomitant aortic regurgitation (AR) or increased cardiac output (e.g. anemia, hyperthyreosis, etc.). In general, underestimation of AS severity by determining VmaxAS and ΔPmeanAS may occur during low flow conditions, reduced LVSV and reduced LV function [17,18,19,20], overestimation during hyperdynamic circulatory states and by pressure recovery [21,22,24]. Pressure recovery describes the phenomenon of conversion of kinetic energy within the AS narrowing into pressure energy in the aortic root and the ascending aorta [23, 25]. The maximum kinetic energy in AS is within the AS narrowing. The recovery of hydrostatic pressure in the aortic root occurs with decreasing velocities downstream the stenosis, because energy is neither created nor destroyed within the circulatory system. Thus, the net pressure gradient (ΔPnetAS) corresponds to the peak-to-peak hydrostatic gradient between left ventricle and aortic root after recurrence of pressure recovery, but does not match with ΔPmeanAS. Pressure recovery is predominant, if transvalvular flow is less turbulent and the aortic root is small. In contrast, conversion of kinetic into thermal energy instead of pressure recovery is present in severe turbulent flow and dilated aortic roots. In conclusion, overestimation of VmaxAS and ΔPmeanAS mainly occurs in AS patients with mild and moderate AS and small aortic root size due to pressure recovery. Small dimensions can be assumed, if the diameter of sinutubular junction (DSTJ) is less than 30 mm [23, 25].

ΔPmeanAS is estimated using information on transvalvular VTI by CWD. The peak gradient obtained from the peak velocity does not add additional information compared with VmaxAS. The peak gradient (ΔPmaxAS) is calculated using VmaxAS in the simplified Bernoulli equation. Because increased prestenotic flow > 1 m/s and pressure recovery is not considered in the ΔPmaxAS calculation, this parameter is highly prone to errors and should not be used for interpretations.

Aortic valve area (AVAeff)

AVAeff is calculated using the continuity equation, which is based on the concept that stroke volume (SV) at the valve orifice level (SVAV) is equal to that at the LVOT (SVLVOT). Velocities and pressure gradients are flow-dependent in rheology. For a specific orifice area, velocities and gradients increase with increasing transaortic flow rate and decrease with decreasing flow rate. The continuity equation (SVLVOT = SVAV) requires the measurement of three parameters: CSALVOT normally determined by using the LVOT diameter (DLVOT) in the equation: CSALVOT = π × (D/2)2, the prestenotic VTI in the LVOT (VTILVOT) determined by pulsed wave Doppler (PWD), and the transvalvular VTI (VTIAS) determined by CWD. AVAeff is calculated by the continuity equation = AVAeff = CSALVOT × VTILVOT/VTIAV.

Several limitations affect determination of AVAeff using the continuity equation. Measurement of DLVOT has to be accurately performed, because 1 mm difference can cause 10% variation of LVSV. Planimetry of CSALVOT performed in 3D TTE and TTE data sets resulted in more accurate SV calculation than estimating SV using DLVOT measured in the long axis view (Fig. 1) [26,27,28,29,30,31]. CSALVOT can often be better determined by 3D transesophageal echocardiography (TEE) due to better spatial resolution in comparison to TTE (Table 1). VTILVOT and VTIAS are mainly influenced by methodological factors including Doppler angulation, the position of the sample volume, Doppler frequency and many more (Table 1). Additionally, calculation of AVAeff produces significant error by overestimating AS severity in the presence of increased LVOT velocities (increased VTILVOT) as well as pressure recovery.

Fig. 1
figure 1

Accurate—objective and transparent—LVOT planimetry and assessment of DLVOT performed in the correct sectional plane at the correct time point using using 3D TTE

Full size image

AVAeff calculated by continuity equation using VmaxLVOT and VmaxAS, the ratio VmaxLVOT/VmaxAS and AVAgeom are described as reasonable parameters if additional information is needed [2]. At least AVAgeom—if planimetry can be correctly performed—should be discussed as a primary parameter for the assessment of AS severity (Figs. 2, 3, 4).

Fig. 2
figure 2

Accurate—objective and transparent—AV planimetry of a bicuspid AV in mild AS performed in the correct sectional plane at the correct time point using using biplane 2D TTE

Full size image
Fig. 3
figure 3

Accurate—objective and transparent—AV planimetry of a tricuspid AV in severe AS performed in the correct sectional plane at the correct time point using using biplane 2D TEE

Full size image
Fig. 4
figure 4

Accurate—objective and transparent—AV planimetry of a tricuspid AV in severe AS performed in the correct sectional plane at the correct time point using using 3D TEE

Full size image

Cardiac alterations for assessing hemodynamic relevance of AS

The assessment of VmaxAS, ΔPmeanAS, and AVAeff recommended as the primary echocardiographic parameters for the evaluation of AS severity is obviously limited by several methodological and hemodynamic factors leading to incongruencies in the characterization of AS severity in clinical routine. Thus, it might be relevant to look for secondary cardiac alterations typically caused by severe AS like left ventricular hypertrophy (LVH), diastolic dysfunction and pulmonary hypertension (PH). These signs underline the severity and are associated with a poor prognosis in AS patients [32,33,34,35,36]. Progressive narrowing of the aortic valve (AV) will lead to concentric LVH, which can be described by increased relative wall thickness (RWT > 0.42) and increased left ventricular mass (LVMi ≥ 95 g/m2 for women and ≥ 115 g/m2 for men). With increasing LVH the diastolic pressure–volume relationship increases leading to diastolic dysfunction and secondary PH. E/Eʹ is known as the surrogate parameter for LV enddiastolic pressure (LVEDP). PH can be detected by increased systolic pulmonary artery pressure (sPAP). Because severe AS is highly unlikely if secondary cardiac alterations are not present, the assessment of RWT, E/Eʹ and sPAP seems to be mandatory to support the diagnosis of severe AS. Furthermore, these parameters are independent of pressure and/or flow conditions.

Valvulo-arterial impedance (Zva) represents a marker of reduced arterial compliance characterizing excessive LV hemodynamic load. Thus, Zva describes hemodynamics and left ventricular dysfunction in severe AS patients [37, 38]. Zva is calculated according to the following equation: Zva = (Psys + ΔPmeanAS)/SVI where Psys is systolic arterial pressure, ΔPmeanAS the mean transvalvular gradient, and SVI the stroke volume index (SVI = SV/BSA) where BSA is body surface area. Reduced arterial compliance is a frequent finding in AS patients and independently contributes to increased afterload. Increased Zva > 3.5 successfully identifies patients with a poor outcome. Therefore, Zva might improve risk stratification and clinical decision making in AS patients. However, the variability of Zva due to the flow dependency is predominant in low-flow than in normal or high-flow conditions.

Myocardial fibrosis in AS patients can be described by the reduction of mitral annular plane systolic excursion (MAPSE) even in the presence of normal LV ejection fraction (LVEF) [39,40,41,42]. MAPSE is measured by M-Mode through the septal mitral annulus in the apical 4-chamber view and serves as a surrogate parameter for myocardial fibrosis. MAPSE < 5 mm indicates AS patients with severe fibrosis. In patients with low-gradient AS MAPSE < 9 mm is able to distinguish between moderate and severe AS [42].

Echocardiographic parameters for assessment of prognosis in AS patients

Prognosis of AS is strongly associated with the symptoms like angina, syncope and LV failure [42,43,44,45,46]. When facing patients with AS and symptoms it is often a challenge to judge whether the symptoms are really caused by the AS or by other reasons. On the other hand, some patients may be subjectively asymptomatic due to a reduced stress level in their daily life and would develop symptoms under exercise. Thus, it seems to be judicious to focus on prognostic relevant echocardiographic parameters in AS patients [42, 44,45,46,47].

Even in patients with mild AS with VmaxAS < 3 m/s and no AV calcification outcome is worse than in controls mainly due to ongoing adherence of the cusps. Rapid progression of aortic jet velocities can be detected by closer follow-ups and is accepted as a marker for high risk patients [35]. Cardiac and non-cardiac mortality is significantly increased in patients with moderate to severe AS with VmaxAS between 3 m/s and 5.5 m/s in comparison to controls and mild AS patients [12, 35, 44, 47,48,49,50,51]. However, it is necessary to correlate gradients with flow conditions and LV contractility, because VmaxAS and ΔPmeanAS are related to SV and LVEF. AS patients with normal LVEF are recently divided into normal flow (NF) and low flow (LF) conditions defined by the cut-off value of 35 ml/m2 for indexed SV and divided into normal or high gradient (HG) and low gradient (LG) conditions by the cut-off value of 40 mmHg for ΔPmeanAS. In AS with normal LVEF the LF–LG AS patients have the worst prognosis, followed by comparable prognoses in LF–HG and NF–HG AS patients and NF–LG AS patients indicating the best prognostic value [11–13, 52, 53]. The definition of flow conditions in AS patients by indexed forward LVSV in the current guidelines [1, 5, 12] can be scrutinized because a parameter describing a volume cannot exactly describe flow conditions, which corespond to volume per time unit. Transvalvular velocities and pressure gradients are influenced by the quantity of LVSV. Thus, increased forward LVSV might be oserved in AS patients with normal LVEF and bradycardia, in AS patients with relevant AR depending on the degree of regurgitation, and in patients with hyperdynamic states. Decreased forward forward LVSV might be present due to multiple factors like concentric LV remodeling with decreased LV cavity size, elevated arterial impedance, increased heart rate and small body height. A cardiac output in AS patients with normal LVEF and low heart rate will result in a higher LVSV with higher transvalvular velocities and pressure gradients than the same cardiac output in AS patients with normal LVEF and fast heart rate. Thus, flow conditions, which are defined by the indexed forward LVSV, describe the transvalvular flow per every particular ejection period, but not the flow conditions through the AV during a predefined time interval like e.g. cardiac output. This discrepancy between a volume and a flow parameter might presumably explain the relatively high number of patients with normal LVEF and severe LG AS in recent trials and registries [14,15,16, 54, 55].

With respect to the prognostic differences of AS subtypes [12, 52, 56, 57] it is further necessary to determine values indicating LV function including LVSV, LVEF and global longitudinal strain (GLS), respectively. LVSV can be determined by SVLVOT as the effective SV (SVeff) if no AR is present and if LVOT flow is not overestimated by hyperdynamic circulatory state or narrowing of the LVOT. Alternatively, SVeff can be determined performing Doppler measurements at the right ventricular outflow tract, if no relevant pulmonary regurgitation is present. LV planimetry—mostly performed by biplane Simpson`s method—is only reliable for determination of SVeff, if no mitral regurgitation (MR) and AR is present. In the presence of MR and/or AR LVSV determined by planimetry (SVplan) represents SVtot. In the presence of mixed aortic valve disease (MAVD) - the combination of AS and AR—without MR forward LVSV as well as SVplan represents SVtot. Finally, SVplan always represents SVtot in the presence of MR and/or AR. Forward LVSV has generally be interpreted with regard to the regurgitant volumes (RV) of concomitant AR and MR (Fig. 5). Flow conditions and SVeff can also be estimated by 3D TTE by volumetric evaluation of both ventricles. LV volume and LVEF measurements by 2D-planimetry or 3D-volumetry are more prone to errors, if the LV cavity becomes smaller. Thus, distinct LV endocardial contour detection is mandatory for reliable results. Different outcome of AS patients with comparable LVEF values is attributed to different degrees of LV fibrosis [42, 50, 53, 58]. Reduction of mitral ring displacement and MAPSE showed a strong correlation to the degree of myocardial fibrosis [39,40,41,42]. Whereas MAPSE might be deemed to be only a surrogate parameter for myocardial fibrosis, GLS seems to be a more sensitive parameter for detecting subclinical myocardial dysfunction as a prognostic marker in severe AS patients with preserved LVEF than the LVEF or MAPSE [42, 59]. This was especially shown in LF–LG AS patients [2]. GLS values derived by 3D TTE were shown to be more robust than GLS values derived by 2D TTE. However, speckle tracking by 2D TTE has a better temporal and spatial resolution than voxel tracking by 3D TTE. Recently, it could be shown that 2D GLS has strong predictive value for prognosis. A meta-analysis of ten studies with severe AS patients with preserved LVEF demonstrates reduced survival predicted by a cut-off value of GLS of—14.7% [60].

Fig. 5
figure 5

Illustration of different compounds of forward LVSV during normal condition, in isolated AS, in MAVD in the absence or presence of MR

Full size image

Practical considerations about image acquisition and data analysis to ensure diagnosis of severe AS

The image documentation of a comprehensive TTE and TEE investigation should be standardized including image optimization, complete including display details presented with current technologies and transparent to be objectively controllable by supervisors.

Beside the standardized TTE documentation described in several recommendations special attention is commanded determining RWT, DLVOT, CSALVOT, the prestenotic mean and maximum velocities (VmeanLVOT, VmaxLVOT), VTILVOT, VmeanAS, VmaxAS, VTIAS, LVEF, DSTJ, E/Eʹ sPAP and AVAgeom by planimetry. The reliability of these parameters is important for correct calculation of SVLVOT and indexed SVLVOT, CSALVOT, AVAeff and ΔPmeanAS, and ΔPmaxAS as well as conclusive interpretation of the data.

The risk of underestimation DLVOT with the consequence of calculation CSALVOT too small can be minimized by measurements in data sets of parasternal biplane image acquisition of the LVOT or in multidimensional 3D data sets of the LVOT and the aortic root. Using these modern approaches correct DLVOT assessment and/or accurate planimetry of CSALVOT is possible within the correct sectional plane and at the corresponding mid systolic time point of the cardiac cycle (Fig. 1). The same options exist for the assessment of DSTJ, and planimetry of the cross sectional area of the aorta (CSAAorta) determined at the level of sinutubular junction (STJ). It is mandatory to assess distance measurements of DLVOT and DSTJ and planimetry of CSALVOT and CSAAorta at maximum values during mid systole.

The position of the sample volume in the LVOT has to be accurately adjusted at the same position and time point of the cardiac cycle where DLVOT and CSALVOT are measured. Minimal variations of the position of the sample volume cause significant alterations of the VTILVOT. Depending on the LVOT size and the intercept angle between the ultrasound beam and the direction of LVOT blood flow it is important to position the sample volume near the septal wall to optimize Doppler angulation (Fig. 6). In general, the apical long axis view enables a better control for checking the correct position of the sample volume at the region of DLVOT measurement (in contrast to the five-chamber view, in which a visual control is not possible). Furthermore, acquisition of the Doppler spectra using the duplex or triplex mode enables to check, whether the sample volume position is at the correct position at mid systole or not.

Fig. 6
figure 6

Illustration of the effect of PWD sample volume position on LVOT spectrum in a normal AV. The yellow circle represents the correct position. The red circle is too far away from the aortic annulus. The blue circle is too near to the anterior mitral leaflet. The green circle is at the aortic annulus between the cusps. In normal AV PWD spectra of LVOT and AV are normally equal, because diameters are equal

Full size image

Inadequate assessment of CWD spectra of the transvalvular flow occurs, if the direction of ultrasound beam is not in line with the direction of the central AS jet stream. Thus, acquisition of CWD spectra derived from the suprasternal and right parasternal acoustic window are necessary, if Doppler angulation is obviously bad using the apical approach.

Methodologically, PWD and CWD spectra should be acquired during the same breathing manoeuvres and at the same heart rate to exclude different filling characteristics of the left ventricle influencing both spectra.

LVEF and left ventricular volumes can conventionally be estimated by planimetry of the two- and four-chamber view using the Simpson’s method. SVLVOT measured by planimetry can only be reliably determined if endocardial contours are correct. Automatic analysis of LVEF often delineate the inner trabecula as the endocardium resulting in too small LV volumes despite correct LVEF. Thus, endocardial contours should be manually corrected. LV contrast imaging is helpful to label the correct endocardial surfaces. It is obvious that triplane or multidimensional data sets can improve LVSV measurements in the presence of missing LV cube geometry and adequate data sets. Volume measurement of the left and right ventricle by 3D TTE enables SVeff determination, if no relevant valve regurgitations are present. In the presence of aortic and mitral regurgitations LV planimetry and volumetry results in total stroke volume (SVtot) assessment, which is important for quantitative assessment of regurgitant fraction in concomitant valvular regurgitations in AS patients [61,62,63]. However, it should be emphasized that high flow conditions will overestimate LVSV by Doppler measurements in comparison to LV planimetry (Fig. 7).

Fig. 7
figure 7

Example of overestimation of AS severity in a small patient at hyperdynamic state (VmaxLVOT > 2 m/s) with small LVOT and aortic root dimensions (DLVOT—16 mm) to document the effects of increased prestenotic flow and pressure recovery. AVAeff is calculated with 1.5 cm2. In contrast, the correct—objective and transparent—AV planimetry results in an AVAgeom = 2.8 cm2

Full size image

Because of the multiple errors in the assessment of the primary recommended target parameters to characterize AS severity, VmaxAS, ΔPmeanAS and AVAeff, the assessment of AVAgeom should be discussed. Beside the necessity of precise information regarding AV cuspidity, AV commissural orientation, AV calcification and the proximity between aortic annulus and coronary ostia [64, 65] to choose the best treatment strategy AVAgeom determination by planimetry is helpful to exclude obvious inconclusive measurements of AVAeff by continuity equation. However, prerequisite is the correct assessment of data sets for AVAgeom planimetry. Planimetry of AVAgeom seems to be underused mainly due to reduced image quality in TTE. Actually, biplane and 3D TTE or TEE provide better orientation for correctly adjusting sectional plane at mid systole for assessment of AVAgeom perpendicular to the central axis of transvalvular flow in AS patients (Figs. 2, 3, 4). However, AVAgeom planimetry in 2D and 3D TTE and TEE is still limited by laws of physics. Ultrasound frequency influence penetration and axial spatial resolution, frame rate spatial and temporal resolution, line density lateral resolution, and post processing rendering of the image contours, and all these factors have impact on the sharpness of AV cusps edges. Modern echocardiographic techniques—especially 3D TEE—can help to ensure reliable measurements of AVAgeom with excellent results (Fig. 4). Thus, planimetry of AVAgeom should be implemented into the group of primary key parameters for grading AS severity in all patients with AS. It is mandatory to perform planimetry of AVAgeom at mid systole at maximum opening of the AV documented by ECG.

Considerations about the interpretation of the parameters VmaxAS, ΔPmeanAS, AVAeff and AVAgeom to avoid under- and overestimation of AS severity

It is well known that alternatives to the three primary haemodynamic parameters recommended for clinical evaluation of AS severity are reasonable when additional information is needed in selected patients. Thus, there is no question to perform a TEE if PWD and CWD spectra by TTE are inadequate to diagnose severe AS and to quantify AS severity. The indication to perform a TEE in AS patients is generally accepted, if diagnosis cannot clearly be made by TTE. In this clinical scenario AVAgeom by planimetry of TEE images is the major additional diagnostic parameter to quantify AS severity—mainly to avoid potential underestimation of AS severity by inadequate TTE images.

This approach raises questions about the general necessity of a TEE in AS patients to fix the correct diagnosis and about the necessary information in AS patients for correct decision making of potential treatment.

If ΔPmaxAS is calculated by the simplified Bernoulli equation ΔPmaxAS = 4 × VmaxAS2, obviously the maximum instantaneous pressure gradients are calculated. It might be considered that this approach is only correct, if VmaxLVOT is < 1 m/s and DSTJ is > 30 mm. In the presence of increased LVOT velocities and relevant pressure recovery due to small aortic root size overestimation of ΔPmeanAS occurs. Especially in smaller patients (body height < 165 cm) the DLVOT is often < 18 mm and the VmaxLVOT lies between 1.2 and 1.6 m/s. Thus, ΔPmeanAS calculated by the simplified Bernoulli equation will be wrong. In addition, the constellation of small patients with a normal size of the aortic root (DSTJ is > 30 mm) is rare causing additional miscalculation due to pressure recovery (Fig. 7).

If VmaxLVOT is > 1 m/s and DSTJ is > 30 mm (normal size of the aortic root), ΔPmeanAS has to be calculated by the modified Bernoulli equation \(\Delta P_{{{\text{mean}}}} {\text{AS }} = \, 4 \, \left( {V_{{{\text{mean}}}} {\text{AS}}^{2} - V_{{{\text{mean}}}} {\text{LVOT}}^{2} } \right).\)

At least if DSTJ is < 30 mm, ΔPnetAS might be corrected according to the following equation:

$$ \Delta P_{{{\text{net}}}} {\text{AS }} = \, \Delta P_{\max } {\text{AS }} - \, \{ \, \Delta P_{\max } {\text{AS }} \times \, 2 \, \times \, ({\text{AVA}}_{{{\text{eff}}}} /{\text{CSA}}_{{{\text{Aorta}}}} ) \, \times \, (1 \, - \, [{\text{AVA}}_{{{\text{eff}}}} /{\text{CSA}}_{{{\text{Aorta}}}} ])\} . $$

Thus, the corrected AVAeff (AVAeff-corr) can be calculated by the equation:

$$ {\text{AVA}}_{{\text{eff - corr}}} = {\text{ AVA}}_{{{\text{eff}}}} \times {\text{ CSA}}_{{{\text{Aorta}}}} / \, \left( {{\text{CSA}}_{{{\text{Aorta}}}} - {\text{ AVA}}_{{{\text{eff}}}} } \right) $$

For better understanding of these calculations and their importance for defining AS severity in the range of moderate to severe AS the following theoretical example of a presumably NFHG-AS in a patient with hyperdynamic flow conditions and small aortic root dimension is presented. The three primary haemodynamic parameters recommended for clinical evaluation of AS severity, VmaxAS, ΔPmeanAS, and AVAeff, are assumed as follows for this hemodynamic scenario:

$$ V_{{\max}} {\text{LVOT }} - \, 2\,{\text{m}}/{\text{s}},V_{{{\text{mean}}}} {\text{LVOT }} - \, 1.4\,{\text{m}}/{\text{s}},{\text{ VTI}}_{{{\text{LVOT}}}} - { 4}0\,{\text{cm}}, $$
$$ V_{{\max}} AS \, - \, 5.5\,{\text{m/s}}, \, V_{{{\text{mean}}}} {\text{AS }} - \, 4.0\,{\text{m/s}},{\text{ VTI}}_{{{\text{AS}}}} { - }130\,{\text{cm}}, $$
$$ D_{{{\text{LVOT}}}} 1.8\,{\text{cm}}, \, D_{{{\text{STJ}}}} 2.4\,{\text{cm}} $$
  1. 1.

    ΔPmaxAS = 4 × VmaxAS2 = 4 × 30.3 = 121 mmHg (simplified Bernoulli equation using VmaxAS)

    • AVAeff = CSALVOT × VmaxLVOT/VmaxAV = 2.5 × 2/5.5 = 0.9 cm2

  2. 2.

    ΔPmeanAS = 4 × VmeanAS2 = 4 × 16 = 64 mmHg (simplified Bernoulli equation using VTI)

    • AVAeff = CSALVOT × VTILVOT/VTIAV = 2.55 × 40/130 = 0.8 cm2

  3. 3.

    ΔPmeanAS = 4 (VmeanAS2 − VmeanLVOT2) = 4(16 − 2) = 56 mmHg (modified Bernoulli equation using VTI)

    • VTIAV, if ΔPmeanAS is 56 mmHg, is about 115 cm

    • AVAeff = CSALVOT × VTILVOT/VTIAV = 2.55 × 40/115 = 0.9 cm2

  4. 4.

    ΔPnetAS = ΔPmaxAS − {ΔPmaxAS × 2 ×  (AVAeff/CSAAorta)  ×  (1 − [AVAeff/CSAAorta])} (equation considering pressure recovery despite all limitations of this equation have to be considered)

    = 121 − {121  × 2  × (0.9/4.50)  ×  (1 − [0.9/4.50])}

    = 121 − {121  × 2  × 0.2  × 0.8} = 121 − 38.7 = 82 mmHg

    • AVAeff-corr = AVAeff × CSAAorta/(CSAAorta − AVAeff) = 0.9 × 4.5/(4.5 − 0.9) = 4.05/3.6 = 1.1 cm2

This example illustrates AVAeff < 1 cm2 in a high gradient AS patient by using the simplified Bernoulli equation classifying the patient as a severe AS patient. Considering pressure recovery reclassification into a moderate AS patient might be necessary due to a calculated AVAeff > the cut-off value of 1 cm2 despite high values of ΔPmeanAS. It is known that contribution of pressure recovery is even more important if the AS is moderate, transvalvular flow is less turbulent and less increased and DSVT is small ( < 30 mm). In consequence, pressure recovery adjustment by CSAAorta assessment should generally be considered for accurate quantification of AS severity—especially in patients with an AVAeff near below the cut-off value between 0.8 and 1.0 cm2. CSAAorta should be estimated at the level of the sinutubular junction [22]. Regarding the fact, that a moderate AS cannot be excluded by the diagnosis of a severe AS based on TTE analysis by VmaxAS, ΔPmeanAS, and AVAeff using the continuity equation, a TEE evaluation in AS patients possibly eligible for AS treatment can be regarded as necessary in nearly all patients—especially because AVAgeom can be determined by 3D TEE in almost all patients. Pressure recovery is more predominant, the more laminar flow is present, and the smaller the aortic root. Thus, the exclusion of moderate AS is almost always necessary by TEE, if AVAeff is in the ranges of 0.8 cm2–1.0 cm2.

There are several additional reasons to establish TEE evaluation in the diagnostic procedure of AS patients if TTE investigation is not conclusive. Because the measurements of LV volumes and LVEF, flow conditions, CSALVOT, AVAgeom, DSTJ and dimensions of the proximal ascending aorta, the detection of cuspidity of the AV and of aortic plaque load, as well as the assessment of additional valvular heart diseases cannot be sufficiently assessed by TTE, a TEE examination seems to be reasonable in nearly all AS patients to clarify the diagnosis and guide decision making and adequate therapy.

Finally, the recommendation to define 1 cm2 as the cut-off value of AVAeff for severe AS is not easily comprehensible in the literature, because prognosis in AS patients was evaluated using all three primary key parameters VmaxAS, ΔPmeanAS or AVAeff with respect to different cut-off values [2, 12, 27, 35, 41, 48,49,50,51]. However, it can be concluded from the results of recent trials that the recommended cut-off value of AVAeff determined by echocardiography is well-founded [24, 66].

Summary and conclusion

VmaxAS, ΔPmeanAS, and AVAeff are accepted as the primary key parameters to quantify AS severity. Due to the fact that AVAeff determination using the continuity equation is highly prone to errors—especially if the image quality in TTE is limited and the acquisition of reliable PWD and CWD spectra is questionable—this expert consensus document addresses the following considerations: (1) TEE should be discussed for reliable measurement of AVAgeom and of the aortic root dimensions at an early stage of the diagnostic procedure. (2) To avoid misinterpretation due to inconsistent results CSALVOT and DLVOT should be assessed using biplane or 3D echocardiography. (3) Practical recommendations of the standardized TTE documentation should provide a verifiable TTE investigation for a transparent check by a supervisor. These recommendations focuss on the correct position of the PWD sample volume in the LVOT, which should be documented in cineloops of the spectra using duplex or triplex mode. The correct angulation of the transvalvular ultrasound beam of the CWD should also be accurately documented using duplex or triplex mode. LVSV has to be measured by Doppler echocardiography as well as by LV planimetry or LV volumetry to enable counterchecking of the LV stroke volumes. Depending on the presence of further valvular heart diseases SVeff should be measured by Doppler echocardiography at the right ventricular outflow tract or by right ventricular volumetry using 3D echocardiography. LVSV and LVEF, MAPSE and GLS should be mandatory implemented into the analysis of AS patients to estimate prognosis of severe AS patients. (4) If the results of a TTE investigation are not conclusive, TEE—mainly 3D TEE—and/or cardiac magnetic resonance (CMR) and/or cardiac computed tomography (CT) should be performed to determine AVAgeom and AV calcification. However, the assessment of the hemodynamic situation is still a TTE domaine in AS patients. TEE, CMR and CT predominantly flank the TTE results by accurate morphological assessment of cardiac and aortic structures.