Cardiovascular magnetic resonance evaluation of aortic stenosis severity using single plane measurement of effective orifice area
- 6.7k Downloads
Transthoracic echocardiography (TTE) is the standard method for the evaluation of the severity of aortic stenosis (AS). Valve effective orifice area (EOA) measured by the continuity equation is one of the most frequently used stenotic indices. However, TTE measurement of aortic valve EOA is not feasible or not reliable in a significant proportion of patients. Cardiovascular magnetic resonance (CMR) has emerged as a non-invasive alternative to evaluate EOA using velocity measurements. The objectives of this study were: 1) to validate a new CMR method using jet shear layer detection (JSLD) based on acoustical source term (AST) concept to estimate the valve EOA; 2) to introduce a simplified JSLD method not requiring vorticity field derivation.
Methods and results
We performed an in vitro study where EOA was measured by CMR in 4 fixed stenoses (EOA = 0.48, 1.00, 1.38 and 2.11 cm2) under the same steady flow conditions (4-20 L/min). The in vivo study included eight (8) healthy subjects and 37 patients with mild to severe AS (0.72 cm2 ≤ EOA ≤ 1.71 cm2). All subjects underwent TTE and CMR examinations. EOA was determinated by TTE with the use of continuity equation method (TTECONT). For CMR estimation of EOA, we used 3 methods: 1) Continuity equation (CMRCONT); 2) Shear layer detection (CMRJSLD), which was computed from the velocity field of a single CMR velocity profile at the peak systolic phase; 3) Single plane velocity truncation (CMRSPVT), which is a simplified version of CMRJSLD method. There was a good agreement between the EOAs obtained in vitro by the different CMR methods and the EOA predicted from the potential flow theory. In the in vivo study, there was good correlation and concordance between the EOA measured by the TTECONT method versus those measured by each of the CMR methods: CMRCONT (r = 0.88), CMRJSLD (r = 0.93) and CMRSPVT (r = 0.93). The intra- and inter- observer variability of EOA measurements was 5 ± 5% and 9 ± 5% for TTECONT, 2 ± 1% and 7 ± 5% for CMRCONT, 7 ± 5% and 8 ± 7% for CMRJSLD, 1 ± 2% and 3 ± 2% for CMRSPVT. When repeating image acquisition, reproducibility of measurements was 10 ± 8% and 12 ± 5% for TTECONT, 9 ± 9% and 8 ± 8% for CMRCONT, 6 ± 5% and 7 ± 4% for CMRJSLD and 3 ± 2% and 2 ± 2% for CMRSPVT.
There was an excellent agreement between the EOA estimated by the CMRJSLD or CMRSPVT methods and: 1) the theoretical EOA in vitro, and 2) the TTECONT EOA in vivo. The CMRSPVT method was superior to the TTE and other CMR methods in terms of measurement variability. The novel CMR-based methods proposed in this study may be helpful to corroborate stenosis severity in patients for whom Doppler-echocardiography exam is inconclusive.
KeywordsParticle Image Velocimetry Cardiovascular Magnetic Resonance Aortic Stenosis Effective Orifice Area Potential Flow Theory
Transthoracic echocardiography (TTE) is the standard method for the evaluation of the severity of aortic stenosis (AS) . One of the parameters that is most frequently used to assess AS severity is the aortic valve effective orifice area (EOA) determined by the continuity equation method. However, TTE measurements of EOA may not be feasible or reliable in a significant proportion of patients due to patients' characteristics, technical limitations or users' experience [1, 2, 3, 4]. When the Doppler-echocardiographic measurements are not feasible or are discordant, it is important to confirm the stenosis severity with other, ideally non-invasive, diagnostic modalities.
Cardiovascular magnetic resonance (CMR) is a non-invasive, non-ionizing technique, with excellent temporal and spatial resolutions and superior measurement reproducibility. CMR may be used to measure the geometric (i.e. anatomic) orifice area (GOA) of the stenotic valve by planimetry [5, 6, 7]. However, the GOA is inferior to EOA to predict hemodynamic and clinical outcomes and its estimation may be difficult in heavily calcified valves [8, 9]. CMR may be used to measure the EOA via the continuity equation. Several studies have shown that EOA obtained using CMR correlates well with the EOA obtained by TTE [10, 11, 12, 13]. However, in a recent study performed by our group , we found that the resulting concordance between TTE and CMR for the EOA computed using the continuity equation is, in large part, due to the fact that the underestimation of ALVOT by TTE is compensated by an overestimation of VTILVOT. We also discussed the potential variability in EOA values obtained using the continuity equation both by TTE and CMR as a result of the multitude of parameters to be measured. There is thus an important need for the development and validation of new simpler, more reproducible but still highly accurate CMR methods to estimate the EOA in AS patients. In a previous in vitro study, using particle image velocimetry measurements, we have shown that EOA can be directly determined using velocity measurements downstream of the stenosis and the application of acoustical source term concept (AST) . Briefly, the fundamental idea behind this concept is that the flow jet created by the stenotic valve generates acoustic noise and the major sources of this sound generation can be determined by computing the acoustical source term. The acoustical source term is a function of the local velocity and the vorticity (a measure of the rate of rotation of fluid elements). Applied to AS, this means that the shear layer surrounding the orifice jet is a major source of acoustic noise. As a consequence, the limits of the jet-like zone downstream of the orifice, and therefore the EOA, can be determined using the AST maps without requiring the knowledge of the flow rate magnitude. In our previous in vitro study, we used particle image velocimetry, an optical technique that cannot be applied to the human body. Interestingly, it has been demonstrated that particle image velocimetry and phase-contrast velocity measure the same velocity map [15, 16, 17, 18, 19]. We can then hypothesize that the EOA of an AS could be determined using AST maps computed from CMR velocity measurements.
The objectives of this study are: 1) to extend the previous method for the determination of the EOA based on acoustical source term to velocity measurements obtained by CMR (here called Jet Shear Layer Detection method (JSLD)); 2) to introduce a simplified JSLD method not requiring vorticity field derivation. Both of the previously mentioned approaches require only a single velocity measurement (downstream of the AS) to determine the EOA. These new methods were evaluated both in vitro and in vivo. In the in vitro study, the EOAs determined by these new CMR methods were compared to the theoretical EOA predicted using the potential flow theory, whereas, in the in vivo study, they were compared to those obtained by standard TTE and CMR methods based on continuity equation.
In vitro study
The in vitro setup consisted of controllable pump generating steady flow (4 to 20 L/min), a compatible module with CMR magnet and a fluid reservoir. Four fixed circular stenoses (sharp-edge orifices with EOA = 0.48, 1.00, 1.38 and 2.11 cm2, with small aspect-ratios) were tested under the same steady flow conditions. Testing sharp-edge orifices, as models of fixed aortic stenosis, is a realistic approach since two (calcified thickened valve and thin fused valve) among the four more common morphological shapes of aortic stenoses can be represented by sharp-edge orifices . Flow rate was measured with a Transonic flow probe 16A415 (accuracy: ± 4%, on full scale) connected to a T206 Transonic flow meter (Transonic, Ithaca, NY, USA) and was calibrated using a standard flow measuring method. A 65% saline and 35% glycerine (in volume) solution at room temperature was used to mimic viscous proprieties of blood at 37°C . The use of such Newtonian fluid is justified in the context of aortic valve and ascending aorta [22, 23, 24]. A similar approach was used by others [25, 26, 27].
Each orifice was placed at the center of a clinical 3 Tesla magnetic resonance scanner with a dedicated phase-array receiver coil (Achieva, Philips Medical Systems, Best, the Netherlands). An ECG patient simulator (model 214B, DNI Nevada Inc, USA) was used to synchronize scanner gating. A standard examination was performed by initial acquisition of images in long-axis and short-axis planes for planning. Phase-contrast retrospective examination was performed in short-axis planes 12 mm upstream and 10 mm downstream of to the orifice plane. Imaging parameters consisted of: TR/TE of 17.99/3.97 ms, flip angle 15°, 50 phases, pixel spacing 1.25 mm, slice thickness 10 mm, acquisition matrix of 256 × 256 and encoding velocity (2 × maximal velocity).
A custom-made research application was developed using Matlab software (Mathworks, Natick, Ma, USA) to process and analyze in vitro and in vivo images .
In vivo study
Eight (8) healthy subjects and 37 patients with mild to severe AS (0.72 cm2 ≤ EOA ≤ 1.71 cm2) were included in this study. Exclusion criteria were: age < 21 years, LV ejection fraction < 50%, atrial fibrillation, mild mitral or aortic regurgitation, poor TTE imaging quality and standard contra-indications to magnetic resonance imaging. All patients provided written informed consent under the supervision of the institutional review board. AS severity classification followed American College of Cardiology/American Heart Association (ACC/AHA) guidelines : mild (1.5 cm2 < EOA ≤ 2.0 cm2), moderate (1.0 cm2 < EOA ≤ 1.5 cm2) and severe (EOA ≤ 1.0 cm2).
Effective orifice area determination using transthoracic echocardiography
Where SVLVOT is the stroke volume measured in the LVOT, ALVOT is the cross-sectional area of the LVOT calculated assuming a circular shape; and VTILVOT and VTIAo are the velocity-time integrals of the LVOT and transvalvular flow, respectively.
Cardiovascular magnetic resonance
EOA determination using CMR
Effective orifice area using continuity equation
Where SVCMR is the stroke volume derived from CMR velocities measured 12 mm upstream from the aortic valve (Simpson's rule was used to integrate flow during systole, Figure 1B) and VTIAo (Figure 1C) is the velocity-time integral of the peak aortic flow velocity measured 10 mm downstream of the aortic valve during systole.
Effective orifice area by Shear Layer Detection
Effective orifice area using single plane velocity truncation (SPVT) measurement method
To evaluate intra- and inter- observer variability related to image analysis by CMR and TTE, the measurements of EOA, using all methods, were repeated in a subset of 15 studies (11 AS patients and 4 control subjects) by two blinded observers with the use of the same set of TTE and CMR images. To further evaluate the intra- and inter- observer- variability related to image acquisition and analysis by TTE and CMR, 5 AS patients were imaged twice within 4 weeks (including image acquisition and analysis).
Results are expressed as mean ± SD. Paired 2-tailed Student's t-test was used to compare EOA measures. Correlations and agreement between CMR and TTE EOA measurements were assessed with the use of Pearson's correlation and Bland-Altman methods, respectively. Statistics were performed with SPSS 17 (SPSS, Chicago, IL).
In vitro study
Absolute and mean relative error for the determination of the EOA in the in vitro study
Effective Orifice area (cm2)
Absolute error (cm2)
Mean relative error (%)
0.10 ± 0.02
23 ± 5
0.02 ± 0.01
4 ± 3
0.02 ± 0.004
5 ± 1
0.03 ± 0.01
3 ± 1
0.04 ± 0.04
4 ± 4
0.05 ± 0.02
5 ± 2
0.09 ± 0.02
6 ± 2
0.04 ± 0.03
3 ± 2
0.04 ± 0.02
3 ± 1
0.06 ± 0.03
3 ± 2
0.03 ± 0.03
1 ± 1
0.08 ± 0.07
4 ± 3
0.07 ± 0.04
9 ± 9
0.03 ± 0.03
3 ± 3
0.05 ± 0.04
4 ± 2
In vivo study
61 ± 18
Male gender n (%)
Heart rate (bpm)
66 ± 11
76 ± 14
169 ± 10
Body surface area (m2)
1.88 ± 0.21
Body mass index (kg/m2)
26 ± 4
Tricuspid n (%)
Bicuspid n (%)
Indeterminate n (%)
In terms of clinical implications, seventeen (37%) patients had a change in AS severity class when using the EOA determined by CMRCONT instead of TTECONT: three (6%) patients were re-classified in a more severe class and 14 (31%) in a less severe class. When using EOA determined by CMRJSLD: nineteen (42%) patients had a change in AS severity: six (13%) patients were re-classified in a more severe class and 13 (29%) in a less severe class. When using the EOA determined by CMRSPVT: twenty-one (46%) patients had a change in AS severity: eight (18%) patients were re-classified in a more severe class and 13 (29%) in a less severe class. Importantly, the severity was changed from severe to moderate in 2 patients with CMRCONT and in 3 patients with CMRJSLD or CMRSPVT.
The intra- and inter- observer variability of EOA measurements was 5 ± 5% and 9 ± 5% for TTECONT, 2 ± 1% and 7 ± 5% for CMRCONT, 7 ± 5% and 8 ± 7% for CMRJSLD, 1 ± 2% and 3 ± 2% for CMRSPVT. When repeating image acquisition, reproducibility of measurements was 10 ± 8% and 12 ± 5% for TTECONT, 9 ± 9% and 8 ± 8% for CMRCONT, 6 ± 5% and 7 ± 4% for CMRJSLD and 3 ± 2% and 2 ± 2% for CMRSPVT, for observer one and two respectively.
Contemporary clinical evaluation of the AS severity is mainly based on the TTE measurements of valve EOA, which corresponds to the minimal cross-sectional area of the transvalvular flow jet downstream of the aortic valve. However, TTE measurements are sometimes not feasible or might lead to discordant results. In particular, the situation where the EOA measured by TTE is in the severe range (e.g. 0.8 cm2) but the gradient (or other stenotic indices) is in the moderate range (i.e. 30 mmHg) poses a challenge for the treating physician, especially if the patient is symptomatic. This discordance may be due to measurement errors, small body size, or low flow state conditions [33, 34]. Low flow state conditions may occur in the setting of a low LV ejection fraction (LVEF) but also in the context of preserved LVEF. This later condition, named paradoxical low flow AS  occurs in patients with pronounced concentric LV remodelling, small LV cavities and impaired LV filling and is characterized by reduced pump function and thus reduced stroke volume and transvalvular flow rate despite preserved LVEF. In patients with low flow states, the transvalvular gradient, which is highly flow-dependent, may be pseudo-normalized and may thus underestimates the severity of AS.
In these situations where the TTE measurements are not feasible or discordant, it is necessary to use another imaging modality to determine the actual AS severity and to confirm or infirm the results of TTE. This information is crucial for therapeutic decision making.
In the present study, we proposed a new method based on direct determination of the valve EOA from a single velocity measurement downstream of the stenosis using AST jet shear layer detection. The results of this study reveal an excellent agreement between the EOA estimated by this new method and the EOA predicted in vitro by the potential flow theory or measured in vivo by TTE with the continuity equation method. We also proposed a simplified version of the JSLD method, which does not require the computation of the vorticity term (included in the definition of the AST).
The main advantage of these methods is that they are simple and require only one image plane and one measurement to calculate the EOA. This minimal requirement for the determination of the EOA contributes to the reduction of the errors and may, at least in part, explain why they have better inter- and intra- observer variability compared to the other TTE or CMR methods. It is also important to note that Yap et al.  have previously introduced a method requiring a single measurement plane by measuring the stroke volume at the level of the aorta, instead of the LVOT. The main originality and interest of the new methods we are proposing in our paper is that the determination of the EOA only requires a single plane velocity measurement and moreover it does not require measurement of stroke volume.
The main limitations of this study are: i) small number of patients with severe AS; ii) absence of gold-standard reference method for EOA measurement in vivo. The determination of valve EOA can be performed by catheterization using the Gorlin formula. However, this method is invasive and not without risk for the patients . Furthermore, it has important limitations and thus cannot be considered as a gold standard reference method .
Another limitation of this study is the potential effect of aliasing on the EOA determined by all the methods. Aliasing may affect flow measurements and EOA proposed methods leading to a systematic overestimation of EOA. Interestingly, EOA obtained using JSLD method will not be affected by aliasing as long as the velocity profile is not truncated below its inflexion points. This represents an extreme case in clinical practice. Finally, it should be mentioned that some unwrapping algorithms [40, 41, 42, 43] allow the correction of aliasing and it is generally avoided in clinical practice.
There was an excellent agreement between the EOA estimated by the CMRSPVT method and: 1) the theoretical EOA in vitro, and 2) the TTECONT EOA in vivo. Furthermore, the CMRSPVT method was superior to the other TTE or CMR methods in terms of measurement variability. This new simple and non-invasive method may be helpful to corroborate stenosis severity in patients for whom Doppler-echocardiography exam is inconclusive.
This work was supported by grants from the Canadian Institutes of Health Research, Ottawa, Ontario, Canada (MOP #79342), the Natural Sciences and Engineering Research Council of Canada, Ottawa (#343165-07), and the Fondation de l'Institut Universitaire de Cardiologie et de Pneumologie de Québec. Dr. Pibarot is the director of the Canada Research Chair in Valvular Heart Diseases, Canadian Institutes of Health Research,. J. Garcia and O.R. Marrufo are supported by CONACYT (Mexico City, Mexico) with PhD grants. Dr. Larose is a Clinical research scholar of the Fonds de la recherché en santé du Québec. We thank Isabel Fortin, Haïfa Mahjoub, Jocelyn Beauchemin, Romain Capoulade, Marie-Annick Clavel and Marc Amyot for their assistance on this study.
- 1.Bonow RO, et al: 2008 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. Circulation. 2008, 118 (15): e523-661.CrossRefPubMedGoogle Scholar
- 2.Baumgartner H, et al: Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice. Journal of the American Society of Echocardiography: official publication of the American Society of Echocardiography. 2009, 22 (1): 1-23. 10.1016/j.echo.2008.11.029. quiz 101-2CrossRefGoogle Scholar
- 3.Johnson MA, Moss RR, Munt B: Determining aortic stenosis severity: what to do when measuring left ventricular outflow tract diameter is difficult. Journal of the American Society of Echocardiography: official publication of the American Society of Echocardiography. 2009, 22 (5): 452-453. 10.1016/j.echo.2009.03.018.CrossRefGoogle Scholar
- 4.Dumesnil JG, Pibarot P, Carabello B: Paradoxical low flow and/or low gradient severe aortic stenosis despite preserved left ventricular ejection fraction: implications for diagnosis and treatment. European heart journal. 2010, 31 (3): 281-289. 10.1093/eurheartj/ehp361.PubMedCentralCrossRefPubMedGoogle Scholar
- 11.Haghi D, et al: A hybrid approach for quantification of aortic valve stenosis using cardiac magnetic resonance imaging and echocardiography: comparison to right heart catheterization and standard echocardiography. Clinical research in cardiology: official journal of the German Cardiac Society. 2006, 95 (3): 162-7.CrossRefGoogle Scholar
- 12.Yap S-C, et al: A simplified continuity equation approach to the quantification of stenotic bicuspid aortic valves using velocity-encoded cardiovascular magnetic resonance. Journal of cardiovascular magnetic resonance: official journal of the Society for Cardiovascular Magnetic Resonance. 2007, 9 (6): 899-906. 10.1080/10976640701693717.CrossRefGoogle Scholar
- 13.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.". Journal of cardiovascular magnetic resonance: official journal of the Society for Cardiovascular Magnetic Resonance. 2011, 13 (1): 25-10.1186/1532-429X-13-25.CrossRefGoogle Scholar
- 22.Nichols W, O'Rourke M: McDonald's blood flow in arteries: theoretical, experimental and clinical principles. 1998, London: Arnold, 4Google Scholar
- 27.Saxena R, Lemmon J, Ellis J, Yoganathan AP: An in vitro assessment by means of laser Doppler velocimetry of the medtronic advantage bileaflet mechanical heart valve hinge flow1. Journal of Thoracic and Cardiovascular Surgery. 2003, 126 (1): 90-98. 10.1016/S0022-5223(03)00581-6.CrossRefPubMedGoogle Scholar
- 28.Garcia J, Kadem L, Pibarot P: "Silicone Rubber Trileaflet Valve Assessment Using Cardiovascular Magnetic Resonance Imaging". 32nd Annual International Conference of the IEEE-EMBS. 2010, 5169-5172.Google Scholar
- 36.Sondergaard L, Ståhlberg F, Thomsen C, Stensgaard A, Lindvig K, Henriksen O: Accuracy and precision of MR velocity mapping in measurement of stenotic cross-sectional area, flow rate, and pressure gradient. Journal of magnetic resonance imaging: JMRI. 1993, 3 (2): 433-7. 10.1002/jmri.1880030220.CrossRefPubMedGoogle Scholar
- 37.Shandas R, Kwon J, Valdes-Cruz L: Real-time 3-dimensional volumetric ultrasound imaging of the vena contracta for stenotic valves with the use of echocardiographic contrast imaging: in vitro pulsatile flow studies. Journal of the American Society of Echocardiography: official publication of the American Society of Echocardiography. 1999, 12 (7): 541-50. 10.1016/S0894-7317(99)70001-5.CrossRefGoogle Scholar
- 40.Salfity MF, Huntley JM, Graves MJ, Marklund O, Cusack R, Beauregard Da: "Extending the dynamic range of phase contrast magnetic resonance velocity imaging using advanced higher-dimensional phase unwrapping algorithms.,". Journal of the Royal Society, Interface/the Royal Society. 2006, 3 (8): 415-427.PubMedCentralCrossRefPubMedGoogle Scholar
- 42.Zhou K, Zaitsev M, Bao S: Reliable two-dimensional phase unwrapping method using region growing and local linear estimation. Magnetic resonance in medicine: official journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine. 2009, 62 (4): 1085-1090.CrossRefGoogle Scholar
- 43.Yang GZ, Burger P, Kilner PJ, Karwatowski SP, Firmin DN: "Dynamic range extension of cine velocity measurements using motion-registered spatiotemporal phase unwrapping". Journal of magnetic resonance imaging: JMRI. 6 (3): 495-502.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.