J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7

ª 2 0 1 7 B Y T H E A M E R I C A N CO L L E G E O F C A R D I O L O G Y F O U N DA T I O N

P U B L I S H E D B Y E L S E V I E R

I S S N 2 4 0 5 - 5 0 0 X / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c e p . 2 0 1 7 . 0 1 . 0 1 5

16 Years of Cardiac ResynchronizationPacing Among Congenital HeartDisease PatientsDirect Contractility (dP/dt-max) Screening When theGuidelines Do Not Apply

Peter P. Karpawich, MSC, MD, Neha Bansal, MD, Sharmeen Samuel, MD, Yamuna Sanil, MD,Kathleen Zelin, MSN, CPNP

ABSTRACT

Fro

Me

dis

All

ins

vis

Ma

OBJECTIVES The purpose of this study was to use direct cardiac resynchronization therapy (CRT)-paced contractility

(dP/dt-max) response as a pre-implantation evaluation among patients with congenital heart disease (CHD) and follow

clinical parameters and contractility indexes after CRT implantation.

BACKGROUND Patients with CHD often develop early heart failure with few therapeutic options, leading to heart

transplantation (HT). Unfortunately, guidelines for CRT do not apply, and function evaluations by cardiac ultrasound are

often inaccurate among CHD anatomies. Therefore, which CHD patients would benefit from CRT remains an enigma.

METHODS From 1999 to 2015, 103 CHD patients with New York Heart Association (NYHA) functional class II to IV were

listed for HT; 40 patients on optimal medical therapy were referred for paced contractility response cardiac catheteri-

zation before CRT consideration. If dP/dt-max improved $15% from baseline, these “responders” were given the option

of CRT with continued follow-up after implantation.

RESULTS Of40patients studied, 26 (65%)(age22�8.2years;9of26[35%]singleor systemic rightventricle; 17of26[65%]

with pacemakers) met criteria for possible hemodynamic benefit and underwent CRT implantation. All 26 patients improved in

NYHA functional classification: 5 of 26 patients (19%) were later relisted for HT (4 to 144 months, mean 55 months) after CRT

implantation,whereas 21 of 26 (81%) continuedwith improvedNYHA functional class (12 to 112months,mean44months) later.

A repeat dP/dt-max study following long-term CRT showed stable function or continued contractility improvement.

CONCLUSIONS Heart failure is common among CHD patients, and therapies are limited. CRT guidelines do not address

clinical and anatomic issues of CHD. Short-term paced contractility response testing identifies those CHD patients who

are likely to respond to CRT regardless of anatomy. (J Am Coll Cardiol EP 2017;3:830–41) © 2017 by the American

College of Cardiology Foundation.

T he congenital heart, even after surgical repair,is not equal to a structurally normal heart.Intrinsic anatomic and physiological variations

often contribute to eventual early myocardial dysfunc-tion. As a result, this ever-increasing number of adultswith congenital heart disease (CHD) constitutes a

m the Section of Cardiology, Department of Pediatrics, The Children’s Ho

dicine, Detroit, Michigan. The authors have reported that they have no re

close.

authors attest they are in compliance with human studies committe

titutions and Food and Drug Administration guidelines, including patien

it the JACC: Clinical Electrophysiology author instructions page.

nuscript received September 16, 2016; revised manuscript received Decem

challenging population for therapeutic intervention.Heart transplantation (HT), at times preceded byimplanted assist devices, is often an eventual commonpathway with a reported 70% survival at 5 years (1).

Biventricular pacing or cardiac resynchronizationtherapy (CRT) to promote cardiac remodeling is a

spital of Michigan, Wayne State University School of

lationships relevant to the contents of this paper to

es and animal welfare regulations of the authors’

t consent where appropriate. For more information,

ber 29, 2016, accepted January 12, 2017.

AB BR E V I A T I O N S

AND ACRONYM S

CHD = congenital heart disease

CRT = cardiac

resynchronization therapy

CRT-neg = ineffective cardiac

resynchronization therapy

response

CRT-pos = effective cardiac

resynchronization therapy

response

CS = coronary sinus

D-TGA = D-transposition of the

great arteries

ECG = electrocardiogram

EF = ejection fraction

HF = heart failure

HT = heart transplantation

LVEDD = left ventricular end-

diastolic diameter

LVEF = left ventricular

ejection fraction

NYHA = New York Heart

Association

PVL = pressure-volume loop

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7 Karpawich et al.A U G U S T 2 0 1 7 : 8 3 0 – 4 1 Congenital Heart Long-Term CRT

831

now-established option for patients with ischemicheart disease and myocardial dysfunction, with pub-lished guidelines for patient selection. These includestable sinus rhythm, QRS duration $120 ms, leftbundle branch QRS morphology, left ventricularejection fraction (LVEF) #35%, increased left ven-tricular diastolic diameter, and New York Heart As-sociation (NYHA) functional class II to IV (2).However, even applying these guidelines, CRT suc-cess or “responder rates” are quite variable with onlya 5% improvement in measured ejection fraction (EF)sufficient enough to indicate a successful response insome studies (3). To compound the problem, there arevery little real long-term clinical data of patient statusfollowing CRT, and these guidelines are really notapplicable to CHD patients, who typically are withoutcoronary disease and often with intrinsicallyabnormal electrocardiograms (ECGs) or have pace-makers. Most have altered anatomy, often with asystemic right (e.g., transposition of great arteries) orsingle-ventricle morphology, and have surgical scarsand/or prosthetic materials, including septal patches,that can alter measurements of EF. So, the clinician,when faced with a CHD patient in heart failure (HF), isessentially in a quandary in attempting to decidewhether any one particular CHD patient, with anygiven diverse congenital anatomy or surgical repair,might be a candidate for CRT. Previous studies of CRTapplications to CHD patients are relatively few, withlimited follow-up, and are not based on any uniformpre-selection criteria. Even the most recent Pediatricand Congenital Electrophysiology Society/HeartRhythm Society expert consensus statement onarrhythmia management among CHD patients wasonly able to modify and adapt these existing guide-lines with no specific issues relating to CHD per se (4).

SEE PAGE 842

This study is the first to our knowledge to specif-ically evaluate CHD patients for CRT efficacy beforeimplantation by using the directly measured intra-cardiac hemodynamic variable of paced ventricularcontractility response (dP/dt-max) as a potential pa-tient selection marker. In addition, long-term clinicalfollow-up, in some cases up to 12 years post-CRTimplantation, was performed to determine whethershort-term clinical improvement equated to morelong-term CRT hemodynamic benefits.

METHODS

From the database of congenital heart patients fol-lowed at the Children’s Hospital of Michigan Sectionof Cardiology, including the Adult Congenital Heart

and Heart Failure programs, information onpatients who developed HF severe enough(NYHA functional class II to IV) to warrantconsideration and discussion for HT listingfrom 1999 to 2015 was reviewed. This studywas approved by the investigational reviewboard of Children’s Hospital of Michigan/Wayne State University/Detroit Medical Cen-ter. Hemodynamic cardiac catheterizationstudies, as part of HT listing evaluations,were reviewed in addition to other clinicallyavailable data. During hemodynamic cathe-terizations, short-term biventricular pacingwith direct measurement of contractility (dP/dt-max) responses was made available to allpatients from 1999 to the present, to deter-mine potential CRT application in addition tostandard HF medical therapies. Patientreferral for pre-CRT testing was at thediscretion of the referring physician.

All patients who agreed to pre-CRT testingfollowed standard cardiac catheterization/el-ectrophysiology study protocols, which incl-uded sedation under the guidance of ananesthetist, with intubation and controlled

ventilation to remove any respiratory variations thatmight affect hemodynamic measurements. After inf-ormed consent was obtained, the femoral vein and ar-tery were entered using standard percutaneoustechniques and catheters were inserted to permit co-ntinuous systemic arterial and ventricular pressuremonitoring as well as to provide temporary pacing asrequired. A heparinized intravenous solution (2,000U/500 ml) was infused. Due to the anticipated shorttime to evaluate systemic ventricular-paced function,combined with a heparinized flush solution, activatedclotting time levels were not routinely obtained. A5- to 6-F pigtail hemodynamic arterial recordingcatheter (Cook Medical, Bloomington, Indiana) wasinserted into the systemic ventricle and attached to aphysiological recording system (Cathcor Model 2.2, orSensis system, Siemens, Munich, Germany). Thispermitted direct measurements of ventricular pres-sures as well as contractility derivatives (dP/dt-max)with a simultaneous ECG recording to determineoptimized paced ventricular function. Followingbaseline hemodynamic measurements, 2 temporary4- to 6-F diameter bipolar or quadripolar pacingcatheters (Boston Scientific, Natick, Massachusetts)were inserted under fluoroscopic guidance to initiatebiventricular pacing. Among patients with otherwisenormal anatomic arrangements (e.g., repaired septaldefects, tetralogy of Fallot), catheters were insertedinto the coronary sinus (CS) and venous ventricle to

FIGURE 1 Exemplary AP Fluoroscopic Appearance of CRT

Leads in a Patient With Congenital AV Block and Normal

Anatomic Relationships

The cardiac resynchronization therapy (CRT) leads are placed in

the venous right ventricle (A), coronary sinus (B), and right

atrium (C). AP ¼ anteroposterior; AV ¼ atrioventricular.

Karpawich et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7

Congenital Heart Long-Term CRT A U G U S T 2 0 1 7 : 8 3 0 – 4 1

832

attain optimal biventricular pacing. As previouslyreported, to optimize paced ventricular contraction,the CS pacing catheter was inserted first, followed byvenous ventricular “site mapping” to maximizebiventricular paced contractility (5). At least 4 ven-tricular sites were tested: apex and low-, mid-, andhigh/inflow-septum. For patients with altered anat-omy with no direct CS connection (e.g., intra-atrialbaffle repair [Mustard] of D-transposition of the greatarteries [D-TGA]), pacing catheters were inserted intoboth the venous as well as the arterial ventricles.Combined pace-mapping was then performedwith theretrograde right ventricular (systemic) catheter posi-tioned at the free wall and left (venous) ventricular atseptal sites mentioned in the preceding text. For thosepatients with pre-existing epicardial or endocardialleads, 1 temporary catheter was placed to approximatethat existing lead and the other catheter was used to“site map” for best biventricular paced contractilityresponse. For single-ventricle patients, 2 4-F arterialpacing catheters were inserted into the ventricle andpositioned at opposite sides, at positions that could beapproximated on the respective epicardial surfaceslater by the implanting surgeon. In each instance,fluoroscopic images were obtained and the best sitewas recorded for permanent lead implant.

Temporary biventricular (VVI mode) pacing wasachieved by pacing (EP Workmate, St. Jude Medical,St. Paul, Minnesota) both venous as well as arterialventricles with an interventricular-paced delayprogrammed between 0 and 50 ms and at least10 pulses/min faster than intrinsic sinus rates or, ininstances of a pre-existing pacemaker, 10 pulses/minfaster than the paced ventricular rate. After estab-lishing intrinsic or pre-existing paced baselinemeasurements of ECG, pressures, and contractility,temporary biventricular pacing was instituted for atleast 5 min to re-establish a CRT baseline. The VVImode was chosen to focus specifically on intrinsicventricular function and remove any atrioventricularcontributions to ventricular contractility. A minimumof 3 contractility measurements (dP/dt-max) werethen obtained and averaged. Each patient served ashis/her own control.

On the basis of established definitions of“responder” showing anywhere from a 5% to 10%change improvement in LVEF by cardiac ultrasound(3), this study was intentionally more conservative.Due to potential CRT implant morbidities especiallyassociated with CHD patients, an arbitrary contrac-tility (dP/dt-max) improvement at least 15% overbaseline values was chosen to show a potentiallyeffective CRT response (CRT-pos). Patient responsesnot demonstrating the $15% improvement were

labeled as ineffective (CRT-neg). After the catheteri-zation study, therapeutic options were discussedwith the patient/family. Those of the CRT-pos groupwere given the option to either undergo anew CRT implant or an upgrade addition to anypre-existing pacing system or continue withestablished medical management. Those of the CRT-neg group continued with medical HF management.Among CRT-pos patients with normal ventricularrelationships (septal defects, tetralogy of Fallot)and a pre-existing transvenous ventricular lead, onlya new CS lead was implanted (Figure 1). Amongpatients with Mustard/D-TGA repair, a combinationepicardial/endocardial (“hybrid”) lead system wasutilized. If there was no previous pacing system, anew epicardial right ventricular (systemic ventricle)lead was implanted first, following standard thora-cotomy techniques, then the transvenous left(venous) ventricular lead was inserted. Among D-TGApatients with a pre-existing epicardial right ventric-ular lead, only a new transvenous venous left ven-tricular lead was implanted. In both instances, theepicardial lead was tunneled subcutaneously to thesubclavicular implant site, and both leads were con-nected to the pulse generator (Figure 2). Among pa-tients with a single ventricle, “biventricular”epicardial leads were implanted intraoperatively atopposing sides of the ventricle as previously

FIGURE 2 Fluoroscopic AP View of CRT Leads in a Patient

With D-TGA/Mustard Repair

An epicardial ventricular lead (A) is on the systemic right, and

endocardial leads (B) and (C) in the venous left ventricle and

atrium, respectively. The epicardial lead is tunneled (D) subcu-

taneously to connect with a left subclavicular generator implant

site. An atrial baffle stent (E) is present. D-TGA ¼ D-transposition

of the great arteries; other abbreviations as in Figure 1.

FIGURE 3 Exemplary Fluoroscopic Lateral View of Epicardial

Lead Placement in a Patient With Post-Repair Single

Ventricle (Atriopulmonary “Fontan” Repair)

Epicardial lead (A) is on the anterior and (B) posterior ventricular

positions. Lead (C) is on the atrium. An intra-atrial catheter is

visible, and contrast material is seen delineating the posterior

wall of the atrium.

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7 Karpawich et al.A U G U S T 2 0 1 7 : 8 3 0 – 4 1 Congenital Heart Long-Term CRT

833

evaluated to optimize paced ventricular contractility(Figure 3). In all instances, standard implant sensingand pacing threshold measurements were obtained.Various leads and dual-chamber/multilead pulsegenerators by different manufacturers were used.

All patients, those with and without CRTpacing, were continually followed up to 12 years, toevaluate their respective clinical conditions, notingimproving/worsening of HF symptoms, any relistingforHT, or death fromany cause. In addition to standardclinical assessments including cardiac echocardiogramand any exercise testing (if performed), a repeat car-diac catheterization to reassess ventricular contrac-tility following implant was performed in all availableCRT recipients from 6 to 14 months later, unless theyhad received an HT during that time interval.

STATISTICS. This was a retrospective study. All datawere reported as mean � SEM for continuous vari-ables and frequency for categorical variables. Patientstypically served as their own controls and werecompared using the paired and nonpaired Student ttest, z-scores, and chi-square analysis as indicated.All statistical analyses were performed using Graph-Pad Prism software (GraphPad Prism Software,version 7, La Jolla, California). Statistical significancewas defined as a p value <0.05.

RESULTS

From 1999 to 2015, a total of 103 CHD patients wereconsidered for HT listing. Of these, 40 (38%) (ages 4 to57 years, mean 25 � 10 years; 60% male) were referredand agreed to undergo short-term CRT pacing efficacyhemodynamic evaluation at the time of their standardHT workup cardiac catheterization study. Of these40 patients, demographics including age, CHD anat-omies, surgical repairs, pre-existing pacemakers,contractility indexes, and follow-up intervals arelisted in Table 1. Only 1 patient exhibited an intrinsicleft bundle branch block QRS morphology. Pre-existing pacemakers were found in 28 of 40 (70%)patients (1 atrial, 2 ventricular, and 25 dual-chamber)associated with their respective congenital heart de-fects. Due to various CHD and surgical repairsinvolving septal prosthetic materials, artificial valves,or pacemakers, “left” ventricular EF could bemeasured with any degree of accuracy in only 14 of 40(35%), whereas EF estimation could be performed in10 of 40 (25%) patients. CHD anatomy precluded LVEFmeasurements in 16 of 40 (40%) patients. Likewise,left ventricular end-diastolic diameter (LVEDD) couldbe measured in only 28 of 40 (70%) patients withsome degree of accuracy using echocardiography.

TABLE 1 Patient Demographics (N ¼ 40)

CHD Diagnosis SexAge(yrs) Pacemaker CRT*

LVEF†(%)

QRS(ms)

NYHAFunctional Class

Follow-Up(Months)

DCM F 4 No Yes 29 98 III, IV 4

AS/DCM F 9 Yes, DDD No 40 180‡ III, IV 24

CAVB F 13 Yes, DDD Yes 22 161‡ II, III 28

TOF/DCM F 13 Yes, DDD No 30§ 200‡ III, IV 84

CCAVB/DCM F 14 Yes, DDD Yes 42 134‡ II 112

WPW, DCM M 14 No Yes 24 172 III 48

CAVB M 14 Yes, DDD Yes 26 147‡ III, IV 30

CAVB M 15 No Yes 65 136k III, IV 72

Single ventricle/”Fontan”-type repair M 16 No No NA 120 III, IV 6

TOF M 17 No No 35§ 157 II, III 6

HCM/DCM M 18 Yes, DDD Yes 45 190‡ II, III 6

Coarctation M 18 No Yes 29 94 II 36

AV canal/AVR/MVR F 19 Yes, DDD Yes 37§ 166‡ II 72

CAVB F 20 Yes, VVI Yes 14 210‡ II, III 36

DCM F 20 No Yes 12 120 II 112

CAVB M 21 Yes, DDD Yes 40 160‡ II, III 48

CAVB F 21 Yes, DDD Yes 28 168‡ II, III 12

TOF/DCM M 23 No Yes 25§ 172 III 72

L-TGA/VSD/PS M 24 Yes, DDD Yes NA 160‡ II, III 144

D-TGA/Mustard M 24 Yes, VVI Yes NA 150‡ II, III 42

Single ventricle/”Fontan”-type repair F 27 Yes, DDD Yes 35§ 167‡ II, III 88

L-TGA/DCM M 27 Yes, DDD Yes NA 164‡ II 35

D-TGA/Mustard M 27 Yes, DDD Yes NA 95 II, III 16

AV canal F 28 Yes, DDD Yes 39§ 120‡ II 24

ASD M 28 Yes, DDD No 36 150‡ II, III 9

D-TGA/Mustard F 28 Yes, DDD No NA 128‡ II, III 42

D-TGA/Mustard M 29 No Yes NA 157 II 36

Single ventricle/”Fontan”-type repair F 29 No Yes NA 122 II, III 12

PA-VSD/DCM M 29 Yes, DDD No 28§ 150‡ III, IV 24

D-TGA/Mustard M 29 No No NA 90 II 78

Truncus M 31 Yes, DDD No 40§ 200‡ II, III 96

DORV/VSD M 32 No Yes 49§ 202 II 12

D-TGA/Mustard M 32 Yes, DDD Yes NA 208‡ II 30

D-TGA/Mustard M 32 Yes, AAI No NA 127 II, III 10

D-TGA/Mustard M 37 Yes, DDD No NA 160‡ II 30

D-TGA-Mustard M 39 Yes, DDD Yes NA 157‡ II, III 12

D-TGA/Mustard M 39 Yes, DDD Yes NA 150‡ II, III 30

Single ventricle F 40 Yes, DDD No NA 60 II 8

D-TGA/Mustard F 45 Yes, DDD No NA 82 II 12

TOF F 57 Yes, DDD No 30§ 120‡ II 12

Mean � SD 25 � 10 33 � 11 148 � 36 40 � 34

*Patient elected to undergo cardiac resynchronization therapy (CRT). †Not applicable (NA) indicates non-”left” ventricular anatomy preventing any accurate ejection fractionmeasurement. ‡Ventricle paced. §Estimated ejection fraction due to congenital heart disease (CHD) anatomy/surgical repair. kIntrinsic left bundle branch block.

AAI ¼ single-chamber atrial pacing mode; AS ¼ aortic stenosis; ASD ¼ atrial septal defect; AV ¼ atrioventricular; AVR ¼ aortic valve replacement; CAVB ¼ congenitalatrioventricular block; DCM¼ dilated cardiomyopathy; DDD ¼ dual-chamber pacing mode; DORV¼ double-outlet right ventricle; D-TGA¼ D-transposition of the great arteries;HCM ¼ hypertrophic cardiomyopathy; L-TGA ¼ L-transposition of the great arteries; LVEF ¼ left ventricular ejection fraction; MVR ¼ mitral valve replacement; NYHA ¼ NewYork Heart Association; PA ¼ pulmonary atresia; PS ¼ pulmonary stenosis; TOF ¼ tetralogy of Fallot; VSD ¼ ventricular septal defect; VVI ¼ single-chamber ventricular pacingmode; WPW ¼ Wolff-Parkinson-White.

Karpawich et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7

Congenital Heart Long-Term CRT A U G U S T 2 0 1 7 : 8 3 0 – 4 1

834

Short-term improvement in ventricular contrac-tility (dP/dt $15%) on cardiac catheterization,as defined, was considered a pre-selection criterionto indicate short-term CRT “responders” versus“nonresponders.” Of the 40 patients who agreed toundergo CRT efficacy hemodynamic evaluation, 26 of40 (65%) showed the required $15% short-term

contractility improvement (mean 597 � 171 mm Hg/svs. 848 � 207 mm Hg/s; p < 0.05), and thus weredefined as CRT-pos responders (Figure 4). After pa-tient/family discussions, all 26 patients elected toundergo CRT implant. The remaining 14 of 40 (35%)did not achieve the prerequisite of $15% improve-ment (696 � 127 mm Hg/s vs. 710 � 137 mm Hg/s;

FIGURE 4 Graph Showing Comparative Responses Among Patients

Demonstrating $15% Improvement in Contractility (dP/dt-max) With

Short-Term CRT Testing

0

200

400

600

800

1000

1200

1400

mm

Hg-

sec

P < 0.006

CRT Short-term RespondersN = 26

597 ± 171

848 ± 207

Pre-CRT Short-term CRT

Ventricular dP/dt-max

With each patient serving as his/her own control, mean contractility in the short term

increased >40% from baseline. These differences were significant. CRT ¼ cardiac

resynchronization therapy.

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7 Karpawich et al.A U G U S T 2 0 1 7 : 8 3 0 – 4 1 Congenital Heart Long-Term CRT

835

p ¼ 0.78) and were defined as nonresponders(CRT-neg) (Figure 5). They continued with standardHF medical management. On comparison of othermeasured patient variables between both CRTresponse groups (Table 2), no significant differenceswere noted in patient age, LVEDD, measured EF, orQRS duration. Of special note, there was no signifi-cant difference in measured baseline contractilityindexes (dP/dt-max) between short-term CRT re-sponders and nonresponders. Pre-existing ventricu-lar pacemakers were found in 65% of CRT-poscompared with 71% of CRT-neg patients. A single or“right” systemic ventricle morphology was found in35% of CRT-pos and 50% of CRT-neg patients.

After the initial pre-CRT evaluation, clinicalfollow-up intervals were comparable between patientgroups: those 26 patients who received a CRT pacingsystem were followed from 4 to 144 months (mean46 � 35 months), whereas those 14 who did not andwere continued on the established HF/HT programwere followed from 2 to 96 months (mean 36 � 32months) (p ¼ 0.40). During those respective intervals,initial clinical improvements were observed in all 26CRT recipients, with each dropping at least 1 grade inNYHA functional classification within several monthsafter implantation and removed from HT consider-ation. Among these patients with clinical improve-ment, pre-CRT compared with post-CRT QRSdurations (mean 153 � 32 ms vs. 139 � 22 ms; p ¼ 0.01)as well as LVEDD measurements and concomitant z-scores decreased significantly (mean 6.3 � 0.9 cm vs.5.6 � 1.0 cm; p ¼ 0.0001; 4.35 � 3.0 vs. 2.5 � 2.2;p ¼ 0.002, respectively). Of these 26 patients, afollow-up cardiac contractility study was performedto measure long-term CRT efficacy from 6 to 14months post-CRT implantation in 21 patients. Again,standard catheterization sedation protocols wererepeated with ventricular pressures, contractilityindexes, and saturations recorded as previously per-formed. However, compared with the pre-CRTimplant study, there was no additional temporarypacing performed. Each patient again served as his orher own control with paced heart rates (with CRTdevices programmed to the VVI mode) comparable toinitial pre-CRT values. This repeat contractility studyconfirmed positive myocardial remodeling with nosignificant differences in dP/dt-max values fromimplant compared with measured values after long-term CRT pacing (Figure 6). Over the continuedpost-CRT implantation follow-up interval in thisstudy, 5 of 26 (19%) patients demonstrated eventualmyocardial deterioration, and were re-listed and un-derwent transplantation from 4 to 144 months (mean55 months) later, whereas 4 died: 1 associated with

elective mitral valve surgery, and 3 associated withmedical noncompliance. Of those who exhibitedeventual myocardial deterioration and underwenttransplantation, 3 of 5 had intrinsic myocellular dis-ease related to hypertrophic and idiopathic dilatedmyopathies, whereas 2 of 5 had repaired CHD (atrio-ventricular canal and ventricular septal defect). Ofnote, those patients with intrinsic muscle diseaseshowed deterioration from 4 to 48 months (mean 20months) after CRT, whereas the other 2 patientsshowed deterioration only after 72 to 144 months(mean 108 months) following CRT. The other 21 of 26(81%) remained off the transplant list 12 to 112 months(mean 44 months) later with stable improved NYHAfunctional class. Of those 14 patients who remainedon the standard HF management protocol, 2 receiveda transplant within 2 years, whereas 5 died awaitingtransplantation. The remaining 7 patients continuedin compensated HF without improvement in NYHAfunctional status.

DISCUSSION

Advancements in the field of HF management haveled to the concept of improving myocardial functionby “synchronizing” right–left ventricular contrac-tility. In this regard, CRT among patients with HFfrom ischemic coronary heart disease is well estab-lished, with published guidelines for patient selectionas noted earlier in the text. The electromechanicaldyssynchrony associated with HF is targeted by CRT

FIGURE 5 Comparative Responses Among Patients Not Demonstrating $15% Improvement in Contractility

Mean dP/dt showed only a 2% change. There was no significant difference. CRT ¼ cardiac resynchronization therapy.

Karpawich et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7

Congenital Heart Long-Term CRT A U G U S T 2 0 1 7 : 8 3 0 – 4 1

836

that simultaneously permits pacing of both ventri-cles. This promotes coordinated biventricularcontraction, decreases myocardial strain andmyocardial energy expenditure, reverses adverseremodeling, and reduces HF symptoms (6,7).Depending on definitions of CRT “responder,”published studies have reported from 20% to50% nonresponders, whereas terms such as “super-responders” add much confusion to the understand-ing of the true efficacy of CRT (8).

Patients with CHD, even after repair, may continueto have altered anatomy and physiology contributingto early HF, often leading to HT consideration. How-ever, transplantation may not be an option for somepatients, and listing does not guarantee organ receipt.As a result, alternative therapies play an importantrole to delay transplantation and/or improve qualityof life. Following the initial published report of suc-cessful CRT application in a CHD patient with HF (9),there have been several additional publications, 2 ofwhich were multicenter, all indicating the feasibilityof CRT application among patients with variable CHDanatomies (10–14). Unfortunately, there was nostandardization for patient selection, follow-upduration was very limited (mean 10 months), anddefinitions of success were variable. An excellentsummary review paper of these and other previouspublications concluded that due to the diverse and

quite unique anatomic and physiological issuesassociated with individual CHD patients, a simpleextrapolation of published guidelines for CHD patientselection was not always efficacious and that otherdiagnostic studies to better define dyssynchrony areneeded. In effect, as the authors indicate, “there arecurrently no acceptable consensus guidelines for pa-tient selection” (15). So, CHD patient selection criteriafor CRT application remains an enigma, with the needfor better patient-specific screening than currentlyexists (16).

Older adults with HF treated by CRT typically havea morphological left ventricular dysfunction andexhibit a left bundle branch block QRS pattern. Onenonrandomized trial, the MADIT-CRT trial (Multi-center Automatic Defibrillator Implantation TrialWith Cardiac Resynchronization Therapy), showedthat patients with a right bundle branch block patterndue to coronary disease may not benefit from CRT(17). Because cardiac electrical synchrony does notequate to mechanical synchrony, relying on QRS axis,duration, and morphology for CRT selection criteriaamong CHD patients can be counter-productive, if notdeceptive. Often, patients with CHD have intrinsicabnormalities of all 3 electrical variables, and notrelated to ischemia. As a result, any concepts ofischemia-related scar burden may not apply. Inaddition, patients with CHD may have complex

TABLE 2 Comparative Patient Data Between Short-Term CRT “Responders” Versus

“Nonresponders”

CHD Diagnosis SexAge(yrs)

Pacemaker,Mode

LVEDD(mm)

LVEF(%)

QRS(ms)

VdP/dt(mm Hg/s)

$15% improved dP/dt

DCM F 4 No 5.9 29 98 483

CAVB F 13 Yes, DDD 5.8 22 161* 625

CAVB/SCD/DCM F 14 Yes, DDD 5.4 43 134* 647

WPW, DCM M 14 No 7.7 24 172 581

CAVB M 14 Yes, DDD 6.9 25 147* 332

CAVB M 15 No 4.8 65 136‡ 735

HCM/DCM/SCD M 18 Yes, DDD 5.4 45 190* 603

s/p coart M 18 No 6.7 29 94 774

AV canal/AVR/MVR F 19 Yes, DDD 6.1 25† 166* 188

CAVB F 20 Yes, VVI 8.1 10 210* 264

DCM F 20 No 6.2 40 120 637

CAVB M 21 Yes, DDD 8.1 40 160* 454

CAVB F 21 Yes, DDD 5.8 28 168* 827

TOF/DCM M 23 No 6.4 35 172 542

L-TGA/VSD/PS M 24 Yes, DDD 5.9 NA 160* 490

D-TGA/Mustard M 24 Yes, VVI NA NA 150* 735

Fontan (DILV) F 27 Yes, DDD 6.2 35† 167* 768

L-TGA, DCM M 27 Yes, DDD 8.2 NA 164* 674

D-TGA/Mustard M 27 Yes, DDD NA NA 95 647

AV canal F 28 Yes, DDD 5.5 14† 120* 618

D-TGA/Mustard M 29 No NA NA 157 769

Single ventricle F 29 No NA NA 122 560

DORV/VSD M 32 No 6.0 40 202 800

D-TGA/Mustard M 32 Yes, DDD NA NA 208* 413

D-TGA/Mustard M 39 Yes, DDD NA NA 157* 530

D-TGA/Mustard M 39 Yes, DDD NA NA 150* 840

Mean � SD 23 � 8 65% pace 6.3 � 1.0 32 � 12 153 � 32 597 � 171

<15% improved dP/dt

subAS/DCM F 9 Yes, DDD 6.4 40 180* 650

TOF/DCM F 13 Yes, DDD 6.1 30 200* 823

Fontan (HLHS) M 16 No NA NA 120 696

TOF M 17 No 6.6 35† 157 759

ASD M 28 Yes, DDD 7.7 20 150* 500

D-TGA/Mustard F 28 Yes, DDD NA NA 128* 798

VSD M 29 Yes, DDD 6.6 20† 150* 432

TGA M 29 No NA NA 90 828

Truncus1/VSD M 31 Yes, DDD 6.3 40 200* 681

D-TGA/Mustard M 32 Yes, AAI NA NA 127 850

D-TGA/Mustard M 37 Yes, DDD NA NA 160 732

Fontan (DILV) F 40 Yes, DDD NA NA 60 590

D-TGA/Mustard F 45 Yes, DDD NA NA 82 698

TOF F 57 Yes, DDD 4.1 30† 120* 832

Mean � SD 29 � 13 71% pace 6.2 � 1.0 30 � 8 137 � 42 696 � 127

LVEF is not applicable to right or single ventricle morphologies. Left ventricular end-diastolic diameter (LVEDD)by echocardiogram is only applicable to anatomic “left” ventricular morphologies. Ventricular contractility index(VdP/dt) is applicable to all ventricular morphologies. *Ventricle paced. †Estimated ejection fraction due to CHDanatomy/surgical repair. ‡Intrinsic left bundle branch block.

DILV ¼ double inlet left ventricle; HLHS ¼ hypoplastic left heart syndrome; SCD ¼ sudden cardiac death; s/pcoart ¼ status-post coarctation repair; subAS ¼ subvalve aortic stenosis; other abbreviations as in Table 1.

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7 Karpawich et al.A U G U S T 2 0 1 7 : 8 3 0 – 4 1 Congenital Heart Long-Term CRT

837

anatomic and physiological variations, different fromthose with an ischemic cardiomyopathy. Such pa-tients may have a systemic “right” or “single”ventricle morphology, and a right bundle branchblock pattern (with or without “left hemi-block”) isnot infrequent, typically due to surgical incisions oraltered conduction system, not coronary disease. Inaddition, patients with repaired CHD often requirepacemaker therapy, and there are no caveats for pa-tients with pre-existing cardiac devices in previouspublications or guidelines. Although QRS durationsdid shorten following CRT in this study, it must beremembered that 70% of patients had pre-existingpacemakers with typically wide paced-QRS com-plexes. So, whether QRS duration can be used as aneffective marker for hemodynamic improvementamong CHD patients, and especially those with pre-existing pacemakers, is unknown. It might also bespeculated that CHD patients with intrinsic muscledisease may not respond as well to CRT as those pa-tients following surgical repair of various defects.However, more extensive studies need to be per-formed in this regard. In addition, a biventricularpacemaker implant among CHD patients is costly andnot without potentially significant morbidities,including the need for a thoracotomy. Venous accessto the coronary sinus may be limited or impossible,and transvenous combined with a surgical epicardiallead placement may be required, creating a “epi–endohybrid” pacing lead implant configuration. In somepatients, only an epicardial lead implant approach isfeasible. And due to post-operative epicardial fat,fibrosis, and tissue concerns, epicardial lead place-ment can be suboptimal.

At this writing, there are no uniform pre-selectioncriteria, nor are there any pre-defined outcomes tomeasure the success rate of CRT among CHD patients.Noninvasive measurement of EF by echocardiogra-phy can easily be deceptive among patients who havea single ventricle or a systemic right ventricle,because EF is often inaccurate. With CHD, ventriculargeometry and loading conditions are significantlyaltered, affecting the validity and reproducibilityof contractility assessment by M-mode as well as2-dimensional echocardiography. Among patientswho have a systemic right ventricle (e.g., the Mustardintra-atrial baffle repair for D-TGA), the decision forinitiating CRT would depend on the systemic ven-tricular function, as per published “guidelines.”However, EF assessment of right ventricular systolicfunction is limited due to the complex shape andgeometry of the anatomic “right” ventricle as it wrapsaround the “left” ventricle. Recently, studies haveshown that contraction in systemic right ventricles is

predominantly due to circumferential rather thanlongitudinal fiber shortening. Moreover, echocardio-graphic visualization of all segments of the rightventricle is quite difficult due to its anterior location

FIGURE 6 Short-Term Versus Follow-Up Results Comparing the Original Short-Term and Later Long-Term Contractility Values

The results are among CRT recipients comparing the original short-term and later long-term (6–14 months post-CRT) contractility (dP/dt-max)

values among patients who demonstrated the initial $15% improvement. As a group, contractility measurements remained stable or continued

to improve indicating favorable long-term remodeling attesting to CRT efficacy. CRT ¼ cardiac resynchronization therapy.

Karpawich et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7

Congenital Heart Long-Term CRT A U G U S T 2 0 1 7 : 8 3 0 – 4 1

838

and near-field artifacts, limiting the utility and accu-racy of circumferential speckle tracking (18–21).Septal patches and conduits additionally can affectaccuracy of EF recordings. A multicenter prospectiveclinical trial (PROSPECT [Predictors of Response toCardiac Re-Synchronization Therapy] trial) used 12echocardiographic parameters of dyssynchrony,based on conventional and tissue Doppler-basedmethods for evaluation of response to CRT amonghearts with normal anatomy, and concluded that nosingle echocardiographic measure may be used toimprove patient selection for CRT because of largevariability in the analysis of dyssynchrony parame-ters (22). Another recent study demonstrated the lackof correlation between echocardiographic indexes ofright ventricular EF and cardiac magnetic resonanceimaging data (23). Although cardiac magnetic reso-nance imaging accurately assesses ventricular func-tion, many of the CHD patients potentially needingCRT evaluation have pre-existing pacemakers, whichmay preclude the use of this imaging modality. Also,EF measurements of a single ventricle are equallyinaccurate. So, in essence, guidelines involving LVEFdo not apply to many CHD patients.

In addition, QRS duration can also be an unreliablemeasure because CHD patients often have intrinsicECG abnormalities. One cross-over trial compared ECG(QRS duration) and echocardiogram parameters forCRT optimization in a pediatric population (mean age

9.1 � 4.3 years) and concluded that echocardiogramoptimization of synchrony was not superior to ECGoptimization (24). Another randomized trial demon-strated that CRT can in fact be harmful if the QRSduration is not prolonged, even in the presence ofdecreased EF and severe symptoms of HF (25). Yet,again, CRT efficacy among patients with a narrow QRShas also been reported (26). This explains why thereare considerable variations in the results of differentstudies and that there is a need for more research onthis subject. In previous studies, there were no pre-determined definition of responders versus non-responders, which obscured the true efficacy of CRT inindividual patients and has led to a nonuniform andnonstandardized measure of successful outcomes.Despite these limitations, the studies done so far in theevaluation of CRT benefits have added important in-formation on the efficacy of CRT as a potential thera-peutic intervention in CHD patients with HF.

Given all these facts, it is difficult to apply thepublished guidelines for the use of CRT to the CHDpatients with heterogeneous cardiomyopathieswithout real evidence of its benefit, especiallyimprovement in contractility, keeping in mind pro-cedure costs and associated potential morbidities.The recent recommendations of CRT indications inadults with CHD, presented by the consensus of thePediatric and Congenital Electrophysiology Societyand the Heart Rhythm Society, largely follow the

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7 Karpawich et al.A U G U S T 2 0 1 7 : 8 3 0 – 4 1 Congenital Heart Long-Term CRT

839

adult criteria of CRT use in ischemic and idiopathiccardiomyopathy patients (4). The recommendationsare based mostly on expert opinion and panelconsensus, lacking research-based evidence. In thatregard, there is a need to test and analyze new pa-rameters to evaluate paced electromechanical dys-synchrony that are least affected by the anatomicheterogeneity of CHD patients.

In this current study, short-term, directlymeasured dP/dt-max hemodynamics were used ascriteria for CRT selection among CHD patients withsymptomatic HF (27). As an isovolumic index ofventricular contractility, dP/dt measures the rate ofventricular pressure rise in early systole and is asensitive measure of systolic function (28). Unlikemany other noninvasive parameters (e.g., EF), itsmeasurement and interpretation are not affected byventricular wall motion abnormalities and variationsin ventricular morphology and anatomy (29).Although implant contractility measurements havebeen shown to be an effective tool to optimize pacinglead placement (5), to date, there are no studies thatutilize direct hemodynamically measured dP/dt as aselection criterion for CRT in CHD patients, per se. Astudy done among patients with a single ventriclecomparing the echocardiographic and pressure-volume loop (PVL) indexes of systolic function(end-systolic elastance, dP/dt-max, stroke workindexed to end-diastolic volume) demonstrated thatPVL indexes are an effective and reliable way tomeasure systolic function in patients with alteredanatomy and can be used to validate echocardio-graphic indexes in CHD patients (30). However, PVLstudies are lengthy and require specialized equip-ment not found in most cardiac catheterizationlaboratories (31). And, although dP/dt values can beobtained by echocardiographic study and have beenutilized among CHD patients to evaluate CRT efficacy(32), values do not always correlate with direct cath-eterization measurements, especially in non–leftventricular morphologies (33). Ventricular regionalwall motion assessment using electroanatomic map-ping has been applied to study right–left delayedactivation and dyssynchrony among patients withtetralogy of Fallot, both in the short term and thoseundergoing CRT, with short-term clinical follow-up(6 months) (34). However, additional studies will berequired to determine whether the observed im-provements at the site of latest activation equate tolonger-term clinical improvements among CHD pa-tients with HF.

For assessing the benefit of CRT, which can be anexpensive therapy associated with potential morbid-ities, it is indispensable to use the parameters that

most accurately determine systolic functionimprovement in CHD patients. In this regard, dP/dtmeasurements can be used as an effective screeningtest regardless of CHD anatomies. This studydemonstrated that most measured patient variables,including EF, QRS, and LVEDD, were not useful toscreen those CHD patients who did and did notrespond to CRT. The study also is the first to ourknowledge to specifically compare pre- and post-CRTcontractility measurements to confirm/refute CRTefficacy among CHD patients over an extended post-implantation time interval up to 12 years. Bymeasuring contractility at different right–left ven-tricular paced activation intervals, additional “fine-tuning” of CRT response can be achieved (35,36).

Patients who develop HF based on individualanatomic/physiological issues typically experienceeventual continuation of myocardial dysfunction,and most therapies act only as a “bridge-to trans-plant.” Delaying the need for transplantation byimproving function, therefore, becomes a primeconcern. Among patients in this study, all 26 CRTrecipients did initially improve their quality of lifeand were removed from transplant consideration.Among these, 7 of 26 (27%) (excluding the patientwho improved but later underwent mitral valve sur-gery and died of surgical complications), not unex-pectedly, either had eventual deterioration offunction requiring transplant or died. This is incomparison to 7 of 14 (50%) of the non-CRT recipientswho followed standard medical management. Amongpatients in this study, the need for transplantationwas delayed, on average more than 4 years with amaximum of 12 years, following CRT. Moreover, 18 of26 (69%) of CRT recipients remained off the trans-plant list with improved function averaging 3 yearsafter CRT implant, attesting to long-term benefits ofpre-selecting those CHD patients who are most likelyto benefit from CRT.

STUDY LIMITATIONS. As a single-center study, pa-tient numbers are limited, and the required referralfor CRT testing added a selection bias depending onphysician perception of worth of CRT as additionalmanagement for CHD patients with HF. The choice ofthe $15% improvement in measured dP/dt-max wasarbitrary and somewhat based on the 5% to 10%improvement in EF as previously published as anindication of CRT success. However, it was shown tobe an effective cutoff for showing significance of CRTeffect. This was also a retrospective study, and due tocompliance issues, not all patients were followed asrecommended. Some patient deaths were directlyrelated to noncompliance. Nevertheless, the use of

PERSPECTIVES

COMPETENCY IN MEDICAL KNOWLEDGE:

Although there are published guidelines on which

patients should benefit from and therefore potentially

receive CRT pacing for HF, these guidelines do not

include patients with CHD. Therefore, clinicians

dealing with CHD patients have attempted to

extrapolate published inclusion criteria with mixed

results. Pre-selecting patients on the basis of

individualized and directly measured paced CRT

contractility provides a more evidence-based and

patient-specific analysis of determining CRT efficacy

regardless of CHD anatomy.

TRANSLATIONAL OUTLOOK: Multicenter clinical

studies are needed to assess the wide application of

directly measured contractility in the pre-selection of

CHD patients for CRT pacing. Correlation with other

noninvasive measured ventricular contractility

variables may facilitate future clinical applications.

Karpawich et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7

Congenital Heart Long-Term CRT A U G U S T 2 0 1 7 : 8 3 0 – 4 1

840

pre-CRT implantation hemodynamic contractilitytesting (dP/dt-max) was shown to be an effectiveclinical tool among CHD patients with HF, regardlessof diverse anatomies and surgical repairs, to pre-determine potential responders.

CONCLUSIONS

A surgically repaired congenital heart is nevernormal, and patients with a congenital heart remaina challenge to all physicians and health care pro-viders. Early HF, unfortunately, can be an eventualoutcome for many patients depending on anatomicand physiological issues. Not every patient improvesafter CRT, and by no means can CRT be considered apanacea or cure for HF, especially among patientswith structurally intrinsic ventricular anatomies thatwill eventually fail. However, clinical symptompalliation and improvement in patient quality of life,while delaying the need for transplant listing, can bean effective therapy. Current published guidelinesfor CRT patient selection simply do not apply to CHDpatients. Proper patient screening is imperative.More studies on this select, but growing, patientpopulation, are required. With the publishedrecommendation to find a better way to improveCHD patient selection, short-term pre-testing of CRTefficacy, as it applies to the most importantmeasured variable of paced ventricular contractility(dP/dt-max) response before CRT implantation, canbe at least 1 clinical tool in CHD patient “responder”selection. Once effectively performed, CRT canimprove clinical outcomes, improve patient

well-being, and act as an effective bridge to delaythe need for heart transplant listing, often for manyyears.

ADDRESS FOR CORRESPONDENCE: Dr. Peter P.Karpawich, Cardiac Electrophysiology, The Children’sHospital of Michigan, 3901 Beaubien Boulevard,Detroit, Michigan 48201. E-mail: pkarpawi@dmc.org.

RE F E RENCE S

1. Dipchand A, Rossano J, Edwards L, et al. Theregistry of the International Society for Heart andLung Transplantation: eighteenth official pediatricheart transplantation report—2015; focus theme:early graft failure. J Heart Lung Transplant 2015;34:1233–43.

2. Auricchio A, Blanc J, Daubert J, et al. Guidelinesfor cardiac pacing and cardiac resynchronizationtherapy. Eur Heart J 2007;28:2256–95.

3. Burns K, Gage R, Curtin A, Bank A. Long-termechocardiographic response to cardiac resynchro-nization therapy in initial nonresponders. J AmColl Cardiol HF 2015;3:990–7.

4. Khairy P, Van Hare GF, Balaji S, et al. PACES/HRS expert consensus statement on the recogni-tion and management of arrhythmias in adultcongenital heart disease. Heart Rhythm 2014;11:e103–62.

5. Karpawich PP, SinghH, Zelin K. Optimizing pacedventricular function in patients with and withoutcongenital heart disease by contractility-guided

lead implant. Pacing Clin Electrophysiol 2015;38:54–62.

6. Prinzen FW, Hunter WC, Wyman BT,McVeigh ER. Mapping of regional myocardialstrain and work during ventricular pacing: experi-mental study using magnetic resonance imagingtagging. J Am Coll Cardiol 1999;33:1735–42.

7. Van Oosterhout MF, Prizen FW, Arts T, et al.Asynchronous electrical activation induces asym-metrical hypertrophy of the left ventricular wall.Circulation 1998;98:588–95.

8. Hsu JC, Solomon SD, Bourgoun M, et al. Pre-dictors of super-response to cardiac resynchroni-zation therapy and associated improvement inclinical outcome: the MADIT-CRT (MulticenterAutomatic Defibrillator Implantation Trial WithCardiac Resynchronization Therapy) study. J AmColl Cardiol 2012;59:2366–73.

9. Rodriquez-Cruz E, Karpawich PP, Lieberman RA,Tantengco MV. Biventricular pacing as alternativetherapy for dilated cardiomyopathy associated

with congenital heart disease. Pacing Clin Elec-trophysiol 2001;24:235–7.

10. Dubin AM, Janousek J, Rhee E, et al. Resynch-ronization therapy in pediatric and congenital heartdisease patients: an international multi-centerstudy. J Am Coll Cardiol 2005;46:2277–83.

11. Janousek J, Gebauer RA, Abdul-Khaliq H,et al., Working Group for Cardiac Dysrhythmiasand Electrophysiology of the Association forEuropean Paediatric Cardiology. Cardiacresynchronization therapy in paediatric andcongenital heart disease: differential effects invarious anatomical and functional substrates.Heart 2009;95:1165–71.

12. Cecchin F, Frangini PA, Brown DW, et al. Car-diac resynchronization therapy (and multisitepacing) in pediatrics and congenital heart disease:five year experience in a single institution.J Cardiovasc Electrophysiol 2009;20:58–65.

13. Khairy P, Fournier A, Thibault B, Dubuc M,Therien J, Vobecky SJ. Cardiac resynchronization

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 8 , 2 0 1 7 Karpawich et al.A U G U S T 2 0 1 7 : 8 3 0 – 4 1 Congenital Heart Long-Term CRT

841

therapy in congenital heart disease. Int J Cardiol2006;109:160–8.

14. Streiper M, Karpawich P, Frias, et al. Initialexperience with cardiac resynchronization therapyfor ventricular dysfunction in young patients withsurgically operated congenital heart disease. Am JCardiol 2004;94:1352–4.

15. Motonaga K, Dubin AM. Cardiac resynchroni-zation therapy for pediatric patients with heartfailure and congenital heart disease: a reappraisalof results. Circulation 2014;129:1879–91.

16. Van der Hulst A, Delgado V, Blom N, et al.Cardiac resynchronization therapy in paediatricand congenital heart disease patients. Eur Heart J2011;32:2236–46.

17. Tompkins C, Kutyifa V, McNitt S, et al. Effecton cardiac function of cardiac resynchronizationtherapy in patients with right bundle branch block(MADIT-CRT) trial. Am J Cardiol 2013;112:525–9.

18. Daubert JC, Saxon L, Adamson PB, et al. forthe European Heart Rhythm Association, EuropeanSociety of Cardiology, Heart Rhythm Society,Heart Failure Society of America, American Societyof Echocardiography, American Heart Association,European Association of Echocardiography, HeartFailure Association. 2012 EHRA/HRS expertconsensus statement on cardiac resynchronizationtherapy in heart failure: implant and follow-uprecommendations and management. HeartRhythm 2012;9:1524–76.

19. Srinivasan C, Sachdeva R, Morrow WR,Greenberg SB, Vyas HV. Limitations of standardechocardiographic methods for quantification ofright ventricular size and function in children andyoung adults. J Ultrasound Med 2011;30:487–93.

20. Davlouros PA, Niwa K, Webb G, Gatzoulis MA.The right ventricle in congenital heart disease.Heart 2006;92 Suppl 1:i27–38.

21. Becker M, Hümpel C, Ocklenburg C, et al. Theright ventricular response to high afterload:comparison between healthy persons and patients

with transposition of the great arteries: a 2D strainstudy. Echocardiography 2010;27:1256–62.

22. Chung ES, Leon AR, Tavazzi L, et al. Results ofthe predictors of response to CRT (PROSPECT)trial. Circulation 2008;117:2608–16.

23. Selly JB, Iriart X, Roubertie F, et al. Multi-variate assessment of the right ventricle byechocardiography in patients with repairedtetralogy of Fallot undergoing pulmonary valvereplacement: a comparative study with magneticresonance imaging. Arch Cardiovasc Dis 2015;108:5–15.

24. Punn R, Hanisch D, Motonaga KS,Rosenthal DN, Ceresnak SR, Dubin AM. A pilotstudy assessing ECG versus ECHO ventriculoven-tricular optimization in pediatric resynchronizationpatients. J Cardiovasc Electrophysiol 2016;27:210–6.

25. Thibauli B, Harel F, Ducharme A, et al.,LESSER-EARTH Investigators. Cardiac resynchro-nization therapy in patients with heart failure anda QRS complex < 120 ms: the evaluation ofresynchronization therapy for heat failure(LESSER-EARTH) trial. Circulation 2013;127:873–81.

26. Bleeker GB, Holman ER, Steendijk P, et al.Cardiac resynchronization therapy in patients witha narrow QRS complex. J Am Coll Cardiol 2006;48:2243–50.

27. Thibault B, Dubuc M, Karst E, et al. Design ofan acute dP/dt hemodynamic measurement pro-tocol to isolate cardiac effect of pacing. J Card Fail2014;20:365–72.

28. Adler D, Monrad ES, Hess OM,Krayenbuehl HP, Sonnenblick EH. Time to dP/dt(max), a useful index for evaluation of contractilityin the catheterization laboratory. Clin Cardiol1996;19:397–403.

29. Rhodes J, Udelson JE, Marx GR, et al. A newnon-invasive method for the estimation of peakdP/dt. Circulation 1993;88:2693–9.

30. Butts RJ, Chowdhury SM, Buckley J, et al.,MOCHA Investigators. Comparison of echocardio-graphic and pressure-volume loop indices of sys-tolic function in patients with single ventriclephysiology: a preliminary report. Congenit HeartDis 2015;10:17–24.

31. Lieberman R, Padeletti L, Schreuder J, et al.Ventricular pacing lead location alters systemichemodynamics and left ventricular functionin patients with and without reduced ejectionfraction. J Am Coll Cardiol 2006;48:1634–41.

32. Janousek J, Tomek V, Chaloupecky V, et al.Cardiac resynchronization therapy: a novel adjunctto treatment and prevention of systemic rightventricular failure. J Am Coll Cardiol 2004;44:1927–31.

33. Plymen C, Finlay M, Tsang V, et al. Haemo-dynamic consequences of targeted single- anddual-site right ventricular pacing in adults withcongenital heart disease undergoing surgical pul-monary valve replacement. Europace 2015;17:274–80.

34. Thambo J, Guillebon M, Xhaet O, et al.Biventricular pacing in patients with Tetralogyof Fallot: non-invasive epicardial mappingand clinical impact. Int J Cardiol 2013;163:170–4.

35. Husain N, Gokhale J, Nicholson L, et al.Comparing echocardiographic assessment of sys-tolic function with catheterization data in patientswith single right ventricles. Acta Cardiol 2014;69:281–8.

36. Sogaard P, Egeblad H, Pedersen A, et al.Sequential versus simultaneous biventricularresynchronization for severe heart failure. Evalu-ation by tissue Doppler imaging. Circulation 2002;106:2078–84.

KEY WORDS congenital heart, dP/dt-max,heart failure, resynchronization pacing,ventricular contractility