Heart, Lung and Circulation (2019) 28, 1400–1410

1443-9506/04/$36.00
https://doi.org/10.1016/j.hlc.2019.04.003

3D Echo in Routine Clinical

Practice – State of the Art in 2019
Jessica Poon, MBBS, FHKCP a, James T. Leung, MBBS, MBiostat b,
Dominic Y. Leung, MBBS, PhD c*
a
Department of Medicine, Ruttonjee and Tang Shiu Kin Hospitals, Hong Kong, China
b
Department of Cardiology, Royal North Shore Hospital, Sydney, NSW, Australia
c
Department of Cardiology, Liverpool Hospital, University of New South Wales, Sydney, NSW, Australia

Three-dimensional (3D) echo has been around for almost five decades. Recent advances in ultrasound,
electronic and computing technologies have moved 3D echo from the research environment to everyday
clinical practice. Real time 3D echo and full volume acquisition are now possible with transthoracic as well
as transoesophageal probes. The main advantages of 3D echo are the infinite cut planes possible, allowing
direct, en face, and anatomical views of cardiac structures, avoiding foreshortening and circumventing the
geometric assumptions of the cardiac chambers inherent in any 2D echo techniques. Three-dimensional
echo is still dependent on image quality, subjected to ultrasound artifacts and faces the compromise
between spatial and temporal resolution. In routine clinical practice in 2019, we recommend a focussed
3D examination after a full 2D echo study. The area where 3D echo has been consistently shown to have
superior accuracy and reproducibility over 2D echo is in the assessment of left ventricular (LV) volumes and
ejection fraction. We recommend obtaining a full volume 3D echo data set from the apical window, from
which LV volumes and LV global longitudinal strain can be measured. Further 3D examination can be
performed depending on the pathologies identified on 2D examination. Three-dimensional echo is superior
to 2D echo in the assessment of mitral valve pathologies and atrial septal defects. Furthermore, real time 3D
transoesophageal echo is a very useful technique in guiding structural cardiac intervention, both before,
during and after the procedure. While 3D echo is not the holy grail of echocardiography, it does represent a
useful technique in selected areas of cardiac imaging.
Keywords Three-dimensional echocardiography  Left ventricular function  Cardiac Imaging

and TOE, enabling its more widespread use in routine clini-

Introduction cal practice.
Since its first description in the 1960s, three-dimensional Numerous studies have demonstrated the utility and, in
echocardiography (3D echo) has gone through major advan- many cases, the superiority of 3D over 2D echo in the assess-
ces. Initially, 3D echo was performed by acquiring a series of ment of almost all cardiac structures and function. These
two-dimensional (2D) tomographic images, either by manual include assessment of left ventricular (LV) and right ventric-
rotation of the transducer with external references or later by ular (RV) volume and function and wall motion, left and
multi-plane transoesophageal echocardiography (TOE). right atrial volume, LV mass, valvular heart disease, congen-
Three-dimensional echo images were then generated by ital heart disease, LV synchrony, guidance of interventional
off-line reconstruction. The process was cumbersome and procedures and in volumetric colour Doppler imaging. In
time consuming, thereby precluding its use in routine clinical this article, we do not aim to give a full review of the technical
practice. Through advances in ultrasound, electronic and aspects and all possible applications of 3D echo. Instead, we
computer technologies, real time display of 3D rendered focus on areas where, in our opinions, 3D echo should be
images of the heart is possible with both transthoracic echo used in routine clinical practice in 2019.

*Corresponding author at: Department of Cardiology, Liverpool Hospital, Elizabeth Drive, Liverpool, New South Wales, 2170 Australia. Tel.: +61287383797;
Fax: +61296025833., Email: [email protected]
© 2019 Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac Society of Australia and New Zealand (CSANZ).
Published by Elsevier B.V. All rights reserved.
Three Dimensional Echo in Clinical Practice 1401

the heart in these tomographic planes from standard echo

Advantages of 3D echo windows. However, as these echocardiographic windows
The main advantages of 3D over 2D echo lie in the fact that a are not anatomical, they may be difficult to comprehend
sector or a full volume dataset can be obtained in three for those not familiar with echo. For example, cardiac sur-
dimensions. Like computed tomography and cardiac mag- geons view the heart differently to echocardiographers and
netic resonance imaging (CMR), this allows infinite cut- finding a common language may be difficult.
planes of the whole dataset both on- and off-line. Therefore, Three-dimensional echo images can be acquired and dis-
limitations of 2D echo like image foreshortening and off-axis played in several ways: real time 3D, zoomed 3D and full
cuts can be circumvented. Together with image cropping and volume. These are fundamentally different to that of 2D echo,
rotation, en face views of cardiac structures can be obtained, therefore, further training and familiarisation are necessary.
allowing more anatomically orientated views easily under- This fundamental difference, together with the additional time
standable by cardiac surgeons and interventional cardiolo- needed for image acquisition, may be an impediment to the
gists. Furthermore, real time 3D volume allows imaging of more widespread uptake of 3D echo in clinical practice. Further-
the left ventricle in a single heart beat allowing simultaneous more, online and/or offline image manipulation and cropping
assessment of all LV walls. This will shorten the time for are often necessary in 3D echo, presenting additional hurdles.
image acquisition post exercise for evaluation of inducible We would suggest cropping, viewing and communicating
ischaemia and for more reliable assessment of LV volume 3D echo in anatomic planes, instead of from where the
and function. images are obtained (as in 2D echo). These anatomic planes,
i.e., transverse, sagittal and coronal, are more readily under-
standable and akin to that obtainable in a computed tomog-
Limitations of 3D Echo raphy (CT) scan or CMR. Additional, helpful, views are those
from anatomical structures: e.g, view of the mitral valve from
It needs to be emphasised early on that we do not consider 3D
the left atrium or the aortic valve from the aorta. These views
echo to be the holy grail of echocardiography. Despite all the
are more easily understood by cardiac surgeons and
technical advances, 3D echo is still an ultrasound-based
interventionists.
technique and all the limitations of echo still apply. The
image quality of 3D echo is dependent on that of 2D echo,
being subjected to the same type of artifacts as dictated by the
physics of ultrasound. 2D or 3D Echo
Despite the advances in transducer and computer technol-
Proponents of 3D echo would suggest that the full echocar-
ogies, one of the main limitations of 3D echo is the frame rate.
diographic examination should be performed in 3D as
Image acquisition and display in echo is always a compro-
numerous studies have demonstrated superiority of 3D over
mise between temporal and spatial resolution. In achieving
2D echo in the assessment of cardiac structures and function.
real time 3D display capabilities, frame rate is often sacri-
With the current day 3D transthoracic echo transducers with
ficed. For 2D echo, a frame rate of >100 Hz is commonly
small footprints, a full 3D examination is feasible with appro-
achieved giving high temporal resolution at a level not
priate scanning protocols and modification of workflow [1].
achievable with other imaging modalities like computed
The image quality obtainable with the state-of-the-art 3D
tomography or CMR. With real time 3D echo, a frame rate
transthoracic transducers is comparable to that from dedi-
of, at most, 20–25 Hz is commonly obtained. With the addi-
cated 2D probes. A complete 3D examination should entail
tion of colour Doppler with 3D echo, frame rate frequently
multiple acquisitions from all standard echocardiographic
drops below 20 Hz, thereby limiting its usefulness. This limi-
windows. The American Society of Echocardiography and
tation is imposed by the physics of ultrasound, and there is a
the European Association of Echocardiography have issued
limit as to how much temporal resolution may be improved
recommendations for the views and the protocol for a full 3D
with technical advances. We feel that this unique advantage
echo examination [1]. The resulting 3D echo dataset can be
of high temporal resolution of echo over CT scan or CMR
cropped either online or offline to display the cardiac struc-
may be compromised in the pursuit of 3D capabilities.
tures in question.
Stitching artifacts are unique to 3D echo. A full volume 90 x
Opponents of 3D echo may claim that 3D echo is just about
90 3D dataset can be obtained with electrocardiograph (ECG)
pretty pictures, without many practical additional uses and
gating and breath hold. Stitching artifacts may be seen in
without incremental clinical benefits. For the most part, they
patients with irregular heart rate and inability to breath hold.
may feel 2D echo is sufficient for their use in clinical practice.
This can be avoided in real time 3D mode or zoomed mode.
This may arise from the fact that performing and interpreting
However, a narrower sector will be available for viewing.
3D echo requires further training. The views obtained are
very different from the familiar, standard 2D echo views.
Furthermore, at least initially, image cropping, manipulation
Image Display and Manipulation and display take extra time.
The standard echo views of 2D echo are familiar to all of us. We would advocate a focussed 3D examination. After a
During our echo training, we were taught to think and view complete 2D echo examination, a gated 3D echo data set
1402 J. Poon et al.

should be obtained from the apical window to allow quanti- echo methods. In CMR, LV volumes are obtained with the
fication of left ventricular volumes and ejection fraction. endocardial border traced inside the trabeculations whereas
Further focussed 3D examination may be performed depend- in 3D echo it is often traced where the trabeculations meet the
ing on the pathologies identified on 2D echo or the areas of LV cavity. The inclusion of basal LV slices may also play a
interests. part in the larger LV volumes obtained with CMR [4].
Full-volume multi-beat acquisitions are traditionally rec-
ommended for LV volume and function assessment as they
Left Ventricular Volumetric and offer the best available 3D resolution and wider sector width.
The wider section width is potentially important in encom-
Functional Assessment passing the whole left ventricle especially in patients whose
Quantification of LV volumes and function is the area where left ventricles are dilated. However, breath-holding and ECG
3D echo has been consistently shown to be superior to 2D gating are required. Inability to breath hold or irregular heart
echo. We recommend a focussed 3D examination of the left rhythm may result in stitching artifacts. Single-beat acquis-
ventricle in the apical window in all cases. There are three itions are feasible and have been shown to provide values
different approaches used for LV volume quantification: 3D similar to those obtained using multi-beat acquisition [13–15]
guided bi-plane analysis, real-time tri-plane quantification A semi-automated algorithm for endocardial border detec-
and direct 3D volumetric analysis. Three-dimensional tion is now available in commercially available echo
guided bi-plane analysis is performed by obtaining anatomi- machines provided by major vendors (Figure 1A and IB).
cally correct, non-foreshortened apical two- and four-cham- This has been showed to be feasible, time-efficient and repro-
ber views from a 3D echo dataset. Left ventricular volumes ducible in LV volumes and ejection measurements [16,17]. A
are then measured with Simpson’s biplane method. Real- recent study showed inter-vendor consistency of measure-
time tri-plane technique simultaneously acquires the three ment of LV volumes and ejection fraction as compared with
standard LV apical views within a single cardiac cycle, and is CMR. However, the use of automated endocardial border
particularly suited for rapid acquisition and analysis or in detection algorithms significantly underestimated LV vol-
patients with an irregular heart rhythm. Direct 3D volumetric umes, and the degree of underestimation was higher with
analysis is based on semi-automatic detection of endocardial larger LV volumes [10].
borders using a deformable shell model [2–4]. Left ventricu- In patients with suboptimal image quality, intravenous
lar volumes can be measured throughout the entire cardiac administration of echo contrast during 3D echo results in
cycle and displayed as a time varying curve with the elec- improved determination of LV volumes with reduced inter-
trocardiogram (Figure 1A and B). reader variability and can thus be proposed as an acceptable
Left ventricular volumes and ejection fraction obtained by alternative when MRI cannot be performed [9,18,19].
3D echo have been consistently shown to show lower intra-
and inter-observer, as well as lower test-retest, variability.
Furthermore, 3D echo derived LV volumes have been dem- Assessment of Mitral Valve
onstrated to have better correlation with that obtained with
CMR [5–10]. In a meta-analysis which included 28 studies
Pathologies
and 1,198 patients comparing 3D and 2D echo with CMR, 2D Assessment of mitral valve pathologies is another area of
echo underestimated LV end diastolic volumes by cardiac imaging where 3D echo should form part of routine
33  10 mL and end systolic volume by 16  5 mL. Underes- clinical practice. Both transthoracic and transoesophageal 3D
timation by 3D echo was consistently less: end diastolic echo allow comprehensive assessment of mitral valve pathol-
volume by 14  5 ml and end systolic volume by 7  3 ml. ogies. In particular, the ability to display en face views and
Of note, despite the underestimation of LV volumes by 3D views from both the left atrial as well as LV aspects of the
echo, there was excellent agreement for 3D echo-derived LV mitral valve apparatus are distinct advantages of 3D echo. In
ejection fraction with CMR-derived values [11]. In a meta- particular, en face views of the mitral valve from the left
analysis of 12 studies comparing 2D and 3D echo in measur- atrium showing its relationships with neighbouring struc-
ing LV volumes and ejection fraction, both the underestima- tures are particularly important for cardiothoracic surgeons
tion of LV volumes and the variabilities were lower with 3D as this is the view of the mitral valve they are very familiar
echo [8]. Due to its lower test-retest variability, 3D echo with. Having these anatomical views are especially helpful in
derived LV volumes and ejection fraction are ideally suited surgical planning.
for serial monitoring of LV function. Changes in LV volumes
assessed by serial 3D echo showed better correlations than Mitral Stenosis
that obtained with CMR [12]. The utilities of 3D echo for the assessment of mitral stenosis
The underestimation of LV volumes by 3D echo compared have been established [20–23]. Compared with 2D echo and
with CMR may be related to the superior image quality of Doppler pressure half time methods, mitral valve areas by
CMR, with better endocardial border definition as well as the direct planimetry of 3D echo images had better agreement
different tracing methods of 3D echo and CMR. This under- with that obtained with the Gorlin equation during cardiac
estimation may not be due to inherent inaccuracies of the 3D catheterisation [20,21]. Three-dimensional echo were also
Three Dimensional Echo in Clinical Practice 1403

Figure 1 (A) Measurement of left ventricular (LV) volumes and ejection fraction by a semi automated algorithm for
endocardial border detection with three-dimensional (3D) echo from the apical window. In this case, the LV end diastolic
volume (EDV) is 177 mL and end systolic volume (ESV) is 77 mL, corresponding to an ejection fraction of 56%. The stroke
volume (SV) and cardiac output (CO) are instantly displayed together with a time varying volume curve shown at the right
lower corner. (B) Measurement of left ventricular (LV) volumes and ejection fraction by a semi automated algorithm for
endocardial border detection with three-dimensional (3D) echo from the apical window in a patient with previous anterior
myocardial infarction. In this case, the LV end diastolic volume (EDV) is 295 mL and end systolic volume (ESV) is 205 mL,
corresponding to an ejection fraction of 30%. The stroke volume (SV) and cardiac output (CO) are instantly displayed
together with a time varying volume curve shown at the right lower corner. Furthermore, it can easily be appreciated that the
normal early diastolic filling, diastasis and late diastolic filling are abolished compared with the example in Figure 1A.
1404 J. Poon et al.

associated with lower intra-observer and inter-observer var- Quantification of severity of mitral regurgitation is where
iability [20,21]. As 3D echo allows direct, perpendicular, en 3D echo is potentially more accurate than 2D echo. Unlike 2D
face views of the mitral valve from both the left atrial and the echo, direct visualisation and measurement of the regurgi-
LV aspects, the cut-planes with the smallest valve area can be tant orifice area and en face views of the vena contracta are
obtained (Figure 2). Furthermore, 3D echo provides detailed possible with 3D echo with colour Doppler. Different aeti-
anatomical information on the commissures and the subvalv- ologies of mitral regurgitation are associated with different
ular apparatus, important in planning an interventional shaped vena contractae [24,25]. While a non-hemispherical
approach by percutaneous transseptal mitral valvuloplasty. shape is observed in the majority of the patients (in contrast
to what was assumed with the proximal isovelocity surface
Mitral Regurgitation area methods with 2D echo), a more elongated semilunar-
Assessment of mitral valve morphology in patients with shaped vena contracta along the line of incomplete leaflet
mitral regurgitation, especially in those with excessive leaflet coaptation is seen with functional mitral regurgitation [25].
motion, is the area where 3D echo, especially 3D TOE, is As 3D echo colour Doppler flow does not involve any geo-
particularly helpful. While most echocardiographers may do metric assumptions, it may be more accurate than 2D echo for
their 3D reconstruction in their heads, having the mitral quantification of severity of mitral regurgitation. It has been
valve morphologies displayed in anatomically understand- shown to be associated with less underestimation of regur-
able views is important for surgeons for their surgical plan- gitant volume for different aetiologies and severity of mitral
ning. The exact location of prolapse, and ruptured chordae regurgitation than 2D flow proximal isovelocity surface area
tendinae if present, can be easily visualised (Figure 3). methods in vitro [26,27], as well as in humans [24,26]. With
Specialised software is available for delineating mitral 3D echo colour Doppler imaging, it is also possible to mea-
valve anatomy in real time for 3D echo. The mitral annulus sure regurgitant jet volumes. While this application of 3D
and the leaflets can be traced and the resultant 3D model echo with colour Doppler is promising, we often find the
shows the saddle-shaped mitral annulus and can display frame rate with real time 3D echo colour Doppler too low for
information like mitral leaflet area, annulus height and diagnostic purposes. Although a full volume 3D dataset with
dimensions, tenting height and area. While the 3D model colour Doppler flow is possible with ECG gating and breath
and the measurements may be useful for surgical planning, hold, it is often hampered by an irregular heart rhythm and
at this stage we do not feel these software add clinically breathless patients, which is commonly seen in patients with
significant incremental benefits. significant mitral regurgitation.

Figure 2 Three-dimensional transoesophageal echo in a patient with mitral stenosis. A direct, perpendicular, en face view of
the mitral valve with the cut-plane at the smallest valve area can be obtained. By planimetry, the mitral valve area is 0.75 cm2.
Three Dimensional Echo in Clinical Practice 1405

Figure 3 (A) Real time live three-dimensional (3D) transoesophageal echo with the zoomed 3D view of the mitral valve from
the left atrium. The aortic valve (red arrow) is continuous with the anterior mitral leaflet showing prolapse of the A2 scallop
(green arrow). There is prolapse of the corresponding P2 scallop (yellow arrow). The anterolateral commissure of the mitral
valve can be identified with its proximity to the left atrial appendage (blue arrow). (B) Real time live three-dimensional (3D)
transoesophageal echo with the zoomed 3D view of the mitral valve from the left atrium. In this case, there is prolapse and
flail of the P2 scallop of the mitral valve with ruptured chordae (yellow arrow). The aortic valve (red arrow) is on the opposite
side of the leaflet identifying the flail segment being part of the posterior leaflet.
1406 J. Poon et al.

Figure 4 Direct en face view of an ostium secundum atrial septal defect from the left atrial side (as indicated by the small
arrows from the left atrium on the two-dimensional (2D) images). It can be immediately appreciated that the defect is not
circular but oval in shape.

the display of the complex spatial relationships between

Congenital Heart Disease cardiac structures more readily than 2D TOE. Real time
Due to its ability to display complex anatomy in 3D, 3D echo 3D TOE is a particularly useful technique during cardiac
is theoretically ideally suited for complex congenital heart interventions. Although 2D TOE is often sufficient in guiding
disease. However, as a result of limitations of image quality, cardiac interventions, the addition of real time 3D TOE does
especially in adults, the spatial resolution obtained with 3D make the appreciation of complex 3D anatomy quicker and
echo is inferior to that obtained with cardiac CT and CMR. easier to comprehend for the interventionist.
One area of assessment of congenital heart disease where 3D Percutaneous transseptal mitral balloon valvuloplasty is
echo, in particular, 3D TOE, is of value is the assessment of an accepted technique in the treatment of rheumatic mitral
ostium secundum atrial septal defects (ASD). stenosis. Three-dimensional TOE guidance is helpful in
Most secundum ASD can be closed percutaneously with transseptal puncture and in aiding balloon crossing of the
an Amplatzer occluder device. Accurately sizing the defect is stenotic mitral valve (Figure 5). Three-dimensional TOE is
essential for choosing the appropriate occluder to be used. As also helpful in guiding percutaneous mitral valve repair with
most ASDs are oval in shape, the diameter of the ASD may be the MitraClip, which has been shown to be effective in
underestimated and hence 2D echo may lead to undersizing selected patients with suitable anatomy.
of the device. As direct perpendicular en face views of ASD Three-dimensional TOE is not only useful in assessing the
can be obtained with 3D echo, more accurate sizing of the shape and size of ASD as discussed earlier, it is helpful in
defect is possible [28] (Figure 4). guiding their percutaneous closure. The correct sizing, posi-
tioning and apposition of the occluder device during the
procedure can be guided by real time 3D echo (Figure 6).
Guidance of Interventional Three-dimensional TOE provides helpful information
before, during and after percutaneous transcatheter aortic
Procedures valve implantation. Selection of the appropriate prosthesis
Transoesophageal echocardiography is a valuable tool to size is important, as under-sizing increases the risks of para-
guide cardiac interventions. Three-dimensional TOE allows valvular aortic regurgitation and over-sizing may increase
Three Dimensional Echo in Clinical Practice 1407

Figure 5 (A) Guiding transseptal puncture with live three-dimensional (3D) transoesophageal echo. (B) Guiding crossing of
a stenotic mitral valve with live three-dimensional (3D) transoesophageal echo with the Inuoe balloon. View from the left
atrium.

the risks of aortic rupture and conduction block. Three- dimensional echo or TOE may not allow accurate assessment
dimensional TOE allows exact characterisation of the aortic of the exact shape and size of the aortic annulus [29]. In
annular geometry and sizes, as it is well documented that the addition, procedural complications such as valvular or para-
aortic annulus is oval-shaped rather than circular. Two- valvular regurgitation, aortic injury, myocardial infarction or
1408 J. Poon et al.

Figure 6 Guiding percutaneous closure of atrial septal defect with real time three-dimensional (3D) transoesophageal echo.
The Amplatzer device (St Jude Medical, Uppsala, Sweden) is viewed from the left atrium before release. The left atrial disc is
well opposed. The device is still seen attached to the delivery mechanism.

pericardial tamponade can be evaluated immediately during software with automatic border detection allows assessment
the procedure. Although cardiac CT also allows accurate of both 3D LV GLS and circumferential and radial strain
assessment of aortic annular size, and the presence of para- without foreshortening (Figure 7).
valvular regurgitation post-procedure can be assessed with
contrast aortogram, the use of 3D TOE does allow assessment
by one imaging technique in real time. Right Ventricular Assessment
Other procedures where guidance with 3D TOE may be of
Echo assessment of right ventricular (RV) volumes and func-
benefit include percutaneous closure of peri-prosthetic mitral
tion remains challenging in clinical practice because of its
regurgitation and left atrial appendage closure.
complex geometry and its unique crescent shape. Its anterior
position in the chest poses additional difficulties because of
suboptimal image quality in many patients. Conventional
Left Ventricular Strain echo parameters such as fractional area change, tricuspid
Left ventricular global longitudinal strain (GLS) has gained annular plane excursion and tissue Doppler S’ are recom-
wide clinical acceptance and is a useful clinical tool. It is a mended measures to assess RV systolic function. However,
sensitive parameter for LV dysfunction and allows early detec- they have major limitations of being angle-dependent, load-
tion of dysfunction when LV ejection fraction is still within the dependent and not fully representing RV global function.
normal range [30]. Furthermore, LV GLS provided incremental Three-dimensional echo has gained increasing acceptance
prognostic value independent of LV ejection fraction in a wide for evaluation of RV volumes and function. Several studies
range of conditions [31]. Traditionally, LV GLS is obtained have shown that it correlated well with CMR-derived vol-
from the three apical views of the left ventricle. One limitation umes in different patient groups [32–35].
inherent in this approach is that the three apical views are Similar to the situation with LV volumes, RV volumes
necessarily from three different heart beats and the global were underestimated by 3D echo, particularly in patients
values are not actually obtained from the same heart beat. This with larger RV volumes and ejection fraction [36]. A volu-
is particularly important in patients with irregular heart metric semi-automated border detection software is available
rhythm. Another pitfall is that foreshortening of the left ven- for RV volume assessement. While 3D echo assessment of RV
tricle may lead to inaccurate assessment of longitudinal strain. volumes and function has shown promise, its incorporation
Three-dimensional echo with a full volume acquisition in the into routine clinical practice is still hampered by the need for
apical windows allows assessment of true LV GLS in the same excellent image quality and, in our opinion, cumbersome
heart beat and avoids foreshortening. Semi-automated software packages.
Three Dimensional Echo in Clinical Practice 1409

Conclusions
Three-dimensional echo has come a long way since its first
inception. With current technologies, a full 3D examination
in routine clinical practice is feasible. The main advantages of
3D echo are the infinite cut planes possible allowing direct en
face, and anatomical views of cardiac structures, thereby
avoiding foreshortened views of cardiac structures. Three-
dimensional echo can circumvent any geometric assump-
tions of the cardiac chambers inherent in any 2D echo tech-
niques. While 3D echo has been shown to be helpful in
assessing most cardiac structures, it is still affected by image
quality, subjected to artifacts seen with 2D echo and the
compromise between spatial and temporal resolution. This
compromise is imposed by the physics of ultrasound which
any future technical advances are unlikely to resolve. In
routine clinical practice in 2019, we recommend a focussed
3D echo examination after a full 2D echo study. The area
where 3D echo has been consistently shown in multiple
studies to have superior accuracy and reproducibility over
2D echo is in the assessment of LV volumes and ejection
fraction. We recommend obtaining a full volume 3D echo
data set from the apical window for measurement of LV
volumes and GLS. Further 3D examination can be performed
depending on the pathologies identified on 2D examination.
Assessment of mitral valve pathologies and ASD with 3D
echo should be used in routine clinical practice. Furthermore,
real time 3D TOE is a very useful technique in guiding
structural cardiac intervention, both before, during and after
the procedure. While 3D echo is not the holy grail of echo-
cardiography, it does represent a useful technique in selected
areas of cardiac imaging.

References
[1] Lang RM, Badano LP, Tsang W, Adams DH, Agricola E, Buck T, et al.
EAE/ASE recommendations for image acquisition and display using
three-dimensional echocardiography. Eur Heart J Cardiovasc Imaging
2012;13:1–46.
[2] Jacobs LD, Salgo IS, Goonewardena S, Weinert L, Coon P, Bardo D, et al.
Rapid online quantification of left ventricular volume from real-time
Figure 7 Semi-automated software with automatic bor- three-dimensional echocardiographic data. Eur Heart J 2006;27:460–8.
[3] Soliman OI, Kirschbaum SW, van Dalen BM, van der Zwaan HB, Mah-
der detection for assessment of three-dimensional (3D) davian Delavary B, Vletter WB, et al. Accuracy and reproducibility of
left ventricular (LV) strain in all three axes in a single quantitation of left ventricular function by real-time three-dimensional
heart beat: longitudinal (7 A), circumferential (7B) and echocardiography versus cardiac magnetic resonance. Am J Cardiol
radial strain (7C). 2008;102:778–83.
[4] Mor-Avi V, Jenkins C, Kuhl HP, Nesser HJ, Marwick T, Franke A, et al.
Real-time 3-dimensional echocardiographic quantification of left ventric-
ular volumes: multicenter study for validation with magnetic resonance
imaging and investigation of sources of error. JACC Cardiovasc Imaging
Other Assessments 2008;1:413–23.
[5] Jenkins C, Bricknell K, Hanekom L, Marwick TH. Reproducibility and
Although 3D echo has been used in assessment of tricuspid accuracy of echocardiographic measurements of left ventricular param-
valve, left and right atrial volume and LV dyssynchrony, we eters using real-time three-dimensional echocardiography. J Am Coll
Cardiol 2004;44:878–86.
feel that their incorporation into routine clinical practice
[6] Caiani EG, Corsi C, Zamorano J, Sugeng L, MacEneaney P, Weinert L,
cannot be recommended in 2019. In particular, the main et al. Improved semiautomated quantification of left ventricular volumes
problem with 3D echo assessment of LV dyssynchrony is and ejection fraction using 3-dimensional echocardiography with a full
matrix-array transducer: comparison with magnetic resonance imaging. J
the limited frame rate, thereby limiting the temporal resolu-
A Soc Echocardiogr 2005;18:779–88.
tion in the assessment leading to significant under-sampling [7] Greupner J, Zimmermann E, Grohmann A, Dubel HP, Althoff TF, Borges
and uncertainties. AC, et al. Head-to-head comparison of left ventricular function
1410 J. Poon et al.

assessment with 64-row computed tomography, biplane left cineventri- balloon mitral valvuloplasty: role of real-time 3D echocardiography. Eur
culography, and both 2- and 3-dimensional transthoracic echocardiog- Heart J 2004;25:2086–91.
raphy: comparison with magnetic resonance imaging as the reference [22] Xie MX, Wang XF, Cheng TO, Wang J, Lu Q. Comparison of accuracy of
standard. J Am Coll Cardiol 2012;59:1897–907. mitral valve area in mitral stenosis by real-time, three-dimensional
[8] Dorosz JL, Lezotte DC, Weitzenkamp DA, Allen LA, Salcedo EE. Perfor- echocardiography versus two-dimensional echocardiography versus
mance of 3-dimensional echocardiography in measuring left ventricular Doppler pressure half-time. Am J Cardiol 2005;95:1496–9.
volumes and ejection fraction: a systematic review and meta-analysis. J [23] Binder TM, Rosenhek R, Porenta G, Maurer G, Baumgartner H.
Am Coll Cardiol 2012;59:1799–808. Improved assessment of mitral valve stenosis by volumetric real-time
[9] Hoffmann R, Barletta G, von Bardeleben S, Vanoverschelde JL, Kasprzak three-dimensional echocardiography. J Am Coll Cardiol 2000;36:1355–61.
J, Greis C, et al. Analysis of left ventricular volumes and function: a [24] Yosefy C, Levine RA, Solis J, Vaturi M, Handschumacher MD, Hung J.
multicenter comparison of cardiac magnetic resonance imaging, cine Proximal flow convergence region as assessed by real-time 3-dimen-
ventriculography, and unenhanced and contrast-enhanced two-dimen- sional echocardiography: challenging the hemispheric assumption. J
sional and three-dimensional echocardiography. J Am Soc Echocardiogr Am Soc Echocardiogr 2007;20:389–96.
2014;27:292–301. [25] Kahlert P, Plicht B, Schenk IM, Janosi RA, Erbel R, Buck T. Direct
[10] Muraru D, Cecchetto A, Cucchini U, Zhou X, Lang RM, Romeo G, et al. assessment of size and shape of noncircular vena contracta area in
Intervendor consistency and accuracy of left ventricular volume meas- functional versus organic mitral regurgitation using real-time three-
urements using three-dimensional echocardiography. J Am Soc Echo- dimensional echocardiography. J Am Soc Echocardiogr 2008;21:912–21.
cardiogr 2018;31. 158-168 e1. [26] Plicht B, Kahlert P, Goldwasser R, Janosi RA, Hunold P, Erbel R, et al.
[11] Rigolli M, Anandabaskaran S, Christiansen JP, Whalley GA. Bias associ- Direct quantification of mitral regurgitant flow volume by real-time
ated with left ventricular quantification by multimodality imaging: a three-dimensional echocardiography using dealiasing of color Doppler
systematic review and meta-analysis. Open Heart 2016;3:e000388. flow at the vena contracta. J Am Soc Echocardiogr 2008;21:1337–46.
[12] Jenkins C, Bricknell K, Chan J, Hanekom L, Marwick TH. Comparison of [27] Little SH, Pirat B, Kumar R, Igo SR, McCulloch M, Hartley CJ, et al. Three-
two- and three-dimensional echocardiography with sequential magnetic dimensional color Doppler echocardiography for direct measurement of
resonance imaging for evaluating left ventricular volume and ejection vena contracta area in mitral regurgitation: in vitro validation and clinical
fraction over time in patients with healed myocardial infarction. Am J experience. JACC Cardiovasc Imaging 2008;1:695–704.
Cardiol 2007;99:300–6. [28] Pepi M, Tamborini G, Bartorelli AL, Trabattoni D, Maltagliati A, De Vita
[13] Yang LT, Nagata Y, Otani K, Kado Y, Otsuji Y, Takeuchi M. Feasibility of S, et al. Usefulness of three-dimensional echocardiographic reconstruc-
One-Beat Real-Time Full-Volume Three-Dimensional Echocardiography tion of the Amplatzer septal occluder in patients undergoing atrial septal
for Assessing Left Ventricular Volumes and Deformation Parameters. J closure. Am J Cardiol 2004;94:1343–7.
Am Soc Echocardiogr 2016;29. 853-860 e2. [29] Ng AC, Delgado V, van der Kley F, Shanks M, van de Veire NR, Bertini
[14] Ren B, Vletter WB, McGhie J, Soliman OI, Geleijnse ML. Single-beat real- M, et al. Comparison of aortic root dimensions and geometries before and
time three-dimensional echocardiographic automated contour detection after transcatheter aortic valve implantation by 2- and 3-dimensional
for quantification of left ventricular volumes and systolic function. Int J transesophageal echocardiography and multislice computed tomogra-
Cardiovasc Imaging 2014;30:287–94. phy. Circ Cardiovasc Imaging 2010;3:94–102.
[15] Spitzer E, Ren B, Zijlstra F, Mieghem NMV, Geleijnse ML. The role of [30] Ng AC, Delgado V, Bertini M, van der Meer RW, Rijzewijk LJ, Shanks M,
automated 3d echocardiography for left ventricular ejection fraction et al. Findings from left ventricular strain and strain rate imaging in
assessment. Card Fail Rev 2017;3:97–101. asymptomatic patients with type 2 diabetes mellitus. Am J Cardiol
[16] Myhr KA, Pedersen FHG, Kristensen CB, Visby L, Hassager C, Mogel- 2009;104:1398–401.
vang R. Semi-automated estimation of left ventricular ejection fraction by [31] Stanton T, Leano R, Marwick TH. Prediction of all-cause mortality from
two-dimensional and three-dimensional echocardiography is feasible, global longitudinal speckle strain: comparison with ejection fraction and
time-efficient, and reproducible. Echocardiography 2018;35:1795–805. wall motion scoring. Circ Cardiovasc Imaging 2009;2:356–64.
[17] Aurich M, Andre F, Keller M, Greiner S, Hess A, Buss SJ, et al. Assess- [32] Sugeng L, Mor-Avi V, Weinert L, Niel J, Ebner C, Steringer-Mascherba-
ment of left ventricular volumes with echocardiography and cardiac uer R, et al. Multimodality comparison of quantitative volumetric analy-
magnetic resonance imaging: real-life evaluation of standard versus sis of the right ventricle. JACC Cardiovasc Imaging 2010;3:10–8.
new semiautomatic methods. J Am Soc Echocardiogr 2014;27:1017–24. [33] Jenkins C, Chan J, Bricknell K, Strudwick M, Marwick TH. Reproduc-
[18] Saloux E, Labombarda F, Pellissier A, Anthune B, Dugue AE, Provost N, ibility of right ventricular volumes and ejection fraction using real-time
et al. Diagnostic value of three-dimensional contrast-enhanced echocar- three-dimensional echocardiography: comparison with cardiac MRI.
diography for left ventricular volume and ejection fraction measurement Chest 2007;131:1844–51.
in patients with poor acoustic windows: a comparison of echocardiogra- [34] van der Zwaan HB, Helbing WA, McGhie JS, Geleijnse ML, Luijnenburg
phy and magnetic resonance imaging. J Am Soc Echocardiogr SE, Roos-Hesselink JW, et al. Clinical value of real-time three-dimen-
2014;27:1029–40. sional echocardiography for right ventricular quantification in congenital
[19] Gurunathan S, Karogiannis N, Senior R. Imaging the heart failure heart disease: validation with cardiac magnetic resonance imaging. J Am
patient-need for accurate measurements of left ventricular volumes Soc Echocardiogr 2010;23:134–40.
and ejection fraction: the role of three-dimensional and contrast echo- [35] Tamborini G, Marsan NA, Gripari P, Maffessanti F, Brusoni D, Muratori
cardiography. Curr Opin Cardiol 2016;31:459–68. M, et al. Reference values for right ventricular volumes and ejection
[20] Zamorano J, Cordeiro P, Sugeng L, Perez de Isla L, Weinert L, Macaya C, fraction with real-time three-dimensional echocardiography: evaluation
et al. Real-time three-dimensional echocardiography for rheumatic in a large series of normal subjects. J Am SocEchocardiogr 2010;23:109–15.
mitral valve stenosis evaluation: an accurate and novel approach. J [36] Shimada YJ, Shiota M, Siegel RJ, Shiota T. Accuracy of right ventricular
Am Coll Cardiol 2004;43:2091–6. volumes and function determined by three-dimensional echocardiogra-
[21] Zamorano J, Perez de Isla L, Sugeng L, Cordeiro P, Rodrigo JL, Almeria C, phy in comparison with magnetic resonance imaging: a meta-analysis
et al. Non-invasive assessment of mitral valve area during percutaneous study. J Am Soc Echocardiogr 2010;23:943–53.