Heart Vessels
DOI 10.1007/s00380-017-0995-2
ORIGINAL ARTICLE
Evaluation of systolic and diastolic properties of hypertensive
heart failure using speckle‑tracking echocardiography with high
volume rates
Shingo Minatoguchi1 · Masanori Kawasaki1   · Ryuhei Tanaka2 · Takashi Yoshizane3 ·
Koji Ono3 · Maki Saeki3 · Maki Nagaya3 · Hidemaro Sato4 · Kazuhiko Nishigaki1 ·
Toshiyuki Noda3 · Michael R. Zile5,6 · Shinya Minatoguchi1 
Received: 19 October 2016 / Accepted: 19 May 2017
© Springer Japan 2017
Abstract  Left ventricular (LV) properties in hypertension
(HTN) could be deteriorated by pressure overload, espe-
cially in endocardium, resulting in hypertensive heart fail-
ure (HHF). We sought to noninvasively examine LV sys-
tolic and diastolic functions at three myocardial layers in
HTN and elucidate features of HHF by speckle-tracking
echocardiography (STE) with high volume rates. We exam-
ined normotensive controls (n = 54), HTN patients without
LV hypertrophy (LVH) (n  = 50), and HTN patients with
LVH (n = 40) and HHF patients (n = 45). The HHF group
was divided into two subgroups based on their LVEF (20
heart failure with preserved ejection fraction: HFpEF and
25 heart failure with reduced ejection fraction: HFrEF).
LV layer systolic function was assessed by strain rate dur-
ing systole. Pulmonary capillary wedge pressure (PCWP)
was estimated (ePCWP) using kinetics-tracking index (KT
index) that we previously reported. HTN patients with
LVH had a significant deterioration of systolic and dias-
tolic properties compared with normotensive controls in
the absence of a significant reduction in LVEF. Patients
* Masanori Kawasaki
masanori@ya2.so‑net.ne.jp
1
Department of Cardiology, Gifu University Graduate School
of Medicine, 1‑1 Yanagido, Gifu 501‑1194, Japan
2
Department of Cardiology, Murakami Memorial Hospital,
Gifu, Japan
3
Department of Cardiology, Gifu Prefectural General Medical
Center, Gifu, Japan
4
Department of Cardiology, Sawada Hospital, Gifu, Japan
5
Medicine‑Cardiology, Medical University of South Carolina,
Charleston, SC, USA
6
Ralph H. Johnson Department of Veterans Affairs Medical
Center, Charleston, SC, USA
with HHF had further deterioration of systolic and dias-
tolic properties compared with HTN patients with LVH.
LV strain at entire myocardium and ePCWP in HFrEF was
deteriorated compared with those in HFpEF. Deterioration
of LV layer SR was more typical during systole, isovolu-
mic relaxation, and early diastole compared with control.
LV dilation was independently associated with LVEF
(r = −0.48, p < 0.001) and ePCWP (r = 0.47, p < 0.001),
and LVH (LV mass index) was independently associ-
ated with E/e′ (r  = 0.37, p  = 0.025), LVEF (r  =  −0.44,
p < 0.001), and ePCWP (r  = 0.67, p < 0.001). LV layer
analysis by STE could detect subtle impairments in systolic
function before the deterioration of LVEF in patients with
HTN. The ePCWP that was estimated using KT index was
the independent factor associated with HHF. The ePCWP
may be useful to noninvasively detect the early stage of
HHF.
Keywords  Speckle-tracking echocardiography · Left
ventricular property · Hypertensive heart failure ·
Pulmonary capillary wedge pressure
Introduction
Left ventricular (LV) systolic and diastolic properties in
patients with hypertension (HTN) could be deteriorated
by pressure overload which causes LV hypertrophy (LVH)
and fibrosis, resulting in structural and functional remod-
eling, and may lead to hypertensive heart failure (HHF)
with reduced or preserved LV ejection fraction (EF) [1].
The LV is thought to adapt to increased afterload due to
sustained HTN by developing LVH. According to the para-
digm of compensatory LV response to pressure overload,
the LV wall thickness should increase in proportion to the
13

increase in blood pressure (BP) to maintain a normal wall
stress, whereas LV dilation is considered an abnormal late
response as the LV transitions toward HF [2]. However,
transition from compensatory remodeling to HHF is not
fully understood.
Recently, layer-specific quantification of myocardial
deformation evaluated by two-dimensional (2D) STE
revealed a significant coronary heart disease in patients
with non-ST-segment elevation acute coronary syndrome
[3]. We also reported that real-time, three-dimensional (3D)
speckle-tracking echocardiography (STE) with high vol-
ume rates allows estimation of LV systolic property and
LV relaxation by strain rate (SR) in different LV layers [4].
Moreover, we reported that pulmonary capillary wedge
pressure (PCWP) that indicates LV diastolic property can
be noninvasively estimated (ePCWP) using the kinetics-
tracking index (KT index) that is defined as the ­log10 [left
atrial (LA) active emptying function (EF)/LA minimum
volume index] [5]. The ePCWP had a strong correla-
tion with PCWP obtained by right heart catheterization in
patients with normal sinus rhythm (r = 0.92) [5] and with
mitral regurgitation (r = 0.70) [6].
The aim of this study was to noninvasively examine LV
systolic and diastolic properties, such as redial SR during
systole and ePCWP in patients with HTN and HHF using
2D and 3D-STE with high volume rates and to elucidate
the features of HHF.
Methods
Subjects and study protocol
We enrolled consecutive 193 patients with HTN who were
treated with anti-hypertensive medications for more than
1 year and 61 normotensive controls. Normotensive con-
trols were recruited from subjects without HTN who were
referred to our hospital because of chest discomfort and
underwent electrocardiography and echocardiography.
Exclusion criteria in patients were the presence of chronic
atrial fibrillation, moderate-to-severe valvular heart dis-
ease, past history of surgery for structural heart disease,
cardiomyopathy, and old myocardial infarction. Patients
with coronary artery disease proven by cardiac catheteri-
zation and diabetes mellitus were also excluded, because
these diseases have might cause deterioration in endocar-
dial layer function. Subjects were also excluded if their
echocardiographic recordings were not adequate to perform
STE analysis. Patients with HTN were divided into groups
according to the presence of LVH (LV mass index >115 g/
m2 in males and >95 g/m2 in females) and HHF [7]. A
diagnosis of HHF was made according to the guidelines of
the European Society of Cardiology for the diagnosis and
13
Heart Vessels
treatment of HF [8]. The HHF group consisted of patients
with both symptoms of HF (New York Heart Association
class II or greater) and objective signs of HF, such as LVEF
<50% or E/e′ > 15 or congestion and cardiomegaly on
chest X-ray, or levels of plasma B type natriuretic peptide
(BNP) >400 pg/ml, according to the recommendations by
the National Institute for Health and Care Excellence clini-
cal guidelines 108 [9]. Furthermore, the HHF group was
divided into two subgroups: HF preserved EF (EF ≥50%
that was measured by bi-plane modified Simpson’s method
using 2D-STE) (HFpEF) and HF reduced EF (EF <50%)
(HFrEF). The present study was approved by the ethics
committee of our institution, and all patients gave informed
consent before participation (IRB Number: G143).
LV properties assessed by the conventional
echo‑Doppler and 2D‑STE examination
A standard echo-Doppler and 2D-STE examination was
performed using an SC2000 ultrasound system (Siemens
Medical Solutions Inc., Mountain View, CA, USA) with
a 4V1c transducer (1.25–4.5 MHz). Echocardiographic
measurements were made according to the American Soci-
ety of Echocardiography criteria [7]. BP was measured just
before echocardiographic examination. Doppler measure-
ments of mitral inflow E-wave and A-wave velocity were
made from the apical four-chamber view, and e′ was meas-
ured at the septal mitral annulus to obtain E/e′. LV end-
diastolic diameter was used as an index of a compensa-
tory response to volume overload by hypertension, and LV
mass was calculated using the 2D area-length method and
indexed for body surface area (LVMI) that was used as an
index of a compensatory response to pressure overload by
hypertension. Relative wall thickness was calculated by the
following formula:
Relative wall thickness = LV posterior wall thickness
× 2/LV end-diastolic diameter.
LV diastolic stress that had a strong relationship with
BNP was calculated by the following formula [10]:
LV diastolic stress = 0.334 × ePCWP
× LV end-diastolic dimension /{LV end-diastolic thickness
× (1 + LV end-diastolic thickness/LV end-diastolic dimension)}
kdynes/cm2 .
We previously reported the concepts of the KT index [5].
In brief, as LV diastolic dysfunction progresses, LA volume
continues to increase [11, 12]. Therefore, we employed
LA volume as the denominator in the KT index to evalu-
ate PCWP. As LV diastolic dysfunction progresses, active
LAEF gradually decreases due to elevated LV stiffness and
LV filling pressure [13]. Therefore, we employed active
Heart Vessels
LAEF as the numerator in the KT index. Therefore, as
diastolic dysfunction progresses, the KT index decreases.
Moreover, Hsiao et al. previously reported the logarithmic
correlation between LV filling pressure and LA distensibil-
ity [(maximum LA volume index − minimum LA volume
index)/minimum LA volume index] that is similar to the
total LAEF [(maximum LA volume index − minimum LA
volume index)/maximum LA volume index] [12]. There-
fore, we determined the logarithmic correlation between
ePCWP and LA function. Finally, ePCWP was calculated
by the following formula as we previously reported [5]
(Fig. 1):
ePCWP = 10.8−12.4
× log (LA active EF/LA minimum volume index),
where LA active EF = (pre-atrial contraction LA vol-
ume  − minimum LA volume)/pre-atrial contraction LA
volume × 100%.
LV properties assessed by 3D‑STE
A full-volume scan can be acquired in only one heart
beat from an apical approach using the SC2000 with the
Fig. 1  Representative measurements of left atrial function. a Rep-
resentative velocity vector imaging of left atrium. LAEF left atrial
emptying function. b Time–left atrial volume curve constructed by
velocity vector imaging. The time–left atrial volume curve is shown
4Z1c 3D volumetric transducer (1.5–3.5 MHz). Data were
stored digitally in a raw format and exported to a separate
workstation equipped with software for off-line analysis.
The recently developed software was applied to automati-
cally divide the LV into 16 segments and to calculate LV
strain and SR with high volume rates at the endocardium,
mid-wall, and epicardium. Using this software, LV pha-
sic peak SR during isovolumic contraction (IVC), sys-
tole, IVR, early diastole, and atrial contraction could be
calculated in one heart beat with a high volume rate [4]
(Fig. 2).
We measured the LV strain in the longitudinal, radial,
and circumferential directions of the entire myocardium.
Regarding SR of the three layers, radial deformation
imaging at the endocardium can yield important infor-
mation about dysfunction in patients with LVH [14] and
radial SR was reported to be closely related to the con-
tractility index even under altered loading conditions [15].
Therefore, we focused on the measurement of radial SR
and examined radial SR at three layers of the myocar-
dium using 3D-STE dividing myocardium into endocar-
dium, mid-wall, and epicardium. In addition, we used the
average radial SR of the three layers as myocardial SR
as an orange line, and time–dV/dt curve is shown as a blue line. The
ECG is shown as a white line below curves. The QRS wave in ECG is
shown as QRS
13

Fig. 2  Full-volume acquisition and automated measurement of the
left ventricle using 3D-STE with high volume rates and novel soft-
ware. a Representative automatic tracing of the endocardial and epi-
cardial borders of a four slice display for full-volume acquisition: one
apical four-chamber view and three short axis views (apical, middle
and basal portions of the left ventricle). Upper panel shows deteri-
orated layer vector imaging. Lower panel shows layer vector imag-
ing in the control group. b Representative time–radial strain rate
(total layer SR). It has been reported that LV contractil-
ity which is load-independent index might be estimated
by LV myocardial peak strain using tissue Doppler echo-
cardiography imaging and 2D-STE [16–18]. Therefore,
we used LV myocardial peak strain as an index of sys-
tolic function. Every measurement was performed in
three sequential cardiac cycles. These parameters were
assessed by 3D-STE and the newly developed software,
which has good reproducibility as we previously reported
[4]. The interobserver correlation coefficient and lim-
its of agreement of SR during systole and IVR in the
endocardium and epicardium were excellent (correlation
13
Heart Vessels
curve in endocardium. c Representative time–radial strain rate curve
in epicardium. Blue arrow shows SR at isovolumic contraction, red
arrow shows SR at systole, yellow arrow shows SR at isovolumic
relaxation, orange arrow shows SR at early diastole, and white arrow
shows SR at atrial contraction. Note that SR at systole (red arrow) in
endocardium was 2.7, whereas SR at systole in epicardium was 2.3.
Note that SR at isovolumic contraction (blue arrow) in endocardium
was 1.0, whereas SR at isovolumic contraction in epicardium was 0.4
coefficient  = 0.74–0.81; average of limits of agreement:
from −0.08 to +0.12; and one standard deviation of lim-
its of agreement, 0.60–0.74) [4].
Statistical analyses
Continuous variables are expressed as the mean ± stand-
ard deviation, and categorical variables are presented as
frequency and percentage. Differences among groups for
the categorical variables were assessed by a Chi-square
test or Fisher’s exact test, and differences among groups
for the continuous variables were analyzed by analysis of
Heart Vessels

variance followed by a Bonferroni test for post hoc com- Results

parisons. Multivariate logistic analysis was performed to
determine the independent echocardiographic features
of HHF in all subjects. Moreover, we performed multi-
variate logistic analysis in patients with HTN, because
increased BP is a risk factor of HHF. Receiver operating
characteristic curve analysis was used to discriminate
patients with HHF, and we determined the sensitivity,
specificity, positive and negative predictive value, and
area under the curve. A p value <0.05 was considered
statistically significant. All statistical analyses were per-
formed using Stat View version 5.0 (SAS Institution Inc.,
Cary, NC, USA).
Study population
We enrolled 193 patients with HTN who were treated
with anti-hypertension medications for more than 1 year
(9 ± 5 years) and 61 normotensive controls. Patients with
chronic atrial fibrillation (n  = 8), moderate-to-severe val-
vular heart disease (n  = 9), past history of surgery for
structural heart disease (n  = 4), cardiomyopathy (n  = 4),
old myocardial infarction (n = 7), and coronary artery dis-
ease (n  = 8) were excluded. HTN patients with diabetes
mellitus (n = 15) and normotensive controls with diabetes

Table 1  Characteristics of the study population

Control (n = 54) HTN without LVH HTN with LVH HFpEF (n = 20) HFrEF (n = 25) p value
(n = 50) (n = 40)

Men, n (%) 33 (61) 30 (60) 23 (58) 12 (60) 15 (60) 0.99

Age (years) 69 ± 9 70 ± 9 69 ± 6 75 ± 13 69 ± 15 0.24
BSA ­(m2) 1.62 ± 0.18 1.64 ± 0.22 1.64 ± 0.19 1.57 ± 0.19 1.67 ± 0.23 0.62
SBP (mmHg) 126 ± 13 140 ± 13* 139 ± 13* 135 ± 15* 134 ± 19* <0.001
DBP (mmHg) 73 ± 9 74 ± 10 73 ± 10 71 ± 12 73 ± 14 0.88
Heart rate (bpm) 60 ± 9 62 ± 10 60 ± 10 70 ± 12 72 ± 15* <0.001
Smoking, n (%) 10 (19) 9 (18) 9 (23) 5 (25) 6 (24) 0.93
Hyperlipidemia, n (%) 18 (33) 19 (38) 14 (35) 10 (50) 8 (32) 0.49
BNP (pg/ml) NA 19 ± 17 79 ± 90 709 ± 552 904 ± 694# <0.001
NYHA classification NA NA NA II: 2 (10) II: 2 (8) 0.73
III: 5 (25) III: 9 (36)
IV: 13 (65) IV: 14 (56)
Medications
 ACEIs or ARBs, n (%) – 32 (64) 29 (73) 13 (65) 17 (68) 0.85
 Beta blockers, n (%) – 9 (18) 10 (25) 12 (60)* 14 (56)* <0.001
 Ca-antagonists, n (%) – 26 (52) 23 (58) 14 (70) 15 (60) 0.71
 Diuretics, n (%) – 9 (18) 7 (18) 13 (65)* 17 (68)* <0.001
 Statins, n (%) 19 (35) 16 (32) 15 (38) 4 (20) 5 (20) 0.44
Echocardiographic parameters
 LV EDD (mm) 45.4 ± 4.2# 44.8 ± 3.9# 48.5 ± 4.4* 47.0 ± 3.7 54.0 ± 7.4*,†,# <0.001
 LV ESD (mm) 28.0 ± 2.9 28.3 ± 3.3 29.6 ± 4.4 30 4 ± 4.7 43.0 ± 8.2*,†,# <0.001
 LVEF (%) 67 ± 6 67 ± 6 69 ± 8 64.7 ± 8.8 41.8 ± 8.6*,†,# <0.001
 Relative wall thickness 0.39 ± 0.05 0.43 ± 0.07* 0.45 ± 0.08* 0.52 ± 0.08*,# 0.43 ± 0.08† <0.001
 LV mass index (g/m2) 87 ± 14† 95 ± 11†,# 132 ± 21* 143 ± 27* 161 ± 33*,# <0.001
 LV EDV by 3D (ml) 84 ± 16 82 ± 18 107 ± 30* 90 ± 32 146 ± 54*,†,# <0.001
 LV ESV by 3D (ml) 35 ± 9 36 ± 11 50 ± 16* 48 ± 20* 99 ± 45*,†,# <0.001
 LVEF by 3D (%) 59 ± 6 57 ± 7 53 ± 6* 47 ± 9* 34 ± 9*,†,# <0.001

LVH left ventricular hypertrophy, HTN hypertension, HHF hypertensive heart failure, HFpEF heart failure with preserved ejection fraction,
HFrEF heart failure with preserved ejection fraction, SBP systolic blood pressure, DBP diastolic blood pressure, BNP B type natriuretic peptide,
HYHA New York Heart Association, NA not applicable, EDD end-diastolic dimension, ESD end-systolic dimension, EF ejection fraction, SR-S
strain rate during systole, E/e′ ratio of early diastolic inflow velocity to annular tissue velocity, ePCWP estimated pulmonary capillary wedge
pressure
* p < 0.05 vs. control; † p < 0.05 vs. HFpEF; # p < 0.05 vs. HTN with LVH

13

mellitus (n  = 7) were also excluded. Three patients with
HTN and one normotensive control were excluded because
of echocardiographic recordings that were not adequate to
perform STE analysis. Finally, we examined LV properties
in 135 patients with HTN and 54 normotensive controls.
The HHF group was divided into two subgroups based on
Fig. 3  Differences in left ventricular diastolic properties and strain
among five groups. LV left ventricular, HTN hypertension, LVH left
ventricular hypertrophy, HFpEF heart failure with preserved ejection
fraction, HFrEF heart failure with reduced ejection fraction, E/e′ ratio
of early diastolic inflow velocity to annular tissue velocity, ePCWP
13
Heart Vessels
their LVEF (20 HFpEF and 25 HFrEF). The characteristics
of the study population are listed in Table 1. Age, gender,
diastolic BP, smoking, hyperlipidemia, and body surface
area were not significantly different among the five groups,
whereas systolic BP was significantly higher in HTN
patients compared with normotensive controls.
estimated pulmonary capillary wedge pressure, ANOVA analysis
of variance. Co: p < 0.05 vs. control; H: p < 0.05 vs. HTN without
LVH group; HH: p < 0.05 vs. HTN with LVH group; P: p < 0.05 vs.
HFpEF group; R: p < 0.05 vs. HFrEF group
Heart Vessels
Echocardiographic LV properties
Left ventricular myocardial peak strain and pha-
sic and layer SR were obtained within three min-
utes (145 ± 11 s) using 3D-STE at a volume rate of
68  ± 6 vps (range 60–86 vps) and the novel software
(Fig.  2). Data assessed by the conventional echocardi-
ography and 3D-STE are shown in Table 1 and Fig. 3.
HTN patients with LVH had a significant deterioration
of systolic and diastolic properties compared with nor-
motensive controls. As shown in Fig. 3, ePCWP in HTN
patients with LVH more increased than HTN patients
without LVH, and further deteriorated in patients with
HFpEF regardless of similar LVEF. In analysis of sub-
groups of HHF, LV circumferential strain at entire myo-
cardium, LV radial strain at entire myocardium, and LV
diastolic stress in HFrEF were deteriorated compared
with those in HFpEF, whereas there was no difference
in LV longitudinal strain at entire myocardium between
HFrEF and HFpEF (Fig. 3).
The 3D-STE assessment of LV phasic and layer SR is
summarized in Fig. 4. The radial SR of the LV endocar-
dium during systole in normotensive controls was sig-
nificantly greater than the corresponding SR of the epi-
cardium by 29%. Deterioration of LV layer SR in HTN
groups was more typical during systole, isovolumic relax-
ation, and early diastole comparing with normotensive
controls. LV SR during isovolumic contraction and sys-
tole in epicardium were reduced comparing with those in
endocardium.
LV dilation, as a compensatory response to volume over-
load, was independently associated with LVEF (r = −0.48,
p < 0.001) and ePCWP (r  = 0.47, p < 0.001). LVH (LV
mass index), as a compensatory response to pressure over-
load, was independently associated with E/e′ (r  = 0.37,
p  = 0.025), LVEF (r  =  −0.44, p < 0.001), and ePCWP
(r = 0.67, p < 0.001) (Table 2).
Independent features of HHF
Multivariate logistic analysis was performed to determine
the independent echocardiographic parameters associated
with HHF in all subjects and in the HTN group (Table 3).
The parameters were adjusted by systolic BP and the use
of beta blockers, because these parameters were affected
by both systolic BP and the use of beta blockers. The
ePCWP and E/e′ were the parameters that showed the
association with HHF in all subjects and in the HTN
group. Receiver operating characteristic curve analysis
for the discrimination of HHF using E/e′ and ePCWP is
shown in Fig. 5.
Discussion
In the present study, we demonstrated that HTN patients
with LVH and HFpEF had a significant deterioration of
systolic and diastolic properties compared with normo-
tensive controls despite the significant absence of a reduc-
tion in LVEF. Patients with HHF had further deterioration
of systolic and diastolic properties compared with HTN
patients with LVH. LV systolic function at endocardium
assessed by radial SR during systole was already reduced
even in HTN patients without LVH compared with that in
normotensive controls. Multivariate regression analysis
revealed that ePCWP was an echocardiographic parameter
that showed the strongest independent factors with HHF.
LV dilatation and LVH were most associated with ePCWP.
Echocardiographic features of patients
with hypertension
In the present study, we demonstrated that LV systolic
function assessed by the LV strain in the radial and cir-
cumferential directions of the entire myocardium and LV
diastolic function assessed by ePCWP were already dete-
riorated even in HTN patients without LVH compared with
normotensive controls despite the significant absence of a
reduction in LVEF. These findings indicated that evaluation
of endocardial and diastolic function is important to detect
subtle deterioration by HTN.
Utility of three‑dimensional strain echocardiography
to assess LV layer function
The heart is a complex mechanical organ that undergoes
cyclic changes in multiple dimensions. The left ventricle
is composed of three myocardial layers and the orienta-
tion of myofibers changes from a right-handed helix in
the subendocardium and circumferential orientation in
the mid-wall to a left-handed helix in the subepicardium.
Therefore, LV function may be different among three myo-
cardial layers. Since the endocardium is most susceptible
to the deleterious effects of interstitial fibrosis and hypop-
erfusion, the endocardial function can be deteriorated at an
earlier stage of HTN [19–22]. LV deformation is charac-
terized by circumferential, longitudinal, and radial motion.
Radial deformation imaging at the endocardium can yield
important information about dysfunction in patients with
LVH [10]. Covell reported that one of the principal pur-
poses of cardiac shearing deformation lies in amplify-
ing the 15% shortening of myocytes into >40% radial LV
wall thickening, which in turn results in a >60% change in
LVEF [23]. Therefore, we focused on the LV deformation
in radial direction and measured LV radial phasic SR in
13
 Heart Vessels

Fig. 4  Left ventricular phasic

and layer radial strain rate using
3D-STE. Analysis of variance
showed significant differences
among all multi-group compari-
son except epicardium during
isovolmic relaxation (p = 0.07)
and epicardium during atrial
contraction (p = 0.09). STE
speckle-tracking echocardi-
ography, SR strain rate, HTN
hypertension, LVH left ventricu-
lar hypertrophy, HFpEF heart
failure with preserved ejection
fraction, HFrEF heart failure
with reduced ejection fraction,
ANOVA analysis of variance.
†
p < 0.05 vs. control; ‡p < 0.05
vs. HTN without LVH;
*p < 0.05 vs. endocardium

13
Heart Vessels

Table 2  Age adjusted determinants of left ventricular end-diastolic composed of three layers seem to be best represented by
dimension and left ventricular mass index in all subjects (n = 189) 3D-STE with high volume rates.
Univariate regression Multivariate regres-
sion Echocardiographic features of hypertensive heart
r p value r p value failure

LVEDD Left ventricular radial SR during systole was reduced in

 SBP (mmHg) – 0.55 patients with HFrEF even in epicardium and this was asso-
 E/e′ – 0.14 ciated with LV enlargement, suggesting an entire myocar-
 LVEF −0.48 <0.001 −0.172 <0.001 dial insult in HHF. In HFrEF, there are abnormalities in the
 ePCWP (mmHg) 0.47 <0.001 0.436 <0.001 pressure–volume relationship during systole that includes
LV mass index decreased LVEF, stroke volume, and LV contractility [24].
 SBP (mmHg) – 0.10 In the present study, LVEF and radial SR during systole
 E/e′ 0.37 <0.001 0.766 0.025 which was reported to be closely related index of contrac-
 LVEF −0.44 <0.001 −0.626 <0.001 tility in HTN patients with HHF was further depressed than
 ePCWP (mmHg) 0.67 <0.001 3.768 <0.001 that in HTN patients without HHF. LV diastolic stress that
LVEDD left ventricular end-diastolic dimension, SBP systolic blood
indicates LV diastolic property, and had a strong relation-
pressure, ePCWP estimated pulmonary capillary wedge pressure, LV ship with BNP in both LV systolic and diastolic dysfunc-
left ventricular, EF ejection fraction tions (r  = 0.942) [10] deteriorated in HTN patients with
LVH and HFpEF comparing with HTN patients without
LVH regardless of similar LVEF. This finding indicated
three myocardial layers. In the present study, we demon- that LV diastolic function has already deteriorated before
strated that radial SR at endocardium during systole was a the deterioration of LVEF.
sensitive property to evaluate the deterioration of LV sys-
tolic function. The radial SR at endocardium during systole Potential clinical applications
reduced in HTN patients with LVH and HFpEF despite the
significant absence of a reduction in LVEF. The technique and methods described in this study have
A transmural insult or progression of disease results in several important potential applications in the clinical care
concomitant myocardial layer dysfunction, leading to a of patients with HTN. Some patients with concentric LVH
reduction in LVEF. We found that the deterioration of three but no symptoms or signs of HF have decreased LV con-
myocardial layer functions including epicardium leads to tractility [1, 25]. LV relaxation is impaired first in patients
HHF, suggesting that the complex mechanics of the heart with diastolic dysfunction and even in mild HTH [26].

Table 3  Beta blocker usage- and systolic blood pressure-adjusted predictors for hypertensive heart failure and their cutoffs and predictive accu-
racies
Odds ratio 95% confidence interval p value

All subjects (n = 189)

 E/e′ 1.248 1.078–1.444 0.003
 ePCWP (mmHg) 1.653 1.247–2.190 <0.001
Hypertension group (n = 135)
 E/e′ 1.227 1.060–1.420 0.006
 ePCWP (mmHg) 1.619 1.027–2.137 <0.001
Cutoff Sensitivity (%) Specificity (%) PPV (%) NPV (%) AUC

All subjects (n = 189)

 E/e′ 12.3 71 78 50 90 0.81
 ePCWP (mmHg) 11.0 93 86 68 98 0.93
Hypertension group (n = 135)
 E/e′ 12.0 71 73 56 83 0.77
 ePCWP (mmHg) 11.0 93 81 71 96 0.90

ePCWP estimated pulmonary capillary wedge pressure, PPV positive predictive value, NPV negative predictive value, AUC area under curve

13

Fig. 5  Receiver operating char-
acteristic curve analysis for the
discrimination of hypertensive
heart failure in patients with
hypertension. AUC area under
the curve, ePCWP estimated
pulmonary capillary wedge
pressure
Thus, it may be important to know the time of deterioration
of systolic and diastolic functions due to pressure overload
to prevent HHF, since systolic and diastolic dysfunctions in
HF are irreversible [27]. Assessment of radial layer-specific
SR measured by STE may allow more accurate estimation
of the onset of the decline in LV systolic function.
In multivariate analysis, ePCWP assessed by echocardiog-
raphy was independent features of HHF. As shown in Fig. 3,
ePCWP in HTN patients with LVH more increased than HTN
patients without LVH, and further deteriorated in patients
with HFpEF regardless of similar LVEF. We have reported
several studies that demonstrated the usefulness of ePCWP in
the clinical settings [28, 29]. In multivariate analysis, ePCWP
assessed during sinus rhythm before AF ablation had the most
significant association with the success of AF ablation (with-
out AF recurrence) [28]. Using 13 mmHg of ePCWP as the
optimal cut-off value, the positive and negative predictive val-
ues were 92 and 44%, respectively (area under curve = 0.81)
[28]. In addition, there have been no previous studies that
evaluated PCWP in healthy subjects in a relatively large
population due to an ethical reason, because the measure-
ment of PCWP requires an invasive method. We reported a
study that demonstrated that the ePCWP was maintained
8.3  ± 1.8 mmHg in healthy male and 8.2 ± 2.3 mmHg in
healthy female with advancing age due to compensation by
13
Heart Vessels
an increase in active LAEF [29]. Likewise, our method to
estimate ePCWP may be helpful to predict the early stage of
HHF as well as our previous reports.
Study limitations
There were several limitations in the present study. First, the
number of patients, particularly the number of patients with
HHF, was small. Although we distinguished the LV proper-
ties in HFpEF from those in HFrEF, because HFpEF should
have different LV properties from HFrEF, the small num-
ber of patients with HHF might not give a strong statistical
power. Second, although we could measure LV phasic SR
by 3D-STE, the role of phasic SR in the assessment of LV
function has not been elucidated in larger population study.
Third, although usefulness of ePCWP has been evaluated in
several clinical studies [5, 6, 28, 29], it should be validated in
a multicenter study with a large number of patients. Finally,
the software used for 3D analysis has not been released and
has not been validated by comparing with magnetic reso-
nance imaging (MRI) as a gold standard to assess LV func-
tion and structure. We did not use MRI to assess LV mass,
particularly to exclude secondary cardiomyopathy. Further
studies including a validation study by both STE and MRI
with a large number of patients will be needed.
Heart Vessels
Conclusions
Left ventricular systolic and diastolic properties evaluated
by STE were more impaired in HTN patients with HHF
than HTN patients without HHF. LV layer analysis by 2D-
and 3D-STE could detect subtle impairment in systolic
function before the deterioration of LVEF in patients with
HTN. The ePCWP was the independent factor associated
with HHF. Real-time one-beat STE with high volume rates
may be useful to noninvasively and comprehensively evalu-
ate LV properties and detect the early stage of HHF.
Acknowledgements  The authors acknowledge the help of Dr. Atsuko
Nishimura and Ms. Mikako Nakata for preparation of the manuscript.
Compliance with ethical standards  atrial volume: a population-based study. J Am Coll Cardiol
Funding  This research received no grant from any funding agency in
the public, commercial or not-for-profit sectors.
Conflict of interest  The authors declare that there is no conflict of
interest.
References
1. Zile MR, Brutsaert DL (2002) New concepts in diastolic dys-
function and diastolic heart failure: part 1: diagnosis, prog-
nosis, and measurement of diastolic dysfunction. Circulation
105:1387–1393
2. Ganau A, Devereux RB, Roman MJ, de Simone G, Pickering
TG, Saba PS, Vargiu P, Simongini I, Laragh JH (1992) Patters of
left ventricular hypertrophy and geometric remodeling in essen-
tial hypertension. J Am Coll Cardiol 19:1550–1558
3. Sarvari SI, Haugaa KH, Zahid W, Bendz B, Aakhus S, Aaberge
L, Edvardsen T (2013) Layer-specific quantification of myocar-
dial deformation by strain echocardiography may reveal signifi-
cant CAD in patients with non-ST-segment elevation acute coro-
nary syndrome. JACC Cardiovasc Imaging 6:535–544
4. Saeki M, Sato N, Kawasaki M, Tanaka R, Nagaya M, Watanabe
T, Ono K, Noda T, Zile MR, Minatoguchi S (2015) Left ventric-
ular layer function in hypertension assessed by myocardial strain
rate using novel one-beat real-time three-dimensional speckle
tracking echocardiography with high volume rates. Hypertens
Res 38:551–559
5. Kawasaki M, Tanaka R, Ono K, Minatoguchi S, Watanabe T,
Iwama M, Hirose T, Arai M, Noda T, Watanabe S, Zile MR,
Minatoguchi S (2015) A novel ultrasound predictor of pulmo-
nary capillary wedge pressure assessed by the combination of
left atrial volume and function: a speckle tracking echocardiog-
raphy study. J Cardiol 66:253–262
6. Kawase Y, Kawasaki M, Tanaka R, Nomura N, Fijii Y, Ogawa
K, Sato H, Miyake T, Kato T, Tsunekawa T, Okubo M, Tsuchiya
K, Tomita S, Matsuo H, Minatoguchi S (2016) Noninvasive esti-
mation of pulmonary capillary wedge pressure in patients with
mitral regurgitation: a speckle tracking echocardiography. J Car-
diol 67:192–198
7. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E,
Pellikka PA, Picard MH, Roman MJ, Seward J, Shanewise JS,
Solomon SD, Spencer KT, Sutton MS, Stewart WJ (2005) Rec-
ommendations for chamber quantification. J Am Soc Echocardi-
ogr 18:1440–1463
8. Swedberg K, Cleland J, Dargie H, Drexler H, Follath F, Koma-
jda M, Tavazzi L, Smiseth OA, Gavazzi A, Haverich A, Hose
A, Jaarsma T, Korewicki J, Levy S, Linde C, Lopez-Sendon
JL, Nieminen MS, Pierard L, Remen WJ (2005) Guidelines for
the diagnosis and treatment of chronic heart failure. Eur Heart J
26:1115–1140
9. National Institute for Health and Care Excellence (NICE) (2010)
Clinical guidelines for chronic heart failure 108 (CG108)
10. Iwanaga Y, Nishi I, Furuichi S, Noguchi T, Sase K, Kihara Y,
Goto Y, Nonogi H (2006) B-type natriuretic peptide strongly
reflects diastolic wall stress in patients with chronic heart fail-
ure: comparison between systolic and diastolic heart failure. J
Am Coll Cardiol 47:742–748
11. Pritchett AM, Mahoney DW, Jacobsen SJ, Rodeheffer RJ,
Karon BL, Redfield MM (2005) Diastolic dysfunction and left
45:87–92
12. Hsiao SH, Huang WC, Lin KL, Chiou KR, Kuo FY, Lin SK,
Cheng CC (2010) Left atrial distensibility and left ventricular
filling pressure in acute versus chronic severe mitral regurgita-
tion. Am J Cardiol 105:709–715
13. Murata M, Iwana S, Tamura Y, Kondo M, Kouyama K, Murata
M, Ogawa S (2008) A real-time three-dimensional echocardio-
graphic quantitative analysis of left atrial function in left ven-
tricular diastolic dysfunction. Am J Cardiol 102:1097–1102
14. Minoshima M, Noda A, Nishizawa T, Hara Y, Sugiura M, Iino
S, Nagata K, Koike Y, Murohara T (2009) Endocardial radial
strain imaging and left ventricular relaxation abnormalities in
patients with hypertrophic cardiomyopathy or hypertensive left
ventricular hypertrophy. Circ J 73:2294–2299
15. Weidemann F, Janmal F, Sutherland GR, Claus P, Kowalski M,
Hatle L, Scheerder I, Bijnens B, Radenakers FE (2002) Myo-
cardial function defined by strain rate and strain during alter-
ations in inotropic states and heart rate. Am J Physiol Heart
Circ Physiol 283:H792–H799
16. Ferferieva V, Van den Bergh A, Claus P, Jasaityte R, Veulemans
P, Pellens M, Gerche A, Rademakers F, Herijgers P, D’hooge J
(2012) The relative value of strain and strain rate for defining
intrinsic myocardial function. Am J Physiol Heart Circ Physiol
302:H188–H195
17. Hoit BD (2011) Strain and strain rate echocardiography and
coronary artery disease. Circ Cardiovasc Imaging 4:179–190
18. Abraham TP, Dimaano VL, Liang HY (2007) Role of tissue
Doppler and strain echocardiography in current clinical prac-
tice. Circulation 116:2597–2609
19. Lumens J, Delhaas T, Arts T, Cowan BR, Young AA (2006)
Impaired subendocardial contractile myofiber function in
asymptomatic aged humans, as detected using MRI. Am J
Physiol Heart Circ Physiol 291:H1573–H1579
20. Mor-Avi Lang RM, Badano L, Belohlavek M, Cardim NM, Der-
umeaux G, Galderisi M, Marwick T, Nagueh SF, Sengupta PP,
Sicari R, Smiseth OA, Smulevitz B, Takeuchi M, Thomas JD,
Vannan M, Voigt U, Zamorano JL (2011) Current amd evolving
echocardiographic techniques for the quantitative evaluation of
cardiac mechanics: ASE/EAE consensus statement on methodol-
ogy and indications. J Am Soc Echocardiogr 24:277–313
21. Monte IP, Mangiafico S, Buccheri S, Bottari VE, Lavanco V,
Arcidiacono AA, Leggio S, Deste W, Tamburino C (2015) Myo-
cardial deformational adaptations to different forms of training: a
real-time three-dimensional speckle tracking echocardiographic
study. Heart Vessels 30:386–395
22. Kowalik E, Kowalski M, Klisiewicz A, Hoffman P (2016) Global
area strain is a sensitive marker of subendocardial damage in
13

adults after optimal repair of aortic coarctation: three-dimen-
sional speckle-tracking echocardiography data. Heart Vessels
31:1790–1797
23. Covell JW (2008) Tissue structure and ventricular wall mechan-
ics. Circulation 118:699–701
24. Carabello BA, Zile MR, Tanaka R, Cooper G IV (1992) Left
ventricular hypertrophy due to volume overload versus pressure
overload. Am J Physiol 263:H1137–H1144
25. Aurigemma GP, Silver KH, Prrest MA, Gaasch WH (1995) Geo-
metric changes allow normal ejection fraction despite depressed
myocardial shortening in hypertensive left ventricular hypertro-
phy. J Am Coll Cardiol 26:195–202
26. Eshoo S, Ross DL, Thomas L (2009) Impact of mild hyperten-
sion on left atrial size and function. Circ Cardiovasc Imaging
2:93–99
13
Heart Vessels
27. Vasan RS, Levy D (1996) The role of hypertension in the patho-
genesis of heart failure. Arch Intern Med 156:1789–1796
28. Kawasaki M, Tanaka R, Miyake T, Matsuoka R, Kaneda M,
Minatoguchi S, Hirose T, Ono K, Nagaya M, Sato H, Kawase Y,
Tomita S, Tsuchiya K, Matsuo H, Noda T, Minatoguchi S (2016)
Estimated pulmonary capillary wedge pressure assessed by
speckle tracking echocardiography predicts successful ablation
in paroxysmal atrial fibrillation. Cardiovasc Ultrasound 14:6.
doi:10.1186/s12947-016-0049-4
29. Kawasaki M, Tanaka R, Ono K, Minatoguchi S, Watanabe T,
Arai M, Nishigaki K, Noda T, Watanabe S, Minatoguchi S
(2016) Impact of gender and healthy aging on pulmonary cap-
illary wedge pressure estimated by the kinetics-tracking index
using two-dimensional speckle tracking echocardiography.
Hypertens Res 39:327–333