Content uploaded by Feroze Mahmood
Author content
All content in this area was uploaded by Feroze Mahmood on May 03, 2016
Content may be subject to copyright.
REVIEW ARTICLES
Paul G. Barash, MD
Giovanni Landoni, MD
Section Editors
APracticalApproachtoanIntraoperativeThree-DimensionalTransesophageal
Echocardiography Examination
Feroze Mahmood, MD,* Jelliffe Jeganathan, MBBS,* Rabya Saraf, BA,† Sajid Shahul, MD,‡
Madhav Swaminathan, MD,§ G. Burkhard Mackensen, MD, PhD, ǁ Ziyad Knio, BS,† and Robina Matyal, MD*
I
NTRAOPERATIVE THREE-DIMENSIONAL (3D) echocar-
diography has been studied since the mid-1990s,
1–3
but its
clinical use has grown exponentially over the last few years.
4–7
Time-consuming and cumbersome acquisition and reconstructive
3D techniques now have been replaced by “live” 3D imaging.
8–10
Post-acquisition manipulation and quantitative analyses also can
be performed on 3D images acquired using the “live” techni-
que.
11
With simultaneous improvements in image processing
techniques and computational speed, quantitative aspects of 3D
echocardiography are now readily available for clinical use.
12,13
Specifically, the technique of multi-planar reformatting (MPR)
can be used for extracting an infinite number of two-dimensional
(2D) slices for accurate linear measures and chamber quantifica-
tion.
14
Therefore, 3D echocardiography has introduced another
dimension, literally and figuratively, to intraoperative transeso-
phageal echocardiography examination (TEE).
Recognizing the need for a uniform methodology for 3D
imaging, the Am erican Society of Echocardiograp hy recently
published guidelines for 3D imaging for transthorac ic echocardio-
graphic (TTE) and TEE imaging.
7
However, the requirements for
the perioper ativ e arena are differe nt in that it is point-of-care in
nature and time limited due to the dynamic nature of the environ-
ment. Specifically, the quantitative aspects of 3D imaging that can
affect intraoperativ e surgical decision making have not been
elaborated on in detail. It is important to approach intraoperative
3D examination in an objective, methodical,andtime-efficie nt
fashion to maxim ize the diagnostic and therapeutic yield from the
acquired data. Based on patient peculiarities and clinical circum-
stances, 3D imaging modes and techniques often need to be indiv-
idualized for data acquisition. Therefore, drawingontheauthors’
experience and available literature, the authors present such an
approach for this Journal.Theauthorshaveattemptedtoincorpo-
rate the qualitative and quantitativeaspectsof3Dechocardiography
data in a routine intraoperative 2D and 3D TEE examination. In
addition, the authors also have suggested the specificmodalities
(Live vs. R-wave gated acquisition)mostsuitableforeachstructure/
view and the technical aspects of 3D data acquisition, analyses,
archiving, and retrieval in the light of current guidel ine s and
regulations. Details of the logistics, personnel, and limitations of 3D
imaging in the perioperative arena also are presented.
CURRENT STATUS OF THREE-DIMENSIONAL
ECHOCARDIOGRAPHY
There has been a gradual increase in the use of 3D imaging
in the perioperative arena. Specifically, 3D echocardiography has
shown value in assisting intraoperative valve repair/replacement
decision making.
15,16
Using 3D TEE imaging, views of intracardiac
structures have been described that resemble surgical views
obtained after valve exposure.
17–20
This has introduced a level of
uniformity in nomenclature and enhancement in communication
across disciplines. With significantly more “raw” Cartesian coor-
dinate data, complex cardiac structures now can be analyzed
quantitatively without geometric assumptions. Intraoperative 3D
imaging also has shown its value during percutaneous interventions
for structural heart disease and has established itself as a vital
procedural adjunct.
7,21
3D TEE is used to assess suitability and for
guidance, to exclude complications, and establish the success of
percutaneous intervention.
22,23
Information obtained from intra-
operative 3D imaging is an adjunct to a comprehensive 2D TEE
examination and provides supplemental information. Specifically, it
improves spatial orientation and accuracy of linear measures, allows
visualization of simultaneous orthogonal views, and cardiac
chambers can be analyzed volumetrically without geometric
assumptions. 2D imaging provides a broad approach to the cardiac
anatomy, physiology, and identifica tion of structures of interest,
whereas 3D TEE currently is used to acquire specificcomplemen-
tary qualitative and quantitative information fromexaminationof
structure(s) of interest.
PRACTICAL ASPECTS OF THREE-DIMENSIONAL IMAGING
A comprehensive intraoperative 3D TEE examination
is used for acquiring and archiving supplementary 3D
From the *Department of Anesthesia, Critical Care and Pain
Medicine; †Department of Surgery, Division of Cardiac Surgery, Beth
Israel Deaconess Medical Center, Harvard Medical School, Boston,
MA; ‡Department of Anesthesia and Critical Care, University of
Chicago, Chicago, IL; §Department of Anesthesiology, Duke Uni-
versity Medical Center, Duke University, Durham, NC; and ǁDepart-
ment of Anesthesiology and Pain Medicine; University of Washington,
Seattle, WA.
Address reprint requests to Feroze Mahmood, MD, Associate Pro-
fessor of Anesthesia, Harvard Medical School, Bet h Israel Deaconess
Medical Center, 1 Deaconess Road, CC-454, Boston, MA, 02215.
E-mail: fmahmood@bidmc.harvard.edu
© 2016 Elsevier Inc. All rights reserved.
1053-0770/2601-0001$36.00/0
http://dx.doi.org/1 0.1053/j.jvca.2015.10.014
Key words: 3D echocardiography, TEE, transesophageal echocar-
diography, live 3D imaging, practical approach, intraoperative
imaging
470 Journal of Cardiothoracic and Vascular Anesthesia, Vol 30, No 2 (April), 2016: pp 470–490
echocardiographic data for real-time and post-acquisition
qualitative and quantitative analyses. There are logistic and
technical aspects of setting up an intraoperative 3D imaging
service. Logistically, it requires availability of equipment and
personnel who are trained in its use and competent in
acquisition and analyses of acquired 3D images. A 3D imaging
laboratory should have the capability to store the acquired 3D
image for later retrieval and post-acquisition analysis. Techni-
cal aspects of 3D imaging require that there be judicious
patient-specific use to supplement information acquired with
2D imaging to improve diagnostic accuracy and patient care.
LOGISTICS
Equipment
There are numerous commercially available ultrasound
systems that are capable of 3D imaging. Currently, there is
only 1 portable system (CX-50; Philips Healthcare, Andover,
MA) that is equipped to perform 3D imaging. However, 3D
imaging in this system is limited to only “live” modes.
Although ultrasound systems differ in their workflow of raw
ultrasound data, they generally follow the same basic method-
ology of imaging of intracardiac structures. Nonetheless, using
an ultrasound system to its full potential requires equipment
familiarity, significant experience, and expertise. An intra-
operative 3D echocardiography service should possess the
following characteristics: (1) A TEE probe capable of 3D
imaging, (2) an ultrasound system capable of the entire
spectrum of 3D imaging modalities (Fig 1), (3) software for
online basic 3D quantification, (4) capacity to archive 3D data,
(5) ability to retrieve 3D data, and (6) ability for offline
quantification and analyses.
Equipment availability constraints may preclude provision of
3D echocardiography services to all cardiac surgery patients. At
the authors’ institution, 3D TEE is performed for all surgical
(cardiac and noncardiac) and percutaneous cardiac interventions.
Minimally, patients undergoing valvular surgery (repair and
replacement), congenital cardiac surgery, and percutaneous
interventions should undergo 3D TEE examination.
Single Versus Multi-Vendor Ultrasound Service
There are various commercially available ultrasound sys-
tems capable of 3D TEE imaging, with accompanying vendor-
specific offline viewing and archiving capabilities (see Fig 1).
The advantage of a single-vendor laboratory is that it is less
expensive to maintain, with uniformity of workflow of data
acquisition, personnel training, archiving, and retrieval of data.
A multi-vendor laboratory is more expensive to maintain and
requires separate service contracts for equipment and training
for each system. Multi-vendor system laboratories are suited for
teaching facilities because they offer the training advantages for
fellows and flexibility to upgrade systems individually. At the
authors’ institution, to comprehensively train residents and
fellows, the authors’ division possesses ultrasound systems
from multiple vendors.
Personnel
Even though credentialing is determined by the local
healthcare facility and institutional policies, most institutions
require evidence of some form of TEE training/education for
anesthesiologists offering services by cardiac anesthesia teams.
In the authors’ department, intraoperative TEE privileges are
approved by the chairman of the department based on the
recommendations of the director of cardiac anesthesia. Even
though all members are certified, a testamur status in the
advanced perioperative transesophageal echocardiography
examination of the National Board of Echocardiography is a
minimum requirement for requesting these privileges. The
authors’ division also requires yearly continuing medical
education related to echocardiography to maintain privileges
as a cardiac anesthesiologist.
Fig 1. Commercially available three-dimensional ultrasound machines currently in the market. (A) Philips Epiq 7. (B) General Electric E9.
(C) Siemens SC2000.
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE
471
Equipment and Personnel Quality Improvement
It also is important to introduce quality improvement
initiatives for equipment and personnel. For example, ultra-
sound companies often introduce software and hardware
upgrades that are not compatible retroactively, making the
already-acquired data not amenable to retrieval and analysis.
Healthcare institutions sometimes choose to change vendors for
multispecialty universal offline viewing of medical images,
creating issues with analyses of “raw” echocardiographic data
due to vendor incompatibilities. Equipment service and
upgrade can be a significant expense unless they are made
part of the contract at the time of equipment acquisition.
Due to evolving technology, it also is important for the
anesthesiologist to be knowledgeable of the most recent
Fig 3. Illustration depicting the trade-offs in 3D image acquisition. (A) Triad of factors needed for an ideal 3D image. (B) For superior spatial
resolution, the trade-off is a decrease in the sector size and a lower temporal resolution. (C) For better temporal resolution, the trade-off is a
decrease in the se ctor size and a lower spatial resolution. (D) For enhanced sector size, the trade-off is a decrease in spatial and temporal
resolutions.
Fig 2. Image showing the location of ECG synchronization ports in different ultrasound machines. (A) ECG port in Philips iE33. (B) Slave box
located above the monitor to record the ECG. (C) ECG port in Philips Epiq series.
MAHMOOD ET AL472
advances in 3D imaging. At the authors’ institution, weekly
echocardiogram reading sessionswithresidents/fellowsand
faculty to review interesting and challenging 3D echocar-
diography studies and quarterly echocardiography review
sessions with faculty and residents from sister institutions
in the city are offered. There are also routine “in-service”
sessions for faculty whenever there are hardware and soft-
ware upgrades.
TECHNICAL ASPECTS OF INTRAOPERATIVE
THREE-DIMENSIONAL IMAGING
Image Acquisition Protocol
Intraoperative 3D imaging is performed as a supplement to
2D imaging. There is no standardized sequence for conducting
an intraoperative 3D examination. Guidelines suggest the
standardized views and their descriptions for specific cardiac
Fig 4. Biplane mode. (A) Illustration depicts the 2 orthogonal planes of imaging in the mitral valve. (B) The planes of imaging as seen in
en-face view of mitral valve. (C and D) Visualization of the individual orthogonal sections of the mitral valve acquired through simultaneous
biplane imaging.
Fig 5. Multiplane mode. Panel D demonstrates simultaneous orthogonal sections acquired through 3D volumetric analysis. Panel A
represents the 2D image derived from Plane 1 in Panel D. Panel B represents the 2D image derived from Plane 2 in Plane D. Panel C represents
the 2D image derived from Plane 3 in Panel D. (Panel D: Plane 1 ¼ Panel A; Plane 2 ¼ Panel B; Plane 3 ¼ Panel C.)
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE
473
structures seen in those views.
7,24
However, the specific
structure and the extent of its examination are dictated by the
context of the clinical situation. Because the workflow of an
intraoperative 3D TEE examination requires lack of motion and
electrical interference and brief apneic periods, ideally it should
be conducted before the start of a surgical procedure. Based on
the authors’ institutional experience, the following sequence of
examination has been developed: (1) Establish ultrasound
equipment availability, readiness, and operation, (2) availability
of echocardiographic (ECG) synchronization capability for R-
wave gated imaging (Fig 2), (3) 3D TEE probe calibration with
the ultrasound system, (4) TEE probe insertion, (5) comple-
tion of a comprehensive 2D and Doppler examination, (6)
identification of structure(s) of interest for 3D examination, (7)
procedure-specific 3D image acquisition and storage, (8)
immediate post-acquisition online access for generating Digital
Imaging and Communications in Medicine (DICOM) loops for
viewing/storage and basic quantification (linear measures), (9)
conclusion and archiving of the study for post-acquisition
analyses, and (10) generation of report.
Spatial Versus Temporal Resolution
The echocardiographic mode used for 3D imaging of a
cardiac structure is dictated by the specific clinical question. A
thorough understanding of the strengths and limitations of
various 3D imaging modes is a prerequisite for optimal 3D data
Fig 6. Incorporation of color-flow Doppler (CFD) during multi-plane imaging. Illustration depicts the multiple planes of the view of the mitral
valve as seen in the panels underneath. Panels 1 and 2 show incorporation of CFD in imaging of the mitral valve.
Fig 7. Narrow sector “live” imaging. Panels show the aortic valve (AV) and the anterior mitral leaflet (AML) using narrow-sector imaging in 2
different planes. Notice the pyramidal shape of the scan plane and the elevation that is visualized in this plane.
MAHMOOD ET AL474
collection and analysis. Even though the computational power
and image formation have improved significantly, there are
various physical impediments common to all systems. For
example, the speed of sound in soft tissues remains constant at
1,540 meters/second. Regardless of the sophistication of the
system, the go-return time for an ultrasound pulse at a given
depth remains constant, limiting the temporal resolution for all
3D images. Because 3D images are composites of 2D images,
predictably, patients with suboptimal 2D images also have
suboptimal 3D images. Furthermore, 3D imaging suffers from
the same limitations imposed by trade-offs between spatial and
temporal resolution (Fig 3). In addition, the sector size is a
major determinant of image quality.
Scan li ne density determines spatial resolution (see Fig 3)
that results in lo nger image acquisition and pr ocessing times,
thus reducing temporal resolu tion. Conversel y, temporal
resolution is based on the volumes/second gener ated by the
system, and hence requires lower line de nsity for faster
processing, the reby reducing spatial resolution (see Fig 3).
Adj ust ment of the sector or volume size can enable the
Fig 8. Piecemeal examination of the mitral valve. Anterior (A1, A2, A3) and posterior (P1, P2, P3) commissures visualized piecemeal using the
narrow-sector imaging mode.
Fig 9. Focused wide-sector “zoom” view. Panels A and B show the region of interest (mitral valve) in 2 orthogonal sections. Panel C shows
the focused en-face view of the mitral valve. Notice the decreases in temporal and spatial resolution with increased sector size. AML, anterior
mitral leaflet; PML, posterior mitral leaflet; AV, aortic valve.
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE
475
echocardiograp her to acquire an image with the highest spatial
and temporal resolution. Al ternatively, R-wave gated recon-
structive techniques can be used to optimize spatial and
temporal resolution s and se ctor siz e (discusse d in the follow-
ing section).
Imaging Methods
There are three modes of 3D imaging: simultaneous multi-
plane imaging, “live” imaging, and R-wave gated multibeat
reconstruction imaging.
Fig 10. “Single-beat” full-volume live mode. Full-volume 3D acquisition of the mitral valve acquired in a single beat. Reduction of the depth
by focusing only on the ROI shows a minimal improvement in spatial resolution.
Fig 11. Multibeat R-wave gated image acquisition and synchronization. Illustration depicts the acquisition of individual R-wave gated
images at exact intervals during the cardiac cycle and the subsequent synchronization of the acquired images.
MAHMOOD ET AL476
Simultaneous Multi-Plane Imaging
This imaging mode refers to the simultaneous generation of
2 or more 2D scan planes at a given probe position. Typically,
the probe is positioned with the desired scan plane rotation to
define the reference plane of the region of interest (ROI).
Activation of the biplane mode allows for simultaneous
visualization of 2 orthogonal sections of the ROI (Fig 4).
Some manufacturers allow multi-plane mode to generate 2 or
more scan planes intersecting the same structure of interest. As
a result, the same structure can be visualized simultaneously in
multiple scan planes (Fig 5). Color-flow Doppler (CFD) also
can be activated during multi-plane imaging for interrogating
regurgitant jets (Fig 6). It also can be used for examination of
cardiac structures (eg, left atrial appendage [LAA]).
Live Imaging
Live imaging can be performed in the following modes:
real-time narrow sector, focused wide sector, and single-beat
full-volume mode. Each of these modes can be acquired with
or without CFD information.
“Live” narrow-sector imaging. Activation of this mode
results in generation of a pyramidal-shaped ultrasound scan
plane, also referred to as a “frustum”. It has an elevational
dimension in addition to length and breadth (Fig 7). It is a very
narrow sector that provides a “probe-responsive,”“live” image
of the ROI (ie, the image moves with TEE probe motion). The
image also has high temporal and spatial resolution. However,
the sector width is insufficient to examine any cardiac structure
in its entirety (see Fig 7). The current generation of ultrasound
systems allows for incorporation of CFD information during
“live” narrow-sector imaging. However, due to the resultant
reductions in temporal and spatial resolution, there is a signifi-
cant deterioration in image quality. “Live imaging” sometimes
can be useful in patients with irregular rhythm when R-wave
gated reconstruction is not possible. When CFD information
needs to be incorporated, it sometimes is helpful to narrow the
sector size to limit imaging only to the ROI (eg, a mitral valve
perforation for precise location of an MR jet). Alternatively, the
mitral valve can be examined piecemeal from the anterolateral to
postero-medial commissure using this mode (Fig 8).
Fig 12. Demonstration of multibeat image synchronization. Panel
A shows the acquisition of R-wave gated images of the mitral valve.
Panel B shows the alignment of the individual images for synchro-
nization. Panel C shows the final synchronization of the individual
images to produce the final output.
Fig 13. Different sources of stitch artifact. Panel 1 shows stitch artifact as the result of misalignment of the acquired mitral valve images due
to pairing of systolic with diastolic images. Panel 2 shows stitch artifact of the mitral valve that could have resulted from either patient
movement or probe motion during image acquisition.
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE
477
Focused wide-sector “zoom” mode. In this mode, an ROI
initia lly is ide ntified in 2 orthogonal planes (Fig 9). Activation of
“zo om” mode limits imaging to the ROI only, thus enlarging the
sector size and allowing for visualization of only the entire ROI in a
“live” and probe-responsi ve mode. The increase in sector size is
achieved at a cost of lower line density and volumes/second with an
expected reduction in spatial and temporal resolution. However,
using this mode, 3D images can be acquired quickly with a
relatively simple workflow. Mitral, tricuspid, and aortic valves can
be visualized en-face in their entirety, even in patients with irregular
rhythms and during electrical interference.
During “live” zoom imaging, CFD information also can be
incorporated, with even further reduction in temporal and
spatial resolution. This imaging mode is particularly suitable
for patients with irregular rhythm and during percutaneous
TEE-guided intracardiac interventions (eg, mitra-clip and
device closure of atrial septal defects).
“Single-beat” full-volume live mode. This imaging mode is
similar to the wide-angle “zoom” mode. When activated, the
Fig 14. R-wave gated multibeat acquisition without color. Region of interest is captured in multiple beats and reconstructed rapidly. Panel A
demonstrates 3D R-wave gated image acquired from 6 heart beats as indicated by the yellow arrow. Frame rate is indicated by the red arrow.
Panel B shows simultaneous orthogonal views to optimize the lateral and elevational width.
Fig 15. R-Wave wave gated mul tibeat acquisition with color.
Integration of color-flow Doppler in multibeat image acquisition.
AV, aortic valve; 3D MR, three-dimensional mitral regurgitation jet.
Fig 16. Depiction of the sagittal, coronal, and transv erse planes.
MAHMOOD ET AL478
entire sector, rather than an ROI, is selected and acquired in 3D
mode (Fig 10). There are expected reductions in spatial and
temporal resolution. This mode also is suitable for patients with
irregular rhythm and when there is electrical interference with
the ECG signal. CFD information can be incorporated during
single-beat full-volume image acquisition. However, adjust-
ments in line density often are required to include the entire
ROI in the 3D image. There is also a consequent reduction in
temporal resolution with CFD incorporation.
Higher spatial resolution can be achieved by reducing the
depth to eliminate unnecessary anatomic detail. For example,
reducing depth to display only the mitral valve in the sector
can result in marginal improvements in spatial resolution
(see Fig 10).
R-Wave Gated Multibeat Acquisition
There are three R-wave gated 3D image acquisition modes:
R-wave gated multiple-beat acquisitions without CFD, R-wave
gated multiple-beat acquisitions with CFD, and wide-angle
“zoom” multibeat acquisition. R-wave gated acquisition is
based on piecemeal examination of a structure (eg, mitral
valve with multiple frustums/volumes that are generated
sequentially in synchrony or gated to the R wave of the
ECG). Because each frustum has a high line density and
temporal resolution, the final reconstructed image is predictably
of a very high quality.
Multibeat acquisition requires temporal and spatial
synchronization (Figs 11 and 12). Temporal and spatial
Fig 18. Various three-dimensional imaging modes for the mitral valve. Either narrow-sector or wide-sector “zoom” mode can be used with
their respective trade-offs. Panel A compromises the sector size for higher resolution in the narrow-sector mode. Panel B compromises
resolution for sector size in wide-sector mode.
Fig 17. Axes of rotation. Panel A depicts the x, y, and z axes. Panel B demonstrates the various axial rotations controlled by the trackball
motion in the ultrasound machine to view the region of interest.
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE
479
synchronization requires that volumes be acquired exactly at
the same time in the cardiac cycle and same location in space
(see Fig 11). Therefore, it can be performed only in patients with
regular cardiac rhythm and during periods when there is no patient
or surgical motion. Irregula r cardiac rhythms and electrical
inter feren ce with the slaved EC G sig nal to the ult raso und system
make it impossible to achieve this synchronization. Spatial
synchronization is achieved by brief suspension of controlled
ventilation or lack of probe or patient motion. Lack of this
synchronization leads to a “stitching artifact” when the volumes
are misaligned in space or when systolic frames are paired with
diastolic frames, leading to a choppy, un-interpretable image
(Fig 13). High volume rate (HVR) mode is another available
proprietary mode that allows live 3D imaging with a high
frame rate and fewer gating artifacts (eg, stitching artifacts).
R-wave gated multibeat acquisition without color. This
mode allows for the highest spatial and temporal resolution
with CFD information. Before initiation of acquisition, the
ROI i s define d. The width of the lateral and elev ation al
planes can be adjusted to include the entire ROI in the
capture bo x. De pending on t he num ber o f beat s sel ected by
the operator, multiple volumes are acquired that are gated to
the R wave of the ECG (Fig 14). The acquired data are
immediately reconstructed and displayed for further manipula-
tion and analysis. Because each volume is acquired independ-
ently, it is possible to achieve the highest spatial and temporal
resolution with the widest sector size. Whereas the reconstructed
multibeat data are not probe responsive after acquisition,
each frustum during acquisition is probe responsive. Therefore,
any probe or patient motion results in a stitching artifact
(see Fig 13).
R-wave gated multibeat acquisition with color. CFD infor-
mation can be acquired simultaneously during multibeat R-
wave gated image acquisition. Due to a reduction in sector size,
Fig 19. Surgical left atrial (LA) and left ventricular (LV) en-face views of the mitral valve. The surgical LA en-face three-dimensional view is
most commonly used to visualize the mitral valve. LAA, left atrial appendage; AV, aortic valve; CS, coronary sinus; A1, A2, A3, anterior and P1,
P2, P3, posterior mitral valve scallops; LVOT, left ventricular outflow tract; AML, anterior mitral leaflet; PML, posterior mitral leaflet.
Table 1. Carpentier Type-I Lesions
Annular Dilation
Requirement Recommendations
Sector size Wide 1. “Live” wide-angle “zoom” mode generally is sufficient
2. All 4 quadrants of annulus should be included in the orthogonal planes
3. R-wave gated imaging also can be performed but does not offer any added advantage
Spatial resolution Low
Temporal resolution Low
Most important requirement: sector size
Leaflet Perforation
Requirement Recommendations
Sector size Narrow/wide 1. “Live” wide-angle” zoom mode is sufficient, but perforation may be difficult to differentiate
from a dropout artifact due to low line density
2. The entire leaflet/region in question has to be visualized within the ROI
3. CFD incorporation in “live” narrow-sector or wide-angle “zoom” mode can identify the
perforation; however, the image has poor spatial resolution
4. R-wave gated imaging provides the highest spatial and temporal resolution
Spatial resolution High
Temporal resolution Low or high with CFD
Most important requirement: high spatial resolution
Abbreviations: CFD, Color-flow Doppler; ROI, region of interest.
MAHMOOD ET AL480
line density and Doppler frequency have to be adjusted to the
lowest level to include the entire ROI in the image (Fig 15). An
optimal display of CFD image requires significant post-
acquisition adjustment and analysis.
Wide-angle “zoom” multibeat acquisition. This 3D imaging
mode is a combination of R-wave multibeat acquisition and the
wide-angle “zoom” mode. In the first step, an ROI is identified
for wide-angle “zoom” 3D imaging. The default setting in the
Table 2. Carpentier Type-II Lesions
Mitral Valve Prolapse: Myxomatous Degeneration
Requirement Recommendations
Sector size Wide 1. Large sector size to include the entire mitral annulus and both leaflets in the ROI is sufficient
2. High spatial resolution for delineation of both anterior and posterior leaflets and coaptation zone
3. Even though temporal resolution is important, it is more important to see the entire extent of the
mitral annulus and leaflets
4. Ideally an R-wave gated multibeat acquisition should be performed
Spatial resolution High
Temporal resolution Intermediate
Most important: High spatial, intermediate temporal resolution
Mitral Valve Flail: Fibroelastic Deficiency (Torn Chordae Tendineae)
Requirement Recommendations
Sector size Narrow/wide 1. R-wave gated multibeat acquisition is ideal for achieving the highest spatial and temporal resolution
to visualize torn chords
2. In patients with irregular rhythm, a “live” narrow-sector image that includes the specific mitral valve
scallop in the ROI
3. Live wide-a ngle “zoom” mode offers satisfactory image quality
Spatial resolution High/intermediate
Temporal resolution High
Most important: High temporal resolution/spatial resolution
Ischemic Rupture of Papillary Muscle
Requirement Recommendations
Sector size Wide 1. R-wave gated multibeat acquisition to achieve the highest spatial and temporal resolution
2. “Live” wide-angle “zoom” mode offers a wider sector size, but temporal resolution is not
satisfactory
3. “Live” narrow sector that inclu des the ROI may be used in patients with irregular rhythms
Spatial resolution High/intermediate
Temporal resolution High
Most important: High temporal resolution/sector size
Abbreviation: ROI, region of interest.
Table 3. Carpentier Type-III Lesions
Rheumatic Restriction
Requirement Recommendations
Sector size Wide 1. Large sector size to include the entire mitral annulus and both leaflets in the ROI is sufficient to
planimeter the mitral valve area at leaflet tips
2. High spatial resolution for delineation of both anterior and posterior leaflets
3. Even though temporal resolution is important, it is more important to see the entire extent of the
mitral annulus and coaptation zone
4. Ideally, an R-wave gated multibeat acquisition should be performed
Spatial resolution High
Temporal resolution Intermediate
Most important: High spatial, intermediate temporal resolution
Ischemic Restriction
Requirement
Recommendations
Sector size Wide 1. R-wave gated multibeat acquisition is ideal for achieving the highest spatial and temporal resolut ion
to include the entire annulus, leaflets, and coaptation zone in ROI
2. Live wide-angle “zoom” mode offers satisfactory image quality
3. Single-beat full-volume acquisition mode also may be used to include the entire mitral valve in
the ROI
Spatial resolution High
Temporal resolution Intermediate/high
Most important: High spatial resolution/intermediate temporal resolution
Abbreviation: ROI, region of interest.
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE
481
equipment is to acquire a “live” probe-responsive zoomed image
of the ROI. However, the operator manually can increase the
number of beats 2–6 beats. Thus, the zoomed ROI is acquired in
an R-wave gated sequential fashion. This results in marginal
improvement in the spatial resolution. Increasing the number of
beats will substantially increase temporal resolution and is
recognized immediately by an increase in frame rate.
Image Display Settings
En-Face Display
It is recommended that the perspective of the echocardiog-
rapher should be through the cardiac chamber that is in
continuity with the ROI.
7,17
For an en-face view of the mitral
valve, the left atrial and left ventricular perspectives should be
acquired by cropping the base and apex of the heart. For an en-
face view of the tricuspid valve, the right atrial and right
ventricular perspective should be acquired by cropping the base
and apex of the heart.
17
Cropping
Cropping refers to the removal of unwanted detail from the
rendered 3D image. Cropping can be performed before acquir-
ing the image. The advantage is that a narrow-sector image of
higher spatial and temporal resolution can be obtained. How-
ever, an image cropped before acquisition and storage cannot
be uncropped after acquisition. On the contrary, an uncropped,
slightly low-resolution 3D image can be acquired with the
added advantage of greater anatomic detail.
7
Multiplanar Tomographic Planes
The acquired 3D data are displayed with the 3D sagittal,
coronal, and transverse planes in perspective (Fig 16). There
are various manufacturer-specific nomenclatures for these
planes. Ultrasound system controls allow for cropping of 3D
volume along any of these axes. There also is provision for
adding another cropping plane for structures that are present at
oblique planes to the ROI. Manufacturers of ultrasound systems
have created workflow arrangements for semi- to completely
automated generation of multiple tomographic planes.
Rotation
Using the trackball function of the system, the volumetric
data also can be rotated along the x, y, and z axes to view the
ROI from multiple perspectives (Fig 17). Different image
display settings may be required to accurately display the
anatomic features. For example, different tissue threshold and
gain settings are required for left atrial/left ventricular perspec-
tives for the mitral valve and for the RA/RV perspective for the
tricuspid valve.
Qualitative Analysis
Mitral Valve Imaging
The mitral valve can be imaged using any of the 3D
imaging modes (Figs 18 and 19). The “surgical” left atrial en-
face 3D view is the most commonly used view of the mitral
valve (see Fig 19).
17,25
The multiplane imaging is useful for
simultaneous 2D visualization of the affected mitral valve
Fig 20. Right atrial (RA) and right ventricular (RV) en-face views of the tricuspid valve. TV, tricuspid valve; LA, left atrium; IVS,
interventricular septum; LV, left ventricle.
Table 4. Tricuspid Valve Examination
Challenges
1.
Thin and poorly echogenic leaflets
2. Location in the far field of the scan plane
3. Artifacts from the presence of intracardiac catheters/wires
4. Artifacts from bubble contrasts due to centrally
administered fluids
Requirement Recommendations
Sector size Wide
1. R-wave gated multibeat acquisition
for achieving the highest spatial and
temporal resolution
2. “Live” wide-angle “zoom” mode
also can be used
3. For localization of vegetations on the
tricuspid valve with a high temporal
resolution, either R-wave gated
multibeat acquisition or a “live”
narrow-angle mode with the affected
leaflet in the ROI also can be used
Spatial
resolution
High
Temporal
resolution
Intermediate
Most important: High spatial,
intermediate temporal resolution
Abbreviation: ROI, region of interest.
MAHMOOD ET AL482
scallops and MR jets from multiple perspectives. The “live”
narrow-sector mode usually does not encompass the entire
valve from commissure to commissure. The most useful views
for mitral valve imaging are the R-wave gated full-volume
(with or without CFD) or wide-angle “ zoom” views with the
highest spatial and temporal resolution (see Fig 18). Alter-
natively, the “live” wide-angle “zoom” or the single-beat full-
volume view also can be used for diagnosis of mitral valve
pathology (Tables 1–3).
Tricuspid valve. Tricuspid leaflets are not as echogenic as are
mitral leaflets. It sometimes is difficult to accurately visualize
the leaflets. However, with appropriate machine and probe
adjustments it is possible to obtain right atrial and right
ventricular en-face views of the tricuspid valve (Fig 20).
19
As
for mitral valve, the R-wave gated modes provide the highest
spatial and temporal resolution (Table 4).
Aortic valve. The aortic valve is visualized en-face from the
ascending aortic perspective (Fig 21).
18
This perspective
provides a view analogous to the midesophageal short-axis
view of the aortic valve. The aortic valve can be visualized in
the “live” narrow- and wide-angle “zoom” modes and the R-
wave gated modes. Often due to the presence of calcification,
there are significant dropout artifacts (Table 5).
Left and right atria. Intraoperative 3D imaging of the left
atrial appendage (LAA) has been established as a procedural
adjunct to LAA ligation procedure.
20,26
Both the LAA and
right atrial appendage can be visualized as en-face views using
the “live” and R-wave gated modes (Fig 22).
27,28
3D imaging
has shown value in differentiating appendage thrombi from
lobes and completeness of surgical LAA ligation.
21,29
Quantitative Analysis
Multiplanar Reformatting
Even though 3D imaging provides high-resolution qualita-
tive images of intracardiac structures, the real value of
volumetric data is in their quantitative analyses. Using the
MPR technique, it is possible to dissect perfect orthogonal
sections and make accurate measurements.
14
Both static and
dynamic annular dimensions feasibly can be derived with semi-
and completely automated analyses.
30
Complex geometric
conformations can be analyzed readily using 3D data.
31
These
quantitative analyses have shown value before and after repair
assessment of intracardiac valves.
16
Mitral Valve
Ready availability of quantitative structural and functional
information of mitral valve function has transformed intra-
operative echocardiographic assessment to a very objective
discipline.
16,32,33
What was once considered a research interest
Fig 22. En-face view of the left (LAA) and right atrial appendages
(RAA). LUPV, left upper pulmonary vein (LUPV).
Fig 21. En-face view of the aortic valve from the ascending aortic
perspective. Corres ponds to the midesophageal short-axis view of
the aortic valve. LA, left atria; TV, tricuspid valve; NCC, noncoronary
cusp; RCC, right coronary cusp; LCC, left coronary cusp; PV, pulmon-
ary valve; LAA, left atrial appendage.
Table 5. Aortic Valve Examination
Challenges
1.
Located in the far field of the scan plane
2. Leaflet curvature often results in dropout artifact
3. Reverberation and acoustic shadowing artifacts due to
calcification
Requirement Recommendations
Sector size Narrow 1. The entire aortic valve can be
included in the ROI using the “live”
narrow-sector imaging mode with a
high spatial and temporal resolution
2. “Live” wide-angle “zoom” mode
also can be used
3. For localization of vegetations on
the aortic valve where high
temporal resolution is required,
either R-wave gated multibeat
acquisition or a “live” narrow-angle
mode with the affected leaflet in the
ROI also can be used
Spatial
resolution
High
Temporal
resolution
Intermediate
Most important: High spatial,
intermediate temporal resolution
Abbreviation: ROI, region of interest.
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE
483
has become routinely available information.
16,34
Mitral valvular
geometric assessment also has transformed from a static single
frame to a dynamic analysis throughout the cardiac cycle. It has
demonstrated value in providing information to assist surgical
decision making by providing objective information regarding
annular dimensions, regurgitation quantification and assess-
ment of a repaired valve (Fig 23).
11,25,35–37
Tricuspid Valve
Due to poor echogenicity of the tricuspid leaflets, annular
measurements are used as surrogate markers of valvular
function. Quantitative analyses of the tricuspid valve were
limited to regurgitation quantification and annular area calcu-
lations using the assumption of a planar circular shape.
38
Volumetric 3D data have confirmed that the tricuspid annulus
is neither circular nor planar in conformation.
38
Using 3D data,
tricuspid annular diameter and area can be calculated with a
degree of precision (Fig 24).
Aortic Valve
With increasing popularity of percutaneous aortic valve
replacement and surgical repair, interest has been renewed
in aortic valve ge ometry (Fig 25).
18
Three-dimensional
anatomy of the left ventricular outflow tract has dem-
onstrated that it is elliptical in shape (Fig 26), and the
assumption of a circular shape has the potential to affect the
accuracy of valve area and other hemodynamic calculations.
39
Fig 23. Quantitative analysis of the mitral valve. Illustrations depicting the various quantitative measurements of the mitral valve that could
be obtained using three-dimensional echocardiography. ITG, intertrigonal distance; AL-PM, anterolateral-posteromedial diameter; AP,
anteroposterior diameter; AL, anterior leaflet length; AL, anterior leaflet area; PL, posterior leaflet area.
Fig 24. Quantitative analysis of the tricuspid valve. Panel A: Illustration shows the measurement of anteroposterior (AP) and septolateral
(SL) diameters of the tricuspid valve in 2 orthogonal planes. Panel B: Quantitative measurements of the tricuspid valv e in orthogonal scan
planes. Panel C: Quantitative measurements visualized in en-face view of the tricuspid valve.
MAHMOOD ET AL484
For aortic valve repair, post-repair echocardiographic cri-
teria have been developed as predictors of recurrence and
reoperation.
40
Right and Left Atria and Ventricles
There are numerous 3D echocardiographic algorithms for
volumetric assessment of both atria and ventricles.
41,42
Most of
these are time consuming and complex off-line techniques that
generally are not practical for the intraoperative environment.
The MPR technique can be used for chamber quantifi-
cation, albeit with an image with lower spatial and temporal
resolution.
Prosthetic Valves/Percutaneous Interventions
Assessment of prosthetic valves and annuloplasty rings
immediately after termination of cardiopulmonary bypass is a
challenging undertaking. For prosthetic valves, qualitative
Fig 25. Geometry of the aortic valve and the left ventricular outflow tract (LVOT). (A) The various quantitative measurements that can be
made using the acquired three-dimensional images of the aortic valve and LVOT. (B) Panel showing coaptation height and depth at the level of
the aortic annulus. IC, inter-commissural; AA, aortic area.
Fig 26. Demonstration of the elliptical shape of the left ventricular outflow tract (LVOT). Panel A shows the LVOT in the long axis. Panel B,
which is a cross-section denoted by the red line in Panel A, demonstrates the elliptical shape of the LVOT in the short axis.
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE
485
imaging to ensure mechanical stability, leaflet excursion, and
lack of impingement of surrounding structures is the mainstay
of intraoperative assessment. Using qualitative CFD informa-
tion, perivalvular regurgitant jets can be localized with a degree
of precision.
25,43
Quantitative analyses have shown value in
accurate assessment of an altered valve geometry area, plani-
metry, and leaflet stress.
44
These remain offline, time-
consuming techniques with little prognostic value.
44
Percutaneous interventions for structural heart disease have
become popular therapeutic options.
45
Perioperative 3D provides
real-time assistance in selecting suitability for the specific
percutaneous intervention and procedural guidance, confirms
the success, and excludes complications. A combination of “live”
narrow-angle and wide-angle “zoom” modes is more useful in
providing real-time procedural guidance. If possible, R-wave
gated multiple-beat acquisition should be used after device
deployment to confirm success with the highest spatial and
temporal resolution (Table 6).
Limitations of Three-Dimensional Imaging
Despite enhanced spatial orientation, there are significant
limitations of 3D imaging. Being composed of ultrasound
pulses, it is subject to the same artifacts as traditional 2D
images. It is extremely important for the echocardiographer to
be aware of these limitations of 3D imaging to prevent
misinterpretation and misdiagnosis.
Resolution
Compared with 2D images, all 3D imaging modes have
lower spatial and temporal resolution. Even though MPR
provides perfect orthogonal 2D sections of anatomic structures
for linear measures, these are comparably lower resolution
images than 3D images.
Artifacts
Images acquired with 3D transducers are composites of
ultrasound pulses and therefore are susceptible to artifacts
(Table 7).
46
Parallax Error
Parallax error is defined as the displacement of or difference
in apparent position of an object viewed along 2 different lines
of sight (Fig 27). During 3D imaging, parallax error is created
while making linear measurements on rendered 3D data
(Fig 28).
Suboptimal Image Quality
Three-dimensional images are a composite of multiple 2D
images. Therefore, 3D imaging is not a substitute for sub-
optimal 2D images.
Data Archiving
Although standards exist for archiving 3D data, they
recommend archiving raw data, resulting in large file sizes.
47
Therefore, these formats have not been very popular. Gener-
ally, manufacturer-specific software is used for handling 3D
data. In case of incompatibilities between ultrasound systems
and viewing stations, 3D images cannot be analyzed as raw
data and lose their 3D functionality. Commercially available
vendor non-specific software (Tomtec Image Arena, GmBH,
Munich, Germany and Mimics Innovations Suite, Materialise,
Leuven, Belgium) can analyze 3D ultrasound data from multi-
ple sources. However, they are not only offline but also
expensive and require a significant learning curve for use.
Incompatibilities of 3D ultrasound system and viewing soft-
ware can be a significant limitation for quantitative analyses in
multi-vendor ultrasound laboratories.
Table 6. Prosthetic Valves
Challenges
1.
Reverberation and acoustic shadowing artifacts
2. Suboptimal image quality
3. Epicardial pacing and irregular rhythms
4. Time constraint
Recommendations
1.
A thorough 2D examination to identify the ROI that may need
further examination with 3D imaging
2. R-wave gated multibeat acquisition for the highest spatial and
temporal resolution
3. It may be possible to achieve R-wave gating with cardiac pacing;
if possible, epicardial pacing should be discontinued to facilitate
R-wave gated acquisition
4. Overdrive pacing may be considered in patients who have
irregular rhythms for R-wave gated imaging
5. Digital subtraction of CFD information from the gray scale image
helps in post-acquisition image optimization
Abbreviations: CFD, Color-flow Doppler; ROI, region of interest.
Fig 27. Illustration describing the occurrence of parallax error.
The star appears to have 2 different backgrounds depending on
where it is viewed—a physics phenomenon that is a common source
of error in three-dimensional echocardiography.
MAHMOOD ET AL486
Fig 28. Parallax error in three-dimensional echocardiography. (A) Illustration demonstrating the angle between the mitral and aortic valves.
(B) The angle visualized in two-dimensional view. (C) Three-dimensional view does not show the angle between the mitral and aortic valves due
to a loss in depth perception. This is a classic example of parallax error. MV, mitral valve; AV, aortic valve.
Table 7. Artifacts
Stitching Artifact
Stitching artifacts occur due to misalignment of isolated images in a multibeat acquisition.
Misalignment of images could either be due to movement of the probe or patient movement during
image acquisition. It also could occur when the systolic and diastolic images are paired together,
leading to a grossly rough and uneven image.
Dropout
Dropout artifacts appear as gaps in the cardiac images and imitate typical perforation in cardiac
structures. It occurs due to improper echocardiogram signals reflecting off the cardiac structures,
which could be due to a decreased gain or improper angulation of the probe in such a way that the
ROI is not perpendicular to the ultrasound beam. Dropout artifacts are much more common in
aortic valve leaflets due to their angulation with the ultrasound beam.
Blurring
Blurring artifacts make structures appear denser and thicker than their original form. Blurring occurs
due to the misalignment of discordant voxels and thus produces an inaccurate projection of the
structures. Blurring commonly is seen as thickened appearance of surgical sutures around the
annulus and likewise presents prosthetic annuli as dense structures.
Blooming
Blooming artifacts occur when the ultrasound beam is reflected off the edges of metallic devices and
appear as an uneven and dense formation localized to the periphery of the metallic device. It gives
an uneven and irregular structure to catheters and guidewires, which make them appear thicker at
some points and narrower at other points.
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE 487
Arrhythmias
R-wave gated multibeat acquisition provides the highest
spatial and temporal resolution. Specifically, for integration of
CFD data, R-wave gated imaging is crucial for optimal image
quality. This can become impossible in patients with arrhyth-
mias, limiting the imaging to “live” or single-beat full-volume
mode. Recently, ultrasound systems have become available that
are capable of providing a very high temporal resolution while
maintaining a wide sector and line density. This is a significant
advance from the currently available systems.
CONCLUSION
Intraoperative 3D echocardiography is a constantly evolving
process. Innovations in technology and their application will be
updated constantly, with improvements in computational power
and processing speed. Currently, there are multiple commer-
cially available platforms with varying workflow and operation,
which unfortunately has resulted in confusion in terminology.
Consequently, there remains a need to develop universal
standards for 3D image acquisition, nomenclature, and analysis
to facilitate communication and enhance diagnosis and treat-
ment using 3D echocardiography.
Table 7 (continued )
Railroad artifact
Railroad-shaped artifacts particularly are seen when catheters with a wide lumen are used. The
surface of the catheter that is perpendicular to the ultrasound beam is seen clearly, whereas the
surfaces that are parallel to the beam do not appear in the image. This leads to a railroad-shaped
artifact, and a single large catheter appears as 2 thin structures parallel to each other.
Reverberation
Reverberation artifacts are collective reflections of the metallic segment of catheters and result in
projecting a more elongated image of a catheter than it actually is. This leads to error in proper
positioning of the catheter unless this artifact is accurately identified.
Shadowing
When a highly reflective material impedes the ultrasound beam from passing beyond it, it forms a
shadow and appears as a cleft or tear in the structure behind it. This could be misinterpreted as a
rupture in the cardiac structure, but can be identified by the fact that it changes form or disappears
when the reflective material changes in position or is withdrawn.
Gain
Incorrect gain settings may lead to an erroneous projection of valve orifices. Increased gain will
present as a narrow orifice, whereas a decreased gain setting presents as a broad orifice.
Increased gain in mitral valve
Decreased gain in mitral valve
MAHMOOD ET AL488
REFERENCES
1. Abraham TP, Warner JG, Kon ND, et al: Feasibility, accuracy, and
incremental value of intraoperative three-dimensional transesophageal
echocardiography in valve surgery. Am J Cardiol 80:1577-1582, 1997
2. Chauvel C, Bogino E, Clerc P, et al: Usefulness of three-
dimensional echocardiography for the evaluation of mitral
valve prolapse: an intraoperative study. J Heart Valve Dis 9:341-349,
2000
3. Pai RG, Jintapakorn W, Tanimoto M, et al: Three-dimensional
echocardiographic reconstruction of the left ventricle by a trans-
esophageal tomographic technique: in vitro and in vivo validation of
its volume measurement. Echocardiography 13:613-622, 1996
4. Scohy TV, Cate Ten FJ, Lecomte PVEL, et al: Usefulness of
intraoperative real-time 3D transesophageal echocardiography in car-
diac surgery. J Card Surg 23:784-786, 2008
5. Schulmeyer MCC, Santelices E, Vega R, et al: Impact of intra-
operative transesophageal echocardiography during noncardiac surgery.
J Cardiothorac Vasc Anesth 20:768-771, 2006
6. Karthik S, Sundar S, Lerner A, et al: Intraoperative assessment of
perivalvular mitral regurgitation: utility of three-dimensional echocar-
diography. J Cardiothorac Vasc Anesth 22:431-434, 2008
7. Lang RM, Badano LP, Tsang W, et al: EAE/ASE recommenda-
tions for image acquisition and display using three-dimensional
echocardiography. J Am Soc Echocardiogr 25:3-46, 2012
8. Sugeng L, Shernan SK, Salgo IS: Live 3-dimensional
transesophageal echocardiography: initial experience using the
fully-sampled matrix array probe. J Am Coll Cardiol 52:446-449,
2008
9. Labib DO, Heshmat H, Gaafar AH, et al: Real-time three-
dimensional transthoracic echocardiography for predicting mitral annu-
loplasty ring size. J Heart Valve Dis 23:583-590, 2014
10. La Canna G, Arendar I, Maisano F, et al: Real-time three-
dimensional transesophageal echocardiography for assessment of mitral
valve functional anatomy in patients with prolapse-related regurgita-
tion. Am J Cardiol 107:1365-1374, 2011
11. Mahmood F, Subramaniam B, Gorman JH, et al: Three-
dimensional echocardiographic assessment of changes in mitral
valve geometry after valve repair. Ann Thorac Surg 88:1838-1844,
2009
12. Mihăilă S, Muraru D, Piasentini E, et al: Quantitative analysis of
mitral annular geometry and function in healthy volunteers using
transthoracic three-dimensional echocardiography. J Am Soc Echocar-
diogr 27:846-857, 2014
13. Biaggi P, Jedrzkiewicz S, Gruner C, et al: Quantification of
mitral valve anatomy by three-dimensional transesophageal echocar-
diography in mitral valve prolapse predicts surgical anatomy and the
complexity of mitral valve repair. J Am Soc Echocardiogr 25:
758-765, 2012
14. Houck RC, Cooke JE, Gill EA: Live 3D echocardiography: a
replacement for traditional 2D echocardiography? Am J Roentgenol
187:1092-1106, 2006
15. Mori M, Yoshimuta T, Ohira M, et al: Impact of real-time three-
dimensional transesophageal echocardiography on procedural success
for mitral valve repair. J Echocardiogr 13:100-106, 2015
16. Mahmood F, Matyal R: A quantitative approach to the intra-
operative echocardiographic assessment of the mitral valve for repair.
Anesth Analg 121:34-58, 2015
17. Mahmood F, Warraich HJ, Shahul S, et al: En face view of the
mitral valve: definition and acquisition. Anesth Analg 115:
779-784, 2012
18. Kim H, Bergman R, Matyal R, et al: Three-dimensional
echocardiography and en face views of the aortic valve: technical
communication. J Cardiothorac Vasc Anesth 27:376-380, 2013
19. Montealegre-Gallegos M, Bergman R, Jiang L, et al: Tricuspid
valve: an intraoperative echocardiographic perspective. J Cardiothorac
Vasc Anesth 28:761-770, 2014
20. Owais K, Mahmood F, Montealegre-Gallegos M, et al: Left atrial
appendage, intraoperative echocardiography, and the anesthesiologist. J
Cardiothorac Vasc Anesth 29:1651-1652, 2015
21. Wunderlich NC, Beigel R, Swaans MJ, et al: Percutaneous
interventions for left atrial appendage exclusion: options, assessment,
and imaging using 2D and 3D echocardiography. JACC Cardiovasc
Imaging 8:472-488, 2015
22. Kijima Y, Akagi T, Nakagawa K, et al: Three-dimensional
echocardiography guided closure of complex multiple atrial septal
defects. Echocardiography 31:E304-E306, 2014
23. Montealegre-Gallegos M, Mahmood F: Intraoperative transeso-
phageal echocardiography: Monere to Decidere. J Cardiothorac Vasc
Anesth 28:1700-1701, 2014
24. Shanewise JS, Cheung AT, Aronson S, et al: Changes in mitral
annular geometry after aortic valve replacement: a three-dimensional
transesophageal echocardiographic study. J Heart Valve Dis 21:
696-701, 2012
25. Mahmood F, Warraich HJ, Gorman JH, et al: Changes in mitral
annular geometry after aortic valve replacement: a three-dimensional
transesophageal echocardiographic study. J Heart Valve Dis 21:
696-701, 2012
26. Yu C-M, Khattab AA, Bertog SC, et al: Mechanical
antithrombotic intervention by LAA occlusion in atrial fibrillation.
Nat Rev Cardiol 10:707-722, 2013
27. Matyal R, Karthik S, Subramaniam B, et al: Real-time three-
dimensional echocardiography for left atrial appendage ligation. Anesth
Analg 108:1467-1469, 2009
28. Jainandunsing JS, Bardia A, Mahmood F: A right atrial
echodensity. J Cardiothorac Vasc Anesth 27:1065-1066, 2013
29. Beigel R, Wunderlich NC, Ho SY, et al: The left atrial
appendage: anatomy, function, and noninvasive evaluation. JACC
Cardiovasc Imaging 7:1251-1265, 2014
30. Mor-Avi V, Sugeng L, Lang RM: Real-time 3-dimensional
echocardiography: an integral component of the routine echocardio-
graphic examination in adult patients? Circulation 119:314-329,
2009
31. Mahmood F, Owais K, Montealegre-Gallegos M, et al: Echo-
cardiography derived three-dimensional printing of normal and abnor-
mal mitral annuli. Ann Card Anaesth 17:279-283, 2014
32. Mahmood F, Hess PE, Matyal R, et al: Echocardiographic
anatomy of the mitral valve: a critical appraisal of 2-dimensional
imaging protocols with a 3-dimensional perspective. J Cardiothorac
Vasc Anesth 26:777-784, 2012
33. Maslow A, Gemignani A, Singh A, et al: Intraoperative assess-
ment of mitral valve area after mitral valve repair: comparison of
different methods. J Cardiothorac Vasc Anesth 25:221-228, 2011
34. Khabbaz KR, Mahmood F, Shakil O, et al: Dynamic 3-
dimensional echocardiographic assessment of mitral annular geometry
in patients with functional mitral regurgitation. Ann Thorac Surg 95:
105-110, 2013
35. Mahmood F, Karthik S, Subramaniam B, et al: Intraopera tiv e
application of geometric three-dimensio nal mitral valve assessment pack-
age: a feasibility study.JCardiothoracVascAnesth 22:292-298, 2008
36. Maslow A, Mahmood F, Poppas A, et al: Three-dimensional
echocardiographic assessment of the repaired mitral valve. J Cardio-
thorac Vasc Anesth 28:11-27, 2014
37. Mahmood F, Gorman JH, Subramaniam B, et al: Changes in
mitral valve annular geometry after repair: saddle-shaped versus flat
annuloplasty rings. Ann Thorac Surg 90:1212-1220, 2010
PRACTICAL APPROACH TO INTRAOPERATIVE 3D TEE 489
38. Mahmood F, Kim H, Chaudary B, et al: Tricuspid annular
geometry: a three-dimensional transesophageal echocardiographic
study. J Cardiothorac Vasc Anesth 27:639-646, 2013
39. Mancuso AJ, Clark J, Mahmood F: Left ventricular outflow tract
obstruction: is it the valve or something else? J Cardiothorac Vasc
Anesth 28:848-849, 2014
40. le Polain de Waroux J-B, Pouleur A-C, Robert A, et al:
Mechanisms of recurrent aortic regurgitation after aortic valve repair:
predictive value of intraoperative transesophageal echocardiography.
JACC Cardiovasc Imaging 2:931-939, 2009
41. Lu X, Xie M, Tomberlin D, et al: How accurately, reproducibly,
and efficiently can we measure left ventricular indices using M-mode,
2-dimensional, and 3-dimensional echocardiography in children? Am
Heart J 155:946-953, 2008
42. Rudski LG, Lai WW, Afilalo J, et al: Guidelines for the
echocardiographic assessment of the right heart in adults: a report
from the American Society of Echocardiography endorsed by
the European Association of Echocardiography, a registered branch
of the European Society of Cardiology, and the Canadian Society
of Echocardiography. J Am Soc Echocardiogr 23:685-713: quiz
786-788, 2010
43. Jaina ndunsing JS , Mah mood F: High tra nsvalvul ar grad-
ients across a prosthetic valve in the mitral position : not i gn-
oring t he bi gger picture . J Cardiot horac Vasc Anesth 2 6:968 -969,
2012
44. Lang RM, Tsang W, Weinert L, et al: Valvular heart disease. The
value of 3-dimensional echocardiography. J Am Coll Cardiol 58:
1933-1944, 2011
45. Lee M-S, Naqvi TZ: A practical guide to the use of
echocardiography in assisting structural heart disease interventions.
Cardiol Clin 31:441-454, 2013
46. Faletra FF, Ramamurthi A, Dequarti MC, et al: Artifacts in three-
dimensional transesophageal echocardiography. J Am Soc Echocar-
diogr 27:453-462, 2014
47. Thomas JD, Adams DB, DeVries S, et al: Guidelines and
recommendations for digital echocardiography: a report from the
Digital Echocardiography Committee of the American Society of
Echocardiography. J Am Soc Echocardiogr 18:287-297, 2005
MAHMOOD ET AL490
