cardiac resynchronization therapy may be...

Listen to this manuscript’s

audio summary by JACC:

Clinical Electrophysiology

Editor-in-Chief

Dr. David J. Wilber.

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 VO L . 2 , N O . 3 , 2 0 1 6

ª 2 0 1 6 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O UN 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 5 . 1 0 . 0 0 7

Cardiac Resynchronization TherapyMay Be Antiarrhythmic Particularlyin RespondersA Systematic Review and Meta-Analysis

Aditya Saini, MD,a Munish Kannabhiran, MD,a Pratap Reddy, MD,a Rakesh Gopinathannair, MD, MA,b

Brian Olshansky, MD,c Paari Dominic, MDa

ABSTRACT

Fro

Un

Lo

be

Bri

Me

Da

Ma

OBJECTIVES This study sought to study the effect of echocardiographic response to cardiac resynchronization therapy

(CRT) on ventricular arrhythmias (VA). The effect of CRT-defibrillator on sustained VA was compared with implantable

cardioverter-defibrillator (ICD)-only therapy.

BACKGROUND CRT is an effective adjunctive therapy in selected patients with advanced congestive heart failure, but

its effect on VA remains controversial.

METHODS PubMed was searched to identify studies. For primary comparison, studies reporting incidence of VA in

patients with congestive heart failure with CRT compared with ICD were included. For secondary comparison, studies

reporting incidence of VA in echocardiographic responders compared with nonresponders were included. Studies

reporting incidence of VA in CRT nonresponders before and after CRT upgrade from ICD were assessed for the third

comparison. Inverse variance method in a random-effects model was used to combine effect sizes.

RESULTS Thirteen studies (4,631 subjects) were included in the primary meta-analysis. Patients with CRT had a signifi-

cantly lower incidence of VA comparedwith patients with ICD only (odds ratio: 0.754; confidence interval: 0.594 to 0.959).

Thirteen studies (n¼ 3,667) were included in themeta-analysis of VA in CRT responders versus nonresponders. Responders

had a significantly lower risk of VA (odds ratio: 0.436; confidence interval: 0.323 to 0.589). Multivariate meta-regression

showed that the percentage beta-blocker use and follow-up duration explained heterogeneity between the studies. Three

studieswere included in the comparison of VA in CRT nonresponders before and after upgrade from ICD. CRT nonresponders

had an elevated risk of VA compared with ICD-only subjects (odds ratio: 1.497; confidence interval: 1.225 to 1.829).

CONCLUSIONS CRT may significantly reduce risk of VA compared with ICDs in patients who meet criteria for CRT. CRT

responders have significant reduction in VA compared with nonresponders. CRT nonresponse might significantly increase

risk of VA. (J Am Coll Cardiol EP 2016;2:307–16) © 2016 by the American College of Cardiology Foundation.

C ardiac resynchronization therapy (CRT) is aneffective adjunctive therapy for patientswith advanced systolic heart failure and

wide QRS duration (QRSd) who are taking guideline-directed medical therapy. CRT can improve

m the aDivision of Cardiology, Department of Medicine, and Center for C

iversity Health Sciences Center- Shreveport, Shreveport, Louisiana; bD

uisville, Louisville, Kentucky; and the cCardiovascular Medicine, Universi

en a consultant for St. Jude Medical, Abiomed, HealthTrust PG, and Boston

stol-Myers Squibb; and has received honoraria from Boston Scientific. Dr. O

dtronic, Biotronik, Amgen, Daiichi Sankyo, Boehringer Ingelheim, Amarin

iichi Sankyo. All other authors have reported that they have no relationsh

nuscript received September 21, 2015; accepted October 22, 2015.

symptoms, exercise capacity, quality-of-life, andlong-term survival, and can reduce congestiveheart failure (CHF) hospitalizations (1,2). The mecha-nism of benefit from CRT is related to improvedleft ventricular (LV) contractile function and

ardiovascular Diseases and Sciences, Louisiana State

ivision of Cardiovascular Medicine, University of

ty of Iowa, Iowa City, Iowa. Dr. Gopinathannair has

Scientific; has received speaker fees from Pfizer and

lshansky has been a consultant for Boston Scientific,

, On-X, BioControl, and Lundbeck; and a speaker for

ips relevant to the contents of this paper to disclose.

https://s3.amazonaws.com/jaccep/jaccep0203/JACCEP0203_wilbersummary_06





http://crossmark.crossref.org/dialog/?doi=10.1016/j.jacep.2015.10.007&domain=pdf

http://dx.doi.org/10.1016/j.jacep.2015.10.007

ABBR EV I A T I ON S

AND ACRONYMS

CHF = congestive heart failure

CI = confidence interval

CRT = cardiac

resynchronization therapy

CRT-D = cardiac

resynchronization therapy–

defibrillator

ICD = implantable

cardioverter-defibrillator

LV = left ventricle

LVEF = left ventricular ejection

fraction

LVESV = left ventricular

end-systolic volume

OR = odds ratio

QRSd = QRS duration

VA = ventricular arrhythmia

VT = ventricular tachycardia

Saini 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 . 2 , N O . 3 , 2 0 1 6

CRT May Be Antiarrhythmic Particularly in Responders J U N E 2 0 1 6 : 3 0 7 – 1 6

308

reverse remodeling (3,4). Although CRT canreduce mortality and CHF hospitalizations,its effect on ventricular arrhythmias (VA)defined as ventricular tachycardia (VT) orfibrillation episodes that stop only after anti-tachycardia pacing and/or shocks (or didnot otherwise stop spontaneously), as docu-mented by device interrogations, remainscontroversial.

Various, mostly nonrandomized, prospec-tive and retrospective trials have studied thisissue, but results have not been consistent.Although some studies (5,6) have demon-strated significant reduction in risk of VAwith active CRT (particularly in subjects withleft bundle branch block morphology), othershave failed to demonstrate any significanteffect of CRT on the incidence of VA whencompared with implantable cardioverter-defibrillator (ICD)-only therapy (3,7,8).

SEE PAGE 317

Individual studies have suffered the limitation ofsmall sample size, nonrandomized and retrospectivestudy designs, and single center experiences. Tocomplicate the issue further, multiple case reportspoint toward a proarrhythmic effect of CRT (9,10).Several recent studies have shown that the effect ofCRT on VA is determined by the response to CRT(8,11–24). Although multiple studies show a signifi-cantly reduced incidence of VA in CRT respondersversus nonresponders, 4 published studies (11–14)show no significant difference or only a trend to-ward reduction in incidence of VA when comparingCRT responders with nonresponders. Three of thesestudies have relatively large sample sizes.

Hence, we performed a systematic review andcombined the data using meta-analytical techniquesto strengthen the level of evidence and providedeeper insight into the issue of response-related dif-ferences in incidence of VA in CRT patients. In thismeta-analysis, we compared: 1) the effect of CRT-defibrillator (CRT-D) versus ICD-only therapy onsustained VA including VT and ventricular fibrilla-tion; 2) the incidence of VA in echocardiographic re-sponders versus nonresponders to CRT; and 3) theincidence of VA among CRT nonresponders beforeand after upgrade from ICD-only status.

METHODS

Our meta-analysis is in accordance with recommen-dations of the Meta-Analysis of Observational studiesin the Epidemiology Group (25).

INCLUSION CRITERIA. For pr imary meta-analys i scompar ing VA in CRT-D with ICD. Studies (retro-spective and prospective; randomized and non-randomized) of CRT-D in patients with CHF withstandard indications (left ventricular ejection fraction[LVEF] #35%; QRSd >120 ms) were included if theyreported the incidence of VA and compared thembetween patients with CRT-D and ICD or betweenpre- and post-CRT-D upgrade from ICD. Studies withmean follow-up of $3 months post-CRT-D implantwere included.For the secondary meta-ana lys i s compar ing VAin CRT responders with nonresponders . Studiescomparing the incidence of sustained VA in patientswith CHF with echocardiographic response to CRTand those with no response to CRT were included.Echocardiographic response definitions shouldinclude 1 of the following: 1) post-implant LVEFof $35%; 2) post-implant improvement in LVEFof $5%; or 3) post-implant left ventricular end-systolic volume (LVESV) or LVESV index reductionby $10%. Studies with mean echocardiographicfollow-up duration of $6 months post-CRT implantwere included.For the secondary meta-ana lys i s compar ing VAin CRT nonresponders with ICD-only pat ients .Studies comparing the incidence of sustained VA inpatients with CHF with echocardiographic nonre-sponse to CRT before and after CRT upgrade of ICDswere included.

EXCLUSION CRITERIA. Studies were excluded ifthey: 1) lacked a control group; 2) were published onlyin abstract form; or 3) were non-English studies withno English translation.Search st rateg ies . We searched MEDLINE (1966 to2014), Cochrane Database of systematic reviews (1999to 2014), and Google Scholar using keywords: cardiacresynchronization therapy, CRT, biventricular pacingAND ventricular arrhythmia, VT, ventricular fibrilla-tion in various combinations. The “Related Article”feature on PubMed, and a manual search of refer-ences was also used to identify additional studies.We reviewed full text of relevant articles. Englishtranslations if necessary were obtained. Titles andabstracts were independently reviewed by 2 re-viewers (A.S. and M.K.) and cross-verified for inclu-sion. Details of search strategy are reported inFigure 1.

DATA EXTRACTION AND ASSESSMENT OF STUDY

QUALITY. For each included study, all data elementsuniformly reported across most studies were extrac-ted by 2 reviewers (A.S. and M.K.) and are shownin Tables 1 and 2. The quality of each study was

FIGURE 1 PRISMA Flow Diagram Illustrating Search Strategies

Search strategies and screening of studies for inclusion in the meta-analysis.

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 . 2 , N O . 3 , 2 0 1 6 Saini et al.J U N E 2 0 1 6 : 3 0 7 – 1 6 CRT May Be Antiarrhythmic Particularly in Responders

309

evaluated in accordance with the guidelines of U.S.Preventive Task Force and the Evidence-Based Man-agement Group (26,27). The following characteristicswere assessed: 1) clear inclusion and exclusioncriteria; 2) study sample representative of the popu-lation; 3) explanation of sample selection; 4) fullspecification of clinical and demographic variables;5) reporting loss of follow-up; 6) clear definition ofoutcomes and outcome assessment; and 7) adjust-ment of possible confounders in multivariate anal-ysis. Studies were graded as “poor” if they met 3criteria, “fair” if they met 3 to 5 criteria, and “good” ifthey met >5 criteria. The quality assessment of indi-vidual studies is reported alongside baseline variablesin Tables 1 and 2. All disagreements between re-viewers were resolved by consensus.

STATISTICAL METHODS. The percentage of patientswith VA in patients with CRT-D and ICD reportedacross studies was used in this meta-analysis togenerate degree of association between CRT and VA.Individual studies did not use Cox proportional haz-ards model or propensity score model to adjust forpotential confounders, which is a limitation of ouranalysis. For all analyses, odds ratio (OR), log OR, andstandard errors were calculated from events andsample sizes.

In the secondary meta-analyses, comparing risk ofVA in CRT nonresponders before and after upgradefrom an ICD, 1 study by Ouellet et al. (5) reported onlyhazard ratios and we therefore adopted the reportedhazard ratio as the best estimate of OR. Also, the hazardratio reported by this study for cumulative incidence ofsubsequent VA after the first event (in cases of elec-trical storm, only the first VA event in any 24-h periodfollow-up period was included), was taken as thesubstitute for general incidence of VA for the purposeof the third analysis comparing VA in CRT non-responders before and after upgrade. We used the in-verse variance method to achieve a weighted estimateof the combined overall effect for all the analyses.

We assessed the results for heterogeneity in ouranalysis by examining the forest plots and thencalculating a Q statistic, which we compared with theI2 index. We considered the presence of significantheterogeneity at the 5% level of significance (for the Qtest) and values of I2 exceeding 56% as an indicator ofsignificant heterogeneity. Although the Q statistic forthe primary analysis showed statistical significance(p ¼ 0.02), I2 index was 49, indicating mild hetero-geneity among studies. For the secondary analysescomparing VA in CRT responders with non-responders, Q statistic (p ¼ 0.003) and I2 index (60)indicated moderate heterogeneity among studies.

This prompted us to adopt the random-effects modelto pool effect sizes and all analyses were performedusing Comprehensive Meta-analysis version 3 (Bio-stat, Inc., Englewood, New Jersey).

The underlying heterogeneity prompted us toperform meta-regression analysis to investigate fac-tors contributing to heterogeneity and if our studyoutcome (VA) was affected by factors other than ourprimary treatment (CRT and response to CRT). Forthis, we adopted a weighted regression random-effects model and carried out a multivariate regres-sion using up to 3 pre-determined factors usingComprehensive Meta-analysis version 3. These vari-ables were selected on the basis of factors shown toaffect VA in individual studies and on availability ofdata for most of the studies included. A 2-sidedp value <0.05 was regarded as significant for all an-alyses. Data were represented as forest plots for pri-mary analysis. Potential publication bias wasassessed with the Egger test and represented graph-ically with Begg funnel plots of the natural log of theOR versus its standard error.

TABLE 1 Characteristics of 13 Studies Included in the Meta-Analysis of CRT-D Versus ICD Therapy in the Incidence of Ventricular Arrhythmias

First Author,Year (Ref. #)

n (Study)/n (Control) Study Design

MeanFollow-Up(Months)

Age(yrs, Mean)

Male(%)

ICM(%)

Mean EF(%)

MeanQRS

Duration(ms)

ACE-I/ARB(%)

BB(%)

Antiarrhythmic(%)

LBBB(%)

QualityAssessment

Higgins et al.,2003 (7)

245 (CRT-D on)/245(CRT off, ICD)

Prospective,randomizedmulticenter study

6 66 84 69 21.5 158 87.5 47 NR 54.5 Good

Young et al., 2003The MIRACLE ICD

Trial (2)

187 (CRT-D on)/182(ICD, CRT off, ICD)

Randomizedcontrolled trial

6 67 76.6 70 24 163.5 90.8 60.16 37.7 NRIsolated

RBBB:13%

Good

Arya et al.,2005 (35)

31 (CRT-D)/34 (ICD) Retrospective,single-centercase-control

11 58 74 55 24.1 147.4 85.6 46.8 57.7 NR Good

Ermis et al.,2005 (36)

18 (pre-CRT upgrade,ICD)/18 (post-CRTupgrade)

Pre-upgrade:retrospective

Post-upgrade:prospective

Pre-upgrade: 47Post-upgrade: 14

69 83 61 21 NR 77 44 39 NR Fair

Voigt et al.,2005 (37)


Retrospective,single-centerstudy

Pre-upgrade: 13.3Post-upgrade: 5.4

67 94.7 84 24 171 95 79 79 NR Fair

Linde et al., 2008REVERSEStudy (3)

345 (CRT-D)/163(CRT off, ICD)

Randomizedcontrolled trial

12 62.2 79.5 57 26.7 153 96.2 95.8 10.8 NR Good

Nordbeck et al.,2009 (38)

168 (CRT-D)/561 (ICD) Retrospectiveanalysis

41 62.4 66.3 79.14 31.74 159 NR NR NR NR Good

Blaschke et al.,2011 (6)

67 (CRT-D)/67 (ICD) Prospective cohort 31 62.5 82 53 24 139 95.5 94 18 CRT 100%ICD: none

Good

Thijssen et al.,2011 (8)


Retrospectiveanalysis ofprospective data

Pre-upgrade: 54Post-upgrade: 37

65 81 75 26 167 90 77 40 NR Good

Timoteo et al.,2011 (18)


12 63 79 50.49 25 NR NR 82.6 15.51 NR Good

Ouellet et al.,2012 (5)

1,089 (CRT-D)/731 (ICD):because of cross-overs, efficacyanalysis wasperformedrecognizing activedevice type duringfollow-up

Substudy of MADITCRT, arandomizedcontrolled trial

28.8 64.6 75.1 55 22 NR 97.6 93.25 7.9 70.4 Good

Gopalamuruganet al., 2014 (29)


24.3 65.76 73.9 48.9 25.62 NR 95 80 18 NR Good

Gopinathannairet al., 2014 (28)

31 (CRT-D, CF-LVAD)/30(ICD, CF-LVAD)

Retrospective,nonrandomizedsingle-centerstudy

22.7 58.52 78.7 60.7 21.06 137.28 62.3 82 42.6 NR Fair

ACE-I ¼ angiotensin-converting enzyme inhibitors; ARB ¼ angiotensin receptor blockers; BB ¼ beta-blockers; CF-LVAD ¼ continuous flow left ventricular assist device; CRT-D ¼ cardiac resynchronization therapy–defibrillator; EF ¼ ejection fraction; ICD ¼ implantablecardioverter-defibrillator; ICM ¼ ischemic cardiomyopathy; LBBB ¼ left bundle branch block; MADIT CRT ¼ Multicenter Automatic Defibrillator Implantation Trial with Cardiac Resynchronization Therapy; MIRACLE-ICD ¼ The Multicenter InSync ICD Randomized ClinicalEvaluation; NR ¼ not reported; RBBB ¼ right bundle branch block; REVERSE ¼ Resynchronization Reverses Remodeling In Systolic Left Ventricular Dysfunction.

Sainietal.

JACC:CLIN

ICAL

ELECTROPHYSIO

LOGY

VOL.2,NO.3,2016

CRTMay

BeAntiarrhythm

icParticularly

inResponders

JUNE

2016:3

07–16

310

TABLE 2 Characteristics of 13 Studies Included in the Meta-Analysis of CRT Responders Versus Nonresponders in the Incidence of Ventricular Arrhythmias

First Author,Year (Ref. #)

No. of Patients inCRT-Responders/Nonresponders Responder Criteria Study Design

MeanFollow-Up(Months)

Age(yrs, Mean)

Male(%)

ICM(%)

Mean EF(%)

MeanLVESV(ml)

QRSDurationMean (ms)

ACE-I/ARB(%)

BB(%)

Antiarrhythmic(%)

QualityAssessment

Di Biase et al., 2008 (11) 227/171 LVESV $10% reductionafter 6 months

Prospective, nonrandomized,cohort (InSync ICD ItalianRegistry)

12 66 88 67 26 178 166 80 60 40 Good

Markowitz et al., 2008 (13) 71/127 LVESV $15% reductionafter 6 months

Retrospective analysis ofprospective, randomizedmulticenter InSync IIIMarquis study

6 66.8 79.8 78 24.8 226.2 155 92 89.4 NR Good

Schaer et al., 2010 (21) 57/213 LVEF >35% at mean20 months follow-up

Retrospective analysis ofprospective registry data

40 60.9 77.4 47.8 21.9 NR 165 93.7 76.7 29.3 Good

Gold et al., 2011 (24) 144/136 LVESVi $15% reductionafter 12 months

Retrospective analysis ofmulticenter, randomizedREVERSE study

12 62.7 79 59 27 NR 152 96 97 10 Good

Thijssen et al., 2011 (8) 70/45 LVESV >15% reductionat 6 months

Retrospective analysis ofprospective data

6 65 81 75 26 169 167 90 77 40 Good

Eickholt et al., 2012 (15) 74/52 LVEF >10% increase after6 months

Prospective, nonrandomizedstudy

28 64 67 52 24.6 NR 159.2 96.8 91.3 27 Good

Shahrzad et al., 2012 (22) 69/50 LVEF >5% increase orLVESV >10%reduction after6 months

Prospective, nonrandomizedcohort

34 59.9 73.9 41 19.87 181.2 NR 97.5 100 2.5 Good

Itoh et al., 2013 (17) 49/35 LVESV $15% reductionafter 6 months

Retrospective cohort study 12 67.5 73 30 26.3 132 170 85 86 21 Good

Manfredi et al., 2013 (19) 49/240 Post-implant EF >45%at follow-up

Retrospective single-centerstudy

7.2 median 71 72 59 20 NR 155 83.7 97 18 Fair

Friedman et al., 2014 (16) 172/156 LVEF >5% increase at6 months

Retrospective analysis ofprospective cohort study

36 68 78 61 25 NR 162 81 90 25.9 Good

Garcia-Lunar et al., 2014 (12) 114/82 LVEF >5% increase after12 months

Retrospective, single-centerstudy

30.1 median 62.8 85.2 46.4 25.6 NR 161.1 84.6 86.7 37.2 Good

Ruwald et al., 2014 (20) 649/103 Post-implant LVEF >35%after 12 months

Retrospective landmarkanalysis of multicenterrandomized MADIT-CRTstudy

26 64 75.3 54.8 24 177 158.7 95 93.8 8.2 Good

van der Heijden et al., 2014 (14) 310/202 LVESVi >15% reductionafter 6 months

Prospective, single-center,nonrandomized

57 median 68 73 53 24 NR 164 89 75 24 Good

LVEF ¼ left ventricular ejection fraction; LVESV ¼ left ventricular end-systolic volume; LVESVi ¼ left ventricular end systolic volume index; other abbreviations as in Table 1.

JACC:CLIN

ICAL

ELECTROPHYSIO

LOGY

VOL.2,NO.3,2016

Sainietal.

JUNE

2016:3

07–16

CRTMay

BeAntiarrhythm

icParticularly

inResponders

311

FIGURE 2 Forest Plot of VA in Patients With CRT-D Compared With Patients With ICDs

Meta-analysis shows a significantly lower incidence of sustained ventricular arrhythmia (VA) in cardiac resynchronization therapy–defibrillator (CRT-D) group in

comparison with implantable cardioverter-defibrillator (ICD)-only group, hence favoring CRT-D therapy. CI ¼ confidence interval.



312

RESULTS

META-ANALYSIS OF CRT-D VERSUS ICD. Thirteenstudies with 4,631 subjects were included in the pri-mary meta-analysis (Table 1). The mean baselineLVEF was 21% to 32%. All studies included patientswith QRSd >120 ms in the CRT-D group and had meanfollow-up duration of 6 to 41 months. The CRT-Dgroup, across all studies, was comparable and metcurrently applicable guidelines for CRT-D but thecontrol group with ICDs only varied across studies(Online Table 1).

Only the study by Gopinathannair et al. (28) in-cludes New York Heart Association class IV patientswith continuous-flow left ventricular assist device inboth study and control population. Because defibril-lator indications included primary and secondaryprevention of sudden cardiac death, patients onconcurrent antiarrhythmic drug therapy included inthe individual studies varied (Table 1) but were notstatistically different between groups within eachstudy.

VA definitions are included in Online Table 2. De-vice settings varied across studies (Online Table 2).Only 3 studies (5–7) reported on subjects with leftbundle branch block in addition to the QRSd (Table 1)and most of these did not report VA incidence byspecific bundle branch block patterns. Hence, toavoid inaccuracies given a small number of studiesreporting bundle branch block patterns, a meta-regression analysis using specific bundle branchblock pattern as a variable was not performed. Of the

studies that showed no statistically significant dif-ference in VA between groups (Figure 2), a significantdifference in improvement of 1 or more echocardio-graphic parameters (LVESV, LVEF, left ventricularend-diastolic volume) between post-CRT and ICD-only group was documented in all but 2 studieswhere post-CRT remodeling data were either not re-ported (29) or were not significantly different be-tween groups (28).

Analysis of the funnel plot for the primary analysisshowed no significant publication bias (OnlineFigure 5). In our pooled analysis, we found a signifi-cantly lower incidence of VA in CRT-D group incomparison with ICD-only group, hence favoring CRTtherapy (OR: 0.754; 95% confidence interval [CI]:0.594 to 0.959; p ¼ 0.021) (Figure 2). We performed asensitivity analysis by excluding the MADIT-CRTsubstudy (5), which had a larger sample size, but asignificant difference in outcome persisted, confirm-ing our finding.

Because only the study by Gopinathannair et al.(28) included left ventricular assist device patients inwhom the indication and programming of CRT re-mains controversial, we performed a second sensi-tivity analysis excluding this study, which showedresults similar to our overall analysis (OR: 0.714; 95%CI: 0.554 to 0.922; p ¼ 0.01). Finally, a meta-regression with duration of follow-up as a covariatedid not show any correlation to HR for VA, suggest-ing that the differences in the duration of follow-upbetween studies did not explain differences inoutcome.

FIGURE 3 Forest Plot of VA in CRT Responders Compared With CRT Nonresponders

Meta-analysis found a significantly lower incidence of VA in echocardiographic responders in comparison with echocardiographic nonresponders, hence favoring CRT

responders. Abbreviations as in Figure 2.


313

META-ANALYSIS OF CRT RESPONDERS VERSUS

NONRESPONDERS. Thirteen studies with 3,667 sub-jects were included in meta-analysis of VA in CRTresponders versus nonresponders (Table 2). Ninestudies were retrospective and 4 were prospective.All studies fulfilled 1 of the 3 echocardiographicresponse criteria mentioned in the inclusion criteria.Baseline study characteristics are shown (Table 2).Follow-up durations ranged from 6 to 57 months. VAdefinitions are included in Online Table 3. Devicesettings varied across studies (Online Table 3). Anal-ysis of the funnel plot for the secondary analysisshowed lack of studies in the right lower corner,suggesting a possible publication bias (OnlineFigure 6). Our meta-analysis found a significantlylower incidence of VA in echocardiographic re-sponders in comparison with echocardiographicnonresponders, hence favoring CRT responders (OR:0.436; 95% CI: 0.323 to 0.589; p < 0.05) (Figure 3).

A multivariate meta-regression analysis showedthat the percentage beta-blocker use (p < 0.01) andduration of follow up (p < 0.01) together explained allthe heterogeneity between the studies with an R2

analog of 1 (Online Figures 1 and 2). Increased beta-blocker use in the sample population increased thepositive effect of CRT on incidence of VA in the CRTresponders. Additionally, the longer the populationwas followed the smaller was the difference inoutcome of VA between CRT responders and non-responders. Meta-regression analysis using QRSdand antiarrhythmic drug use were not signifi-cant in explaining heterogeneity. Criteria for CRTresponse did not add to the heterogeneity between

the studies in univariate or multivariate analysis(Online Figure 3).

META-ANALYSIS OF CRT NONRESPONDERS VERSUS

ICD-ONLY PATIENTS. Three studies (5,8,24) wereincluded in the pooled analysis of CRT nonrespondersversus ICD-only subjects. In the pooled analysis ofthese 3 studies, we found a significant difference inoccurrence of VA among CRT nonresponders com-pared with ICD-only subjects with higher incidence ofVA in CRT nonresponders (OR: 1.497; 95% CI: 1.225 to1.829; p < 0.05) (Online Figure 4). This is an importantfinding that may be clinically relevant because veryfew previous studies have explored this aspect andfurther randomized studies are required to clarify.

DISCUSSION

Our meta-analysis, the largest and most complete todate, assessing the impact of CRT therapy on VAshows that CRT decreases risk of VA compared withICD therapy alone. Within this group of patients, in-dividuals with echocardiographic response to CRTtherapy may have increased benefits with lowerincidence of sustained VA compared with non-responders. Finally, patients who do not show evi-dence of response by echocardiographic parametersto CRT after an upgrade from an ICD-only therapymay have higher incidence of VA compared with theirpre-upgrade status.

The impact of CRT on VA has been a matter ofdebate for more than a decade. Experimental datahave demonstrated proarrhythmic effects of CRT.



314

LV epicardial pacing in CRT has been shown toincrease the transmural dispersion of refractorinessand QT intervals (30). In addition, there are multiplecase reports of increased monomorphic and poly-morphic drug-refractory VT that occurred soon afterinitiation of CRT and resolved only after cessation ofbiventricular pacing (9,10).

On the other end of the spectrum, there is evidencethat acute biventricular pacing decreases the induc-ibility of sustained monomorphic VT in patients withischemic cardiomyopathy (31). Major randomizedcontrolled clinical trials (MIRACLE ICD [2], REVERSE[3], CONTAK CD [7]) did not detect a difference inoccurrence of VA between CRT-D and ICD. Although asubstudy of MADIT-CRT (5) showed that active CRTtreatment was associated with significant reductionin risk of life-threatening VA, evidence from othersmaller studies has been conflicting.

Given the neutral results obtained in most studiesdiscussed previously, recent studies have focused onassessing whether reverse remodeling or LVEFimprovement (echocardiographic response) with CRTtherapy is associated with reduction in VA occur-rence. Haugaa et al. (32) showed that persistent LVmechanical dyssynchrony by speckle-tracking echo-cardiography after CRT placement was associatedwith increased VA. A substudy of the REVERSE trialby Gold et al. (24) revealed that subjects whoresponded to CRT by reverse remodeling had areduced incidence of VA compared with those whodid not. The results of our meta-analysis help supportthe findings of these studies that suggest that anti-arrhythmic effects of CRT may be a function ofechocardiographic response to pacing.

Although no mechanistic conclusions can be made,our meta-analysis suggests that the reduced VAoccurrence in CRT patients occurs concurrently withfavorable echocardiographic responses to CRT, such asreverse remodeling and improved contractile func-tion. Increased beta-blocker use has been shown to beindependently associated with reduced LVESV andcardiac events after receiving CRT (33). Consequently,increased beta-blocker use in the sample populationincreased the positive effect of CRT on incidence of VAin the CRT responders in our meta-regression analysis.

If antiarrhythmic benefits of CRT are a function ofechocardiographic response, the outcome in subjectswithout a favorable echocardiographic response toCRT becomes important. We address this issue in ourstudy and demonstrate increased risk of VA in CRTnonresponders when compared with ICD-only ther-apy, underscoring the importance of selection ofcandidates for CRT. This outcome highlights the needfor careful selection of patients with borderline

indications for CRT, where there could be a signifi-cant risk of increasing VA with inadequate responseto CRT. Large randomized controlled clinical trials areneeded in the future to confirm the observations ofthis meta-analysis.STUDY LIMITATIONS. Most of the study designs wereobservational, retrospective, single center and hadinherent limitations. These studies did not reportadjusted hazard ratios from Cox proportional hazardsor propensity score models and therefore the effectsize data could suffer from known and unknownconfounders.

Device type, CRT lead location, and device settings,such as detection zone and tachyarrhythmia therapyalgorithms, were different across different studiesand were not uniformly reported. We recognize this asan important but inherent limitation of this meta-analysis because differences in device programmingparameters within a study and across individualstudies can potentially affect overall outcomes.

In the primary meta-analysis comparing VA in CRTwith ICD-only patients, the patients in the ICD armvaried between studies with some studies includingpatients with narrow QRS (meeting standard guide-lines for ICD) and others including patients with a wideQRS (>120ms). In themeta-analysis of CRT respondersversus nonresponders, we used 3 different responsecriteria on the basis of previously performed studies;however, a meta-regression analysis by responsecriteria did not find any significant contribution toheterogeneityby response criteriaused. Finally, for thethird analysis assessing VA risk in CRT nonresponderscompared with ICD-only therapy, we included only3 studies limiting the power of the analysis.

CONCLUSIONS

CRT may significantly reduce risk of sustained VAcompared with ICDs in patients who meet criteria forCRT therapy. CRT recipients with echocardiographicevidence of favorable response have a reduced risk ofVA versus nonresponders. CRT nonresponders have ahigher risk of VA compared with similar patients whoreceive ICD only. The potential clinical implicationsof these results need to be further explored withproperly powered, prospective randomized trials in acontrolled setting where device types and program-ming parameters can be standardized.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Paari Dominic, Division of Cardiology, Department ofMedicine, LSU Health Sciences Center, 1501 KingsHighway, Shreveport, Louisiana 71103. E-mail:[email protected].

mailto:[email protected]

PERSPECTIVES

COMPETENCY IN MEDICAL KNOWLEDGE: CRT

therapy may reduce the risk of sustained ventricular ar-

rhythmias compared with ICDs in patients who meet

criteria for CRT therapy. CRT recipients with echocardio-

graphic evidence of favorable response have significant

reduction in the risk of ventricular arrhythmias compared

with nonresponders.

TRANSLATIONAL OUTLOOK: Large prospective ran-

domized controlled trials are required to explore our ob-

servations that CRT nonresponders may have a

significantly higher risk of VA compared with patients

who receive only ICD.


315

RE F E RENCE S

1. Abraham WT, Fisher WG, Smith AL, et al. Cardiacresynchronization in chronic heart failure. N Engl JMed 2002;346:1845–53.

2. Young JB, Abraham WT, Smith AL, et al. Com-bined cardiac resynchronization and implantablecardioversion defibrillation in advanced chronicheart failure: the MIRACLE ICD Trial. JAMA 2003;289:2685–94.

3. Linde C, Abraham WT, Gold MR, et al., REVERSEStudy Group. Randomized trial of cardiacresynchronization in mildly symptomatic heartfailure patients and in asymptomatic patients withleft ventricular dysfunction and previous heartfailure symptoms. J Am Coll Cardiol 2008;52:1834–43.

4. Cleland JG, Daubert JC, Erdmann E, et al.,Cardiac Resynchronization-Heart Failure (CARE-HF) Study Investigators. The effect of cardiacresynchronization on morbidity and mortality inheart failure. N Engl J Med 2005;352:1539.

5. Ouellet G, Huang DT, Moss AJ, et al. Effect ofcardiac resynchronization therapy on the risk offirst and recurrent ventricular tachyarrhythmicevents in MADIT-CRT. J Am Coll Cardiol 2012;60:1809–16.

6. Blaschke F, Knaus T, Celebi O, et al. Ventriculartachycardia or ventricular fibrillation occurs lessoften in patients with left bundle branch block andcombined resynchronization and defibrillatorsthan in patients with narrow QRS and conventionaldefibrillators. Europace 2012;14:224–9.

7. Higgins SL, Hummel JD, Niazi IK, et al. Cardiacresynchronization therapy for the treatment ofheart failure in patients with intraventricular con-duction delay and malignant ventricular tachyar-rhythmias. J Am Coll Cardiol 2003;42:1454–9.

8. Joep Thijssen C, Jan Willem Borleffs, Delgado V,et al. Implantable cardioverter-defibrillator patientswho are upgraded and respond to cardiac resynch-ronization therapy have less ventricular arrhythmiascompared with nonresponders. J Am Coll Cardiol2011;58:2282–9.

9. Medina-Ravell VA, Lankipalli RS, Yan GX, et al.Effect of epicardial or biventricular pacing to pro-long QT interval and increase transmural dispersionof repolarization: does resynchronization therapy

pose a risk for patients predisposed to long QT ortorsade de pointes? Circulation 2003;107:740–4.

10. Shukla G, Chaudhry GM, Orlov M, et al. Po-tential proarrhythmic effect of biventricular pac-ing: fact or myth? Heart Rhythm 2005;2:951–6.

11. Di Biase L, Gasparini M, Lunati M, et al. Anti-arrhythmic effect of reverse ventricular remodel-ing induced by cardiac resynchronization therapy:the InSync ICD (Implantable Cardioverter-Defibrillator) Italian Registry. J Am Coll Cardiol2008;52:1442–9.

12. García-Lunar I, Castro-Urda V, Toquero-Ramos J, et al. Ventricular arrhythmias in super-responders to cardiac resynchronization therapy.Rev Esp Cardiol 2014;67:883–9.

13. Markowitz SM, Lewen JM, Wiggenhorn CJ,et al. Relationship of reverse anatomical remod-eling and ventricular arrhythmias after cardiacresynchronization. J Cardiovasc Electrophysiol2009;20:293–8.

14. van der Heijden AC, Höke U, Thijssen J, et al.Super-responders to cardiac resynchronizationtherapy remain at risk for ventricular arrhythmiasand benefit from defibrillator treatment. Eur JHeart Fail 2014;16:1104–11.

15. Eickholt C, Siekiera M, Kirmanoglou K, et al.Improvement of left ventricular function undercardiac resynchronization therapy goes along witha reduced incidence of ventricular arrhythmia.PLoS One 2012;7:e48926.

16. Friedman DJ, Upadhyay GA, Rajabali A, et al.Progressive ventricular dysfunction among non-responders to cardiac resynchronization therapy.Baseline predictors and associated clinical out-comes. Heart Rhythm 2014;11:1991–8.

17. Itoh M, Yoshida A, Fukuzawa K, et al. Time-dependent effect of cardiac resynchronizationtherapy on ventricular repolarization and ventric-ular arrhythmias. Europace 2013;15:1798–804.

18. Timóteo AT, Oliveira MM, Silva MN, et al.Incidence of ventricular arrhythmias in patientswith severe left ventricular systolic dysfunction: isthere a benefit after cardiac resynchronizationtherapy? Rev Port Cardiol 2011;30:823–8.

19. Manfredi JA, Al-Khatib SM, Shaw LK, et al.Association between left ventricular ejection

fraction post-cardiac resynchronization treatmentand subsequent implantable cardioverter defibril-lator therapy for sustained ventricular tachyar-rhythmias. Circ Arrhythm Electrophysiol 2013;6:257–64.

20. Ruwald MH, Solomon SD, Foster E, et al. Leftventricular ejection fraction normalization in car-diac resynchronization therapy and risk of ven-tricular arrhythmias and clinical outcomes: resultsfrom the Multicenter Automatic Defibrillator Im-plantation Trial With Cardiac ResynchronizationTherapy (MADIT-CRT) trial. Circulation 2014;130:2278–86.

21. Schaer BA, Osswald S, Di Valentino M, et al.Close connection between improvement in leftventricular function by cardiac resynchronizationtherapy and the incidence of arrhythmias in car-diac resynchronization therapy-defibrillator pa-tients. Eur J Heart Fail 2010;12:1325–32.

22. Shahrzad S, Soleiman NK, Taban S, et al. Theeffect of left ventricular (LV) remodeling onventricular arrhythmia in cardiac resynchroniza-tion therapy (CRT-D) patients (antiarrhythmiceffect of CRT). Pacing Clin Electrophysiol 2012;35:592–7.

23. Van Boven N, Bogaard K, Ruiter J, et al.Functional response to cardiac resynchronizationtherapy is associated with improved clinicaloutcome and absence of appropriate shocks.J Cardiovasc Electrophysiol 2013;24:316–22.

24. Gold MR, Linde C, Abraham WT, et al. Theimpact of cardiac resynchronization therapy on theincidence of ventricular arrhythmias in mild heartfailure. Heart Rhythm 2011;8:679–84.

25. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology:a proposal for reporting. JAMA 2000;283:2008–12.

26. Harris RP, Helfand M, Woolf SH, et al. Currentmethods of the US Preventive Services Task Force:a review of the process. Am J Prev Med 2001;20:21–35.

27. Levine M, Walter S, Lee H, et al. Users’ guidesto the medical literature. IV. How to use an articleabout harm. Evidence-Based Medicine WorkingGroup. JAMA 1994;271:1615–9.

http://refhub.elsevier.com/S2405-500X(15)00377-1/sref1












































































































































316

28. Gopinathannair R, Birks EJ, Trivedi JR, et al.Impact of cardiac resynchronization therapy onclinical outcomes in patients with continuous flowleft ventricular assist devices. J Cardiac Fail 2015;21:226–32.

29. Gopalamurugan AB, Ganesha Babu G,Rogers DP, et al. Is CRT pro-arrhythmic?A comparative analysis of the occurrence of ven-tricular arrhythmias between patients implantedwith CRTs and ICDs. Front Physiol 2014;5:334.

30. Fish JM, Brugada J, Antzelevitch C. Potentialproarrhythmic effects of biventricular pacing.J Am Coll Cardiol 2005;46:2340–7.

31. Zagrodzky JD, Ramaswamy K, Page RL, et al.Biventricular pacing decreases the inducibility ofventricular tachycardia in patients with ischemiccardiomyopathy. Am J Cardiol 2001;87:1208–10.

32. Haugaa KH, Marek JJ, Ahmed M, et al. Me-chanical dyssynchrony after cardiac resynchroni-zation therapy for severely symptomatic heart

failure is associated with risk for ventricular ar-rhythmias. J Am Soc Echocardiogr 2014;27:872–9.

33. Nabeta T, Inomata T, Iida Y, et al. Prognosticsignificance of beta-blocker up-titration in con-junction with cardiac resynchronization therapy inheart failure management. Heart Vessels 2015 Aug8 [E-pub ahead of print].

34. Lin G, Rea RF, Hammill SC, et al. Effect ofcardiac resynchronisation therapy on occurrenceof ventricular arrhythmia in patients withimplantable cardioverter defibrillators undergoingupgrade to cardiac resynchronisation therapy de-vices. Heart 2008;94:186–90.

35. Arya A, Haghjoo M, Dehghani MR, et al. Effectof cardiac resynchronization therapy on the inci-dence of ventricular arrhythmias in patients withan implantable cardioverter-defibrillator. HeartRhythm 2005;2:1094–8.

36. Ermis C, Seutter R, Zhu A, et al. Impact ofupgrade to cardiac resynchronization therapy on

ventricular arrhythmia frequency in patients withimplantable cardioverter-defibrillators. J Am CollCardiol 2005;46:2258–63.

37. Voigt A, Barrington W, Ngwu O, et al. Biven-tricular pacing reduces ventricular arrhythmicburden and defibrillator therapies in patients withheart failure. Clin Cardiol 2005;29:74–7.

38. Nordbeck P, Seidl B, Fey B, et al. Effect ofcardiac resynchronization therapy on the incidenceof electrical storm. Int J Cardiol 2010;143:330–6.

KEY WORDS biventricular pacing, CRT,ICD, ventricular arrhythmia, ventricularfibrillation, ventricular tachycardia

APPENDIX For supplemental tables andfigures, please see the online version of thisarticle.



















































cardiac resynchronization therapy may be...

Documents