Secondary Logo

Journal Logo

Perioperative Echocardiography and Cardiovascular Education

Identification of Severe Mitral Stenosis Using Real-Time Three-Dimensional Transesophageal Echocardiography During an Left Ventricular Assist Device Insertion

Fierro, Michael A. MD; Welsby, Ian J. MBBS

Author Information
doi: 10.1213/ANE.0000000000001551

In the article, Identification of Severe Mitral Stenosis Using Real-Time Three-Dimensional Transesophageal Echocardiography During a Left Ventricular Assist Device Insertion,” that appeared in the November 2016 issue on page 1089, contained an error. The title was incorrectly printed as “Three-Dimensional Transthoracic Echocardiography for Evaluation of Mitral Stenosis Identification of Severe Mitral Stenosis Using Real-Time Three-Dimensional Transesophageal Echocardiography During an Left Ventricular Assist Device Insertion.” The correct title is “Identification of Severe Mitral Stenosis Using Real-Time Three-Dimensional Transesophageal Echocardiography During a Left Ventricular Assist Device Insertion.” This has been corrected on the website. The publisher regrets this error.

Anesthesia & Analgesia. 124(1):374, January 2017.


Patient consent was obtained, for publication of a case report with images.


We present a 78-year-old woman with progressive nonischemic cardiomyopathy and a history of rheumatic mitral stenosis (MS) with a baseline cardiac index of 1.5 L/min/m2 who underwent destination left ventricular assist device (LVAD) insertion. The patient had balloon valvuloplasty of the mitral valve 20 years before presentation and had been followed for moderate MS, with a mean transmitral gradient of 9 mm Hg measured by transthoracic echocardiography (TTE) 6 weeks before surgery. On intraoperative 2D transesophageal echocardiography (TEE), she had severely depressed left ventricular systolic function, mild right ventricular dysfunction, severe biatrial enlargement, and moderate mitral regurgitation with Carpentier class IIIa leaflet restriction (Supplemental Digital Content 1, Supplemental Video, The patient also had moderate tricuspid regurgitation, trace pulmonic regurgitation, and no aortic regurgitation. She was pacemaker-dependent with a ventricular pacing at a rate of 80 bpm.

Figure 1.
Figure 1.:
Continuous wave Doppler velocity-time integral tracing of mitral valve inflow. This Doppler tracing was acquired in the midesophageal long-axis view, where beam alignment was parallel to the direction of blood flow. A mean pressure gradient of 7 mm Hg correlates with moderate mitral stenosis.
Figure 2.
Figure 2.:
Measurement of mitral valve pressure half-time (PHT) using a continuous wave Doppler tracing of mitral inflow. This Doppler tracing was acquired in the midesophageal long-axis view, where beam alignment was parallel to the direction of blood flow. A PHT of 150 milliseconds was acquired when averaging the value across multiple measurements, which correlated with a mitral valve area (MVA) of 1.47 cm2.
Figure 3.
Figure 3.:
Measurement of the mitral valve area (MVA) by 3D transesophageal echocardiography (TEE). Panel A, The anterior-posterior view of the mitral valve, as would be seen in a midesophageal 4-chamber TEE view. The blue slice plane, depicted by the blue arrow, projects in parallel to the mitral valve orifice. This line should ideally be orthogonal to the red and green 2 and should be adjusted to image the narrowest point of the mitral valve. The red slice plane, depicted by the red arrow, projects through the center of the mitral valve and is directed toward the ventricular apex. Panel B, The medial-lateral projection of the valve, as would be seen in a midesophageal 2-chamber TEE view. The blue slice plane, depicted by the blue arrow, projects in parallel to the mitral valve orifice. The green slice plane, depicted by the green arrow, projects through the center of the mitral valve and is directed toward the ventricular apex. Panel C, A multiplanar reconstruction of the mitral valve orifice. Planimetry was used to trace the orifice and measure its area as 0.55 cm2 (yellow arrow). Panel D, The 3D image as a compilation of the 3 slice planes (red, green, and blue lines) and demonstrates the spatial orientation of the planes.

Intraoperatively, the patient had a mean transmitral diastolic gradient of 7 mm Hg and a peak gradient of 13 mm Hg (Figure 1). The pressure half-time (PHT) was averaged across several beats to be 150 milliseconds, which correlates with a mitral valve area (MVA) of 1.47 cm2 (Figure 2). Although the Doppler indices correlated with moderate MS, the mitral valve visually appeared to be severely stenosed in both 2D and 3D Zoom (Philips, Andover, MA) examinations (Supplemental Digital Content 1, Supplemental Video,, and Supplemental Digital Content 2, Supplemental Video, To resolve the discrepancy in assessments, we acquired a full-volume loop of the mitral valve and utilized the Philips QLab software’s multiplanar reconstruction (MPR) feature to measure an MVA of 0.55 cm2, which was consistent with severe MS (Figure 3). Although MPR is proprietary to Philips, similarly functioning software allowing for 3D reconstruction and planimetry is available from alternative manufacturers. The patient subsequently underwent mitral valve replacement with a 25-mm stented bioprosthetic valve in addition to HeartMate III LVAD insertion (Thoratec Corporation, Pleasanton, CA) and tricuspid valve repair.


The prebypass TEE for a patient undergoing LVAD insertion should specifically look for intracardiac shunting, aortic insufficiency, ascending aortic atherosclerosis or aneurysm, tricuspid regurgitation, right ventricular function, intracardiac thrombus, and ventricular scar, particularly at the cannulation site.1 MS is also important to recognize because it can impair left ventricular filling, thus limiting LVAD flow. A mean diastolic pressure gradient ≥10 mm Hg across the mitral valve is considered an indication for valve replacement during LVAD insertion.2

Doppler-derived transvalvular pressure gradients are dependent on cardiac output with an inverse relationship between velocity and MVA, leading to overestimation of valve area during low flow states. The presence of aortic insufficiency or decreased left ventricular compliance will independently decrease the PHT, leading to overestimation of MVA. The PHT method has also been found to be ineffective in representing MVA immediately after commissurotomy, especially in older patients.3

Three-dimensional planimetry has been proposed as a measurement tool that can bypass these Doppler limitations (Table). In 2011, Schlosshan et al4 demonstrated that the use of real-time 3D TEE mitral valve planimetry was technically feasible in 95% of patients with rheumatic disease and that the technology had a low interobserver variability. In a TTE study, Doppler overestimated rheumatic MVA by 0.19 ± 0.2 cm2 when compared with 3D planimetry in 87 consecutive patients.6 In a series of 80 patients, MVA measurements using 3D TTE planimetry had better correlation with cardiac catheter-derived MVA using the Gorlin method than those derived from Doppler measurements.7 These studies suggest that 3D planimetry using MPR is an accurate method of assessing MVA and is potentially superior to Doppler modalities when cardiac rhythm or dysfunction limits the accuracy of velocity-derived measurements. MPR can provide details about valve pathology, especially calcification and leaflet fusion, which provides supportive information when it is unclear whether the valve opening is limited because of low cardiac output versus valvular disease. In cases where an MPR measurement is believed to be low because of poor contractility, especially in the absence of significant valvular pathology, a volume challenge or administration of inotrope can be utilized to increase cardiac output to potentially elicit greater valve opening.

Advantages, Disadvantages, and Limitations of Transesophageal Echocardiography-Derived Methods of Assessment of MS

To acquire adequate resolution in 3D imaging, gated acquisition is used to image discrete segments of the full image over multiple sequential beats, which are then merged together to complete the full image loop. This process of “stitching” requires a regular cardiac rhythm and the absence of respirations or electrocautery artifact for proper alignment of the segments occur. If the duration of the cardiac cycle is variable across the acquired images, such as in patients with atrial fibrillation, which is common in advanced MS, significant stitching artifact will be present and cause the individual segments to appear dyssynchronous. An alternative in such patients is the utilization of the Philips high-volume rate 3D mode or a similar modality from an alternative manufacturer, where high temporal resolution (or high frame rate image) can be achieved from a single-loop image, at the expense of decreased lateral resolution, which causes a more granular image appearance.

Philips MPR (Figure 3) is a specific type of 3D processing that combines image reconstruction and multiplanar cropping to identify the narrowest orifice of a valve. To perform MPR of the mitral valve, first a 4-beat or greater midesophageal 4-chamber view full-volume loop must be acquired (to have adequate temporal resolution). The loop is then processed into the individual frames that compose the cardiac cycle, and the user is presented with adjustable still images of orthogonal planes, which are correlated with the electrocardiogram (ECG) tracing. Using visual inspection and the ECG to provide reference, the frame in which maximal diastolic valve opening occurs can be identified by scrolling through the sequential frames.

Once the appropriate frame is selected, it will be divided into 3 quadrants with adjustable lines or slice planes. Adjusting a line in 1 quadrant will affect the images in the other 2 quadrants. The green and red lines should be placed in the middle of the mitral orifice and rotated to align with the flow across the mitral valve. Then, the blue line should be placed orthogonal to the other 2 lines in the red and green quadrants, at the narrowest point of the mitral valve. The green (Figure 3, panel A) and red (Figure 3, panel B) quadrants are analogous to the midesophageal 4-chamber (0°) and 2-chamber (90°) TEE views, respectively. Adjustment of the blue line in either quadrant will automatically refresh the view in the orthogonal quadrant. A processed, en face cross-sectional view of the mitral valve is then displayed in the blue quadrant (Figure 3, panel C), allowing the orifice to be measured by either automatic or manual planimetry. The precision of this method is superior to direct planimetry of a 2D or uncropped 3D image as MPR facilitates measurement of the valve area during the temporal period of maximal opening with spatial alignment along the narrowest axis at the valve leaflets, whereas direct planimetry cannot reliably identify the narrowest orifice of the valve.

This case highlights the limitations of utilization of Doppler indices to measure MVA in a patient with severely depressed cardiac output and rheumatic mitral disease and demonstrates the utility of 3D MPR planimetry as a less flow-dependent method of valve area measurement.

Clinician’s Key Teaching Points

By Kent H. Rehfeldt, MD, Andre-Stephane Lambert, MD, Nikolaos J. Skubas, MD

  • A comprehensive intraoperative transesophageal echocardiography examination should be performed before insertion of a left ventricular assist device (LVAD) to rule out intracardiac shunts and masses, characterize aortic pathology at the site of the outflow cannula, evaluate right ventricular systolic function, and quantify aortic and tricuspid regurgitation.
  • Although relatively uncommon in this patient group, significant mitral stenosis (MS) can restrict inflow to the LVAD. A mean mitral diastolic pressure gradient >10 mm Hg indicates severe MS and concomitant mitral replacement should be considered. However, in low flow states, the severity of MS may be underestimated. The calculation of mitral valve area by pressure half-time may similarly underestimate MS severity in the presence of aortic regurgitation or decreased left ventricular compliance.
  • In this case, the 2D and 3D echocardiographic appearance of the mitral valve was consistent with severe MS, but the mean pressure gradient and pressure half-time were suggestive of moderate MS. To resolve the discrepancy, a full-volume 3D data set of the mitral valve was acquired and off-line planimetry of the diastolic mitral orifice confirmed the presence of severe MS. This prompted mitral valve replacement in addition to LVAD implantation.
  • Full-volume 3D imaging requires electrocardiogram (ECG)-gated data sets from 4 to 6 sequential cardiac cycles to be “stitched”ttogether to provide high temporal and lateral resolution. Manufacturer-provided software allows the user to manipulate the 3D imaging planes to transect the image in any desired orientation. In cases of valve stenosis, the operator can identify the narrowest level of the valve and accurately measure the opening by planimetry. When a regular ECG signal is lacking, a high-volume rate can capture 3D data from a single heartbeat, although at the expense of lateral resolution.


Name: Michael A. Fierro, MD.

Contribution: This author helped design the study and write the manuscript.

Name: Ian J. Welsby, MBBS.

Contribution: This author helped write the manuscript.

This manuscript was handled by: Martin London, MD.


1. Chumnanvej S, Wood MJ, MacGillivray TE, Melo MF. Perioperative echocardiographic examination for ventricular assist device implantation. Anesth Analg. 2007;105:583601.
2. Wang TS, Hernandez AF, Felker GM, Milano CA, Rogers JG, Patel CB. Valvular heart disease in patients supported with left ventricular assist devices. Circ Heart Fail. 2014;7:215222.
3. Messika-Zeitoun D, Meizels A, Cachier A, et al. Echocardiographic evaluation of the mitral valve area before and after percutaneous mitral commissurotomy: the pressure half-time method revisited. J Am Soc Echocardiogr. 2005;18:14091414.
4. Schlosshan D, Aggarwal G, Mathur G, Allan R, Cranney G. Real-time 3D transesophageal echocardiography for the evaluation of rheumatic mitral stenosis. JACC Cardiovasc Imaging. 2011;4:580588.
5. Baumgartner H, Hung J, Bermejo J, et al.; American Society of Echocardiography; European Association of Echocardiography. Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice. J Am Soc Echocardiogr. 2009;22:123.
    6. Min SY, Song JM, Kim YJ, et al. Discrepancy between mitral valve areas measured by two-dimensional planimetry and three-dimensional transoesophageal echocardiography in patients with mitral stenosis. Heart. 2013;99:253258.
    7. Zamorano J, Cordeiro P, Sugeng L, et al. Real-time three-dimensional echocardiography for rheumatic mitral valve stenosis evaluation: an accurate and novel approach. J Am Coll Cardiol. 2004;43:20912096.

    Supplemental Digital Content

    Copyright © 2016 International Anesthesia Research Society