C Johnson HWbDO Speckle tracking 6:3 R87–R98

echocardiography: tips and tricks

REVIEW

Practical tips and tricks in measuring strain,

strain rate and twist for the left and right
ventricles

&KULVWRSKHU-RKQVRQb06F1, .DWKHULQH.X\Wb%6F2, 'DYLG2[ERURXJKb3K'1,† and 0DUWLQ6WRXWb3K'2,†

1Research institute for Sport and Exercise Sciences, Liverpool John Moores University, Liverpool, UK
2School of Healthcare Science, Manchester Metropolitan University, Manchester, UK

&RUUHVSRQGHQFHVKRXOGEHDGGUHVVHGWR'2[ERURXJKGOR[ERURXJK#OMPXDFXN

*(C Johnson and K Kuyt contributed equally as joint lead authors)

†(D Oxborough and M Stout contributed equally as joint senior authors)

$EVWUDFW
Strain imaging provides an accessible, feasible and non-invasive technique to assess Key Words
cardiac mechanics. Speckle tracking echocardiography (STE) is the primary modality with f left ventricle
the utility for detection of subclinical ventricular dysfunction. Investigation and adoption of f right ventricle
WKLVWHFKQLTXHKDVLQFUHDVHGVLJQLȴFDQWO\LQERWKWKHUHVHDUFKDQGFOLQLFDOHQYLURQPHQWΖW f strain
is therefore important to provide information to guide the sonographer on the production f strain rate
of valid and reproducible data. The focus of this review is to (1) describe cardiac physiology f twist
and mechanics relevant to strain imaging, (2) discuss the concepts of strain imaging and f deformation
STE and (3) provide a practical guide for the investigation and interpretation of cardiac f speckle tracking
mechanics using STE. echocardiography

ΖQWURGXFWLRQ coronary blood flow. Electrophysiological processes

propagate contraction of cardiac muscle filaments via
The assessment of myocardial function provides a core
excitation–contraction coupling. However, myocytes are
component of the echocardiographic examination.
known to only contract by approximately 15–20% (1). To
Traditionally this has been underpinned by the
maximise efficiency, myocardial fibre arrangement and
assessment of left ventricular ejection fraction (LVEF),
interaction becomes integral to produce adequate stroke
but increasingly speckle tracking echocardiography (STE)
volume for each cardiac cycle.
is being used in both research and clinical environments
Each myocyte is arranged in a ‘scaffold’ that maintains
to directly assess myocardial wall deformation (strain).
the arrangement of cardiac cells and the muscle fibres in
This article reviews the underlying physiology, introduces
the required configuration (2). In the left ventricle (LV),
the concepts of strain and provides a practical guide for
the myocardium is arranged in three distinct layers. The
acquisition, analysis and subsequent interpretation.
sub-endocardium (innermost), the mid-myocardium
and the sub-epicardium (outermost). The interaction of
&DUGLDFPHFKDQLFVDQDWRP\DQGSK\VLRORJ\ these layers plays a key role in ensuring stroke volume
and hence overall cardiac output. The LV uniquely
In order to maintain metabolic demands, cardiac possesses a double helical myocardial fibre arrangement.
architecture links closely with electrical activation and That is, fibres in the sub-endocardium are arranged in a

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 R88
echocardiography: tips and tricks

left-handed helix, then smoothly transition to a transverse

circular arrangement in the midmyocardium and then
finally a right-handed helix in the epicardium (3) (Fig. 1).
During ventricular systole, there is shortening of the
muscle fibres in all three layers of the LV myocardium
subsequently leading to reductions in both LV length and
circumference. As a result of the myocardium being an
incompressible structure, there is resultant thickening
of the LV myocardial walls (the thickening of the LV
myocardium as seen on 2D echocardiography). This
myocardial deformation can be termed longitudinal,
circumferential and radial. In addition, as all three layers of
fibres shorten in ventricular systole (but largely governed
by the larger radius of the subepicardial fibres), there is
rotation of the LV base in a clockwise direction and LV )LJXUHb
apex in a anticlockwise direction, commonly termed LV 0\RFDUGLDOGHIRUPDWLRQUHODWLQJWRȴEUHDUFKLWHFWXUH$GDSWHGZLWK
permission, from Nakatani 2011 (4).
twist (Fig. 2) (4).
LV deformation largely governs the overall cardiac
comprises multiple fibre layers, which can be split into
contraction and as such impacts significantly on the
the superficial and deep. The deep muscle fibres in the
deformation of the other chambers. The interventricular
RV are arranged longitudinally, whereas the superficial
septum is shared by both ventricles and septal deformation
(subepicardial) layers are arranged circumferentially,
properties are largely controlled by those of the LV, but
and parallel to the atrioventricular groove (7). The RV
septal movement is also integral to right ventricle (RV)
superficial fibres connect to the myofibers of the LV via
contraction and stroke volume (5).
the cardiac apex, thereby adding further LV influence
The RV has a complex geometry that varies in
over RV deformation (5).
appearance depending on the plane of imaging. The RV
The RV also deforms longitudinally, circumferentially
appears triangular when viewed from the side, yet in
and radially. However, the key overall deformation is
cross section, it appears crescent shaped (5). Physiological
shortening of the longitudinal fibres during ventricular
loading conditions result in the interventricular septum
systole (8). RV free wall radial contraction can be noted
being shaped concavely towards the LV in both systole
visibly during transthoracic echocardiography (TTE) but
and diastole. The volume of the RV is larger than that
is less in magnitude than the LV because of its higher
of the LV in a normal healthy adult, but the RV walls
surface-to-volume ratio. Finally, there is traction of the
are much thinner with overall RV mass approximately
RV-free wall at its points of LV attachment, secondary to LV
one-sixth of LV mass (6). The RV myocardium also
deformation (5). There is no evidence of a significant layer
of obliquely orientated myocardial fibres associated with
the RV; therefore, there is no powerful influence of twist
mechanics to overall RV contraction. Circumferential and
radial RV deformation is currently not routinely measured
due partially to less influence on RV function, and the
limitations of commercially available technology that can
accurately track the thinner walls of the RV.

6WUDLQPHDVXUHPHQWDQG67(

Strain is a measure of deformation of the myocardium

during the cardiac cycle in multiple directions;
lengthening, shortening and thickening. STE is a common
)LJXUHb
0\RFDUGLDOȴEUHDUUDQJHPHQWRIWKHVXEHQGRPLGDQGHSLFDUGLXP technique used to measure strain. STE is based on grey-
Reproduced, with permission, from Nakatani 2011 (4). scale images obtained during echocardiography and has

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020
 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 R89
echocardiography: tips and tricks

been developed to be angle independent, a key advantage

over previous methods used to derive strain. These grey-
scale echocardiographic images are composed of several
bright speckles resulting from the ultrasound-myocardial
tissue interaction. Specialist software is able to identify the
speckles and track them frame by frame during the cardiac
cycle using either of two processes known as the sum of
)LJXUHb
the absolute differences or Fourier analysis (Fig. 3) (9).
Simple diagram showing the principle of Lagrangian strain. Following
The sum of absolute differences algorithm fits correlation- deformation L(t) there is a change in length from initial reference L(t0).
weighted data into a spatial polynomial curve in order to Resulting in an increased length by ΔL. This increase in length will be
expressed as a positive percentage value.
calculate regional strain at approximately 3 mm intervals
(10). Fourier analysis assumes coherence of the geometry
This figure is then divided by original reference length
tracked using a sequence of intermediate passages from
and then expressed as a percentage.
2 cm down to 5 pixel bands (11). Software platforms are
SR simply describes the rate of deformation (i.e.
then able to resolve the magnitude of deformation in the
how quickly the deformation occurs within the cardiac
directions mentioned earlier to generate strain and strain
cycle). SR can also be calculated using a derivative of
rate (SR) curves (12).
the Lagrangian strain equation. As such, the following
During STE, deformation is calculated using the
equation explains SR based on Lagrangian strain:
Lagrangian method (12). Lagrangian strain is defined as
dSL (t ) 1 dL (t )
deformation from an original length and calculated using εR L (t ) = =
the following equation: dt L0 dt

⎡ L (t ) − L (t 0 )⎤⎦
ε L (t ) = ⎣ where, SRL is Lagrangian strain rate. L0 is the reference
L (t 0 ) length at time t0 (usually end diastole). Natural strain may
where, L(t) is the length of the object at time instance also be used to assess deformation and deformation rate.
t following deformation and L(t0) ≈ L0, that is, the length Natural strain uses a constantly changing reference length
of the object when not subject to eternal forces. ε1 is integrated from SR:
Lagrangian strain. t

In more simple terms, strain is expressed as a fractional ε N (t ) =

∫dε N (t )
length change, where shortening is a negative value and t0
lengthening a positive value. Figure 4 shows the concept
of Lagrangian strain. In order to calculate strain at any However, STE is calculated using Lagrangian strain
given time point, the difference between length at the because the baseline length (end-diastole) is always
reference point (end-diastole during echocardiography) known and utilised as a reference (12). Natural strain will
and the current length (usually end-systole) is calculated. therefore not be referred to further in this review.
It must be emphasised that as the heart is a 3D
structure, there are six ‘strains’ placed on the myocardium.
These include the three strains already mentioned
(longitudinal, radial and circumferential) and in addition,
three directions of ‘shear’ strain produced from the
movement of plane layers over each other (13). When
the myocardium contracts, the three layers of myocardial
muscle fibres each contract, and due to the different
orientation of the muscle fibres in each layer, they move
over one another in a shearing motion. Shear strain
plays a key role in both radial thickening and twisting
of the LV, enabling systolic thickening of the myocardial
wall significantly beyond the 10% thickening of each
)LJXUHb
The theoretical concept of speckle tracking echocardiography. Data from individual myocyte (14). Shear strain gradient increases
Bansal & Kasliwal 2013 (9). as you progress from the sub-epicardium towards the

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 R90
echocardiography: tips and tricks

sub-endocardium and correlates with increasing radial This is also problematic during higher HR’s and therefore it
thickening during systole. is important to maximise image quality and other settings
whilst subjectively monitoring myocardial tracking.

2SWLPDOLPDJHDFTXLVLWLRQ ΖPDJHTXDOLW\

All images used for LV and RV strain analysis are acquired STE relies on the acquisition of high-quality images.
as part of the minimum dataset for a standard adult TTE Images should be acquired using the optimal gain settings
(15). It is important however to consider the technical and breath-hold techniques to clearly delineate the
aspects of image acquisition that are specific to STE endocardial and epicardial borders and to avoid artefact
providing optimal images for post processing and hence related to excess noise, rib or lung movements and
maximising validity and reproducibility. It is essential to translational motion of the heart (18). Furthermore, image
acknowledge that there are numerous software platforms width and depth should be focused on the chamber of
that provide the functionality to post-process raw grey- interest. It is import to note that although STE is less angle
scale images for STE, each with individual nuances. The dependent than TDI measures of strain, the components
following sections provide insight into these technical of longitudinal and circumferential strains are opposite
considerations allowing for a standardised approach in polarity to radial strain therefore any deviation from
but with the aim to provide theoretical underpinning the major axis will result in a progressive reduction in
irrespective of vendor differences. strain values in the relative axis (19). In view of this, it
is essential to avoid apical foreshortening in the apical
views and ensure circular LV chamber in the parasternal
(OHFWURFDUGLRJUDSKLF (&* JDWLQJ
short axis (PSAX) views whilst avoiding the use of non-
ECG gating is required for timing of events throughout standardised views (20). Suboptimal image acquisition
the cardiac cycle and is of utmost importance in STE may result in poor speckle tracking. If in a single view two
analysis. An optimal ECG signal with minimal heart rate or more segments are not adequately tracked, then the
(HR) variability should be present across three cardiac calculation of global strain values is restricted (21).
cycles in patients with normal sinus rhythm. The presence
of significant HR variability will limit the calculation of
6WDQGDUGLVHGYLHZV
global strain values, which is especially problematic in
patients with atrial fibrillation (16). The acquisition of STE analysis to produce LV global longitudinal strain
three cardiac cycles ensures at least one full cardiac cycle (GLS), global circumferential strain (GCS), global radial
is present, and the cycle with the optimal endocardial strain (GRS) and rotation, twist and torsion curves relies
delineation can be selected for speckle tracking. on the acquisition of specific views (Fig. 5). Three apical
views including the four chamber (A4CH), two chamber
(A2CH) and long axis (APLAX) allow the analysis of GLS.
)UDPHUDWH
GCS and GRS are processed from the PSAX view at the
Images should be maintained at a frame rate between basal level, defined at the level of the mitral valve (MV)
40 and 90 frames per second for analysis of cardiac leaflet tips and papillary muscle (PM) level, whilst a PSAX
deformation (17). As HR increases mechanical events view at the apical level, defined as the level just above the
throughout the cardiac cycle are shortened and therefore point of systolic cavity obliteration, allows the assessment
require a higher frame rate to allow optimal STE. of rotation/twist. RV GLS requires the acquisition of a
Therefore, investigations involving increased HR, such RV focused A4CH view. This view is achieved by lateral
as exercise and pharmacological stress testing, require a translation of the probe from the conventional A4CH
proportional increase in frame rate. Lower frame rates view. The optimal RV focused A4CH view should provide
may cause unsatisfactory STE due to reduced temporal the maximum RV basal diameter, visualising the whole
resolution and possible speckle drop out resulting in an of the RV-free wall, from the tricuspid lateral annulus to
underestimate strain values (18). Conversely, too high a the RV apex, whilst avoiding foreshortening of the RV
frame rate (>100 frames per second) may cause the STE apex, throughout the cardiac cycle (21, 22). In addition
algorithm to be unable to identify absolute change in the to 2D images, acquisition of spectral Doppler traces of the
speckle pattern and inadequately track the myocardium. MV, tricuspid valve (TV), aortic valve (AV) and pulmonary

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020
 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 5
echocardiography: tips and tricks

)LJXUHb
Standardised views required for speckle tracking
echocardiography. Left ventricular focused apical
views include (A) four–chamber, (B) two-chamber
and (C) long axis. Parasternal short axis views at
the level of (D) mitral valve, (E) papillary muscle
and (F) apex. Right ventricular focused apical view
(G) four chamber. LV, left ventricle; PSAX,
parasternal short axis; RV, right ventricle.

valve (PV), allow the definition of true LV and RV end- but their correlation was good (ICC 0.65, 95% CI
diastolic and end-systolic event timing. 0.42–0.78) (23). Medvedofsky et al. reported good
agreement between both contrast and non-contrast
images (ICC r = 0.85), and between contrast STE- and CMR-
&RQWUDVWHFKRFDUGLRJUDSK\
derived strain (ICC r = 0.83) (24). Nagy et al. investigated
The feasibility and utility of STE strain analysis in whether deformation analysis provided additional
contrast TTE studies is a controversial topic. Current diagnostic sensitivity beyond just wall motion scoring
EACVI guidelines for contrast TTE do not refer to (WMS) during dobutamine stress echocardiography.
strain measurements, and commercially available echo They concluded that although STE analysis was feasible,
machines do not allow the user to measure strain, whilst it did not add diagnostic benefit over expert WMS alone
the contrast protocol is active. However, there have (25). All of the above studies performed the STE analysis
been some studies investigating the use and feasibility retrospectively using offline specialist software.
of measuring strain during a contrast-enhanced TTE.
These have produced varied results. Zoppellaro et al. used
the flash replacement technique, a form of myocardial ΖPDJHDQDO\VLV
contrast echocardiography to visualise the myocardium
in 40 patients. Non-contrast and contrast images were Once optimal images have been acquired they should
then analysed for longitudinal strain. They found that be transferred to the appropriate workstation for post
the longitudinal strain calculated from the non-contrast processing. During post processing, it is important to
and contrast-enhanced images were statistically different replicate the analysis used in previous investigations of
(−18.8 ± 4.5% and −22.8 ± 5.4%, respectively; P < 0.001), the same patient.

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 5
echocardiography: tips and tricks

(YHQWWLPLQJ allows simultaneous visualisation of event timing (AVC

and MVC) and myocardial function (26). The optimal
The accurate definition of two reference points in each
surrogate, with the closest accuracy to 2D interrogation
region throughout the cardiac cycle is imperative for
of MVC and AVC, is spectral Doppler of the MV and AV,
successful tracking and subsequent analysis. These
respectively (26).
reference points relate to onset and termination of
myocardial contraction (end diastole and end systole,
respectively). Small deviations in definition of end
Region of interest
diastole and end systole have been reported to influence
strain values significantly and at a clinically relevant The assessment of STE is based on subjective selection of the
magnitude (26). optimal ROI when tracking the myocardial wall (Fig. 6). Due
End diastole is defined as the reference position/ to the complex anatomical structure of the myocardium
length to which systolic deformation is measured. End and the distinct contribution of each myocardial layer to
diastole and end systole correspond to mitral/tricuspid contraction, it is important to encompass the whole of
valve closure (MVC/TVC) and aortic/pulmonary valve the myocardium from the subendocardial border to the
closure (AVC/PVC) for the LV and RV, respectively. It is subepicardial border (13). In normal healthy individuals,
common for the software to automatically define or utilise the inner (subendocardial) layer contributes the most to
surrogate markers of event timings, such as automated longitudinal and circumferential strain, with a reduction
AVC detection algorithms or peak R wave detection on in the mid-myocardium, and the lowest contribution
the ECG for MVC. This can cause deviations from true from the outer (subepicardial) layer (27). Therefore, the
event timing in patients with conduction delays, and width of the ROI has significant effects on the strain values
therefore, differences in subsequent strain values. It is reported (20, 28). If the tracking area width is too narrow,
therefore recommended that the sonographer visually encompassing only the endocardial layer, higher strain
interrogates the APLAX view frame by frame as this view values will be reported (28). A tracking area width larger

)LJXUHb
Correct region of interest placement. Left
ventricular focused apical (A) four–chamber, (B)
two–chamber, (C) long axis, parasternal short axis
at the levels of the (D) mitral valve, (E) papillary
muscles, (F) apex and (G) right ventricular focus
apical four chamber.

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020
 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 5
echocardiography: tips and tricks

than the myocardium, encompassing the pericardium /9UDGLDOVWUDLQDQG65

and excess tissue (PMs, trabeculation, etc.), can result
Radial strain represents the change in wall thickness
in an underestimation of strain (28). Recommendations
(thickening in systole, represented as a positive strain value) of
suggest it is important for vendors to provide accurate
the myocardium perpendicular to the long axis and directed
information pertaining to the spatial extent of the ROI
towards the centre of the LV cavity. Radial SR represents the
(13). Further to this, care should be taken in the apical
rate of thickening of the LV, producing a positive systolic
views when selecting start and end points of the ROI
value and negative diastolic values. GRS is averaged from the
(Fig. 6). In the A4CH view the ROI should begin at the
same 12 regional segments as circumferential strain using
septal MV annulus, progress to the apex and end at the
the PSAX views at MV and PM level.
lateral MV annulus. In the APLAX view the ROI should
begin at the MV annulus of the posterior wall, progress
to the apex and end at the base of the septal wall /9URWDWLRQWZLVWDQGWRUVLRQ
taking care not to extend into the LV outflow tract. The
As well as the three previously defined LV mechanics
automatically generated ROI should be visually assessed
rotation of the myocardial architecture is present,
by the sonographer with manual correction of inadequate
expressed in degrees. Rotation occurs around the long
contouring recommended.
axis of the LV, with basal rotation in a clockwise direction
(negative value) and apical rotation in an anticlockwise
/9ORQJLWXGLQDOVWUDLQDQG65 direction (positive value), when viewed from the apex
during systole. From rotation values, twist and torsion are
Longitudinal strain represents the change in length calculated. Twist (°) is defined as the difference in apical
(shortening in systole, represented as a negative strain and basal systolic rotation when viewed from the apex,
value) of the myocardium along the long axis of the LV with torsion (°/cm) calculated as the twist angle divided
(base to apex). Longitudinal SR represents the rate at by distance between base and apex.
which the deformation occurs, producing a negative value
in systole and positive values in diastole. GLS is averaged
from 16, 17 or 18 regional segments across three apical 59ORQJLWXGLQDOVWUDLQDQG65
views (21). The segmental division at basal- and mid- There are discrepancies present within the literature
level are homogeneous across the three models, with the regarding the optimal ROI and subsequent segmental
inclusion of six segments at each level. Differences across model to investigate longitudinal strain within the RV,
the segmental models relate to the division of the apical with consensus yet to be achieved. These models include
level. With the 17-segment model dividing the apical a full RV chamber six-segmental model (incorporating
level into five segments (anterior, septal, inferior and the interventricular septum into the calculation of RV
lateral, with the addition of an ‘apical cap’). Alternatively global longitudinal strain; RV GLS) and a three-segmental
the 16- and 18-segment models divide the apex into model (encompassing only the RV free wall from lateral
four and six equal segments, respectively. A consensus tricuspid annulus to RV insertion point on the LV in the
recommendation is yet to be addressed; however, local calculation of RV free wall longitudinal strain; RV FWLS)
standardisation is recommended. (8). When comparing the tracking of the myocardium
using the three-segment and six-segment model, it was
reported that the latter was feasible in a larger portion of
/9FLUFXPIHUHQWLDOVWUDLQDQG65
the subjects due to an increased frequency of inadequate
Circumferential strain represents the change in length free wall segment tracking when using the three-
(shortening in systole, represented as a negative strain segmental model (8). Absolute strain values derived from
value) of the myocardium along the circumferential axis the septum and free wall of the RV differ significantly,
of the LV as viewed in the short axis. Circumferential SR with higher (more negative) strain values produced by
represents the rate of circumferential deformation, with free wall deformation. Subsequently calculations of
a negative systolic value and positive diastolic values. longitudinal strain differ, with higher RV FWLS than RV
GCS is averaged from 12 regional segments across 2 GLS (8). Although the septum contributes to the systolic
PSAX views; 6 segments at MV level and 6 segments at function of the RV (29), it is traditionally seen as part of
PM level. the LV architecture. Furthermore, regional assessment of

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 5
echocardiography: tips and tricks

septal function is reported in LV regional and GLS, with of clinically relevant values can be measured. Strain curves
the LV bullseye (BE) plot allowing easy visualisation of offer values of peak strain, time to peak strain and post-
septal deformation. It is therefore suggested that the ROI systolic index, with SR curves providing values of peak
should encompass both the RV free wall and septum, with systolic, peak early diastolic and peak late diastolic SRs.
longitudinal strain averaged from only the three free wall Rotation curves provide peak systolic apical and basal
segments. RV FWLS is the default parameter to report (30). rotation, twist and torsion values, with rotation rate
curves representing peak twist and untwist rate.

0HDVXUHPHQWUHVXOWVLQWHUSUHWDWLRQ
%(SORWVDQGFRORXUHG0PRGH
6WUDLQDQG65FXUYHV
BE plots of the LV provide regional values of peak strain, time
The analysis software generates global strain, SR, rotation to peak strain and post-systolic strain with corresponding
and twist curves (Figs 7 and 8). From these curves a variety colour-coded visual plot (Fig. 9). Conventionally the

)LJXUHb
Strain and strain rate curves. (A) Left ventricular
longitudinal strain and (B) strain rate. (C)
Circumferential strain and (D) strain rate. (E)
Radial strain and (F) strain rate. (G) Right
ventricular longitudinal strain and (H) strain rate.
Orange arrow indicates peak strain values. White
arrow measures time to peak strain. SRs, systolic
strain rate; Sre, early diastolic strain rate; Sra, late
diastolic strain rate.

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020
 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 5
echocardiography: tips and tricks

)LJXUHb
Left ventricular rotational mechanics curves. Left
ventricular (A) rotation and (B) rotation rate. Blue
lines represent apical rotation, pink lines
represent basal rotation and white lines are twist
GHVFULEHGSUHYLRXVO\DV7ZLVWɋ ɋ$SLFDOURWDWLRQ
− Basal rotation). Blue arrow indicates peak apical
rotation, pink arrow indicates peak basal rotation
and the white arrow highlights peak twist. PT,
peak twist rate; PUT, peak untwist rate.

peak strain spectrum consists of red representing normal in a healthy adult population. From 24 eligible studies
peak strain to pink highlighting severely reduced peak and 28 valid datasets comprising 2597 participants,
strain and light blue presenting paradoxical deformation. normal values for GLS varied from −15.9 to −22.1% (mean
Other colour maps to represent time to peak strain exist −19.7%, CI −20.4 to −18.9%). Meta-regression analyses
and highlight early time to peak strain (preceding or at showed that age, gender, body mass index, frame rate and
AVC), late time to peak strain (proceeding AVC + 250 ms) vendor were not deemed significant sources of variation
and intermediate time to peak strain. The post-systolic among normal ranges of GLS. Mean blood pressure was
index spectrum can be used to demonstrate deformation independently associated with higher values of strain.
before AVC and delay of peak strain after AVC. The BE plot Prospective GLS data were obtained in the EACVI
therefore allows easy visualisation of regional dysfunction NORRE study (32). NORRE was a large multi-centre
and dyssynchrony. A colour m-mode map also provides European study performed within the realms of the
colour coded grading of regional strain using the same European Association of Cardiovascular Imaging (EACVI).
spectrum as the BE plot. The study recruited 549 participants with mean age
45.6 ± 13.3 years, 227 males and 322 females establishing
normal ranges for GLS of −22.5 ± 2.7%. Similar to
1RUPDWLYHYDOXHV
conventional measures of LV function, there was a
The most commonly measured strain parameter in both significantly (P < 0.05) higher reference value for GLS
research and clinical practice is currently longitudinal in females when compared to males (−23 ± 2.7% versus
strain of the LV. Normal values for GLS have been reported −21.7 ± 2.5%). The absolute lowest expected values for
in numerous studies either when comparing against GLS were −16.7% in males and −17.8% in females. Unlike
pathology or when attempting to define normality in a the previous study (31), there was a significant decline in
healthy population. A comprehensive meta-analysis (31) GLS in females between age ranges 20 and 40, 40 and 60
attempted to define the normal expected ranges for GLS and >60 years.

)LJXUHb
Left ventricular bullseye plots. (A) Peak systolic strain, (B) time to peak longitudinal strain and (C) post-systolic index. The outermost ring represents basal
level segments, the second ring represents mid-level segments and the inner most ring represents apical level segments, with the centre circle
representing the apical cap. ANT, anterior; INF, inferior; LAT, lateral; POST, posterior; SEPT, septal.

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 5
echocardiography: tips and tricks

Both these studies have also attempted to classify in mean twist between males and females with females
normal circumferential and radial strain parameters for showing higher values (7.4 ± 2.6° males and 8.3 ± 3.3° for
the LV. The previously referenced meta-analysis presented females). The lowest expected values were 2.2° for males
normal GCS reference values from −20.9 to −27.8% and 1.9° for females. There was no significant change in
(mean −23.3%, 95% CI −24.6 to −22.1%). These data LV twist with increasing age.
were obtained from 14 independent studies consisting of There is conflicting agreement regarding the best
a total of 599 patients. GRS values ranged from 35.1 to method to measure RV longitudinal strain, in that the
59% (mean 47.3%, 95% CI 43.6–51%), from 568 patients normative data are somewhat variable. A prospective study
from 12 studies. NORRE (32) measured GCS and GRS in collected RV longitudinal strain data in 116 participants.
their 549 participants. Average GCS was −31.9 ± 4.5% and Mean participant age was 48 ± 16 years and 58% of the
average GRS 37.4 ± 8.4%. Again, there was higher GCS cohort was female. Participants were found to have a
and GRS in females when compared to males (difference mean value for RV FWLS of −26 ± 4% (37). These data
in GCS non-significant). Lowest expected values for GCS were supported by a meta-analysis conducted by the same
were −22.3% for males and −23.6% in females, whilst the team using ten studies and a total of 486 patients meeting
lowest cut-off values for GRS were 20.6% for males and strict inclusion criteria. For these studies, the mean age
21.5% for females. ranged from 43 to 57 years and 59% of the cohort was
The reproducibility of strain is variable with suggested female. Normal values for STE measured RV FWLS of
inter-vendor variability in the measurement of GLS (33). −27.2% (95% CI −29 to −24%) were suggested. RV GLS
Inter- and intra-observer reproducibility of GLS (33) and produced a normal value of −20.1 (95% CI −20 to −19%)
GCS (34) has been reported to be very good, with radial (37). In a further prospective study conducted by leading
strain proving less so (34). Therefore, it is recommended European experts, 276 healthy participants were recruited
that serial measurements are taking using the same aged between 18 and 76 years and 55% female (8). This
machine to acquire the images and software to analyse. study attempted to define the feasibility of 3-segment
Where this is not possible, the other alternative is to use model (RV FWLS) and the 6-segment model (RV GLS)
vendor-independent analysis software to analyse the of RV longitudinal strain together with definition of
images. This is supported by NORRE who suggest lower reference ranges for both techniques. RV GLS analysis was
radial and higher values of circumferential strain will be feasible in 92% of the population. RV FWLS demonstrated
obtained using GE equipment when compared to Phillips. higher magnitude than RV GLS. Males demonstrated
There were no significant differences seen between lower RV longitudinal strain values independent of the
vendors in GLS (32). 3- or 6-segment model. Reference limits of normality of
Normal values for SR using STE are less well studied −20% for males and −20.3% for females were suggested
and therefore there is no clear clinical consensus for these for RV GLS, with limits of −22.5% for males and −23.3%
values. Normal resting values of between 1.0/s and 1.4/s for females suggested for RV FWLS.
(SD 0.5–0.6/s) have been suggested by leading experts for It is important to note the limitations of 2D STE. Firstly,
longitudinal deformation (35). as previously mentioned, the chambers of the heart are 3D
Normal values for LV rotation and twist show structures that have sophisticated mechanics produced
some variation in the literature and are dependent on by complex myofiber orientation in several directions.
the technique used for measurement, the location of 2D STE does not allow tracking of ‘out-of-plane’ speckle
the region of interest (e.g. sub-endocardium or sub- motion. Due to this 2D STE requires the acquisition of
epicardium, participant age and loading haemodynamics multiple images from several views. This gives rise to
of the LV (17)). Early studies using 118 healthy volunteers potential variation in plane slices and HR. 3D STE has
(36) have reported mean peak LV twist values of the potential to eradicate these inherent limitations using
7.7 ± 3.5°. This study also noted that peak LV twist was a single apical view for image acquisition and allowing
significantly higher in participants >60 years (10.8 + 4.9°) tracking of speckles across planes in multiple directions
compared to those aged <40 years (6.7 ± 2.9°) and those during post processing. 3D STE is not without its own
aged 40–60 years (8.0 ± 3.0°). NORRE (32) were able to limitations. 3D STE requires adequate temporal resolution
collect LV twist data within their population in order to with the suggested optimal frame rate of 35–50 FPS (38).
establish the reference ranges based on age and gender. Acquisition of a 3D data set with adequate frame rates
Total average mean twist was 7.9 + 3.1° over the entire requires compliance of the patient in multi-beat breath-
population. There was, however, a significant difference hold techniques. If breath-hold and therefore multi-beat

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020
 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 5
echocardiography: tips and tricks

imaging is not appropriate, 3D data sets acquired over Circulation 2008 117 1436–1448. (https://doi.org/10.1161/
CIRCULATIONAHA.107.653576)
less beats with frame rates below 18 FPS are reported to
6 Lorenz CH, Walker ES, Morgan VL, Klein SS & Graham TP Jr.
underestimate strain values (39). Normal human right and left ventricular mass, systolic function,
and gender differences by cine magnetic resonance imaging. Journal
of Cardiovascular Magnetic Resonance 1999 1 7–21. (https://doi.
org/10.3109/10976649909080829)
&RQFOXVLRQ 7 Ho SY & Nihoyannopoulos P. Anatomy, echocardiography, and
normal right ventricular dimensions. Heart 2006 92 i2–i13. (https://
doi.org/10.1136/hrt.2005.077875)
STE provides an accessible, feasible and reproducible
8 Muraru D, Onciul S, Peluso D, Soriani N, Cucchini U, Aruta P,
investigation for cardiac deformation and rotational Romeo G, Cavalli G, Iliceto S & Badano LP. Sex- and method-specific
mechanics. Efforts towards standardisation of STE image reference values for right ventricular strain by 2-dimensional speckle-
tracking echocardiography. Circulation: Cardiovascular Imaging 2016 9
acquisition, post processing and interpretation should be
e003866. (https://doi.org/10.1161/CIRCIMAGING.115.003866)
a priority in order to achieve inter-vendor, inter-software 9 Bansal M & Kasliwal RR. How do I do it? Speckle-tracking
and inter-organisational compliance. This coupled echocardiography. Indian Heart Journal 2013 65 117–123. (https://
doi.org/10.1016/j.ihj.2012.12.004)
with increased ease of use and rapid analysis through
10 Reisner SA, Lysyansky P, Agmon Y, Mutlak D, Lessick J & Friedman Z.
technological development provides an opportunity for Global longitudinal strain: a novel index of left ventricular systolic
an effective and efficient routine clinical investigation. function. Journal of the American Society of Echocardiography 2004 17
630–633. (https://doi.org/10.1016/j.echo.2004.02.011)
Future implementation of STE in clinical practice could
11 Nesbitt GC, Mankad S & Oh JK. Strain imaging in echocardiography:
offer a further non-invasive method of investigation into methods and clinical applications. International Journal of
cardiac function within subclinical disease populations, Cardiovascular Imaging 2009 25 9–22. (https://doi.org/10.1007/
s10554-008-9414-1)
allowing a window of opportunity to alter patient therapy/
12 Hoit BD. Strain and strain rate echocardiography and coronary artery
management before pronounced dysfunction. disease. Circulation: Cardiovascular Imaging 2011 4 179–190. (https://
doi.org/10.1161/CIRCIMAGING.110.959817)
13 Voigt JU, Pedrizzetti G, Lysyansky P, Marwick TH, Houle H,
Baumann R, Pedri S, Ito Y, Abe Y, Metz S, et al. Definitions for
'HFODUDWLRQRILQWHUHVW a common standard for 2D speckle tracking echocardiography:
7KH DXWKRUV GHFODUH WKDW WKHUH LV QR FRQȵLFW RI LQWHUHVW WKDW FRXOG EH consensus document of the EACVI/ASE/Industry Task Force
perceived as prejudicing the impartiality of this review. to standardize deformation imaging. European Heart Journal:
Cardiovascular Imaging 2015 16 1–11. (https://doi.org/10.1093/ehjci/
jeu184)
14 Yuan LJ, Takenaka K, Uno K, Ebihara A, Sasaki K, Komuro T,
Funding Sonoda M & Nagai R. Normal and shear strains of the left ventricle
7KLVZRUNGLGQRWUHFHLYHDQ\VSHFLȴFJUDQWIURPDQ\IXQGLQJDJHQF\LQWKH in healthy human subjects measured by two-dimensional speckle
SXEOLFFRPPHUFLDORUQRWIRUSURȴWVHFWRU tracking echocardiography. Cardiovascular Ultrasound 2014 12 7.
(https://doi.org/10.1186/1476-7120-12-7)
15 Wharton G, Steeds R, Allen J, Phillips H, Jones R, Kanagala P,
Lloyd G, Masani N, Mathew T, Oxborough D, et al. A minimum
$XWKRUFRQWULEXWLRQVWDWHPHQW dataset for a standard adult transthoracic echocardiogram: a
C Johnson and K Kuyt: Joint lead authors. D Oxborough and M Stout: Joint guideline protocol from the British Society of Echocardiography.
senior authors. Echo Research and Practice 2015 2 G9–G24. (https://doi.org/10.1530/
ERP-14-0079)
16 Kusunose K, Yamada H, Nishio S, Tomita N, Hotchi J, Bando M,
Niki T, Yamaguchi K, Taketani Y, Iwase T, et al. Index-beat assessment
of left ventricular systolic and diastolic function during atrial
5HIHUHQFHV fibrillation using myocardial strain and strain rate. Journal of the
1 Spotnitz HM. Macro design, structure, and mechanics of the left American Society of Echocardiography 2012 25 953–959. (https://doi.
ventricle. Journal of Thoracic and Cardiovascular Surgery 2000 119 org/10.1016/j.echo.2012.06.009)
1053–1077. (https://doi.org/10.1016/S0022-5223(00)70106-1) 17 Mor-Avi V, Lang RM, Badano LP, Belohlavek M, Cardim NM,
2 Buckberg G, Hoffman JIE, Mahajan A, Saleh S & Coghlan C. Cardiac Derumeaux G, Galderisi M, Marwick T, Nagueh SF, Sengupta PP,
mechanics revisited. Circulation 2008 118 2571–2587. (https://doi. et al. Current and evolving echocardiographic techniques for the
org/10.1161/CIRCULATIONAHA.107.754424) quantitative evaluation of cardiac mechanics: ASE/EAE consensus
3 Zhukov L & Barr AH. Heart-muscle fiber reconstruction from statement on methodology and indications endorsed by the
diffusion tensor MRI. IEEE Visualization, 2003. VIS 2003, 597–602. Japanese Society of Echocardiography. Journal of the American Society
(https://doi.org/10.1109/VISUAL.2003.1250425) of Echocardiography 2011 24 277–313. (https://doi.org/10.1016/j.
4 Nakatani S. Left ventricular rotation and twist: why should we echo.2011.01.015)
learn? Journal of Cardiovascular Ultrasound 2011 19 1–6. (https://doi. 18 Rösner A, Barbosa D, Aarsæther E, Kjønås D, Schirmer H & D'hooge J.
org/10.4250/jcu.2011.19.1.1) The influence of frame rate on two-dimensional speckle-tracking
5 Haddad F, Hunt SA, Rosenthal DN & Murphy DJ. Right ventricular strain measurements: a study on silico-simulated models and images
function in cardiovascular disease, Part I: anatomy, physiology, recorded in patients. European Heart Journal: Cardiovascular Imaging
aging, and functional assessment of the right ventricle. 2015 16 1137–1147. (https://doi.org/10.1093/ehjci/jev058)

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.
 C Johnson HWbDO Speckle tracking 6:3 R98
echocardiography: tips and tricks

19 Forsha D, Risum N, Rajagopal S, Dolgner S, Hornik C, Barnhart H, 29 Buckberg G & Hoffman JI. Right ventricular architecture responsible
Kisslo J & Barker P. The influence of angle of insonation and target for mechanical performance: unifying role of ventricular septum.
depth on speckle-tracking strain. Journal of the American Society Journal of Thoracic and Cardiovascular Surgery 2014 148 3166.e1–3171.
of Echocardiography 2015 28 580–586. (https://doi.org/10.1016/j. e1. (https://doi.org/10.1016/j.jtcvs.2014.05.044)
echo.2014.12.015) 30 Badano LP, Kolias TJ, Muraru D, Abraham TP, Aurigemma G,
20 Stoebe S, Tarr A, Pfeiffer D & Hagendorff A. The impact of the width Edvardsen T, D'Hooge J, Donal E, Fraser AG, Marwick T, et al.
of the tracking area on speckle tracking parameters – methodological Standardization of left atrial, right ventricular, and right atrial
aspects of deformation imaging. Echocardiography 2014 31 586–596. deformation imaging using two-dimensional speckle tracking
(https://doi.org/10.1111/echo.12440) echocardiography: a consensus document of the EACVI/ASE/
21 Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Industry Task Force to standardize deformation imaging. European
Ernande L, Flachskampf FA, Foster E, Goldstein SA, Kuznetsova T, Heart Journal: Cardiovascular Imaging 2018 19 591–600. (https://doi.
et al. Recommendations for cardiac chamber quantification by org/10.1093/ehjci/jey042)
echocardiography in adults: an update from the American Society of 31 Yingchoncharoen T, Agarwal S, Popović ZB & Marwick TH. Normal
Echocardiography and the European Association of Cardiovascular ranges of left ventricular strain: a meta-analysis. Journal of the
Imaging. European Heart Journal: Cardiovascular Imaging 2015 16 American Society of Echocardiography 2013 26 185–191. (https://doi.
233–270. (https://doi.org/10.1093/ehjci/jev014) org/10.1016/j.echo.2012.10.008)
22 Rudski LG, Lai WW, Afilalo J, Hua L, Handschumacher MD, 32 Sugimoto T, Dulgheru R, Bernard A, Ilardi F, Contu L, Addetia K,
Chandrasekaran K, Solomon SD, Louie EK & Schiller NB. Guidelines Caballero L, Akhaladze N, Athanassopoulos GD, Barone D, et al.
for the echocardiographic assessment of the right heart in adults: a Echocardiographic reference ranges for normal left ventricular 2D
report from the American Society of Echocardiography endorsed by strain: results from the EACVI NORRE study. European Heart Journal:
the European Association of Echocardiography, a registered branch Cardiovascular Imaging 2017 18 833–840. (https://doi.org/10.1093/
of the European Society of Cardiology, and the Canadian Society of ehjci/jex140)
Echocardiography. Journal of the American Society of Echocardiography 33 Nagata Y, Takeuchi M, Mizukoshi K, Wu VC, Lin FC, Negishi K,
2010 23 685–713; quiz 786. (https://doi.org/10.1016/j. Nakatani S & Otsuji Y. Intervendor variability of two-dimensional
echo.2010.05.010) strain using vendor-specific and vendor-independent software.
23 Zoppellaro G, Venneri L, Khatter RS, Li W & Senior R. Simultaneous Journal of the American Society of Echocardiography 2015 28 630–641.
assessment of myocardial perfusion, wall motion, and deformation (https://doi.org/10.1016/j.echo.2015.01.021)
during myocardial contrast echocardiography: a feasibility study. 34 Oxborough D, George K & Birch KM. Intraobserver reliability of two-
Echocardiography 2016 33 889–895. (https://doi.org/10.1111/ dimensional ultrasound derived strain imaging in the assessment of
echo.13190) the left ventricle, right ventricle, and left atrium of healthy human
24 Medvedofsky D, Lang RM, Kruse E, Guile B, Weinert L, Ciszek B, hearts. Echocardiography 2012 29 793–802. (https://doi.org/10.1111/
Jacobson Z, Negron J, Volpato V, Prado A, et al. Feasibility of left j.1540-8175.2012.01698.x)
ventricular global longitudinal strain measurements from contrast- 35 Marwick TH. Measurement of strain and strain rate by
enhanced echocardiographic images. Journal of the American Society echocardiography: ready for prime time? Journal of the American
of Echocardiography 2018 31 297–303. (https://doi.org/10.1016/j. College of Cardiology 2006 47 1313–1327. (https://doi.org/10.1016/j.
echo.2017.10.005) jacc.2005.11.063)
25 Nagy AI, Sahlén A, Manouras A, Henareh L, Da Silva C, Günyeli E, 36 Takeuchi M, Nakai H, Kokumai M, Nishikage T, Otani S & Lang RM.
Apor AA, Merkely B &Winter R. Combination of contrast-enhanced Age-related changes in left ventricular twist assessed by two-
wall motion analysis and myocardial deformation imaging during dimensional speckle-tracking imaging. Journal of the American Society
dobutamine stress echocardiography. European Heart Journal: of Echocardiography 2006 19 1077–1084. (https://doi.org/10.1016/j.
Cardiovascular Imaging 2015 16 88–95. (https://doi.org/10.1093/ echo.2006.04.011)
ehjci/jeu171) 37 Fine NM, Chen L, Bastiansen PM, Frantz RP, Pellikka PA, Oh JK &
26 Mada RO, Lysyansky P, Daraban AM, Duchenne J & Voigt JU. How Kane GC. Reference values for right ventricular strain in patients
to define end-diastole and end-systole?: impact of timing on strain without cardiopulmonary disease: a prospective evaluation and
measurements. JACC: Cardiovascular Imaging 2015 8 148–157. meta-analysis. Echocardiography 2015 32 787–796. (https://doi.
(https://doi.org/10.1016/j.jcmg.2014.10.010) org/10.1111/echo.12806)
27 Leitman M, Lysiansky M, Lysyansky P, Friedman Z, Tyomkin V, 38 Negishi K, Negishi T, Agler DA, Plana JC & Marwick TH.
Fuchs T, Adam D, Krakover R & Vered Z. Circumferential and Role of temporal resolution in selection of appropriate strain
longitudinal strain in 3 myocardial layers in normal subjects and technique for evaluation of subclinical myocardial dysfunction.
in patients with regional left ventricular dysfunction. Journal of the Echocardiography 2012 29 334–339. (https://doi.org/10.1111/j.1540-
American Society of Echocardiography 2010 23 64–70. (https://doi. 8175.2011.01586.x)
org/10.1016/j.echo.2009.10.004) 39 Yodwut C, Weinert L, Klas B, Lang RM & Mor-Avi V. Effects of frame
28 Spriestersbach H, Oh-Icí D, Schmitt B, Berger F & Schmitz L. The rate on three-dimensional speckle-tracking-based measurements
influence of the region of interest width on two-dimensional of myocardial deformation. Journal of the American Society of
speckle tracking-based measurements of strain and strain rate. Echocardiography 2012 25 978–985. (https://doi.org/10.1016/j.
Echocardiography 2015 32 89–95. (https://doi.org/10.1111/echo.12589) echo.2012.06.001)

Received in final form 3 June 2019

Accepted 13 June 2019
Accepted Manuscript published online 13 June 2019

KWWSVHUSELRVFLHQWLȴFDFRP © 2019 The authors This work is licensed under a Creative Commons
https://doi.org/10.1530/ERP-19-0020
 3XEOLVKHGE\%LRVFLHQWLȴFD/WG Attribution-NonCommercial 4.0 International
License.