Int J Cardiovasc Imaging

DOI 10.1007/s10554-017-1217-9

ORIGINAL PAPER

Normal references of right ventricular strain values by two-

dimensional strain echocardiography according to the age
and gender
Jae‑Hyeong Park1 · Jin‑Oh Choi2 · Seung Woo Park2 · Goo‑Yeong Cho3 ·
Jin Kyung Oh1 · Jae‑Hwan Lee1 · In‑Whan Seong1

Received: 11 May 2017 / Accepted: 22 July 2017

© Springer Science+Business Media B.V. 2017

Abstract Right ventricular (RV) strain values by
2-dimensional strain echocardiography (STE) can be used
as objective markers of RV systolic function. However,
there is little data about normal reference RV strain val-
ues according to age and gender. We measured normal
RV strain values by STE. RV strain values were analyzed
from the digitally stored echocardiographic images from
NORMAL (Normal echOcardiogRaphic diMensions and
functions in KoreAn popuLation) study for the measure-
ment of normal echocardiographic values performed in
23 Korean university hospitals. We enrolled total 1003
healthy persons in the NORMAL study. Of them, we
analyzed 2-dimensional RV strain values in 493 subjects
(261 females, mean 47 ± 15 years old) only with echocar-
diographic images by GE machines. Their LV systolic and
diastolic functions were normal. RV fractional area change
was 48 ± 6% and tricuspid annular plane systolic excursion
was 23 ± 3 mm. Total RV global longitudinal peak sys-
tolic strain ­(RVGLStotal) was −21.5 ± 3.2%. Females had
higher absolute ­RVGLStotal (−22.3 ± 3.3 vs −20.7 ± 2.9%,
p < 0.001) than males. Younger (<50 years old) females had
higher absolute ­RVGLStotal (−22.9 ± 3.2 vs −20.5 ± 2.8%,
* Jae‑Hyeong Park
jaehpark@cnu.ac.kr
1
Division of Cardiology, Department of Internal Medicine,
Chungnam National University Hospital, Chungnam National
University School of Medicine, Daejeon 35015, South Korea
2
Division of Cardiology, Department of Medicine, Samsung
Medical Center, Sungkyunkwan University School
of Medicine, Seoul, South Korea
3
Division of Cardiology, Department of Internal Medicine,
Seoul National University and Cardiovascular Center,
Seoul National University Bundang Hospital, Seongnam,
South Korea
p < 0.001) than age matched males. R ­ VGLStotal in females
gradually increased according to age (p for trend = 0.002)
and becomes almost similar in age ≥50 years. However,
this trend was not seen in males (p for trend = 0.287), and
younger males had similar ­ RVGLStotal value to that of
older males (age ≥50 years, −20.5 ± 2.8 vs −20.9 ± 3.1%,
p = 0.224). We calculated normal RVGLS values in normal
population. Females have higher absolute strain values than
males, especially in younger age groups (<50 years old).
Keywords Echocardiography · Strain · Normal
population · Reference value
Introduction
Because the presence of right ventricular (RV) dysfunction
is a poor prognostic factor in several cardiovascular dis-
eases [1–3], measurement of RV systolic function and iden-
tification of RV systolic dysfunction is important. However,
measurement of RV systolic function is limited and chal-
lenging in routine clinical practice due to its complex mor-
phology, complex systolic motion and prominent myocar-
dial trabeculations [4].
In standard echocardiography, fractional area change
of RV (RVFAC), tricuspid annular plane systolic excur-
sion (TAPSE), and RV myocardial performance index can
be used as objective indicators of RV systolic function [5].
However, all of these parameters do not always represent
intrinsic myocardial function.
Strain echocardiography is a reliable echocardiographic
modality that can measure myocardial mechanical defor-
mation and evaluate regional and global myocardial sys-
tolic function [6]. Also, it can detect subclinical myocardial
dysfunction in the early stages of various cardiovascular

13
Vol.:(0123456789)

diseases [7]. Since the clinical application of RV strain
values requires a normal reference range, it is necessary
to define normal range using a normal population without
any cardiovascular diseases. However, there is little data
on normal reference values for RV strains. We had per-
formed a nation-wide multicenter trial to find normal echo-
cardiographic parameters in the Korean population called
the NORMAL (Normal echOcardiogRaphic diMensions
and functions in KoreAn popuLation) study [8]. Previously,
we published normal left ventricular strain values by two-
dimensional strain analysis [9]. In this substudy, we wished
to define the normal reference of RV strain values accord-
ing to age and gender.
Methods
Study populations
The NORMAL study was conducted to establish normal
reference values for echocardiographic examination in
Koreans [8]. From January 2011 to March 2014, a total
of 23 higher education hospitals participated in this study.
We prospectively included normal adults (aged 20–79
years) without significant cardiac disorders or clinical ill-
nesses that could affect cardiac structures and function
such as hypertension and diabetes. All participants were
assessed using comprehensive echocardiography. After
echocardiographic studies, subjects with structural or func-
tional abnormalities in heart valves or ventricles were also
excluded from this study. All study participants agreed
to provide their information for research purposes. If the
subject refused to participate in the study, he or she was
excluded. Written consent of this study was waived if the
patient verbally stated their wish to participate. The study
protocol was approved by the Institutional Review Boards
of all included institutions.
Standard conventional echocardiography
Echocardiography was performed and measured in each
institution according to the standard method outlined by
the American Society of Echocardiography [5]. All con-
ventional echocardiographic parameters were measured
and averaged over three cardiac cycles. All echocardio-
graphic data was stored in DICOM format and transmitted
electronically to the Echocardiography Core Laboratory
(ECL) at the Samsung Medical Center. Trained ECL staffs
reviewed and reanalyzed all echocardiographic images with
commercially available software (EchoPAC PC software,
GE Medical Systems, Horten, Norway). Basal and mid
cavity RV minor dimensions and RV longitudinal diam-
eter were measured at end-diastolic frame of RV focused
13
Int J Cardiovasc Imaging
apical four-chamber view. Conventional echocardiographic
parameters indicating RV systolic function were measured
in a standard manner according to the guideline [10]. The
RVFAC was calculated using the percent area change of the
end-diastolic and end-systolic areas of the RV in the apical
four-chamber views. TAPSE measured the contraction dis-
tance of the RV tricuspid annular plane along its longitudi-
nal plane during systole.
Two‑dimensional strain echocardiography
We performed this substudy only with patients who under-
went echocardiography using GE machines and measured
the RV strain values using an apical four-chamber view.
We analyzed digitally stored images in cineloop format at
approximately 60 frames/s using offline software (EchoPAC
PC BT20, GE Medical Systems, Horten, Norway). We
selected the best 2-dimensional echocardiographic image
and manually traced RV endocardial border at the end-
systolic frame. After the tracing, a speckle-tracking region
of interest was automatically selected. Because the RV has
thinner free wall than the LV, the width of region of interest
was adjusted to accommodate the total thickness of the RV
free wall and exclude the pericardium. The software tracked
echo-dense speckles in each frame using the sum of abso-
lute differences. RV free wall and interventricular septum
were divided into three segments (basal, mid and apical).
Because the software automatically checked the tracing
quality of each myocardial segment, the endocardial bor-
der or the region of interest was readjusted to obtain better
tracing results in cases of poor tracing quality. If the qual-
ity of tracing was consistently poor even after readjustment,
we used another echocardiographic loop. RV strain analy-
sis was feasible in about 98.4% of our subjects (493/501).
Myocardial velocities were derived as the ratio between
frame to frame displacements in apical four-chamber view.
RV global longitudinal peak systolic strain (RVGLS) and
strain rate (RVGLSR) were measured from the averaging
values of the segments. Strain values of the RV free wall
­(RVGLSfree wall and ­RVGLSRfree wall) and interventricular
septum ­(RVGLSseptum and ­RVGLSRseptum) were measured
by the averages of three segmental values (base, mid and
apex). Total strain values (­RVGLStotal and R ­ VGLSRtotal)
were calculated by the average of six segmental values of
the RV free wall and interventricular septum. Because tis-
sue shortening has negative strain values, a smaller value
(that is, a higher absolute value) indicates better RV sys-
tolic function (Fig. 1).
Statistical analysis
All statistical analyses were performed using commer-
cial software including SPSS Statistics version 22 (IBM
Int J Cardiovasc Imaging
Fig. 1  Comparison of total right ventricular global longitudinal peak
systolic strain ­(RVGLStotal) and total right ventricular global longitu-
dinal peak systolic strain rate ­(RVGLSRtotal) according to the gender
and the age groups. Females (open circles) shows better ­RVGLStotal
Co., Chicago, IL, USA) and MedCalc (version 12.3.0.0,
Mariakerke, Belgium). Continuous data is expressed as
mean ± standard deviation, and 95% confidence intervals
(CIs) for each variable are provided. For defining normal
reference values, we presented upper (97.5th percentile)
and lower (2.5th percentile) limit of values. An independ-
ent t test was performed to compare mean values between
males and females. One-way analysis of variance was per-
formed to evaluate whether mean values differed according
to age groups. Interobserver and intraobserver variabilities
in RVGLS were evaluated with intraclass correlation coef-
ficient (ICC) in 20 randomly selected subjects by two inde-
pendent observers (Park and Oh). To determine reproduc-
ibility, the same observer who was blinded to the former
results measured RV strain values for each selected subject
again at a separate time (at least 2 weeks later). p values
<0.05 were considered statistically significant.
Results
Baseline characteristics of the study population
We originally enrolled a total of 1003 healthy persons in
NORMAL study. Of them, 501 subjects had measurable
echocardiographic images done by GE echocardiographic
machines. Because of poor tracking quality of the RV, eight
subjects were excluded in this study. Finally, we analyzed
2-dimensional RV strain in 493 subjects (261 females,
mean 47 ± 15 years old). Their clinical characteristics are
listed in the Table 1. Females had lower anthropometric
measurements (weight, height, body mass index and body
surface area) than males (p < 0.001). Also, systolic and
diastolic blood pressures were slightly lower in females.
However, all values were within normal limits. There was
than males (closed rectangles) in younger age group (<50 years old).
Also, females have better R
­ VGLSRtotal than males in the age groups
<60 years old. *p value <0.01, ns no statistical significance
no significant difference in heart rate between males and
females.
Conventional echocardiographic findings
Conventional echocardiographic findings were summarized
in Table 1. Their LV ejection fraction (LVEF) was 62 ± 4%,
and there was a significant difference in LVEF between
males and females. RV basal and mid ventricular dimen-
sions were 34 ± 4 and 26 ± 4 mm, and RV long axis dimen-
sion was 66 ± 11 mm. Females had lower RV dimensions
than males (p < 0.001). Also, RV areas were significantly
smaller in females than in males (p < 0.001). Conventional
echocardiographic indicators of RV systolic function,
RVFAC and TAPSE, were within normal limits. RVFAC
in females was higher than in males (49 ± 6 vs 47 ± 6%,
p = 0.001). However, there was no significant difference in
TAPSE between males and females (23 ± 3 vs 22 ± 3 mm,
p = 0.734).
Strain echocardiographic findings of RV
Strain values are listed in Table 2. ­RVGLStotal was
−21.5 ± 3.2%, ­RVGLSfree wall was −26.4 ± 4.2%, and
­RVGLSseptum was −16.6 ± 3.7%. Females had statistically
higher absolute ­RVGLStotal (−22.3 ± 3.3 vs −20.7 ± 2.9%,
p < 0.001), ­RVGLSfree wall (−27.2 ± 4.8 vs −25.5 ± 3.8%,
p < 0.001), and ­RVGLSseptum (−17.4 ± 4.0 vs −15.8 ± 3.2%,
p < 0.001) than males.
Strain values according to gender and age group are sum-
marized in Tables 3 and 4. ­RVGLStotal and R ­ VGLSfree wall
showed statistical difference between age groups, especially
in females. However, there was no statistical difference in
male group. Younger females (<50 years old) had higher
absolute ­RVGLStotal and ­RVGLSfree wall than older females
(≥50 years old, −22.9 ± 3.2 vs −21.6 ± 3.3%, p = 0.002,
13
 Int J Cardiovasc Imaging

Table 1  Demographic and Variables Total (n = 493) Females (n = 261) Males (n = 232) p value
conventional echocardiographic
characteristics Baseline characteristics
Age (years) 47 ± 15 47 ± 15 47 ± 15 0.660
Weight (kg) 62 ± 10 56 ± 8 68 ± 9 <0.001
Height (cm) 164 ± 9 158 ± 6 170 ± 7 <0.001
2
BSA ­(m ) 1.7 ± 0.2 1.6 ± 0.1 1.8 ± 0.1 <0.001
BMI (kg/m2) 23 ± 3 22 ± 3 23 ± 2 <0.001
SBP (mmHg) 120 ± 12 117 ± 13 123 ± 11 <0.001
DBP (mmHg) 73 ± 10 71 ± 10 74 ± 9 0.002
Heart rate (bpm) 68 ± 10 69 ± 9 68 ± 10 0.087
Conventional echocardiographic findings
LVEF (%) 62 ± 4 63 ± 4 62 ± 4 0.054
RVDbase (mm) 34 ± 4 33 ± 4 35 ± 4 <0.001
RVDmidventricle (mm) 26 ± 4 25 ± 3 27 ± 4 <0.001
RVDlong axis (mm) 66 ± 11 64 ± 10 69 ± 10 <0.001
RVEDA ­(cm2) 15 ± 3 14 ± 3 17 ± 3 <0.001
RVESA ­(cm2) 8±2 7±2 9±2 <0.001
RVFAC (%) 48 ± 6 49 ± 6 47 ± 6 0.001
TAPSE (mm) 23 ± 3 23 ± 3 22 ± 3 0.734

SD standard deviation, BSA body surface area, SBP systolic blood pressure, DBP diastolic blood pressure,
LVEF left ventricular ejection fraction, RVDbase right ventricular dimension, base, RVDmidventricle right ven-
tricular dimension, midventricle, RVDlong axis right ventricular dimension, long axis, RVEDA right ventricu-
lar end-diastolic area, RVESA right ventricular end-systolic area, RVFAC right ventricular fractional area
change, TAPSE tricuspid annular plane systolic excursion

Table 2  Reference values of right ventricular 2-dimensional longitudinal peak systolic strain and strain rate in the normal population and sepa-
rated by gender
Variable Total (n = 493) Females (n = 261) Males (n = 232) p value
Mean ± SD 95% CI Mean ± SD 95% CI Mean ± SD 95% CI

RVGLS (%)
RVGLSseptum −16.6 ± 3.7 −17.0 to −16.3 −17.4 ± 4.0 −17.9 to −16.9 −15.8 ± 3.2 −16.2 to −15.4 <0.001
RVGLSfree wall −26.4 ± 4.2 −26.8 to −26.0 −27.2 ± 4.8 −27.8 to −26.7 −25.5 ± 3.8 −26.0 to −25.1 <0.001
RVGLStotal −21.5 ± 3.2 −21.8 to −21.2 −22.3 ± 3.3 −22.7 to −21.9 −20.7 ± 2.9 −21.0 to −20.3 <0.001
RVGLSR (1/s)
RVGLSRseptum −1.1 ± 0.3 −1.12 to −1.07 −1.1 ± 0.3 −1.18 to −1.12 −1.0 ± 0.2 −1.06 to −1.00 <0.001
RVGLSRfree wall −1.6 ± 0.4 −1.66 to −1.60 −1.7 ± 0.4 −1.76 to −1.67 −1.5 ± 0.3 −1.57 to −1.49 <0.001
RVGLStotal −1.4 ± 0.3 −1.38 to −1.34 −1.4 ± 0.3 −1.46 to −1.40 −1.3 ± 0.2 −1.31 to −1.25 <0.001

SD standard deviation, 95% CI 95% confidence interval, RVGLSseptum right ventricular global longitudinal peak systolic strain, septum, RVGLS-
right ventricular global longitudinal peak systolic strain, free wall, RVGLStotal right ventricular global longitudinal peak systolic strain,
free wall
total, RVGLSRseptum right ventricular global longitudinal peak systolic strain rate, septum, RVGLSRfree wall right ventricular global longitudinal
peak systolic strain rate, free wall, RVGLSRtotal right ventricular global longitudinal peak systolic strain rate, total

and −27.9 ± 4.2 vs −26.3 ± 4.6%, p = 0.006). ­RVGLStotal
and ­RVGLSfree wall in females gradually decrease accord-
ing to age (p for trend = 0.002 and <0.001, respectively)
and becomes almost similar in age ≥50 years. However,
this trend was not seen in males (p for trend = 0.246,
and 0.287, respectively), and younger males had similar
­RVGLStotal and ­RVGLSfree wall values to that of older males
(−20.5 ± 2.8 vs −20.9 ± 3.1%, p = 0.224 and −25.2 ± 3.6
vs −26.0 ± 3.9%, p = 0.127, respectively). Females had
similar age- and gender difference of R ­ VGLSRtotal and
­RVGLSRfree wall like R
­ VGLStotal and ­RVGLSfree wall.
Gender difference of R­ VGLStotal was observed only in
the younger population (<50 years old) (−22.9 ± 3.2% in
females vs −20.5 ± 2.8% in males, p < 0.001). However,
there was no gender difference in the older population
(≥50 years old, −21.6 ± 3.3% in females vs −20.9 ± 3.1%
in males, p = 0.128).

13
Int J Cardiovasc Imaging

Table 3  Right ventricular 2-dimensional longitudinal strain and strain rates according to gender and age
Age (years) RVGLSseptum (%) RVGLSfree wall (%) RVGLStotal (%)
Mean ± SD 95% CI Mean ± SD 95% CI Mean ± SD 95% CI

Female (n = 261)
~30 (n = 46) −17.5 ± 3.4 −18.5 to −16.5 −28.2 ± 3.8 −29.3 to −27.1 −22.8 ± 2.5 −23.6 to −22.1
31–40 (n = 51) −17.8 ± 3.7 −18.9 to −16.8 −28.5 ± 4.7 −29.9 to −27.2 −23.2 ± 3.6 −24.2 to −22.2
41–50 (n = 57) −17.8 ± 4.2 −18.9 to −16.7 −27.3 ± 4.0 −28.3 to −26.2 −22.5 ± 3.1 −23.4 to −21.7
51–60 (n = 51) −16.4 ± 3.5 −17.4 to −15.5 −27.1 ± 4.2 −28.3 to −25.9 −21.8 ± 3.1 −22.6 to −20.9
61~ (n = 56) −17.4 ± 4.7 −18.7 to −16.2 −25.2 ± 4.9 −26.5 to −23.9 −21.3 ± 3.7 −22.3 to −20.3
p value for trend 0.412 <0.001 0.002
Male (n = 232)
~30 (n = 38) −15.8 ± 2.9 −16.7 to −14.9 −25.8 ± 3.7 −27.0 to −24.6 −20.8 ± 2.9 −21.8 to −19.9
31–40 (n = 51) −15.5 ± 2.6 −16.6 to −14.8 −24.7 ± 3.5 −25.7 to −23.7 −20.1 ± 2.5 −20.8 to −19.4
41–50 (n = 45) −15.5 ± 3.7 −16.7 to −14.4 −25.3 ± 3.6 −26.4 to −24.2 −20.4 ± 3.0 −21.3 to −19.5
51–60 (n = 46) −16.1 ± 3.4 −17.2 to −15.1 −25.9 ± 4.2 −27.2 to −24.7 −21.0 ± 3.3 −22.0 to −20.0
61~ (n = 52) −15.9 ± 3.5 −16.9 to −14.9 −26.1 ± 3.8 −27.1 to −25.0 −21.0 ± 3.0 −21.8 to −20.1
p value for trend 0.580 0.246 0.287

p value from analysis of variance test with regression

RVGLSseptum right ventricular global longitudinal peak systolic strain, septum, RVGLSfree wall right ventricular global longitudinal peak systolic
strain, free wall, RVGLStotal right ventricular global longitudinal peak systolic strain, total, SD standard deviation, CI confidence interval

Table 4  Right ventricular 2-dimensional longitudinal peak systolic strain rate according to gender and age
Age (years) RVGLSRseptum (%) RVGLSRfree wall (%) RVGLSRtotal (%)
Mean ± SD 95% CI Mean ± SD 95% CI Mean ± SD 95% CI

Female (n = 261)
~30 (n = 46) −1.20 ± 0.22 −1.27 to −1.14 −1.78 ± 0.33 −1.88 to −1.68 −1.49 ± 0.20 −1.55 to −1.43
31–40 (n = 51) −1.18 ± 0.26 −1.26 to −1.11 −1.87 ± 0.47 −2.00 to −1.74 −1.52 ± 0.33 −1.62 to −1.43
41–50 (n = 57) −1.19 ± 0.32 −1.27 to −1.10 −1.68 ± 0.30 −1.76 to −1.60 −1.43 ± 0.29 −1.51 to −1.36
51–60 (n = 51) −1.08 ± 0.24 −1.15 to −1.02 −1.70 ± 0.32 −1.79 to −1.61 −1.39 ± 0.24 −1.46 to −1.32
61~ (n = 56) −1.09 ± 0.26 −1.16 to −1.02 −1.57 ± 0.34 −1.66 to −1.47 −1.33 ± 0.26 −1.37 to −1.26
p value for trend 0.006 <0.001 <0.001
Male (n = 232)
~30 (n = 38) −1.04 ± 0.24 −1.12 to −0.96 −1.56 ± 0.30 −1.65 to −1.46 −1.30 ± 0.26 −1.38 to −1.21
31–40 (n = 51) −1.04 ± 0.23 −1.11 to −0.98 −1.51 ± 0.31 −1.60 to −1.42 −1.28 ± 0.24 −1.34 to −1.21
41–50 (n = 45) −1.04 ± 0.23 −1.11 to −0.97 −1.54 ± 0.30 −1.63 to −1.45 −1.29 ± 0.23 −1.36 to −1.22
51–60 (n = 46) −1.01 ± 0.24 −1.08 to −0.94 −1.49 ± 0.30 −1.58 to −1.40 −1.25 ± 0.25 −1.32 to −1.17
61~ (n = 52) −1.03 ± 0.24 −1.09 to −0.96 −1.57 ± 0.30 −1.65 to −1.49 −1.30 ± 0.24 −1.36 to −1.23
p value for trend 0.579 0.871 0.866

p value from analysis of variance test with regression

RVGLSRseptum right ventricular global longitudinal peak systolic strain rate, septum, RVGLSRfree wall right ventricular global longitudinal peak
systolic strain rate, free wall, RVGLSRtotal right ventricular global longitudinal peak systolic strain rate, total, SD standard deviation, CI confi-
dence interval

Variability ­ VGLStotal were 0.810, 0.917, and 0.897, respectively.

R
Intraobserver ICCs of R­ VGLSRseptum, ­RVGLSRfree wall and
Intraobserver ICCs of R
­ VGLSseptum, ­RVGLSfree wall and ­RVGLSRtotal were 0.785, 0.850, and 0.816, respectively.
­ VGLStotal were 0.919, 0.877, and 0.937, respectively.
R Interobserver ICCs of ­RVGLSRseptum, ­RVGLSRfree wall and
Interobserver ICCs of R
­ VGLSseptum, ­RVGLSfree wall and ­RVGLSRtotal were 0.666, 0.824, and 0.790, respectively.

13

Discussion
In this study, we showed normal references of RV strain
values according to age and gender. Females had lower
RV strain values than males, especially in the younger age
group (<50 years old). However, this gender difference was
not evident in the older age group (≥50 years old).
Because strain measurement can give active myocardial
contraction, strain values of RV based on 2-dimensional
echocardiography have advantages over other conventional
echocardiographic parameters [11, 12]. Also, strain can
detect subtle myocardial changes in subclinical cardiovas-
cular diseases [12, 13]. Therefore it seems reasonable to
adapt the strain echocardiography technique to measure RV
deformation. However, there are several problems of sim-
ply applying strain echocardiography to RV strain measure-
ment. One is lack of normal reference values. Also, there
are several algorithms and different vendors of echocardio-
graphic machines that can be used to measure RV strain.
Thus, the types of echocardiographic machines and strain
algorithms being used should be considered when measur-
ing RV strain.
In this study, we analyzed RV strain with speckle track-
ing algorithm in 493 normal subjects free of cardiovascu-
lar diseases. R­ VGLStotal, ­RVGLSfree wall and ­RVGLSseptum
were −21.5 ± 3.2, −26.4 ± 4.2, and −16.6 ± 3.7%, respec-
tively. Our results seems similar to the data published by
Meris et al. [14] and Fine et al. [15]. Meris et al. included
100 normal subjects (46 females, 43 ± 13 years old) with
GE algorithm and reported that normal RVGLS of RV
free wall and RVGLS of septum were −28.7 ± 4.1% and
−19.8 ± 3.4%, respectively. Fine et al. reported their refer-
ence range of RV free wall strain by speckle tracking algo-
rithm from 116 patients without cardiopulmonary disease.
Their results showed mean global RV free wall strain was
−26 ± 4% with a corresponding reference range of −18% to
−34% [15].
Females had lower (better) ­RVGLStotal (−22.3 ± 3.3 vs
−20.7 ± 2.9%, p < 0.001), ­RVGLSfree wall (−27.2 ± 4.8 vs
−25.5 ± 3.8%, p < 0.001), and ­RVGLSseptum (−17.4 ± 4.0
vs −15.8 ± 3.2%, p < 0.001) than males. There are several
reports showing females to have lower RV strain values
than males [16]. Muraru et al. published a similar gen-
der difference showing that females have better RV strain
values than males. They reported that there was little age-
related difference in RV strain values.
Younger females had better RV strain values than older
females in our study. The gender difference was observed
only in the younger age group (<50 years old). However,
the age-related difference was not observed in the male
group. The exact mechanism of age-and gender difference
is not yet known. The effect of sex hormones on cardiac
function can be one possible explanation of this gender
13
Int J Cardiovasc Imaging
difference [17, 18]. Because estrogen and estrogen recep-
tor signaling can have favorable cardioprotective effects
including modulation of inflammation, cardiac hypertro-
phy and vascular function [19], estrogen can influence
myocardial function especially in the premenopausal
period. Estrogen can protect cardiac fibrosis by inhibiting
matrix metalloproteinase-9 expression and increasing tis-
sue inhibitor-I expression [20]. Because the effect of sex
hormones in females usually decreases after menopause,
and menopause usually occurs in their fifties, the insig-
nificant difference in gender in the older group may be
explained through the decrease of the favorable effects
of female sex hormones. This finding also supports the
effect of sex hormones on cardiac function. Another pos-
sible explanation is the sex-related differences in age-
associated downregulation of β1-adrenergic receptors
in the human ventricular myocardium [21]. Female RVs
have more pronounced decreases of β1-adrenergic recep-
tors with increasing age and show a closer relationship
between β1-adrenergic receptor density and age than in
males.
Limitations
This study has several limitations. First, we analyzed
echocardiographic images only with GE echocardio-
graphic machines and GE algorithm. Because there can
be a vender difference in the measurement of RV strain,
it is possible that these results cannot be generalized to
other algorithms. To overcome this limitation, other
strain algorithms need to be used to calculate RV strain
values. Secondly, we excluded patients with significant
cardiovascular diseases such as hypertension or diabetes
only based on past medical histories obtained from the
study subjects. Other clinical tests including bloodwork
were not obtained. Therefore, subjects with subclinical
cardiovascular diseases may have been included in this
study. On the other hand, it is also possible that the effects
of such conditions on cardiac structure are insignificant.
Conclusions
In this study, we present age- and gender-specific refer-
ence values of RV strains identified by two-dimensional
speckle tracking echocardiography. Females have higher
absolute strain values than males, especially in the
younger age group (age <50 years old).
Compliance with ethical standards
Conflict of interest The authors have nothing to declare.
Int J Cardiovasc Imaging

References Canadian Society of Echocardiography. J Am Soc Echocardiogr

23:685–713
11. Smiseth OA, Torp H, Opdahl A, Haugaa KH, Urheim S (2016)
1. Park SJ, Park JH, Lee HS et al (2015) Impaired RV global lon-
Myocardial strain imaging: how useful is it in clinical decision
gitudinal strain is associated with poor long-term clinical out-
making? Eur Heart J 37:1196–1207
comes in patients with acute inferior STEMI. JACC Cardiovasc
12. Negishi K, Negishi T, Hare JL, Haluska BA, Plana JC, Marwick
Imaging 8:161–169
TH (2013) Independent and incremental value of deformation
2. Zornoff LA, Skali H, Pfeffer MA et al (2002) Right ventricular
indices for prediction of trastuzumab-induced cardiotoxicity. J
dysfunction and risk of heart failure and mortality after myocar-
Am Soc Echocardiogr 26:493–498
dial infarction. J Am Coll Cardiol 39:1450–1455
13. Plana JC, Galderisi M, Barac A et al (2014) Expert consensus
3. Park JH, Park MM, Farha S et al (2015) Impaired global right
for multimodality imaging evaluation of adult patients during
ventricular longitudinal strain predicts long-term adverse out-
and after cancer therapy: a report from the American Society of
comes in patients with pulmonary arterial hypertension. J Car-
Echocardiography and the European Association of Cardiovas-
diovasc Ultrasound 23:91–99
cular Imaging. Eur Heart J Cardiovasc Imaging 15:1063–1093
4. Haddad F, Hunt SA, Rosenthal DN, Murphy DJ (2008) Right
14. Meris A, Faletra F, Conca C et al (2010) Timing and magnitude
ventricular function in cardiovascular disease, part I: anatomy,
of regional right ventricular function: a speckle tracking-derived
physiology, aging, and functional assessment of the right ventri-
strain study of normal subjects and patients with right ventricular
cle. Circulation 117:1436–1448
dysfunction. J Am Soc Echocardiogr 23:823–831
5. Lang RM, Badano LP, Mor-Avi V et al (2015) Recommenda-
15. Fine NM, Chen L, Bastiansen PM et al (2015) Reference values
tions for cardiac chamber quantification by echocardiography in
for right ventricular strain in patients without cardiopulmonary
adults: an update from the American Society of Echocardiogra-
disease: a prospective evaluation and meta-analysis. Echocardi-
phy and the European Association of Cardiovascular Imaging. J
ography 32:787–796
Am Soc Echocardiogr 28(1–39):e14
16. Muraru D, Onciul S, Peluso D et al (2016) Sex- and method-
6. Pirat B, McCulloch ML, Zoghbi WA (2006) Evaluation of global
specific reference values for right ventricular strain by 2-dimen-
and regional right ventricular systolic function in patients with
sional speckle-tracking echocardiography. Circ Cardiovasc
pulmonary hypertension using a novel speckle tracking method.
Imaging 9:e003866
Am J Cardiol 98:699–704
17. Pelzer T, Jazbutyte V, Hu K et al (2005) The estrogen receptor-
7. Sarvari SI, Haugaa KH, Anfinsen OG et al (2011) Right ventric-
alpha agonist 16alpha-LE2 inhibits cardiac hypertrophy and
ular mechanical dispersion is related to malignant arrhythmias:
improves hemodynamic function in estrogen-deficient spontane-
a study of patients with arrhythmogenic right ventricular car-
ously hypertensive rats. Cardiovasc Res 67:604–612
diomyopathy and subclinical right ventricular dysfunction. Eur
18. Gallinelli A, Angioni S, Matteo ML, Montaldo PL, Fenu
Heart J 32:1089–1096
MA, Volpe A (1999) Variations of cardiac performance and
8. Choi JO, Shin MS, Kim MJ et al (2015) Normal echocardio-
inotropism in healthy postmenopausal women treated with estro-
graphic measurements in a Korean Population Study: part I.
progestin replacement therapy. Menopause 6:49–55
Cardiac chamber and great artery evaluation. J Cardiovasc Ultra-
19. Deroo BJ, Korach KS (2006) Estrogen receptors and human dis-
sound 23:158–172
ease. J Clin Invest 116:561–570
9. Park JH, Lee JH, Lee SY et al (2016) Normal 2-dimensional
20. Voloshenyuk TG, Gardner JD (2010) Estrogen improves TIMP-
strain values of the left ventricle: a substudy of the normal echo-
MMP balance and collagen distribution in volume-overloaded
cardiographic measurements in Korean Population Study. J Car-
hearts of ovariectomized females. Am J Physiol Regul Integr
diovasc Ultrasound 24:285–293
Comp Physiol 299:R683–R693
10. Rudski LG, Lai WW, Afilalo J et al (2010) Guidelines for the
21. Lindenfeld J, Cleveland JC Jr., Kao DP et al (2016) Sex-related
echocardiographic assessment of the right heart in adults: a
differences in age-associated downregulation of human ventricu-
report from the American Society of Echocardiography endorsed
lar myocardial beta1-adrenergic receptors. J Heart Lung Trans-
by the European Association of Echocardiography, a regis-
plant 35:352–361
tered branch of the European Society of Cardiology, and the

13