European Heart Journal - Cardiovascular Imaging (2018) 19, 1372–1379

doi:10.1093/ehjci/jey024

Estimation of myocardial work from

pressure–strain loops analysis: an
experimental evaluation
Arnaud Hubert1,2,3, Virginie Le Rolle2,3, Christophe Leclercq1,2,3, Elena Galli1,2,3,

Downloaded from https://academic.oup.com/ehjcimaging/article/19/12/1372/4911133 by guest on 15 June 2022

Eigil Samset4, Cyrille Casset5, Philippe Mabo1,2,3, Alfredo Hernandez2,3 and
Erwan Donal1,2,3*
1
CHU Rennes, Service de Cardiologie et Maladies Vasculaires et CIC-IT 1414, 2 Rue Henri Le Guilloux, CHU Pontchaillou, 35000 Rennes, France; 2Université de Rennes 1, LTSI,
campus Beaulieu, 35000 Rennes, France; 3INSERM, U1099, campus Beaulieu, 35000 Rennes, France; 4Institute for Surgical Research; Center for Cardiological Innovation;
Department of Cardiology, Oslo University Hospital, Rikshospitalet, Oslo, Norway; Institute of Clinical Medicine, University of Oslo, Oslo, Norway; and 5Abbott, 15900 Valley
view court, Sylmar, 91342-3577 CA, USA

Received 16 December 2017; editorial decision 27 January 2018; accepted 2 February 2018; online publish-ahead-of-print 26 February 2018

Purpose The area of left ventricular (LV) pressure–strain loop (PSL) is used as an index of regional myocardial work. The
purpose of the present work is to compare the main segmental PSL markers and the derived global work indices,
when they are calculated using an estimated pressure signal or an observed pressure signal.
...................................................................................................................................................................................................
Methods In nine patients implanted with a bi-ventricular pace-maker (CRT), LV pressure was invasively measured in five conditions:
and results CRT-off, LV-pacing, right ventricular-pacing and two different CRT-pacing. For each condition, systolic blood pressure
was measured by brachial artery cuff-pressure and transthoracic echocardiography loops were recorded simultaneously.
The error and relative root mean square error (rRMSE) between measured and estimated pressure were calculated for
each patient and each configuration. Correlation coefficient (R2) and Bland–Altman (BA) analysis were performed for PSL
area and work indices. A total of 43 different haemodynamic conditions were compared (774 segmental PSL). The global
rRMSE between estimated and measured LV-pressure was 12.3 mmHg. The estimated and measured segmental LV-PSL
were strongly correlated, with an R2 of 0.98. BA analysis shows that the mean bias for the estimation of segmental LV-PSL
area is 86.0 mmHg.%. A significant bias effect with linearly increasing error with pressure values is observed. R2 >_ 0.88 and
a mean bias in BA analysis <_41.4 mmHg.% was observed for the estimation of global myocardial work indices.
...................................................................................................................................................................................................
Conclusion The non-invasive estimation for LV pressure–strain loop area and the global myocardial work indices obtained
from LV-PSL strongly correlates with invasive measurements.
䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏

Keywords myocardial work • strain • invasive haemodynamic

..
Introduction .. analysis of LV-PSL. They interestingly demonstrate a strong correla-
.. tion of LV-PSL area with cardiac metabolism, assessed by fluorine 18
..
The evaluation of left ventricular (LV) function is challenging in car- .. fluorodeoxyglucose-positron emission tomography. Preliminary
.. results about LV function in ischaemic cardiomyopathy4 and in car-
diac imaging, particularly in echocardiography.1 This difficulty is mainly ..
due to the lack of consideration for load conditions. Russell et al.,2–5 .. diac resynchronization therapy6 (CRT) are encouraging. Two distinct
..
developed a new tool, based on LV pressure–strain loops (LV-PSL), .. methods were used to calculate cardiac work: pressure–strain loop
to assess myocardial work, which explore LV function, balanced by .. area5 and segmental work.7 These two methods have same units
..
afterload with an estimation of intraventricular pressure during a car- .. (mmHg.%) and provide similar results because they both reflect a
diac cycle. The originality of their study was to create a non-invasive .. surrogate estimation of the power over the cardiac cycle, when the
..
estimation of LV pressure during a cardiac cycle allowing for the . force cannot be measured. Indices of global and segmental cardiac

* Corresponding author. Tel: þ33 2 99 28 25 07, Fax: þ33 2 99 28 25 29, Email: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. V
C The Author(s) 2018. For permissions, please email: [email protected].
Pressure–strain loop clinical experimental validation 1373

Downloaded from https://academic.oup.com/ehjcimaging/article/19/12/1372/4911133 by guest on 15 June 2022

Figure 1 Schema of the global method.

work, calculated from LV-PSL, may thus provide a new opportunity

.. (AF), significant aortic stenosis or aortic prosthesis or absence of
..
to better predict CRT-response.8 However, the accuracy of this new .. femoral and radial artery access were excluded. Patient echogenicity
..
tool, in different haemodynamic conditions, has not been verified. In .. in supine position had to be satisfying. The study was approved by
this work, we analyse LV-PSL acquired from patients implanted with .. local ethics committee (validation number: 35RC14-9767). Patients
..
a CRT system, while studying different CRT stimulation configura- .. were included from January 2016 to October 2017 with nine patients
tions, in order to induce reproducible and significant modifications of
.. (two excluded for an absence of invasive pressure, one excluded due
..
their cardiac electro-mechanical activity. .. to significant artifacts on the invasive pressure signal) finally analysed
The purpose of the present work is to compare the main segmen-
.. (61 ± 5 years, four males) for a total of 43 different haemodynamic
..
tal PSL markers and the derived global work indices, when they are .. profiles [two patients had only four CRT configurations with satisfy-
.. ing transthoracic echocardiography (TTE) loops], 774 segmental PSL
calculated using an estimated pressure signal or an observed pressure ..
signal. .. analysed.
..
..
.. Invasive pressure measurement
..
Methods .. The left heart catheterization (LHC) was performed via a retrograde
.. access from the femoral artery with a PressureWire CertusV cathe-
R

The first step was to obtain synchronous representations of the .. ter (Abbott, St Paul, Mi, USA) which permit pressure measures from
..
measured and estimated LV pressure curves (Figure 1). .. -30 to þ300 mmHg with an error ±1 mmHg. It was linked with a
.. R
.. RadiAnalyzer XpressV (Abbott, St Paul, MI, USA). Pressure waves
Data acquisition and calculation .. were recorded on a dedicated workstation for offline analysis.
Population and CRT configuration
..
.. Pressures data were saved in an Excel file. During a procedure, five
In this work, we included 12 patients >_18 years old recently .. different CRT configurations were tested, as previously described.
(<1 month) implanted with a CRT in accordance to current interna-
..
.. For each configuration, pressure data were acquired, at the same
tional recommendations. In order to induce different haemodynamic .. time as TTE.
..
conditions, five CRT configurations were tested for each patient: ..
CRT off, right ventricular (RV) pacing alone, LV pacing alone, and two .. Calculation of estimated LV pressure curve
..
different biventricular (BIV) pacing: ‘standard BIV pacing’ (BIV1) and .. A complete TTE, with a frame rate > 60frames/s, was acquired with
‘BIV with multipoint pacing’ (BIV2). Patients with atrial fibrillation
.. simultaneous invasive pressure recording. As previously described by
1374 A. Hubert et al.

Downloaded from https://academic.oup.com/ehjcimaging/article/19/12/1372/4911133 by guest on 15 June 2022

Figure 3 Myocardial work curves and method of calculation.
Figure 2 Example of a comparison of pressure curves (upper Positive (Pi) and negative (Nj) work are marked respectively as grey
panel) and the corresponding observed strain trace (lower panel). and black. Subscripts i (respectively j) corresponds to the ascending
(respectively descending) parts of the curve.

Russel et al., the analysis of aortic and mitral valve events during a
loop [mitral valve closure (MVC), aortic valve opening (AVO), aortic
..
.. synchronized to the maximum value of systolic pressure (Figure 2).
valve closure (AVC), mitral valve opening (MVO)] was performed in .. Signals were oversampled to 1 kHz, using spline interpolation, in
..
an APLAX view. As suggested by Russel et al.5,7,9; valvular timings .. order to: (i) harmonize sampling frequencies and (ii) facilitate all proc-
obtained from TTE are used to estimate a normalized, patient specific .. essing steps (differentiation, comparison with valve timing, . . .) by
..
LV pressure curve. The instantaneous systolic pressure value esti- .. manipulating a time vector expressed in millisecond.
mated by a brachial artery cuff was then used to scale the normalized ..
..
pressure signal. This analysis was realized using an echocardiography .. Calculation of segmental work
work station (EchoPAC version 202, General Electric Healthcare,
.. Segmental myocardial work was calculated from strain and from
..
Horton, Norway). Data of the curve were exported to an ExcelV file
R
.. measured and estimated LV pressure, as proposed by Russell et al.7
.. The instantaneous power was first obtained by multiplying the strain
(Office Suite, Microsoft Corp., Redmond, WA, USA). ..
.. rate, obtained by differentiating the strain curve, and the instantane-
.. ous LV pressure. Then, segmental myocardial work was calculated by
2D-speckle tracking echocardiography ..
TTE recordings were processed using an acoustic-tracking dedicated .. integrating the power over time, during the cardiac cycle from mitral
.. valve closure until mitral valve opening (Figure 3).
software (EchoPAC) which allowed for an off-line analysis of speckle- ..
based strain. To calculate the LV global and segmental longitudinal
.. Myocardial work was then used to evaluate positive (Wp) and neg-
..
strain (GLS), a line was traced along the LV endocardium’s inner bor- .. ative (Wn) segmental work. Positive and negative works were first
der in each of the three apical views (in the same loop than estimated
.. determined as the ascending and descending parts of the curves.
..
curve) on an end-systolic frame, and a region of interest was auto- .. Then, positive segmental work Wp (respectively Wn) is finally defined
.. as the sum of positive (respectively negative) variations for each
matically defined between the endocardial and epicardial borders ..
with GLS then automatically calculated from the mid-wall strain in .. segment:
.. X X
three apical views. All strain data (mean values, curves) were .. Wp = Pi ; W n = Nj ;
R
exported in ExcelV file. ..
.. i j
..
Data analysis .. where Pi (respectively Nj) is the variation associated with each
.. ascending (respectively descending) parts i (respectively j) of the seg-
All data was processed using custom-made functions created with ..
R
OctaveV (GNU Octave, version 4.2.1). Both measured and estimated
.. mental work (Figure 3). The indices i (respectively j) are comprised
.. between 1 and the total number of ascending (respectively descend-
signals were fully analysed with Octave, in order to calculate segmen- ..
.. ing) parts. Finally, global (GW), positive (GWp), and negative (GWn)
tal works and pressure–strain loops area: .. work indices are defined as mean values over all segments:
Only loops in sinus rhythm in TTE were analysed. ..
..
Concerning invasive pressure, data were saved in about 10 succes- .. 1X N
1X N
GWn
sive cycles. Only cycles with correlation >0.99 were selected in order .. GWp = W p;s ; GWn = W n;s ; GW= ;
.. N s=1 N s=1 GWp
to avoid the effect of premature beats and the mean pressure from ..
all selected cycles was calculated. Both the estimated and the meas- .. where N is the total number of segments. Positive work represents
..
ured pressure signals were extracted at a sampling rate of 250 Hz .. segmental shortening during the systole, i.e. effective energy for
and were then resampled at 1000 Hz. Both signals were then
.. blood ejection. Negative work represents segmental lengthening
Pressure–strain loop clinical experimental validation 1375

Table 1 Comparison of estimated vs. measured left ventricular pressure curves

Total cycle 0_MVC MVC_AVO AVO_AVC AVC_MVO MVO_end

....................................................................................................................................................................................................................
Error (%) 41.9 ± 21.1 65.1 ± 27.3 28.8 ± 20.7 5.0 ± 2.7 48.6 ± 16.2 69.9 ± 44.6
rRMSE (mmHg) 12.3 ± 4.5 12.9 ± 5.5 12.4 ± 8.5 6.5 ± 3.5 18.2 ± 9.5 9.2 ± 3.3

0, beginning of the cycle; MVC, mitral valve closure; AVO, aortic valve opening; AVC, aortic valve closure; MVO, mitral valve opening; end, end of the cycle; rRMSE, relative
root mean square error.

..
during the systole, i.e. energy loss for blood ejection. Figure 3 shows .. variables in y-axis. The mean difference with his double of standard
positive (grey) and negative (black) parts of segmental works during .. deviation is used as limits to assess visually the good agreement.
..

Downloaded from https://academic.oup.com/ehjcimaging/article/19/12/1372/4911133 by guest on 15 June 2022

systole for a septal wall segment before CRT in a representative ..
patient. .. Statistical analysis
..
.. Continuous data are presented as the mean ± standard deviation,
Pressure–strain loop area
.. and categorical data are presented as a number.
..
Pressure–strain loops were determined for each segmental strain ..
..
using both measured and estimated pressures. The area of each loop ..
was calculated using rectangle method, considering small increments .. Results
..
of ventricular pressure in order to insure a precise determination. ..
The same method was used for the calculation of the loop area for .. Comparison of pressure waveforms
.. The error and the RMSE for invasive and measured LV pressure on a
measured and estimated pressures. ..
.. whole cycle were equal to 41.9 ± 21.1% and 12.3 ± 4.5 mmHg.
.. Maximal error was found between AVC and MVC because the refer-
Evaluation of differences between ..
.. ence measured pressure is low, while the RMSE remains stable
estimated and measured data .. between 9.2 ± 3.3 mmHg, and 18.2 ± 9.5 mmHg. As illustrated in
Comparison of pressure waveforms ..
.. Figure 2, minimal error was evaluated around the systolic peak,
To compare pressure curves, the error is first calculated as the mean ..
difference between measured and estimated pressures expressed as
.. between AVO and AVC, with an error of 5.0 ± 2.7% for a RMSE of
.. 6.5 ± 3.5 mmHg (Figure 2, Table 1).
a percentage of the measured pressure: ..
..
N   .. Comparison of segmental
1X 
Pmeasured ðkÞ  Pestimated ðkÞ; ..
E% =100    .. pressure–strain loops area
N k=1 Pmeasured ðkÞ
..
.. The global intraclass coefficient (ICC) was equal to 0.975
where Pmeasured(k) and Pestimated(k) are respectively the measured and .. (P < 0.0001). On BA analysis, mean bias of segmental PSL area was
estimated pressures at the sample k. Root mean square error
..
.. 86 mmHg.% (-42.9 to 214.8 mmHg.%) (Figure 2). Of note, bias
(RMSE) is then evaluated as: .. increased with the size of PSL. In fact, majority of plot outside the
..
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi .. interval -2DS to þ2DS are for LV-PSL greater than 1000. Visually
u N
u1 X .. proportion of plots outside vs. inside increased with the size of PSL
RMSE=t ðPmeasured ðkÞ  Pestimated ðk ÞÞ2 : ..
N k=1 .. (Figure 4).
..
..
Both errors and RMSE were calculated for the total cardiac cycle .. Comparison of global cardiac work indices
and for intervals between each valve events. ..
.. For global work index, correlation between measures and estima-
.. tions was 0.982 (P < 0.0001). In BA analysis, mean bias was
Comparison of myocardial work and pressure–strain loop
..
.. 25.1 mmHg.% (-34.6 to 84.8 mmHg.%). For global positive work, cor-
area .. relation between measures and estimations was 0.971 (P < 0.0001).
..
Work indices (positive, negative, and global work) were evaluated .. In BA analysis, mean bias was -16.4 mmHg.% (-98.4 to 65.7 mmHg.%).
for each patient and each configuration. Furthermore, segmental .. For global negative work, correlation between measures and estima-
..
pressure–strain loop areas were calculated for each segment. All .. tions was 0.878 (P < 0.0001). In BA analysis, mean bias was
these invasive and measured indices were compared using linear ..
.. -41.4 mmHg.% (-87.5 to 4.7 mmHg.%) (Figure 5).
regression method7 and Bland–Altman (BA) analysis.10,11 Linear ..
regression is a linear approach for modelling the relationship
.. Variations between CRT configurations
..
between an index variable and another independent variable, using .. There is no statistical difference in the accuracy of the estimated
..
linear predictor functions to model relationships. It gives the intra- .. pressure between the 5 different CRT configurations with a total
class correlation coefficient. BA analysis consists in a method of data .. error between 40.2 and 50.6%, total RMSE between 11.5 and
..
plotting to assess a good agreement between two variables. Average .. 13.3 mmHg. For global positive work and global work index, ICC
value of the two variables is in x-axis and difference between the
.. between measured and estimated indices is all superior to 0.99.
1376 A. Hubert et al.

Downloaded from https://academic.oup.com/ehjcimaging/article/19/12/1372/4911133 by guest on 15 June 2022

Figure 4 Results of segmental pressure–strain loops comparison, included all patients and all CRT configurations. PSL, pressure–strain loops.

Only two configurations (LV pacing alone and BIV pacing 2) were ... determination of the exact timing of mitral opening and closure is
found to be associated with a mean bias >100 mmHg.%. These two
.. more difficult. Another reason is the method of synchronization of
..
configurations had also the lower ICC between estimated and meas- .. pressure used. In fact, estimated and measured pressure curve were
ured global negative work (0.893 and 0.886, respectively) (Table 2).
..
.. synchronized on the time of systolic peak, which was physiologically
.. between AVO and AVC. The observed difference in magnitude of
..
.. systolic pressure measured by blood pressure cuff and pressure cath-
Discussion .. eter is not surprising since arterial peak pressure is augmented when
..
The main findings of this study were (i) the estimation of pressure
.. the arterial waves travel towards the periphery. The method of
.. Russell et al.5 does not provide a means to estimate LV diastolic pres-
curve and therefore of PSL area by the algorithm developed by ..
.. sure, and the marked difference between estimated and measured
Russel et al.2,4 is imperfect with an increased bias for greater area and .. pressures during diastole reflects this limitation.
(ii) Although this estimation is imperfect, consequences on global ..
..
work indices are limited with good correlations between estimated .. . . . for a precise work estimation
and measured work, regardless of CRT configuration. ..
.. While our results about myocardial work are consistent with the
..
An imprecise pressure prediction. . . .. study of Russel et al.2,4, a significant bias in BA analysis of PSL area esti-
.. mation was found. It reveals a paradox: although the estimation of
The mean value for all patient of RMSE was found equal to ..
12.3 mmHg for the estimated pressure wave curve. Differences were .. LV-pressure is imprecise, the deducted estimation of LV- work is
.. accurate. The major reason is that the temporal integration induces a
observed between maximum systolic values of measured and esti- ..
mated pressures. This observation points to a flaw of this method:
.. smoothing of the difference between measured and estimated works.
.. Another reason is the relative precision of the estimation of the pres-
the arterial systolic pressure measure by a brachial cuff is imprecise ..
and this imprecision grows when the arterial pressure is high. On top
.. sure between AVO and AVC (Table 2). This moment of the cardiac
..
of it, arterial pressure could be false if the patient has brachial vascular .. cycle is (i) the period when the pressure is maximal and (ii) the period
disease.
.. when the LV strain is maximal (Figure 2). As myocardial work value
..
But the accuracy of the estimated curve is not equal along the car- .. resulted in pressure and strain, his value provides from this interval
.. (i.e. AVC–AVO), when the estimation of the pressure is the best.
diac cycle. In fact, the precision of the estimation of the pressure ..
between AVO and AVC is quite excellent, with a RMSE of only .. This questions on the added value of LVP estimation on top of
.. strain curves. In fact, other indices, based only on strain curves analy-
6.5 mmHg. On the contrary, when we observe the pressure estima- ..
tion away from the peak, its precision is worse and especially before .. sis,12 were developed to explore myocardial work, in particular for
.. patients with mechanical dyssynchrony.13–15
MVC and after MVO (in diastole). However, although the error ..
increases, the RMSE is stable. Indeed, pressure level is very low ..
.. Influence of CRT configuration and
before MVC and after MVO so the absolute error, in mmHg is quite ..
low. A possible explanation is that aortic exact timings are quite easy .. clinical additive value
..
to determine visually in APLAX view so errors in this interval are low .. We didn’t really find any significant difference between the different
whereas the mitral valve had a visually lower kinetic so the
.. hemodynamic conditions but there were two configurations with a
Pressure–strain loop clinical experimental validation 1377

Downloaded from https://academic.oup.com/ehjcimaging/article/19/12/1372/4911133 by guest on 15 June 2022

Figure 5 Results of global work indices comparison, on all patients and all CRT configurations. (Panel A) Intraclass correlation and Bland–Altman
analysis of global positive work (GPW); (Panel B) Intraclass correlation and Bland–Altman analysis of global negative work (GNW); (Panel C)
Intraclass correlation and Bland–Altman analysis of global work index (GWI).

Table 2 Comparison of estimated vs. measured markers as a function of CRT configuration

LV pressure Segmental area Global work indices ICC

................................................ ...................................... .................................................................
Total Total RMSE ICC Mean bias GPW GNW GWI
error (%) (mmHg) (mmHg.%)
....................................................................................................................................................................................................................
CRT OFF 50.6 ± 17.8 13.3 ± 5.4 0.983 53.9 ± 126.4 0.996 0.961 0.997
LV pacing 40.4 ± 24.0 11.9 ± 4.4 0.992 141.9 ± 121.3 0.991 0.893 0.997
RV pacing 41.2 ± 25.2 12.7 ± 5.9 0.985 66. 6 ± 145.7 0.986 0.928 0.992
BIV pacing 1 40.2 ± 17.7 11.5 ± 4.6 0.988 86.8 ± 114.1 0.992 0.991 0.993
BIV pacing 2 43.6 ± 19.8 12.0 ± 3.6 0.991 108.9 ± 127.0 0.993 0.886 0.994
1378 A. Hubert et al.

..
lower accuracy of estimated PSL area and work: LV pacing and multi- .. Conclusion
point BIV pacing. These two configurations had the best strain (and ..
.. Although the estimation of the LV pressure is imperfect, the
probably the best synchronization of deformation of all the LV- ..
segments) but with the same accuracy of the estimated wave curve .. deducted estimation of LV myocardial work is accurate, which con-
..
than others. As PSL and work resulted in the multiplication of LV .. firms previous studies. The added value of LV pressure estimation on
pressure and strain, a great strain increases the effect of the error in
.. top of strain curves, in regard of these results, could be debated but
..
the pressure curve estimation. In the present study, we focused on .. further works and prognostic data vs. simple strain information will
.. have to be collected.
the comparison the main segmental PSL markers and the derived ..
global work indices, when they are calculated using an estimated ..
.. Acknowledgments
pressure signal or an observed pressure signal, we did not have the ..
power to look at the clinical value associated with all the measure-
.. To the researches Nurses CIC-IT 1414: Valerie Le Moal and Brigitte
.. Lelièvre, to the nurses in the cath lab of the Cardiology Department,
ment that have been done. It is the next step. The correlation being ..

Downloaded from https://academic.oup.com/ehjcimaging/article/19/12/1372/4911133 by guest on 15 June 2022

.. to Christian Serot, (Abbott) for his assistance in organizing the study.
clinically relevant, it is meaningful to consider the next step: a clinically .. The authors acknowledge financial support from the French National
driven study based on the optimization of the pacing modality using ..
.. Research Agency (ANR) and the grant got from General Electric
the iteration of the cardiac work indices.16–19 .. Healthcare.
Because of the consistency between measured and estimated ..
..
works over CRT configurations, global indices appear as a robust sur- .. Funding
rogate estimation for myocardial work. Of note, it has been demon-
.. French National Research Agency (ANR) (ANR-16-CE19-0008-01) (proj-
..
strated that despite few methodological imperfections, the .. ect MAESTRo); and the grant got from General Electric Healthcare.
..
estimation of myocardial work indices looks like to be extremely .. Conflict of interest: E.D. received a research grant from General
promising tools for estimating the response to CRT.6,8 If septal flash .. Electric Healthcare. E.S. is employed by General Electric Healthcare. C.C.
..
and apical rocking are very probably a first step simple way to look .. is a research engineer employed by Abbott (Sylmar, CA, USA).
for mechanical dyssynchrony, the patterns of septal strain and the
..
..
myocardial work indices looks promising enough to encourage new .. References
.. 1. Galli E, Leclercq C, Donal E. Mechanical dyssynchrony in heart failure: still a valid
large multicentre studies trying to rehabilitate the imaging approach ..
in the field of CRT (selection and optimization of patients).1,8,12,20,21 .. concept for optimizing treatment? Arch Cardiovasc Dis 2017;110:60–8.
.. 2. Russell K, Opdahl A, Remme EW, Gjesdal O, Skulstad H, Kongsgaard E et al.
For the clinical practice, one can assume that the visual assessment of .. Evaluation of left ventricular dyssynchrony by onset of active myocardial force
rocking/septal flash could provide a good idea about the patients that .. generation: a novel method that differentiates between electrical and mechanical
..
could be suitable for CRT, but myocardial work analysis provides a .. 3. etiologies. Circ Cardiovasc Imaging 2010;3:405–14.
Gjesdal O, Remme EW, Opdahl A, Skulstad H, Russell K, Kongsgaard E et al.
way to quantify the wasted work in the septum this is expressed as
..
.. Mechanisms of abnormal systolic motion of the interventricular septum during
apical rocking. It is a quite automatic measurement suitable for any .. left bundle-branch block. Circ Cardiovasc Imaging 2011;4:264–73.
.. 4. Russell K, Smiseth OA, Gjesdal O, Qvigstad E, Norseng PA, Sjaastad I et al.
machine learning approach and it could be use even by non-experts. .. Mechanism of prolonged electromechanical delay in late activated myocardium
..
.. 5. during left bundle branch block. Am J Physiol Heart Circ Physiol 2011;301:H2334–43.
Russell K, Eriksen M, Aaberge L, Wilhelmsen N, Skulstad H, Remme EW et al. A
.. novel clinical method for quantification of regional left ventricular pressure-strain
Limits ..
.. loop area: a non-invasive index of myocardial work. Eur Heart J 2012;33:724–33.
The number of patients included was limited but we explored four .. 6. Vecera J, Penicka M, Eriksen M, Russell K, Bartunek J, Vanderheyden M et al.
or five really different haemodynamic conditions for each patient,
.. Wasted septal work in left ventricular dyssynchrony: a novel principle to predict
.. response to cardiac resynchronization therapy. Eur Heart J Cardiovasc Imaging
which permit to increase the number of situations observed. ..
.. 7. 2016;17:624–32. Russell K, Eriksen M, Aaberge L, Wilhelmsen N, Skulstad H, Gjesdal O et al.
We cannot exclude a discrete difference in the haemodynamic ..
reflected by the invasive pressure extracted from the RadiAnalyzer .. Assessment of wasted myocardial work: a novel method to quantify energy loss
.. due to uncoordinated left ventricular contractions. Am J Physiol Heart Circ Physiol
and the haemodynamic of the cycle used using TTE for the estimation .. 2013;305:H996–1003.
of the LV pressure. To limit this bias, we used a mathematical process .. 8. Galli E, Leclercq C, Hubert A, Bernard A, Smiseth OA, Mabo P et al. Role of
.. myocardial constructive work in the identification of responders to CRT. Eur
for the invasive pressure analysis (cycle selected only if they had cor- .. Heart J Cardiovasc Imaging 2018;19:1010–8.
relation of 0.99), and only cycles in stable sinus rhythm were analysed
.. 9. Boe E, Russell K, Eek C, Eriksen M, Remme EW, Smiseth OA et al. Non-invasive
..
in TTE. It is also important to keep in mind that the approach tested .. myocardial work index identifies acute coronary occlusion in patients with non-
.. ST-segment elevation-acute coronary syndrome. Eur Heart J Cardiovasc Imaging
here is considering myocardial work in the period from mitral valve .. 2015;16:1247–55.
closure to mitral valve opening, so inaccuracies before mitral valve .. 10. Bland JM, Altman DG. Statistical methods for assessing agreement between two
..
closure and after mitral valve opening has no impact on the results. .. 11. methods of clinical measurement. Lancet 1986;327:307–10.

The estimates of diastolic pressure are not accurate, which is why no

.. Bland JM, Altman DG. Comparing methods of measurement: why plotting differ-
.. ence against standard method is misleading. Lancet 1995;346:1085–7.
attempt is made to try to estimate (or use) the values in diastole. .. 12. Menet A, Bernard A, Tribouilloy C, Leclercq C, Gevaert C, Guyomar Y et al.
.. Clinical significance of septal deformation patterns in heart failure patients receiv-
Further works are required to quantify the influence of pressure .. ing cardiac resynchronization therapy. Eur Heart J Cardiovasc Imaging 2017;18:
morphology on segmental work and loop area. In addition, the pro- ..
.. 13. 1388–97.
posed work indicators could be compared to integral-based markers .. Stankovic I, Aarones M, Smith HJ, Voros G, Kongsgaard E, Neskovic AN et al.

of regional mechanical activity such as in Bernard et al.,22 since the .. Dynamic relationship of left-ventricular dyssynchrony and contractile reserve in
.. patients undergoing cardiac resynchronization therapy. Eur Heart J 2014;35:
later markers do not rely on the estimation of a pressure waveform. . 48–55.
Pressure–strain loop clinical experimental validation 1379

14. Mada RO, Lysyansky P, Duchenne J, Beyer R, Mada C, Muresan L et al. New .. 18. Kosmala W, Marwick TH. Meta-analysis of effects of optimization of cardiac
automatic tools to identify responders to cardiac resynchronization therapy. .. resynchronization therapy on left ventricular function, exercise capacity, and
J Am Soc Echocardiogr 2016;29:966–72.
.. quality of life in patients with heart failure. Am J Cardiol 2014;113:988–94.
..
15. Stankovic I, Prinz C, Ciarka A, Daraban AM, Kotrc M, Aarones M et al. .. 19. Daubert C, Behar N, Martins RP, Mabo P, Leclercq C. Avoiding non-responders to
Relationship of visually assessed apical rocking and septal flash to response and .. cardiac resynchronization therapy: a practical guide. Eur Heart J 2017;38:1463–72.
long-term survival following cardiac resynchronization therapy (PREDICT-CRT). .. 20. Bernard A, Menet A, Marechaux S, Fournet M, Schnell F, Guyomar Y et al.
Eur Heart J Cardiovasc Imaging 2016;17:262–9. .. Predicting clinical and echocardiographic response after cardiac resynchroniza-
16. Auger D, Hoke U, Bax JJ, Boersma E, Delgado V. Effect of atrioventricular and .. tion therapy with a score combining clinical, electrocardiographic, and echocar-
ventriculoventricular delay optimization on clinical and echocardiographic out- .. diographic parameters. Am J Cardiol 2017;119:1797–802.
comes of patients treated with cardiac resynchronization therapy: a meta-analy-
.. 21. Donal E, Delgado V, Magne J, Bucciarelli-Ducci C, Leclercq C, Cosyns B et al.
..
sis. Am Heart J 2013;166:20–9. .. Rational and design of EuroCRT: an international observational study on multi-
17. Brignole M, Auricchio A, Baron-Esquivias G, Bordachar P, Boriani G, Breithardt .. modality imaging and cardiac resynchronization therapy. Eur Heart J Cardiovasc
OA et al. 2013 ESC Guidelines on cardiac pacing and cardiac resynchronization .. Imaging 2017;18:1120–7.
therapy: the Task Force on cardiac pacing and resynchronization therapy of .. 22. Bernard A, Donal E, Leclercq C, Schnell F, Fournet M, Reynaud A et al. Impact of
the European Society of Cardiology (ESC). Developed in collaboration .. cardiac resynchronization therapy on left ventricular mechanics: understanding
with the European Heart Rhythm Association (EHRA). Eur Heart J 2013;34: .. the response through a new quantitative approach based on longitudinal strain
..

Downloaded from https://academic.oup.com/ehjcimaging/article/19/12/1372/4911133 by guest on 15 June 2022

2281–329. integrals. J Am Soc Echocardiogr 2015;28:700–8.