Download Report

Clinical research
European Heart Journal (2007) 28, 2627–2636
doi:10.1093/eurheartj/ehm072
Imaging
Changes in systolic left ventricular function in isolated
mitral regurgitation. A strain rate imaging study
Anna Marciniak1†, Piet Claus2†, George R. Sutherland1, Maciej Marciniak1, Tiia Karu1,
Aigul Baltabaeva1, Elisa Merli1, Bart Bijnens1,2, and Marjan Jahangiri1*
1
Department of Cardiology and Cardiothoracic Surgery, St George’s Hospital, London, UK; and 2University of Leuven, Belgium
Received 23 May 2006; revised 24 January 2007; accepted 8 March 2007; online publish-ahead-of-print 25 May 2007
This paper was guest edited by Prof. Genevieve Anne Derumeaux, University Hospital Lyon, INSERM EMI 0226, Lyon, France
KEYWORDS
Mitral regurgitation;
Echocardiography;
Strain rate imaging
Introduction
Mitral regurgitation (MR) is common and the severity of
regurgitation tends to increase with age.1 Previous outcome
studies have shown that patients with isolated MR who have
symptoms or a reduced ejection fraction (EF) are at high
risk of progressive left ventricular (LV) dysfunction and have
a higher late mortality despite valve repair or replacement.2,3
However, the clinical outcome among patients with asymptomatic MR is poorly deﬁned and the criteria deﬁning the highrisk subgroups are uncertain.4 The timing of mitral surgery has
remained one of the most vexing clinical problems as patients
can be minimally symptomatic even where there is signiﬁcant
MR with impaired ventricular dysfunction.
Currently, standard grey-scale ultrasound parameters
reﬂecting global LV systolic function, such as LVEF, endsystolic short-axis diameter (ESD), and end-diastolic shortaxis diameter (EDD)s are used in clinical practice to
* Corresponding author: Department of Cardiothoracic Surgery, St George’s
Hospital, Blackshaw Road, London SW17 0QT, UK. Tel: þ44 208 725 3565;
fax: þ44 208 725 2049.
E-mail address: marjan.jahangiri@stgeorges.nhs.uk
†
The ﬁrst two authors contributed equally to this study.
monitor LV function in patients with volume overload.
However, these volume-based functional parameters have
important limitations in assessing myocardial contractile
function where either regurgitant volume (RV) or increased
cavity pressure can mask any underlying changes in myocardial force development.5
In an attempt to improve the assessment of early changes
in contractility in these patients, myocardial velocities have
been assessed.5–8 However, these seem to parallel changes in
stroke volumes (SV) rather than contractility and are inﬂuenced by the exaggerated overall motion of these hyperdynamic hearts. Only reduced velocities, with increased SV,
are associated with a reduced LV contractile reserve6 or a
postoperative abnormal reduction in EF.7 Despite the
ability to identify changes in global LV function in MR,9–13
none have been able to identify subclinical LV dysfunction.
Recently, the assessment of local myocardial deformation
[using strain rate (SR) and strain (S) imaging] has been shown
to detect changes in regional systolic function at an earlier
sub-clinical stage than either conventional echocardiography or velocity imaging.14–16
The aim of this explorative study was to understand the
changes in LV regional systolic deformation in patients with
& The European Society of Cardiology 2007. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
Downloaded from by guest on October 21, 2014
Aims The aim of the present study is to understand the changes in left ventricular (LV) regional systolic
deformation based on strain rate (SR) imaging in patients with isolated mitral regurgitation (MR). Progressive LV dilatation and irreversible myocardial damage as a result of chronic isolated MR are important
causes of morbidity and mortality in patients following valve surgery. To date, there is no speciﬁc diagnostic method to detect subclinical changes in systolic function before irreversible dysfunction occurs.
Methods and results Seventy-seven individuals were studied: 54 asymptomatic patients (age 56 + 12)
with isolated non-ischaemic MR divided into three groups: mild, moderate, and severe and 23 healthy
subjects. All underwent a standard echo examination and a tissue Doppler study. A mathematical
study was carried out to predict how SR should alter with increasing dimensions and due to irreversible
myocardial damage. Radial as well as longitudinal peak systolic SR was signiﬁcantly decreased in patients
with severe MR compared to the other groups (LV posterior wall: P ¼ 0.0006, septum: P ¼ 0.0004, LV
lateral wall: P ¼ 0.0003). From both modelling and in our patients, deformation correlated inversely
with LV end-diastolic diameter and end-systolic diameter (ESD). Deformation measurements (corrected
for increased geometry) enabled the identiﬁcation of patients classically referred to as at risk of irreversible myocardial damage (ESD 4.5 cm).
Conclusion In patients with a wide range of MR, deformation remains unchanged due to a balance of
increased dimensions and increased stroke volume. Only when contractility is expected to change, deformation will signiﬁcantly decrease. SR imaging indices, corrected for geometry, might potentially be
useful in detecting subclinical deterioration in LV function in asymptomatic patients with severe MR.
2628
A. Marciniak et al.
isolated MR and relate these to changes in geometry and
stroke volume in order to assess its potential for detecting
subclinical changes in systolic function. For this, we combined theoretical mathematical modelling with measurements in patients with various degrees of regurgitation.
Methods
Study population
Standard echocardiography
All echocardiographic studies were performed using a Vivid 7 ultrasound scanner (General Electric—GE Vingmed). The images were
acquired from standard parasternal and apical views. Standard LV
M-mode measurements included the estimation of LVEDD, LVESD,
and lateral and septal atrioventricular plane displacement and
velocities. EF, EDV, ESV, and SV were measured using the biplane
Simpson’s method. LV ﬁlling was assessed by measuring inﬂow at
the tips of the leaﬂets of the mitral valve using pulsed Doppler.
The following Doppler-derived parameters were measured: early
diastolic peak ﬂow velocity (E), late diastolic velocity (A), and
deceleration time of early ﬁlling (E-dec).
Colour Doppler myocardial velocity imaging—data
acquisition
For each patient, parasternal long axis and apical 4 chamber views
were acquired. For longitudinal deformation, real time twodimensional MVI data were recorded from the septum and lateral
wall. For radial deformation data were recorded from the LV posterior wall (LVPW).
A frame rate of 200–300 frames per second was used to acquire the
data. An image sector angle of 158 and an optimal depth of imaging
Table 1 Clinical and standard echocardiographic parameters
Age (years)
Male (%)
SBP (mmHg)
DBP (mmHg)
HR (b.p.m.)
LV EDD (cm)
LV ESD (cm)
IVS (cm)
LVPW (cm)
LV mass (g)
EDV (mL)
ESV (mL)
FS (%)
EF (%)
SVMV (mL)
SVLVOT (mL)
RV (mL)
Control
n ¼ 23
Mild
n ¼ 10
Moderate
n ¼ 14
Severe
n ¼ 30
Correlation w. RV—r,
P-value
Severe vs. other,
P-value
50 + 12
27
128 + 16
76 + 10
67 + 9
4.6 + 0.4
2.9 + 0.4
0.9 + 0.2
0.8 + 0.2
140 + 50
107 + 23
35 + 11
36 + 4
68 + 7
–
–
–
54 + 12
50
131 + 5
70 + 6
65 + 9
4.6 + 0.5
3.0 + 0.4
0.9 + 0.2
0.9 + 0.2
152 + 50
111 + 28
36 + 12
35 + 4
67 + 6
57 + 9
40 + 9
20 + 7
59 + 12
40
136 + 11
76 + 7
71 + 11
5.3 + 0.7
3.3 + 0.5
1.0 + 0.2
1.0 + 0.2
247 + 83
140 + 39
45 + 15
36 + 9
65 + 9
83 + 17
36 + 9
47 + 11
55 + 11
50
138 + 17
78 + 13
76 + 15
6.2 + 0.7
4.0 + 0.6
1.1 + 0.2
1.1 + 0.2
350 + 123
197 + 61
72 + 24
37 + 7
61 + 12
122 + 30
34 + 9
88 + 24
0.19, P ¼ 0.1
–
0.28, P ¼ 0.01
0.14, P ¼ 0.2
0.39, P ¼ 0.0005
0.73, P , 0.0001
0.71, P , 0.0001
0.46, P , 0.0001
0.51, P , 0.0001
0.64, P , 0.0001
0.65, P , 0.0001
0.70, P , 0.0001
0.06, P ¼ 0.6
20.40, P ¼ 0.0004
–
–
–
P ¼ 0.7
–
P ¼ 0.06
P ¼ 0.2
P ¼ 0.008
P , 0.0001
P , 0.0001
P , 0.0001
P , 0.0001
P , 0.0001
P , 0.0001
P , 0.0001
P ¼ 0.4
P ¼ 0.03
–
–
–
Values are mean + SD; r, the Pearson correlation coefﬁcient; MR, mitral regurgitation; SBP, systolic blood pressure; DBP, diastolic blood pressure; HR,
heart rate; LV, left ventricle; EDD, end-diastolic diameter; ESD, end-systolic diameter; IVS, interventricular septum thickness; PWT, posterior wall thickness;
EF, ejection fraction; EDV, end-diastolic volume; ESV, end-systolic volume; FS, fraction shortening; SVMV, mitral valve stroke volume; SVLVOT, left ventricular
outﬂow tract stroke volume; RV, regurgitant volume.
Downloaded from by guest on October 21, 2014
The study population consisted of 77 individuals: 54 patients with
isolated non-ischaemic MR and 23 healthy aged matched subjects
examined in the Department of Cardiology, St George’s Hospital,
London. The patients with MR were a series of consecutive patients
referred to the Department of Echocardiography with a provisional
diagnosis of isolated MR between October 2004 and September
2005. All patients were asymptomatic or minimally symptomatic
and the echocardiographic examination was requested on the
basis of the presence of a pan-systolic murmur.
Patients were excluded if they had MR due to ischaemic heart
disease or cardiomyopathy, had associated mitral stenosis, or any
other form of valve disease, which was more than trivial, atrial
ﬁbrillation, bundle branch block, or a history of previous cardiac
surgery. In addition, during the same inclusion period, healthy controls were recruited from the local population via advertisements
using posters and announcements in the local media. From the 35
controls initially assessed in this way, all subjects (n ¼ 23) with an
age comprised within +1.5 standard deviations of the mean
patient age were included for further data analysis. An informed
written consent was obtained from all subjects.
From the 82 patients with MR initially assessed for the inclusion
into the study, 28 patients were excluded due to coronary artery
disease based on the coronary angiogram. Fifty-four patients with
MR were included and in these, all echocardiographic parameters
could be assessed. None of these patients showed any echocardiographic evidence of ischaemic or structural heart disease. In 24/30
of the patients (81%) with severe MR, co-existing coronary artery
disease was excluded based on coronary angiograms within the
preceding 3 month period. Neither the control group nor the
patients with mild or moderate MR or the remaining six patients
(19%) with severe MR had a history of ischaemic heart disease,
nor did they have signiﬁcant risk factors. All had normal physical
examinations and no evidence of coronary artery disease on their
resting 12 lead ECGs and their echocardiographic examination.
The aetiology of the isolated moderate and severe MR was as
follows: mitral valve prolapse (32 patients), leaﬂet non-coaptation
(seven patients) and chordal rupture (ﬁve patients).
Patients were sub-divided into three groups based on their RVs:
mild MR (a RV ,30 mL; n ¼ 10), moderate MR (RV: 30–59 mL;
n ¼ 14), and severe MR (RV . 60 mL; n ¼ 30). Mitral RVs were quantiﬁed according to previously published guidelines.17,18 Mitral/aortic
stroke volumes were obtained by multiplying the area of the
mitral annulus/outﬂow tract with the velocity time integral of the
pulsed Doppler trace of the ﬂow through the respective valve.
Changes in systolic LV function in isolated MR
2629
Figure 1 Scatter plots of the end-diastolic diameter (EDD) and the end-systolic diameter (ESD) vs. regurgitant volume (RV) in all patients. The horizontal grey
line indicates the mean of the control group and grey shading shows 1.5 standard deviations around this mean. Grouping: †control group, Vmild mitral regurgitation, B moderate mitral regurgitation, O severe mitral regurgitation (ESD , 4.5 cm), 4 severe mitral regurgitation (ESD 4.5 cm).
Table 2 Doppler mitral inﬂow and ring displacement
Mild
n ¼ 10
Moderate
n ¼ 14
Severe
n ¼ 30
0.8 + 0.2
0.5 + 0.1
169 + 31
0.9 + 0.1
0.5 + 0.1
178 + 47
1.1 + 0.4
0.8 + 0.4
192 + 43
1.3 + 0.4
0.7 + 0.3
182 + 41
15 + 2
13 + 2
15 + 2
13 + 1
14 + 3
13 + 2
14 + 3
12 + 3
Correlation w. RV—r,
P-value
0.64, P , 0.0001
0.29, P ¼ 0.01
0.14, P ¼ 0.2
20.26, P ¼ 0.02
20.19, P ¼ 0.1
Severe vs. other,
P-value
P , 0.0001
P ¼ 0.2
P ¼ 0.9
P ¼ 0.2
P ¼ 0.2
Values are mean +SD; r, the Pearson correlation coefﬁcient; E-dec, deceleration time of early transmitral ﬁlling.
were used to increase temporal resolution. Special attention was paid
to the colour Doppler velocity range setting in order to avoid any
aliasing within the image. For this purpose and to simultaneously
optimize velocity resolution, pulsed repetition frequency (PRF)
values were set as low as possible, just avoiding aliasing.
Strain rate imaging study—data analysis
All data were analysed ofﬂine using a dedicated workstation (GE
Echopac). For the evaluation of longitudinal function, midventricular segment shortening was analysed for the septum and
LV lateral walls. For LV radial function, mid-ventricular segment
thickening of the LVPW was analysed. Peak systolic SR and S
during the ejection period were assessed for each segment analysed.16 Computational areas of 10 mm (longitudinal) and 5 mm
(radial) with a width of 1 mm (to avoid averaging different ultrasound beams) were used. Frame by frame manual tracking was performed to maintain the computational area within the myocardial
region of interest throughout the cardiac cycle. Values were averaged over three consecutive cycles. Aortic valve opening and
closure were deﬁned using pulsed wave blood pool Doppler tracings
acquired during the same examination and with a similar R–R
interval.
Modelling the relation between deformation and
ventricular geometry
In order to study the resulting changes in deformation (SR and S)
when ventricular size changes but intrinsic contractility of the
myocardium is not altered, we constructed a simpliﬁed model of
the LV and used ﬁnite difference techniques to solve the dynamics
during the cardiac cycle.
Despite the fact that this model has intrinsic limitations and
cannot be used to study all aspects of cardiac mechanics, it is
able to simulate qualitatively the regional deformation patterns of
the myocardium. The choice of a simple model enables us to solve
the equations of motion for the segments of the myocardium in a
straightforward manner using Matlab (The MathWorks, Inc.,
Natick, MA, USA). The details of this model have been described
previously.19,20 In summary, in this model a mid-ventricular
segment of the LV is described by a chain of elastic elements. The
thickness of the elements is deﬁned to be proportional to the reciprocal of the length of the elements (based on local mass conservation). The elements are described as Maxwell elements splitting
active and passive properties. It describes the element as a contractile component (CE) in series with an elastic component (SE),
together in parallel with another elastic component (PE) to take
into account the extracellular collagen matrix. The active developed contraction force of the segments is modelled by a Gaussian
shape, allowing variation of the amplitude of contraction. Again,
this is a simpliﬁcation, but should be sufﬁcient to study the relationship between deformation and geometry for a given active force
development. The intra-cavity pressure, based on a representative
measured pressure curve, is used as a boundary condition.
To simulate radial deformation during a complete cardiac cycle,
the cycle was divided into 800 time steps of 1 ms and the opening
and closure of the valves were used to sub-divide the cardiac
cycle into different time periods. The equations of motion were
Downloaded from by guest on October 21, 2014
Transmitral Doppler data
Peak E velocity (m/s)
Peak A velocity (m/s)
E-dec (ms)
Systolic ring Displacement
Mitral–lateral (mm)
Mitral–septal (mm)
Control
n ¼ 23
2630
A. Marciniak et al.
Figure 2 Scatter plots of peak systolic tissue velocity from the mitral lateral (left) and the mitral septal (right) ring in all patients vs. regurgitant volume (RV).
The horizontal grey line indicates the mean of the control group and grey shading shows 1.5 standard deviations around this mean. Grouping: cf. Figure 1.
Table 3 Radial and (absolute) longitudinal systolic deformation
Mild
n ¼ 10
Moderate
n ¼ 14
Severe
n ¼ 30
Correlation w. RV—r,
P-value
Severe vs. other,
P-value
2.9 + 0.6
51 + 10
3.1 + 0.5
51 + 11
2.8 + 0.6
53 + 9
2.2 + 0.9
40 + 14
20.48, P , 0.0001
20.39, P ¼ 0.0004
P ¼ 0.0006
P ¼ 0.0008
1.5 + 0.3
22 + 6
1.6 + 0.2
21 + 5
1.6 + 0.3
20 + 5
1.6 + 0.2
18 + 2
1.5 + 0.3
22 + 6
1.5 + 0.4
20 + 6
1.2 + 0.5
16 + 7
1.2 + 0.5
16 + 6
20.49, P ,
20.45, P ,
20.49, P ,
20.43, P ,
P ¼ 0.0004
P ¼ 0.002
P ¼ 0.0003
P ¼ 0.002
0.0001
0.0001
0.0001
0.0001
Values are mean +SD; r, the Pearson correlation coefﬁcient. SR, peak systolic strain rate; S, peak systolic strain; LV, left ventricle.
integrated at each time step. In order to study the dependency of
deformation on cavity size, a range of ventricular diameters, corresponding to the measurements made in our study population, were
simulated. For each of these, the amplitude of the active force
development was also changed in order to attempt to understand
the interaction of changes in contractility with changes in ventricular volume, as would occur in patients with MR if the myocardium
were irreversibly damaged by the underlying physiology.
Statistical analysis
Results are expressed as means + SD (standard deviation). Given the
exploratory nature of this study, no formal sample size was calculated and data was collected over a 1 year period. Statistical analysis was performed with Statistica (version 7.1, StatSoft Inc., Tulsa,
OK). Given the deﬁnition of the disease’s severity based on RV,
which is a continuous parameter, all relevant parameters were
ﬁrst correlated with RV. Secondly, to compare differences in parameters between the clinically relevant groups, a pre-planned contrast was setup to compare the severe MR group to the combination
of the other groups (ControlþMild MRþModerate MR). For correlations between variables, the Pearson correlation was calculated.
A two-tailed P-value of ,0.05 was considered statistically signiﬁcant.
17 + 11 mL for mild MR, 47 + 10 mL for moderate MR, and
88 + 28 mL for severe MR. There was a strong correlation
between EDD, EDV, and RV. Patients with severe MR had LV
diameters (EDD, ESD) and volumes (EDV, ESV) which were
signiﬁcantly higher while their EF was signiﬁcantly lower
compared to the other groups.
Scatter plots of the LV EDD and LV ESD vs. RV for all
patients are presented in Figure 1. Also LV mass increased
signiﬁcantly with degree of MR and was signiﬁcantly higher
in patients with severe MR compared to the other groups.
There was a relatively high correlation between early
peak diastolic velocity in mitral inﬂow and RV and this parameter was signiﬁcantly increased for patient with severe
MR compared to the other groups. However, there were no
other signiﬁcant differences in transmitral Doppler ﬂow velocities or mitral ring (lateral and septal) displacement
between patients with MR vs. Controls (Table 2).
Figure 2 shows a scatter plot of lateral (left) and septal
(right) ring velocities vs. RV.
Strain and strain rate imaging data
Results
The clinical and echocardiographic data of all patients are
presented in Table 1. RVs for the individual groups were
Each segmental data set acquired allowed the processing of
regional deformation traces which were interpretable. The
parasternal short axis was used to quantify regional radial
systolic function of the LVPW (77 segments were analysed).
Downloaded from by guest on October 21, 2014
Radial deformation
SR–LV posterior wall (1/s)
S–LV posterior wall (%)
Longitudinal deformation
SR–septum (1/s)
S–septum (%)
SR–LV lateral wall (1/s)
S–LV lateral wall (%)
Control
n ¼ 23
Changes in systolic LV function in isolated MR
2631
Downloaded from by guest on October 21, 2014
Figure 3 Scatter plots of radial (LV posterior wall–A) and longitudinal (LV lateral wall–B) peak systolic strain rate (SR) and strain (S) in all patients vs. regurgitant
volume (RV). For longitudinal deformation, absolute values are shown (SR, ABS; S, ABS). LV, left ventricle. The horizontal grey line indicates the mean of the
control group and grey shading shows 1.5 standard deviations around this mean. Grouping: cf. Figure 1.
Figure 4 Correlation between longitudinal peak systolic strain in the septum
(shown as absolute value: S, ABS) and ejection fraction (EF) in all patients.
r adjusted for regurgitant volume: 0.23 (P ¼ 0.052). Grouping: cf. Figure 1.
The apical four-chamber view was used to quantify longitudinal systolic deformation of the mid segments of
septum and LV lateral walls (154 segments were analysed).
Radial and longitudinal peak systolic SR values (Table 3)
were signiﬁcantly lower in the severe MR group compared
to the other groups. Similarly, radial and longitudinal peak
systolic S values (Table 3) were signiﬁcantly lower in the
severe MR group compared to the other groups. All deformation parameters showed a moderate correlation with
RV. Scatter plots of the peak systolic SR and S values for
radial and longitudinal deformation for all patients and Controls are shown in Figure 3. SR/S values show a wide spread
and although the values were signiﬁcantly lower for severe
MR, there still is an important overlap when compared to
the other groups.
As used in clinical practice, a cut off value of ESD4.5 cm
was taken to identify patients who are at severe risk for irreversible myocardial damage. These patients were shown as a
separate group. These individuals showed an even greater
reduction in deformation compared to the control group
with less overlap in the scatterplot (Figure 3).
Figure 4 shows the relationship of deformation and EF.
Although, when including all individuals, there is a correlation between deformation and EF, this is mainly inﬂuenced
by patients with severe MR. Mild and moderate showed
almost no correlation. Although there is a tendency
2632
A. Marciniak et al.
Figure 5 A theoretical model of strain rate and strain as a function of ventricular diameter and its relationship with stroke volume and contractility. See online
supplementary material for a colour version of this ﬁgure.
Downloaded from by guest on October 21, 2014
Figure 6 Correlation between radial peak systolic strain rate (SR) and strain (S) and left ventricular diameters in all patients. EDD, end-diastolic diameter; ESD,
end-systolic diameter. Grouping: cf. Figure 1.
towards lower deformation for lower EF, several of the
patients with severe MR and low deformation have an
EF.60%, while some individuals with EF,60% show normal
deformation.
Figure 5 shows the results of the mathematical modelling
study used to determine the dependency of deformation on
ventricular dimensions and contractility. Figure 5 (left)
shows the changes in SV that occur with changing EDD,
with lines of constant contractility. When ventricular function is not impaired, an increase in SV, to compensate for
the increased regurgitation, is achieved by ventricular dilatation. If ventricular function becomes increasingly
impaired, SV can only be maintained by a further increase
in EDD. There is a clear inverse dependency of both SR
and S on EDD for a certain SV. An increase in LV size with
no change in SV would lead to a decrease in deformation.
Increased regurgitation, resulting in increased SV, without
a change in ventricular dimensions, will lead to increased
deformation.
This modelling predicted inter-relationship of deformation
and geometry was conﬁrmed in the measurements in our
clinical study. Radial and longitudinal peak systolic SR and S
were inversely correlated with EDD as well as with ESD for
both normals and patients with MR (Figures 6 and 7).
From these ﬁgures it can be seen that the relationship
between SR/S and EDD, ESD begins to deviate for patients
Changes in systolic LV function in isolated MR
2633
who were clinically considered as having decreased myocardial contractility. This is in agreement with the predictions
of the modelling study where a decrease in contractility
resulted in even more decrease in deformation.
Figure 8 shows the geometry compensated deformation
indices (calculated by dividing deformation by diameter:
SR/EDV and S/EDV). These two parameters were signiﬁcantly
reduced in patients with severe MR. This change was most
marked in the severe MR group with an ESD4.5 cm.
Reproducibility
The approach for the calculation of strain and strain-rate
was similar to the study of Pena et al.21 For longitudinal
deformation, there was ,3% intra-observer variability and
,5% inter-observer variability. For radial deformation, this
was ,8% and ,16%, respectively.
Examples of radial SR and strain curves of a patient with
severe MR and a control subject are presented in Figure 9.
Discussion
In the compensated phase of chronic MR, with increasing RV,
SV has to increase to maintain forward cardiac output. This is
affected by an increase in LV end-diastolic volume brought
about by a combination of changes in geometry and spherical
dilatation along the LV short axis.22,23 In this situation, the
combination of an augmented preload and reduced or
normal afterload with maintained intrinsic contractility, preserve LV ejection. In this compensated phase, patients are
frequently asymptomatic.4 The duration of the compensated
phase of isolated MR can vary but may last for many years.
However, in chronic MR, progressive LV remodelling, with
increasing wall stresses due to dilatation, can ultimately
lead to irreversible changes in the myocardium, resulting
in the development of LV contractile dysfunction. In this
phase, the contractile dysfunction impairs ejection and
EDV will start to increase further in order to maintain the
required SV. Since this makes the local wall stress even
worse, it will hasten tissue damage, resulting in an effective
reduction in forward output, leading to a failing ventricle.4
According to the ACC/AHA guidelines, asymptomatic MR
patients with an ESD4.5 cm and an EF60% beneﬁt from
valve surgery to protect ventricular function.4,24,25
However, if ESD4.5 cm and EF60% is used as a cut-off
point, there still is a high incidence of heart failure26 and
poor survival27 after surgery. Thus, the measurement of
ESD and EF is less than optimal measures of subclinical ventricular dysfunction in asymptomatic patients. From our
study, we also found that a large proportion of patients
with severe MR and reduced deformation have an EF.60%,
indicating that EF might not be the most sensitive parameter
to evaluate these patients.
Downloaded from by guest on October 21, 2014
Figure 7 Correlation between [the absolute values (ABS) of] longitudinal peak systolic strain rate (SR) and strain (S) in the left-ventricular lateral wall and the
left-ventricular diameters in all patients. EDD, end-diastolic diameter; ESD, end-systolic diameter. Grouping: cf. Figure 1.
2634
A. Marciniak et al.
Downloaded from by guest on October 21, 2014
Figure 8 Scatter plots of the peak systolic SR/EDV and the peak systolic S/EDV indices for radial (A) and longitudinal (B) deformation vs. regurgitant volume
(RV). For longitudinal deformation, absolute values are shown (SR, ABS; S, ABS). LV, left ventricle; SR, strain rate; S, strain; EDV, end-diastolic volume; ESV, endsystolic volume. The horizontal grey line indicates the mean of the control group and grey shading shows 1.5 standard deviations around this mean. Grouping: cf.
Figure 1.
Myocardial deformation (S), and the speed at which this
deformation takes place (SR) reﬂect systolic function and
are the result of the intrinsic contractility (force development) of the myocardium acting to overcome loading conditions and thus ejecting the required blood volume into
the circulation. Previous studies have shown that deformation (strain) increases with increasing SV (if geometry is
not altered)15 and that deformation rate (strain-rate) parallels change in contractility. In the mathematical simulations
described in this paper, we show that deformation also
depends on ventricular size. An increase in size, without
any change in SV or contractility will lead to a decrease in
regional deformation. This ﬁnding is of importance when
interpreting deformation values in patient groups where
geometry changes during follow up. Taking this knowledge
into account, changes in deformation can be used to
detect changes in systolic function.
Our clinical ﬁndings, combining data from all subjects,
conﬁrms the predictions from mathematical modelling that
radial and longitudinal SR and S decrease as EDD increases,
thus conﬁrming that regional myocardial deformation
indices change with the changing geometry of the heart
and that a decrease in deformation indices does not necessarily mean that intrinsic myocardial function is decreased.
The initial LV response to increased SV is to increase both
contractility and LV diameter, thus increasing SV at the cost
of an increase in wall stress. However, such a chronic
increase in wall stress will result in myocardial damage
and a reduction in contractility, associated with a further
fall in peak systolic deformation indices when LV contractile
function is no longer preserved.
This suggests that correcting deformation parameters for
changes in geometry could be a sensitive way of detecting
early changes in contractile function. The value of such a
correction was clearly evident in our patients, in whom, as
ESD increased in severe MR, (which is traditionally associated with a reduction in contractility), SR and S values
decreased further. This was more than would be expected
when only taking the EDD into account. In all patients with
an ESD4.5 cm, regional deformation (rate) values were
signiﬁcantly reduced but importantly it should be noted
that the decrease in SR and S values, even more than predicted by the geometry, occurred before the ESD reached
4.5 cm.
Changes in systolic LV function in isolated MR
2635
In summary in patients with MR, deformation is not directly
related to the degree of MR itself. Deformation is determined
by the size, the SV, and the contractility of the muscle. The
increase in deformation with increasing SV is compensated
by the decrease due to the bigger size. Only an additional
decrease in contractility, as will be observed in some patients
with severe MR, will decrease the deformation.
The ﬁndings of our study, using a combination of modelling and measurements in patients with a wide range of MR,
provides an initial impression that SR and S imaging (corrected for geometry), might potentially be used as a risk
stratiﬁcation tool in diagnosing changes in LV dysfunction
at a sub-clinical level in patients with severe asymptomatic
MR. This has to be conﬁrmed in further outcome studies.
Conclusions
Our results show that myocardial deformation changes with
changing geometry of the ventricle. When the observed
changes deviate from the predicted changes then myocardial contractility is reduced. The clinical study conﬁrmed
the changes induced by isolated MR which were predicted
by the modelling study. Thus, SR imaging may be a sensitive
tool in detecting subclinical changes in LV function in asymptomatic patients with severe MR.
Limitations
The major limitation in this study is that we were not able to
measure a parameter reﬂecting true contractility in our
patients. The gold standard for assessing true contractility
is the measurement of end-systolic elastance using
pressure–volume measurement. This requires the introduction of a conductance catheter in the LV of the patients.
For ethical reasons, this is not possible in patients who are
not undergoing cardiac catheterisation or surgery.
Echocardiography is known to be not the most accurate
method for the quantiﬁcation of the degree of MR. We
might have over- or underestimated the severity of MR.
However, from our ﬁndings, strain seems to be not mainly
determined by the degree of MR itself since we do not ﬁnd
a direct relationship between deformation and the grade
of MR over a very wide range. For this reason, potential
over- or underestimation would not change the conclusions
of this paper.
We presumed that none of these patients had segmental
dysfunction due to coronary artery disease. The majority
of patients with severe MR had a coronary angiogram
which excluded co-existing coronary artery disease but this
procedure was not performed in all patients. However,
none of these patients had any clinical features of angina
Downloaded from by guest on October 21, 2014
Figure 9 An example of radial strain rate and strain curves of a patient with severe mitral regurgitation and a control subject. SR, strain rate; S, strain;
MR, mitral regurgitation. See online supplementary material for a colour version of this ﬁgure.
2636
and their physical examination and ECG did not show any
evidence of coronary artery disease. The standard echocardiographic images also showed no signs of regional dysfunction due to ischaemia.
The image artefacts such, as reverberations, can degrade
the calculation of the regional deformation. Signal noise
could be a potential problem in this group of patients as
this is ampliﬁed during the SR calculation. To minimize this
problem, all data were acquired at high frame rate and
with a narrow sector angle with the wall placed in the
centre of the image. Finally, 231 segments (154 segments
describing longitudinal deformation and 77 segments radial
deformation) were obtained and analysed and none were
excluded from the study on the basis of uninterpretable
curves.
A. Marciniak et al.
10.
11.
12.
13.
14.
15.
Supplementary material
Supplementary material is available at European Heart
Journal online.
16.
Acknowledgement
17.
This work was supported by British Heart Foundation. Ref. PG/06/
009/20255.
18.
References
1. Singh JP, Evans JC, Levy D, Larson MG, Freed LA, Fuller DL, Lehman B,
Benjamin EJ. Prevalence and clinical determinants of mitral, tricuspid,
and aortic regurgitation (the Framingham Heart Study). Am J Cardiol
1999;83:897–902.
2. Ling LH, Enriquez-Sarano M, Seward JB, Tajik AJ, Schaff HV, Bailey KR,
Frye RL. Clinical outcome of mitral regurgitation due to ﬂail leaﬂet.
N Engl J Med 1996;335:1417–1423.
3. Avierinos JF, Gersh BJ, Melton LJ 3rd, Bailey KR, Shub C, Nishimura RA,
Tajik AJ, Enriquez-Sarano M. Natural history of asymptomatic mitral
valve prolapse in the community. Circulation 2002;106:1355–1361.
4. Bonow RO, Carabello B, de Leon AC, Edmunds LH Jr, Fedderly BJ,
Freed MD, Gaasch WH, McKay CR, Nishimura RA, O’Gara PT,
O’Rourke RA, Rahimtoola SH, Ritchie JL, Cheitlin MD, Eagle KA,
Gardner TJ, Garson A Jr, Gibbons RJ, Russell RO, Ryan TJ, Smith SC Jr.
ACC/AHA Guidelines for the Management of Patients With Valvular
Heart Disease. Executive Summary. A report of the American College
of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients With Valvular Heart
Disease). Circulation 1998;98:1949–1984.
5. Bruch C, Stypmann J, Gradaus R, Breithardt G, Wichter T. Impact of
stroke volume on mitral annular velocities derived from tissue Doppler
imaging. Heart 2005;91:243–244.
6. Haluska BA, Short L, Marwick T. Relationship of ventricular longitudinal
function to contractile reserve in patients with mitral regurgitation.
Am Heart J 2003;146:183–188.
7. Agricola E, Galderisi M, Oppizzi M, Schinkel AF, Maisano F, De Bonis M,
Margonato A, Maseri A, Alﬁeri O. Pulsed tissue Doppler imaging detects
early myocardial dysfunction in asymptomatic patients with severe
mitral regurgitation. Heart 2004;90:406–410.
8. Tsutsui H, Uematsu M, Shimizu H, Yamagishi M, Tanaka N, Matsuda H,
Miyatake K. Comparative usefulness of myocardial velocity gradient in
detecting ischemic myocardium by a dobutamine challenge. J Am Coll
Cardiol 1998;31:89–93.
9. Hiro T, Katayama K, Miura T, Kohno M, Fujii T, Hiro J, Matsuzaki M. Stroke
volume generation of the left ventricle and its relation to chamber shape
19.
20.
21.
22.
23.
24.
25.
26.
27.
Downloaded from by guest on October 21, 2014
Conﬂict of interest: none declared.
in normal subjects and patients with mitral or aortic regurgitation. Jpn
Circ J 1996;60:216–227.
Osbakken MD, Bove AA, Spann JF. Left ventricular regional wall motion
and velocity of shortening in chronic mitral and aortic regurgitation.
Am J Cardiol 1981;47:1005–1009.
Ohi H, Uchida M, Sato H, Yamaguchi H, Kawai S, Okada R. Differences in
left ventricular shape between aortic and mitral regurgitation: an echocardiographic study. J Cardiol 1989;19:823–830.
Badke FR, Covell JW. Early changes in left ventricular regional dimensions and function during chronic volume overloading in the conscious
dog. Circ Res 1979;45:420–428.
Vokonas PS, Gorlin R, Cohn PF, Herman MV, Sonnenblick EH. Dynamic geometry of the left ventricle in mitral regurgitation. Circulation 1973;48:
786–796.
Weidemann F, Kowalski M, D’hooge J, Bijnens B, Sutherland GR. Doppler
myocardial imaging. A new tool to assess regional inhomogeneity in
cardiac function. Basic Res Cardiol 2001;96:595–605.
Weidemann F, Jamal F, Sutherland GR, Claus P, Kowalski M, Hatle L,
De Scheerder I, Bijnens B, Rademakers FE. Myocardial function deﬁned
by strain rate and strain during alterations in inotropic states and heart
rate. Am J Physiol Heart Circ Physiol 2002;283:H792–H799.
D’hooge J, Heimdal A, Jamal F, Kukulski T, Bijnens B, Rademakers F,
Hatle L, Suetens P, Sutherland GR. Regional strain and strain rate
measurements by cardiac ultrasound; principles, implementation and
limitations. Eur J Echocardiogr 2000;1:154–170.
Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD,
Levine RA, Nihoyannopoulos P, Otto CM, Quinones MA, Rakowski H,
Stewart WJ, Waggoner A, Weissman NJ. American Society of Echocardiography. Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography.
J Am Soc Echocardiogr 2003;16:777–802.
Dujardin KS, Enriquez-Sarano M, Bailey KR, Nishimura RA, Seward JB,
Tajik AJ. Grading of mitral regurgitation by quantitative Doppler echocardiography: calibration by left ventricular angiography in routine clinical
practice. Circulation 1997;96:3409–3415.
Claus P, Bijnens B, Weidemann F, Dommke C, Bito V, Heinzel F, Sipido KR,
De Scheerder I, Rademakers FE, Sutherland GR. Post-systolic thickening
in ischaemic myocardium: a simple mathematical model for simulating
regional deformation. Functional imaging and modeling of the heart.
In: Katila T, Magnin IE, Clarysse P, Montagnat J, Nenonen J eds, Lecture
Notes in Computer Science 2230. 2001. p134–139.
Claus P, Bijnens B, Breithardt O, Herbots L, Sutherland GR. Why Ischemic
Hearts Respond Less to Cardiac Resynchronisation Therapy. A Modeling
Study Functional Imaging and Modeling of the Heart. In: Magnin IE,
Montagnat J, Clarysse P, Nenonen J, Katila T eds, Lecture Notes in Computer Science 2674. 2003. p287–294.
Pena JLB, Silva MG, Faria SCC, Salemi VMC, Mady C, Sutherland GR.
Sequential changes in longitudinal and radial myocardial deformation in
the normal neonate heart. (Abstracts supplement). Eur J Echocardiogr
2005;6:52.
Vokonas PS, Gorlin R, Cohn PF, Herman MV, Sonnenblick EH. Dynamic geometry of the left ventricle in mitral regurgitation. Circulation 1973;48:
786–796.
Hiro T, Katayama K, Miura T, Kohno M, Fujii T, Hiro J, Matsuzaki M. Stroke
volume generation of the left ventricle and its relation to chamber shape
in normal subjects and patients with mitral or aortic regurgitation. Jpn
Circ J 1996;60:216–227.
Wisenbaugh T, Skudicky D, Sareli P. Prediction of outcome after valve
replacement for rheumatic mitral regurgitation in the era of chordal
preservation. Circulation 1994;89:191–197.
Enriquez-Sarano M. Timing of mitral valve surgery. Heart 2002;87:
79–85.
Enriquez-Sarano M, Schaff HV, Orszulak TA, Bailey KR, Tajik AJ, Frye RL.
Congestive heart failure after surgical correction of mitral regurgitation.
A long-term study. Circulation 1995;92:2496–2503.
Enriquez-Sarano M, Tajik AJ, Schaff HV, Orszulak TA, Bailey KR, Frye RL.
Echocardiographic prediction of survival after surgical correction of
organic mitral regurgitation. Circulation 1994;90:830–837.