Hmptsuaispncbtm

ma

0d

Relation of Optimal Lead Positioning as Defined byThree-Dimensional Echocardiography to Long-Term Benefit of

Cardiac ResynchronizationMichael Becker, MD*, Rainer Hoffmann, MD, Fabian Schmitz, MD, Anne Hundemer, MD,

Harald Kühl, MD, Patrick Schauerte, MD, Malte Kelm, MD, and Andreas Franke, MD

We sought to define the impact of echocardiographically defined left ventricular (LV) leadposition on the efficacy of cardiac resynchronization therapy (CRT) in a serial study using3-dimensional echocardiography. Fifty-eight consecutive patients (53 � 9 years of age; 37men) with heart failure were included in the study. Echocardiograms were obtained beforeCRT, within 7 days after implantation, and at 12 � 2 months of follow-up using a3-dimensional digital ultrasound scanner (iE33, Philips, Andover, Massachusetts). Analysisof the temporal course of contraction in 16 LV segments was performed offline using asemiautomatic contour tracing software (LV Analysis, TomTec, Unterschleissheim, Ger-many). Based on the resulting volume/time curves the segment with the latest minimum ofsystolic volume in each patient was identified preoperatively (segment A). In addition, thetemporal difference between the pre- and postoperative (within 7 days) minimum ofsystolic volume was determined for each segment. The segment with the longest temporaldifference was defined to show the greatest effect of CRT. Location of the LV lead tip wasassumed to be within this segment (segment B). LV lead position was defined as optimalwhen segments A and B were equal and as nonoptimal when they were far from each other.Using this definition, 26 patients had a nonoptimal and 32 patients an optimal LV lead position.Before CRT ejection fraction (32 � 4% vs 31 � 6%), LV end-systolic and end-diastolic volumes(242 � 92 vs 246 � 88 ml, 315 � 82 vs 323 � 90 ml), and peak oxygen consumption (14.3 �1.4 vs 14.6 � 1.5 ml/min/kg) were equal in the 2 groups. At 12 � 2 months of follow-up, patientswith an assumed optimal LV lead position showed greater increases of ejection fraction (10 �2% vs 6 � 3%) and peak oxygen consumption (2.4 � 0.3 vs 1.5 � 0.4 ml/min/kg) and greaterdecreases of LV end-systolic (32 � 7 vs 21 � 5 ml) and end-diastolic (20 � 7 vs 13 � 6 ml)volumes. In conclusion, correspondence of the segment with the latest preoperative LV con-traction with the segment with the greatest effect based on CRT results in a significantly greaterbenefit of ejection fraction and peak oxygen consumption and a greater improvement in LVremodeling. Thus, there is an optimal LV lead position that should be obtained. © 2007

Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:1671–1676)

rmcnppc

M

FfmAcectfipA

eart failure is associated with substantial mortality andorbidity and remains the most common diagnosis at hos-

ital admission in older patients. Cardiac resynchronizationherapy (CRT) has been proved beneficial for patients withymptomatic chronic heart failure associated with left ventric-lar (LV) asynchrony.1–10 Because as many as 1/3 of patientsre inadequate responders to CRT, several studies have tried todentify predictors of a positive response to CRT.11–14 Previoustudies have analyzed the effect of CRT on echocardiographicarameters, focusing on the identification of patients with sig-ificant mechanical asynchrony.12,13 Dependence of CRT suc-ess on technical factors, in particular LV lead position, haseen less extensively evaluated. Based on experimental elec-rophysiologic studies,15 a LV lead position in the area of the

ost recent electromechanical activation before CRT should

Department of Cardiology, University RWTH Aachen, Aachen, Ger-any. Manuscript received March 20, 2007; revised manuscript received

nd accepted July 1, 2007.*Corresponding author: Tel: 49-241-808-9301; fax: 49-241-808-2303.

tE-mail address: mibecker@ukaachen.de (M. Becker).

002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved.oi:10.1016/j.amjcard.2007.07.019

esult in a maximum resynchronization effect. However, place-ent of the LV lead is often not selectable because of anatomic

onditions.16 Impossibility to reach a desired vein or scar tissueot allowing sufficient electrical response in the desired leadosition is the most seen technical problem. The aim of theresent study was to assess the impact of an echocardiographi-ally defined LV lead position on the efficacy of CRT.

ethods

or this study, 67 consecutive patients with end-stage heartailure scheduled for implantation of a biventricular pace-aker were screened from June 2004 to December 2005.ll patients were in New York Heart Association functional

lass III or IV despite optimal pharmacologic therapy, anjection fraction �35%, and a QRS width �120 ms asriteria for CRT implantation. Three patients refused par-icipation in the present study, and 6 patients with atrialbrillation were excluded from the study. The remaining 58atients (53 � 9 years of age; 37 men; New York Heartssociation classes III [n � 40] and IV [n � 18]) formed

he study group and gave written informed consent. Thirty-

www.AJConline.org

noiT

ivlirbtvalsVDt

fiiatp

dmudtAeoTigo

Fcica of CRT

1672 The American Journal of Cardiology (www.AJConline.org)

ine patients had a myocardial infarction within the previ-us year (13 anterior, 9 lateral, 10 inferior, 7 posterior), andn 19 patients the cause of heart failure was nonischemic.his study was approved by the local ethical committee.

The LV pacing lead was placed through the coronary sinusnto a cardiac vein of the free wall. An average of 2.3 � 0.4eins was evaluated intraoperatively to reach an optimal LVead position. The aim was a preferably short and reproduc-ble width of the QRS complex and a preferably high andeproducible increase in arterial systolic blood pressure withiventricular pacing. In addition, adequate pacing parame-ers (sensing and pacing threshold) were required. The rightentricular pacing lead was implanted transvenously at thepex. Patients received a biventricular cardioverter-defibril-ator (Attain-System with InSync Marquis, Medtronic, Dus-eldorf, Germany n � 37; or Aescula-System with Epic HF-339, St. Jude Medical, Eschborn, Germany n � 21).oppler echocardiography guaranteed an optimal atrioven-

igure 1. Top, volume/time curve of each segment before CRT is displayed. Thontraction was identified before CRT and defined as the segment with maxims displayed. For each segment the temporal difference (�ts) between pre-ontraction delay with resynchronization therapy compared with before therapmong all 16 LV segments was defined as the segment with greatest benefit

ricular interval by obtaining maximal transmitral diastolic L

lling without premature termination of atrial filling. Thenterventricular timing was set to 0 in all patients. There wasnother control of the device at 6 � 1 months of follow-upo ensure that no LV lead dislocation or change of the CRTrogramming existed.

Echocardiograms were obtained before CRT, within 7ays after implantation of the CRT system, and at 12 � 2onths of follow-up using a 3-dimensional (3-D) digital

ltrasound scanner (iE33 and Sonos 7500 by Philips, An-over, Massachusetts). A full-volume loop of the left ven-ricle was acquired using an apical position of the probe.nalysis of LV ejection fraction and LV end-systolic and

nd-diastolic volumes was determined offline with the aidf semiautomatic contour tracing software (LV Analysis,omTec, Unterschleissheim, Germany) using and compar-

ng preimplantation and 12-month follow-up echocardio-rams. This technique has been described in detail previ-usly.17 In addition, the temporal course of contraction in 16

nt with the latest minimum of systolic volume as an indicator for latest systolicnchrony (segment A). Bottom, volume/time curve of each segment with CRTstoperative minimum of systolic volume was determined. The decrease inis segment is shown (�ts) (red arrow). The segment showing the greatest �tsand assumed to be the segment with the LV lead position (segment B).

e segmeum asyand poy for th

V segments was analyzed using this program. Based on

rpvimsawmlpwoatllr

tmwde

oobopqfll(ac

af

pCgFv1tdgtttAlnm9

R

I(wsla5aspa3p

lma

Fao

1673Heart Failure/Lead Position–Dependent Remodeling in CRT

esulting segmental volume/time curves (Figure 1), in eachatient, the segment with the latest minimum of systolicolume as an indicator for latest systolic contraction wasdentified preoperatively and defined as the segment withaximum mechanical delay (segment A). In addition, the

egment with the largest temporal difference between pre-nd postoperative minimum of systolic volume (Figure 1)as determined and defined as the segment with greatestechanical effect of CRT (segment B). Location of the LV

ead position was assumed to be in this segment. LV leadosition was defined as optimal when segments A and Bere equal and as nonoptimal when they were far from eachther. The physician performing echocardiographic analysisnd classification of LV lead position as optimal or nonop-imal was blinded against possible difficulties during LVead placement and anatomic limitations. Vice versa, LVead implantation was performed by a physician blinded toesults of the echocardiographic examination.

All patients underwent bicycle cardiopulmonary exerciseesting (10 W/min increments) at baseline and after 12 � 2onths of CRT. Peak oxygen consumption at peak exerciseas defined as the highest oxygen consumption measureduring the last step of the symptom-limited exercise test andxpressed as milliliters per kilogram per minute.

After CRT, implantation biplane fluoroscopy in orthog-nal views (left anterior oblique at 60° and right anteriorblique at 30°) was performed. These images were analyzedy 2 independent readers to determine the anatomic locationf the LV lead (interobserver variability 8%). For that pur-ose a resized 16-segment schema (modified from Cer-ueira et al18) was projected onto the left anterior obliqueuoroscopic image. In addition, the right anterior oblique

evel was divided into basal, medial, and apical sectionsFigure 2). Physicians who evaluated the fluoroscopic im-ge were blinded to the physician who obtained the echo-ardiographically assumed LV lead position.

Outcomes were measured at baseline and at 12 monthsfter CRT implantation. Outcome was change in ejection

igure 2. To determine anatomic LV lead position, fluoroscopic orthogonalfter CRT implantation. Right, for that purpose, the resized 16-segment sblique level was divided into basal, medial, and apical sections.

raction, LV end-systolic and end-diastolic volumes, and b

eak oxygen consumption from baseline to 12 months.ontinuous variables are presented as mean � SD. Cate-orical data are presented as frequencies and percentages.or analysis of outcome a repeated measures analysis ofariance was performed and included values at baseline and2 months as dependent variables and the fixed effect ofime, LV lead group, and time-by-group interaction as in-ependent variables. Model assumptions were checkedraphically by qq plots. We calculated a Cohen � coefficiento evaluate the agreement between anatomic LV lead posi-ion determined by fluoroscopy and assumed LV lead posi-ion determined by detailed circumferential strain analysis.ll tests were 2-sided and assessed at the 5% significance

evel. Because of the exploratory nature of the end points,o adjustment was made to significance level to account forultiple testing. All analyses were performed with SAS

.1.3 (SAS Institute, Cary, North Carolina).

esults

n 32 patients, the location of latest contraction before CRTsegment A, the segment with maximum mechanical delay)as equal with the largest effect of CRT (segment B, the

egment with the assumed LV lead position). Here the LVead position was defined as “optimal.” The LV lead wasssumed to be in the following segments: 16 lateral (8 basal,medial, and 3 apical), 9 anterior (5 basal, 3 medial, and 1

pical), and 7 posterior (4 basal and 3 medial). In 26 patientsegment B did not match with segment A. Here the LV leadosition was defined as “nonoptimal.” The LV lead wasssumed to be in the following segments: 10 lateral (6 basal,medial, and 1 apical), 8 anterior (5 basal and 3 apical), 6

osterior (3 basal and 3 medial), and 2 inferior (2 basal).Location of the segment with maximum mechanical de-

ay before CRT was as follows: 27 lateral (12 basal, 11edial, and 4 apical), 18 anterior (10 basal, 4 medial, and 4

pical), 9 posterior (5 basal and 4 medial), and 4 inferior (2

right anterior oblique at 30° and left anterior oblique at 60°) were acquiredwas projected onto the left anterior oblique level. Left, the right anterior

views (chema

asal and 2 medial). Distribution of these segments between

pn

cablffcpn

bCmdcicwicBdw

3a

ngv

otpmpaIemipl

D

TdcctdCfp

TCa

V

AMIQEPLLC

%)

TC

����

1674 The American Journal of Cardiology (www.AJConline.org)

atients with an assumed optimal LV lead position andonoptimal LV lead position was not different.

At baseline before CRT there were no differences inlinical characteristics, ejection fraction, LV end-systolicnd end-diastolic volumes, and peak oxygen consumptionetween patients with optimal and those with nonoptimaleft LV position (Table 1). Comparison of baseline withollow-up results demonstrated greater increases in ejectionraction and peak oxygen consumption and greater de-reases in LV end-systolic and end-diastolic volumes inatients with an optimal LV lead position compared with aonoptimal LV lead position (Table 2).

A more detailed analysis on the impact of the distanceetween the segment with latest mechanical delay beforeRT (segment A) and the assumed LV lead segment (seg-ent B) in patients with a nonoptimal LV lead position

emonstrated a decreasing effectiveness of CRT with in-reasing distance. Improvement of ejection fraction andncrease of peak oxygen consumption decreased with in-reasing distance between segments A and B. Comparedith patients with optimal LV lead placement, improvement

n ejection fraction and peak oxygen consumption was de-reased by 19% and accordingly 21% when segments A and

were adjacent, by 34% and accordingly 29% with aistance of 1 segment, and by 52% and accordingly 41%ith a distance of �2 segments.There was 1 patient with an optimal LV lead position and

patients with a nonoptimal LV lead who were categorized

able 1linical characteristics, ejection fraction, left ventricular end-systolic andnd nonoptimal left ventricular lead positions

ariable Optimal LV Le(n � 3

ge (yrs) 55.6 �en 20 (63

schemic cardiomyopathy 21 (66RS duration (ms) 161 �jection fraction (%) 31 �eak oxygen consumption (ml/kg/min) 14.6 �V end-sysolic volume (ml) 246 �V end-diastolic volume (ml) 323 �oncomitant therapyAngiotensin-converting enzyme inhibitors 24 (76Angiotensin II receptor blockers 4 (12� Blockers 27 (84Digitalis 12 (36Diuretics 19 (60Aldosterone antagonists 18 (56

able 2omparison of baseline and follow-up results in patients with optimal ver

Optimal LV Lead(n � 32

Ejection fraction (%) 10 � 2Peak oxygen consumption (ml/kg/min) 2.4 � 0.LV end-systolic volume (ml) 32 � 7LV end-diastolic volume (ml) 20 � 7

� � difference.

s inadequate responders based on clinical information with

o subjective clinical benefit and based on echocardio-raphic information with a decrease in LV end-systolicolume of �15%.19

Anatomic position could be determined by fluoroscopy inrthogonal views. In 54 of 58 patients, the assumed LV posi-ion by echocardiography corresponded to anatomic LV leadosition by fluoroscopy; 25 patients showed a lateral place-ent of the LV lead (14 basal, 8 medial, and 3 apical), in 16

atients there was a posterior position (10 basal and 6 medial),nd in 17 patients an anterior position (9 basal and 8 medial).n 4 patients the anatomic position did not match with thechocardiographic position; there was a distance of 1.1 seg-ents between these results. All these patients were graded

nto the group with a nonoptimal LV lead position; 1 of theseatients would have been classified as having an optimal LVead position using the location determined by fluoroscopy.

iscussion

he major findings of this study are that (1) 3-D echocar-iography allows evaluation of changes in the temporalourse of segmental LV contraction with CRT, and (2)onformity of the segment with an assumed LV lead posi-ion based on greatest temporal difference in contractionelay and segment with the latest systolic contraction beforeRT results in significantly greater improvement in ejection

raction, LV remodeling, and cardiopulmonary exercise ca-acity with CRT.

stolic volumes, and peak oxygen consumption in patients with optimal

ition Nonoptimal LV Lead Position p Value(n � 26)

51.4 � 8 0.0817 (65%) 0.1118 (69%) 0.27158 � 18 0.4632 � 4 0.21

14.3 � 1.4 0.77242 � 92 0.18315 � 82 0.33

21 (79%) 0.183 (11%) 0.33

22 (84%) 0.0810 (37%) 0.6216 (63%) 0.0915 (58%) 0.10

optimal left ventricular lead position

on Nonoptimal LV Lead Position(n � 26)

p Value

6 � 3 �0.011.5 � 0.4 �0.0121 � 5 �0.0113 � 6 �0.01

end-dia

ad Pos2)

9%)%)1961.58890

%)%)%)%)%)

sus non

Positi)

3

Analysis of the temporal course of LV contraction has

bshdraAtcttisoTa

CusiaCtphswmsasdnsclrLrtptetnea

esaatotesttt

diwtiIdtp

1

1

1

1675Heart Failure/Lead Position–Dependent Remodeling in CRT

een an important element of multiple echocardiographictudies.20–25 Tissue Doppler echocardiographic methodsave been especially applied to describe and assess theegree of asynchrony in LV contraction. This approachefers to longitudinal motion, which may be less sensitive tosynchrony than circumferential deformation analysis.26

pplied 3-D echocardiographic analysis allows determina-ion of segmental volume, a parameter that relies on cir-umferential, radial, and longitudinal motions. In addition,his system enables analysis of the entire LV segment func-ion using 4 consecutive heart beats with the same RRntervals. This is in contrast to analysis of single points inpace using serial heart beats with different RR intervals inther methods that can lead to nonconformity of results.hus, this technique may be optimal for analysis of cardiacsynchrony in patients with planned CRT.

Numerous studies have demonstrated the efficacy ofRT in treatment of patients with advanced heart fail-re.3,4,6 The rate of approximately 30% of inadequate re-ponders remains an unsolved problem. One approach tomprove outcome may be determination of the degree ofsynchrony before CRT as a predictor for CRT response.onversely, the focus may be on an improved positioning of

he LV lead. The current definition of an optimal LV leadosition in many centers is based on invasively determinedemodynamic effects and shortening of QRS duration mea-ured intraoperatively. However, measurement of QRSidth as a surrogate for an optimal LV lead position is aatter of debate because some studies1 have shown that LV

timulation can be as effective as biventricular stimulation,lthough decrease of QRS width is usually less. In contrast,ome groups have found that a maximal decrease of QRSuration may be an electrical marker of optimal resynchro-ization.27 Our study demonstrates that concordance of theegment with the greatest effect based on CRT-relatedhanges in the contraction with CRT and the segment withatest contraction before CRT results in the best functionalesponse. Thus, it should be the aim to obtain this optimalV lead position to minimize the number of inadequate

esponders to a CRT treatment. Certainly it must be men-ioned that there are some factors that may limit optimallacement: impossibility to reach a desired vein or to anchorhe LV lead in the desired place and scar tissue not allowinglectrical response are the most seen problems. In view ofhese technical difficulties, this study indicates that even aeighborhood to the optimal LV lead site results in greaterffectiveness of the CRT system. Thus, every effort tochieve a most optimal LV lead position is desirable.

Some other limitations must be mentioned. Comparingchocardiograms before CRT with those after CRT, theegment with the maximal temporal difference between pre-nd postoperative systolic contraction was determined. An-tomic location of the LV lead position was assumed withinhis segment. The definition was verified by analyzing flu-roscopic loops in orthogonal views with a rate of misin-erpretation about 8%. This approach, although showingxcellent agreement, can only be an approximation; furthertudies using 3-D computed tomography have to confirmhese results. The 3-D echocardiography used has pooremporal (33 to 40 ms) and spatial (16 segments) resolu-

ions. Regarding the LV lead position, we defined concor-

ance of the segment with the maximal temporal differencen systolic contraction before CRT with CRT to the segmentith the latest systolic contraction as optimal. This defini-

ion of optimal is arbitrary. This introduced analysis methods quite fast but actually not feasible in the operating theater.t is desirable to optimize this procedure by further technicalevelopments to achieve a helpful clinical tool for applica-ion during LV lead implantation for the best effect inatients after CRT.

1. Auricchio A, Stellbrink C, Sack S, Vogt J, Bocker D, Block M, KirkelsJH, Ramdat-Misier A. Pacing Therapies in Congestive Heart Failure(PATH-CHF) Study Group. Long-term clinical effect of hemo-dynam-ically optimized cardiac resynchronization therapy in patients withheart failure and ventricular conduction delay. J Am Coll Cardiol2002;39:2026–2033.

2. Linde C, Leclercq C, Rex S, Garrigue S, Lavergne T, Cazeau S,McKenna W, Fitzgerald M, Deharo JC, Alonso C, et al. Long-termbenefits of biventricular pacing in congestive heart failure: results fromthe Multisite Stimulation In Cardiomyopathy (MUSTIC) study. J AmColl Cardiol 2002;40:111–118.

3. Bristow MR, Saxon LA, Boehmer J, Krueger S, Krass DA, DeMarcoT, Carson P, DiCarlo L, DeMets D, White BG, DeVries DW, FeldmanAM. Comparison of Medical Therapy, Pacing, and Defibrillation inHeart Failure (COMPANION) Investigators. Cardiac-resynchroniza-tion therapy with or without an implantable defibrillator in advancedchronic heart failure. N Engl J Med 2004;350:2140–2150.

4. Cazeau S, Leclercq C, Lavergne T, Walker S, Varma C, Linde C,Garrique S, Kappenberger L, Haywood GA, Santini M, Bailleul C,Daubert JC. Multisite Stimulation in Cardiomyopathies (MUSTIC)Study Investigators. Effects of multisite biventricular pacing in pa-tients with heart failure and intraventricular conduction delay. N EnglJ Med 2001;344:873–880.

5. Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E,Kocovic DZ, Packer M, Clavell AL, Hayes DL, et al. MIRACLEStudy Group. Multicenter InSync Randomized Clinical Evaluation.Cardiac resynchronization in chronic heart failure. N Engl J Med2002;346:1845–1853.

6. Young JB, Abraham WT, Smith AL, Leon AR, Liberman R, WilkhoffB, Canby RC, Schroeder JS, Liem LB, Hall S, Wheelan K. MulticenterInSync ICD Randomized Clinical Evaluation (MIRACLE ICD) TrialInvestigators. Combined cardiac resynchronization and implantable car-dioversion defibrillation in advanced chronic heart failure: the MIRACLEICD trial. JAMA 2003;289:2685–2694.

7. Leclerq C, Cazeau S, Le Breton H, Ritter P, Mabo P, Gras D, Pavin D,Lazarus A, Daubert JC. Acute hemodynamic effects of biventricularDDD pacing in patients with end-stage heart failure. J Am Coll Cardiol1998;32:1825–1831.

8. Kass DA, Chen CH, Curry C, Talbot M, Berger R, Fetics B, Nevo E.Improved left ventricular mechanics drom acute VDD pacing in pa-tients with dilated cardiomyopathy and ventricular conduction delay.Circulation 1999;99:1567–1573.

9. Gras D, Mabo P, Tang T, Luttikuis O, Chatoor R, Pedersen AK,Tscheliessnigg HH, Deharo JC, Puglisi A, Silvestre J, et al. Multisitepacing as a supplemental treatment of congestive heart failure: pre-liminary results of the Medtronic Inc in sync study pacing. ClinElectrophysiol 1998;21:2249–2255.

0. Cazeau S, Leclercq C, Lavergne T, Walker S, Varma C, Linde C,Garrigue S, Kappenberger L, Haywood GA, Santini M, Bailleul C, Daub-ert JC, for the Multisite Stimulation In Cardiomyopathies (MUSTIC)Study Investigators. Effects of multisite biventricular pacing in patientswith heart failure and intraventricular conduction delay. N Engl J Med2001;344:873–880.

1. Molhoek SG, Van Erven L, Bootsma M, Steendijk P, Van Der WallEE, Schalij MJ. QRS duration and shortening to predict clinical re-sponse to cardiac resynchronization therapy in patients with end-stageheart failure. Pacing Clin Electrophysiol 2004;27:308–313.

2. Bax JJ, Ansalone G, Breithardt OA, Derumeaux G, Leclercq C, SchalijMD, Soogard P, St John Sutton M, Nihoyannopoulos P. Echocardio-graphic evaluation of resynchronization therapy: ready for routine

clinical use? A critical appraisal. J Am Coll Cardiol 2004;44:1–9.

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

1676 The American Journal of Cardiology (www.AJConline.org)

3. Pitzalis MV, Iacoviello M, Romito R, Guida P, De Tommasi M, LuzziG, Anaclerio M, Forleo C, Rizzon P. Ventricular asynchrony predictsa better outcome in patients with chronic heart failure receiving cardiacresynchronization therapy. J Am Coll Cardiol 2005;45:65–69.

4. Auricchio A, Stellbrink C, Butter C, Vogt J, Misier AR, Bocker D,Kirkels JH, Kramer A, Huvelle E. Clinical efficacy of cardiac resyn-chronization therapy using left ventricular pacing in heart failurepatients stratified by severity of ventricular conduction delay. J AmColl Cardiol 2003;42:2109–2116.

5. Auricchio A, Fantoni C, Regoli F, Carbucicchio C, Goette A, Geller C,Kloss M, Klein H. Characterization of left ventricular activation inpatients with heart failure and left bundle-branch block. Circulation2004;109:1133–1139.

6. Singh JP, Houser S, Heist EK, Ruskin JN. The coronary venousanatomy. A segmental approach to aid cardiac resynchronization ther-apy. J Am Coll Cardiol 2005;46:68–74.

7. Kuhl HP, Schreckenberg M, Rulands D, Katoh M, Schafer W, Schum-mers G, Bucker A, Hanrath P, Franke A. High-resolution transthoracicreal-time three-dimensional echocardiography: quantitation of cardiacvolumes and function using semi-automatic border detection and com-parison with cardiac magnetic resonance imaging. J Am Coll Cardiol2004;43:2083–2090.

8. Cerquera MD, Weismann NJ, Dilsizian V, Jacobs AK, Kaul S, LaskeyWK, Pennell DJ, Rumberger JA, Ryan T, Verani MS. American HeartAssociation writing group on myocardial segmentation and registra-tion for cardiac imaging. Standardized myocardial segmentation andnomenclature for tomographic imaging of the heart: a statement forhealthcare professionals from the Cardiac Imaging Committee of theCouncil on Clinical Cardiology of the American Heart Association.Circulation 2002;105:539–542.

9. Bleeker GB, Bax JJ, Fung JW, van der Wall EE, Zhang Q, Schlij MJ,Chan JY, Yu CM. Clinical versus echocardiographic parameters toassess response to cardiac resynchronization therapy. Am J Cardiol

2006;97:260–263.

0. Ansalone G, Giannantoni P, Ricci R, Trambaiolo P, Fedele F, SantiniM. Doppler myocardial imaging to evaluate the effectiveness of pacingsites in patients receiving biventricular pacing. J Am Coll Cardiol2002;39:489–499.

1. Yu CM, Chau E, Sanderson JE, Fan K, Tang MO, Fung WH, Lin H,Kong SL, Lam YM, Hill MR, Lau CP. Tissue Doppler echocardio-graphic evidence of reverse remodeling and improved synchronicityby simultaneously delaying regional contraction after biventricularpacing therapy in heart failure. Circulation 2002;105:438–445.

2. Kapetanakis S, Kearney MT, Siva A, Gall N, Cooklin M, MonaghanMJ. Real-time three-dimensional echocardiography. A novel tech-nique to quantify global left ventricular mechanical dyssynchrony.Circulation 2005;112:992–1000.

3. Capasso F, Guinta A, Stabile G, Turco P, La Rocca V, Grimaldi G, DeSimone A. Left ventricular functional recovery during cardiac resyn-chronization therapy: predictive role of asynchrony measured by strainrate analysis. Pacing Clin Electrophysiol 2005;28(suppl 1):1–4.

4. Breithardt OA, Stellbrink C, Franke A, Balta O, Diem BH, BakkerP, Sack S, Auricchio H, Pochet T, Salo R. Acute effects of cardiacresynchronization therapy on left ventricular Doppler indices inpatients with congestive heart failure. Am Heart J 2002;143:34 – 44.

5. Sogaard P, Egeblad H, Kim WY, Jensen HK, Pedersen AK, KristensenBO, Mortensen PT. Tissue Doppler imaging predicts improved systolicperformance and reversed left ventricular remodeling during long-termresynchronization therapy. J Am Coll Cardiol 2002;40:723–730.

6. Helm RH, Leclercq C, Faris OP, Ozturk C, McVeigh E, Lardo AC,Kass DA. Cardiac dyssynchrony analysis using circumferential versuslongitudinal strain: implications for assessing cardiac resynchroniza-tion. Circulation 2005;111:2760–2767.

7. Lecoq G, Leclercq C, Leray E, Crocq C, Alonso C, de Place C, MaboP, Daubert C. Clinical and electrocardiographic predictors of a positiveresponse to cardiac resynchronization therapy in advanced heart fail-

ure. Eur Heart J 2005;26:1094–1100.