LEFT VENTRICULAR MECHANICS

Velocity Vector Imaging: Standard Tissue-Tracking

Results Acquired in Normals—The VVI-STRAIN Study
Shemy Carasso, MD, Patric Biaggi, MD, Harry Rakowski, MD, Diab Mutlak, MD, Jonathan Lessick, MD,
Doron Aronson, MD, Anna Woo, MD, and Yoram Agmon, MD, Haifa, Israel; Toronto, Ontario, Canada

Background: Echocardiographic imaging assessment of left ventricular mechanics is a new technology that is of-
fered by various vendors. Different software algorithms have at times yielded conflicting results. The aim of this study
was to determine normal myocardial mechanical parameters in a healthy population using Velocity Vector Imaging.

Methods: One hundred twenty subjects were selected for this study, including healthy subjects referred for
echocardiography to evaluate minor symptoms or murmurs, who had normal echocardiographic findings
and healthy volunteers. Study subjects were recruited in Haifa, Israel and Toronto, Canada. Echocardiography
was performed using commercially available systems to analyze archived studies. Endocardial and epicardial
longitudinal and circumferential strain and strain rate were calculated as well as rotational mechanical param-
eters. Age and gender differences were evaluated.

Results: Average endocardial longitudinal, circumferential, and radial strains and twist were
19.6 6
2.0%,
27.6 6 3.9%, +30.1 6 7.5%, and 9.6 6 3.9 , respectively. Epicardial circumferential strain and twist
were
11.3 6 2.2% and 4.0 6 1.9 , respectively. Shortening increased from base to apex longitudinally (10%)
and circumferentially (33%). Thickening at the apex was 16% lower than at the base. Men and older subjects
had increased endocardial circumferential strain and apical rotation.

Conclusions: Mechanical parameters differ with location (endocardial vs epicardial, basal vs apical strain
gradients), age, and gender. Care should be taken when comparing regional strain measurements between
systems, and gender and age should be matched between and within two-dimensional strain systems. (J
Am Soc Echocardiogr 2012;25:543-52.)

Keywords: Myocardial mechanics, Strain, Rotation, Normal, Speckle tracking

Echocardiographic assessment of left ventricular (LV) mechanics uses
relatively new technology that is now offered by various vendors.1-6
Although methodology has been validated, processing algorithms
differ among vendors, and results may not be comparable and at
times have even yielded conflicting findings.7-10 Thus, system-
specific standards measured in a normal population are required.
Such a study was recently published for longitudinal strain11 using
EchoPAC (GE Vingmed Ultrasound AS, Horten, Norway), but results
have not been published for Velocity Vector Imaging (VVI; Siemens
Medical Solutions USA, Inc., Mountain View, CA). An advantage of
VVI software is its ability to analyze myocardial mechanics from any
ultrasound system stored in Digital Imaging and Communications
in Medicine format. This substantially enlarges the number of studies
available for analysis in echocardiography laboratories with machines
From Rambam Health Care Campus, Haifa, Israel (S.C., D.M., J.L., D.A., Y.A.); the
School of Medicine, The Technion, Israel Institute of Technology, Haifa, Israel
(S.C., J.L., D.A., Y.A.); the University of Toronto, Toronto, Ontario, Canada (S.C.,
P.B., H.R., A.W.); and Toronto General Hospital, Peter Munk Cardiac Centre,
Toronto, Ontario, Canada (P.B., H.R., A.W.).
Reprint requests: Shemy Carasso, MD, Rambam Health Care Campus, PO Box
9602, Haifa 31096, Israel (E-mail: s_carasso@rambam.health.gov.il).
0894-7317/$36.00
Copyright 2012 by the American Society of Echocardiography.
doi:10.1016/j.echo.2012.01.005
from multiple vendors, provided that no machine-specific differences
are found. A new version of VVI that allows selective measurement
of both subendocardial and subepicardial strain parameters was avail-
able to us. Our goals were to describe endocardial and epicardial lon-
gitudinal and circumferential mechanics analyzed by VVI in a normal
population, exploring age and gender differences as well as acquisition
and storage (machine source and frame rates) differences.
METHODS
Study Population
Subjects selected for this study included consecutive patients referred
for elective echocardiography to evaluate minor symptoms or benign
murmurs during 2009. Conventional echocardiographic studies had
to be reported as having normal results, and clinically, patients had to
be healthy, without major illnesses, and free of cardiovascular risk
factors. A second group of patients included healthy volunteers free
of cardiovascular risk factors. Study subjects were recruited in Haifa,
Israel, and Toronto, Canada.
Conventional Echocardiography
Echocardiography was performed using Sequoia (Siemens Medical
Solutions USA, Inc.), iE33 (Philips Medical Systems, Best, The
Netherlands), and Vivid 7 (GE Vingmed Ultrasound AS) systems.
Studies were done and reported according to accepted guidelines of
543
544 Carasso et al
Abbreviations
the American Society of
Echocardiography12,13 and saved
CI = Confidence interval for further analysis in the
FEARR = Fractional early laboratory Digital Imaging and
apical reverse rotation Communications in Medicine
archive.
LV = Left ventricular
SR = Strain rate LV Mechanics
2D = Two-dimensional Measurements were performed
using VVI software version
VVI = Velocity Vector Imaging 2.5.1 from archived studies
(Figures 1 and 2). Using VVI’s
clip editor, QRS frames were marked and two or three consecutive
cycles were selected for analysis for each view. Endocardial and epi-
cardial contours were then traced and processed. Averaged endocar-
dial and epicardial circumferential strain, strain rate (SR), and rotation
velocities and angles and radial strain and SR were measured at three
parasternal short-axis planes (basal, mid, and apical) in six segments
per short-axis section. Endocardial and epicardial longitudinal strain
and SR were measured from the apical four-chamber, two-chamber,
and three-chamber views in six segments per view.
Ventricular Twist. Averaged myocardial rotation angles were used
to calculate LV twist, defined as the maximal instantaneous basal-to-
apical angle difference. Rotation and twist timing was standardized
to cycle length.
Diastolic Mechanics
Systolic and early diastolic peak SR were measured in the longitudinal
and circumferential axes. We showed in a previous report that peak
systolic and early diastolic SR were both lower longitudinally and
higher circumferentially.10,14 Assuming that systolic and diastolic
function are interrelated, we needed a new parameter to identify
systolic-independent diastolic abnormalities. The ratio of peak early
diastolic to peak systolic SR ratio (SR E/S ratio) was calculated
(Figure 2A) to assess whether the diastolic and systolic extent of
changes were disproportionate.
We used fractional early apical reverse rotation (FEARR) to assess
early LV relaxation.14 We measured the fractional decrease in rotation
angle from its peak value to its value 10% of the cycle length later,
using the equation shown in Figure 2B. The threshold of 10% diastolic
time was selected because it was previously reported to demonstrate
the largest decrease in fractional reverse rotation for moderate
compared with mild LV hypertrophy.9
Reproducibility
For intraobserver variability, 10 randomly assigned patients were
reanalyzed by the same observer (S.C.) several months after the initial
analysis. For interobserver variability, the same patients and the exact
same loops were analyzed by a second observer (P.B.).
Statistical Analysis
Continuous data are reported as mean 6 SD. Assuming a non-normal
distribution, nonparametric tests were used. Gender subgroups as well
as strain gradients (endocardial and epicardial) were compared using
Wilcoxon’s rank-sum test. Differences among age subgroups, seg-
ments, short-axis strain gradients (basal, mid, and apical), and images
from various echocardiographic machines were assessed using analy-
sis of variance with Tukey’s post hoc test. P values < .05 were consid-
ered statistically significant. For test performance, intraobserver and
Journal of the American Society of Echocardiography
May 2012
Figure 1 LV mechanical parameter orientation. LV segmenta-
tion and myocardial mechanical parameters analyzed by VVI.
LAX, Long-axis; SAX, short-axis.
interobserver variability, intraclass correlation coefficients with 95%
confidence intervals (CIs) were calculated, because they take the
bias between observations into account.15,16 Statistical analyses were
performed using SPSS release 11.0 (SPSS, Inc., Chicago, IL).
RESULTS
Patients
Patients were consecutively recruited during 2009. Of a potential 142
normal subjects who met the inclusion criteria, 120 patients had
VVI-analyzable studies. Their age ranged from 19 to 84 years
(median, 39 years), and 55% were men. The distribution of referral
diagnoses is shown in Figure 3A. Assignment to echocardiographic
machines was random and reflected machine availability
(Figure 3B), except for a group of Canadian volunteers who were im-
aged on the Vivid 7 system but analyzed using VVI software. Most of
the subjects (n = 88) were enrolled in Israel and the rest in Canada.
Two-Dimensional (2D) Doppler Echocardiography
All conventional echocardiographic parameters were within the
accepted normal ranges for gender, age, and heart rate (Table 1).
Women demonstrated significantly lower LV mass of about 10%,
while ejection fraction, Doppler, and tissue Doppler parameters
were similar. Age was not associated with 2D echocardiographic
changes. Expected age-related significant changes in the mitral inflow
parameters and annular velocities were demonstrated.
Two-Dimensional Strain Imaging
Acquired clips had 14 to 57 frames per heart cycle (median, 24
frames per heart cycle), corresponding to a heart rate ranging from
41 to 90 beats/min (median, 64 beats/min). Of 2,160 potential seg-
ments, 2,061 (95%), 2,021 (94%), 1,990 (92%), and 1,945 (90%)
were analyzable for longitudinal, endocardial circumferential,
epicardial circumferential, and radial strain assessment, respectively.
Longitudinal Mechanics. The average longitudinal strain was

20 6 2.0% and demonstrated a small yet significant apical-to-basal
gradient that was similar by gender and age (Table 2, Figure 4A).
Basal inferior septal and inferior wall strain were lowest (
16 6 3%).
Women had a slightly higher average longitudinal strain. Systolic and
early diastolic SR followed the same pattern, with a significant decrease
in the early diastolic to systolic SR ratio (SR E/S) with older age. There
Journal of the American Society of Echocardiography
Volume 25 Number 5
Figure 2 Strain, SR, and rotation. (A) Strain (top) and SR (bottom) parametric maps and curves. (B) Apical rotation curve and
calculation of FEARR. CL, Cycle length; q, rotation angle.
was also a significant increase in the late diastolic SR (SR A) with age.
Epicardial longitudinal strain was significantly lower than endocardial
strain only in apical segments (up to 50% lower; Figure 4), resulting
in a nonsignificantly lower average epicardial strain (
20 6 2% vs

18 6 4% for endocardial and epicardial strain, respectively, P > .10).
Circumferential Endocardial Mechanics. The average endo-
cardial circumferential strain was
28 6 4%, with a large and signif-
icant apical-to-basal gradient (
32 6 6% vs
24 6 4% at the apex
and base, respectively; Table 3). Here again, basal inferior and inferior
septal segments demonstrated the lowest strain (
22 6 5%). Women
had mildly but significantly lower systolic strain values, and larger
reductions were observed in SR S and SR E, with an unchanged SR
E/S ratio. With older age, systolic strain and SR S increased, resulting
in a significant decrease in the SR E/S ratio, similar to its longitudinal
value. SR A was similar in all the subgroups analyzed.
Circumferential Epicardial Mechanics. Epicardial strain was
about 40% lower than endocardial circumferential strain, averaging
11 6 2% (Table 4). There were no differences related to gender.
Epicardial strain and SR followed the endocardial pattern of changes
that were not statistically significant.
Radial Mechanics. The average radial strain was 30 6 7.5% and
demonstrated a reversed apical-to-basal gradient (26 6 10% vs
32 6 12%, apical and basal, respectively; Table 5). The only other
significant change was a decrease in the SR E/S ratio in the oldest
subgroup, again similar to its longitudinal and circumferential values.
Carasso et al 545
Rotational Mechanics. Basal endocardial rotation was in the
clockwise direction, and the average peak rotation angle was

3.4 6 2.0 (Table 6). Mid and apical rotation were both counter-
clockwise, and peak angles were 2.2 6 1.9 and 7.1 6 3.3 , respec-
tively. The opposing orientations of the basal and mid levels placed
the rotation equatorial (null rotation level) between these levels.
This pattern was consistent between genders and did not change
with age. The overall peak endocardial twist (peak instantaneous
apical-to-basal rotation angle difference) was 9.9 6 4.0 .
Epicardial rotation demonstrated the same level rotation orienta-
tion, although mid rotation was more variable in its orientation.
Peak rotation angles were <50% of endocardial angles, as was
epicardial twist.
FEARR was 38 6 9% and was not affected by gender or age. A
similarly calculated fraction for the overall twist yielded the same
fraction.
Frame Rate and Heart Rate. Because studies were archived
before the application of VVI, they were subject to reduction of frame
rate due to archive storage policy. As a result, lower heart rates resulted
in a larger number of frames per cycle. We divided patients according
to the number of frames available for analysis (<25 frames/cycle [n =
69; heart rate, 72 6 9 beats/min] vs $25 frames/cycle [n = 51; heart
rate, 57 6 6 beats/min]). There were no gender or age differences be-
tween the groups, and measured strain parameters were similar.
However, the ability to identify SR A was 70% in the group with
low frame rates compared with 93% in the group with high frame
546 Carasso et al
Figure 3 Distribution of referral diagnoses and acquisition
machines. (A) Pie chart of the echocardiographic referral
diagnoses. (B) Pie chart of the echocardiographic acquisition
machines. ECG, Electrocardiographic.
rates. Whether this was because of decreased VVI sensitivity or higher
heart rates resulting in fusion of SR E and A was unclear.
Echocardiographic System. Comparison of VVI-analyzed strain
parameters from studies acquired by the various available machines
did not demonstrate any differences.
Reproducibility. Intraobserver agreement intraclass correlation
coefficients were 0.93 for strain (95% CI, 0.84–0.97), 0.97 for SR
S (95% CI, 0.94–0.99), 0.95 for SR E (95% CI, 0.88–0.98), and
0.97 for peak rotation angle (95% CI, 0.94–0.99). Interobserver
intraclass correlation coefficients were somewhat lower: 0.89 for
strain (95% CI, 0.75–0.96), 0.91 for SR S (95% CI, 0.78–0.97),
0.95 for SR E (95% CI, 0.74–0.96), and 086 for peak rotation an-
gle (95% CI, 0.63–0.96). There was 100% agreement between
observers regarding the rotation direction.
DISCUSSION
Evaluation of myocardial mechanics can play an important role in
the evaluation of disease states. Comparisons are typically made to
normal controls, so it is important to understand differences in
myocardial mechanics by region of the myocardium studied, age,
gender, and instrumentation. In this study, we have shown where
important differences in measurements occur, on the basis of the
Journal of the American Society of Echocardiography
May 2012
layer of myocardium assessed, apical-to-basal gradients, and the
effects of gender and age. These findings are summarized in Table 7.
Apical-to-Basal Strain Gradients, Regional Strain Dispersion
Previous studies using various modalities (magnetic resonance imag-
ing tagging, echocardiography) to assess regional strain have demon-
strated conflicting results regarding the presence of apical-to-basal
strain gradients.8,11,17-20 Circumferential gradients have been more
consistently demonstrated, because apical strain is nearly double
basal strain. Longitudinal strain gradients are much smaller, and in
some studies such gradients could not be demonstrated.11,17 Using
the first version of VVI, we also could not identify a longitudinal
strain gradient in both normal controls and a study group of
patients with hypertrophic cardiomyopathy.8 This probably relates
to the tracking algorithm and its sensitivity.
We have also demonstrated a significant regional pattern of strain
variability, showing that basal inferior and inferior septal segmental
strain was significantly lower than opposite wall and mid segmental
strain. This finding is consistent with the well-known difficulty in the
visual assessment of hypokinesis of the same specific segments.
Whether this is an artifact due to acquisition parameters (i.e., foreshort-
ened images, overshadowing by more proximal wall region) or
represents truly decreased shortening (i.e., inclusion of part of the
membranous septum) is not known, but both are possible explanations.
Endocardial-Epicardial Strain Gradients
We have shown a decrease of about 50% in circumferential tran-
smural strain from the subendocardial to the subepicardial layer.
Circumferential strain gradients have been consistently demonstrated
by magnetic resonance imaging and lately by 2D strain as well.21-23
Regarding longitudinal transmural strain gradients, these have been
found to be very small and not significant, except in apical
segments. The transmural gradients in apical segments resemble
circumferential transmural strain gradients. The cause for this ‘‘apical
anomaly’’ is not clear, but it may be due to the increased apical
curvature and/or acquisition problems concerning near-field reduced
2D resolution. Moreover, the presence of an endocardial apical-to-
basal gradient and an opposite epicardial basal-to-apical gradient
may be the reason longitudinal gradients were not demonstrated
when full myocardial thickness longitudinal strain was measured,11
while global longitudinal strain remains very similar when the same
cycle is analyzed by either system.24 The high endocardial-epicardial
circumferential gradient seen here may also provide an explanation
for the generally higher circumferential endocardial strain compared
with full-thickness averaged circumferential strain.24 Moreover,
abnormalities in strain gradients (myocardial, regional) could be
clinically important. They may reflect ischemia (which may be more
subendocardial), scarring patterns that are disease specific, and
changes in fiber orientation and architecture. This may aid in the
differential diagnosis as well as in the long-term follow-up of patients.
Gender-Related Differences
We found women to have slightly higher average endocardial lon-
gitudinal strain and lower circumferential strain, SR S, and SR E.
This probably represents a slightly more vertical orientation of strain
without an actual difference in the extent of shortening. This is supported
by the similar global LVejection fractions and identical radial strain (thick-
ening) in both groups. Global diastolic mechanics assessed using global
SR E/S ratio (longitudinal, circumferential, and radial) and FEARR also
did not demonstrate gender-related differences.
Journal of the American Society of Echocardiography Carasso et al 547
Volume 25 Number 5

Table 1 Conventional 2D Doppler echocardiographic characteristics

By gender By age (y)

All patients Men Women <40 40–59 >60

Variable (n = 120) (n = 66) (n = 54) (n = 62) (n = 37) (n = 21)

Male gender 66 (55%) 36 (50%) 21 (57%) 9 (43%)

Age (y) 41 6 13 41 6 14 45 6 14 32 6 6 48 6 5† 66 6 6‡
Body surface area (m2) 1.85 6 0.19 1.94 6 0.16 1.71 6 0.14* 1.86 6 0.19 1.87 6 0.18 1.81 6 0.20
2D echocardiography
LV diastolic diameter (cm) 4.8 6 0.4 4.9 6 0.4 4.7 6 0.3* 4.9 6 0.4 4.8 6 0.3 4.6 6 0.3†
LV systolic diameter (cm) 3.1 6 0.4 3.1 6 0.3 2.9 6 0.3* 3.0 6 0.4 3.1 6 0.2 2.8 6 0.3†,‡
Septal thickness (cm) 0.8 6 0.1 0.9 6 0.1 0.8 6 0.1* 0.8 6 0.1 0.9 6 0.1 1.0 6 0.1†,‡
Posterior wall thickness (cm) 0.8 6 0.1 0.8 6 0.1 0.7 6 0.1 0.7 6 0.1 0.8 6 0.1 0.8 6 0.1
Ejection fraction (%) 66 6 9 65 6 10 67 6 6 66 6 7 65 6 11 67 6 4
LV mass (g) 128 6 25 138 6 25 114 6 20* 128 6 26 126 6 24 136 6 28
LV mass index (g/m2) 69 6 11 71 6 12 65 6 9 69 6 12 67 6 11 76 6 12†
LA size (cm) 3.5 6 0.3 3.5 6 0.3 3.4 6 0.3 3.4 6 0.3 3.5 6 0.3 3.7 6 0.4†
Mitral inflow
E/A ratio 1.3 6 0.4 1.4 6 0.5 1.3 6 0.4 1.5 6 0.4 1.3 6 0.3 1.0 6 0.3†,‡
Mitral E velocity (cm/sec) 77 6 19 75 6 18 80 6 20 78 6 22 79 6 15 73 6 17
Mitral E deceleration time (msec) 196 6 47 202 6 54 190 6 35 186 6 44 191 6 30 231 6 57†,‡
Tissue Doppler
E0 septal velocity (cm/sec) 11 6 3 10 6 3 11 6 3 14 6 3 10 6 2† 8 6 1†,‡
E0 lateral velocity (cm/sec) 13 6 4 13 6 4 13 6 4 16 6 5 12 6 3† 10 6 3†,‡
Mitral E/E0 ratio 7.4 6 2.9 7.5 6 2.8 7.2 6 3 4.5 6 2 8.8 6 2.2† 9.0 6 2.3†
Pulmonary pressure (mm Hg) 25 6 4 25 6 4 25 6 3 25 6 4 25 6 4 28 6 4†

Data are expressed as number (percentage) or as mean 6 SD.

*P < .05 versus men.
†
P < .05 versus age < 40 years.
‡
P < .05 versus age 40 to 59 years.

Table 2 Longitudinal endocardial mechanics

By gender By age (y)

All patients Men Women <40 40–59 >60

Variable (n = 120) (n = 66) (n = 54) (n = 62) (n = 37) (n = 21)

Strain (%)
19.6 6 2
19.1 6 1.9
20.0 6 2.4*
19.6 6 2.1
19.5 6 1.9
19.4 6 2.7
Base
19.2 6 2.3
18.7 6 2.1
19.7 6 2.5*
19.2 6 2.2
19.3 6 2.6
18.8 6 2.6
Mid
18.1 6 2.5
17.7 6 2.3
18.5 6 2.6
18.5 6 2.5
17.6 6 2.0
17.5 6 3.0
Apex
21.3 6 3.2§
21.0 6 2.6‡
21.8 6 3.8§
21.1 6 3.0§
21.5 6 3.4§
21.7 6 3.5§
Systolic SR (sec
1)
1.02 6 0.12
1.02 6 0.12
1.01 6 0.13
1.02 6 0.12
1.04 6 0.11
1.01 6 0.15
Base
1.00 6 0.13
1.00 6 0.13
1.03 6 0.12
1.00 6 0.12
1.03 6 0.14
0.97 6 0.15
Mid
0.93 6 0.14
0.92 6 0.12
0.94 6 0.15
0.95 6 0.13
0.92 6 0.11
0.90 6 0.17
Apex
1.14 6 0.20§
1.13 6 0.19§
1.14 6 0.21§
1.11 6 0.19‡
1.17 6 0.21§
1.16 6 0.19§
Early diastolic SR (sec
1) 1.04 6 0.28 1.01 6 0.32 1.09 6 0.2 1.06 6 0.32 1.09 6 0.23 0.93 6 0.17‡
Base 0.98 6 0.28 0.94 6 0.33 1.04 6 0.19 1.00 6 0.32 1.02 6 0.24 0.87 6 0.16‡
Mid 0.94 6 0.27 0.90 6 0.30 0.99 6 0.22 0.98 6 0.31 0.94 6 0.20 0.80 6 0.17‡
Apex 1.21 6 0.36§ 1.18 6 0.38§ 1.26 6 0.33§ 1.19 6 0.39‡ 1.30 6 0.36§ 1.13 6 0.25‡,§
SR E/S ratio 1.04 6 0.13 1.02 6 0.13 1.05 6 0.12 1.07 6 0.1 1.04 6 0.14 0.93 6 0.11§
Late diastolic SR (sec
1) 0.37 6 0.12 0.36 6 0.12 0.39 6 0.13 0.31 6 0.09 0.42 6 0.13† 0.46 6 0.10†
Base 0.40 6 0.14 0.40 6 0.15 0.42 6 0.14 0.34 6 0.12 0.46 6 0.16† 0.48 6 0.11†
Mid 0.37 6 0.13 0.35 6 0.13 0.38 6 0.13 0.31 6 0.10 0.41 6 0.15† 0.45 6 0.11†
Apex 0.35 6 0.14§ 0.33 6 0.12§ 0.37 6 0.16§ 0.28 6 0.10‡ 0.38 6 0.14§,† 0.45 6 0.15†

Data are expressed as mean 6 SD.

*P < .05 versus men.
†
P < .05 versus age < 40 years.
‡
P < .05 versus age 40 to 59 years.
§
P < .05 versus base.
548 Carasso et al Journal of the American Society of Echocardiography
May 2012

Figure 4 Longitudinal strain gradients. Example of subendocardial and subepicardial segmental strain curves. Note that peak apical
strain is highest in the subendocardial trace and lowest in the subepicardial trace, resulting in a minimally lower average (black curve)
epicardial strain.

Table 3 Circumferential endocardial mechanics

By gender By age (y)

All patients Men Women <40 40–59 >60

Variable (n = 120) (n = 66) (n = 54) (n = 62) (n = 37) (n = 21)

Strain (%)
27.9 6 4.0
28.3 6 4.2
27.2 6 3.7
26.9 6 3.7
27.9 6 4.2
31.2 6 4.3†,‡
Base
24.4 6 4.1
24.4 6 4.4
24.3 6 3.8
24.0 6 3.9
24.6 6 3.6
26.1 6 5.4
Mid
26.2 6 4.6
26.5 6 4.9
25.9 6 4.4
25.3 6 3.6
25.4 6 4.7
30.6 6 5.1†,‡
Apex
32.4 6 6.2§
33.5 6 6.2§
31.2 6 5.8*,§
30.9 6 5.7§
33.3 6 6.5†
35.9 6 6.2‡,§
Systolic SR (sec
1)
1.66 6 0.33
1.71 6 0.36
1.58 6 0.28*
1.58 6 0.28
1.67 6 0.34
1.90 6 0.39†,‡
Base
1.40 6 0.27
1.42 6 0.29
1.38 6 0.24
1.38 6 0.26
1.38 6 0.26
1.50 6 0.31
Mid
1.50 6 0.33
1.55 6 0.5
1.38 6 0.25*
1.45 6 0.27
1.45 6 0.33
1.78 6 0.38†,‡
Apex
2.01 6 0.54§
2.12 6 0.56§
1.82 6 0.44*,§
1.88 6 0.44§
2.12 6 0.62†,§
2.27 6 0.61†,§
Early diastolic SR (sec
1) 1.66 6 0.34 1.71 6 0.37 1.57 6 0.28* 1.62 6 0.3 1.72 6 0.41 1.74 6 0.46
Base 1.34 6 0.29 1.38 6 0.31 1.30 6 0.27 1.34 6 0.27 1.36 6 0.32 1.30 6 0.31
Mid 1.50 6 0.36 1.56 6 0.35 1.44 6 0.29 1.46 6 0.3 1.52 6 0.46 1.64 6 0.39
Apex 2.09 6 0.61§ 2.16 6 0.58‡ 1.83 6 0.44*,§ 2.01 6 0.52§ 2.20 6 0.73§ 2.09 6 0.57§
SR E/S ratio 1.01 6 0.12 0.99 6 0.13 1.02 6 0.12 1.03 6 0.11 1.03 6 0.14 0.90 6 0.09†,‡
Late diastolic SR (sec
1) 0.42 6 0.16 0.40 6 0.15 0.46 6 0.16 0.36 6 0.13 0.45 6 0.13† 0.56 6 0.31†,‡
Base 0.42 6 0.19 0.41 6 0.17 0.45 6 0.19 0.38 6 0.17 0.43 6 0.17 0.55 6 0.19†,‡
Mid 0.39 6 0.15 0.37 6 0.15 0.44 6 0.19* 0.34 6 0.16 0.41 6 0.15† 0.52 6 0.18†,‡
Apex 0.46 6 0.22 0.44 6 0.20 0.49 6 0.24 0.37 6 0.16 0.52 6 0.23† 0.57 6 0.24†

Data are expressed as mean 6 SD.

*P < .05 versus men.
†
P < .05 versus age < 40 years.
‡
P < .05 versus age 40 to 59 years.
§
P < .05 versus base.

Age-Related Differences
Although longitudinal strain remained similar with aging, endocardial
circumferential strain and SR S gradually increased. These minor
changes are probably reflected by the small trend toward increased
LV ejection fraction in the elderly. Although diastolic SR E did
not significantly change with age, the SR E/S ratio decreased (in all strain
orientations), and diastolic SR A increased (longitudinal and endocardial
circumferential). The lack of early relaxation abnormality in the elderly
despite mitral inflow changes is probably supported by the unchanged
FEARR. The lower E/S ratio and increased SR A probably represent in-
creased myocardial stiffness (reduced passive compliance).
Technical Issues Using VVI
Endocardial Trace. The single most important step in strain anal-
ysis using VVI is drawing the endocardial contour before processing
the echocardiographic clip. Care should be taken to draw a contour
that is parallel to shortening orientation to measure the peak strains.
Figure 5 demonstrates a common pitfall that should be avoided in
tracing the endocardium. Papillary muscle systolic shortening
direction is perpendicular to endocardial shortening and actually
stretch (expressing their radial strain) within their endocardial inser-
tion region. This results in the creation of very low peak strain values
in these areas if the drawn trace includes papillary insertion areas.
Common sites for this artifact are the lower mid posterior and mid
anterior segments in the apical long-axis views. A somewhat more
epicardial trace, going behind the papillary muscle, usually avoids
inclusion its radial strain in shortening analyses. Similar reasoning
should be applied tracing abnormal ventricular geometries, as in
cardiomyopathies (hypertrophic or dilated).
The existence of an endocardial-epicardial strain gradient, espe-
cially in the short axis, must be considered when drawing the trace.
Journal of the American Society of Echocardiography Carasso et al 549
Volume 25 Number 5

Table 4 Circumferential epicardial mechanics

By gender By age (y)

All patients Men Women <40 40–59 >60

Variable (n = 120) (n = 66) (n = 54) (n = 62) (n = 37) (n = 21)

Strain (%)
11.2 6 2.3
11.2 6 2.5
11.1 6 1.9
11.6 6 2.2
10.9 6 2.2
10.6 6 2.5
Base
11.3 6 2.9
11.2 6 3.1
11.3 6 2.8
11.9 6 2.9
10.7 6 2.8
10.5 6 2.9
Mid
10.6 6 3.0
10.6 6 3.1
10.5 6 3.0
11.0 6 3
9.8 6 2.9
10.6 6 3.4
Apex
11.8 6 3.4
12.1 6 3.6
11.5 6 3.1
11.9 6 3
12.3 6 4§
10.7 6 3.0
Systolic SR (sec
1)
0.62 6 0.12
0.64 6 0.14
0.59 6 0.1*
0.65 6 0.12
0.60 6 0.13
0.52 6 0.12†,‡
Base
0.62 6 0.16
0.63 6 0.17
0.60 6 0.14
0.65 6 0.16
0.57 6 0.15†
0.57 6 0.14
Mid
0.58 6 0.16
0.59 6 0.18
0.57 6 0.14
0.61 6 0.15
0.55 6 0.15
0.56 6 0.19
Apex
0.66 6 0.19
0.70 6 0.20§
0.62 6 0.17*
0.68 6 0.17
0.70 6 0.22‡
0.57 6 0.17
Early diastolic SR (sec
1) 0.60 6 0.13 0.61 6 0.14 0.62 6 0.12 0.63 6 0.12 0.59 6 0.13 0.55 6 0.15
Base 0.61 6 0.17 0.62 6 0.18 0.60 6 0.17 0.63 6 0.17 0.61 6 0.18 0.52 6 0.15
Mid 0.57 6 0.16 0.57 6 0.18 0.57 6 0.15 0.61 6 0.16 0.54 6 0.16 0.52 6 0.16
Apex 0.63 6 0.21 0.64 6 0.20 0.62 6 0.22 0.65 6 0.18 0.66 6 0.26 0.52 6 0.16†,‡
SR E/S ratio 0.97 6 0.12 0.94 6 0.12 1.02 6 0.12 0.98 6 0.11 0.97 6 0.12 0.92 6 0.11†,‡
Late diastolic SR (sec
1) 0.20 6 0.07 0.20 6 0.15 0.21 6 0.13 0.18 6 0.05 0.23 6 0.09† 0.22 6 0.05†
Base 0.19 6 0.08 0.19 6 0.09 0.20 6 0.07 0.20 6 0.09 0.17 6 0.07 0.21 6 0.08
Mid 0.19 6 0.09 0.18 6 0.1 0.19 6 0.07 0.16 6 0.06 0.20 6 0.11† 0.23 6 0.07†
Apex 0.22 6 0.14§ 0.22 6 0.14 0.24 6 0.13§ 0.18 6 0.06 0.30 6 0.19†,§ 0.21 6 0.11‡

Data are expressed as mean 6 SD. All epicardial mechanical variables in this table are significantly different from their corresponding endocardial
values (Table 3) except for SR E/S ratio.
*P < .05 versus men.
†
P < .05 versus age < 40 years.
‡
P < .05 versus age 40 to 59 years.
§
P < .05 versus base.

Table 5 Radial mechanics

By gender By age (y)

All patients Men Women <40 40–59 >60

Variable (n = 120) (n = 66) (n = 54) (n = 62) (n = 37) (n = 21)

Strain (%) 30.0 6 7.5 29.9 6 7.6 30.2 6 7.7 30.3 6 8 29.4 6 6.9 30.4 6 7.6
Base 31.7 6 11.6 31.4 6 10.8 31.7 6 12.1 31.6 6 11.2 32.2 6 13.5 29.8 6 7.4
Mid 33.4 6 10.5 32.9 6 10.2 34.7 6 11.5 33.6 6 11 31.0 6 9.0 38.7 6 11.6‡
Apex 25.4 6 9.8§ 26.3 6 10.3§ 24.5 6 9.3§ 26.5 6 10.8‡ 24.4 6 9.1§ 24.0 6 7.3§
Systolic SR (sec
1) 1.33 6 0.28 1.36 6 0.28 1.30 6 0.29 1.33 6 0.28 1.35 6 0.27 1.30 6 0.31
Base 1.43 6 0.44 1.42 6 0.47 1.43 6 0.41 1.42 6 0.42 1.45 6 0.54 1.40 6 0.31
Mid 1.43 6 0.41 1.46 6 0.37 1.40 6 0.47 1.45 6 0.39 1.40 6 0.32 1.43 6 0.62
Apex 1.18 6 0.38§ 1.24 6 0.39§ 1.11 6 0.36§ 1.16 6 0.23‡ 1.21 6 0.33§ 1.17 6 0.39§
Early diastolic SR (sec
1)
1.29 6 0.29
1.32 6 0.30
1.26 6 0.28
1.32 6 0.31
1.31 6 0.28
1.18 6 0.22†,§
Base
1.29 6 0.42
1.27 6 0.40
1.32 6 0.45
1.33 6 0.37
1.30 6 0.51
1.16 6 0.36
Mid
1.37 6 0.40
1.40 6 0.44
1.34 6 0.34
1.37 6 0.47
1.39 6 0.41
1.38 6 0.31
Apex
1.23 6 0.42
1.30 6 0.44
1.15 6 0.38§
1.28 6 0.42
1.21 6 0.35
1.05 6 0.31†,§
SR E/S ratio 0.97 6 0.12 0.97 6 0.15 0.98 6 0.13 0.99 6 0.11 0.96 6 0.12 0.91 6 0.11†
Late diastolic SR (sec
1)
0.51 6 0.23
0.47 6 0.21
0.56 6 0.25
0.46 6 0.22
0.53 6 0.26
0.58 6 0.21
Base
0.65 6 0.37
0.62 6 0.36
0.69 6 0.37
0.58 6 0.33
0.68 6 0.45
078 6 0.30†
Mid
0.49 6 0.29
0.42 6 0.32
0.57 6 0.24*
0.49 6 0.27
0.46 6 0.35
0.54 6 0.27
Apex
0.41 6 0.24§
0.39 6 0.23§
0.43 6 0.26§
0.36 6 0.23‡
0.49 6 0.26§
0.42 6 0.21†,§

Data are expressed as mean 6 SD.

*P < .05 versus men.
†
P < .05 versus age < 40 years.
‡
P < .05 versus age 40 to 59 years.
§
P < .05 versus base.
550 Carasso et al Journal of the American Society of Echocardiography
May 2012

Table 6 Rotational mechanics

By gender By age (y)

All patients Men Women <40 40–59 >60

Variable (n = 120) (n = 66) (n = 54) (n = 62) (n = 37) (n = 21)

Endocardial rotation ( )
Base
3.4 6 2.1
3.2 6 1.9
3.5 6 2.3
3.1 6 2.2
3.6 6 1.9
4.0 6 2.3
Mid 2.2 6 2.0‡ 2.4 6 2.1‡ 2.0 6 1.9‡ 2.1 6 1.8‡ 2.3 6 1.8‡ 2.5 6 2.9‡
Apex 7.01 6 3.3§ 7.3 6 3.0§ 6.9 6 3.5§ 6.5 6 3.1§ 7.7 6 3.1†,§ 8.0 6 3.8†,§
FEARR 0.38 6 0.09 0.39 6 0.09 0.37 6 0.09 0.38 6 0.09 0.37 6 0.08 0.38 6 0.9
Twist 9.9 6 4.0 9.9 6 3.8 9.9 6 4.3 9.1 6 3.9 10.5 6 3.6 11.0 6 4.5†
Fraction of early untwist 0.37 6 0.09 0.38 6 0.1 0.36 6 0.1 0.38 6 0.09 0.35 6 0.1 0.38 6 0.1
Epicardial rotation ( )
Base
1.5 6 1.0
1.3 6 0.8
1.6 6 1.1*
1.5 6 1.0
1.4 6 0.9
1.4 6 1.0*
Mid 0.8 6 1.2‡ 0.9 6 1.3‡ 0.6 6 1.0‡ 0.9 6 1.0‡ 0.65 6 1.4‡ 0.68 6 1.4‡
Apex 2.9 6 1.5§ 3.1 6 1.4§ 2.6 6 1.5§ 2.72 6 1.4§ 3.1 6 1.6§ 2.8 6 1.4§
Twist 4.1 6 1.9 4.2 6 1.7 3.9 6 2.1 3.9 6 1.9 4.3 6 2.0 3.9 6 1.8

Data are expressed as mean 6 SD. All epicardial values are significantly different from endocardial values.
*P < .05 versus men.
†
P < .05 versus age < 40 years.
‡
P < .05 versus age 40 to 59 years.
§
P < .05 versus mid.

Table 7 Summary of major observations attempted. This resulted in increased segmental dropout (missing,
nonanalyzable, or excluded segments) in the epicardial and radial
Variable Men Women Young Old strain analysis.
Longitudinal Analyzing and Averaging Multiple Beats. The VVI algorithm
Endocardial strain [ can compensate for mild translational motion resulting from patient
Apical-to-basal gradient [ [ [ [ and respiratory cycle movements. Because beat-to-beat variability is
SR E/S ratio 1 1 1 <1 expected, analyzing and averaging consecutive beats creates more re-
Epicardial strain (apex) Y Y Y Y producible results. Stable-rhythm two-beat or three-beat acquisitions
Basal-to-apical gradient [ [ [ [ are probably best for strain analysis.
Circumferential
Endocardial strain Y [ Heart Rate and Frame Rate. More important than the frame rate
Apical-to-basal gradient [[ [[ [[ [[ per se is the number of frames per cycle. Low frame rates actually
SR E/S ratio 1 1 1 <1 pose two problems: one is the ability of the strain algorithm to track
Epicardial strain Y Y Y Y and the other is its time resolution and identification of fast changes
Radial in deformation. VVI can definitely track cycles of #20 frames. For
Strain data transferred to digital archives at 25 frames/sec, at heart rates of
Basal-to-apical gradient [[ [[ [[ [[ 80 to 60 beats/min, the time resolution would be between 33 and
SR E/S ratio 1 1 1 <1 25 msec, probably enough for the identification of most of fast defor-
Rotation apex mation changes. As we have shown here, the number of frames per
Endocardial Y [ cycle did not influence our strain measurements, apart from the ability
Epicardial Y Y Y Y to identify an ‘‘independent’’ SR A wave. This is an important message,
as most studies are usually automatically archived at a basic rate of
25 frames/sec for data storage space considerations. Remembering
Because VVI does not average strain of the full myocardial thickness, that most tagging magnetic resonance imaging gold-standard mec-
placing the endocardial trace in variable distances from the endo- hanics studies had only 12 to 16 frames/cycle is also reassuring.17-20
cardial border may result in inconsistent measurements that directly Having said that, we would recommend storing a minimum of 35
affect reproducibility and ability to assess changes between consecu- frames/sec in digital archives. For very high heart rate studies as
tive studies. We established a set of tracing rules that enhanced our stress tests, full–frame rate archiving should be considered. This
reproducibility using VVI. would avoid storage of too small a number of frames per cycle at
higher heart rates.
Segmental Dropout. This study was a challenging endeavor, as
we aimed to produce standards for most of the parameters Limitations
calculated by VVI. Although patients were recruited prospectively, This prospective study was conducted at two tertiary medical centers.
with rather good quality studies, not all segments were analyzable. Thus, subjects referred for echocardiography to these centers may not rep-
Moreover, as a rule, acquired images were focused on the best en- resent the normal general population. Epicardial strain is not validated for
docardial demarcation, and epicardial visualization was not regularly VVI, yet the numbers we report for circumferential epicardial strain are
Journal of the American Society of Echocardiography Carasso et al 551
Volume 25 Number 5

Figure 5 Common tracing error analyzing endocardial strain. (Left) Trace position and (right) strain parametric maps and average
trace strain. (A) Papillary muscle long-axis shortening. Note that papillary strain is in the normal range for longitudinal shortening.
(B) Papillary muscle thickening, perpendicular to the papillary long axis but parallel to the endocardial shortening direction. (C) Mixed
papillary thickening and endocardial shortening at the papillary insertion zone resulting in lower shortening. (D) True endocardial
strain behind the papillary muscle insertion.
comparable with previously measured values by magnetic resonance im- REFERENCES
aging tagging and 2D strain.21-23 Our study included a rather small number
of patients aged >60 years. Although characterizing normal values in this 1. Leitman M, Lysyansky P, Sidenko S, Shir V, Peleg E, Binenbaum M, et al. Two-
age group is important, completely normal 2D Doppler results in a patient dimensional strain-a novel software for real-time quantitative echocardiographic
assessment of myocardial function. J Am Soc Echocardiogr 2004;17:1021-9.
without comorbidities and risk factors are not at all common.
2. Amundsen BH, Helle-Valle T, Edvardsen T, Torp H, Crosby J, Lyseggen E,
et al. Noninvasive myocardial strain measurement by speckle tracking
echocardiography: validation against sonomicrometry and tagged mag-
CONCLUSIONS
netic resonance imaging. J Am Coll Cardiol 2006;47:789-93.
3. Helle-Valle T, Crosby J, Edvardsen T, Lyseggen E, Amundsen BH,
We have presented comprehensive values for the gamut of param- Smith HJ, et al. Noninvasive method for assessment of left ventricular ro-
eters measured in the assessment of systolic and diastolic myocardial tation speckle tracking echocardiography. Circulation 2005;112:3149-56.
mechanics using VVI software. When research studies are per- 4. Pirat B, Khoury DS, Hartley CJ, Tiller L, Rao L, Schulz DG, et al. A novel
formed, it is important to recognize differences in analysis software feature-tracking echocardiographic method for the quantitation of re-
by different vendors so that normal values for each specific system gional myocardial function: validation in an animal model of ischemia-re-
should be used for comparison. Parameters of myocardial mechan- perfusion. J Am Coll Cardiol 2008;51:651-9.
ics also differ significantly with age and gender. Therefore, studies 5. Mizuguchi Y, Oishi Y, Miyoshi H, Iuchi A, Nagase N, Oki T. The functional
role of longitudinal, circumferential, and radial myocardial deformation
should incorporate age-matched and gender-matched controls for
for regulating the early impairment of left ventricular contraction and
comparisons. In addition, VVI demonstrates significant and substan-
relaxation in patients with cardiovascular risk factors: a study with two-
tial longitudinal, circumferential, and myocardial (endocardial to epi- dimensional strain imaging. J Am Soc Echocardiogr 2008;21:1138-44.
cardial) strain gradients, which may be shown to have clinical 6. Zhang Q, Fang F, Liang YJ, Xie JM, Wen YY, Yip GW, et al. A novel
relevance. Care should be taken in the technical acquisition of me- multi-layer approach of measuring myocardial strain and torsion by 2D
chanical data, taking into account possible pitfalls that we have iden- speckle tracking imaging in normal subjects and patients with heart
tified. diseases. Int J Cardiol 2011;147:32-7.
552 Carasso et al
7. Serri K, Reant P, Lafitte M, Berhouet M, Le Bouffos V, Roudaut R, et al.
Global and regional myocardial function quantification by two-
dimensional strain: application in hypertrophic cardiomyopathy. J Am
Coll Cardiol 2006;47:1175-81.
8. Carasso S, Yang H, Woo A, Vannan MA, Jamorski M, Wigle ED, et al. Systolic
myocardial mechanics in hypertrophic cardiomyopathy: novel concepts
and implications for clinical status. J Am Soc Echocardiogr 2008;21:675-83.
9. Delgado V, Tops LF, van Bommel RJ, van der Kley F, Marsan NA,
Klautz RJ, et al. Strain analysis in patients with severe aortic stenosis and
preserved left ventricular ejection fraction undergoing surgical valve
replacement. Eur Heart J 2009;30:3037-47.
10. Carasso S, Cohen O, Mutlak D, Adler Z, Lessick J, Reisner SA, et al.
Differential effects of afterload on left ventricular long- and short-axis
function: insights from a clinical model of patients with aortic valve steno-
sis undergoing aortic valve replacement. Am Heart J 2009;158:540-5.
11. Dalen H, Thorstensen A, Aase SA, Ingul CB, Torp H, Vatten LJ, et al.
Segmental and global longitudinal strain and strain rate based on echocar-
diography of 1266 healthy individuals: the HUNT study in Norway. Eur
J Echocardiogr 2010;11:149-56.
12. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA,
et al. Recommendations for chamber quantification: a report from the
American Society of Echocardiography’s Guidelines and Standards
Committee and the Chamber Quantification Writing Group. J Am Soc
Echocardiogr 2005;18:1440-63.
13. Qui~ nones MA, Otto CM, Stoddard M, Waggoner A, Zoghbi WA. Recom-
mendations for quantification of Doppler echocardiography: a report
from the Doppler Quantification Task Force of the Nomenclature and
Standards Committee of the American Society of Echocardiography. J
Am Soc Echocardiogr 2002;15:167-84.
14. Carasso S, Yang H, Woo A, Jamorski M, Wigle ED, Rakowski H. Diastolic
myocardial mechanics in hypertrophic cardiomyopathy. J Am Soc
Echocardiogr 2010;23:164-71.
Journal of the American Society of Echocardiography
May 2012
15. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability.
Psychol Bull 1979;86:420-8.
16. Gayat E, Ahmad H, Weinert L, Lang RM, Mor-Avi V. Reproducibility
and inter-vendor variability of left ventricular deformation measurements
by three-dimensional speckle-tracking echocardiography. J Am Soc
Echocardiogr 2011;24:878-85.
17. Kramer CM, Reichek N, Ferrari VA, Theobald T, Dawson J, Axel L.
Regional heterogeneity of function in hypertrophic cardiomyopathy.
Circulation 1994;90:186-94.
18. Young AA, Kramer CM, Ferrari VA, Axel L, Reichek N. Three-dimensional
left ventricular deformation in hypertrophic cardiomyopathy. Circulation
1994;90:854-67.
19. Tseng WY, Dou J, Reese TG, Wedeen VJ. Imaging myocardial fiber disarray
and intramural strain hypokinesis in hypertrophic cardiomyopathy with
MRI. J Magn Reson Imaging 2006;23:1-8.
20. Ennis DB, Epstein FH, Kellman P, Fananapazir L, McVeigh ER, Arai AE. As-
sessment of regional systolic and diastolic dysfunction in familial hypertro-
phic cardiomyopathy using MR tagging. Magn Reson Med 2003;50:638-42.
21. Azhari H, Weiss JL, Rogers WJ, Siu CO, Zerhouni EA, Shapiro EP.
Noninvasive quantification of principal strains in normal canine hearts
using tagged MRI images in 3-D. Am J Physiol 1993;264:H205-16.
22. Ishizu T, Seo Y, Enomoto Y, Sugimori H, Yamamoto M, Machino T, et al.
Experimental validation of left ventricular transmural strain gradient
with echocardiographic two-dimensional speckle tracking imaging. Eur
J Echocardiogr 2010;11:377-85.
23. Leitman M, Lysiansky M, Lysyansky P, Friedman Z, Tyomkin V, Fuchs T,
et al. Circumferential and longitudinal strain in 3 myocardial layers in
normal subjects and in patients with regional left ventricular dysfunction.
J Am Soc Echocardiogr 2010;23:64-70.
24. Biaggi P, Carasso S, Garceau P, Greutmann M, Gruner C, Tsang W, et al.
Comparison of two different speckle tracking software systems: does the
method matter? Echocardiography 2011;28:539-47.