Assessment of Aortic Stenosis Severity by

Cardiovascular Magnetic Resonance

J. Garcia*
M. Markl*,**
A.J. Barker*

*Department of Radiology, Northwestern University,
737 N Michigan Av, Suite 1600, Chicago, IL, USA, 60611. 
**Department of Biomedical Engineering, Northwestern University, 2145 Sheridan Road, Evanston, IL, USA, 60208.


Cardiovascular magnetic resonance (CMR) has become a valuable tool to corroborate aortic stenosis (AS) severity when echocardiography assessment is discordant. Moreover, CMR can provide useful complementary information about AS severity and hemodynamic markers. In particular, the use of advanced 4D flow CMR allows a comprehensive assessment of complex flow alterations produced by AS. This review provides an overview of the added value obtained by standard 2D flow and advanced 4D flow quantification for AS severity assessment and discusses the advantages and disadvantages of current clinical metrics. This includes an introduction of promising new hemodynamic markers, and discusses how these novel makers may identify potential complications and disease progression in patients with AS. 

Keywords: aortic stenosis, cardiovascular magnetic resonance, flow quantification.

Julio Garcia
Department of Radiology, Northwestern University, 737 N Michigan, Suite 1600, Chicago, IL, USA, 60611.
Email address:
Fecha de recepción:
8 de Agosto de 2013
Fecha de aceptación:
18 de Noviembre de 2013


La imagenología de resonancia magnética cardiovascular (RMC) se ha establecido como una importante herramienta para corroborar la severidad de la estenosis aórtica (EA) cuando el examen por ecocardiografía es contradictorio. Además, la RMC puede proveer importante información complementaria con respecto a la severidad de la EA y diversos indicadores hemodinámicos. En particular, el uso de técnicas avanzadas de flujo en 4D por RMC permite una extensiva evaluación de las complejas alteraciones de flujo provocadas por la presencia de la EA. Este artículo de revisión describe de manera detallada el valor agregado obtenido en la práctica clínica con el uso de las técnicas de medición de flujo bidimensionales, así como las técnicas avanzadas de flujo en 4D para la cuantificación y evaluación de la severidad de la EA. De igual modo, se discuten las ventajas y desventajas de los parámetros clínicos comúnmente utilizados para la estratificación de la severidad de la EA. Además, incluye una introducción a nuevos y prometedores índices hemodinámicos, discute su utilidad para la identificación de potenciales complicaciones y de progresión de la EA in vivo. 

Palabras clave: estenosis aórtica, resonancia magnética cardiovascular, cuantificación de flujo.


Aortic stenosis (AS) is a multifaceted disease, with a prevalence of 2-3% in populations older than 80 years old [1], and which involves atherosclerotic- and elastocalcinosis-like processes affecting the aortic valve opening (via narrowing or obstruction) and motility [2], [3]. Transthoracic echocardiography (TTE) is the primary imaging technique for the assessment of AS severity as indicated in the ACC/AHA/ESC guidelines [4]-[6]. However, in patients with inadequate TTE quality or discordant results, cardiovascular magnetic resonance (CMR) can be used to corroborate the AS severity, to assess ventricular function and volume, and to estimate myocardial fibrosis/hypertrophy [7]-[9]. In particular, flow imaging by ECG-gated 2D phase contrast (PC) CMR offers the opportunity to quantify flow-derived parameters with higher reproducibility than TTE [8]. It has also been shown that CMR is more diagnostic than 2D echocardiography in determining the presence of congenital defects, such as bicuspid aortic valve [10].

CMR flow imaging techniques and analytic tools have rapidly evolved in recent years and today permit a comprehensive assessment of changes in aortic hemodynamics associated with aortic valve disease. This paper reviews the current clinical metrics using 2D flow velocity measurements, introduces emerging 2D flow and advanced 4D flow hemodynamics markers, and future perspectives in the assessment of AS severity. 

Standard of care: AS assessment

Cardiac auscultation remains the most widely used method of screening by evaluating the cardiac murmurs related to valvular heart diseases. The production of acoustic noise related to murmurs is due to 3 main factors: (1) high blood flow rate through normal or abnormal orifices; (2) forward flow through a narrowed or irregular orifice into a dilated vessel or chamber; (3) backward or regurgitant flow through an incompetent valve. Heart murmurs are an important clue for the diagnosis of AS in asymptomatic patients. In particular midsystolic (systolic ejection) murmurs, often crescendo-decrecendo, occurs when the blood is ejected across the aortic valve. An increase of intensity depends in part on the velocity of the blood flow across a narrowed area, thus it may be a sign of stenosis. On detection, the imaging modality of choice is TTE which can evaluate cardiac morphology, dimension, volumes, function, and the severity of the valve obstruction. In comparison to other imaging modalities, TTE is fast, cheap, portable, and widely available in the clinic. In general, the main parameters of interest are peak transvalvular velocity, peak/mean pressure gradient (PG), and aortic valve effective orifice area (EOA) [4], [6], [11], Table 1. Nonetheless, a comprehensive evaluation of valve morphology is essential to fully characterize the presence and/or severity of valve stenosis and to understand disease progression. This includes the documentation of the presence of congenital anomalies, degree of leaflet thickening and calcification, presence and extent of commissural fusion, and any fibrocalcific remodelling of the subvalvular apparatus. Therapeutic decisions are also guided by assessment of left ventricular function which is accomplished with the measurement of: systolic and diastolic diameters, wall thickening and motion, ejection fraction, masse, and geometrical remodelling. 

The utility of TTE must be balanced with the known challenges of this imaging powerful modality. For example, care must be exercised when measuring morphologic and hemodynamic parameters. Morphological measurements are limited by acoustical windows and inappropriate transducer alignment, such as for valve planimetry. These limitations are also especially important in the left ventricular outflow tract (LVOT) where the lumen dimensions are necessary for EOA computation [3], [12]. Additional care must be exercised for the measurement of LVOT and transvalvular hemodynamic parameters which are also affected by transducer position and acoustic window [4], [6], [13]-[15]. Since velocity measurements are used to estimate pressure gradients using the simplified Bernoulli equation (4 × V 2) [4], any error in the velocity measurement is propagated by the square when calculating pressure gradient. Moreover, the clinical measurement of EOA has a high variability with TTE for each of the three measurements (LVOT area, LVOT and transvalvular velocity-time integrals) required for its estimation [4]-[6]. Studies have shown that in experienced laboratories LVOT and transvalvular velocity measurements have a very low intra- and inter-observer variability (3-4%) [4]; however, LVOT dimensional variability is higher (5-8%) and often requires corroboration by other imaging modalities. Finally, stroke volume (SV) measurement in the LVOT assumes laminar flow and a flat velocity profile, which is often not the case in patients with AS. As result, SV measurement may be under- or over-estimated in such scenarios [4], [6], [13]-[15]. 

Keeping these measurement challenges in mind, EOA assessment by TTE may be not feasible in up to 20-30% of patients due to the described limitations. Furthermore, often there are discrepancies between EOA severity and pressure gradient [16], [17]. Thus, the care provider should be aware of the intrinsic limitations and uncertainty of the technique regarding stenosis severity and thus therapeutic management strategies. In addition, if TTE measurements are not feasible or are discordant, it is important to confirm the stenosis severity with other, ideally non-invasive, non-ionizing, diagnostic modalities. For this reason, CMR has become an alternative imaging technique to corroborate TTE measurements and may provide additional information concerning stenosis severity. 

Role of CMR for the assessment of aortic valve stenosis

For the assessment of AS, CMR offers a range of different pulse sequences. For example, steady-state free precession (SSFP) allows for the visualization of valve/ventricle anatomy and motion (Fig. 1). Turbo spin echo (T1 weighted, T2 weighted, fat saturation) and inversion      

Table 1. Current and emerging flow-derived parameters in the assessment of aortic stenosis.
Parameter Criteria for Utility and Limitations Image
Severity advantage modality
Valvular obstruction
Peak jet velocity > 4 m/s Easy to measure Highly flow TTE, CMR
Low inter/intra- dependant
observer variability Overestimates LV
High specificity energy loss in
patients with
small aortas
May underestimate
stenosis severity in
low-flow conditions
Mean pressure > 40 mmHg Same as peak Same as peak TTE, CMR
gradient jet velocity jet velocity
Valve jet angle/ NA Same as peak Same as peak CMR
displacement * jet velocity jet velocity
Reflects stenosis
severity and LV
Effective orifice 1 cm2 Less flow dependant Susceptible to TTE, CMR
area (EOA) than pressure gradientmeasurement errors
or peak velocity using continuity
May under/over-
estimate stenosis
severity in
patients with
hypertension, and
low-flow states
EOAi= 0.6 cm2/m2 Represent intrinsic May overestimate TTE, CMR
EOA/BSA severity of valve stenosis severity
obstruction in obese patients
EOA kinetic * NA Same as EOA.  Same as EOA TTE, CMR
Opening and closing Needs high
slopes characterize temporal
stenosis severity resolution
and its effect
on LV function
Vorticity NA Quantify rotational May be noise CMR
magnitude * flow magnitude due to velocity
Characterize shear derivative
layer separation computation
regions Needs high spatial
AAo: Ascending aorta surface; BSA: Body surface area; CMR: Cardiovascular magnetic resonance; CO: Cardiac output; DBP: Diastolic blood pressure; LV: left ventricle; MBP: Mean blood pressure; NA: Not Available; SBP: Systolic blood pressure; SV i: Stroke volume indexed to BSA; TTE: Transthoracic echocardiography. Included in international guidelines for the aortic stenosis assessment; * emerging parameter using cardiovascular magnetic resonance.
Energy loss index 0.5 - 0.6 Less flow dependant than Same as EOA TTE,
(ELI) = cm2/m2 gradient or peak velocity CMR
[(EOA × AAo)/ Considers pressure recovery
(EOA + AAo)]∕BSA and is similar to EOA
measured by catheter
Reflects LV energy loss
caused by stenosis
Should be measured in
patients with small aortas
Stroke work loss > 25% Less dependent May underestimate TTE,
(SWL) = than gradient stenosis severity and CMR
100 × Pmean or peak velocity LV energy loss in
∕SBP + ΔPmean) patients with hypertension
Turbulent kinetic NA Similar to ELI May be affected by partial CMR
energy * Local measurement volume velocity
of irreversible measurements
turbulent energy Needs balanced 4D
dissipation flow measurements
Viscous energy NA Quantifies viscous energetic May be affected by CMR
loss * dissipation due to stenosis velocity derivative
severity. Independent of computation
pressure recovery
Vascular load
Systemic arterial 0.6 ml. Most frequent cause of Susceptible to TTE,
compliance (SAC) = mmHg-1 increased arterial load measurement errors CMR
SVi/(SBP-DBP) .m-2 Can unmask hypertension
Systemic vascular > 2,000 in patients pseudo- Susceptible to TTE,
resistance (SVR) = normalized blood pressure measurement errors CMR
Global hemodynamic load
Valvulo arterial > 4.5 mmHg. Represents global Susceptible to TTE,
impedance (Zva) = ml-1.m2 (valvular+arterial) measurement errors CMR
(SBP + ΔPmean)∕SV i load imposed on LV Does not
May be superior differentiate load
to predict occurrence of contribution
symptoms and events (valvular vs. arterial)
Flow pattern
Helicity * NA Characterize complex Typically visually CMR
rotation flow patters quantified
Vortical feature * NA Visualize structural Same as vorticity CMR
flow organization magnitude
Wall shear stress * NA Quantifies the friction May be affected by CMR
force of the flowing velocity derivative
blood at the computation. Needs
arterial wall high spatial resolution


Fig. 1. Fluid mechanics of the aortic valve. Schematic representation of the left ventricle, aortic valve and ascending aorta. The blood flows from the left ventricular outflow tract (LVOT) through the aortic valve (AVA indicates anatomic aortic valve area) and is spatially accelerated to the vena contracta (VC) position, where the blood then decelerates and diverges within the ascending aorta (AAo). The cross-sectional area of VC corresponds to the valve effective orifice area (EOA). Magnetic resonance velocity measurements at VC (10 mm from the aortic valve) are indicated in a blue square. Notice that AVA » EOA.

recovery techniques are used to characterize valve masses [18], [19]. Phase-contrast (PC) MRI is employed to quantify blood flow velocity in flexibly selectable 2D imaging planes above, and below the aortic valve. The primary use of CMR flow velocity measurements is to corroborate the standard measures obtained by TTE such as peak velocity, transvalvular peak/mean PG, and valve EOA [4], [7]-[9], Table 1. 

PC-MRI relies on the intrinsic motion sensitivity of MRI which can be used to image vessels as that employed by phase contrast MR-angiography, but also to quantify blood flow velocities. Using appropriate velocity encoding gradients, flow or tissue motion dependant phase effects can be used to measure two datasets with different velocity dependant signal phases at otherwise identical acquisition parameters. Subtraction of the two resulting phase images allows the quantitative assessment of the underlying blood velocities (Fig. 2). The velocity encoding gradients can be applied along arbitrary directions to capture the nature of blood flow in any orientation within the imaging slice. Thus, stationary objects (e.g. static tissue) within the slice have a null net phase and moving objects (e.g. blood flow) have a net phase or phase shift proportional to blood velocity in the measured direction. Measured velocities in the predominant blood flow direction appear bright and flow opposite direction in dark (Fig. 2B). Velocity mapping requires an adequate selection of velocity encoding sensitivity (also termed ‘V enc’) to avoid velocity aliasing (phase shift > 180o). To synchronize phase contrast measurements with pulsatile flow, data acquisition is gated to the cardiac cycle and time resolved (CINE) images are collected to depict the dynamics of blood flow during the cardiac cycle. Following data acquisition, PC-MRI generates 2 set of images: time-resolved magnitude and phase difference (velocity) images that depict vessel anatomy and blood flow over the cardiac cycle. Magnitude images are used for anatomic orientation and boundary vessel identification for the quantification of peak/mean velocity and blood flow from the velocity images (Fig. 3). 

Pressure gradients across the aortic valve can be estimated by the simplified Bernoulli equation (4 ×V peak2), where mean PG is obtained by averaging V peak from each time frame over systole. It has been demonstrated that PC velocity mapping can accurately measure velocities over 5 m/s selecting adequate velocity encoding sensitivity [20]. However, in clinical practice 2D PC often underestimates PG measured by TTE [7]-[9], [21], due to local signal loss, background noise, velocity aliasing, inadequate plane positioning and temporal resolution [22]-[26]. A promising metric for AS severity assessment is valve EOA [8], [9], [27], [28]. Valve EOA can be estimated by CMR using the continuity equation (EOA=SV/VTIAo) [8], [9], [29], where SV is the stroke volume and VTIAo is the velocity-time integral over valve ejection period downstream of the aortic valve (Fig. 3). The left ventricular SV can be estimated based on multi-slice short axis CINE SSFP images covering the entire left ventricle, which is considered the gold standard [30]. Alternatively, PC MRI in the LVOT can be used by multiplying each pixel velocity and area to estimate the instantaneous flow volume (Fig. 3). It is important to differentiate the valve EOA from the anatomic valve area (AVA). The AVA corresponds to the physical opening of aortic valve, often measured by valve planimetry, and EOA corresponds to the hemodynamic opening of aortic valve at vena contracta position where V peak is located (Fig. 1). AVA is usually greater than EOA and they are physically related by the contraction coefficient (CC=EOA/AVA). It should be noted that a similar AVA may have a different EOA. This is relevant for differentiating tricuspid from bicuspid valve hemodynamics and AS severity [31]. 


Fig. 2. Standard 2D CINE PC MRI with one-directional through-plane (Z) velocity encoding. Panel A: A reference and velocity sensitive scan (bipolar encoding gradient) are acquired in direct succession. Magnitude images are calculated by averaging both scans and the subtraction provides phase difference images that contain quantitative blood flow velocities, as shown in a 2D slice above and parallel to the aortic valve (AoV), pulmonary artery (PA) and left atrium (LA). Due to time constraints, the MR data cannot be acquired during a single heartbeat, thus velocity data are collected over several cardiac cycles. The measurement is synchronized with the cardiac cycle using an ECG-gated k-space segmented data acquisition. For each heartbeat and time-frame only a subset of N-segments of all required phase-encoding steps are measured. The procedure is repeated until the entire dataset is acquired. The selection of the number of phase-encoding lines (N-segments) determines the temporal resolution (i.e., time to collect data for a single time-frame) and a total scan time of the acquisition. Panel B: The presence of aortic stenosis will require the selection of higher velocity sensitivities (V enc), from 200 to 500 cm/s, for a proper flow measurement. Blood flow velocities in the predominant blood flow direction will appear bright and blood flow velocities in the opposite direction will appear dark. Notice that velocities exceeding V enc range will produce aliasing within the image. 


Fig. 3. CMR image planes used for aortic valve measurements. Panel A: Flow velocity map was acquired at two image planes, one located at left ventricular outflow tract (LVOT) and the second at aortic level (Ao) downstream of the aortic valve plane (reference). Red contours in Ao and LVOT planes define the region of interest (ROI) for flow velocity measurements. Panel B: Measurement of stroke volume (SV) during systole at LVOT, ROI appears red in panel A. The change in instantaneous flow (Q) at the ROI is calculated as follows: Q(t) = average velocity (t) ×ALV OT , where ALV OT is the cross-sectional area of the LVOT. The SV is estimated by the flow-time integral during systole. Panel C: Peak velocity measurement over systole used for aortic velocity-time integral (VTIAo) at Ao, ROI appears red in panel A. The VTIAo is the area under the velocity curve. Both SV and VTIAo are needed for the valve effective orifice area estimation by continuity equation. Ascending aorta: AAo; Left atrium : LA; Left ventricle: LV.

Emerging 2D hemodynamic markers

Previous TTE studies have suggested that valve opening and closing kinetic analysis, i.e. the temporal changes of EOA during systole, can provide incremental prognostic information beyond standard EOA as computed by the continuity equation [32]-[34]. However this analysis is cumbersome, time consuming, and may lead to measurement errors using TTE. Aortic valve PC velocity measurements allow the instantaneous computation of EOA using the time-resolved version of the continuity equation (EOA[t]=Q[t]/VAo-peak[t]), where Q[t] is the instantaneous flow at LVOT and VAo-peak[t] is the instantaneous peak velocity of transvalvular flow [28], [35]. In particular, EOA opening slope has been associated with plasma level of NT pro-brain natriuretic peptide (BNP) which has been shown to be a powerful predictor of outcome in patients with AS [36], [37]. Recently, it has been shown that the estimation of aortic valve EOA by CMR using a vorticity-derived jet shear layer detection (JSLD) method avoids the need for SV measurement and is less variable than other flow-derived EOA approaches [27]. This vorticity-derived method shows the potential usefulness of advanced fluid mechanics parameters in the assessment of AS severity using 2D PC measurements.

A recent study suggests that aortic valve flow jet angle/displacement may provide complementary hemodynamic information in patients with isolated AS severity [38]. Angle and displacement were associated to left ventricular function, geometric remodelling and valvulo-arterial impedance, a powerful marker of AS prognosis. In particular, a Cox risk analysis suggested that valve angle may be closely related to aortic valve replacement event. This parameter may play an important role in valve-related aortic diseases; further details are presented in the 4D flow section.

It should be noted that standard 2D PC techniques provide one-directional “through plane” velocity encoding measurements, the image quality may be degraded by noise and/or due to inadequate selection of the V enc.V enc and signal-to-noise ratio (SNR) in the corresponding magnitude images are inversely related (V noise V enc∕SNR). It is recommended to keep a V enc as low as possible to optimize velocity noise and improve image quality but above the maximal expected velocity to avoid aliasing [39], [40]. 

Advanced 4D flow hemodynamic markers

One of the major limitations of 2D PC measurements is the need to select of a 2D image plane. Full volumetric 3D coverage with three-directional velocity measurements as accomplished by recently introduced 4D flow MRI techniques can help eliminating these limitations [39], [40]. The acquisition of a 3D data volume in combination with 3-directional velocity encoding requires longer acquisition times (up to 15-20 mins) during free-breathing (Fig. 4). Respiration control strategies such as navigator gating are thus necessary to avoid motion effects. 

4D flow measurements allow a retrospective plane quantification analysis of imaged cardiac structures. Furthermore, the use of advanced visualization tools with 4D flow data facilitates analysis of complex blood flow patterns, such as highly helical flow commonly observed in the presence of valve disease [41]. The volumetric interrogation of 3D blood flow velocities permits the computation of advanced parameters capable of characterizing valve-related flow effects such as vorticity, JSLD, helicity (in the form of localized normalized helicity, LNH), flow angle, wall shear stress (WSS), and energy loss (turbulent and viscous). It has been suggested that complex flow patterns may play a role in endothelial cell signaling and valvular fiber organization by inducing functional changes within the cells and altering the stimulation of other internal structures such as G-protein and kinase receptors or iron channels [42]-[44]. 

Flow helicity is typically assessed using 3D flow visualization strategies such streamlines, flow vectors or time-resolved 3D pathlines (Fig. 5). A more effective and representative analysis may be performed using vortical and LNH features. Vorticity-derived features allow for the quantification of JSLD and LNH parameters, and visualization of complex flow patterns. The direct estimation of EOA using a 4D flow JSLD method is an example of a vorticity-derived parameter (Fig. 5) [45], [46]. A recent study demonstrated that LNH features can visualize plug flow organization in different numerical setups [47], [48]. Similar results can be obtained using 4D flow MRI measurement in the context of valve-related aortic diseases (Fig. 6), allowing a more comprehensive analysis of flow organization due to AS severity. As indicated in the previous section, recent studies have suggested an association between transvalvular jet flow angle and the aortic dilation rate and WSS alteration using 4D flow MRI measurements in patients with bicuspid valves [49], [50]. A recent sub-study of multicenter SEAS study demonstrated that transvalvular energy loss, measured by TTE, may provide independent and additional prognosis information in asymptomatic AS patients rather than conventional TTE measures [51]. In this context, recent studies proposed two methods, turbulent kinetic energy (TKE) dissipation and viscous energy losses (VEL), to compute of energy loss using 4D flow MRI data [52], [53]. Both TKE and VEL may provide further information that traditional energy loss computed with TTE in AS patients. However, larger studies are needed to assess the diagnostic value of both parameters. 


Fig. 4. Schematic of 4D flow of the thoracic aorta. For each time frame, four 3D raw datasets are collected to measure three-directional blood flow velocities (Vx, Vy, Vz) with a reference scan and three velocity-encoded acquisitions. For applications in the aorta or pulmonary systems a typical TR on the order of 5-6 ms, spatial resolution = 2 × 2 × 2 mm3, Venc = 100-150 ms, Segments = 2, parallel imaging with R = 2, navigator efficiency = 50%-80% results in a total scan time of approximately 15-20 minutes with a temporal resolution of 40-50 ms.


Fig. 5. Aortic valve effective orifice area assessment using jet shear layer detection method. Figure shows three different cases (control, moderate and severe aortic stenosis) using valve area estimation with the 4D flow jet shear layer detection (JSLD) method at peak systole. The first column illustrates the aortic flow velocity streamlines at peak systole; the second column shows a 3D lateral view and top valve view of JSLD structure (red iso-surface) computed from 4D flow MRI data at peak systole for a control subject; the third column shows a 3D lateral view and top valve view of JSLD at peak systole for a moderate aortic stenosis patient; the fourth column shows a 3D lateral view and top valve view of JSLD at peak systole for a severe aortic stenosis patient.

Current challenges and future perspectives

Depending on the choice of imaging modality, different limitations and challenges are encountered during the acquisition and analysis of flow for the assessment of AS severity and valve-related diseases. The TTE estimation of most hemodynamic parameters of AS severity (i.e., peak velocity, peak/mean PG, and valve EOA) requires the measurement of SV, which might be subject to several measurement errors, such as image foreshortening or poor image quality of LVOT. Hence, the accurate measurement of flow ismandatory to assess and interpret AS severity parameters. In addition, the standard TTE hemodynamic parameters of AS severity do not take in consideration the influence of pressure recovery phenomenon, the interaction with systemic arterial hypertension, and transvalvular flow rate variability [3]. New TTE parameters, such as the energy loss index (Table 1), have been proposed to consider the pressure recovery that might occurs downstream of the stenotic valve. In particular, patients with moderate to severe AS and small aortas are subject of pressure recovery effect [3], [54]. Patients with AS often have concomitant valve regurgitation, and in both scenarios aortic compliance may be reduced thereby increasing the hemodynamic burden of the LV and the mechanical stress on the aortic valve. An emerging TTE measurement of the arterial load on the LV is the valvulo-arterial impedance, Table 1. It has been proposed that the measurement of aortic compliance and valvulo-arterial impedance by TTE or CMR may better assess the interaction between the ventricular, valvular and arterial factors, and therefore improve the risk stratification in patients with AS [3]. The main pitfall of all hemodynamic parameters of AS severity, regardless the imaging technique, is the dependence on the trasvalvular flow magnitude which may vary patient from patient and/or follow-up visit of the same patient [55]. In particular the PG, the most frequent used parameter, is directly related to the square of the transvalvular velocity magnitude and may lead to a significant underestimation of AS severity in patients with low-flow rate. Patients with “pseudo-severe” AS at low-flow conditions have the tendency to mask the “true” AS severity and represent a challenging population for therapeutic decision [3]. In these cases, stress testing (exercise or dobutamine challenge) can further aid in the stratification of this patient group. Beyond the assessment of the native aortic valve, same hemodynamic parameters (i.e., PG and EOA) can be used to evaluate implanted prosthetic valves (bioprosthetic valves, new generation of transcatheter valves or TAVIs, and mechanical monoleaflet and bileaflet valves) performance using both 2D and 4D flow MRI [56]-[60]. In general, bioprosthetic valves can be scanned as the native valve. However, a specific limitation may exist for mechanical valves and TAVIs, the metallic components of some prosthesis will present a challenge due to the signal void that they can create. Several studies have shown that signal void was found within the valve but no further downstream of the valve jet where the measurements are performed [56]-[60]. In particular, the single use of a single plane downstream of the valve is suggested [60] due to the difficulty of flow measurement at the LVOT with mechanical heart valves. The flow MRI clinical assessment of prosthetic valves, as a complement of TTE follow-up, may be useful for the early detection of malfunction [58] or valve hemodynamic deterioration. 


Fig. 6. Aortic flow helicity. The horizontal panels show a control subject and a patient with bicuspid aortic valve (BAV) and aortic (Ao) dilation (> 4 cm). The first column illustrates the aortic flow velocity streamlines at peak systole; the second column shows 3D localized normalized helicity (LNH, positive spin in red, negative spin in blue) features at peak systole; the third column shows 3D LNH features during systole deceleration and fourth column shows 3D LNH features mid-diastole. Localized, tightly coherent, and temporally long LNH structures (red arrow) illustrate the high complex vortex flow alterations which occur during cardiac cycle.

The emerging 2D and advanced 4D hemodynamic markers obtained using CMR, Table 1, represent an initial effort to overcome TTE parameter limitations, may be used to further understand of AS severity hemodynamics, and enhance risk stratification and clinical decision management of these patients. However, it is important to emphasise that further validation in large prospective studies are needed before implementing them in clinical routine. Regarding 4D flow MRI, new strategies focus on sequence design and hardware development (e.g. short echo time, k-space sampling and parallel imaging [61]-[64]), which have the potential to improve acquisition time and make 4D flow measurement fit clinical schedules and time demands. In addition to long acquisition times, advanced 2D and 4D flow parameters often need time-consuming dedicated post-processing for data analysis. The need of more automated assessment for clinical workflow is crucial. Tools and dedicated software most be developed to standardize the measurement of advanced hemodynamic markers and, more important, the translation of those measurements into longitudinal clinical studies evaluating AS severity outcome. 


Quantitative assessment of aortic stenosis severity using cardiovascular magnetic resonance flow imaging is rapidly progressing and will positively impact clinical practice in the near future. In particular, advanced flow techniques such as 4D flow MRI provide a unique and intuitive flow visualization and quantification of several hemodynamic markers. However, current acquisition time and processing strategies need to be streamlined in order to be incorporated in clinical practice. The data presented in this review provides an overview of the potential for these new flow-derived parameters to further the aid in the assessment of AS severity. Nonetheless, larger prospective studies are needed to evaluate the association of advanced hemodynamic markers with patient outcome in valve-related diseases.


Authors are supported by NIH R01HL115828 grant, AHA 13SDG14360004, and CONACyT 203355 grant.


  1. Lindroos M, Kupari M, Heikkilä J, Tilvis R. “Prevalence of aortic valve abnormalities in the elderly: an echocardiographic study of a random population sample”. J. Am. Coll. Cardiol., 1993;21(5):1220-5.
  2. Rajamannan NM, Evans FJ, Aikawa E, et al. “Calcific aortic valve disease: not simply a degenerative process: A review and agenda for research from the National Heart and Lung and Blood Institute Aortic Stenosis Working Group. Executive summary: Calcific aortic valve disease-2011 update”. Circulation, 2011;124(16):1783-91.
  3. Pibarot P, Dumesnil JG. “Improving assessment of aortic stenosis”. J. Am. Coll. Cardiol., 2012;60(3):169-80.
  4. Bonow RO, Carabello BA, Chatterjee K, et al. “Focused update incorporated into the ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines” (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease): endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. Circulation, 2008;118(15):e523-661.
  5. Vahanian A, Baumgartner H, Bax J, et al. “Guidelines on the management of valvular heart disease: The Task Force on the Management of Valvular Heart Disease of European Society of Cardiology”. Eur. Heart J., 2007;28(2):230-68.
  6. Baumgartner H, Hung J, Bermejo J, Chambers JB, Evangelista A, Griffin BP, Iung B, Otto CM, Pellikka PA, Quiñones M. “Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice”. J. Am. Soc. Echocardiogr., 2009;22(1):1-23.
  7. Cawley PJ, Maki JH, Otto CM. “Cardiovascular magnetic resonance imaging for valvular heart disease: technique and validation”. Circulation, 2009;119(3):468-78.
  8. Garcia J, Kadem L, Larose E, Clavel M-A, Pibarot P. “Comparison between Cardiovascular Magnetic Resonance and Transthoracic Doppler Echocardiography for the Estimation of Effective Orifice Area in Aortic Stenosis”. J. Cardiovasc. Magn. Reson., 2011;13:25.
  9. Caruthers SD, Lin SJ, Brown P, Watkins MP, Williams TA, Lehr KA, Wickline SA. “Practical value of cardiac magnetic resonance imaging for clinical quantification of aortic valve stenosis: comparison with echocardiography”. Circulation, 2003;108(18):2236-43.
  10. Malaisrie SC, Carr J, Mikati I, Rigolin V, Yip BK, Lapin B, McCarthy PM. “Cardiac magnetic resonance imaging is more diagnostic than 2-dimensional echocardiography in determining the presence of bicuspid aortic valve”. J. Thorac. Cardiovasc. Surg., 2012;144(2):370-6.
  11. Vahanian A, Alfieri O, Andreotti F, et al. “Guidelines on the management of valvular heart disease (version 2012)”. Eur. Heart J., 2012;33(19):2451-96.
  12. Pibarot P, Larose E, Dumesnil J. “Imaging of Valvular Heart Disease”. Can. J. Cardiol., 2013;29(3):337-49.
  13. Zhou YQ, Faerestrand S, Matre K. “Velocity distributions in the left ventricular outflow tract in patients with valvular aortic stenosis. Effect on the measurement of aortic valve area by using the continuity equation”. Eur. Heart J., 1995;16(3):383-93.
  14. Leye M, Brochet E, Lepage L, Cueff C, Boutron I, et al. “Size-adjusted left ventricular outflow tract diameter reference values: a safeguard for the evaluation of the severity of aortic stenosis”. J. Am. Soc. Echocardiogr., 2009;22(5):445-51.
  15. Johnson MA, Moss RR, Munt B. “Determining aortic stenosis severity: what to do when measuring left ventricular outflow tract diameter is difficult”. J. Am. Soc. Echocardiogr., 2009;22(5):452-3.
  16. Malouf J, Le Tourneau T, Pellikka P, Sundt TM, Scott C, Schaff H V, Enriquez-Sarano M. “Aortic valve stenosis in community medical practice: Determinants of outcome and implications for aortic valve replacement”. J. Thorac. Cardiovasc. Surg., 2012;144(6):1421-7.
  17. Minners J, Allgeier M, Gohlke-Baerwolf C, Kienzle R-P, Neumann F-J, Jander N. “Inconsistent grading of aortic valve stenosis by current guidelines: haemodynamic studies in patients with apparently normal left ventricular function”. Heart 2010;96(18):1463-8.
  18. Tizon-Marcos H, Larose É, Fuchs C, Mathieu P, Pibarot P, Charest A. “Human valve leaflet tissue characterization in severe aortic stenosis”. Circulation 2010;122:A19916.
  19. Wintersperger BJ, Becker CR, Gulbins H, Knez A, Bruening R, Heuck A, Reiser MF. “Tumors of the cardiac valves: imaging findings in magnetic resonance imaging, electron beam computed tomography, and echocardiography”. Eur. Radiol 2000;10(3):443-449.
  20. Mostbeck GH, Caputo GR, Higgins CB. “MR measurement of blood flow in the cardiovascular system” Am. J. Roentgenol., 1992;159(3):453-61.
  21. Garcia J, Capoulade R, Le Ven F, Gaillard E, Kadem L, Pibarot P, Larose É. “Discrepancies between cardiovascular magnetic resonance and Doppler echocardiography in the measurement of transvalvular gradient in aortic stenosis: the effect of flow vorticity”. J. Cardiovasc. Magn. Reson., 2013;15:84.
  22. Gao JH, Gore JO. “Turbulent flow effects on NMR imaging: measurement of turbulent intensity”. Med. Phys., 1991;18(5):1045-51.
  23. Kilner PJ, Manzara CC, Mohiaddin RH, Pennell DJ, Sutton MG, Firmin DN, Underwood SR, Longmore DB. “Magnetic resonance jet velocity mapping in mitral and aortic valve stenosis”. Circulation, 1993;87(4):1239-48.
  24. Kilner P, Firmin D, Rees R, Martinez J, Pennell D, Mohiaddin R, Underwood S, Longmore D. “Valve and great vessel stenosis: assessment with MR jet velocity mapping”. Radiology, 1991;178(1):229-235.
  25. Kuethe DO, Gao JH. “NMR signal loss from turbulence: models of time dependence compared with data”. Phys. Rev. E, 1995;51(4):3252-3262.
  26. O’Brien KR, Cowan BR, Jain M, Stewart RA, Kerr AJ, Young AA. “MRI phase contrast velocity and flow errors in turbulent stenotic jets”. J. Magn. Reson. Imaging, 2008;28(1):210-8.
  27. Garcia J, Marrufo OR, Rodriguez AO, Larose E, Pibarot P, Kadem L. “Cardiovascular magnetic resonance evaluation of aortic stenosis severity using single plane measurement of effective orifice area”. J. Cardiovasc. Magn. Reson., 2012;14(6):23.
  28. Garcia J, Pibarot P, Capoulade R, Le Ven F, Kadem L, Larose E. “Usefulness of cardiovascular magnetic resonance imaging for the evaluation of valve opening and closing kinetics in aortic stenosis”. Eur. Heart J. Cardiovasc. Imaging, 2013;14(8):819-826.
  29. Yap S-C, van Geuns R-J, Meijboom FJ, Kirschbaum SW, McGhie JS, Simoons ML, Kilner PJ, Roos-Hesselink JW. “A simplified continuity equation approach to the quantification of stenotic bicuspid aortic valves using velocity-encoded cardiovascular magnetic resonance”. J. Cardiovasc. Magn. Reson., 2007;9(6):899-906.
  30. Longmore D, Underwood S, Hounsfield G, et al. “Dimensional accuracy of magnetic resonance in studies of the heart”. Lancet, 1985;1(8442):1360-1362.
  31. Pibarot P, Larose E. “What our eyes see is not necessarily what our heart feels”. Cardiology, 2008;109(2):122-5.
  32. Bermejo J, Antoranz JC, García-Fernandez MA, Moreno MM, Delcan JL. “Flow Dynamics of Stenotic Aortic Valves Doppler Spectrograms”. Am J Cardiol, 2000;85(5):611-617.
  33. Lester SJ, McElhinney DB, Miller JP, Lutz JT, Otto CM, Redberg RF. “Rate of change in aortic valve area during a cardiac cycle can predict the rate of hemodynamic progression of aortic stenosis”. Circulation, 2000;101(16):1947-52.
  34. Bakhtiary F, Dzemali O, Steinseiffer U, Schmitz C, Glasmacher B, Moritz A, Kleine P. “Opening and closing kinematics of fresh and calcified aortic valve prostheses: an in vitro study”. J. Thorac. Cardiovasc. Surg., 2007;134(3):657-62.
  35. Garcia J, Capoulade R, Kadem L, Larose É, Pibarot P. “Valve opening and closing kinematic assessment in patients with aortic stenosis”. In: Proc. Intl. Soc. Mag. Reson. Med. 21. Vol. 101. ; 2013. p. 1328.
  36. Hachicha Z, Dumesnil JG, Pibarot P. “Usefulness of the valvuloarterial impedance to predict adverse outcome in asymptomatic aortic stenosis”. J. Am. Coll. Cardiol., 2009;54(11):1003-11.
  37. Gerber IL, Legget ME, West TM, Richards AM, Stewart RA. “Usefulness of serial measurement of N-terminal pro-brain natriuretic peptide plasma levels in asymptomatic patients with aortic stenosis to predict symptomatic deterioration”. Am. J. Cardiol., 2005;95(7):898-901.
  38. Garcia J, Kadem L, Larose É, Pibarot P. “Assessment of Transvalvular Flow Jet Eccentricity in Aortic Stenosis”. In: Proc Intl Soc Mag Res Med 20. Vol. 4. ; 2012. p. 1186.
  39. Markl M, Kilner PJ, Ebbers T. “Comprehensive 4D velocity mapping of the heart and great vessels by cardiovascular magnetic resonance”. J. Cardiovasc. Magn. Reson., 2011;13:7.
  40. Markl M, Frydrychowicz A, Kozerke S, Hope M, Wieben O. “4D flow MRI”. J. Magn. Reson. Imaging, 2012;36(5):1015-36.
  41. Bissell MM, Hess AT, Biasiolli L, et al. “Aortic dilation in bicuspid aortic valve disease: flow pattern is a major contributor and differs with valve fusion type”. Circ. Cardiovasc. Imaging, 2013;6(4):499-507.
  42. Lehoux S, Tedgui A. “Cellular mechanics and gene expression in blood vessels”. J. Biomech., 2003;36(5):631-643.
  43. Chiu JJ, Chen LJ, Chen CN, Lee PL, Lee CI. “A model for studying the effect of shear stress on interactions between vascular endothelial cells and smooth muscle cells”. J. Biomech., 2004;37(4):531-9.
  44. Balachandran K, Sucosky P, Yoganathan AP. “Hemodynamics and mechanobiology of aortic valve inflammation and calcification”. Int. J. Inflam., 2011;2011:263870.
  45. Garcia J, Barker AJ, Schnell S, Entezari P, Mahadevia R, Pibarot P, Carr JC, Markl M. “4D Flow Jet Shear Layer Detection Method for the Measurement of Effective Orifice Area and Assessment of Aortic Stenosis Severity”. J Cardiovasc Magn Reson. 2013;15(Suppl 1):P241.
  46. Garcia J, Markl M, Schnell S, Entezari P, Mahadevia R, Wu C, Malaisrie SC, Pibarot P, Carr J, Barker AJ. “A novel method for the assessment of valve effective orifice area using 4D flow shear layer detection method in patients with aortic stenosis”. In: Proc. Intl. Soc. Mag. Reson. Med. 21. Vol. 25. ; 2013. p. 1329.
  47. Morbiducci U, Ponzini R, Gallo D, Bignardi C, Rizzo G. “Inflow boundary conditions for image-based computational hemodynamics: Impact of idealized versus measured velocity profiles in the human aorta”. J. Biomech., 2013;46(1):102-109.
  48. Morbiducci U, Ponzini R, Rizzo G, Cadioli M, Esposito A, Montevecchi FM, Redaelli A. “Mechanistic insight into the physiological relevance of helical blood flow in the human aorta: an in vivo study”. Biomech. Model. Mechanobiol., 2011;10(3):339-55.
  49. Hope MD, Hope T a, Crook SES, Ordovas KG, Urbania TH, Alley MT, Higgins CB. “4D flow CMR in assessment of valve-related ascending aortic disease”. JACC. Cardiovasc. Imaging, 2011;4(7):781-7. 
  50. Della Corte A, Bancone C, Conti CA, Votta E, Redaelli A, Del Viscovo L, Cotrufo M. “Restricted cusp motion in right-left type of bicuspid aortic valves: a new risk marker for aortopathy”. J. Thorac. Cardiovasc. Surg., 2012;144(2):360-9, 369.e1.
  51. Bahlmann E, Gerdts E, Cramariuc D, Gohlke-Baerwolf C, Nienaber CA, et al. “Prognostic Value of Energy Loss Index in Asymptomatic Aortic Stenosis”. Circulation, 2013;127(10):1149-1156.
  52. Dyverfeldt P, Hope MD, Tseng EE, Saloner D. “Turbulence mapping: a new CMR approach for assessment of aortic stenosis”. J. Cardiovasc. Magn. Reson., 2013;15(Suppl 1):P110.
  53. Barker AJ, Bandi KC, Garcia J, van Ooij P, Mccarthy P, Carr J, Malaisrie SC, Markl M. “A Direct Calculation of Hemodynamic Energy Loss in the Presence of Abnormal Aortic Flow”. In: Proc. Intl. Soc. Mag. Reson. Med. 21. Vol. 3. ; 2013. p. 62.
  54. Garcia D, Pibarot P, Dumesnil JG, Sakr F, Durand L-G. “Assessment of Aortic Valve Stenosis Severity A New Index Based on the Energy Loss Concept”. Circulation 2000;101(7):765-771.
  55. Pibarot P, Dumesnil JG. “Low-Flow, Low-Gradient Aortic Stenosis With Normal and Depressed Left Ventricular Ejection Fraction”. J. Am. Coll. Cardiol., 2012;60(19):1845-53.
  56. Von Knobelsdorff-Brenkenhoff F, Dieringer MA, Greiser A, Schulz-Menger J. “In vitro assessment of heart valve bioprostheses by cardiovascular magnetic resonance: four-dimensional mapping of flow patterns and orifice area planimetry”. Eur. J. Cardiothorac. Surg., 2011;40(3):736-42.
  57. Kozerke S, Hasenkam JM, Pedersen EM, Boesiger P. “Visualization of flow patterns distal to aortic valve prostheses in humans using a fast approach for cine 3D velocity mapping”. J. Magn. Reson. Imaging, 2001;13(5):690-8.
  58. Garcia J, Marrufo O, Rodriguez A, Pibarot P, Kadem L. “A New Index Measured by Cardiovascular Magnetic Resonance Imaging to Detect Mechanical Heart Valve Malfunction”. Conf Proc IEEE Eng Med Biol Soc., 2011;2011:1347-1350.
  59. Kvitting J-PE, Dyverfeldt P, Sigfridsson A, Franzén S, Wigström L, Bolger AF, Ebbers T. “In vitro assessment of flow patterns and turbulence intensity in prosthetic heart valves using generalized phase-contrast MRI”. J. Magn. Reson. Imaging, 2010;31(5):1075-80.
  60. Le Ven F, Garcia Flores J, Capoulade R, Larose E, Pibarot P. “Comparison of Magnetic Resonance Imaging Versus Echocardiography in the Evaluation of Hemodynamic Function of Bileaflet Mechanical Valves”. Canadian Journal of Cardiology., 2013;29(10):S160.
  61. Thunberg P, Karlsson M, Wigström L. “Accuracy and reproducibility in phase contrast imaging using SENSE”. Magn. Reson. Med., 2003;50(5):1061-8.
  62. Baltes C, Kozerke S, Hansen MS, Pruessmann KP, Tsao J, Boesiger P. “Accelerating cine phase-contrast flow measurements using k-t BLAST and k-t SENSE”. Magn. Reson. Med., 2005;54(6):1430-8.
  63. Jung B, Honal M, Ullmann P, Hennig J, Markl M. “Highly k-t-space-accelerated phase-contrast MRI”. Magn. Reson. Med., 2008;60(5):1169-77.
  64. O’Brien KR, Gabriel RS, Greiser A, Cowan BR, Young AA, Kerr AJ. “Aortic valve stenotic area calculation from phase contrast cardiovascular magnetic resonance: the importance of short echo time”. J. Cardiovasc. Magn. Reson., 2009;11:49.