Download Evaluating strict and conventional left bundle

BASIC SCIENCE
Europace (2013) 15, 1816–1821
doi:10.1093/europace/eut132
Evaluating strict and conventional left
bundle branch block criteria using
electrocardiographic simulations
Loriano Galeotti 1, Peter M. van Dam 2, Zak Loring 1,3, Dulciana Chan 1, and
David G. Strauss 1*
1
Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, U.S. Food and Drug Administration, Silver Spring, MD 20993, USA; 2Cognitive Neuro Science,
Radboud University Medical Center, Nijmegen, The Netherlands, 6525EZ; and 3Duke University School of Medicine, Durham, NC 27710, USA
Received 18 December 2012; accepted after revision 15 April 2013; online publish-ahead-of-print 23 May 2013
Aims
Left bundle branch block (LBBB) is a critical predictor of patient benefit from cardiac resynchronization therapy (CRT),
but recent studies suggest that one-third of patients diagnosed with LBBB by conventional electrocardiographic (ECG)
criteria may have a false-positive diagnosis. In this study, we tested the hypothesis that recently proposed strict LBBB ECG
criteria improve specificity in cases of left ventricular hypertrophy (LVH) /dilatation and incomplete LBBB.
.....................................................................................................................................................................................
Methods
We developed five heart models based on a healthy male with increasing degrees of LV hypertrophy and/or dilation. With
each model, we simulated six conduction types: normal conduction, four increments of delayed initiation of LV activation
and results
(incomplete LBBB), and complete LBBB. Simulated ECGs were evaluated for the presence of LBBB by conventional (LV
conduction delay and QRSd ≥120 ms) and strict ECG criteria (LV conduction delay, QRSd ≥140 ms men or ≥130 ms
women, and mid-QRS notching in at least two of the leads I, aVL, V1, V2, V5, and/or V6). Both conventional and strict LBBB
criteria had 100% sensitivity. However, conventional criteria falsely diagnosed LBBB in cases with LVH + LV dilated
10 mm, LVH or LV dilated 10 mm combined with LV initiation ≥6 ms after the right ventricle (RV), and with LV
dilated 5 mm combined with LV initiation ≥12 ms after RV (48% specificity). Strict LBBB criteria resulted in no false positives (100% specificity).
.....................................................................................................................................................................................
Conclusions
New strict LBBB criteria increase the specificity of complete LBBB diagnosis in the presence of LV hypertrophy/dilatation
and incomplete LBBB, which is critical for selecting CRT patients.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Left bundle branch block † Cardiac resynchronization therapy † Electrocardiography † Simulations
Introduction
Cardiac resynchronization therapy (CRT) improves left ventricular
(LV) mechanical function by reducing dyssynchrony between the interventricular septum and LV lateral wall. Cardiac resynchronization
therapy devices differ from conventional pacemakers and implantable
defibrillators by having an additional lead for LV epicardial stimulation.
The additional LV lead needed for CRT carries an additional risk of
complications; clinical trials have reported an overall complication
rate of 15%, of which 10% is related to the LV lead.1 – 4 Although
CRT devices improve heart failure symptoms, quality of life, heart
failure hospitalizations, and mortality in the populations studied,
many individual patients do not benefit. Recent studies5 – 9 have
demonstrated that patients with left bundle branch block (LBBB) are
more likely to benefit from CRT than patients without LBBB. Separate
studies have suggested that one-third of patients diagnosed with LBBB
by conventional ECG criteria may not actually have complete LBBB,
but instead have a combination of LV enlargement and/or incomplete
LBBB (slowed conduction in left bundle branch resulting in a delayed
activation of the LV).10 – 13
Conventional criteria for LBBB were initially developed in the early
20th century based on animal studies.14 However, 15 years later
* Corresponding author. David G. Strauss, 10903 New Hampshire Avenue, WO62 1126, Silver Spring, MD 20993, USA. Tel: +301 796 6323; fax: +301 796 9927,
Email: [email protected]
Simulations were done at Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, U.S. Food and Drug Administration, Silver Spring, MD, USA.
Published by Oxford University Press on behalf of European Society of Cardiology 2013.
1817
Evaluating LBBB criteria using simulations
What’s new?
† Recently proposed strict left bundle branch block (LBBB) electrocardiographic (ECG) criteria have been tested with simulations of different combinations of left ventricular (LV)
hypertrophy and LV dilatation combined with incomplete
and complete LBBB.
† Conventional electrocardiographic (ECG) criteria for LBBB
have poor specificity in the presence of LV hypertrophy and
dilatation.
† The strict ECG criteria for LBBB had 100% sensitivity and specificity, while conventional LBBB criteria had 100% sensitivity
and 48% specificity.
† Utilizing the strict LBBB criteria may improve LBBB diagnostic
specificity and better identify appropriate candidates for
cardiac resynchronization therapy.
research showed that the original criteria for LBBB and right bundle
branch block were reversed.15 In 1956, a study by Grant and Dodge16
suggested that more than one-third of patients diagnosed with newonset LBBB by conventional ECG criteria were misdiagnosed, as the
patients did not have the change in the direction of electrical forces in
the beginning of the QRS complex that should occur if activation only
spreads from the right ventricle (RV) to the LV. They also noted that
the QRS duration usually prolonged by 70 –80 ms with the onset of
LBBB, instead of the 40 ms suggested by canine studies. More recently, two endocardial mapping studies have demonstrated that
one-third of patients with LBBB diagnosed by conventional ECG criteria do not have endocardial activation consistent with LBBB.7,17 In
addition, prior simulation studies highlighted the importance of
mid-QRS notching/slurring, which begins after the first 40 ms and
ends at approximately two-thirds through the QRS duration (corresponding to the time of breakthrough of activation to the LV endocardium and then to the LV lateral wall epicardium, respectively, when
the left bundle is completely blocked).14 Furthermore, differences
in baseline QRS duration between men and women suggest the
need to differentiate the LBBB QRS duration threshold by sex.14
Based on the above experimental and clinical observations, along
with computer simulations, ‘strict’ LBBB criteria were proposed.14,18
The criteria include a terminal negative deflection in V1, QRS duration ≥140 ms for men or ≥130 ms for women, and mid-QRS
notching/slurring in at least two of the leads V1, V2, V5, V6, I, and/
or aVL. The aim of this study is to evaluate both strict and conventional ECG criteria for complete LBBB using computer simulated combinations of left ventricular hypertrophy (LVH) and dilatation with
complete and incomplete LBBB.
Methods
ECG simulations
Simulations were performed using ECGSIM19,20 version 2.2.0-Pre19.01.12. ECGSIM is an Open Source interactive ECG simulation
program, based on the equivalent double-layer source model, which
assumes that the electric potentials at the body surface generated by
the heart are determined completely by the trans-membrane potentials
at the myocardial surface as described by van Dam et al.21 The heart is
modeled as a mesh network of interconnected nodes describing the
endocardial and epicardial surfaces. Other anatomic regions such as
lungs, liver, and torso surfaces are also represented in the model as
mesh networks. The strength of the equivalent double layer at each
cardiac node is characterized by a trans-membrane potential (TMP),
defined by a depolarization and repolarization time, and a waveform of
the TMP in time. The surface ECG can be computed directly from the
given equivalent double layer strength and an appropriate transfer function.20 All the settings needed to obtain ECGs are obtained using appropriate custom software and embedded in the ECGSIM model. For this
study, we modified the nodes’ depolarization time (which is calculated
knowing the activation foci as described below), while the other settings
were left at their default values.
Anatomic models
The anatomic heart model is based on magnetic resonance imaging of a
healthy volunteer (male, 40 years old, 80 kg, 188 cm height), segmented
and converted to an ECGSIM data file using custom software. The normal
anatomy model was subsequently morphed (Figure 1A) to represent two
different amounts of left ventricular dilatation (LV dilated 5 mm and
10 mm models), left ventricular hypertrophy (LVH model), and a combination of left ventricular dilatation and hypertrophy (LVH + LV
dilated 10 mm model). The characteristics of the five anatomic models
used in this study are summarized in Table 1. According to the American
Society of Echocardiography reference limits and partition values of LV
mass and geometry,22 the normal anatomic model is within normal
limits, while the dilated 5 mm and dilated 10 mm models are moderately
and severely abnormal, respectively. The LVH and LVH + dilated 10 mm
models are classified as severely abnormal and hypertrophied.
Activation sequences
The activation sequence describes the timing of depolarization (activation) for all the nodes in the ventricles. It is automatically computed
after defining the foci and timing of activation initiation, corresponding
to the nodes representing the insertion points of the left and right
bundle branches into the LV and RV endocardium. Activation starts
from activation foci at a given time and spreads to adjacent nodes via circular wavefronts. The conduction speed is defined for the endocardium,
epicardium, and mid-myocardium (transmural conduction) independently.
The normal activation sequence (Figure 2) was modeled as described in
previous studies based on observations by Durrer and coworkers.23,24 In
the normal activation sequence, there are three activation foci in LV (midseptum, activated at 0 ms; anterior paraseptal, activated at 5 ms; LV midanterior wall activated at 15 ms) and three in RV (mid-septum activated at
26 ms; anterior paraseptal activated at 6 ms; inferior free wall activated at
16 ms). Thus, LV activation begins 6 ms before RV. Additional activation
sequences modeling incomplete LBBB were simulated by delaying activation of the LV initiation foci by 6, 12, 18, and 24 ms, so that activation in LV
and RV are simultaneous, LV starts activating 6 ms after RV, LV 12 ms after
RV, and LV 18 ms after RV. Left ventricular foci were also inactivated completely to simulate complete LBBB (Figure 2). Conduction velocity was set
at 0.9 m/s in the endocardium, 0.7 m/s in the epicardium, and 0.6 m/s
transmurally. These values are in between experimental values measured
in cardiac muscle fibres and Purkinje fibres and were chosen to model anisotropy of fibre orientation and the effect of intramural insertion of
Purkinje fibres. Similar activation speeds can be measured on the activation maps reported by Durrer et al.23 in isolated human hearts.
1818
L. Galeotti et al.
Simulations and electrocardiographic criteria
The five anatomic models and the six activation sequences were used to
simulate 30 ECGs. Simulated ECGs were then evaluated for meeting conventional ECG criteria for LBBB (LV conduction delay based on QS or rS
in V1 and QRS duration ≥120 ms) and strict ECG criteria for LBBB [QS
or rS in V1, QRS duration ≥140 ms and mid-QRS (beginning after 40 ms)
notching/slurring in at least two of the leads V1, V2, V5, V6, I, and/or aVL].
QRS durations were measured using the MATLAB version of the algorithm described by Martinez et al.25
Results
The performance of conventional ECG criteria to diagnose LBBB is
shown in Table 2. Conventional LBBB criteria correctly identified
all true complete LBBB activation sequence cases (Tables 2 and 3);
however, conventional ECG criteria falsely diagnosed LBBB in 13
of the 25 non-LBBB cases (100% sensitivity and 48% specificity).
Like conventional LBBB criteria, the strict LBBB criteria correctly
identified all complete LBBB activation sequences cases (Table 3);
however, the strict LBBB criteria did not falsely diagnose any LBBB
cases (100% sensitivity and 100% specificity).
QRS duration, morphology, and notching
Figure 1 Anatomic case diagrams (A) and corresponding simulated ECGs with normal conduction (B) and LBBB (C ). Notice the
appearance of clear mid-QRS notching in the presence of complete
LBBB. Left ventricular models are not to scale.
The baseline normal case had a QRS duration of 102 ms (Table 2).
The complete LBBB activation resulted in a 60 ms QRS prolongation
to a QRS duration of 162 ms. When complete LBBB activation was
combined with anatomic models having either LV dilatation or
hypertrophy, QRS duration increased to 176–179 ms; and when
the complete LBBB activation was combined with anatomic models
having both LV dilatation and hypertrophy, QRS duration increased
to 199 ms.
In contrast, activation sequences having delays of LV initiation
(incomplete LBBB) by 6 ms increments resulted in increases of
QRS duration by 0, 5, 11, and 17 ms, respectively. Thus, with the
normal anatomic model, LV activation beginning 18 ms after RV activation (the maximum observed in an endocardial mapping study by
Auricchio et al.7) resulted in a QRS duration of 119 ms, just below
the threshold for conventional LBBB ECG criteria. However, in
anatomic models with modest LV dilation of 5 and 10 mm (which
frequently occurs in patients referred for CRT), QRS duration was
above the 120 ms threshold when LV activation was ≥12 or
≥6 ms relative to the RV.
Table 1 Anatomic ventricular model characteristics
Anatomic model
Lateral and septal LV wall thickness (cm)
(classificationa)
LV end-diastolic diameter (cm)
(classification)
LV mass/BSA (g/m2)
(classification)
...............................................................................................................................................................................
a
Normal anatomy
1.0 (normal)
5.3 (normal)
LV dilated 5 mm
LV dilated 10 mm
1.0 (normal)
1.0 (normal)
6.6 (moderately abnormal)
7.7 (severely abnormal)
142 (moderately abnormal)
186 (severely abnormal)
98 (normal)
LVH
1.8 (severely abnormal)
5.3 (normal)
227 (severely abnormal)
LVH + LV dilated
10 mm
1.5 (moderately abnormal)
7.7 (severely abnormal)
314 (severely abnormal)
Classification is based on American Society of Echocardiography guidelines22 Table 4: ‘Reference limits and partition values of left ventricular mass and geometry.’
1819
Evaluating LBBB criteria using simulations
QRS morphology differed only slightly between the normal
activation sequence and the delayed LV initiation (incomplete
LBBB) activation sequences, with increasing presence of small notching/slurring in some leads as the LV delay increased. When going
from incomplete LBBB with a 24 ms LV delay to complete LBBB
activation sequences, the QRS morphology changed abruptly, and
wide mid-QRS notching appeared in most or all of leads V1, V2,
V5, V6, I, and/or aVL (Figure 2B). This trend was observed in all the
anatomic models (Figure 1B and C ).
The LV hypertrophy and dilation anatomic cases had increased
QRS amplitude in frontal (V1, V2) and leftward leads (V5, V6) compared with the normal anatomic models (Figure 1B). For example,
in the normal activation case, the R-wave amplitude in V6 increased
from 0.8 mV in the normal anatomic model to 1.0 mV in the LV
dilated 5 mm model, 1.2 mV in the LV dilated 10 mm model, and
1.3 mV in the LVH + LV dilated 10 mm model.
Discussion
Figure 2 Activation sequences of the 6 conduction types on the
normal anatomic model (A) and the corresponding simulated ECG
(B). Case 1 has normal activation, cases 2 – 5 have increasing delay in
initiation of LV activation (incomplete LBBB) and case 6 has complete LBBB. The red circle on the LBBB case highlights the
mid-QRS notching.
This study demonstrates that the strict LBBB criteria are more
specific (100% specificity) than conventional LBBB criteria (48%
specificity), while both sets of criteria result in 100% sensitivity in
diagnosing complete LBBB. Improvement in the diagnostic specificity
for LBBB is likely to increase the response rate of CRT. In fact,
although CRT improves patients’ quality of life and reduces heart
failure hospitalizations, 30% of patients do not respond to
CRT.26 – 28 Consequently, this has created interest in the development of better methods to identify likely responders. Approximately
65 500 CRT-defibrillator (CRT-D) devices are implanted yearly in
the USA;29 however 31% of patients (20 000) receiving CRT-D
devices are not diagnosed with LBBB,30 and one-third (15 500)
of the 45 000 LBBB patients (by conventional ECG criteria) may
be misdiagnosed. In addition, outcomes of some clinical studies
have shown response close to 50% in the placebo arm1 and complication rates up to 15%.1 – 4 This highlights the need for improved criteria to identify which patients will benefit from CRT.
In the current study, we observed that QRS duration increased in
the presence of LV dilation and hypertrophy. These enlarged LV
models have larger diameters; therefore, the distance between
points in the LV increases. Since the conduction velocities and activation times were unchanged, the time needed for the activation wavefronts to reach the last node increased, resulting in an increased QRS
duration. In cases simulating LVH there was an additional increase in
QRS duration due to the increased wall thickness, requiring even
more time for the activation wavefront to reach the epicardial
surface.
The strict LBBB criteria used in this study require a QRS duration
≥140 ms for men and ≥130 ms for women.14 The difference in QRS
duration threshold is necessary to account for sex differences in
heart size. We used a 140 ms threshold because our model was
based on a male heart. In the current study, all complete LBBB activation cases had a QRS duration of at least 162 ms, while only one noncomplete LBBB activation case had a QRS duration greater than
140 ms. This is consistent with an endocardial mapping study performed by Auricchio et al.,7 which observed an average QRS duration
of 179 ms in patients with LV activation beginning at least 40 ms after
RV activation (suggesting true complete LBBB) and an average QRS
1820
L. Galeotti et al.
Table 2 Simulated QRS duration and diagnostic performance of conventional LBBB ECG criteria in 30 simulations
Simulated activation sequence
............................................................................................................................................
Normal activation
(RV initiation 6 ms
after LV)
LV initiation
delayed 6 ms
LV initiation
delayed 12 ms
LV initiation
delayed 18 ms
LV initiation
delayed 24 ms
Complete
LBBB
Normal anatomy
QRSd 102
True Neg
QRSd 102
True Neg
QRSd 107
True Neg
QRSd 113
True Neg
QRSd 119
True Neg
QRSd 162
True Pos
LV dilated 5 mm
QRSd 109
True Neg
QRSd 109
True Neg
QRSd 115
True Neg
QRSd 121
False Pos
QRSd 127
False Pos
QRSd 177
True Pos
LV dilated 10 mm
QRSd 117
True Neg
QRSd 117
True Neg
QRSd 122
False Pos
QRSd 128
False Pos
QRSd 134
False Pos
QRSd 179
True Pos
LVH
QRSd 114
True Neg
QRSd 127
False Pos
QRSd 114
True Neg
QRSd 126
False Pos
QRSd 120
False Pos
QRSd 131
False Pos
QRSd 126
False Pos
QRSd 137
False Pos
QRSd 132
False Pos
QRSd 143
False Pos
QRSd 176
True Pos
QRSd 199
True Pos
...............................................................................................................................................................................
Anatomic model
LVH + LV dilated 10 mm
LBBB, left bundle branch block; LV, left ventricular; LVH, left ventricular hypertrophy; Neg, negative; Pos, positive; QRSd, QRS duration (ms); RV, right ventricle.
Table 3 Diagnostic performance of conventional (A)
and strict (B) ECG criteria for LBBB
ECG diagnosis
............................................
LBBB present
LBBB absent
surface ECGs in sinus rhythm, in particular when considering only
the presence of notching and the duration of the QRS
complex.31,32 This could be verified using more complex simulation
models;10 however, ECGSIM has been used for multiple research
studies.33 – 37
................................................................................
A: Conventional LBBB criteria
simulated conduction
LBBB simulated
Non-LBBB simulated
Conclusion
5
0
13
12
................................................................................
B: Strict LBBB criteria
simulated conduction
LBBB simulated
Non-LBBB simulated
5
0
0
25
LBBB, left bundle branch block; ECG, electrocardiogram.
duration of 129 ms in patients with LV activation beginning ,20 ms
after RV activation (suggesting a false-positive diagnosis of LBBB).
Limitations
This study included a modest number of simulated cases. Although
the anatomic model was derived from a real patient, the activation sequence and the anatomic modifications were artificially produced.
However, it was consistently demonstrated that there is a large increase in QRS duration and development of mid-QRS notching
when LV activation foci were completely blocked (complete LBBB)
compared with normal or incomplete LBBB activation sequences.
The anisotropy due to the cardiac fibre orientation is not represented in the present model, thus the resulting activation wavefronts
on the cardiac surfaces were circular instead of elongated.23 While
these limitations may be critical for simulating arrhythmias such as
ventricular fibrillation, this is not so crucial for simulating body-
This simulation study supports that strict LBBB criteria have greater
specificity and equivalent sensitivity compared with conventional
LBBB criteria in the presence of LV dilation and hypertrophy. Utilizing
the strict criteria could potentially improve LBBB diagnostic specificity and better identify appropriate candidates for CRT, as demonstrated by Mascioli et al.9 This study also suggests that LV
enlargement in the absence of complete LBBB may cause QRS duration to increase well beyond 120 ms, which is the threshold often
used to select patients for CRT.38,39 Consequently, this may cause
patients without true LBBB to undergo a CRT implant despite a
low likelihood of response.9,26 – 28
Conflict of interest: none declared.
Funding
This work was supported by an FDA Critical Path Initiative Grant and in
part by an appointment to the Research Participation Program at the
Center for Devices and Radiological Health administered by the Oak
Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the U.S. Food and
Drug Administration.
Ethics
The present study utilizes only simulated data and publicly available
numerical models.
1821
Evaluating LBBB criteria using simulations
References
1. Mehra MR, Greenberg BH. Cardiac resynchronization therapy: caveat medicus! J Am
Coll Cardiol 2004;43:1145 –8.
2. Strickberger SA, Conti J, Daoud EG, Havranek E, Mehra MR, Pina IL et al. Patient selection for cardiac resynchronization therapy: from the Council on Clinical Cardiology Subcommittee on Electrocardiography and Arrhythmias and the Quality of
Care and Outcomes Research Interdisciplinary Working Group, in collaboration
with the Heart Rhythm Society. Circulation 2005;111:2146 –50.
3. Cognis CRT-D; Livian CRT-D and Contack Renewal 3 RF HE CRT-D-P010012/S230—
Summary of Safety and Effectiveness Data (SSED). http://www.accessdata.fda.gov/
cdrh_docs/pdf/P010012S230b.pdf (20 September 2012 date last accessed).
4. Medtronic CRDM Cardiac Resynchronization Therapy Defibrillators-P010031/S232—
Summary of Safety and Effectiveness Data (SSED). http://www.accessdata.fda.gov/
cdrh_docs/pdf/P010031S232b.pdf (20 September 2012 date last accessed).
5. Rickard J, Kumbhani DJ, Popovic Z, Verhaert D, Manne M, Sraow D et al. Characterization of super-response to cardiac resynchronization therapy. Heart Rhythm 2010;
7:885 –9.
6. Adelstein EC, Saba S. Usefulness of baseline electrocardiographic QRS complex
pattern to predict response to cardiac resynchronization. Am J Cardiol 2009;103:
238 –42.
7. 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 bundlebranch block. Circulation 2004;109:1133 –9.
8. Zareba W, Klein H, Cygankiewicz I, Hall WJ, McNitt S, Brown M et al. Effectiveness of
Cardiac Resynchronization Therapy by QRS Morphology in the Multicenter Automatic Defibrillator Implantation Trial-Cardiac Resynchronization Therapy
(MADIT-CRT). Circulation 2011;123:1061 –72.
9. Mascioli G, Padeletti L, Sassone B, Zecchin M, Lucca E, Sacchi S et al. Electrocardiographic criteria of true left bundle branch block: a simple sign to predict a better clinical and instrumental response to CRT. Pacing Clin Electrophysiol 2012;35:927 –34.
10. Bacharova L, Szathmary V, Mateasik A. Electrocardiographic patterns of left bundlebranch block caused by intraventricular conduction impairment in working myocardium: a model study. J Electrocardiol 2011;44:768–78.
11. Strauss DG. Understanding ventricular activation. J Electrocardiol 2011;44:282 – 4.
12. Bleeker GB, Schalij MJ, Molhoek SG, Verwey HF, Holman ER, Boersma E et al. Relationship between QRS duration and left ventricular dyssynchrony in patients with
end-stage heart failure. J Cardiovasc Electrophysiol 2004;15:544 –9.
13. Molhoek SG, van Erven L, Bootsma M, Steendijk P, Van Der Wall EE, Schalij MJ. QRS
duration and shortening to predict clinical response to cardiac resynchronization
therapy in patients with end-stage heart failure. Pacing Clin Electrophysiol 2004;27:
308 –13.
14. Strauss DG, Selvester RH, Wagner GS. Defining left bundle branch block in the era of
cardiac resynchronization therapy. Am J Cardiol 2011;107:927 – 34.
15. Barker PS, Macleod AG, Alexander JA. The excitatory process observed in the
exposed human heart. Am Heart J 1930;5:720 –42.
16. Grant RP, Dodge HT. Mechanisms of QRS complex prolongation in man; left ventricular conduction disturbances. Am J Med 1956;20:834 –52.
17. Vassallo JA, Cassidy DM, Marchlinski FE, Buxton AE, Waxman HL, Doherty JU et al.
Endocardial activation of left bundle branch block. Circulation 1984;69:914 –23.
18. Loring Z, Chelliah S, Selvester RH, Wagner G, Strauss DG. A detailed guide for quantification of myocardial scar with the Selvester QRS score in the presence of electrocardiogram confounders. J Electrocardiol 2011;44:544 –54.
19. ECGsim. http://www.ecgsim.org (23 November 2012, date last accessed).
20. van Oosterom A, Oostendorp TF. ECGSIM: an interactive tool for studying the
genesis of QRST waveforms. Heart 2004;90:165–8.
21. van Dam P, van Oostendorp T. Interactive Simulation of the Activation Sequence:
replacing effect by cause. Comput Cardiol 2011;38:657 – 60.
22. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber
Quantification Writing Group, developed in conjunction with the European
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
Association of Echocardiography, a branch of the European Society of Cardiology.
J Am Soc Echocardiogr 2005;18:1440 –63.
Durrer D, van Dam RT, Freud GE, Janse MJ, Meijler FL, Arzbaecher RC. Total excitation of the isolated human heart. Circulation 1970;41:899 –912.
Surawicz B, Childers R, Deal BJ, Gettes LS, Bailey JJ, Gorgels A et al. AHA/ACCF/HRS
recommendations for the standardization and interpretation of the electrocardiogram: part III: intraventricular conduction disturbances: a scientific statement from
the American Heart Association Electrocardiography and Arrhythmias Committee,
Council on Clinical Cardiology; the American College of Cardiology Foundation;
and the Heart Rhythm Society. Endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol 2009;53:976 – 81.
Martinez JP, Almeida R, Olmos S, Rocha AP, Laguna P. A wavelet-based ECG delineator: evaluation on standard databases. IEEE Trans Biomed Eng 2004;51:570 –81.
Zaca V, Mondillo S, Gaddi R, Favilli R. Profiling cardiac resynchronization therapy
patients: responders, non-responders and those who cannot respond—the good,
the bad and the ugly? Int J Cardiovasc Imaging 2011;27:51 –7.
Boriani G, Biffi M, Martignani C, Valzania C, Diemberger I, Bertini M et al. Is cardiac
resynchronization therapy cost-effective? Europace 2009;11(Suppl 5):v93 –7.
McAlister FA, Ezekowitz J, Hooton N, Vandermeer B, Spooner C, Dryden DM et al.
Cardiac resynchronization therapy for patients with left ventricular systolic dysfunction: a systematic review. JAMA 2007;297:2502 – 14.
Hammill SC, Kremers MS, Stevenson LW, Heidenreich PA, Lang CM, Curtis JP et al.
Review of the registry’s fourth year, incorporating lead data and pediatric ICD procedures, and use as a national performance measure. Heart Rhythm 2010;7:1340 –5.
Bilchick KC, Kamath S, Dimarco JP, Stukenborg GJ. Bundle-branch block morphology and other predictors of outcome after cardiac resynchronization therapy in
medicare patients. Circulation 2010;122:2022 –30.
Potse M, Dube B, Vinet A. Cardiac anisotropy in boundary-element models for the
electrocardiogram. Med Biol Eng Comput 2009;47:719 –29.
Wei D, Okazaki O, Harumi K, Harasawa E, Hosaka H. Comparative simulation of excitation and body surface electrocardiogram with isotropic and anisotropic computer heart models. IEEE Trans Biomed Eng 1995;42:343 –57.
Galeotti L, Strauss DG, Ubachs JF, Pahlm O, Heiberg E. Development of an automated method for display of ischemic myocardium from simulated electrocardiograms. J Electrocardiol 2009;42:204 – 12.
van Huysduynen BH, Swenne CA, Bax JJ, Bleeker GB, Draisma HH, van Erven L et al.
Dispersion of repolarization in cardiac resynchronization therapy. Heart Rhythm
2005;2:1286 – 93.
Van Huysduynen BH, Swenne CA, Draisma HH, Antoni ML, Van De Vooren H, Van
Der Wall EE et al. Validation of ECG indices of ventricular repolarization heterogeneity: a computer simulation study. J Cardiovasc Electrophysiol 2005;16:1097 –103.
Kesek M, Gustavsson O, Wiklund U. Nondipolar content of T wave derived from a
myocardial source simulation with increased repolarization inhomogeneity. Ann
Noninvasive Electrocardiol 2009;14:185 –92.
Arthur RM, Wang S, Trobaugh JW. Changes in body-surface electrocardiograms
from geometric remodeling with obesity. IEEE Trans Biomed Eng 2011;58:1565 –73.
Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA III, Freedman RA, Gettes LS et al.
ACC/AHA/HRS 2008 Guidelines for Device-Based Therapy of Cardiac Rhythm Abnormalities: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/
AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and
Antiarrhythmia Devices): developed in collaboration with the American Association
for Thoracic Surgery and Society of Thoracic Surgeons. J Am Coll Cardiol 2008;51:
e1 –62.
Dickstein K, Vardas PE, Auricchio A, Daubert JC, Linde C, McMurray J et al. 2010
Focused Update of ESC Guidelines on device therapy in heart failure: an update of
the 2008 ESC Guidelines for the diagnosis and treatment of acute and chronic
heart failure and the 2007 ESC Guidelines for cardiac and resynchronization
therapy. Developed with the special contribution of the Heart Failure Association
and the European Heart Rhythm Association. Europace 2010;12:1526 –36.