Download Relationship between endocardial activation sequences defined by

Europace (2012) 14, 99–106
doi:10.1093/europace/eur235
CLINICAL RESEARCH
Pacing and Resynchronization Therapy
Relationship between endocardial activation
sequences defined by high-density mapping
to early septal contraction (septal flash)
in patients with left bundle branch block
undergoing cardiac resynchronization therapy
Simon G. Duckett 1,2*, Oscar Camara3,4, Matthew R. Ginks 1,2, Julian Bostock 2,
Phani Chinchapatnam 1, Maxime Sermesant1,5, Ali Pashaei 3,4, Pier D. Lambiase 1,2,
Jaswinder S. Gill 1,2, Gerry S. Carr-White 2, Alejandro F. Frangi 3,4,6, Reza Razavi 1,
Bart H. Bijnens 6, and C. Aldo Rinaldi 1,2
1
Department of Imaging Sciences, The Rayne Institute, Kings College London, UK; 2The Department of Cardiology, Guy’s and St Thomas’ Hospital, London, UK; 3Information &
Communication Technologies Department, Centre for Computational Imaging & Simulation Technologies in Biomedicine, Universitat Pompeu Fabra, Barcelona, Spain; 4Networking
Research Center on Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Barcelona, Spain; 5INRIA, Asclepios Research Project, Sophia Antipolis, France; and 6Institució
Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain
Received 15 May 2011; accepted after revision 9 June 2011; online publish-ahead-of-print 13 July 2011
Aims
Early inward motion and thickening/thinning of the ventricular septum associated with left bundle branch block is
known as the septal flash (SF). Correction of SF corresponds to response to cardiac resynchronization therapy
(CRT). We hypothesized that SF was associated with a specific left ventricular (LV) activation pattern predicting a
favourable response to CRT. We sought to characterize the spatio-temporal relationship between electrical and
mechanical events by directly comparing non-contact mapping (NCM), acute haemodynamics, and echocardiography.
.....................................................................................................................................................................................
Methods
Thirteen patients (63 + 10 years, 10 men) with severe heart failure (ejection fraction 22.8 + 5.8%) awaiting CRT
and results
underwent echocardiography and NCM pre-implant. Presence and extent of SF defined visually and with M-mode
was fused with NCM bull’s eye plots of endocardial activation patterns. LV–dP/dtmax was measured during different
pacing modes. Five patients had a large SF, four small SF, and four no SF. Large SF patients had areas of conduction
block in non-infarcted regions, whereas those with small or no SF did not. Patients with large SF had greater acute
response to LV and biventricular (BIV) pacing vs. those with small/no SF (% increase dP/dt 28 + 14 vs. 11 + 19% for
LV pacing and 42 + 28 vs. 22 + 21% for BIV pacing) (P , 0.05). This translated into a more favourable chronic
response to CRT. The lines of conduction block disappeared with LV/BIV pacing while remaining with right ventricle
pacing.
.....................................................................................................................................................................................
Conclusion
A strong association exists between electrical activation and mechanical deformation of the septum. Correction of
both mechanical synchrony and the functional conduction block by CRT may explain the favourable response in
patients with SF.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Cardiac resynchronization therapy † Septal flash † Non-contact mapping † Electro-mechanical † Heart failure
* Corresponding author. Simon G. Duckett, Division of Imaging Sciences, The Rayne Institute, 4th Floor, Lambeth Wing, St Thomas’ Hospital London SE1 7EH, UK.
Tel: +44 20 7188 7188 8450; fax: +44 207188 5442, Email: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2011. For permissions please email: [email protected].
100
Introduction
Cardiac resynchronizsation therapy (CRT) is a standard treatment
for patients with severe heart failure and conduction delay usually
demonstrated by left bundle branch block (LBBB). Although CRT
is well established, the non-responder rate is 30%, which poses
significant challenges in improving patient selection.1 A favourable
response to CRT is known to be associated with a specific
‘U-shape’ activation pattern within the left ventricle (LV), which
is not readily apparent from the surface electrocardiogram
(ECG).2 – 4 Non-contact mapping (NCM) is able to characterize
this endocardial activation pattern5 but is an invasive procedure
and not suitable for all CRT candidates. Previous work in our institution demonstrated that NCM can identify regions of slow conduction and LV pacing outside these areas leads to significant
haemodynamic improvement.6,7
Certain heart failure patients demonstrate an early septal contraction and inward motion, stretching the lateral wall, followed by a
septal lengthening, while the lateral wall starts to contract.8 This
fast inward and outward motion of the septum is within the isovolumic contraction time (IVCT) and is referred to as a septal flash (SF).
The presence of a SF is visualized on echocardiography in all views
showing the septum and can be appreciated using either (slowed
down) grey scale or (anatomical) M-mode. Elimination of the SF
has been shown to result in a favourable response to CRT.8 Previous
studies have reported the SF as a pathophysiological mechanism
underlying heart failure in patients with LBBB.9 – 13 Mechanical deformation changes the electrical properties of myocardial tissues14 with
stretching of myocytes decreasing their conductivity15 and computer models have shown that electro-mechanical interaction may
play a role in LBBB.16
We hypothesized that SF is associated with a particular (U
shaped) LV activation pattern that may predict a favourable
response to CRT. We sought to characterize the spatio-temporal
relationship between electrical and mechanical events that may
explain why patients with SF on echocardiography respond to
CRT by directly comparing NCM, acute haemodynamics, and
echocardiographic imaging.
Methods
Study population
All patients had severe heart failure fulfilling standard criteria for CRT
[New York Heart Association (NYHA) class III– IV, QRS .120 ms, left
ventricular ejection fraction (LVEF) ≤35%, and LV end-diastolic diameter ≥55 mm]. The study complied with Declaration of Helsinki and
the local ethics committee approved the research. All patients gave
written informed consent to undergo the non-routine invasive procedures prior to their CRT implant.
Echocardiographic acquisition
Standard echocardiograms with tissue Doppler imaging (TDI) were
acquired on a GE Vivid 7 scanner (General Electric-Vingmed, Milwaukee, WI, USA). Analysis was performed using EchoPac version 6.0.1
(General Electric-Vingmed). Ejection fraction and LV dimensions
were measured using 2D modified biplane Simpson’s method. Transmitral flow velocities were obtained from the apical-4CH view with
S.G. Duckett et al.
pulsed-wave Doppler positioned at the tip of the leaflets. End diastole
(onset of isovolumic contraction) and end systole (aortic valve closure)
were determined using transmitral and aortic Doppler profiles.
Assessment of dyssynchrony and septal flash
The inter-ventricular delay was calculated as the difference between
the LV and right ventricle (RV) pre-ejection periods.17 Intraventricular dyssynchrony was measured with TDI.18,19 To determine
the presence and extent of the SF, two independent clinical experts,
blinded to the electrophysiology studies reviewed all echocardiographic images. Existence of an SF was defined as the presence of
early inward and outward motion within the isovolumic contraction
period seen both visually from the grey scale SAX and 4CH views as
well as on M-mode in the PSLAX, SAX, and 4CH. Bull’s eye plots of
the LV indicated the presence and spatial extent (segments involved)
of the SF. Patients were categorized according to the SF appearance
and extent: a large SF if there was marked fast early septal inward/
outward motion with a prominent displacement involving ≥50% of
the septal segments in an American Heart Association model; a
small SF if there was early inward/outward motion, but with a
very limited displacement; or no SF.
Cardiac magnetic resonance imaging
Cardiac magnetic resonance was performed on a Philips Achieva 1.5T
scanner (Philips, Healthcare, Best, The Netherlands) for delayed
enhancement (DE-MR) imaging to assess myocardial scar. Delayed
enhancement -MR was performed 15 – 20 min following the administration of 0.1– 0.2 mmol/kg gadopentetate dimeglumine (Magnevistw,
Bayer Healthcare, Dublin, Ireland) using conventional inversion recovery techniques.20
Non-contact mapping and acute
haemodynamic study
Non-contact mapping was performed 11 + 8.6 days prior to CRT
implant. Bilateral femoral venous access was used to place 5F
Supreme JSN 401443 quadripolar electrode catheters (St Jude
Medical, Minnetonka, MN, USA) to the high right atrium and RV
apex to perform atrial and RV sensing and pacing, a 5F Supreme
CRD-2 401860 quadripolar electrode catheter (St Jude Medical) was
positioned at the His bundle. The coronary sinus was intubated
using a pre-shaped 8F Fast-Cath SL3 406842 guiding sheath (St Jude
Medical) and a multipolar Pathfinder 16 electrode catheter
01-16-1003 (Cardima Inc., Fremont, CA, USA) was passed to a
postero-lateral or lateral branch of the coronary sinus to perform epicardial LV pacing to replicate standard CRT. A NCM Ensite Array
EC1000 (St Jude Medical, St Paul, MN, USA) was passed via the
femoral artery retrogradely across the aortic valve to the LV cavity.
Via the other femoral artery, a decapolar steerable 6F Livewire
401915 electrode catheter (St Jude Medical) was passed to the LV
cavity along with a high fidelity pressure wire (Radi Medical Systems,
Uppsala, Sweden) with a 5Fr multipurpose catheter. Chamber geometry was reconstructed using a locator signal from the decapolar
catheter.21
The following pacing protocol was performed: (100 bpm with consistent capture, AV delay 100 ms, VV simultaneous): atrium sensed and
paced (AAI), DDDRV, DDDLV, and DDDBIV (conventional CRT) AAI
pacing at 100 bpm was used as baseline to assess LV dP/dtmax measurements and performed prior to each change in pacing mode.22 Ensite
data, pacing parameters and percentage rise in LV dP/dtmax for each
pacing mode were recorded once steady-state pacing had been
achieved for at least 1 min. Capture without fusion was confirmed
101
Electromechanical interaction in LBBB
using VVI pacing and examining the paced QRS morphology. Endocardial maps were obtained in sinus rhythm and in each pacing configuration. Electrograms were acquired at 1200 Hz, giving a temporal
resolution of 0.83 ms. High-pass filter was set at 8 Hz. Virtual unipolar
electrograms recorded from the endocardial surface were analysed to
measure the duration of LV activation. Peak negative dv/dt is a universally accepted indicator of the onset of activation in a region of myocardium. Thus, the onset of activation was taken as the time to the first
peak negative 8 Hz high-passed unipolar virtual electrogram anywhere
in the chamber. The end of LV activation was defined as the time of the
latest peak negative 8 Hz unipolar electrogram on the virtual endocardial surface.23 Lines of block were interpreted from patterns of activation observed in isopotential maps and the associated
morphological features of electrograms. Bull’s eye activation maps
were generated from the NCM data sets by computing the long axis
from the manual identification of the ventricular apex and the centre
of the mitral valve in at least two views. We estimated the short
axis by identifying the ventricular septum, and lateral wall and
drawing a perpendicular line through the long axis. We constructed
temporal bull’s eye plots representing the percentage of activated
tissue (black in the figures) through time. In parallel, we generated
bull’s eye plots of the presence and location of mechanical septal
motion visually identified with echocardiography as well as the position
and extent of scar determined by DE-MR imaging. These visualization
tools allowed a multimodal analysis and integration to directly
compare different factors involved in the studied pathophysiology,
mainly mechanical septal motion, presence and location of scar, and
presence and location of lines of conduction blocks. Lines of conduction block where propagation of the activation wavefront is halted is
visualized as a stagnation of the black– white transition during several
frames in the series of bull’s eyes plots (Figure 1). Acute haemodynamic
responders were considered as an increase in LV dP/dtmax of ≥10%
from baseline AAI pacing.24
Following the invasive protocol patients subsequently underwent
standard CRT implantation. Patients were followed up after 6
months with a repeat echocardiogram and deemed to have remodelled if there was a ≥15% reduction in LV end-systolic volume (ESV).
Statistical analysis
Statistical analysis was performed using the SPSS software package (IBM,
Chicago, IL, USA). Continuous variables were expressed as mean values
+ standard deviation (SD), where appropriate continuous variables
were assessed with a Student’s t-test and one-way analysis of variance.
P values of ,0.05 were considered statistically significant.
Results
Thirteen patients were studied (aged 63 + 10 years, 10 males,
NYHA class 3 + 0). All had LBBB (QRS duration 158 + 24 ms).
Eight patients had non-ischaemic dilated cardiomyopathy (DCM),
and five ischaemic dilated cardiomyopathy (ICM). Mean LV EF
22.8 + 5.8%, inter-ventricular delay 47.8 + 23.6 ms, septal-lateral
delay from TDI 79.2 + 40.3 ms, and SD of 12 segments TDI
50.2 + 13.5 ms (Table 1). Five patients had a large SF, four small
SF, and four no SF (Table 1). Notably, patients with a large SF
had a significantly broader QRS (172.4 + 25.0 ms vs. small
155.5 + 18.0 or no SF 142.8 + 21.0, P , 0.05) and also significantly more inter-venricular mechanical delay but less intraventricular delay.
Figure 1 Patient with a large septal flash. (A) Fused activation and echo bull’s eye plots. The black area indicates activation pattern, red segments area with septal flash. There is initial depolarization of the ventricular septum, followed by slowed conduction (20 – 40 ms) both anteriorly and inferiorly. Areas of block indicated with white arrows. There is breakout of activation inferiorly although slowed depicted by the
dashed arrows. Block remains anteriorly. Activation continues inferiorly until the anterior wall is retrogradely activated. (B) Unipolar isochronal
map. Top row anterior: Bottom row posterior. There is a line of anterior block with activation spreading inferiorly. (C) Left ventricular noncontact mapping unipolar activation map. The activation breaks out in the ventricular septum, instead of spreading uniformly there is an anterior
line of block show by white arrows. The activation spreads inferiorly, leading to a U-shaped activation pattern. Ant, anterior; Sep, septum; Inf,
inferior; Lat, lateral.
102
S.G. Duckett et al.
Table 1 Comparison of patients with varying size of septal flash
All patients
Large SF
Small SF
Absent SF
...............................................................................................................................................................................
Number
13
5
4
4
Age
Sex
63.1 + 10.2
10 M:3 F
66 + 7.1
4 M:1F
57 + 13.0
4M
66.5 + 11.8
2 M:2F
Aetiology of HF
8 DCM
5 ICM
4 DCM
1 ICM
1 DCM
3 ICM
3 DCM
1 ICM
QRS duration (ms)
158 + 24
172.4 + 25.0*
155.5 + 18.0
142.8 + 21.0
Ejection fraction (%)
EDV (mL)
22.8 + 5.8
255 + 67
22.4 + 6.5
269.8 + 65.8
22.5 + 4.4
248.0 + 88.9
23.5 + 7.5
245.3 + 61.6
ESV (mL)
198.2 + 59.1
211.2 + 64.7
195.5 + 83.4
184.5 + 29.8
IVMD (ms)
Septal/lateral delay TDI (ms)
47.8 + 23.6
79.2 + 40.3
69 + 13.9*
56 + 53.2
40.5 + 17.2
82.5 + 22.2
21.8 + 29.1
105 + 20.8
TDI 12 segment SD (ms)
50.2 + 13.5
50.9 + 16.0
47.4 + 14.0
52.2 + 13.0
TIVT (ms)
IVCT (ms)
186 + 128
68 + 63
304 + 72*
133 + 29
222 + 57
59 + 55
125 + 72
38 + 49
IVRT (ms)
118 + 88
171 + 68*
163 + 81
89 + 41
*Significant difference between patients with a large septal flash and small/absent septal flash (P , 0.05).
EDV, end-diastolic volume; ESV, end-systolic volume; IVCT, isovolumic contract time; IVMD. interventricular mechanical delay; IVRT, isovolumic relaxation time;
TIVT,total isovolumic time.
Table 2 Acute haemodynamic response with different pacing modes
AAI baseline dP/dtmax
DDDLV pacing dP/dtmax
% increase
DDDBIV pacing dP/dtmax
% increase
782 + 214
828 + 240
915 + 219*
1038 + 229*
20 + 18
28 + 14
998 + 254*
1141 + 256*
31 + 24
42 + 28
...............................................................................................................................................................................
All patients
Large SF
Small SF
713 + 216
876 + 182*
25 + 14
835 + 141
21 + 22
No SF
Combined small and no SF
880 + 112
797 + 179
856 + 179
866 + 161
–4 + 9
11 + 19
1063 + 152
949 + 181*
22 + 24
22 + 21
*Significant difference from AAI pacing (P , 0.05).
Acute haemodynamic response to
cardiac resynchronization therapy
Overall, there was a significant increase in acute response to CRT
(baseline AAI pacing LV dP/dtmax 782 + 214 mmHg/s, DDDLV
pacing 916 + 219 mmHg/s (P , 0.05), and DDDBIV pacing
998 + 254 mmHg/s (P , 0.05)). Patients with a large SF had a
larger haemodynamic response than those with a small or no SF
(Table 2). All patients with a large SF were acute haemodynamic
responders for DDDLV and Biventricular (BIV) pacing. One
patient with a small SF did not respond to DDDLV pacing and
two did not respond to DDDBIV pacing. No patients without a
SF responded to DDDLV pacing and one did not respond to
BIV pacing.
Clinical response and remodelling
At 6 months one patient with no SF had died. All except for one
patient (small SF) improved by at least one NYHA class [responder
rate 92%, (P , 0.05)]. The SF was no longer present after CRT.
With respect to remodelling the overall responder rate was 64%
[average decrease in ESV 19.8 + 21.4%, (P , 0.05)]. All patients
with a large SF remodelled (decrease in ESV 37.1 + 21.2%), two
patients remodelled with a small SF (decrease in ESV 11.5 +
7.4%), and none with no SF (decrease in ESV 1.9 + 13.7%)
(Table 3). Comparing patients with a large SF to small or no SF
there was a greater improvement in ejection fraction (P , 0.05)
and larger decrease in ESV (P , 0.05), and NYHA score
(P , 0.05). For isovolumic times (IVTs) patients with a large
SF had a greater decrease in both IVT (P , 0.05) and IVCT
(P , 0.05), there was no difference in isovolumic relaxation time
(P ¼ 0.2).
Activation maps
Figures 1 and 2 shows examples of the NCM and echo bull’s eye
plots from a patient with and without a SF. While the activation
spreads homogenously over the endocardial surface when no SF
is present, there were definite areas of conduction block in the
cases with a large SF (Figure 1). In the large SF patients a specific
103
Electromechanical interaction in LBBB
Table 3 New York Heart Association class and echocardiographic parameters pre- and post-cardiac resynchronization
therapy
All patients
Large SF
Small SF
Absent SF
N 5 13
N55
N54
N 5 3a
...................................
..................................
..................................
..................................
...............................................................................................................................................................................
Pre-CRT
Post-CRT
Pre-CRT
Post-CRT
Pre-CRT
Post-CRT
Pre-CRT
Post-CRT
3+0
1.6 + 0.8b
3+0
1.0 + 0.7
3+0
2.3 + 0.5
3+0
1.7 + 0.6
Ejection fraction (%)
EDV (mL)
22.7 + 5.8
255.5 + 67.1
31.6 + 10.5b
227.5 + 71.3b
22.4 + 6.5
269.8 + 65.8
37.2 + 13.1
211.4 + 83
22.5 + 4.4
248.0 + 88.9
30.8 + 4.6
245.3 + 83.1
23.5 + 7.5
245.3 + 61.6
23.3 + 6.7
230.7 + 16.0
ESV (mL)
198.2 + 59.1
157.9 + 55.6b
211.2 + 64.7
137.0 + 61.1
195.5 + 83.4
171.3 + 69.0
184.5 + 29.8
175.0 + 6.6
TIVT(ms)
IVCT(ms)
186 + 128
68 + 63
147 + 107
48 + 47
304 + 72
171 + 68
173 + 61
33 + 19
222 + 57
59 + 55
144 + 98
53 + 37
127 + 72
38 + 49
139 + 120
64 + 62
IVRT(ms)
118 + 88
99 + 74
133 + 29
140 + 65
163 + 81
91 + 63
89 + 41
74 + 69
NYHA class
a
One patient died.
Significant difference from pre-cardiac resynchronization therapy measurements.
EDV, end-diastolic volume; ESV, end-systolic volume; IVCT, isovolumic contract time; IVRT, isovolumic relaxation time; TIVT, total isovolumic time.
b
Figure 2 Patient with no septal flash. (A) Fused activation and echo bull’s eye plots. There is uniform activation both anterior and inferiorly
with no block. (B) Unipolar isochronal map. Top row anterior: Bottom row posterior. Uniform activation with no block. (C) Left ventricle
non-contact mapping unipolar activation map. The activation breaks out in the ventricular septum, and spreads uniformly with no areas of
block. Ant, anterior; Sep, septum; Inf, inferior, Lat, lateral.
activation pattern was seen with an initial activation of a
small septal region followed by a delay of further activation
during 2–3 frames on the activation map corresponding to 20 –
40 ms delay. Subsequent activation continued either anteriorly or
inferiorly and the opposite area remained blocked until it was activated retrogradely. This resulted in a U-shaped activation pattern
with a line of conduction block in either the anterior or inferior
region. In patients with a small or no SF there was no conduction
block except in one ICM patient with a small SF who had an area of
block associated with an anterior-lateral scar. In patients with a
large SF the conduction block was present with RV pacing but
was seen to disappear with LV and BIV pacing suggesting the
lines of block were functional in nature.2
Discussion
Our results show that patients with LBBB and a large SF have areas
of conduction block with a U-shaped activation pattern on NCM
104
and this is associated with favourable acute and chronic response
to CRT. With LV or BIV pacing there was disappearance of the
SF which was associated with a more homogenous pattern of electrical activation and improved haemodynamic response. The disappearance of the lines of conduction block during LV and BIV pacing
would suggest that the areas of block were functional. Patients
without a prominent SF did not exhibit such an abnormal electrical
activation pattern and their response to CRT was less marked.
Although these results suggest an electro-mechanical interaction
during the activation period of the LV we are unable to establish a
causal relationship between the presence of a large SF and the
presence of conduction block. Our study design does not permit
us to ascertain whether the conduction block was produced by
the presence of a large SF. In fact the opposite could equally be
the case and the presence of conduction block may produce a
SF. A likely explanation of the observed activation patterns is the
different level of conduction block that may exist within the left
bundle conduction system. The surface ECG is inexact in characterizing the location and extent of specific ventricular delays.2 In
patients with a large SF there may be block in the proximal left
bundle (Figure 1) accordingly the endocardium of the LV would
be expected to be activated from the RV with a delay of
40 –50 ms caused by slow muscle-to-muscle trans-septal conduction. Once the impulse reaches the LV endocardium, it
enters the subendocardial conduction system and may propagate
relatively rapidly within the LV. This has been demonstrated in previous mapping studies2 and would explain the greater degree of
echocardiographic inter-ventricular mechanical delay in the large
SF patients.
In a patient with no SF (Figure 2) the activation pattern in the LV
could be explained by preserved activation within the proximal left
bundle that causes early activation of the LV endocardium of the
septum from where it propagates slowly through the LV myocardium which would correlate more with intra-ventricular delay. The
explanation that the difference between activation patterns in
patients with and without SF is caused by the type of conduction
defect and presence/absence of early LV endocardial activation is
supported by the fact that the patients with SF had a significantly
wider QRS duration at baseline. This may also explains the
better acute and chronic response to CRT as the large SF group
has primarily a conduction defect that can be rectified by CRT.
The placement of the LV lead in the lateral wall opposite to the
origin of septal contraction by the activating wavefront compensated for the delay introduced by the U-shaped activation
sequence hence creating optimal segmental recruitment of viable
myocardium to maximize effective cardiac work. Indeed this activation pattern may demarcate a group of patients most likely to
respond to optimally synchronized segmental recruitment.
Left bundle branch block: measures of
dyssynchrony and the septal flash
All patients had LBBB, however, a large SF was only seen in five.
Left bundle branch block is a complex electrical disease resulting
from conduction delay located at several anatomical levels of the
activation sequence2 which may not be apparent from the
surface ECG.4 The group with a large SF had a longer QRS
S.G. Duckett et al.
duration (large SF, 172.4 + 25.0 ms, small SF, 155.5 + 18.0 ms,
and no SF, 142.8 + 21.0 ms) (P , 0.05). Looking at interventricular dyssynchrony, patients with a large SF had a larger
delay (large SF, 69 + 13.9 ms, small SF, 40.5 + 17.2 ms, and no
SF, 21.8 + 29.1 ms). This fits the hypothesis that in the presence
of the SF complete LBBB is present. Measuring intra-ventricular
dyssynchrony using the septal/lateral delay show the opposite.
Patients with a large SF appear to have less intra-ventricular dyssynnchrony. This may be due to patients without SF having less
transeptal conduction delay but more delay within the LV. Alternatively, it may be due to the measurement method. First, the peak
velocity in patients with an SF is often outside the ejection
period (AVO/AVC) and therefore excluded from the measurements. Secondly, since the SF induces stretching/unstretching of
the different segments of the LV, this is not necessarily reflected
in the velocity of the basal segments. When a 12-segment standard
deviation is used there is no difference between the groups,
emphasizing that measurements based on velocities are not appropriate ways to assess the intra-ventricular dyssynchrony.25
Comparison with previous studies
Auricchio previously demonstrated a heterogeneity of ventricular
activation in patients with LBBB.2 Some patients had a specific
‘U-shaped’ activation sequence with a line of block the location
and length of which was highly variable and was related to the
site and time of LV breakthrough. Prolongation of transseptal
time in patients with LBBB would be expected. However, Auricchio found that approximately one-third of patients with LBBB
had a normal transseptal time and slightly prolonged or nearnormal LV endocardial activation times. The abrupt change in
the transseptal time correlated to a change in the LV breakthrough
site suggesting that LV breakthrough at anterior or septobasal was
via one or more septal branches of the His–Purkinje system,
whereas patients with a significantly prolonged transseptal time
had a mid-septal or septoapical breakthrough site likely to indicate
a cell-to-cell activation sequence from RV to LV. This is analogous
to our findings in patients with and without a large SF. In this study
the ability to change the location and length of the line of block
with pacing supports a functional basis for the lines of block. The
majority of patients with a large SF (80%) had idiopathic cardiomyopathy, thus lacking the presence of ischaemic myocardial
scar associated with morphologically based conduction delay and
block. Previous haemodynamic data showed that patients with
QRS duration ,150 ms have less acute benefit with CRT than
patients with a QRS duration .150 ms. Our data support this
and that of Auricchio showing evidence of a more homogeneous
electrical activation process is less likely to respond to CRT in
patients with QRS complex ,150 ms. Our findings build on
those of Auricchio with the addition of acute and chronic response
to CRT being dependent on the activation pattern which in turn
can be recognized from a relatively simple non-invasive demonstration of the SF. Our findings also give insight into the underlying
pattern of the LV activation in patients with LBBB.
Septal flash and myocardial stretch
One speculative hypothesis for the conduction abnormality seen
with a large SF is shown in Figure 3. Although this is a potential
105
Electromechanical interaction in LBBB
Figure 3 Summary of potential hypothesis with regard to the electrical activation, mechanical events, and their interaction, in patients with left
bundle branch block and a septal flash. Shortly, after the onset of the activation (breaking trough at the mid/basal-septum, coming from the right
side) (A), the fast and large contraction of the septum during the initial phase of the septal flash (B) leads to mechanical stretching of the surrounding
tissues, thus changing its conductivity and slowing down or halting further activation. When activation slowly surpasses the stretched region
(around half of the total activation time, while only a limited area of the left ventricle is activated at this point), either anterior or inferior of the
septal flash region (C/D), it rapidly conducts further over the left ventricle endocardium, activating the rest of the ventricle, and initiating mechanical
activity in the free walls, which start to contract and exhibit force on the contracted septum, which in its turn is stretched, resulting in the outward/
lengthening portion of the septal flash. (E/F). A, anterior wall; AVO, aortic valve opening; I, inferior wall; L, lateral wall; MVC, mitral valve closure; RV,
right ventricle; S, septum.
explanation for the finding of conduction block in SF patients we
cannot substantiate this hypothesis as we can only show an association rather than causation.
Study limitations
Non-contact mapping relies on unipolar signal detection and may
not reliably distinguish between signals from the opposite site of
the septum since they are very sensitive and reflect electrograms
from the entire wall. Non-contact mapping may also be less accurate in enlarged LVs. We studied a small number of patients with
only five having a large SF. There are definite trends that patients
with a large SF have a greater acute response to LV and BIV
pacing and greater remodelling at 6 months. However, due to
the small numbers it is not possible to determine whether these
differences are clinically significant. The design of the bull’s eye
plots was crude. A better method of quantifying the SF is required,
including a quantitative analysis of the extent and location of the
deformation. We included both ICM and DCM patients. There is
evidence that patients with ICM compose a spectrum of different
conduction abnormities due to variable locations and extent of
scar; however, 80% of our patients with a large SF were nonischaemic. A further factor that is very difficult to control for in
such a small mechanistic study is the volume of viable myocardium
available for recruitment during pacing. To fully investigate many of
these limitations would require a large population and normalization of the haemodynamic/echocardiographic remodelling data;
however, given the invasive and complex nature of these clinical
measurements this would not be feasible.
Clinical implications for cardiac
resynchronization therapy
This work concurs with previous studies that show correction of
the SF by CRT to be associated with significant remodelling.8
The identification that a large SF correlates with a LV activation
pattern that is associated with a favourable response to CRT
may have important clinical implications. Non-contact mapping
has been advocated as a way to determine responders to CRT;
however, this is a highly invasive and not practical. If the SF is
able to act as a surrogate for a favourable activation pattern this
may represents a simple, reproducible, and non-invasive tool to
106
predict CRT response and aid patient selection. If the SF is proven
to be indicative of U-shaped activation it may allow better selection of optimal lateral wall LV lead placement in CRT patients.
Conclusion
In LBBB with a SF there is a strong association between electrical
activation and mechanical deformation. This information may be
valuable in understanding and predicting response to CRT.
Acknowledgements
The author would like to thank Dr Tarik Hussain.
Conflict of interest: M.G. receives research funding from
St. Jude Medical. R.R. receives research funding from Philips
Healthcare. G.C.-W. receives support from Medtronic and St.
Jude Medical. C.A.R is a consultant to St. Jude Medical.
Funding
This work was supported by European Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement n. 224495
(euHeart project) and CENIT Program from Spanish MICINN-CDTI
under grant CEN20091044 (cvREMOD project).
References
1. Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L et al.
The effect of cardiac resynchronization on morbidity and mortality in heart
failure. N Engl J Med 2005;352:1539 –49.
2. Auricchio A, Fantoni C, Regoli F, Carbucicchio C, Goette A, Geller C et al.
Characterization of left ventricular activation in patients with heart failure and
left bundle-branch block. Circulation 2004;109:1133 – 9.
3. Fung JW, Chan JY, Yip GW, Chan HC, Chan WW, Zhang Q et al. Effect of left
ventricular endocardial activation pattern on echocardiographic and clinical
response to cardiac resynchronization therapy. Heart 2007;93:432 – 7.
4. Rodriguez LM, Timmermans C, Nabar A, Beatty G, Wellens HJ. Variable patterns
of septal activation in patients with left bundle branch block and heart failure.
J Cardiovasc Electrophysiol 2003;14:135 –41.
5. Gornick CC, Adler SW, Pederson B, Hauck J, Budd J, Schweitzer J. Validation of a
new noncontact catheter system for electroanatomic mapping of left ventricular
endocardium. Circulation 1999;99:829 –35.
6. Lambiase PD, Rinaldi A, Hauck J, Mobb M, Elliott D, Mohammad S et al. Noncontact left ventricular endocardial mapping in cardiac resynchronisation
therapy. Heart 2004;90:44–51.
7. Ginks MR, Lambiase PD, Duckett SG, Bostock J, Chinchapatnam P, Rhode KS et al.
Simultaneous X-MRI and Non Contact Mapping Study of the Acute Hemodynamic Effect of Left Ventricular Endocardial and Epicardial Cardiac Resynchronization Therapy in Humans. Circ Heart Fail 2011 1;4:170 –9.
8. Cikes M, Bijnens B, Durić Z, Bencic ML, Gosev I, Velagić V et al. Detecting volume
responders prior to implantation of a cardiac resynchronization therapy device via
minithoracotomy: the septal flash as a predictor of immediate left ventricular
reverse remodeling. Heart Surg Forum 2009;12:E362 – 7.
S.G. Duckett et al.
9. Parsai C, Bijnens B, Sutherland GR, Baltabaeva A, Claus P, Marciniak M et al.
Toward understanding response to cardiac resynchronization therapy: left ventricular dyssynchrony is only one of multiple mechanisms. Eur Heart J 2009;30:
940 –9.
10. De Boeck BW, Teske AJ, Meine M, Leenders GE, Cramer MJ, Prinzen FW et al.
Septal rebound stretch reflects the functional substrate to cardiac resynchronization therapy and predicts volumetric and neurohormonal response. Eur J Heart
Fail 2009;11:863 – 71.
11. Kirn B, Jansen A, Bracke F, van Gelder B, Arts T, Prinzen FW. Mechanical discoordination rather than dyssynchrony predicts reverse remodeling upon cardiac
resynchronization. Am J Physiol Heart Circ Physiol 2008;295:H640 –6.
12. Voigt JU, Schneider TM, Korder S, Szulik M, Gürel E, Daniel WG et al. Apical
transverse motion as surrogate parameter to determine regional left ventricular
function inhomogeneities: a new, integrative approach to left ventricular asynchrony assessment. Eur Heart J 2009;30:959 –68.
13. Kerckhoffs RC, Faris OP, Bovendeerd PH, Prinzen FW, Smits K, McVeigh ER et al.
Electromechanics of paced left ventricle simulated by straightforward mathematical model: comparison with experiments. Am J Physiol Heart Circ Physiol 2005;289:
H1889 –97.
14. Lab MJ. Mechanosensitive-mediated interaction, integration, and cardiac control.
Ann NY Acad Sci 2006;1080:282 – 300.
15. Kuijpers NH, ten Eikelder HM, Bovendeerd PH, Verheule S, Arts T, Hilbers PA.
Mechanoelectric feedback leads to conduction slowing and block in acutely
dilated atria: a modeling study of cardiac electromechanics. Am J Physiol Heart
Circ Physiol 2007;292:H2832 –53.
16. Kerckhoffs RC, Lumens J, Vernooy K, Omens JH, Mulligan LJ, Delhaas T et al.
Cardiac resynchronization: insight from experimental and computational
models. Prog Biophys Mol Biol 2008;97:543 –61.
17. Ghio S, Constantin C, Klersy C, Serio A, Fontana A, Campana C et al. Interventricular and intraventricular dyssynchrony are common in heart failure patients,
regardless of QRS duration. Eur Heart J 2004;25:571 – 8.
18. Bax JJ, Bleeker GB, Marwick TH, Molhoek SG, Boersma E, Steendijk P et al. Left
ventricular dyssynchrony predicts response and prognosis after cardiac resynchronization therapy. J Am Coll Cardiol 2004;44:1834 –40.
19. Yu C-M, Fung W-H, Lin H, Zhang Q, Sanderson JE, Lau C-P. Predictors of left
ventricular reverse remodeling after cardiac resynchronization therapy for heart
failure secondary to idiopathic dilated or ischemic cardiomyopathy. Am J Cardiol
2003;91:684 –8.
20. Lima JAC, Judd RM, Bazille A, Schulman SP, Atalar E, Zerhouni EA. Regional heterogeneity of human myocardial infarcts demonstrated by contrast-enhanced
MRI: potential mechanisms. Circulation 1995;92:1117 –25.
21. Schilling RJ, Peters NS, Davies DW. Simultaneous endocardial mapping in
the human left ventricle using a noncontact catheter : comparison of
contact and reconstructed electrograms during sinus rhythm. Circulation 1998;
98:887 – 98.
22. Derval N, Steendijk P, Gula LJ, Deplagne A, Laborderie J, Sacher F et al. Optimizing hemodynamics in heart failure patients by systematic screening of left ventricular pacing sites: the lateral left ventricular wall and the coronary sinus are rarely
the best sites. J Am Coll Cardiol 2010;55:566 – 75.
23. Bleeker GB, Kaandorp TA, Lamb HJ, Boersma E, Steendijk P, de Roos A et al.
Effect of posterolateral scar tissue on clinical and echocardiographic improvement
after cardiac resynchronization therapy. Circulation 2006;113:969 –76.
24. van Gelder BM, Bracke FA, Meijer A, Pijls NHJ. The hemodynamic effect of intrinsic conduction during left ventricular pacing as compared to biventricular pacing.
J Am Coll Cardiol 2005;46:2305 –10.
25. Chung ES, Leon AR, Tavazzi L, Sun J-P, Nihoyannopoulos P, Merlino J et al. Results
of the Predictors of Response to CRT (PROSPECT) Trial. Circulation 2008;117:
2608 –16.