Ann Thorac Surg 2004;78:96-102
© 2004 The Society of Thoracic Surgeons
Original article: cardiovascular
Regurgitant jet evaluation using three-dimensional echocardiography and magnetic resonance
Joerg Albers, MDa*,
Tobias Nitschea,
Jan Boese, PhDb,
Raffaele De Simone, MDa,
Ivo Wolfc,
Antje Schroedera,
Christian F. Vahl, MDa
a Department of Cardiac Surgery, University of Heidelberg, Heidelberg, Germany
b Biophysics, Heidelberg, Germany
c Medical and Biological Informatics, German Cancer Research Centre, Heidelberg, Germany
Accepted for publication November 25, 2003.
* Address reprint requests to Dr Albers, Department of Cardiac Surgery, University of Heidelberg, INF 110, 69120 Heidelberg, Germany
e-mail: oerg.albers{at}urz.uni-heidelberg.de
 |
Abstract
|
|---|
BACKGROUND: Three-dimensional assessment of regurgitant jet volume is the prerequisite for stratifying valve insufficiency. However, systematic comparison of three-dimensional methods is lacking. Therefore, we evaluated magnetic resonance imaging and three-dimensional echocardiography experimentally.
METHODS: An insufficiency chamber (22 x 18.5 x 27 cm; ostia 10, 16, and 20 mm; regurgitant volumes 2.3 to 25 mL) within experimental circulation (BioMedicus pump, tubes, pulsatile flow 0.2 to 1.9 L/min) was used for three-dimensional echocardiography (HP Sonos 2500) and magnetic resonance imaging (Siemens Magnetom Vision). Doppler flowmeter served as a gold standard. Segmentation used thresholding and surface integration of velocity vectors. Jet volume was evaluated qualitatively by polynom fitting.
RESULTS: Jet volume calculated by magnetic resonance (r = 0.99, p < 0.0001) and by echocardiography (r = 0.99, p < 0.0001) correlated identically to the gold standard. Jet volume derived from imaging correlated with each other by r = 0.98 (p < 0.0001). Polynom fits indicated a more paraboloid shape of magnetic resonance jet volume.
CONCLUSIONS: Experimentally, three-dimensional echocardiography and magnetic resonance imaging possess identical accuracy for determining regurgitant jet volume. Magnetic resonance imaging seems to provide qualitatively better image data for three-dimensional reconstruction.
|
Introduction
|
|---|
Major determinants of survival and left ventricular (LV) performance after regurgitant valve replacement are volumes: left ventricular end-diastolic volume [1] and volume of the regurgitant jet [2]. Thus, there is substantial clinical need for volumetric approaches in stratifiying patients suffering from valve regurgitation.
It is unfortunate that clinically, widely accessible imaging tools without the need for radiation exposure (like 2D Doppler) can only semiquantitatively assess regurgitant jets. On the other hand, sophisticated technology (like modern magnetic resonance) enables real time four-dimensional visualization of flow [3]. However, the technique is not yet widely accessible due to hardware limitations. Thus, there is a substantial need for choosing imaging modalities capable of volumetry in addition to the potential of clinical applicability.
Velocity encoded cine magnetic resonance imaging (VEC-MRI) was validated in vitro and in vivo for accuracy to measure blood flow velocity [4]. In patients suffering from mitral regurgitation, velocity encoded MRI as a noninvasive modality shows nearly identical results in determining cardiac volumes and function when compared to invasive measures [5]. The sequences needed for VEC-MRI are available on nearly all MRI scanners. Additionally, using phase velocity mapping, two images are obtained: one morphologic and one velocity image enabling integration of the two informations.
Three-dimensional (3D) color Doppler is a relatively young imaging modality developed by our group. It is clinically available, and provides intraoperative applicability [2, 6]. The acquisition of rotational tomograms differs consistently from the "true" tomographic approach used by MRI.
The purpose of the present study was to validate and to compare the accuracy of two clinically important 3D imaging modalities, 3D color Doppler and VEC-cine MR, in determining regurgitant jet volumes at varying experimental flow rates.
|
Material and methods
|
|---|
Experimental circulation
A regurgitant chamber was constructed with the dimensions of a truncated cone (r1 = 22 cm, r2 = 18.5 cm, h = 27 cm) fitting in a head coil routinely used for MRI. At the center point of the chamber, an adapter was placed prepared for different regurgitant orifices. Circular orifice diameters of 10, 16+, and 20 mm were used. Outside the chamber, a linear conduit (length 25 cm, inner diameter 3 cm) was placed leading to the orifice adapter for laminarization purposes (Fig 1).
The material of the chamber, the conduit, and the orifices consisted of Delrin 150, a nonmagnetic polyoxymethylene. As pump unit, a centrifugal pump (BioMedicus, model 540, Medtronic, Inc., Eden Prairie, MN) was used. Tubes consisted of polyvinylchloride with the dimensions 3/8 x 3/32 inches. Pulsatility was achieved by integration of a magnetic valve (self-construction) into the circulation. The valve was triggered by a generator (self-construction) with a 5 V rectangular signal. Cycle times were chosen, resulting in an experimental systole of 300 ms with a total cycle time of 830 ms representing an experimental "heart" rate of 72/min. The fluid used consisted of Ringer solution (Ringer DAB 7, Braun, Melsungen, Germany) and hydroxyethylene starch solution (HAES steril 10%, Fresenius, Bad Homburg, Germany) in a ratio of 1:3. The fluid viscosity was adjusted to 3.75 mPa · s (Reynolds number: 56.36) using a capillary viscosimeter (Schott GmbH, Hofheim, Germany), representing the viscosity of blood at normal body temperature. Total priming volume was 10 L. Contrast enhancement was achieved by the addition of 2% cornstarch particles for echocardiography, or 10 mL gadolinium-diethylenetriamine-pentaacetic acid (DTPA) (Magnevist, Schering AG, Germany) for MRI, respectively. A single channel Doppler flow meter with an ultrasonic flow probe (Transonic Systems Inc., Models T106 and 10C, Ithaca, NY) were integrated into the circulation for validation purposes. The acquired values of flow velocity were integrated over time after the experiments using the software Matlab (The Math Works Inc., Natick, MA). The resulting flowmeter volumes referring to experimental systole were regarded as the gold standard jet volumes. Sixty consecutive jet volumes were calculated by integrating the acquired flow velocity measures acquired from the Doppler flowmeter over time. Ten different flow rates were used ranging from 170 mL/min to 1,823 mL/min while leaving constant the period of the experimental heart cycle. Mean, standard deviation, and standard error were calculated for each flow rate. Accuracy of the measures of the flowmeter was tested by measuring ten times consecutively total volume collected in an external cylinder over time (volume range, 1,000 to 8,000 mL). Digital data acquisition of flow, pressure, and trigger signal was performed using analog to digital (A/D) conversion (DAQCard Al-16E-4, National Instruments Inc., Austin, TX) on a personal computer (Pentium II, 200 MHz) running the software Labview (National Instruments Inc., Austin, TX). For the method evaluation experiment, nine different volume values (range, 3.11 to 25.85 mL) were measured ten times by either imaging method.
View larger version (105K):
[in this window]
[in a new window]
|
Fig 1. Experimental circulation. (A) General setup allowing for pulsatile production of regurgitant volumes into a regurgitant chamber. (B) View into the regurgitant chamber with affixed regurgitant ostium. 1 = centrifugal pump; 2 = magnetic valve triggered by artificial electrocardiogram; 3 = straight conduit leading to regurgitant chamber; 4 = regurgitant chamber; 5 = regurgitant ostium.
|
|
Jet volume determination using magnetic resonance imaging
A whole body MR scanner with a field strength of 1.5 Tesla was used (Magnetom Vision, Siemens, Erlangen, Germany). The spoiled gradient echo sequence (FLASH) used a repetition time (TR) of 24 ms, an echo time (TE) of 5 ms, a flip angle of 40 degrees, and an image matrix of 128 x 128 pixel with a pixel size of 1.875 mm x 1.875 mm. Beginning with the slice directly adjacent to the regurgitant orifice, 12 slices were acquired using a slice thickness of 5 mm. The method of phase-shift velocity mapping was applied. The maximum flow velocity to be measured before aliasing occurs (Venc) was set to 150 cm/s. Temporal resolution was 24 ms. For each image slice, 34 acquisitions were performed. Acquisition time was 25 minutes for one slice position. Quantification of volume flow during experimental systole was performed in two steps: First, segmentation of jet pixels was calculated resulting in an area of pixels. The acquired gray values were converted to velocity values using equation [1] (see Appendix). Second, mean velocities were integrated over time. Multiplication of areas and mean velocities resulted in volume flow. For a given time period –experimental systole – equation [2] was used (see Appendix). All values of MRI jet volumes compared to three-dimensional echocardiography (3DE) and gold standard refer to the first slice beneath jet orifice.
Jet volume determination using 3DE
Three-dimensional echocardiography was performed using the HP Sonos 2500 device (Hewlett-Packard, Andover, MA) and a multiplane transesophageal transducer (5 MHz). Rotational acquisition of Doppler data was performed in steps of 5 degrees. A rotation of 180 degrees resulted in 36 images. The maximum depth of the Doppler beam was 8 cm and the Nyquist limit was adjusted to 117 cm/s. Temporal resolution was 120 ms, therefore one complete experimental heart cycle was sampled by 6 echocardiography frames. Acquisition time was 3 minutes. The positioning of the transesophageal echocardiography (TEE) transducer was fixed directly opposite to the regurgitant orifice at a distance of 10 cm within the regurgitant chamber.
Image processing
For the determination of jet areas in MRIs, classification of jet pixels was performed using an adapted thresholding algorithm. When pixels lying outside the experimental regurgitant jet possessed a gray value higher than the threshold, a connected component analysis was performed to prevent those pixels from being assigned to the jet. For determination of jet volumes by 3DE, the surface integration of velocity vectors (SIVV) method was used [7]. Integration of velocity vectors took place on a spherical surface with its center at the tip of the Doppler transducer; raytracing was performed.
Statistical methods
The velocity values (n = 90) derived from both imaging modalities were compared to the gold standard using a linear model of regression. Pearsons coefficients of correlation were calculated to assess the quality of the linearity assumption. Variables of t testing were determined by 
= 0.05, resulting in significant values when p value is less than . Approximation of the velocity profiles to an ideal paraboloid pattern was quantified by fitting the profile to a mathematical polynom of second order, the value of ideal paraboloid shape being 1.0. Relative symmetry was calculated with ideal symmetry = 1.
|
Results
|
|---|
Variability and accuracy of experimental jet volume
Comparison of flowmeter and external cylinder volumes showed nearly exact correlation (r = 0.99, p < 0.0001) indicating high accuracy. The resulting values of experimental jet volume are listed in Table 1.
Maximum standard deviation was 0.2 mL (range, 0.02 to 0.2; median, 0.1). Thus, the experimental circulation was able to produce jet volumes with high reproducibility. Maximum jet volume tested was 25.05 ± 0.15 mL.
Jet volume determination using MRI
The jet volumes derived from calculations on MRIs correlated with the jet volumes acquired by Doppler flowmeter (gold standard): r = 0.99, p less than 0.0001. Assuming linear correlation, the equation was y = 1.11 x –0.54. Jet volumes were slightly underestimated (Fig 2).
Range of 95% confidence interval (CI) was 1.09 to 1.14. The results of polynom fitting are listed in Table 2.
Range of R2 values was 0.80 to 0.93 (median, 0.90). Range of jet symmetry was 0.70 to 0.97 (median, 0.91).
View larger version (12K):
[in this window]
[in a new window]
|
Fig 2. Comparison of MRI and 3D Echo with flowmeter. (A) MRI: linear correlation. Regurgitant volume was underestimated by 0.54 mL (R2 = 0.99, p < 0.0001). (B) 3DE: linear correlation. Regurgitant volume was overestimated by 0.44 mL (R2 = 0.99, p < 0.0001). (3D Echo = three-dimensional echocardiography; MRI = magnetic resonance imaging.)
|
|
Jet volume determination using 3DE
The jet volumes derived from calculations on 3DE images correlated with the jet volumes acquired by Doppler flowmeter (gold standard): r = 0.99, p less than 0.0001. Assuming linear correlation, the equation was y = 1.07 x + 0.45. Jet volumes were slightly overestimated by the 3DE method for all volumes investigated (Fig 2). Range of 95% CI was 1.05 to 1.10. The results of polynom are listed in Table 2. Range of R2 values was 0.35 to 0.64 (median, 0.45). Range of jet symmetry was 0.49 to 0.97 (median, 0.94).
Three-dimensional representation of jet volumes
The values of flow velocity acquired by the two different imaging modalities were visualized. Both flow profiles of the slice nearest to the regurgitant orifice, and 3D reconstruction of the JV in toto, are depicted in Figure 3.
View larger version (38K):
[in this window]
[in a new window]
|
Fig 3. Three-dimensional visualization. (A) Three-dimensional flow velocity profiles. Peak systolic flow velocity profile is depicted at the slice nearest to regurgitant orifice. Left: MRI; right: 3DE. Regurgitant orifice diameters/jet volume from top to bottom row: 10 mm/3.11 mL, 10 mm/7.94 mL, 20 mm/15.7 mL, 16 mm/18.34 mL, 20 mm/20.66 mL, 20 mm/25.85 mL. (B) Three-dimensional regurgitant jets. Left: MRI; right: 3DE. Three-dimensional reconstructed regurgitant jets are shown at a lateral view with the regurgitant ostium at the left side. Five time frames during experimental systole (top to bottom: 0 ms/62 ms/130 ms/200 ms of systole) are chosen from a 3D movie. JV 25.85 mL. (3DE = three-dimensional echocardiography; JV = jet volume; MRI = magnetic resonance imaging.)
|
|
|
Comment
|
|---|
In the described experimental setting, we found a high accuracy of both imaging modalities in determining regurgitant volumes. When color Doppler techniques became clinically available, Helmcke and colleagues [8] showed that the sensitivity and specificity of 2D color Doppler for the simple detection of mitral regurgitation was 100%. The jet signals were assessed in the left atrium from three different views. This was one of the first steps in assessing regurgitant jets by Doppler echocardiography. However, the two-dimensional approach enabled only semiquantification. Correlation with angiography was feasible for patient groups with mild, moderate, and severe mitral regurgitation. Unfortunately, there was considerable overlap between the severity groups. The maximal jet area assessed by color Doppler flow imaging could predict angiographic grade of mitral regurgitation (r = 0.76). However, there was only weak correlation with regurgitant volume as measured in angiography, and did not predict hemodynamic functional status as measured by thermodilution [9].
In previous studies, our group could show that, using the 3D color Doppler approach, jet volumes correlated with regurgitant volumes (r = 0.90, p < 0.01) and regurgitant fraction (r = 0.76, p < 0.01) even when assessing eccentric jets [2, 10]. In a comparative in vitro study, Simpson and colleagues [11] showed that the turbulence area surrounding a laminar jet core present in stenotic and regurgitant jets, correlated with flow rate (r = 0.98). Additionally, the distance from orifice to subsequent onset of flow relaminarization correlated with flow rate (r = 0.96). In the present study using 3D color Doppler, the superiority of the 3D approach was demonstrated by nearly exact correlation of 3DE-JV with gold standard volume. When using eccentric jets, however, even superiority in accuracy would have increased.
When no additional valvular defects are present, determination of stroke volumes is feasible by volumetric assessment of the cardiac cavities without measuring flow velocity. This approach uses endocardial border tracing on MRIs obtained in end diastole and end systole for both ventricles. The difference in ventricular volume from diastole to systole represents the stroke volume. In mitral regurgitation, the stroke volume difference between the left and right ventricle yields a measure of the regurgitant volume. Underwood and colleagues [12] determined the ratio of LV and RV stroke volumes by radionuclide ventriculography, magnetic resonance volumetry, and cardiac catheterization in 10 patients with isolated mitral regurgitation and 18 patients with isolated aortic regurgitation. For MRI, they used spin echo sequence (TE 24 ms) on a 0.5 Tesla scanner. For isolated regurgitation, magnetic resonance volumetry resulted in accurate measures of ventricular volumes, ejection fraction (EF), and LV to RV stroke volume ratios. In the present study, volumetry of cavities was not performed due to the ex vivo setting. However, in single valve disease volumetry represents a clinical alternative to relatively time consuming flow velocity measurements.
Sechtem and colleagues [13] described a cine MRI approach to valvular regurgitation. They used low TE (12 ms) on a 1.5 Tesla scanner. Areas of low signal intensity within the regurgitant chamber were visualized. Regurgitant fraction (RF) was determined by volumetry in cine MRIs. In 17 patients with mitral regurgitation, RF was 4% (healthy), 12% (mild), 35% (moderate), and 63% (severe regurgitation), and correlated well with RF values determined by Doppler or angiography. However, the distinction of healthy subjects and mild MR was not feasible using cine MRI. Utz and colleagues [14] examined 12 patients by MRI (TE 12 ms) and cardiac catheterization, 14 patients by MRI and Doppler. Cine magnetic resonance imaging highlighted intraventricular blood with increased signal intensity. In valvular regurgitation, an area of decreased signal within the high-signal intracardiac blood pool was noted. It is assumed that the reason for this signal void was turbulence resulting in a rapid dephasing of spins due to very rapid T2 decay. Therefore, another possible method for regurgitation imaging was found. However, in 9 of 36 cases there was signal decrease without valvular regurgitation. Moreover, it has to be stressed that cine MR is extremely sensitive to field inhomogeneity due to the necessary gradient refocused echoes. Theoretically, a shorter TE may permit less time for dephasing of the spins to occur. Suzuki and colleagues [15] demonstrated that variations in echo time and display settings in fact cause differences in the measured area of regurgitant signal voids in cine MRIs. A good correlation was found clinically between the volume of the signal void and the regurgitant volume calculated for cine MRI volume measurements (r = 0.84). The volumes derived from MRI (TE 17 ms) differentiated significantly (p < 0.01) between mild, moderate, and severe lesions [16]. Signal voids are also observed in aortic regurgitation. The volumes of the LV cavity derived from MR correlated with echo volumes (r = 0.92, p > 0.0001) [17]. In the present study, analysis of signal void volumes was not performed, because only a semiquantitation is possible. In addition, many factors may cause variation in signal void volume.
Fujita and colleagues [18] proved the feasibility of velocity encoded cine MR (Venc 250 cm/s, TE 8.7 ms) for measurement of mitral regurgitation. They calculated the difference between left ventricular inflow volume and aortic outflow volume. When the slice for ventricular inflow was selected, 1 cm inwards LV, inflow volume was significantly less (5.4 L/min vs 5.9 L/min, p < 0.01). The measures of MRI derived regurgitant fraction and the semiquantitative estimation by Doppler echocardiography were correlated (r = 0.87). Hundley and colleagues [5] validated clinically the method of velocity encoded cine MR. In 23 patients suffering from mitral regurgitation, they found a high correlation of MRI derived left ventricular volumes and regurgitant fraction (r = 0.96). In the present study, the method of VEC-MRI was used because of the widely accepted validation data existing for in vitro and in vivo settings. Therefore, this method was chosen to be compared with the clinically widely accepted method of 3DE.
Limitations of the study
The use of velocity encoded MRI for intracardiac flow is theoretically limited by the occurrence of complex, turbulent flow as in poststenotic or high-flow pulsatile flow. In a phantom study, it could be demonstrated that improvement of accuracy can then be achieved by reduction of TE, voxel size, and velocity-encoded strength [19]. However, loss of phase signal cannot be totally avoided. Turbulent flow is a major problem, especially for signal void quantification as described above. The flow changes to a velocity-dependent paraboloid, mixed or eddy current pattern [20]. In the present study, the maximum velocity values were below 90 cm/s at a pulsatile flow of 1.4 L/min, representing laminar flow. In addition, no aliasing occurred, and 3D flow profiles showed paraboloid pattern. Future studies of intracardiac flow, in particular at the left ventricular outflow tract level, will have to deal with turbulent patterns. Magnetic resonance imaging velocity vectors were acquired in two dimensions over time, although the method is capable of providing all three vector components. However, the acquisition of multidirectional data is time consuming, and is shown not to be needed for accuracy in the setting presented in this study. For evaluation of turbulent flow patterns, multidirectional vectors have to be acquired.
Magnetic resonance velocity mapping also reveals new insights to intracardiac flow phenomena. A diastolic vortex formation dependent on the middiastolic motion of the anterior mitral leaflet in the human left ventricle was visualized [21]. Vortices may play a beneficial role in terms of energy preservation. In contrast to jet momentum and flow convergence methods, a new control volume approach did not require simplification of the flow field. Additionally, measuring all three components of the velocity, MRI phase velocity encoding was superior to Doppler ultrasound as could be shown in a pig model (TE 12 ms, Venc 20 cm/s). The potential of the control volume method for quantification of mitral regurgitation was investigated. Usually, acceleration causes signal loss or velocity overestimation. However, when choosing a control volume outside the acceleration field, accurate measurements were performed. Because no assumptions about converging flow fields are necessary in the control volume method, measurements were accurate for any shape of regurgitant orifices. Aortic outflow did not affect accuracy [22].
In summary, the presented results are consistent with prior studies comparing VEC-MRI or Doppler echocardiography alone with gold standard. We demonstrate, in this study, identical values for accuracy of 3DE and VEC-MRI in determining experimental regurgitant volumes.
|
Acknowledgments
|
|---|
This work was supported by the Deutsche Forschungsgemeinschaft (DFG), SFB 414 "Computer and Sensor Aided Surgery," project Q2.
|
Appendix
|
|---|
 | (1) |
vel = velocity value (cm/s); Venc = maximum velocity encoded in flow direction; gvold = original grey value of acquired phase-shift image; gvmax = maximum absolute grey value in acquired phase shift image (4096);
 | (2) |
V = volume; P = time period; 
i = mean velocity at time point i; A = segmented jet area; n = number of points of time
|
References
|
|---|
- Crawford M.H., Souchek J., Oprian C.A., et al. Determinants of survival and left ventricular performance after mitral valve replacement. Circulation 1990;81:1173-1181.[Abstract/Free Full Text]
- De Simone R., Glombitza G., Vahl C.F., Albers J., Meinzer H.P., Hagl S. Three-dimensional color Doppler. A new approach for quantitative assessment of mitral regurgitant jets. J Am Soc Echocardiogr 1999;12:173-185.[Medline]
- Guilfoyle D.N., Gibbs P., Ordidge R.J., Mansfield P. Real-time flow measurements using echo-planar imaging. Magn Res Med 1991;18:1-8.[Medline]
- Kondo C., Caputo G.R., Semelka R., Foster E., Shimakawa A., Higgins C.B. Right and left ventricular stroke volume measurements with velocity-encoded cine MR imaging: in vitro and in vivo validation. Am J Roentgenol 1991;157:9-16.[Abstract/Free Full Text]
- Hundley W.G., Hillis L.D., Hamilton C.A., et al. Assessment of coronary arterial restenosis with phase-contrast magnetic resonance imaging measurements of coronary flow reserve. Circulation 2000;101:2375-2381.[Abstract/Free Full Text]
- De Simone R., Glombitza G., Vahl C.F., Albers J., Meinzer H.P., Hagl S. Assessment of mitral regurgitant jets by three-dimensional color Doppler. Ann Thorac Surg 1999;67:494-499.[Abstract/Free Full Text]
- Chew M.S., Brandberg J., Bjarum S., et al. Pediatric cardiac output measurement using surface integration of velocity vectors: an in vivo validation study. Crit Care Med 2000;28:3664-3671.[Medline]
- Helmcke F., Nanda N.C., Hsiung M.C., et al. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175-183.[Abstract/Free Full Text]
- Spain M.G., Smith M.D., Grayburn P.A., et al. Quantitative assessment of mitral regurgitation by Doppler color flow imaging: angiographic and hemodynamic correlations. J Am Coll Cardiol 1989;13:585-590.[Abstract]
- De Simone R., Glombitza G., Vahl C.F., Albers J., Meinzer H.P., Hagl S. Three-dimensional color Doppler: a new approach for quantitative assessment of mitral regurgitant jets. J Am Soc Echocardiogr 1999;12:173-185.
- Simpson I.A., Maciel B.C., Moises V., et al. Cine magnetic resonance imaging and color Doppler flow mapping displays of flow velocity, spatial acceleration, and jet formation: a comparative in vitro study. Am Heart J 1993;126:1165-1174.[Medline]
- Underwood S.R., Klipstein R.H., Firmin D.N., et al. Magnetic resonance assessment of aortic and mitral regurgitation. Br Heart J 1986;56:455-462.[Abstract/Free Full Text]
- Sechtem U., Pflugfelder P.W., Cassidy M.M., et al. Mitral or aortic regurgitation: quantification of regurgitant volumes with cine MR imaging. Radiology 1988;167:425-430.[Abstract/Free Full Text]
- Utz J.A., Herfkens R.J., Heinsimer J.A., Shimikawa A., Glover G., Pelc N. Valvular regurgitation: dynamic MR imaging. Radiology 1988;168:91-94.[Abstract/Free Full Text]
- Suzuki J.I., Caputo G.R., Kondo C., Higgins C.B. Cine MR imaging of valvular heart disease: display and imaging parameters affect the size of the signal void caused by valvular regurgitation. Am J Roentgenol 1990;155:723-727.[Abstract/Free Full Text]
- Wagner S., Auffermann W., Buser P., et al. Diagnostic accuracy and estimation of the severity of valvular regurgitation from the signal void on cine magnetic resonance images. Am Heart J 1989;118:760-767.[Medline]
- Pflugfelder P.W., Landzberg J.S., Cassidy M.M., et al. Comparison of cine MR imaging with Doppler echocardiography for the evaluation of aortic regurgitation. Am J Roentgenol 1989;152:729-735.[Abstract/Free Full Text]
- Fujita N., Chazouilleres A.F., Hartiala J.J., et al. Quantification of mitral regurgitation by velocity-encoded cine nuclear magnetic resonance imaging. J Am Coll Cardiol 1994;23:951-958.[Abstract]
- Stahlberg F., Sondergaard L., Thomsen C., Henriksen O. Quantification of complex flow using MR phase imaging —a study of parameters influencing the phase/velocity relation. Magn Res Imag 1992;10:13-23.
- Stahlberg F., Söndergaard L., Thomsen C. MR flow quantification with cardiovascular applications: a short overview. Acta Paediatr Suppl 1995;410:49-56.[Medline]
- Walker P.G., Oyre S., Pedersen E.M., Houlind K., Guenet F.S.A., Yoganathan A.P. A new control volume method for calculating valvular regurgitation. Circulation 1995;92:579-586.[Abstract/Free Full Text]
- Chatzimavroudis G.P., Oshinski J.N., Pettigrew R.I., Walker P.G., Franch R.H., Yoganathan A.P. Quantification of mitral regurgitation with MR phase-velocity mapping using a control volume method. J Magn Res Med 1998;8:577-582.
Top
Abstract
Introduction
