Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 May 9.
Published in final edited form as: Ultrasound Med Biol. 2008 Feb 6;34(4):647–654. doi: 10.1016/j.ultrasmedbio.2007.08.009

THREE-DIMENSIONAL ULTRASOUND IMAGING MODEL OF MITRAL VALVE REGURGITATION: DESIGN AND EVALUATION

Stephen H Little *, Stephen R Igo †, Marti McCulloch *, Craig J Hartley †, Yukihiko Nosé †, William A Zoghbi *
PMCID: PMC3348787  NIHMSID: NIHMS374576  PMID: 18255217

Abstract

We describe the development of a cardiac flow model and imaging chamber to permit Doppler assessment of complex and dynamic flow events. The model development included the creation of a circulatory loop with variable compliance and resistance; the creation of a secondary regurgitant circuit; and incorporation of an ultrasound imaging chamber to allow two-dimensional (2D) and three-dimensional (3D) Doppler characterization of both simple and complex models of valvular regurgitation. In all, we assessed eight different pulsatile regurgitant volumes through each of four rigid orifices differing in size and shape: 0.15 cm2 circle, 0.4 cm2 circle, 0.35 cm2 slot and 0.4 cm2 arc. The achieved mean (and range) hemodynamic measures were: peak trans-orifice pressure gradient 117 mm Hg (40 to 245 mm Hg), trans-orifice peak Doppler velocity 560 cm/s (307 to 793 cm/s), Doppler time-velocity integral 237 cm (111 to 362 cm), regurgitant volume 43 mL (11 to 84 mL) and orifice area 0.32 cm2 (0.15 to 0.4 cm2). The model was designed to optimize Doppler signal quality while reflecting anatomic structural relationships and flow events. The 2D color Doppler, 3D color Doppler and continuous wave Doppler quality was excellent whether the data were acquired from the imaging window parallel or perpendicular to the long-axis of flow. This model can be easily adapted to mimic other intracardiac flow pathology or assess future Doppler applications.

Keywords: Circulatory loop, Cardiac ultrasound model, Mitral regurgitation, Real-time 3D color Doppler echocardiography

INTRODUCTION

Mitral valve regurgitation (MR) can lead to atrial arrhythmias, pulmonary artery hypertension, congestive heart failure and death. It has recently been recognized that even asymptomatic MR can have a profound impact on life expectancy (Enriquez-Sarano et al. 2005). The decision to proceed with surgical valve repair or replacement is based on an assessment of symptoms and valve regurgitation severity. Although echocardiography is the primary tool to assess MR, the accurate quantification of regurgitation severity is hampered by the lack of a true diagnostic gold standard. Published clinical guidelines recommend use of a synthesis of several two-dimensional (2D) Doppler, color Doppler and spectral Doppler parameters to grade MR severity (Zoghbi et al. 2003).

Real-time hand-held three-dimensional (3D) color Doppler echocardiography (3D-CD) has recently become commercially available. The use of 3D-CD technology permits entirely novel measures of complex intracardiac flows. However, clinical use of these new parameters has not been validated for quantification of valvular regurgitation.

Our objective was to develop an in vitro circulatory loop to model the hemodynamic conditions encountered in patients with MR. Important prior work by others have described nonpulsatile models of valve regurgitation (Simpson et al. 1989; Thomas et al. 1990), models created for optimization and validation of 2D Doppler techniques (Thomas et al. 1988; Vandenburg et al. 1988; Cagniot et al. 1992) or models permitting 3D color Doppler reconstruction from a single imaging window (Li et al. 1999; Guo et al. 2001; Coisne et al. 2002; Sitges et al. 2003). We hereby describe the development of a model of pulsatile regurgitant flow which allows both 2D Doppler and hand-held, real-time 3D color Doppler interrogation of flow events from multiple ultrasound imaging windows.

Essential to the model is an ultrasound (US) imaging chamber to allow interrogation of the regurgitant jet from multiple windows, similar to the clinical situation and the ability to compare 2D Doppler and novel 3D color Doppler technologies to an ultrasonic flow meter gold standard. Specific requirements of the flow model include the capacity for a broad range of easily tailored hemodynamic conditions that closely mimic the clinical situations, as well as optimal 2D and 3D Doppler image quality.

METHODS

Circulatory loop

A pulsatile in vitro circulatory loop was designed to achieve up to 7 L/min forward flow. A schematic representation of the circulatory loop is shown in Fig. 1. To mimic blood viscosity (4 cP) the circulatory fluid was 30% glycerin, 70% water, with 1% corn starch added as ultrasound scattering particles (DeGroot et al. 2000). With a T-junction connection, a variable speed bellows pump (Baylor College of Medicine, Houston, TX, USA) was attached to a plastic connecting tubing, 2.5 cm diameter. Prosthetic mechanical heart valves (Carbo-Medics Inc., Austin, TX, USA) placed proximally and distally to the pump connection ensured unidirectional forward flow. To mimic arterial compliance and provide wave dampening an air/fluid chamber was incorporated into the loop and positioned immediately downstream to the pump. A three-way stopcock (Baylor College of Medicine Fabrication Laboratory, Houston, TX, USA) vented to room air allowed adjustment of the air/fluid ratio within the chamber and permitted the circulatory compliance to be readily increased or decreased. Circulatory loop flow resistance was adjusted using screw-clamps to externally compress flow tubing distal to the regurgitant limb junction. Total forward flow was assessed using a calibrated ultrasonic flow meter (Tran-sonic Systems, Ithaca, NY, USA). Positioned downstream of the compliance chamber, was a large (8L) unpressurized (open) fluid reservoir representing the venous system (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Circulatory loop (A) incorporating a regurgitant limb with ultrasound imaging chamber (B). Arrowheads indicate unidirectional flow.

Regurgitant limb and imaging chamber

Using constant pump volume displacement and frequency, flow volume tailored to experimental need was directed into the regurgitant limb by increasing downstream resistance within the circulatory loop. The regurgitant loop incorporated an imaging chamber composed of two acrylic cylinders partitioned by a divider plate containing a geometric orifice. High fidelity pressure transducers (Merit Medical, South Jordan, UT, USA) were positioned on either side of the divider plate to record peak chamber pressure and trans-orifice pressure gradient. Incorporated into the imaging chamber were ultrasound windows at standard cardiac anatomic position and distance to the flow orifice mimicking the apical, parasternal and pulmonary or transesophageal clinical imaging windows (Fig. 2). The distance from the imaging widows to the regurgitant orifice also mimicked the clinical situation. For example, the distance from the apex of the imaging chamber to the flow orifice was 8 cm, representing the typical distance from the cardiac apex to the mitral valve annulus.

Fig. 2.

Fig. 2

Open in a new tab

Ultrasound imaging chamber. Arrows indicate pulsatile flow into chamber, across the divider plate containing central orifice, then returning to the circulatory loop. Arrow-heads indicate imaging windows.

Standard transthoracic echocardiogram probes have a “footprint” of 18 to 20 mm in largest dimension. The current commercially available real-time 3D color Doppler echocardiogram probe (X4 transducer; Philips Medical Systems, Andover, MA, USA) has a footprint of 25 mm. To capture peak flow velocity, ultrasound windows needed to be large enough to permit free rotation and angulation of the US probe. We positioned three circular (30 mm diameter) US windows as two opposing long-axis and one short-axis orientation representing in the clinical setting the apical, pulmonary and parasternal imaging windows, respectively. During the development process, we identified that the material used to create the imaging windows must provide the least US interference while providing the tensile strength required to withstand intrachamber pressures in excess of 200 mm Hg. After evaluating several materials of varying thickness, we identified that imaging windows made of 3 mm thick silicone rubber provided the best combination of US image quality and minimal distensibility.

To model the variations in clinically encountered mitral regurgitation, it was important that the regurgitant orifice size and shape be changed as readily as the regurgitant volume. To achieve this goal, each regurgitant orifice was machined into a central interchangeable divider plate. The material chosen for the divider plate was important because it needed to be rigid and nondeformable to maintain orifice size and geometry under high flow and pressure, yet thin enough to limit interference of US transmission and avoid significant Doppler image degradation and artifact. Use of a transparent acrylic plate (2 mm thickness) provided the best combination of rigidity and image quality. The orifices tested were designed to assess idealized flow conditions using small simple shapes as well as through areas of increasing size and geometric complexity to more closely model clinical MR. Ultimately, we were able to mount a bio-prosthetic tri-leaflet valve within the divider plate by securing the annulus with sutures and cyanoacrylate glue.

Testing protocol

All experiments were performed at a pump frequency of 60 cycles per min; frequency, however, could be easily varied. Flow into the secondary regurgitant circuit was achieved by increasing flow resistance within the primary circulatory loop. Experimental regurgitant volumes were chosen to approximate the hemodynamic conditions of clinical MR (10 to 80 mL/beat). In all, we assessed eight different pulsatile regurgitant volumes through each of four rigid orifices differing in size and shape: 0.15 cm2 circle, 0.4 cm2 circle, 0.35 cm2 slot (1.6 mm wide × 22 mm long) and 0.4 cm2 arc (width 1.6 mm; radius 14.3 mm from the center of the test housing at a subtended angle of 53.5 degrees).

2D and spectral Doppler data were acquired using a commercial ultrasound scanner (Sonos 7500, Philips Medical Systems, Andover, MA, USA) with 2 to 4 MHz transducer. Without changing flow conditions or scan depth, an ×4 transducer (Philips Medical Systems, Andover, MA, USA) was used to digitally acquire 3D color Doppler data immediately following the 2D Doppler study. The regurgitant bioprosthetic valve was assessed for 2D and 3D Doppler image quality.

For each orifice tested, we recorded peak transorifice pressure gradient (as assessed by high-fidelity pressure transducer), peak continuous wave Doppler velocity, Doppler time-velocity integral (TVI), regurgitant volume and derived orifice area. For the quantitative assessment of regurgitant flow, 2D-color Doppler (2DCD) and continuous wave Doppler data across the model orifice were acquired from an apical-equivalent view, parallel to the regurgitant flow and digitally transferred to an off-line quantitative station (DigiView; Digisonics Inc., Houston, TX, USA). For each test orifice and flow condition, three continuous wave Doppler signals across the regurgitant orifice were optimized and digitally acquired. Measurements were performed off-line, blinded to the flow measurements, regurgitant orifice characteristics and 3D data. As previously reported, the spectral Doppler data were used in conjunction with both 2D color Doppler and novel 3D color Doppler assessments of the proximal flow convergence zone to compare the relative accuracy of either 2D or 3D method to estimate effective regurgitant orifice area and regurgitant volume (Little et al. 2007).

RESULTS

Hemodynamics achieved

For the four geometric regurgitant orifices tested, the mean (and range) achieved hemodynamic measures were: peak trans-orifice pressure gradient 117 mm Hg (40 to 245 mm Hg), regurgitant volume 43 mL (11 to 84 mL) and orifice area 0.32 cm2 (0.15 to 0.4 cm2). As shown in Fig. 3, the high fidelity pressure transducers demonstrated only a mild degree of system noise, which was likely related to the use of rigid inflow and outflow chambers. As expected, the pressure waveforms varied with orifice size and flow rate. Peak flow recorded by ultrasonic probe demonstrated a flow profile similar to that expected in clinical MR.

Fig. 3.

Fig. 3

Open in a new tab

High fidelity transvalvular pressure and flow recordings. Upper panels depict the inflow and outflow pressure difference across a circular orifice (A) and slot-shaped orifice (B). Lower panels depict the peak flow rate for each orifice, respectively.

Spectral doppler

Using either of the two axial imaging windows, the spectral Doppler (continuous wave) signal was qualitatively very good (Fig. 4). Peak velocity spectral “bleeding” was minimized by filter and gain adjustment. For all flow conditions tested, the peak and mean Doppler velocities were easily discernable. The trans-orifice peak Doppler velocity averaged 560 cm/s (307 to 793 cm/s) and Doppler TVI, 237 cm (111 to 362 cm). Typical resting cardiac cycle timing (50% systole) was reflected in the trans-orifice Doppler signal.

Fig. 4.

Fig. 4

Open in a new tab

Continuous wave spectral Doppler recording across a circular flow orifice.

2D color Doppler

Color Doppler depictions of flow events across the regurgitant orifice were qualitatively very good (Fig. 5). In all cases, proximal flow events such as flow convergence, as well as events in the receiving chamber such as color jet propagation and direction could be easily appreciated.

Fig. 5.

Fig. 5

Open in a new tab

Three-dimensional color Doppler depiction of proximal flow convergence, orifice plate and downstream jet. Panel A shows the 2D color Doppler long-axis (sagittal-equivalent), panel B the orthogonal long-axis (coronal-equivalent) and panel C the short-axis (transverse-equivalent). Panel D demonstrates the 3D orifice surface as well as the proximal flow convergence volume and jet volume. Solid arrow indicates the flow convergence.

3D color Doppler

Full volume 3D ultrasound images require that “real-time” data packets be stitched together for rapid reconstruction based on assigned timing within the cardiac cycle. This task is usually achieved by gating the 3D data set to a simultaneous electrocardiogram (ECG). Since our model did not mimic cardiac depolarization, we used the peak of the intrachamber pressure signal to trigger appropriate 3D image acquisition. This mock-ECG worked well with qualitatively very good 3D color Doppler depictions of flow events across the regurgitant orifice (Fig. 5D). The proximal flow convergence zone and color Doppler regurgitant jet characteristics were well demonstrated from long-axis views (Fig. 6A) while flow at or nearest the regurgitant orifice (vena contracta) were best viewed from the short-axis window (Fig. 6B). For the prosthetic valve, its flow events were well demonstrated when viewed from an orientation parallel or perpendicular to the flow axis. In addition, the depiction of 3D color Doppler flow through a volume rendered bioprosthetic valve was very clear (Fig. 7).

Fig. 6.

Fig. 6

Open in a new tab

Three-dimensional color Doppler reconstruction depicting a long-axis (sagittal-equivalent) view of asymmetric color Doppler flow convergence through an arc-shaped regurgitant orifice (A) and a orthogonal long-axis (coronal-equivalent) view demonstrating a discrete vena contracta (with area measurement) through a circular orifice. Broken arrow indicates direction of flow.

Fig. 7.

Fig. 7

Open in a new tab

Bioprosthetic valve 3D reconstruction. Panel A shows the long-axis (sagittal-equivalent) view, panel B the orthogonal long-axis (coronal-equivalent) view, and panel C the short-axis (transverse-equivalent) view. Panel D is the 3D reconstruction view demonstrating three distinct valve struts. Panel E shows 3D color Doppler flow through the 3D rendered bioprosthetic valve. Broken arrow indicates direction of flow.

DISCUSSION

Mild MR does not lead to remodeling of cardiac chambers and has a benign clinical course, whereas severe MR is associated with significant adverse remodeling of cardiac chambers, morbidity and mortality. Although clinically very important, the characterization of regurgitant valve severity is often difficult. In fact, it has been suggested that the quantitative estimation of the severity of MR is “possibly the most elusive calculation for modern cardiac imaging” (Simpson et al. 1996). Fueling the need for an accurate volumetric evaluation of MR are the improvements in surgical valve repair techniques and clinical outcomes. As surgical intervention for MR becomes an option at an earlier stage in the disease natural history, patient selection based on quantitative assessment of MR severity has become increasingly important. Contributing to the difficulty in assessing clinical MR severity is the lack of a true diagnostic gold standard as well as the association between MR severity and the hemodynamic conditions at the time of evaluation.

The objective of this project was to develop an in vitro flow model of MR so that estimates of regurgitation severity, derived from novel 3D-CD measures, could be compared against classic 2D Doppler parameters and a flow meter gold standard, across a wide range of hemo-dynamic flow conditions. The essential components of the flow model development included the creation of a circulatory loop with variable compliance and resistance, the creation of a secondary regurgitant circuit and incorporation of an US imaging chamber to allow 2D and 3D Doppler characterization of both simple and complex models of valvular regurgitation.

The hemodynamic variables were designed to mimic those variables encountered in the clinical assessment of MR severity. The trans-orifice pressure gradient range of 40 to 245 mm Hg represents the clinical spectrum including patients with severe left ventricle (LV) systolic dysfunction (low LV systolic pressure) to those with marked systemic arterial hypertension or aortic stenosis (high LV systolic pressure). Similarly, the modeled regurgitant volume range (11 to 84 mL) and orifice area range (0.15 to 0.4 cm2) reflect the clinical spectrum of mild to severe MR, based on varied leaflet morphology.

The model was designed to optimize Doppler signal quality while reflecting anatomic structural relationships and flow events. The 2D color and continuous wave Doppler quality was excellent whether the data were acquired from the imaging window parallel or perpendicular to the long-axis of flow. The 3D color Doppler and full volume images were excellent. Both the 2D and 3D Doppler images were of better quality than can usually be expected in a clinical study. This superior image quality is likely related to the lack of respiratory motion and overlying tissue, large US windows permitting free probe rotation and angulation, and a regular and relatively slow pump frequency (60 bpm).

One of our primary goals was to permit a head-to-head comparison of classic 2D Doppler and novel 3D Doppler assessments of complex flow such as MR. Unlike a comparison of methods in the clinical setting, these methods were compared in an in vitro setting against a flow meter gold standard for assessment of actual regurgitant volume. Employing the model described here, we have recently reported (Little et al. 2007) that the classic 2D color Doppler method was associated with greater underestimation of regurgitant volume for both asymmetric regurgitant orifice shape and larger regurgitant volume. In contrast, the 3D-CD method demonstrated excellent correlation with the flow meter standard. The 2D Doppler methods were accurate for small symmetrical orifice shapes (circle) with decreasing accuracy in estimating true regurgitant volume as greater regurgitant flow was modeled through more complex orifice shapes (slot, arc). In contrast, the 3D method, which does not incorporate any geometric assumptions of flow events, was able to accurately estimate regurgitant volume across a wide range of flow conditions. This is an important finding that will be tested in the clinical setting, albeit with a less than ideal gold standard.

The model described is designed to mimic the hemodynamic conditions of MR. Since essentially all major components of the system are modular, the model can easily be adapted to mimic other valvular disorders such as mitral stenosis or aortic valve regurgitation. Indeed, we foresee application of this model in the future to assess prosthetic valve function, echocardiography contrast agents and other novel Doppler applications as they become available.

Limitations

As in any in vitro assessment of Doppler technique, the successful translation of our head-to-head comparison of 2D and 3D Doppler applications to the clinical realm is not guaranteed. A model of a complex hemo-dynamic event such as MR can only be as good as the variables considered. We believe that by adjusting regurgitant volume, trans-orifice pressure gradient and flow orifice geometry, we have created a robust model of the important hemodynamic variables. However, we did not incorporate dynamic motion of either the orifice plate (mitral valve annulus) or orifice (mitral leaflet). In some clinical situations such as mitral valve prolapse, this dynamic motion may be important.

SUMMARY

The in vitro model of MR performed as intended, achieving pressure gradients and regurgitant volumes within the clinically encountered range across regurgitant orifices of varying size and shape. We describe the development of an ultrasound imaging chamber to permit Doppler assessment of complex and dynamic flow events. Both 2D and 3D Doppler image quality was excellent and Doppler-derived estimates of flow correlated well against the in vitro flow meter standard. This model can be easily adapted to mimic other intracardiac flow pathology or assess future Doppler applications.

Acknowledgments

The authors thank Mr. Juan Fernandez and Mr. Joe Martinez for fabrication of the ultrasound imaging chamber and regurgitant valve orifices and Mrs. Jo Ann Rabb for her secretarial assistance.

Footnotes

Presented in part at the annual meeting of the Scientific Sessions of the American Heart Association, Dallas, TX, November 13-16, 2005.

REFERENCES

  1. Cagniot A, Cape EG, Walker PG, Yoganathan AP, Levine RA. Effect of heart rate on centerline velocities of pulsatile intracardiac jets: An in vitro study with laser Doppler anemometry and pulsed Doppler ultrasound. J Am Soc Echocardiogr. 1992;5:393–404. doi: 10.1016/s0894-7317(14)80272-1. [DOI] [PubMed] [Google Scholar]
  2. Coisne D, Erwan D, Christiaens L, Blouin P, Allal J, Barraine R. Quantitative assessment of regurgitant flow with total digital three-dimensional reconstruction of color Doppler flow in the convergent region: In vitro validation. J Am Soc Echocardiogr. 2002;15:233–240. doi: 10.1067/mje.2002.117901. [DOI] [PubMed] [Google Scholar]
  3. DeGroot C, Drangova M, Fenster A, Zhu S, Pflugfelder PW, Boughner DR. Evaluation of 3-D colour Doppler Ultrasound for the measurement of proximal isovelocity surface area. Ultrasound Med Biol. 2000;26:989–999. doi: 10.1016/s0301-5629(00)00245-3. [DOI] [PubMed] [Google Scholar]
  4. Enriquez-Sarano M, Avierinos J, Messika-Zeitoun D, Detaint D, Capps M, Nkomo V, Scott C, Schaff HV, Tajik AJ. Quantitative determinants of the outcome of asymptomatic mitral regurgitation. N Engl J Med. 2005;352:875–883. doi: 10.1056/NEJMoa041451. [DOI] [PubMed] [Google Scholar]
  5. Guo Z, Boughner DR, Dietrich JM, Pflugfelder PW, Durand LG, Loew M, Fenster A. Quantitative assessment of in vitro jets based on three-dimensional color Doppler reconstruction. Ultrasound Med Biol. 2001;27:235–243. doi: 10.1016/s0301-5629(00)00337-9. [DOI] [PubMed] [Google Scholar]
  6. Li X, Shiota T, Delabays A, Teien D, Zhou X, Sinclair B, Pandian NG, Sahn DJ. Flow convergence flow rates from three-dimensional reconstruction of color Doppler flow maps for computing transvalvular regurgitant flows without geometric assumptions: An in vitro quantitative flow study. J Am Soc Echocardiogr. 1999;12:1035–1044. doi: 10.1016/s0894-7317(99)70099-4. [DOI] [PubMed] [Google Scholar]
  7. Little S, Igo S, Pirat B, McCulloch M, Hartley C, Nose Y, Zoghbi W. In vitro validation of real-time three-dimensional color Doppler echocardiography for direct measurement of proximal isovelocity surface area in mitral regurgitation. Am J Cardiol. 2007;99:1440–1447. doi: 10.1016/j.amjcard.2006.12.079. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Simpson IA, Valdes-Cruz LM, Sahn DJ, Murillo A, Tamura T, Chun KJ. Doppler color flow mapping of simulated in vitro regurgitant jets: Evaluation of the effects of orifice size and hemodynamic variables. J Am Coll Cardiol. 1989;13:1195–1207. doi: 10.1016/0735-1097(89)90284-2. [DOI] [PubMed] [Google Scholar]
  9. Simpson IA, Shiota T, Gharib M, Sahn DJ. Current status of flow convergence for clinical applications: Is it a leaning tower of “PISA”? J Am Coll Cardiol. 1996;27:504–509. doi: 10.1016/0735-1097(95)00486-6. [DOI] [PubMed] [Google Scholar]
  10. Sitges M, Jones M, Shiota T, Qin JX, Tsujino H, Bauer F, Kim YJ, Agler DA, Cardon LA, Zetts AD, Panza JA, Thomas JD. Real-time three-dimensional color Doppler evaluation of the flow convergence zone for quantification of mitral regurgitation: Validation experimental animal study and initial clinical experience. J Am Soc Echocardiogr. 2003;16:38–45. doi: 10.1067/mje.2003.37. [DOI] [PubMed] [Google Scholar]
  11. Thomas JD, Wilkins GT, Choong CY, Abascal VM, Palacios IF, Block PC, Weyman AE. Inaccuracy of mitral pressure half-time immediately after percutaneous mitral valvotomy. Dependence on transmitral gradient and left atrial and ventricular compliance. Circulation. 1988;78:980–993. doi: 10.1161/01.cir.78.4.980. [DOI] [PubMed] [Google Scholar]
  12. Thomas JD, Liu CM, Flachskampf FA, O'Shea JP, Davidoff R, Weyman AE. Quantification of jet flow by momentum analysis. An in vitro color Doppler flow study. Circulation. 1990;81:247–259. doi: 10.1161/01.cir.81.1.247. [DOI] [PubMed] [Google Scholar]
  13. Vandenberg BF, Dellsperger KC, Chandran KB, Kerber RE. Detection, localization, and quantifification of bioprosthetic mitral valve regurgitation. An in vitro two-dimensional color Doppler flow-mapping study. Circulation. 1988;78:529–538. doi: 10.1161/01.cir.78.3.529. [DOI] [PubMed] [Google Scholar]
  14. Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD, Levine RA, Nihoyannopoulos P, Otto CM, Quinones MA, Rakowski H, Stewart WJ, Waggoner A, Weissman NJ, American Society of Echocardiography Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr. 2003;16:777–802. doi: 10.1016/S0894-7317(03)00335-3. [DOI] [PubMed] [Google Scholar]

RESOURCES