Journal of the American College of Cardiology Vol. 46, No.

12, 2005
© 2005 by the American College of Cardiology Foundation ISSN 0735-1097/05/$30.00
Published by Elsevier Inc. doi:10.1016/j.jacc.2005.09.019

FOCUS ISSUE: CARDIAC RESYNCHRONIZATION THERAPY

STATE-OF-THE-ART PAPERS

Cardiac Resynchronization Therapy

Part 1—Issues Before Device Implantation
Jeroen J. Bax, MD,* Theodore Abraham, MD, FACC,† S. Serge Barold, MD, FACC,‡
Ole A. Breithardt, MD,§ Jeffrey W. H. Fung, MD,储 Stephane Garrigue, MD, PHD,¶
John Gorcsan III, MD, FACC,# David L. Hayes, MD, FACC,** David A. Kass, MD,†
Juhani Knuuti, MD, PHD,†† Christophe Leclercq, MD, PHD,‡‡ Cecilia Linde, MD, PHD,§§
Daniel B. Mark, MD, PHD, FACC,储储 Mark J. Monaghan, PHD,¶¶
Petros Nihoyannopoulos, MD, FRCP, FACC, FESC,*** Martin J. Schalij, MD,*
Christophe Stellbrink, MD,††† Cheuk-Man Yu, MD储
Leiden, the Netherlands; Baltimore, Maryland; Tampa, Florida; Mannheim and Bielefeld, Germany;
Hong Kong, China; Pessac and Rennes, France; Pittsburgh, Pennsylvania; Rochester, Minnesota; Turku, Finland;
Stockholm, Sweden; Durham, North Carolina; and London, United Kingdom
Cardiac resynchronization therapy (CRT) has been used extensively over the last years in
the therapeutic management of patients with end-stage heart failure. Data from 4,017
patients have been published in eight large, randomized trials on CRT. Improvement in
clinical end points (symptoms, exercise capacity, quality of life) and echocardiogra-
phic end points (systolic function, left ventricular size, mitral regurgitation) have been
reported after CRT, with a reduction in hospitalizations for decompensated heart failure
and an improvement in survival. However, individual results vary, and 20% to 30% of
patients do not respond to CRT. At present, the selection criteria include severe heart
failure (New York Heart Association functional class III or IV), left ventricular ejection
fraction ⬍35%, and wide QRS complex (⬎120 ms). Assessment of inter- and particularly
intraventricular dyssynchrony as provided by echocardiography (predominantly tissue
Doppler imaging techniques) may allow improved identification of potential responders
to CRT. In this review a summary of the clinical and echocardiographic results of the
large, randomized trials is provided, followed by an extensive overview on the currently
available echocardiographic techniques for assessment of LV dyssynchrony. In addition,
the value of LV scar tissue and venous anatomy for the selection of potential candidates
for CRT are discussed. (J Am Coll Cardiol 2005;46:2153– 67) © 2005 by the American
College of Cardiology Foundation

Cardiac resynchronization therapy (CRT) has changed the Severe heart failure, New York Heart Association (NYHA)
treatment of patients with end-stage, drug-refractory heart functional class III and IV
failure. To date, eight large randomized, clinical trials in Depressed systolic left ventricular (LV) function, LV ejec-
CRT have been completed (1– 8). A summary of these tion fraction (LVEF) ⬍35%
trials, with a total of 4,017 patients, is provided in Table 1. Wide QRS complex: ⬎120 ms with interventricular con-
The inclusion criteria in these trials were: duction disorder

From the *Department of Cardiology, Leiden University Medical Center, Leiden,
the Netherlands; †Johns Hopkins University, Baltimore, Maryland; ‡University of
South Florida, Tampa, Florida; §University of Klinikum Mannheim, Mannheim,
Germany; 储The Chinese University of Hong Kong, Hong Kong, China; ¶Hopital
Cardiologique du Haut-Leveque, Pessac, France; #University of Pittsburgh, Pitts-
burgh, Pennsylvania; **Mayo Clinic, Rochester, Minnesota; ††Turku PET Center,
University of Turku, Turku, Finland; ‡‡Hopital Pontchaillou, Rennes, France;
§§Karolinska University Hospital, Stockholm, Sweden; 储储Duke Clinical Research
Institute, Durham, North Carolina; ¶¶King’s College Hospital, London, United
Kingdom; ***Hammersmith Hospital, London, United Kingdom; and †††Stadtische
Kliniken Bielefeld, Bielefeld, Germany.
Dr. Abraham receives honoraria from GE, Guidant, Medtronic, St. Jude and
receives research support from Guidant; Dr. Barold received lecture fees from
Medtronic; Dr. Breithardt has been a consultant for Medtronic and Guidant and has
research affiliations with Medtronic, Guidant, and GE Vingmed; Dr. Hayes is on the
advisory board of Guidant Inc. and has been a speaker for Guidant Inc., Medtronic
Inc., St. Jude Medical, and ELA Medical, and has received royalties from Blackwell
Futura; Dr. Gorcsan received research grant support from GE, Toshiba, Siemens,
Medtronic, and St. Jude; Dr. Kass has been a consultant for Guidant Inc.; Dr. Mark
has been a consultant and received grants from Medtronic, Inc. Dr. Monaghan has
received support from Philips, GE, Siemens, Guidant, Medtronic, and Accusphere;
Dr. Schalij is on the advisory board of Guidant and has received research grants from
Medtronic, Guidant, and St. Jude; Dr. Stellbrink is a sponsored investigator for
Guidant, Medtronic, St. Jude, and Biotronik and is also an advisor to Guidant and
Biotronik; Dr. Nihoyannopoulos received research grants and consultant fees from
Medtronic.
Manuscript received April 19, 2005; revised manuscript received September 19,
2005, accepted September 19, 2005.
2154 Bax et al.
Issues Before Device Implantation
Abbreviations and Acronyms
CMR ⫽ cardiovascular magnetic resonance
CRT ⫽ cardiac resynchronization therapy
LV ⫽ left ventricle/ventricular
LVEF ⫽ left ventricular ejection fraction
MIRACLE ⫽ Multicenter InSync Randomized Clinical
Evaluation study
MSCT ⫽ multislice computed tomography
NYHA ⫽ New York Heart Association
TDI ⫽ tissue Doppler imaging
TSI ⫽ tissue synchronization imaging
The Cardiac Resynchronization-Heart Failure (CARE-
HF) study is the only study that, to some extent, has included
the presence of LV dyssynchrony in the inclusion criteria.
In these large trials, the most frequently used primary end
points mainly reflect the functional status (6-min walk test,
NYHA functional class, quality of life score, and peak O2)
and should be considered clinical end points. The vast
majority of studies demonstrated improvement in 6-min
walking distance, NYHA functional class, and quality of
life score. In addition, the majority of studies that evalu-
ated peak O2 reported an improvement in this parameter
after CRT.
Few studies have focused on morbidity and mortality. In
particular, the recent CARE-HF study has focused primar-
ily on morbidity and mortality (6). Inconsistency on the
reduction in hospital admissions has been reported. While a
reduction in hospitalizations was observed in the Multisite
Simulation in Cardiomyopathies (MUSTIC) and Multicenter
InSync Randomized Clinical Evaluation (MIRACLE) trials
(2,3), this was not the case in the CONTAK-Cardiac
Defibrillator (CONTAK-CD) and MIRACLE-Implantable
Cardioverter Defibrillator trials (4,8). The Comparison of
Medical Therapy, Pacing and Defibrillation in Heart Fail-
ure (COMPANION) trial demonstrated a reduction in the
composite end point of all-cause mortality or hospitalization
during 16 months follow-up (5). The CARE-HF study
subsequently demonstrated a clear survival benefit after
CRT as compared to optimized medical therapy (6).
Secondary end points in the studies were predominantly
echocardiographic measurements including:
LV systolic function, LVEF,
LV reverse remodeling, LV volumes,
Mitral regurgitation
In these large randomized trials, detailed information on
echocardiographic data is not consistently available, but
detailed echocardiographic studies have been performed. In
general, LV systolic function improved, LV reverse remod-
eling occurred, and mitral regurgitation decreased. How-
ever, the extent of improvement in LVEF and reduction in
LV volumes varied substantially among studies.
Based on the available trials, the current American College
of Cardiology/American Heart Association/North Ameri-
JACC Vol. 46, No. 12, 2005
December 20, 2005:2153–67
can Society of Pacing and Electrophysiology guidelines state
that CRT is beneficial in patients with heart failure, severe
systolic LV dysfunction, and wide QRS complex (level of
evidence class IB). Although the results are indeed prom-
ising, analysis of individual responses revealed that 20% to
30% of patients do not respond to CRT (3,9). Accordingly,
emphasis has shifted towards selection of potential respond-
ers to CRT, before device implantation. Before discussing
the identification of responders, the definition of a re-
sponder should be addressed.
WHO IS A RESPONDER TO CRT?
Small studies have initially used invasive methods to
assess acute hemodynamic response to CRT. Long-term
response (usually assessed at three to six months of CRT) is
mainly evaluated by clinical or echocardiographic parame-
ters (Table 2). The relationship between acute hemody-
namic response and chronic outcomes is still not entirely
clear.
Clinical parameters are subjective and mainly reflect
symptoms; a substantial placebo effect may be present in
40% of individuals (3). Still, these are the criteria that are
traditionally used to evaluate patients with heart failure, and
may be most relevant from a patient perspective. The
ultimate clinical end points include a reduction in hospital-
ization and mortality rate.
Echocardiographic parameters may be more objective.
Improvements in LV systolic function (mainly expressed as
LVEF) have been reported in most studies, but the absolute
increase in LVEF varied substantially among studies (3,6,8).
Reverse LV remodeling (indicated by a decrease in LV systolic
and diastolic diameters and volumes) has been reported
consistently in CRT studies and often used as an indicator
of response (2,3,6,8). Indeed, molecular changes in the
lateral wall of the LV such as increased stress kinase levels
may decrease after CRT. Whether these changes contribute
to the reverse LV remodeling or are merely a result of the
reverse remodeling is yet unclear. Echocardiography has also
been used to demonstrate improvements in inter- and
intraventricular dyssynchrony (10).
Clinical non-responders have also been reported when
correlated with echocardiographic results. Yu et al. (11) used
a 15% improvement in LV end-systolic volume index after
CRT and demonstrated non-response in 13 of 30 (43%)
patients. Moreover, clinical and echocardiographic response
to CRT may not always appear simultaneously, and patients
who respond clinically may not exhibit reverse LV remod-
eling and vice versa. It appears that more patients exhibit
improvement in clinical parameters as compared to echo-
cardiographic markers. This discrepancy may even further
complicate the definition of response to CRT.
QRS DURATION TO PREDICT RESPONSE TO CRT
The response to CRT was initially considered to result in
part from resynchronization of interventricular dyssyn-
JACC Vol. 46, No. 12, 2005 Bax et al. 2155
December 20, 2005:2153–67 Issues Before Device Implantation

Table 1. CRT in Randomized Clinical Trials

End Points

Trials Design Patients (n) Primary Secondary Results Summary

PATH-CHF (1) Crossover 41 6MWT NYHA functional class Improvement in
Peak VO2 QOL 6MWT
Hospitalizations NYHA functional class
QOL
Less hopitalizations
MUSTIC-SR (2) Crossover 58 6MWT NYHA functional class Improvement in
QOL 6MWT
Peak VO2 NYHA functional class
LV volumes QOL
MR Peak VO2
Hospitalizations LV volumes
Total mortality MR
Less hospitalizations
MIRACLE (3) Parallel arms 453 6MWT Peak VO2 Improvement in
NYHA functional class LVEF 6MWT
QOL LVEDD NYHA functional class
MR QOL
Clinical composite response LVEF
LVEDD
MR
MIRACLE-ICD (4) Parallel arms 555 6MWT Peak VO2 Improvement in
NYHA functional class LVEF NYHA functional class
QOL LV volumes QOL
MR
Clinical composite response
COMPANION (5) Parallel arms 1,520 All-cause mortality or All-cause mortality and cardiac Reduced all-cause mortality/
hospitalization morbidity hospitalization
CARE-HF (6) Open label, 814 All-cause mortality NYHA functional class Reduced mortality/morbidity
randomized QOL Improvement in
LVEF NYHA functional class
LVESV QOL
Hospitalization for heart failure LVEF
LVESV
PATH-CHF II (7) Crossover (no pacing 86 6MWT NYHA functional class Improvement in
vs. LV pacing) Peak VO2 QOL 6MWT
QOL
Peak VO2
CONTAK-CD (8) Crossover, parallel 490 6MWT LVEF Improvement in
controlled NYHA functional class LV volumes 6MWT
QOL Composite of mortality, NYHA functional class
hospitalizations, VT/VF QOL
LVEF
LV volumes
CARE-HF ⫽ Cardiac Resynchronization-Heart Failure; CONTAK-CD ⫽ CONTAK-Cardiac Defibrillator; COMPANION ⫽ Comparison of Medical Therapy, Pacing and
Defibrillation in Heart Failure; CRT ⫽ cardiac resynchronization therapy; LV ⫽ left ventricular; LVEDD ⫽ left ventricular end-diastolic dimension; LVEF ⫽ left ventricular
ejection fraction; LVESV ⫽ left ventricular end-systolic volume; MIRACLE ⫽ Multicenter InSync Randomized Clinical Evaluation; MIRACLE-ICD ⫽ Multicenter InSync
Implantable Cardioverter Defibrillator trial; MR ⫽ mitral regurgitation; MUSTIC ⫽ Multisite Simulation in Cardiomyopathies; NYHA ⫽ New York Heart Association;
PATH-CHF ⫽ Pacing Therapies in Congestive Heart Failure trial; QOL ⫽ quality-of-life score; VF ⫽ ventricular fibrillation; VO2 ⫽ volume of oxygen; VT ⫽ ventricular
tachycardia; 6MWT ⫽ 6-min walk test.

chrony (dyssynchrony between the left and right ventricle).
Thus, patients with interventricular dyssynchrony were
selected for CRT. This selection was based on the QRS
duration, because this parameter is considered to reflect
interventricular dyssynchrony. Indeed, Rouleau et al. (12)
demonstrated a good relation between interventricular dys-
synchrony (assessed by tissue Doppler imaging [TDI]) and
QRS duration. Accordingly, patients with wide QRS were
considered candidates for CRT. In general, studies used
QRS duration ⬎120 to 130 ms as a selection criterion. The
initial studies required the presence of a left bundle branch
block pattern on the electrocardiogram, whereas more recent
studies also included patients with non-specific interventric-
ular conduction delay (a poorly defined entity) or even right
bundle branch block pattern. The beneficial effect of CRT
on symptoms, exercise capacity, systolic LV function, and
hospitalization rate was demonstrated in these patients with
wide QRS complex, as outlined in the preceding text (Table
1). In addition, data from the Pacing Therapies in Conges-
tive Heart Failure (PATH-CHF) II trial demonstrated that
the benefit of CRT was most pronounced in patients with
QRS duration ⬎150 ms (as compared to patients with QRS
2156 Bax et al. JACC Vol. 46, No. 12, 2005
Issues Before Device Implantation December 20, 2005:2153–67

Table 2. Markers of Chronic Response to CRT

Clinical parameters
NYHA functional class
Quality-of-life score
6-min walk distance, exercise capacity peak VO2
(Heart failure) hospitalizations
(Cardiac) mortality
Echocardiographic parameters
Left ventricular ejection fraction
Left ventricular dimensions/volumes
Mitral regurgitation
Interventricular resynchronization
Left ventricular resynchronization
CRT ⫽ cardiac resynchronization therapy; NYHA ⫽ New York Heart Association;
VO2 ⫽ volume of oxygen.

Figure 2. The QRS duration does not relate with the extent of left
duration 120 to 150 ms) (13). These observations tend to ventricular (LV) dyssynchrony as assessed with tissue Doppler imaging.
support the use of the QRS duration for patient selection. Adapted from Bleeker et al. (17).
However, careful analysis of the individual patients in many
CRT studies demonstrated that 20% to 30% of the patients
It has subsequently been suggested that intraventricular
failed to respond to CRT, despite prolonged QRS duration.
dyssynchrony may predict response to CRT more accurately
In particular, Reuter et al. (14) evaluated 102 patients
(10). In this respect, studies using TDI demonstrated that
undergoing CRT and reported non-response in 18% of
patients with intraventricular dyssynchrony had a high
patients. These observations prompted Molhoek et al. (15)
likelihood of a positive response to CRT (10). Bleeker et al.
to analyze the precise value of the QRS duration to predict
(17) evaluated the relation between QRS duration and LV
response to CRT. Their study included 61 patients, and 45
dyssynchrony (assessed by TDI) in 90 patients with severe
(74%) responded to CRT. The QRS duration at baseline
heart failure; LVEF ⬍35%; and narrow (⬍120 ms), inter-
before pacing was similar between the responders and
mediate (120 to 150 ms), or wide (⬎150 ms) QRS com-
non-responders (179 ⫾ 30 ms vs. 171 ⫾ 32 ms; p ⫽ NS).
plexes. Substantial LV dyssynchrony on TDI was present in
However, a significant shortening in QRS duration after six
27%, 60%, and 70% of patients, respectively. When QRS
months of CRT was observed only in responders. Receiver
duration was considered as a continuous variable, no relation
operating characteristic curve analysis showed that a reduc-
between QRS duration and LV dyssynchrony could be dem-
tion in QRS duration ⬎10 ms had a high sensitivity (73%)
onstrated (Fig. 2). Ghio et al. (18) confirmed the absence of
with low specificity (44%) (Fig. 1) in prediction of respond-
LV dyssynchrony in 48% of patients with an intermediate
ers. Conversely, a reduction in QRS duration ⬎50 ms was
(120 to 150 ms) QRS complex and in 28% of patients with
highly specific (88%) but not sensitive (18%) to predict
a wide (⬎150 ms) QRS complex. These observations
response to CRT. Comparable findings were recently re-
indicate that patients with a wider QRS complex have a
ported by Lecoq et al. (16).
higher likelihood of LV dyssynchrony, although 30% of
patients with wide (⬎150 ms) QRS complex lack LV
dyssynchrony. This 30% may partially explain a similar
percentage of non-responders in the large trials. These
observations have resulted in many echocardiographic stud-
ies evaluating different echocardiographic parameters to
detect LV dyssynchrony and predict response to CRT.

Figure 1. Receiver operating characteristic curve analysis on the change in
QRS duration after six months of cardiac resynchronization therapy (CRT)
demonstrated an optimal sensitivity of 58% and 56% (using a cutoff value
of 30 ms) to predict response to CRT. Adapted from Molhoek et al. (15).
ECHOCARDIOGRAPHIC SELECTION
OF RESPONDERS TO CRT: ROLE OF
INTERVENTRICULAR DYSSYNCHRONY
Echocardiography is the most practical approach to evaluate
dyssynchrony and predict response to CRT. Initial studies
have focused on assessment of interventricular (left-right)
dyssynchrony to predict response. Interventricular dyssyn-
chrony can be evaluated by pulsed-wave Doppler echocar-
diography assessing the extent of interventricular mechani-
cal delay defined as the time-difference between left and
right ventricular pre-ejection intervals; a delay of 40 ms or
JACC Vol. 46, No. 12, 2005
December 20, 2005:2153–67
more has been proposed as a marker of interventricular
dyssynchrony (10). Bordachar et al. (19), however, reported
that interventricular dyssynchrony was not related to hemo-
dynamic improvement during CRT.
Tissue Doppler imaging has also been used to assess
interventricular dyssynchrony (11,20,21). Three studies used
TDI to compare the delay between peak systolic velocity
of the right ventricular free wall and the LV. Although
Penicka et al. (21) suggested that assessment of interventric-
ular delay by TDI contributed to the prediction of response to
CRT, this was not confirmed by two different studies where
TDI was used (11,20). Bax et al. (20) evaluated 80 patients
undergoing CRT and demonstrated a similar extent of inter-
ventricular dyssynchrony in the 59 responders and the 21
non-responders (47 ⫾ 34 ms vs. 49 ⫾ 29 ms; p ⫽ NS). Yu
et al. (11) demonstrated that interventricular dyssychrony
was not predictive of response to CRT in 54 patients
undergoing CRT. Thus, most evidence suggests that inter-
ventricular dyssynchrony is not useful in the prediction of
response to CRT.
ECHOCARDIOGRAPHIC SELECTION OF RESPONDERS
TO CRT: ROLE OF LV DYSSYNCHRONY ON M-MODE,
TWO-DIMENSIONAL ECHOCARDIOGRAPHY, AND TDI
A variety of echocardiographic techniques have been sug-
gested for the assessment of LV dyssynchrony and predic-
tion of response to CRT. These techniques include M-mode
assessment, two-dimensional echocardiography using phase
imaging or intravenous contrast, TDI, and potentially
three-dimensional echocardiography (10). Tissue Doppler
imaging is the most extensively tested technique, and
different methods have been proposed, including pulsed-
wave TDI, color-coded TDI, tissue tracking, displacement
mapping, strain and strain rate imaging, and, most recently,
tissue synchronization imaging (TSI). A comprehensive
summary of the merits of the different techniques to predict
response to CRT is provided in Table 3.
M-mode echocardiography. Using an M-mode recording
from the parasternal short-axis view (at the level of the
papillary muscles), the septal-to-posterior wall motion delay
can be obtained, and a cutoff value of 130 ms or more was
proposed as a marker of intraventricular dyssynchrony. Pitzalis
et al. (22) evaluated 20 patients and reported a sensitivity and
specificity of 100% and 63%, respectively, to predict re-
sponse to CRT. In addition, the authors recently demon-
strated the prognostic value of the septal-to-posterior wall
motion delay (23). However, this parameter is often difficult
to obtain. Rose et al. (24) evaluated the feasibility of
obtaining this parameter in the patients of the
CONTAK-CD database. A clear definition of the systolic
deflection of both the septal and posterior wall was possible
in only 45% of 79 patients. Moreover, the septal-to-
posterior wall motion delay did not correlate with LV
reverse remodeling. Most importantly, the sensitivity and
specificity to predict response (both clinical and echocardio-
Bax et al. 2157
Issues Before Device Implantation
graphic) were 24% and 66%. These findings were similar in
patients with ischemic and dilated cardiomyopathy.
Two-dimensional echocardiography. The first use of
two-dimensional echocardiography was that of a semiauto-
matic method for endocardial border delineation (25). The
degree of LV dyssynchrony was quantified in two-
dimensional echocardiographic sequences from the apical
four-chamber view, focusing on the septal-lateral relation-
ships. Computer-generated regional wall movement curves
were compared by a mathematical phase analysis, based on
Fourier transformation. The resulting septal-lateral phase
angle difference is a quantitative measure for intraventricular
(dys)synchrony. Using this approach, patients with exten-
sive LV dyssynchrony between the septum and lateral wall
exhibited an immediate improvement in hemodynamics
after CRT. The second approach utilized echo contrast
(Optison, Mallinckrodt, Hazelwood, Missouri) to optimize
LV border detection (26). With the improved LV border
detection, regional fractional area changes were determined
and plotted versus time, yielding displacement maps. From
these maps, the LV dyssynchrony between the septum and
lateral wall was determined. The authors observed an acute
reduction in LV dyssynchrony after biventricular pacing,
which correlated with an acute increase in LVEF. No
studies on the two-dimensional techniques for prediction of
long-term outcome have been published.
TDI, strain, and strain rate imaging. Tissue Doppler
imaging measures the velocity of longitudinal cardiac mo-
tion and allows comparison of timing of motion in relation
to electrical activity (QRS complex). Different parameters
can be derived, and the most frequently used include the
peak systolic velocity, the time to onset of systolic velocity,
and the time to peak systolic velocity.
The TDI measurements can be obtained directly using
pulsed-wave TDI and using color-coded TDI, which needs
post-processing. With pulsed-wave TDI, only one region
can be interrogated at a time making the procedure time-
consuming and precludes comparison of segments simulta-
neously. Because measurements are influenced by differences
in heart rate, loading conditions and respiration measure-
ments that are not simultaneous may be less meaningful. In
addition, the timing of peak systolic velocity is often difficult
to identify, resulting in imprecise information on LV
dyssynchrony. There is limited evidence of pulsed-wave
TDI to predict response to CRT. Two studies have dem-
onstrated a relation between LV dyssynchrony on pulsed-
wave TDI and improvement in symptoms and/or LVEF
after CRT (27,28), but prediction of response was not
addressed. Bordachar et al. (19) showed that LV dyssyn-
chrony assessed by pulsed-wave TDI correlated with an
improvement in cardiac output and reduction in mitral
regurgitation after CRT. Penicka et al. (21) used pulsed-
wave TDI (with an integration of interventricular and LV
dyssynchrony) and reported a sensitivity of 96% with a
specificity of 77% to predict response to CRT.
2158 Bax et al. JACC Vol. 46, No. 12, 2005
Issues Before Device Implantation December 20, 2005:2153–67

Table 3. Echocardiographic Studies on LV Dyssynchrony to Predict Response to CRT

Follow-Up
Patients Period Echo Methodology for LV Sensitivity Specificity
Authors (n) (month) Technique Dyssynchrony Main Findings (%) (%)
Pitzalis et al. (22) 20 1 M-mode Septal-to-posterior wall Septal-to-posterior delay ⱖ130 ms 100 63
motion delay predicted LV reverse
remodeling
Pitzalis et al. (23) 60 14 M-mode Septal-to-posterior wall Septal-to-posterior delay ⱖ130 ms NA NA
motion delay predicted event-free survival
Rose et al. (24) 79 6 M-mode Septal-to-posterior wall Septal-to-posterior delay ⱖ130 ms 24 66
motion delay insufficient to predict LV
reverse remodeling
Breithardt et al. (25) 34 Acute 2D phase Difference between the Delayed lateral wall motion NA NA
imaging lateral and septal predicted acute hemodynamic
wall motion phase improvement
angles
Kawaguchi et al. (26) 10 9 Contrast Septal and lateral 1 Septal inward motion, NA NA
echo fractional area 2 spatial and temporal LV
changes dyssynchrony by 40%, and
correlated with 1 LVEF
Penicka et al. (21) 49 6 Pulsed-wave Ts(onset) of 3 basal Summation of inter- and 96 71
TDI LV and 1 basal RV intra(LV)-ventricular delay
segments ⬎102 ms predicted 1 LVEF
Ansalone et al. (27) 21 1 Pulsed-wave Systolic dyssynchrony Extensive LV dyssynchrony NA NA
TDI among 5 basal resulted in 1 in symptoms, and
segments 1 LVEF
Garrigue et al. (28) 12 12 Pulsed-wave Ts(onset) between Post-CRT reduction in LV NA NA
TDI septum and lateral dyssynchrony and improvement
wall in symptoms
Bordachar et al. (19) 41 3 Pulsed-wave Maximal difference in Maximal difference in 12 LV NA NA
TDI, 12 LV segments for segments for Ts, T(onset),
M-mode, Ts, Ts(onset), Ts-SD closely correlated with
pulsed- Ts-SD, and DLC improvement of MR and CO,
wave but not interventricular delay by
Doppler pulsed-wave Doppler echo
Yu et al. (29) 25 3 Color-coded Ts of 12 LV segments Improve LV dyssynchrony by NA NA
TDI in ejection phase delaying Ts in 12 LV segments
globally resulting in
homogenous Ts
Bax et al. (30) 25 Acute Color-coded Septal-to-lateral delay Septal-to-lateral delay of Ts ⱖ60 76 88
TDI of Ts in ejection ms predicted 1 LVEF
phase
Yu et al. (31) 30 3 Color-coded Ts-SD of 12 LV Ts-SD of 12 LV segments ⬎33 NA NA
TDI segments in ejection ms predicted LV reverse
phase remodeling
Bax et al. (20) 85 12 Color-coded Septal-to-lateral delay Septal-to-lateral delay of Ts ⱖ65 92 92
TDI of Ts in ejection ms predicted LV reverse
phase remodeling; and associated with
lower event rate
Notabartolo et al. (32) 49 3 Color-coded Maximal difference in Maximal difference in Ts in 97 55
TDI Ts in 6 basal 6 basal segments ⬎110 ms
segments (both predicted reverse remodeling
ejection phase and
post-systolic
shortening)
Yu et al. (11) 54 3 Color-coded Ts-SD of 12 LV Ts-SD of 12 LV segments ⬎31 96 78
TDI, SRI segments in ejection ms predicted LV reverse
phase (and 17 other remodeling
parameters)
Sogaard et al. (33) 20 12 Color-coded DLC Number of basal segments with NA NA
TDI, SRI DLC predicted 1 LVEF
Sogaard et al. (34) 20 Acute Color-coded TT and DLC Optimization of V-V timing by NA NA
TDI, TT, TT resulted in 2 DLC and
SRI 1 LVEF

Continued on next page

JACC Vol. 46, No. 12, 2005 Bax et al. 2159
December 20, 2005:2153–67 Issues Before Device Implantation

Table 3. Continued
Follow-Up
Patients Period Echo Methodology for LV Sensitivity Specificity
Authors (n) (month) Technique Dyssynchrony Main Findings (%) (%)
Breithardt et al. (35) 18 Acute Strain, SRI Strain and SRI at Baseline: lateral wall strain and NA NA
septal and lateral strain rate higher than septal;
walls CRT: septal wall strain and
strain rate higher than lateral
Sun et al. (36) 34 Acute Displacement, Strain, SRI, and 1 LVEF and 1 peak strain rate NA NA
strain, displacement at at lateral wall by CRT; 1
and SRI septal and lateral Septal and inferior wall
walls displacement by LV pacing
Popović et al. (37) 22 1–12 Displacement Strain at septal and 1 Global LV peak strain, 2 NA NA
and strain lateral walls coefficient of change in LV
strain, no change in septal or
lateral wall displacement
Dohi et al. (38) 38 Acute TSI and Peak radial strain Delay in septal-posterior strain 95 88
strain septum versus ⱖ130 ms, predicted 1 stroke
posterior wall volume
Gorcsan III et al. (39) 29 Acute TSI Septal-posterior delay Septal-posterior delay ⱖ65 ms 87 100
(both ejection phase predicted 1 stroke volume
and post-systolic
shortening)
Yu et al. (40) 56 3 TSI Ts-SD of 12 LV Ts-SD of 12 LV segments in 87 81
segments in ejection ejection phase ⬎34 ms
phase (inclusion of predicted reverse LV
post-systolic remodeling
shortening)
Zhang et al. (43) 13 3 3D echo Time to minimal Improvement of Tmsv parameters NA NA
systolic volume during CRT Tmsv of 16 LV
(Tmsv) in 6, 12, and segments by 3D echo correlated
16 LV segments closely with Ts-SD of 12 LV
segments by TDI
CO ⫽ cardiac output; CRT ⫽ cardiac resynchronization therapy; DLC ⫽ delayed longitudinal contraction; LV ⫽ left ventricular; LVEF ⫽ left ventricular ejection fraction;
MR ⫽ mitral regurgitation; RV ⫽ right ventricular; SRI ⫽ strain rate imaging; TDI ⫽ tissue Doppler imaging; TMSV ⫽ time to minimal systolic volume; Ts ⫽ time to
peak myocardial systolic velocity; TSI ⫽ tissue synchronization imaging; Ts(onset) ⫽ time to onset of myocardial systolic velocity; Ts-SD ⫽ standard deviation of time
to peak myocardial systolic velocity; TT ⫽ tissue tracking; V-V interventricular.

Eight studies have used color-coded TDI to assess LV
dyssynchrony and predict outcome (Table 3) (11,20,29 –34).
From these color-coded images, TDI tracings can be obtained
by post-processing, and the majority of studies have used
time to peak systolic velocity to assess LV dyssynchrony.
Initially, investigators focused on the four-chamber view to
identify LV dyssynchrony by color-coded TDI (Fig. 3).
Velocity tracings were derived from the basal septal and
lateral segments, and the septal-to-lateral delay was mea-
sured. It was shown that a delay ⱖ60 ms was predictive of
acute response to CRT (30). Subsequently, a four-segment
model was applied, which included four basal segments
(septal, lateral, inferior, and anterior) (20). It was shown
that a delay ⱖ65 ms allowed prediction of response to CRT.
Using this cutoff value, sensitivity/specificity for prediction
of clinical improvement (defined by an improvement in
NYHA functional class and 6-min walking distance) were
both 80% whereas sensitivity/specificity for LV reverse
remodeling (defined as a ⱖ15% reduction in LV end-
systolic volume) were both 92%. In addition, patients with
LV dyssynchrony ⱖ65 ms had a favorable prognosis after
CRT.
The extensive studies by Yu et al. (11,29,31) have used a
12-segment model. Tracings were derived from 12 seg-
ments, and an LV dyssynchrony index was derived from the
standard deviation of all 12 time intervals demonstrating
that a dyssynchrony index ⱖ31 ms yielded a sensitivity and
specificity of 96% and 78% to predict LV reverse remodel-
ing (11). In general, prediction of response to CRT based
on time to peak systolic velocity (using a varying number of
segments) yields a high sensitivity (ranging from 76% to
97%) and specificity (ranging from 55% to 92%) (Table 3).
Seven studies have used tissue tracking, strain, and/or
strain rate imaging (11,33–38). Tissue tracking (GE Ving-
med, Horten, Norway) provides a color-coded display of
myocardial displacement, allowing for easy visualization LV
dyssynchrony and the region of latest activation. Sogaard et
al. (34) pioneered this approach and demonstrated that the
number of segments with delayed longitudinal contraction
was related to the improvement in LVEF during CRT.
Strain and strain rate analysis is performed by off-line
analysis of the color-coded tissue Doppler images. Strain
analysis allows direct assessment of the extent and timing of
myocardial deformation during systole and is expressed as
2160 Bax et al. JACC Vol. 46, No. 12, 2005
Issues Before Device Implantation December 20, 2005:2153–67

Figure 3. Color-coded four-chamber tissue Doppler image (upper left). Post-processing yields velocity tracings (right); severe left ventricular dyssynchrony
is present as indicated by the delay in the peak systolic velocity of the septum (yellow curve) as compared to the lateral wall (green curve).

the percentage of segmental shortening or lengthening in important in patients with ischemic cardiomyopathy with
relation to its original length (10). The main advantage over the presence of scar tissue. Breithardt et al. (35) showed that
TDI is that strain analysis allows differentiation between CRT reversed the pathologic septal-lateral strain relation-
active systolic contraction and passive motion. This is ships and reduced the incidence of early systolic pre-stretch

Figure 4. Tissue synchronization imaging, four-chamber view. (Left) The color represents timing: green ⫽ normal timing; red ⫽ severe delay.
Color-guided visual identification of the site of latest activation facilitates sample-placement to derive the tissue Doppler imaging (TDI) velocity tracings.
In this patient, the earliest activation (green) is in the lateral wall, and the latest activation is in the lateral wall (red); the TDI velocity tracings (right)
confirm the delay between the septum and lateral wall.
JACC Vol. 46, No. 12, 2005 Bax et al. 2161
December 20, 2005:2153–67 Issues Before Device Implantation

Figure 5. Four-dimensional tissue synchronization imaging (TSI) of a normal individual. From the different views (left), a three-dimensional impression
of the left ventricle is reconstructed (middle). The entire three-dimensional image of the left ventricle is green indicating no dyssynchrony. Post- processing
allows the display of the different segments in a polar map format (right) to further facilitate identification of the site of latest activation.

in the late activated wall and of post-systolic shortening.
However, no study with strain or strain rate imaging has so
far reported on the actual prediction of response to CRT
(Table 3), except for Yu et al. (11), who demonstrated that
strain rate imaging could not predict LV reverse remodel-
ing. Despite the technical advantages of the strain imaging,
the technique has not become routine practice in the
evaluation of patients considered for CRT. The main
limitations are the time-consuming aspect of the technique,
the high operator dependency, and the moderate reproduc-
ibility. Only one direct comparison between TDI and strain
rate imaging has been reported in 54 patients undergoing

Figure 6. Four-dimensional tissue synchronization imaging (TSI) of a patient with left ventricular dyssynchrony. From the different views (left), a
three-dimensional impression of the left ventricle is reconstructed (middle). The region of latest activation is indicated in red. Post-processing yields the
polar-map format (upper right) indicating in red the site of latest activation (anteroseptal), and further calculations can be performed (lower right) to
further quantify the extent of left ventricular dyssynchrony using different parameters.
2162 Bax et al. JACC Vol. 46, No. 12, 2005
Issues Before Device Implantation December 20, 2005:2153–67

Figure 7. Real-time three-dimensional full volume analysis of regional left ventricular function showing (top left) the reconstructed left ventricular cast and
the bulls-eye display (bottom left) of the 16 segments. Changes in regional volume (color-matched) for each of the 16 segments is displayed in the upper
right. The anteroseptal and septal segments (light blue and green) show poor function and significant delay in achieving a minimum volume compared
to other segments. A first derivative display of the regional volume curves is shown in the lower right panel and also demonstrates significant dispersion
in the timing of minimal regional volume (indicated by the zero crossing points).

CRT (11). Left ventricular dyssynchrony on TDI was
predictive of LV reverse remodeling, but strain rate imaging
failed to predict response to CRT (11).
TSI TO ASSESS LV DYSSYNCHRONY
A recent addition to the tissue Doppler approach to quan-
tify LV dyssynchrony has been the automated color-coding
of time to peak longitudinal velocities. This color-coding of
temporal velocity data is superimposed on the routine
two-dimensional echocardiographic images to provide vi-
sual mechanical information on the anatomical regions.
Tissue synchronization imaging (GE Vingmed) (38 – 40) is
a signal-processing algorithm of the tissue Doppler data to
automatically detect peak positive velocity and then color-
code the time to peak velocities in green for normal timing,
yellow-orange for moderate delay, and red for severe delays
in peak longitudinal velocity (Fig. 4). Interval start time is
manually fine-tuned to begin with aortic valve opening to
exclude isovolumic contraction velocity and extended to
rapid filling (E-wave) to include post-systolic LV dyssyn-
chrony; TSI color-coding is then used to guide placement of
7 ⫻ 15 mm oval regions of interest in the basal and
mid-segments from apical views. Regions of interest are
localized where the color-coding of timing is most repre-
sentative for the anatomical segment for time-velocity curve
analysis. Although the color-coding information is very
useful, it is considered important to continue to use the
time-velocity tracings to correctly identify the peak veloci-
ties for LV dyssynchrony analysis. Left ventricular dyssyn-
chrony can be defined as difference in time to peak velocity
of opposing walls: inferoseptal to lateral wall (four-chamber
view), anterior to inferior wall (two-chamber view), and
anteroseptal to posterior wall (long-axis view). In a pilot
series of 29 patients, TSI was assessed before CRT, and
acute response (defined as an immediate increase in LV
stroke volume by 15% or more) was observed in 21 patients
(39). Differences in baseline TSI time to peak velocity of
opposing LV walls (three views average) were greater in
acute responders than non-responders: 120 ⫾ 148 ms versus
35 ⫾ 153 ms (p ⬍ 0.05). When delays between individual
walls were compared, dyssynchrony between the anterosep-
tal and posterior wall (assessed from the apical long-axis
JACC Vol. 46, No. 12, 2005 Bax et al. 2163
December 20, 2005:2153–67 Issues Before Device Implantation

Figure 8. Parametric polar-map displays (lower left and lower right panels) of left ventricular dyssynchrony of the real-time three-dimensional images;
blue indicates early activation, red indicates late activation. Left ventricular dyssynchrony is present before cardiac resynchronization therapy (CRT) in the
anteroseptal region (red, lower left panel); almost complete resynchronization has occurred after CRT, with disappearance of “red” regions (lower right).

view) had the greatest ability to separate acute responders
from non-responders after CRT. A cutoff value of ⱖ65 ms
had a sensitivity and specificity of 87% and 100%, respec-
tively, to predict acute response to CRT (39). The cutoff
value of 65 ms is in agreement with other results (20) using
color-coded TDI. In addition, a recent study by Yu et al.
(40) used TSI in 56 patients to predict LV reverse remod-
eling after three months of CRT, and a sensitivity of 87%
with a specificity of 81% was demonstrated.
Recent technological enhancements include multiplane
TSI imaging to keep frame rates high with three-
dimensional reconstruction of color-coded temporal LV
activation; imaged in the time-domain, this allows actual
four-dimensional information. Examples of four-dimensional
TSI in a normal individual and a patient with LV dyssyn-
chrony are demonstrated in Figures 5 and 6. Angle-
corrected tissue Doppler data for color-coded temporal
activation, known as Dyssynchrony Imaging (Toshiba, To-
kyo, Japan), also have the potential to provide radial
dyssynchrony data that may be additive to dyssynchrony
analysis of longitudinal velocities alone. In 38 patients under-
going CRT, Dohi et al. (38) showed a sensitivity of 95% with
a specificity of 88% to predict acute response to CRT using
radial dyssynchrony.
THREE-DIMENSIONAL ECHOCARDIOGRAPHY
TO ASSESS LV DYSSYNCHRONY
Three-dimensional echocardiography has evolved from a
technique based on reconstruction of multiple two-
dimensional scan planes to an almost real-time methodology.
Left ventricular volumes and LVEF can be assessed with
high accuracy (41). In the context of LV dyssynchrony,
analysis of regional function in the time-domain is impor-
tant (10), and a series of plots is obtained representing the
change in volume for each segment (usually 16 or 17
segments) throughout the cycle (Fig. 7). With synchronous
contraction of all segments, each segment would be ex-
pected to achieve the minimum volume at almost the same
point in the cardiac cycle. In LV dyssynchrony, dispersion
exists in the timing of the point of minimum volume for
each of the segments. The degree of dispersion reflects the
severity of LV dyssynchrony (42).
Parametric “polar-map” displays (of the three-dimensional
data) of the timing of LV contraction have been developed to
facilitate interpretation of data. This methodology examines
regional LV contraction at approximately 3,000 points over
the endocardial surface rather than in 16 or 17 segments.
Color-coding is used to identify the region/site of latest
2164 Bax et al. JACC Vol. 46, No. 12, 2005
Issues Before Device Implantation December 20, 2005:2153–67

Figure 9. Noninvasive multislice computed tomography of the venous anatomy. (Left) Three-dimensional volume rendered reconstruction. (Right)
Multiplanar curved reconstruction (MPR) of the coronary sinus (CS). Indicated on the three-dimensional reconstruction (left) are the CS, the posterior
interventricular vein (PIV), the posterior vein of the left ventricle (PVLV), and the great cardiac vein (GCV). LA ⫽ left atrium; LV ⫽ left ventricle; RA
⫽ right atrium; RV ⫽ right ventricle.

activation, and this is potentially useful for electrophysiolo-
gists to select the optimal LV lead position. Examples of
parametric polar-map images (of the three-dimensional
data) pre- and post-CRT are displayed in Figure 8. Zhang
et al. (43) used three-dimensional echocardiography in 13
patients who had previously received CRT; when CRT was
withheld, significant LV dyssynchrony occurred, associated
with a decrease in LVEF. Currently, no extensive data are
available on the prediction of response to CRT using
three-dimensional echocardiography.
OTHER FACTORS RELATED TO RESPONSE:
VENOUS ANATOMY AND SCAR TISSUE
Besides LV dyssynchrony, other factors are important for
success of CRT. In particular, the venous anatomy is
important. Considerable variability exists in the venous
system (44) as well as variability in the nomenclature used to
designate specific coronary veins. The authors evaluated
venous anatomy by retrograde venography in 129 consecu-
tive patients referred for implantable cardioverter defibril-
lator implantation. In 5 (4%) patients coronary sinus cannu-
lation failed, and in 38 (29%) visualization was suboptimal
and/or venograms were incomplete. In the remaining 86
patients, the anterior interventricular vein and middle car-
diac vein could be visualized in 99% and 100% of patients,
respectively. The veins (posterior or left marginal veins)
between these two veins are mostly used for LV lead
placement in CRT. Only one prominent vein was present in
51% of patients, two veins in 46% of patients, whereas more
than two veins were present in only 2% of patients. When
patients would be considered for CRT only in the presence
of the posterior veins, 55% would be acceptable. When
patients would be considered for CRT in the presence of the
posterior or the left marginal veins, 99% would be accept-
able. Ideally, venous anatomy should be assessed non-
invasively, at the outpatient clinic, to determine whether a
transvenous approach is feasible, or whether a (minimal
invasive) surgical approach should be used for LV lead

Figure 10. Resting single-photon emission computed tomography images (using technetium-99m tetrofosmin) of a patient with a previous inferopos-
terolateral infarction. A severe defect in tracer uptake is visible on the mid-ventricular short-axis slice (SA) (left) in the inferior and posterolateral regions,
which is confirmed on the horizontal long-axis (HLA) (middle) and vertical long-axis (VLA) (right) projections.
JACC Vol. 46, No. 12, 2005
December 20, 2005:2153–67
Figure 11. Contrast-enhanced cardiovascular magnetic resonance short-axis slices illustrating the presence and transmural extent of scar tissue. (Left)
Non-transmural scar tissue in the posterolateral wall as indicated by the hyperenhanced (white) region. (Right) Transmural scar formation in the
posterolateral region.
implantation. The feasibility of multislice computed tomog-
raphy (MSCT) to visualize the venous anatomy was recently
demonstrated (45). The variability in venous anatomy was
confirmed, and findings were in line with previous invasive
observations; MSCT allowed not only precise determina-
tion of coronary sinus and its tributaries, but also assessment
of distances between veins and ostial size of the coronary
sinus. An example of an MSCT depicting venous anatomy
is demonstrated in Figure 9.
Although information on venous anatomy is needed, it is
really the integrated information on the site of latest
activation (maximum LV dyssynchrony) and venous anat-
omy that determines the approach for LV lead implantation
(transvenous versus surgical). Accordingly, it would be
desired for example to fuse the three-dimensional TSI
images (Figs. 5 and 6) with the MSCT images (Fig. 9), in
order to co-registrate electromechanical delays with venous
anatomy.
Another factor that is important for success of CRT is
whether scar tissue is present in the region of latest
activation (where the LV lead should be positioned).
Currently, no solid data on this issue are available. One
anecdotal report, however, demonstrated that initial re-
sponse to CRT was reversed after acute infarction in the
territory of the LV lead (46). It would, therefore, be of
potential interest to evaluate non-invasively, before CRT
implantation, whether the target region for LV lead
positioning contains viable tissue or whether scar tissue is
present. Various techniques are available, including nu-
clear imaging techniques, echocardiographic techniques,
or cardiovascular magnetic resonance (CMR). Routine
resting single-photon emission computed tomography
imaging with a technetium-99m-labeled tracer would
provide the requested information, as indicated in Figure
10, showing a large defect in the inferior and posterolat-
eral regions, indicating scar tissue. Contrast-enhanced
CMR may eventually provide the optimal information,
Bax et al. 2165
Issues Before Device Implantation
because it allows precise depiction of the extent, and
transmurality of scar tissue (Fig. 11).
CONCLUSIONS
In large clinical trials, the beneficial effect of CRT has been
demonstrated. On an individual basis, however, 20% to 30%
of patients still do not respond to CRT. Observational
echocardiographic studies have clearly demonstrated that
the presence of LV dyssynchrony is an important factor
determining response to CRT, whereas interventricular
dyssynchrony appears of less importance. Various studies
using different echocardiographic approaches have subse-
quently aimed at prediction of response to CRT, and the
available literature suggests that TDI using 2-, 4-, or
12-segment models of LV dyssynchrony may represent the
best method to predict response to CRT. Careful analysis of
the literature also reveals the limitations of published
studies. The number of patients is small in most studies, a
lack of consensus exists in assessment of response (in
particular clinical parameters vs. echocardiographic param-
eters), the measurements of LV dyssynchrony are different
among studies, and direct comparisons between LV dyssyn-
chrony parameters to predict response are virtually absent.
The Predictors of Response to Cardiac Resynchronization
Therapy (PROSPECT) study is underway to examine if
prospectively defined echocardiographic parameters of sys-
tolic dyssynchrony are able to predict a favorable response to
CRT, and the latter includes both clinical composite end
points and LV reverse remodeling (47). Still, from the
existing evidence, it appears mandatory to expand current
guidelines for patient selection for CRT, and to include
assessment of LV dyssynchrony.
In addition to LV dyssynchrony, information on venous
anatomy and the presence of scar tissue may further opti-
mize response to CRT, but further studies are needed to
fully appreciate the clinical importance of these issues.
2166 Bax et al.
Issues Before Device Implantation
Reprint requests and correspondence: Dr. Jeroen J. Bax, Depart-
ment of Cardiology, Leiden University Medical Center, Albinusdreef
2, 2333 ZA Leiden, the Netherlands. E-mail: jbax@knoware.nl.
REFERENCES
1. Auricchio A, Stellbrink C, Sack S, et al., Pacing Therapies in
Congestive Heart Failure (PATH-CHF) Study Group. Long-term
clinical effect of hemodynamically optimized cardiac resynchronization
therapy in patients with heart failure and ventricular conduction delay.
J Am Coll Cardiol 2002;39:2026 –33.
2. Cazeau S, Leclercq C, Lavergne T, et al., Multisite Simulation in
Cardiomyopathies (MUSTIC) Study Investigators. Effects of multisite
biventricular pacing in patients with heart failure and intraventricular
conduction delay. N Engl J Med 2001;344:873– 80.
3. Abraham WT, Fisher WG, Smith AL, et al., MIRACLE Study
Group. Multicenter InSync Randomized Clinical Evaluation. Cardiac
resynchronization in chronic heart failure. N Engl J Med 2002;346:
1845–53.
4. Young JB, Abraham WT, Smith AL, et al., Multicenter InSync ICD
Randomized Clinical Evaluation (MIRACLE ICD) Trial Investiga-
tors. Combined cardiac resynchronization and implantable cardiover-
sion defibrillation in advanced chronic heart failure: the MIRACLE
ICD trial. JAMA 2003;289:2685–94.
5. Bristow MR, Saxon LA, Boehmer J, et al., Comparison of Medical
Therapy, Pacing and Defibrillation in Heart Failure (COMPANION)
Investigators. Cardiac resynchronization therapy with or without an
implantable defibrillator in advanced chronic heart failure. N Engl
J Med 2004;350:2140 –50.
6. Cleland J, Daubert JC, Erdmann E, et al. The effect of cardiac
resynchronization on morbidity and mortality in heart failure. N Engl
J Med 2005;352:1539 – 49.
7. Auricchio A, Stellbrink C, Butter C, et al., Pacing Therapies in
Congestive Heart Failure II Study Group, Guidant Heart Failure
Research Group. Clinical efficacy of cardiac resynchronization therapy
using left ventricular pacing in heart failure patients stratified by
severity of ventricular conduction delay. J Am Coll Cardiol 2003;42:
2109 –16.
8. Lozano I, Bocchiardo M, Achtelik M, et al., VENTAK CHF/
CONTAK CD Investigators Study Group. Impact of biventricular
pacing on mortality in a randomized crossover study of patients with
heart failure and ventricular arrhythmias. Pacing Clin Electrophysiol
2000;23:1711–2.
9. Bax JJ, Van der Wall EE, Schalij MJ. Cardiac resynchronization
therapy for heart failure. N Engl J Med 2002;347:1803– 4.
10. Bax JJ, Ansalone G, Breithardt OA, et al. Echocardiographic evalua-
tion of cardiac resynchronization therapy: ready for routine clinical
use? A critical appraisal. J Am Coll Cardiol 2004;44:1–9.
11. Yu CM, Fung JW, Zhang Q, et al. Tissue Doppler imaging is superior
to strain rate imaging and postsystolic shortening on the prediction of
reverse remodeling in both ischemic and nonischemic heart failure
after cardiac resynchronization therapy. Circulation 2004;110:66 –73.
12. Rouleau F, Merheb M, Geffroy S, et al. Echocardiographic assessment
of the interventricular delay of activation and correlation to the QRS
width in dilated cardiomyopathy. Pacing Clin Electrophysiol 2001;24:
1500 – 6.
13. Auricchio A, Stellbrink C, Butter C, et al. Clinical efficacy of cardiac
resynchronization therapy using left ventricular pacing in heart failure
patients stratified by severity of ventricular conduction delay. J Am
Coll Cardiol 2003;42:2109 –16.
14. Reuter S, Garrigue S, Barold SS, et al. Comparison of characteristics
in responders versus nonresponders with biventricular pacing for
drug-resistant congestive heart failure. Am J Cardiol 2002;89:346 –50.
15. Molhoek SG, Bax JJ, Van Erven L, et al. QRS duration and
shortening to predict clinical response to cardiac resynchronization
therapy in patients with end-stage heart failure. Pacing Clin Electro-
physiol 2004;27:308 –13.
16. Lecoq G, Leclercq C, Leray E, et al. Clinical and electrocardiographic
predictors of a positive response to cardiac resynchronization therapy
in advanced heart failure. Eur Heart J 2005;26:1094 –100.
JACC Vol. 46, No. 12, 2005
December 20, 2005:2153–67
17. Bleeker GA, Schalij MJ, Molhoek SG, et al. Relationship between
QRS duration and left ventricular dyssynchrony in patients with
end-stage heart failure. J Cardiovasc Electrophysiol 2004;15:544 –9.
18. Ghio S, Constantin C, Klersy C, et al. Interventricular and intraven-
tricular dyssynchrony are common in heart failure patients, regardless
of QRS duration. Eur Heart J 2004;25:571– 8.
19. Bordachar P, Lafitte S, Reuter S, et al. Echocardiographic parameters
of ventricular dyssynchrony validation in patients with heart failure
using sequential biventricular pacing. J Am Coll Cardiol 2004;44:
2157– 65.
20. Bax JJ, Bleeker GB, Marwick TH, et al. Left ventricular dyssynchrony
predicts response and prognosis after cardiac resynchronization ther-
apy. J Am Coll Cardiol 2004;44:1834 – 40.
21. Penicka M, Bartunek J, de Bruyne B, et al. Improvement of left
ventricular function after cardiac resynchronization therapy is pre-
dicted by tissue Doppler imaging echocardiography. Circulation 2004;
109:978 – 83.
22. Pitzalis MV, Iacoviello M, Romito R, et al. Cardiac resynchronization
therapy tailored by echocardiographic evaluation of ventricular asyn-
chrony. J Am Coll Cardiol 2002;40:1615–22.
23. Pitzalis MV, Iacoviello M, Romito R, et al. Ventricular asynchrony
predicts a better outcome in patients with chronic heart failure
receiving cardiac resynchronization therapy. J Am Coll Cardiol 2005;
45:65–9.
24. Marcus G, Rose E, Viloria EM, et al. Septal to posterior wall motion
delay fails to predict reverse remodeling or clinical improvement in
patients undergoing cardiac resynchronization therapy. J Am Coll
Cardiol 2005;46:2208 –14.
25. Breithardt OA, Stellbrink C, Kramer AP, et al. Echocardiographic
quantification of left ventricular asynchrony predicts an acute hemo-
dynamic benefit of cardiac resynchronization therapy. J Am Coll
Cardiol 2002;40:536 – 45.
26. Kawaguchi M, Murabayashi T, Fetics BJ, et al. Quantitation of basal
dyssynchrony and acute resynchronization from left or biventricular
pacing by novel echo-contrast variability imaging. J Am Coll Cardiol
2002;19:2052– 8.
27. Ansalone G, Giannantoni P, Ricci R, et al. Doppler myocardial
imaging in patients with heart failure receiving biventricular pacing
treatment. Am Heart J 2001;142:881–96.
28. Garrigue S, Reuter S, Labeque JN, et al. Usefulness of biventricular
pacing in patients with congestive heart failure and right bundle
branch block. Am J Cardiol 2001;88:1436 – 41.
29. Yu CM, Chau E, Sanderson JE, et al. Tissue Doppler echocardio-
graphic evidence of reverse remodeling and improved synchronicity by
simultaneously delaying regional contraction after biventricular pacing
therapy in heart failure. Circulation 2002;105:438 – 45.
30. Bax JJ, Marwick TH, Molhoek SG, et al. Left ventricular dyssyn-
chrony predicts benefit of cardiac resynchronization therapy in patients
with end-stage heart failure before pacemaker implantation. Am J
Cardiol 2003;92:1238 – 40.
31. Yu CM, Fung WH, Lin H, Zhang Q, Sanderson JE, Lau CP.
Predictors of left ventricular reverse remodeling after cardiac resyn-
chronization therapy for heart failure secondary to idiopathic dilated or
ischemic cardiomyopathy. Am J Cardiol 2003;91:684 – 8.
32. Notabartolo D, Merlino JD, Smith AL, et al. Usefulness of the peak
velocity difference by tissue Doppler imaging technique as an effective
predictor of response to cardiac resynchronization therapy. Am J
Cardiol 2004;94:817–20.
33. Sogaard P, Egeblad H, Kim WY, et al. Tissue Doppler imaging
predicts improved systolic performance and reversed left ventricular
remodeling during long-term cardiac resynchronization therapy. J Am
Coll Cardiol 2002;40:723–30.
34. Sogaard P, Egeblad H, Pedersen AK, et al. Sequential versus simul-
taneous biventricular resynchronization for severe heart failure: evalu-
ation by tissue Doppler imaging. Circulation 2002;106:2078 – 84.
35. Breithardt OA, Stellbrink C, Herbots L, et al. Cardiac resynchroni-
zation therapy can reverse abnormal myocardial strain distribution in
patients with heart failure and left bundle branch block. J Am Coll
Cardiol 2003;42:486 –94.
36. Sun JP, Chinchoy E, Donal E, et al. Evaluation of ventricular
synchrony using novel Doppler echocardiographic indices in patients
with heart failure receiving cardiac resynchronization therapy. J Am
Soc Echocardiogr 2004;17:845–50.
JACC Vol. 46, No. 12, 2005
December 20, 2005:2153–67
37. Popovic ZB, Grimm RA, Perlic G, et al. Noninvasive assessment of
cardiac resynchronization therapy for congestive heart failure using myo-
cardial strain and left ventricular peak power as parameters of myocardial
synchrony and function. J Cardiovasc Electrophysiol 2002;13:1203– 8.
38. Dohi K, Suffoletto MS, Schwartzman D, Ganz L, Pinsky MR,
Gorcsan J, III. Utility of echocardiographic radial strain imaging to
quantify left ventricular dyssynchrony and predict acute response to
cardiac resynchronization therapy. Am J Cardiol 2005;96:112– 6.
39. Gorcsan J, III, Kanzaki H, Bazaz R, Dohi K, Schwartzman D.
Usefulness of echocardiographic tissue synchronization imaging to
predict acute response to cardiac resynchronization therapy. Am J
Cardiol 2004;93:1178 – 81.
40. Yu CM, Zhang Q, Fung JW, et al. A novel tool to assess systolic
asynchrony and identify responders of cardiac resynchronization ther-
apy by tissue synchronization imaging. J Am Coll Cardiol 2005;45:
677– 84.
41. Kühl HP, Schreckenberg M, Rulands D, et al. High-resolution
transthoracic real-time three-dimensional echocardiography: quantita-
tion of cardiac volumes and function using semi-automatic border
detection and comparison with cardiac magnetic resonance imaging.
J Am Coll Cardiol 2004;43:2083–90.
Bax et al. 2167
Issues Before Device Implantation
42. Kapetenakis S, Siva A, Corrigan N, Cooklin M, Kearney MT,
Monaghan MJ. Real-time three-dimensional echocardiography: a
novel technique to quantify global left ventricular mechanical dyssyn-
chrony. Circulation 2005;112:992–1000.
43. Zhang Q, Yu CM, Fung JW, et al. Assessment of the effect of
cardiac resynchronization therapy on intraventricular mechanical
synchronicity by regional volumetric changes. Am J Cardiol 2005;
95:126 –9.
44. Meisel E, Pfeiffer D, Engelmann L, et al. Investigation of coronary
venous anatomy by retrograde venography in patients with malignant
ventricular tachycardia. Circulation 2001;104:442–7.
45. Jongbloed MR, Lamb HJ, Bax JJ, et al. Noninvasive visualization of
the cardiac venous system using multislice computed tomography.
J Am Coll Cardiol 2005;45:749 –53.
46. Kanhai SM, Viergever EP, Bax JJ. Cardiogenic shock shortly after
initial success of cardiac resynchronization therapy. Eur J Heart Fail
2004;6:477– 81.
47. Yu CM, Abraham WT, Bax JJ, et al. Predictors of response to cardiac
resynchronization therapy (PROSPECT)-study design. Am Heart J
2005;149:600 –5.