European Journal of Echocardiography (2010) 11, 834–844
doi:10.1093/ejechocard/jeq084
Systolic time intervals as simple
echocardiographic parameters of left ventricular
systolic performance: correlation with ejection
fraction and longitudinal two-dimensional strain
Patricia Reant 1*, Marina Dijos 1, Erwan Donal 2,3,4, Aude Mignot 1, Philippe Ritter 1,
Pierre Bordachar 1, Pierre Dos Santos 1, Christophe Leclercq2,3,4, Raymond Roudaut 1,
Gilbert Habib 5, and Stephane Lafitte 1
1
Département de Cardiologie, CHU de Bordeaux, Université de Bordeaux, CIC-0005, Inserm U828, Plateforme Technologique d’Innovation Biomédicale, Bordeaux-Pessac, France;
CHU Pontchaillou, Rennes, France; 3Université de Rennes 1 France, CIC-IT 804 Rennes, France; 4Inserm, U 642, Rennes, France; and 5Département de Cardiologie, Hôpital La
Timone, Marseille, France
2
Received 13 April 2010; accepted after revision 1 July 2010; online publish-ahead-of-print 26 July 2010
Conventionally, the evaluation of left ventricular (LV) systolic function is based on ejection fraction assessment, which
may be supplemented by other echocardiographic techniques, such as tissue Doppler imaging, 3D evaluation, and
speckle tracking strains. However, these imaging modalities have a high technicity and are time-consuming, while
being associated with reproducibility limitations. In this context, the usefulness of simpler measurements such as systolic time intervals (STI) by pulsed Doppler echocardiography must be emphasized.
.....................................................................................................................................................................................
Methods
In this multicentre study, left ventricular ejection fraction (LVEF), dP/dtmax, LV stroke volume, myocardial longitudinal
and results
deformation, aortic pre-ejectional period (PEP, ms), and left ventricular ejection time (LVET, ms) were prospectively
investigated and compared in 134 consecutive heart failure (HF) patients and 43 control subjects. Feasibility of STI
measurements was 100%. Intra-observer reproducibility was 98% for PEP, 96% for LVET, 87% for LVEF, and 93%
for global longitudinal strain (GLS). By subgroup analyses, with increasingly altered LVEF or GLS, PEP significantly
increased, whereas significantly LVET decreased, resulting in a significantly increased PEP/LVET ratio (P , 0.001).
In the HF patients group, a correlation between LVEF and PEP/LVET was found, with r ¼ 0.55
(y ¼ 20.0083x + 0.75, P , 0.001). Based on receiver operating curve analyses, the area under the curve was 0.91
for PEP/LVET . 0.43, which allowed us to detect LVEF , 35% with a sensitivity of 87%, and a specificity of 84%.
.....................................................................................................................................................................................
Conclusion
STI can be easily and accurately measured in clinical practice, and may be used for detecting alterations in LV systolic
function. Moreover, this method is likely to have potential applications in the management of HF patients.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Left ventricular function † Echocardiography † Systolic time intervals † 2D strain
Introduction
Echocardiography is commonly used in the evaluation of left ventricular (LV) systolic function. However, the reproducibility of LV
ejection fraction has been a matter of controversy. More recently,
new techniques such as tissue Doppler imaging (TDI), 3D
evaluation, and speckle tracking echocardiography have been proposed to more precisely quantify LV systolic function.1 – 4 Yet,
these methods are technically complex, time-consuming, and
user-dependent.
Timing of mechanical cardiac events, particularly LV ejection,
described 40 years ago using a phonocardiogram, an
* Corresponding author: Hôpital Cardiologique Haut-Lévêque, Avenue de Magellan, 33605 Pessac, France. Tel: +33 557 65 65 65; fax: +33 557 65 64 10, Email: patricia.reant@chubordeaux.fr
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2010. For permissions please email: journals.permissions@oxfordjournals.org.
Downloaded from by guest on September 30, 2016
Aims
835
Doppler systolic time intervals
Methods
Study population
Consecutive HF patients with chronic dilated or ischaemic cardiomyopathy and left ventricular ejection fraction (LVEF) of ,50%, who prospectively underwent echocardiography between January and
November 2007 in three cardiologic centres: Rennes, Bordeaux, and
Marseille (France), were included. The exclusion criteria were haemodynamic instability, HF event ,3 months, atrial fibrillation, valvular
regurgitation .grade 2/4 or more than mild valvular stenosis, valvular
prosthesis, and cardiac stimulator. A control group was selected on a
population of subjects without heart disease, diabetes, or hypertension
and presenting normal echocardiography and EKG. All subjects gave
their consent to participating in the study.
Echocardiographic analysis
All patients underwent echocardiographic evaluation using a Vivid 7
machine (GE Medical Systems-Vingmed, Horten, Norway) with a
4-MHz transducer. Doppler gains were adjusted at a 100 mm s21
sweep speed. Standard echocardiogram included assessment of LV
diameters as well as volumes, left atrial end-systolic area, LVEF by
biplane Simpson method, LV peak tissue Doppler systolic velocity
(S′ , cm s21) at the lateral part of the mitral annulus and at the right ventricular free wall, dP/dtmax (mmHg s21) by Bargiggia method, aortic
time velocity integral (cm), LV cardiac output, and systolic pulmonary
artery pressure (mmHg), according to American Society of Echocardiography guidelines.22 LV diastolic function was evaluated using
mitral inflow (early diastolic peak E wave velocity, cm s21), early diastolic tissue Doppler velocity (E′ , cm s21) at the lateral part of the mitral
annulus, and mitral propagation flow velocity in colour M-mode (Vp,
cm s21). Mitral regurgitation severity was quantified using the PISA
method [effective regurgitant orifice area (EROA), mm2]. All measurements were performed off-line, blindly of clinical data, and averaged
from three cardiac cycles on digital stored images with dedicated
software (EchoPac, version BT08, GE-Vingmed, Horten, Norway) by
the same operator.
Systolic time intervals
Based on pulsed Doppler aortic acquisitions, STI were obtained: aortic
pre-ejectional period (PEP: delay from Q wave of QRS to aortic valve
opening, ms) and LV ejection time (LVET, ms) (Figure 1). In simple
terms, PEP is the interval from the onset of ventricular depolarization
to the beginning of aortic ejection. LVET represents the interval from
beginning to termination of aortic flow. PEP/LVET ratio was also calculated. Figure 1 depicts PEP and LVET measurements in both a control
subject (Figure 1A) and an HF patient (Figure 1B).
Speckle tracking strain analysis
For speckle tracking longitudinal strain analysis, 2D greyscale cardiac
cycle acquisitions were recorded in apical 4-, 2-, and 3-chamber
views, with a frame rate adjusted between 70 and 80 Hz. Using
EchoPac software, strain was measured as previously described.23
Two-dimensional strain is a non-Doppler-based method to evaluate
systolic myocardial deformations based on standard 2D acquisitions.
Global longitudinal strain (GLS) was averaged on three consecutive
cardiac cycles.
Reproducibility
A first observer performed all echocardiograms, all measurements
being repeated under blinded conditions. Intra-observer reproducibilities of STI, LVEF, and GLS were calculated by averaging the differences
between 10 measures. Intra-observer reliability of PEP and LVET was
equally performed on two different acquisitions for 20 subjects. The
studies were analysed off-line by a second observer blinded to the
values obtained for 10 patients. Inter-observer reproducibilities were
also computed.
Statistical analysis
Statistical Package for Social Sciences, version 11.5 (Chicago, IL), was
used for statistical evaluation. All values were expressed as mean +
SD. The Student t-test, ANOVA, and Friedman tests were used to
compare the patient data to those of control subjects. A P-value of
,0.05 was considered statistically significant. Pearson test and linear
regression analysis were used to seek correlations. Using the area
under the receiver operating characteristic (ROC) curve analysis, the
optimal cut-off value, sensitivity, and specificity of STI in detecting
LVEF , 35% and LVEF , 45% were determined.
Results
Population characteristics
The study population included 134 patients (63 + 17 years; 73%
men) with dilated (59%) or ischaemic cardiomyopathy (41%).
Sixteen per cent of them were in NYHA functional class I, 44.7%
in NYHA II, 16.4% in NYHA III, and 5.2% in NYHA IV; 83.1% of
them were treated with beta-blockers, while 89.4% were given
angiotensin-converting enzyme inhibitors or angiotensin receptors
blockers. Forty-three control subjects were recruited (43 + 14
years; 39% men) (Table 1).
Downloaded from by guest on September 30, 2016
electrocardiogram, and an external carotid pulse,5 – 12 can be
measured by echocardiography.13 This approach is likely to be of
benefit for several practical and clinical applications, such as LV
function evaluation under difficult conditions. Moreover, this
method could help to optimize different therapies such as
cardiac resynchronization in heart failure (HF) patients. To this
purpose, usefulness, accuracy, and reproducibility of systolic time
intervals (STI) (aortic pre-ejectional period [PEP] and left ventricular ejection time [LVET]) in detecting LV systolic dysfunction were
tested and compared with conventional and newer techniques.
Clinical applications for STI have been previously reported in
mitral valve stenosis, coronary artery disease, arterial hypertension,
or atrial fibrillation.14 – 17 Moreover, some therapies improved only
patients with abnormal PEP/LVET.18 – 20 Since PEP is a time interval
based on a fine balance between intrinsic myocardial contractility,
LV preload and afterload as well as electrical activation, LVET is
dependent on intrinsic contractility and influenced by LV preload
and afterload.5,21 However, both parameters will be dependent
to some degree on heart rate or QRS width and it is crucial to
consider the validity of STI measurements in regard to loading conditions, heart rate, and electrical activation.
This study aimed to prospectively compare STI echocardiographic measurements with conventional LV function parameters,
including myocardial strain by speckle tracking in consecutive unselected HF patients and control subjects from three cardiologic
centres.
836
P. Reant et al.
Downloaded from by guest on September 30, 2016
Figure 1 Measurements of pre-ejectional period and left ventricular ejection time. (A) In a control subject: LVEF ¼ 63%, GLS ¼ 222%,
PEP ¼ 67 ms, LVET ¼ 273 ms, PEP/LVET ¼ 0.24. (B) In a heart failure patient: LVEF ¼ 10%, GLS ¼ 23.4%, PEP ¼ 171 ms, LVET ¼ 245 ms,
PEP/LVET ¼ 0.70.
Echocardiography
Table 1 displays echocardiographic characteristics of HF and control
subjects. LVEF was 30.3 + 9.0%, and GLS was 29.3 + 3.4% in HF
patients vs. 65.5 + 4.4% (P , 0.001), and 222.0 + 1.8% (P ,
0.001), respectively, for controls. STI measurement was feasible in
100% of cases. PEP was longer, whereas LVET was shorter in
HF patients than controls (P , 0.001). Consequently, PEP/LVET
was significantly increased in HF patients (0.50 + 0.14 vs. 0.24 +
0.03 in controls, P , 0.001).
59%, n ¼ 25), and group 4 (LVEF ≥ 60%, n ¼ 36). Figure 2B
shows values of the timing time measurements in group 1
(GLS ≥ 26.25%, n ¼ 23), group 2 (212.5% ≤ GLS ≤ 26.26%,
n ¼ 85), group 3 (218.74%≤ GLS ≤ 212.6%, n ¼ 25), and
group 4 (GLS ≤ 218.75%, n ¼ 33). In both cases, PEP increased
from group 4 to group 1, while LVET decreased from group 4
to group 1, and consequently, PEP/LVET significantly increased
from group 4 to group 1.
Correlation analysis
Subgroup analysis
A subgroup analysis was performed. STI was evaluated by LV systolic function levels. Figure 2A provides values of the timing
measurements in group 1 (patients with LVEF ≤ 20%, n ¼ 21),
group 2 (21% ≤ LVEF ≤ 40%, n ¼ 95), group 3 (41% ≤ LVEF ≤
Pearson correlations were performed to compare STI measurements with LVEF, dP/dtmax, stroke volume, LV output, LV S′ TDI,
and GLS (Table 2). All STI correlated with LV systolic performance
parameters. High correlation levels were observed between STI
and LVEF (r ¼ 20.71 for PEP [P , 0.001], r ¼ 0.64 for LVET
837
Doppler systolic time intervals
Table 1
Population characteristics
Heart failure (n 5 134)
Controls (n 5 43)
P
...............................................................................................................................................................................
63 + 17
43 + 14
Male gender (%)
Left ventricular end-diastolic diameter (mm)
73
67.7 + 10.4
39
49.4 + 4.6
,0.001
Left ventricular end-systolic diameter (mm)
57.3 + 10.8
31.8 + 3.6
,0.001
Left ventricular end-diastolic volume (mL)
Left ventricular end-systolic volume (mL)
177.3 + 78.1
125.7 + 61.4
73.6 + 19.7
26.0 + 8.3
,0.001
,0.001
30.3 + 9.0
65.5 + 4.4
,0.001
3.3 + 1.0
5.2 + 1.4
4.5 + 0.7
10.6 + 2.1
,0.001
,0.001
222.0 + 1.8
72 + 9
,0.001
,0.001
Mean age (years)
Left ventricular ejection fraction (%)
Left ventricular output (mL min21)
Left ventricular lateral S′ TDI (cm s21)
dP/dtmax (mmHg s21)
Global longitudinal strain (%)
Aortic pre-ejectional period (ms)
,0.001
657 + 215
29.3 + 3.4
126 + 29
Left ventricular ejection time (ms)
259 + 34
302 + 22
,0.001
PEP/LVET
RR interval (ms)
0.50 + 0.14
902 + 187
0.24 + 0.03
860 + 139
,0.001
0.18
Heart rate (bpm)
69.6 + 15.5
71.5 + 11.7
0.44
E/E′
Mitral propagation velocity (cm s21)
10.9 + 5.7
35.9 + 11.2
6.4 + 1.7
56.7 + 16.5
,0.001
,0.001
Mitral regurgitation EROA (mm2)
,0.05
23.6 + 7.1
10.5 + 2.7
7.0 + 14.1
14.0 + 2.9
13.6 + 3.1
,0.001
,0.001
Systolic pulmonary artery pressure (mmHg)
33.8 + 12.2
21.7 + 3.2
,0.05
[P , 0.001], and r ¼ 20.78 for PEP/LVET [P , 0.001]), and
between STI and GLS (r ¼ 0.68 for PEP [P , 0.001], r ¼ 20.61
for LVET [P , 0.001], and r ¼ 0.75 for PEP/LVET [P , 0.001]).
Figure 3 depicts linear regression analysis between PEP/LVET and
LVEF in the entire population (Figure 3A) (y ¼ 20.0074x + 0.72,
r ¼ 0.78, P , 0.001) and the HF group (Figure 3B)
(y ¼ 20.0083x + 0.75, r ¼ 0.55, P , 0.01).
Linear regression analyses were undertaken to search for a
potential correlation between STI and heart rate or QRS width.
Figure 4 illustrates that no correlation was observed between
PEP and heart rate (Figure 4A), whereas a correlation was found
between LVET and heart rate (r ¼ 0.61, P , 0.001) (Figure 4B).
Moreover, no significant correlation between PEP/LVET and
heart rate was found (Figure 4C). Figure 5 illustrates that no correlation was observed between LVET and QRS width (Figure 5B),
whereas a correlation was found between PEP or PEP/LVET and
QRS width (r ¼ 0.56, P , 0.001 and r ¼ 0.43, P , 0.001, respectively) (Figure 5A and 5C ). PEP and PEP/LVET are both significantly
related to QRS width whereas LVET is significantly related to heart
rate. Similar to Weissler et al.,11 we derived data from regression
equation and we can propose these corrections to heart rate
and to QRS width: LVETc ¼ 1.5 HR + LVET; PEPc ¼ 0.46 QRS
width + PEP; PEP/LVETc ¼ 0.017 QRS width + PEP/LVET. Correlations between corrected STI and other parameters of LV systolic
function are presented on Table 3. Finally, no significant correlation
was found between mitral regurgitation EROA and STI
measurements.
Receiver operating characteristic curves
analysis
Areas under ROC curves were analysed for both PEP and PEP/
LVET in order to identify predictors of LVEF , 35% and LVEF ,
45%. Figure 6A depicts ROC curves for detecting LVEF , 35%.
The best area under the ROC curves was observed for PEP/
LVET (AUC ¼ 0.91, optimal cut-off . 0.43, sensitivity 87%, and
specificity 84%). PEP and PEP/LVET had similar and strong capabilities for detecting LVEF , 45%, with areas under ROC curves of
0.96 and 0.98, respectively (Figure 6B).
Reproducibility
Mean intra-observer variability was 13% for LVEF, 7% for GLS, 2%
for PEP, and 4% for LVET. Based on two different acquisitions,
mean intra-observer variability was 7.6% for PEP, and 4.3% for
LVET. Inter-observer variability was 15% for LVEF, 8% for GLS,
3% for PEP, and 4.5% for LVET.
Discussion
Despite recent developments in echocardiography for estimating
LV systolic function, this prospective study demonstrates that STI
measurements are precise, reproducible, and probably the simplest
method to perform. Numerous studies demonstrated that external
measurements of STI directly reflected the intra-cardiac events
measured simultaneously from high-fidelity catheters in the LV
Downloaded from by guest on September 30, 2016
1.4 + 3.0
Left atrium systolic area (cm2)
Right ventricle S′ TDI (cm s21)
838
P. Reant et al.
Downloaded from by guest on September 30, 2016
Figure 2 Systolic time intervals by left ventricular performance levels. (A) Systolic time intervals by left ventricular ejection fraction level.
Group 1 (n ¼ 21): LVEF ≤ 20%, group 2 (n ¼ 95): LVEF 21 – 40%, group 3 (n ¼ 25): LVEF 41– 59%, and group 4 (n ¼ 36): LVEF ≥ 60%.
*P , 0.05 group 4 vs. group 1, †P , 0.05 group 4 vs. group 2, ‡P , 0.05 group 4 vs. group 3, $P , 0.05 group 3 vs. group 1, £P , 0.05
group 3 vs. group 2, and §P , 0.05 group 2 vs. group 1. (B) Systolic time intervals by global longitudinal strain level. Group 1 (n ¼ 23):
GLS ≥ 26.25%, group 2 (n ¼ 85): GLS 26.26% to 212.5%, group 3 (n ¼ 25): GLS 212.6% to 218.74%, and group 4 (n ¼ 33):
GLS ≤ 218.75%. *P , 0.05 group 4 vs. group 1, †P , 0.05 group 4 vs. group 2, ‡P , 0.05 group 4 vs. group 3, $P , 0.05 group 3 vs. group
1, £P , 0.05 group 3 vs. group 2, and §P , 0.05 group 2 vs. group 1.
and the aorta8,9,11,24 – 28 but STI can be also evaluated by conventional echocardiographic techniques such as M-mode or pulsed
Doppler.13 Our study evaluated STI using Doppler measurements
in subjects with different LVEF levels, and compared them with
conventional parameters of LV systolic function, as well as to
newer technologies such as longitudinal deformations assessed
by speckle tracking methods.
Factors influencing systolic time
intervals
STI provide a temporal description of sequential phases of the
cardiac cycle physiologically influenced by the same variables as
those affecting other global performance. LV performance
indices are based on capacity to pump blood (stroke volume
cardiac output), ability to generate force (pressure, dP/dtmax) and
to shorten with each contraction (ejection fraction, longitudinal
deformation), the temporal relationships of contraction (STI), as
well as on a combination of these variables.
In subgroup analyses by systolic performance levels, with
decreasing LVEF or GLS values, PEP significantly increased,
whereas LVET significantly decreased, resulting in a significantly
increased PEP/LVET. Several studies using external measurements
of STI have shown that interventions altering LV performance
produce directionally similar changes in STI as well as in other
LV performance indices.5 – 12,24 The diminished rate of dP/dtmax
839
Doppler systolic time intervals
results in a prolongation of isovolumic contraction time and PEP.
This is accounted for by that PEP is equal to electromechanical
delay plus isovolumic contraction time. Therefore, changes in
PEP are largely dependent on isovolumic contraction time. In
our study, a better correlation existed between dP/dtmax and
PEP than with LVET. The shortening of LVET with LV dysfunction
Table 2 Pearson correlations between systolic
intervals and other left ventricular systolic performance
indices
PEP
LVET
PEP/LVET
LVEF
GLS
20.71*
0.68*
0.64*
20.61*
20.78*
0.75*
dP/dtmax
................................................................................
20.41*
0.32†
20.49*
′
LV S TDI
LV stroke volume
20.58*
20.34*
0.39*
0.65*
20.59*
20.53*
LV output
20.47*
0.36*
20.53*
*P , 0.001, †P , 0.01.
is more complex. In case of LV failure, there is a delay in the onset
of ejection, while the velocity of myocardial fibre shortening is
reduced. As the extent of fibre shortening is also reduced, LVET
tends to be shortened. LVET was shown to be related to stroke
volume.11,12,25,29,30 Because GLS is influenced by both the isovolumic contraction period and the degree of myocardial shortening,
this parameter correlated with both PEP and LVET, and even
more with PEP/LVET ratio. As PEP is lengthened, and LVET is shortened in the case of LV dysfunction, PEP/LVET is a more useful
index of overall LV performance.11,12 In our study, we observed
that PEP/LVET was more strongly correlated to LVEF and GLS
than either PEP or LVET.
All STI components have been known for long to vary inversely
with heart rate, and corrections must thus be made.21 PEP/LVET
correlated better with other LV performance measurements
than either PEP or LVET and was considered to be independent
of heart rate. However, this is still a matter of debate.31,32 In our
study, we observed a correlation between QRS width and both
PEP and PEP/LVET and between heart rate and LVET. Other
factors known to influence PEP are positive and negative inotropes
as well as preload changes, while those influencing LVET include
positive and negative inotropes, LV failure, reduced afterload, and
increased preload.5,21
Downloaded from by guest on September 30, 2016
Figure 3 Linear regression analysis between pre-ejectional period/left ventricular ejection time and left ventricular ejection fraction.
840
P. Reant et al.
Downloaded from by guest on September 30, 2016
Figure 4 Linear regression analysis between systolic time intervals and heart rate. (A) Linear regression analysis between pre-ejectional
period and heart rate. (B) Linear regression analysis between left ventricular ejection time and heart rate. (C) Linear regression analysis
between pre-ejectional period/left ventricular ejection time and heart rate.
Interest of systolic time intervals
compared with more time-consuming
techniques
Although STI is less attractive than new techniques such as 2D
strain or 3D evaluation of LV systolic function, it presents some
advantages. First, it can be obtained with each ultrasound machines
whereas new technologies require the last generation of
echographs. Second, it is characterized by its simplicity, reliability,
and rapidity to obtain whereas other tools need experience, learning curve, and require much time for analysis. Third, it is particularly helpful in case of poor quality window because pulsed
Doppler does not require a 2D image.
Recently, the Hunt study, conducted on healthy volunteers,
reported intra- and inter-observer variability of LVEF of 5% and
6% and for GLS of 3% and 6%.33 However, in clinical practice,
Doppler systolic time intervals
841
Downloaded from by guest on September 30, 2016
Figure 5 Linear regression analysis between systolic time intervals and QRS width. (A) Linear regression analysis between pre-ejectional
period and QRS width. (B) Linear regression analysis between left ventricular ejection time and QRS width. (C) Linear regression analysis
between pre-ejectional period/left ventricular ejection time and QRS width.
more important variations on the reproducibility of LVEF or GLS
are reported in HF patients: mean 18% for LVEF in the VALLIANT
study.34
In Figure 7, we propose a diagnostic algorithm allowing a
multiparametrical approach of LV systolic function in clinical
practice accordingly quality of acoustic window and available
techniques. STI measurements appear particularly helpful when
LVEF evaluation is suboptimal. Use of STI to qualitatively
appreciate LV systolic function should be particularly helpful
for initial evaluation and follow-up of patients with poor
quality window in circumstances which need this evaluation
for therapeutic reasons such as before implantation of cardiac
defibrillator or resynchronization therapy or to optimize parameters of stimulation.
842
P. Reant et al.
Table 3 Pearson correlations between systolic intervals (corrected to heart rate or QRS width) and other left
ventricular systolic performance indices
PEPc to QRS width
LVETc to heart rate
PEP/LVETc to QRS width
PEPc/LVETc
20.38*
0.36*
0.74*
20.70*
20.57*
0.47*
20.48*
0.43*
...............................................................................................................................................................................
LVEF
GLS
dP/dtmax
20.61*
0.25
20.59*
20.64*
LV S′ TDI
LV stroke volume
20.17
20.12
0.57*
0.46*
20.20
20.43*
20.23
20.24
LV output
20.15
0.52*
20.23
20.24
*P , 0.001, †P , 0.01.
Figure 7 Algorithm for evaluation of left ventricular systolic function.
Downloaded from by guest on September 30, 2016
Figure 6 Receiver operating characteristic curve analyses. (A) Prediction of LVEF , 35%. Dark grey: pre-ejectional period
(AUC ¼ 0.86, optimal cut-off . 117 ms, sensitivity 67%, specificity 89%). Light grey: pre-ejectional period/left ventricular ejection
time (AUC ¼ 0.91, optimal cut-off . 0.43, sensitivity 87%, specificity 84%). (B) Prediction of LVEF , 45%. Dark grey: pre-ejectional
period (AUC ¼ 0.96, optimal cut-off . 89 ms, sensitivity 98%, specificity 85%). Light grey: pre-ejectional period/left ventricular ejection
time (AUC ¼ 0.98, optimal cut-off . 0.33, sensitivity 95%, specificity 82%).
843
Doppler systolic time intervals
Potential clinical applications
Clinical applications for STI have been reported in patients with
mitral valve stenosis, coronary artery disease, arterial hypertension,
or atrial fibrillation.14 – 17 STI may be used to define the severity of
LV muscle dysfunction for initial assessment and for long-term
follow-up and therapy evaluation. Patients with coronary disease
and abnormal PEP/LVET are reported to exhibit a significantly
poorer 5-year prognosis.18,19 Moreover, STI measurements have
allowed physicians to demonstrate that medical therapy such as
beta-blockers or coronary bypass surgery improved only in patients
with abnormal PEP/LVET.20 STI may also be used to investigate
changes in LV performance occurring during haemodialysis.35 In
our study, ROC curve analysis illustrates the usefulness of STI, particularly PEP/LVET, in detecting LVEF , 35% (AUC 0.91, optimal
cut-off value .0.43), with an 87% sensitivity and an 84% specificity.
This could prove relevant in the selection of patients requiring
cardiac resynchronization or cardiac implantable defibrillator,
especially when precise LVEF evaluation by echocardiography
proves difficult or if the evaluation is done by novice sonographers.
STI measurements could be another echocardiographic parameters
in guiding management in dilated cardiomyopathy.36
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Limitations
15.
16.
17.
18.
19.
Conclusion
This study demonstrates the easiness of obtaining an accurate LV
systolic performance evaluation using a simple tool such as STI
measurement by Doppler echocardiography. STI correlated well
with conventional LV systolic performance indices, yet showing
less measurement variability. This method may be particularly
useful in the case of poor quality windows, for detecting LVEF ,
35%, especially when the question of resynchronization therapy
or cardiac implantable defibrillators is raised for refractory HF
patients.
20.
21.
22.
23.
24.
Acknowledgement
25.
We deeply thank Serge Cazeau for his large contribution in the
conceptualization of time intervals.
26.
Conflict of interest: none declared.
27.
References
28.
1. Reant P, Labrousse L, Lafitte S, Bordachar P, Pillois X, Tariosse L et al. Experimental validation of circumferential, longitudinal, and radial 2-dimensional strain during
dobutamine stress echocardiography in ischemic conditions. J Am Coll Cardiol
2008;51:149 –57.
2. Leitman M, Lysyansky P, Sidenko S, Shir V, Peleg E, Binenbaum M et al. Twodimensional strain—a novel software for real-time quantitative echocardiographic
assessment of myocardial function. J Am Soc Echocardiogr 2004;17:1021 –9.
3. Langeland S, D’Hooge J, Wouters PF, Leather HA, Claus P, Bijnens B et al. Experimental validation of a new ultrasound method for the simultaneous assessment
29.
30.
31.
Downloaded from by guest on September 30, 2016
Since PEP and PEP/LVET are linked to QRS width and LVET is
linked to heart rate, corrections as described in the present
study or this one of Weissler et al.11 should be applied. Moreover,
all STI measurements are dependent on loading conditions. Consequently, important precautions must be taken into account for
interpretation of the data.
of radial and longitudinal myocardial deformation independent of insonation
angle. Circulation 2005;112:2157 –62.
Donal E, Bergerot C, Thibault H, Ernande L, Loufoua J, Augeul L et al. Influence of
afterload on left ventricular radial and longitudinal systolic functions: a twodimensional strain imaging study. Eur J Echocardiogr 2009;10:914 –21.
Boudoulas H, Karayannacos PE, Lewis RP, Leier CV, Vasko JS. Effect of afterload
on left ventricular performance in experimental animals. Comparison of the preejection period and other indices of left ventricular contractility. J Med 1982;13:
373 –85.
Leier CV, Desch CE, Magorien RD, Triffon DW, Unverferth DV, Boudoulas H
et al. Positive inotropic effects of hydralazine in human subjects: comparison
with prazosin in the setting of congestive heart failure. Am J Cardiol 1980;46:
1039 –44.
Nakamura Y, Wiegner AW, Gaasch WH, Bing OH. Systolic time intervals: assessment by isolated cardiac muscle studies. J Am Coll Cardiol 1983;2:973 –8.
Ahmed SS, Levinson GE, Schwartz CJ, Ettinger PO. Systolic time intervals as
measures of the contractile state of the left ventricular myocardium in man. Circulation 1972;46:559 –71.
Talley RC, Meyer JF, McNay JL. Evaluation of the pre-ejection period as an estimate of myocardial contractility in dogs. Am J Cardiol 1971;27:384 –91.
Weissler AM. Interpreting systolic time intervals in man. J Am Coll Cardiol 1983;2:
1019 –20.
Weissler AM, Harris WS, Schoenfeld CD. Systolic time intervals in heart failure in
man. Circulation 1968;37:149 –59.
Lewis RP, Rittogers SE, Froester WF, Boudoulas H. A critical review of the systolic time intervals. Circulation 1977;56:146–58.
Stefadouros MA, Witham AC. Systolic time intervals by echocardiography. Circulation 1975;51:114 – 7.
Weissler AM, Leonard JJ, Warren JV. Observations on the delayed first heart
sound in mitral stenosis and hypertension. Circulation 1958;18:165 –8.
Lewis RP, Boudoulas H, Welch TG, Forester WF. Usefulness of systolic time
intervals in coronary artery disease. Am J Cardiol 1976;37:787 –96.
Dodek A, Burg JR, Kloster FR. Systolic time intervals in chronic hypertension:
Alterations and response to treatment. Chest 1975;68:51–5.
Boudoulas H, Lewis RP, Sherman JA, Bush CA, Dalamangas G, Forester WF. Systolic time intervals in atrial fibrillation. Chest 1978;74:629 –34.
Gillilan RE, Parnes WP, Khan MA, Bouchard RJ, Warbasse JR. The prognostic
value of systolic time intervals in angina pectoris patients. Circulation 1979;60:
268 –75.
Northover BJ. Estimation of the risk of death during the first year after acute myocardial infarction from systolic time intervals during the first week. Br Heart J 1989;
62:429–37.
Boudoulas H, Sohn YH, O’Neill W, Brown R, Weissler AM. Identification of
patients with improved survival following coronary bypass surgery. Cardiology
1984;71:247 –54.
Boudoulas H. Systolic time intervals. Eur Heart J 1990;11:93 –104.
Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R, Feigenbaum H et al.
Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. American Society of Echocardiography Committee on Standards,
Subcommittee on Quantitation of Two-Dimensional Echocardiograms. J Am Soc
Echocardiogr 1989;2:358 –67.
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.
Garrard CL Jr, Weissler AM, Dodge HT. The relationship of alterations in systolic
time intervals to ejection fraction in patients with cardiac disease. Circulation 1970;
42:455–62.
Harley A, Starmer CF, Greenfield JC Jr. Pressure-flow studies in man. An evaluation of the duration of the phases of systole. J Clin Invest 1969;48:895–905.
Leighton RF, Zaron SJ, Robinson JL, Weissler AM. Effects of atrial pacing on left
ventricular performance in patients with heart disease. Circulation 1969;40:
615 –22.
Martin CE, Shaver JA, Thompson ME, Reddy PS, Leonard JJ. Direct correlation of
external systolic time intervals with internal indices of left ventricular function in
man. Circulation 1971;44:419 –31.
Van de Werf F, Piessens J, Kesteloot H, De Geest H. A comparison of systolic
time intervals derived from the central aortic pressure and from the external
carotid pulse tracing. Circulation 1975;51:310–6.
Weissler AM, Peeler RG, Roehll WH Jr. Relationships between left ventricular
ejection time, stroke volume, and heart rate in normal individuals and patients
with cardiovascular disease. Am Heart J 1961;62:367 –78.
Greenfield JC Jr, Harley A, Thompson HK, Wallace AG. Pressure-flow studies in
man during atrial fibrillation. J Clin Invest 1968;47:2411 –21.
Cokkinos DV, Heimonas ET, Demopoulos JN, Harralambakis A, Tsartsalis G,
Gardikas CD. Influence of heart rate increase on uncorrected pre-ejection
844
period/left ventricular ejection time (PEP/LVET) ratio in normal individuals. Br
Heart J 1976;38:683 –8.
32. Spodick DH, Doi YL, Bishop RL, Hashimoto T. Systolic time intervals reconsidered. Reevaluation of the preejection period: absence of relation to heart rate.
Am J Cardiol 1984;53:1667 –70.
33. Thorstensen A, Dalen H, Amundsen BH, Aase SA, Stoylen A. Reproducibility in
echocardiographic assessment of the left ventricular global and regional function,
the HUNT study. Eur J Echocardiogr 11:149 –56.
P. Reant et al.
34. Thune JJ, Kober L, Pfeffer MA, Skali H, Anavekar NS, Bourgoun M et al. Comparison of regional versus global assessment of left ventricular function in patients
with left ventricular dysfunction, heart failure, or both after myocardial infarction:
the valsartan in acute myocardial infarction echocardiographic study. J Am Soc
Echocardiogr 2006;19:1462 –5.
35. Geleris P, Raidis C, Papadimitriou M, Boudoulas H, Metaxas P. Effect of hemodialysis on left ventricular performance. J Med 1983;14:211 –22.
36. Thomas DE, Wheeler R, Yousef ZR, Masani ND. The role of echocardiography in
guiding management in dilated cardiomyopathy. Eur J Echocardiogr 2009;10:iii15–21.
Downloaded from by guest on September 30, 2016