Effect of pacing mode and pacing site on torsional and strain parameters and on coronary flow

transcript

From the De

Athens, ‘‘Alex

Reprint reque

peutics, Medi

Ave & Lourou

0894-7317/$3

Copyright 201

http://dx.doi.o

Effect of Pacing Mode and Pacing Site on Torsionaland Strain Parameters and on Coronary Flow

Elektra Papadopoulou, MD, PhD, Anna Kaladaridou, MD, PhD, John Mattheou, MD, PhD,Constantinos Pamboucas, MD, PhD, Sophia Hatzidou, MD, PhD, Anna Antoniou, MD, PhD,

and Savvas Toumanidis, MD, PhD, Athens, Greece

Background: Right ventricular apical pacing may induce detrimental effects on left ventricular function andcoronary flow. In this study, the effects of pacing site and mode on cardiac mechanics and coronary bloodflow were evaluated.

Methods: This prospective study included 25 patients who received dual-chamber pacemakers with the ven-tricular lead placed in the right ventricular apex and presented in sinus rhythm (SR) at their regularly scheduledvisits at the pacemaker clinic. Patients underwent complete transthoracic echocardiographic examinationswhile in SR, followed by noninvasive Doppler assessment of coronary flow in the left anterior descendingcoronary artery (LAD) and speckle-tracking echocardiography of short-axis planes in SR, atrial pacing (AAI-P), atrioventricular (dual-chamber) pacing (DDD-P), and ventricular pacing (VVI-P).

Results: Rotation of the base was significantly decreased with VVI-P compared with AAI-P. Left ventriculartwist decreased significantly with DDD-P compared with AAI-P. Circumferential strain of the base significantlydecreased with DDD-P and VVI-P compared with SR. The velocity-time integral of diastolic flow in the LADdecreased significantly with DDD-P compared with SR (10.7 6 2.2 vs 10.2 6 2.2 vs 8.9 6 1.6 vs8.7 6 2.6 cm in SR and with AAI-P, DDD-P, and VVI-P, respectively, P = .003). Basal rotation and time fromonset of the QRS complex to peak basal rotation as a percentage of systole were independently associatedwith the velocity-time integral of diastolic flow in the LAD during SR and the three pacing modes.

Conclusions: Acute right ventricular apical pacing showed a detrimental effect on left ventricular twist andbasal mechanics, with the latter being independently associated with decreased LAD diastolic flow velocityparameters. (J Am Soc Echocardiogr 2014;-:---.)

Keywords: Pacemaker, Speckle-tracking echocardiography, Strain, Twist, LAD flow

Cardiac pacing is the established treatment of choice for various typesof bradyarrhythmias and especially for sick-sinus syndrome and atrio-ventricular conduction disorders. However, long-term pacing fromthe right ventricular apex (RVA) where, the ventricular lead is typicallyplaced, can cause harmful effects on cardiac perfusion, metabolism,and structure, leading to a deterioration in left ventricular (LV) systolicand diastolic performance.1,2

Abnormal electrical activation changes the pattern of mechanicalactivation, leading to intraventricular dyssynchrony and regional alter-ations of myocardial strain and work. These result in less effectivecontraction and abnormal LV relaxation, as well as in changes ofmyocardial perfusion, ultimately leading to ventricular remodeling.3

Newer techniques such as speckle-tracking echocardiographyallow the quantification of myocardial function in a more efficient

partment of Clinical Therapeutics, Medical School, University of

andra’’ Hospital, Athens, Greece.

sts: Elektra Papadopoulou, MD, PhD, Department of Clinical Thera-

cal School, University of Athens, ‘‘Alexandra’’ Hospital, 80 Vas Sofias

St, Athens 11528, Greece (E-mail: elektrapap@yahoo.gr).

4 by the American Society of Echocardiography.

rg/10.1016/j.echo.2014.10.014

and comprehensive way. LV strain and torsion are key parametersof cardiac performance and can help us better understand cardiacmechanics in normal individuals and in disease4–6. Therefore,speckle-tracking echocardiography can assist our efforts to evaluatethe changes in cardiac mechanics in complex clinical situations,such as patients with pacemakers.

Furthermore, it seems that long-term RVA pacing results in changesin oxygen demand, because of the altered metabolic needs, but alsoleads to a high incidence of regional myocardial perfusion defects andabnormalities of microvascular flow associated with impaired globalLV function.1–3,7

The purpose of this study was to evaluate the interaction betweenthe disturbed LV mechanics during RVA pacing and changes in coro-nary blood flow.

METHODS

Study Population

Twenty-five patients who had dual-chamber (DDD) pacemaker im-planted either for sick-sinus or carotid-sinus syndrome were includedin this prospective study. All patients presented in sinus rhythm (SR)at their regularly scheduled visits at the pacemaker clinic of our

Abbreviations

CRT = Cardiacresynchronization therapy

d-VTI = Velocity-time integralof diastolic coronary flow

LAD = Left anterior

descending coronary artery

LV = Left ventricular

perTTP = Time to peak as apercentage of systolic

duration

RVA = Right ventricular apical

SR = Sinus rhythm

2 Papadopoulou et al Journal of the American Society of Echocardiography- 2014

hospital, and only patients withatrial and/or ventricular pacing< 20% on interrogation of thepacemaker were included. Allpatients had the ventricular leadplaced at the right ventricularapex and the atrial electrode atthe right atrial appendage.Patients with structural heart dis-ease, more than mild valvularregurgitation and any degree ofvalve stenosis, known coronaryheart disease or symptoms sug-gestive of coronary heart disease,atrioventricular block, bundlebranch block, or atrial fibrillation,symptomatic heart failure or an

ejection fraction < 50%were excluded from the study. Informed con-sent was given by all study participants, and the study protocol wasapproved by the scientific committee of ‘‘Alexandra’’ UniversityHospital.

Study Protocol

Interrogation of the pacemaker was performed and the baseline char-acteristics and functional parameters of the pacemaker were noted. Acomprehensive baseline echocardiographic study in SR was alsoperformed. Atrial (AAI) pacing at 10 beats/min above baseline (toensure continuous atrial pacing) was then performed for 5min beforea second echocardiographic evaluation. After 5 min in SR, DDD pac-ing was applied at 10 beats/min above the baseline sinus rate, with anatrioventricular delay 20msec shorter than the intrinsic rate to ensurecontinuous atrial and ventricular pacing. Further echocardiographicimages were acquired after 5 min of DDD pacing, which was againfollowed by a 5-min period in SR. Finally, VVI pacing was performedfor 5 min at 10 beats/min above the baseline SR heart rate, and a lastechocardiographic evaluation was performed.

Echocardiography

In each individual, a standard, comprehensive baseline two-dimensional, M-mode, and Doppler study was performed in SR.Study participants were imaged in the left lateral decubitus positionwith a commercially available system (GE Vivid 7 Dimension; GEVingmed Ultrasound AS, Horten, Norway) using a 3.5-MHz (M4S)transducer.

Parasternal long-axis views were used to measure LV internaldimensions at end-diastole and end-systole, interventricular septaland posterior wall thickness, and left atrial end-systolic diameteraccording to the recommendations of the American Society ofEchocardiography.8 LV end-diastolic and end-systolic volumes, aswell as ejection fraction, were derived from the apical four- andtwo-chamber views using the biplane Simpson’s rule. TheDoppler ex-amination included interrogation of mitral inflow, and early (E) andlate (A) peak diastolic velocities and deceleration time weremeasured. Tissue Doppler analysis included pulsed-wave interroga-tion of the medial and lateral mitral annulus, and the mean valuewas calculated. Peak diastolic early e0 and late a0 annular velocitieswere obtained, and the E/e0 ratio was calculated.

Speckle-Tracking Echocardiography. Speckle-tracking anal-ysis was applied to estimate LVrotational mechanics and circumferen-

tial strain parameters. Parasternal short-axis views at the level of themitral valve and apex and standard apical views (four-, two-, andthree-chamber) were recorded for each study participant duringeach pacing mode, according to the recommendations of theEuropean Association of Echocardiography and the AmericanSociety of Echocardiography.9 Five consecutive beats in each viewwere stored digitally for offline analysis. The frame rate was set at50 to 100 frames/sec, the sector width was set as narrow as possible,and gain settings were optimized. Offline analysis was performed us-ing EchoPAC PC 08 version 7.0.0 (GE Medical Systems, Milwaukee,WI). The endocardial border was traced manually in an end-systolicframe, and the region of interest was adjusted to include the entiremyocardium. Optimal visualization of themyocardial walls with mini-mization of dropout and clear delineation of myocardial tissue wassought in every individual. Image acquisition was performed duringan end-expiratory breath-hold. Only subjects with optimal trackingquality, automatically validated by software, were included for furtheranalysis. For each view, three consecutive beats were analyzed, andmean values were calculated for all parameters derived. The followingparameters were measured: peak systolic rotation of the base andapex, peak twist, peak systolic twisting rate, peak untwisting rate,and peak systolic circumferential strain of the base and apex. Thetime from QRS onset to the peak value was measured for each ofthe above parameters in each pacing mode, and the time to peak asa percentage of systolic duration (perTTP) was calculated (Figure 1).

Coronary Blood Flow in the Left Anterior DescendingCoronary Artery (LAD)

In each pacing mode, the blood flow in the mid-distal part of the LADwas assessed with Doppler echocardiography, as previouslydescribed.10–12 Briefly, a modified foreshortened two- or three-chamber view was obtained by sliding the transducer more superiorlyand medially than the standard views, and the distal LAD was soughtwith color flowmapping guidance over the epicardial part of the ante-rior wall or the interventricular septum. Color Doppler echocardiog-raphy was performed with the velocity range set at 12 to 16 cm/sec.

When adequate visualization of flow in the LAD had beenachieved, pulsed-wave Doppler echocardiography was applied byplacing the sample volume (3–4 mm in size) on the color signal inthe LAD, taking into account the diastolic position of the vessel.

Adequate measurement of coronary blood flow velocity wasensured when the angle between the color flow and Doppler beamwas <20� and was kept as low as possible. Every effort was madeto obtain the pulsed-wave Doppler signal at the same position ineach patient for every pacing mode. A spectral trace of the coronaryflow and determination of peak and mean diastolic velocities, as wellas the velocity-time integral, was performed offline by an experiencedinvestigator. Diastolic components of the coronary flow during threecardiac cycles were taken into account, because Doppler signalsacquired during systole were inadequate for analysis as a result ofcardiac motion (Figure 2).

Statistical Analysis

Quantitative variables are presented as mean 6 SD. Repeated-measurements analysis of variance was used to evaluate differencesin echocardiographic parameters between SR and the AAI, DDD,and VVI pacing modes, and Bonferroni correction was applied.Pearson correlation coefficients (r) and random-effects regressionanalysis were used to explore the association between diastolic flow

Figure 1 Assessment of basal (A) and apical (B) rotation and twist (C) by using speckle-tracking imaging. The endocardial border istraced manually in an end-systolic frame, and the region of interest is adjusted to include the entire myocardium. LV twist is depictedwith the white line in (C) and is calculated as the net difference between LV apical (green line) and basal (purple line) rotation.

Journal of the American Society of EchocardiographyVolume - Number -

Papadopoulou et al 3

Figure 2 Spectral trace of mid-LAD diastolic flow and assess-ment of the d-VTI (centimeters) in SR (A), AAI pacing (B), DDDpacing (C), and VVI pacing (D) in the same patient.

velocity parameters in the LAD and the other echocardiographicparameters.

Regression coefficients (b) with their standard errors werecomputed from the results of the random-effects regression analysis.For 10 participants, the analysis of peak systolic strain and LV twistdata was repeated after 2 weeks by the same observer on the same

two-dimensional echocardiographic loop and the same cardiac cycleto define the intraobserver variability in the analysis. In addition, asecond independent observer analyzed the same cardiac cycle todefine the interobserver variability in the analysis of tissue tracking–derived deformation and rotational parameters. For each segment,the differences in strain and twist data were calculated and given asthe relative deviation between these two measurements. All P valuesreported are two tailed. Statistical significance was set at .05, and an-alyses were conducted using Stata version 9.0 (StataCorp LP, CollegeStation, TX).

RESULTS

The study population consisted of 23 patients, whose baselinecharacteristics are shown in Table 1. Two patients were excludedfrom the analysis because of inadequate imaging quality. Mean valuesof the study parameters for SR and the AAI, DDD, and VVI pacingmodes are shown in Table 2. Heart rate was significantly lower inSR compared with the AAI, DDD, and VVI pacing modes.Rotation of the base was significantly higher with AAI comparedwith VVI pacing mode. PerTTP rotation of the base decreased signif-icantly in DDD and VVI modes compared with SR and AAI pacing,circumferential strain of the base decreased significantly in DDD andVVI modes compared with SR, and velocity-time integral of diastoliccoronary flow (d-VTI) in the LAD decreased significantly in DDDmode compared with SR. Additionally, perTTP circumferential strainof the base was significantly increased in the DDD and VVI modescompared with AAI. Twist was significantly decreased between AAIand DDD pacing modes. Untwisting rate remained unchanged(Figure 3). No significant differences in any of the study parameterswere found between the DDD and VVI modes.

Pearson correlation revealed that the d-VTI of LAD flow waspositively correlated with rotation of the base, perTTP rotation ofthe base, and twist (Table 3).

When random-effects regression analysis was conducted with thed-VTI of LAD flow as the dependent variable, and all the other echo-cardiographic parameters as independent variables, it was found thatrotation of the base and perTTP rotation of the base were indepen-dently associated with the d-VTI of LAD flow during SR or the threepacing modes. For a one-unit absolute decrease in rotation of thebase, the d-VTI of LAD flow decreased by about 0.15 units (adjustedb = �0.15, SE = 0.76, P = .048), while for a one-unit decrease inperTTP rotation of the base, the d-VTI of LAD flow decreased byabout 0.05 units (adjusted b = 0.05, SE = 0.014, P = .001).

Intraobserver variability for peak systolic circumferential strain was4.5 6 1.5% and for twist was 6.5 6 5.5%. Interobserver variabilitywas 7.5 6 2.5% and 8.5 6 5.5%, respectively.

DISCUSSION

The main findings of our study are as follows: (1) Twist of the leftventricle was significantly lower during RVA pacing in DDD modethan during AAI pacing. This reduction was due to a significant reduc-tion in basal rotation, because apical rotation was not significantlychanged. Moreover, peak basal rotation occurred much earlier duringRVA pacing. (2) Circumferential strain of the base was also signifi-cantly decreased and occurred significantly later. (3) The d-VTI inthe LAD was consistently decreased during RVA pacing compared

Table 1 Baseline clinical characteristics andechocardiographic findings

Variable Value

Age (y) 68.9 6 7.8

Men/women 12/11

Pacing duration (mo) 69 6 49

SSS/CSS 14/9

LVDD (mm) 46.4 6 3.1

LVDS (mm) 27.8 6 2.8

IVS thickness (mm) 9.4 6 1.8

PWT (mm) 7.8 6 1.1

LA diameter (mm) 38.6 6 4.2

EF (%) 59 6 5

E wave (cm/sec) 64.1 6 14.3

A wave (cm/sec) 83.0 6 20.9

E/A ratio 0.7 6 0.1

E/E0 ratio 8.3 6 1.7

CSS, Carotid-sinus syndrome; EF, ejection fraction; IVS, interventric-

ular septal;LA, left atrial;LVDD, LVdiastolicdiameter;LVDS, LVsystolic

diameter; PWT, posterior wall thickness; SSS, sick-sinus syndrome.

Data are expressed as mean 6 SD or as numbers.

with SR. It was correlated significantly with LV twist, but basal rotationand time to peak basal rotation were the only independent predictorsof LAD flow diastolic velocity parameters (i.e., the d-VTI of LAD flowin SR and with the three pacing modes).

The adverse effects of chronic RVA pacing are becoming increas-ingly recognized. Recent large randomized trials provided robustevidence concerning the clinical sequelae of chronic RVA pacing.The Mode Selection Trial13 included >2,000 patients with sinus-node dysfunction and randomized them to dual-chamber pacingversus ventricular pacing. In that study, RVA pacing > 40%, despitemaintaining atrioventricular synchrony, seemed to confer anincreased risk for heart failure hospitalizations, and the average riskfor heart failure hospitalizations was 10%. Moreover, trials such asthe Dual Chamber and VVI Implantable Defibrillator trial14 and theMulticenter Automatic Defibrillator Implantation Trial II,15 whichincluded patients who received implantable defibrillator devices,showed that frequent RVA pacing resulted in increased hospitaliza-tion for heart failure, new or worsened heart failure, and death. Ithas been speculated that the abnormal electrical activation leads toa distorted mechanical activation with ensuing inter- and intraventric-ular dyssynchrony, which is associated with systolic and diastolicdysfunction and overt heart failure in a subset of patients.

Abnormal electrical activation from the right ventricular apexincludes not only an altered sequence of regional electrical activation,starting from the apex and ending at the basal posterolateral wall, butalso delay in the propagation of the wave front through workingmyocardium rather than the dedicated conduction system.7 Theensuing mechanical activation is thus altered, following the patternof electrical activation. Therefore, the basal regions of the left ventricleare among the last to be activated and contracted.

Few data exist regarding the effect of RVA pacing on the rotationalproperties of the left ventricle. In a study by Matsuoka et al.,16 whichincluded 30 patients with permanent dual-chamber pacemakersimplanted for sick-sinus syndrome, echocardiographic and rotational

parameters of the left ventricle were measured in AAI and DDD pac-ingmodes. The authors reported a significant decrease in LV twist dur-ing RVA pacing (i.e., DDD pacing mode compared with AAI mode),whichwas due to significant decreases in basal and apical rotation andin untwisting rate. In a study by Delgado et al.,17 involving 25 patientsreferred for electrophysiologic study who had normal results on base-line echocardiography, rotational data were extracted during SRbefore the electrophysiologic study and during constant RVA pacing.Again, a significant reduction in LV twist was reported, mainly attrib-utable to the significant reduction in basal rotation. In this prospectivestudy, the rotational properties of the left ventricle and circumferen-tial strain were evaluated consecutively during SR and with all threebasic pacing modes. Basal rotation and circumferential strain weresignificantly reduced with DDD or VVI pacing compared with SRor AAI pacing. However, apical rotation and circumferential strainwere not significantly affected by ventricular pacing. Therefore, theobserved reduction in twist can be attributed to the disorganizedand reduced rotational properties of the base. Previous studies haveunderlined the correlation of twist with indexes of systolic function,such as +dP/dt, and this parameter is emerging as a more accurate in-dicator of global systolic performance than ejection fraction.18

Moreover, twist seems to convey more information about systolicfunction than standard clinical indices in a variety of cardiac dis-eases.19–22 Therefore, reduced twist points to affected globalsystolic function during RVA pacing, as a result of the disorganizedelectrical and mechanical activation and the compromisedperformance of the base even only after a few minutes of pacing.

Blood Flow in the LAD

Several animal studies23–26 have already described the effect ofpacing on myocardial blood flow, regional wall motionabnormalities, and regional myocardial perfusion, suggesting thatdisorganized contraction and regional differences in metabolicneeds are impeding flow near the site of early contraction. In ananimal model27 of heart failure induced by rapid pacing, variationsin myocardial blood flow were apparent immediately after initiationof pacing. Myocardial flow was reduced at the wall adjacent to thesite of activation and showed a progressive reduction in wall thick-ening over the follow-up period of the study. However, on inactiva-tion of the pacemaker during follow-up, subendocardial flowshowed no regional differences.

Clinical studies have confirmed these results in humans. Severalstudies have shown that RVA pacing can induce dyssynchrony,impairment of LV function, and reduced coronary and myocardialblood flow.28–31

In a study by Tse and Lau,32 201Tl exercise scintigraphy was used toassess myocardial perfusion and radionuclide ventriculography toassess systolic function in patients undergoing long-term pacingfrom the right ventricle apex with a DDD pacing system. Perfusiondefects were more common in an inferior or apical segment, andpatients with perfusion defects had significantly lower ejection frac-tions and a higher incidence of apical wall motion abnormalities.Similar findings were recently reported in an echocardiographic studythat included 74 patients with pacemakers implanted for sick-sinussyndrome who underwent long-term RVA pacing. Wall motionabnormalities became apparent after 1 year of RVA pacing, mostfrequently involving the LV apex. Patients with high-volume RVApacing (>50%) presented a gradual decrease of ejection fractionand a gradual increase in LV dimensions.33 Moreover, a recentstudy34 including 14 patients with permanent RVA pacing who

Table 2 Mean values and comparison of study parameters among the SR and the AAI, DDD, and VVI pacing modes withBonferroni correction

Variable SR AAI DDD VVI

HR (beats/min) 67.1 6 11.4 77 6 9.7* 77.2 6 9.8* 77.7 6 10.4*

Rotbase (�) �6.9 6 2.8 �7.1 6 2.3 �5.5 6 3.2 �4.9 6 2.3§

PerTTP rotbase (%) 105.2 6 17.6 99.7 6 7.7 92.3 6 12†,jj 88.3 6 14.5§,‡

Rotapex (�) 11.1 6 4.2 11.5 6 4.7 10.4 6 3.7 11.6 6 3.0

PerTTP rotapex (%) 97.4 6 12.6 95.8 6 7.5 97.3 6 10.4 94.7 6 5.7

Twist (�) 17.0 6 4.8 17.8 6 3.9 14.6 6 4.6jj 15.6 6 3.7

PerTTP twist (%) 97.9 6 11.2 96.2 6 7.3 97.3 6 8.1 93.2 6 7.4

UTR (�/sec) �111.2 6 42.5 �115.2 6 34.3 �88.9 6 39.5 �116.2 6 39.2

PerTTP UTR (%) 120.6 6 13.3 117.3 6 6.7 116.1 6 7.3 116.1 6 10.4

CSbase �17.8 6 2.8 �16.4 6 2.3 �14.9 6 4.3† �15.4 6 3.7‡

perTTP CSbase (%) 101.2 6 9.5 97.9 6 5.1 102.9 6 7.8jj 104.8 6 8.1§

CSapex �22.0 6 13 �24.1 6 6.2 �24.4 6 5.1 �22.6 6 4.8

PerTTP CSapex (%) 98.1 6 3.8 98.1 6 3.5 98.5 6 8 92.5 6 7.2{

d-VTI LAD flow (cm) 10.7 6 2.2 10.2 6 2.2 8.9 6 1.6† 8.7 6 2.6

CSapex, Peak circumferential strain of the apex; CSbase, peak circumferential strain of the base; HR, heart rate; perTTP CSapex, time from QRS

onset to peak circumferential strain of the apex, as a percentage of systole; perTTP CSbase, time from QRS onset to peak circumferential strain of

the base, as a percentage of systole,perTTP rotapex, time fromQRSonset to peak rotation of the apex, as a percentage of systole;perTTP rotbase,

time from QRS onset to peak rotation of the base, as a percentage of systole; perTTP twist, time from QRS onset to peak twist, as a percentage ofsystole; perTTP UTR, time from QRS onset to peak untwisting rate, as a percentage of systole; rotapex, peak rotation of the apex; rotbase, peak

rotation of the base; UTR, untwisting rate.

*P < .05 for individual comparisons versus SR.†P < .05 significant difference between SR and DDD.‡P < .05 significant difference between SR and VVI.§P < .05 significant difference between AAI and VVI.jjP < .05 significant difference between AAI and DDD.{P < .05 significant difference between VVI and SR, AAI, and DDD.

Figure 3 Velocity-time integral (centimeters) of LAD diastolicflow in SR and under AAI, DDD, and VVI pacing.

underwent 201Tl scintigraphy and coronary angiography within 2months concluded that there are alterations in flow velocitiesmeasured in the LAD and in the dominant artery in paced patientsversus control subjects, along with impairment of microvascularflow as detected by perfusion defects mainly in the inferior, apical,and inferoseptal wall on stress myocardial scintigraphy and reducedcoronary flow reserve in the defect-related artery.

In this study, the d-VTI of flow in the LADwas consistently reducedin all patients when they were acutely paced from the right ventricleapex, compared with physiologic conduction.

The primary cause of these adverse effects has long been debated.Most authors agree that an abnormal sequence of electrical activationleads to abnormal mechanical activation, with early contraction of theearly-activated regions and prestretch of the late-activated regions.The abnormal sequence leads to redistribution of wall stress andless effective contraction of the earlier activated region. In an experi-mental study35 with seven anesthetized dogs that underwent ventric-ular pacing and magnetic resonance tagging imaging, systolic fiberstrain and work were significantly decreased near the pacing site,whereas these parameters were more than doubled in remotemyocardial areas. Therefore, redistribution of stress and unloadingof the ventricular muscle leads to a reduction in perfusion, meta-bolism, and structure.36 Comparable results in humans have alsobeen presented.37 In our study, the d-VTI of flow in the LAD wascorrelated with LV twist, implying a relation between decreased cor-onary flow and decreased systolic performance in the acute setting.Furthermore, rotation of the base and time from QRS onset topeak rotation of the base were the only independent predictors ofthe d-VTI of flow in the LAD. This finding reinforces the hypothesisthat an altered mechanical sequence of activation, and ultimatelyless effective contraction as reflected in our study by the significantreduction of LV twist, may be responsible for the altered myocardialflow. Moreover, reduced systolic performance through reducedwringing movement of the heart may affect the regulation of vascular

Table 3 Pearson correlation coefficients of d-VTI in the LADwith the rotational and strain parameters

Variable

d-VTI-LAD flow

HR �0.09 .384

Rotbase 0.24 .021

perTTP rotbase 0.21 .046

Rotapex 0.09 .397

perTTP rotapex 0.06 .587

Twist 0.23 .030

perTTP twist 0.14 .177

perTTP UTR �0.03 .803

CSbase �0.06 .605

perTTP CSbase �0.04 .725

CSapex 0.01 .895

perTTP CSapex 0.10 .365

CSapex, Peak circumferential strain of the apex; CSbase, peak

circumferential strain of the base; HR, heart rate; perTTP CSapex,time from QRS onset to peak circumferential strain of the apex, as

a percentage of systole; perTTP CSbase, time from QRS onset to

peak circumferential strain of the base, as a percentage of systole,

perTTP rotapex, time from QRS onset to peak rotation of the apex,as a percentage of systole; perTTP rotbase, time from QRS onset

to peak rotation of the base, as a percentage of systole; perTTP twist,

time from QRS onset to peak twist, as a percentage of systole;perTTP UTR, time from QRS onset to peak untwisting rate, as a per-

centage of systole; rotapex, peak rotation of the apex; rotbase, peak

rotation of the base; UTR, untwisting rate.

tone through alteration of the externally induced pressure on thevascular wall and consequently alter flow.38 This hypothesis couldexplain our finding of reduced flow velocity parameters in the mid-LAD despite no apparent change in deformation parameters of theapex.

On the other hand, in our study, diastolic parameters such asuntwisting rate were not significantly altered between pacing modes,and heart rate, although significantly increased in the three pacingmodes, did not correlate with change in the d-VTI of flow in LAD.Therefore, diastolic properties as characterized by untwist anddiastolic time duration do not seem to be as important in this studyas systolic performance and LV twist in relation to LAD velocityparameters.

In keeping with the previously mentioned hypothesis of an interac-tion between organized contraction of the left ventricle and flow inthe LAD, a recent study examining the effect of cardiac resynchroni-zation therapy (CRT) on coronary blood flow in patients with dilatedcardiomyopathy, evaluated by transthoracic Doppler echocardiogra-phy, showed that LAD flow was increased in CRT responders whentheir CRT devices were programmed to biventricular pacing with LVpreexcitation compared with intrinsic conduction or right ventricularpacing. These results, according to the authors, suggest that theacutely increased coronary blood flow in LAD during CRT is associ-ated with a more synchronous activation pattern.39

To our knowledge, this is the first time changes in rotational param-eters of the left ventricle have been correlated with changes inmyocardial blood flow diastolic velocity parameters assessednoninvasively with Doppler echocardiography in a prospective

study. Our findings underline the importance of a normal wringingsystolic movement in the even distribution of stress across themyocardiumwhich is necessary for normal coronary flow and systolicperformance.

Further prospective studies are needed to identify paced patientswho are at risk for developing heart failure and to clarify whether rota-tional and strain parameters of systolic function, along with alterationsin myocardial blood flow, can have any predictive value toward thisend.

Limitations

In our study population, the exclusion of coronary artery disease wasmade on the basis of clinical criteria only. Therefore, we were not ableto identify cases with silent coronary artery disease. Another limita-tion of our study is that the changes in epicardial coronary flow areactually changes in the d-VTI of flow in the LAD measured by trans-thoracic Doppler echocardiography. Coronary flow is calculated asthe product of the velocity-time integral of flow and cross-sectionalarea of the vessel, which in this study we were not able to measure.Moreover, there was a lack of perfusion imaging in the present studyto examine whether abnormal epicardial flow translates intoabnormal subendocardial perfusion. Data on dyssynchrony werenot included, and a further limitation may be the fact that the orderof the different pacing modes was not randomized, although a5 min interval in SR was allowed between the different modes.

CONCLUSIONS

This study shows that LV twist, rotation of the base, and diastolicvelocity parameters of flow in the LAD, as expressed by the d-VTIof flow, are affected significantly by RVA pacing. These findings un-derline the importance of the normal wringing systolic movementof the heart for normal systolic performance and coronary perfusion.

ACKNOWLEDGMENTS

We would like to thank Chara Tzavara, biostatistician, Centre forHealth Services Research, Department of Hygiene, Epidemiologyand Medical Statistics, Athens University Medical School, Athens,Greece, for her assistance in the statistical analysis of data. We alsothank Philip Lees, a native English speaker experienced in the editingof medical texts, who checked the manuscript.

REFERENCES

1. Manolis AS. The deleterious consequences of right ventricular apical pac-ing: time to seek alternate site pacing. Pacing Clin Electrophysiol 2006;29:298-315.

2. De Sisti A, M�arquez MF, Tonet J, Bonny A, Frank R, Hidden-Lucet F.Adverse effects of long-term right ventricular apical pacing and identifica-tion of patients at risk of atrial fibrillation and heart failure. Pacing ClinElectrophysiol 2012;35:1035-43.

3. Tops LF, Schalij MJ, Bax JJ. The effects of right ventricular apical pacing onventricular function and dyssynchrony implications for therapy. J Am CollCardiol 2009;54:764-76.

4. Sengupta PP, Tajik AJ, Chandrasekaran K, Khandheria BK. Twist me-chanics of the left ventricle: principles and application. JACC CardiovascImaging 2008;1:366-76.

5. Phillips AA, Cote AT, Bredin SS, Warburton DE. Heart disease and leftventricular rotation—a systematic review and quantitative summary.BMC Cardiovasc Disord 2012;12:46-65.

6. Gorcsan J 3rd, Tanaka H. Echocardiographic assessment of myocardialstrain. J Am Coll Cardiol 2011;58:1401-13.

7. Brenyo A, Goldenberg I, Barsheshet A. The downside of right ventricularapical pacing. Indian Pacing Electrophysiol J 2012;12:102-13.

8. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA,et al, Chamber Quantification Writing Group, American Society of Echo-cardiography’s Guidelines and Standards Committee, European Associa-tion of Echocardiography. Recommendations for chamberquantification: a report from the American Society of Echocardiography’sGuidelines and Standards Committee and the Chamber QuantificationWriting Group, developed in conjunction with the European Associationof Echocardiography, a branch of the European Society of Cardiology. JAm Soc Echocardiogr 2005;18:1440-63.

9. Mor-Avi V, Lang RM, Badano LP, Belohlavek M, Cardim NM,Derumeaux G, et al. Current and evolving echocardiographic techniquesfor the quantitative evaluation of cardiac mechanics: ASE/EAE consensusstatement on methodology and indications endorsed by the Japanese So-ciety of Echocardiography. Eur J Echocardiogr 2011;12:167-205.

10. Hozumi T, Yoshida K, Ogata Y, Akasaka T, Asami Y, Takagi T, et al. Nonin-vasive assessment of significant left anterior descending coronary arterystenosis by coronary flow velocity reserve with transthoracic colorDoppler echocardiography. Circulation 1998;97:1557-62.

11. CaiatiC,MontaldoC,ZeddaN,MontisciR,RuscazioM,LaiG, et al.Validationof a new noninvasive method (contrast-enhanced transthoracic second har-monic echoDoppler) for the evaluation of coronary flow reserve: comparisonwith intracoronary Doppler flowwire. J AmColl Cardiol 1999;34:1193-200.

12. Meimoun P, Tribouilloy C. Non-invasive assessment of coronary flow andcoronary flow reserve by transthoracic Doppler echocardiography: amagic tool for the real world. Eur J Echocardiogr 2008;9:449-57.

13. Sweeney MO, Hellkamp AS, Ellenbogen KA, Greenspon AJ,Freedman RA, Lee KL, et al. Adverse effect of ventricular pacing on heartfailure and atrial fibrillation among patients with normal baseline QRSduration in a clinical trial of pacemaker therapy for sinus node dysfunction.Circulation 2003;107:2932-7.

14. Wilkoff BL, Cook JR, Epstein AE, Greene HL, Hallstrom AP, Hsia H, et al.,Dual Chamber and VVI Implantable Defibrillator Trial Investigators. Dual-chamber pacing or ventricular backup pacing in patients with an implant-able defibrillator: the Dual Chamber and VVI Implantable Defibrillator(DAVID) Trial. JAMA 2002;288:3115-23.

15. Barsheshet A, Moss AJ, McNitt S, Jons C, Glikson M, Klein HU, et al.MADIT-II Executive Committee Long-term implications of cumulativeright ventricular pacing among patients with an implantable cardi-overter-defibrillator. Heart Rhythm 2011;8:212-8.

16. Matsuoka K, Nishino M, Kato H, Egami Y, Shutta R, Yamaquchi H, et al.Right ventricular apical pacing impairs left ventricular twist as well as syn-chrony: acute effects of right ventricular apical pacing. J Am Soc Echocar-diogr 2009;22:914-9.

17. Delgado V, Tops LF, Trines SA, Zeppenfeld K, Marsan NA, Bertini M, et al.Acute effects of right ventricular apical pacing on left ventricular synchronyand mechanics. Circ Arrhythm Electrophysiol 2009;2:135-45.

18. Kim WJ, Lee BH, Kim YJ, Kang JH, Jung YJ, Song JM, et al. Apical rotationassessed by speckle-tracking echocardiography as an index of global leftventricular contractility. Circ Cardiovasc Imaging 2009;2:123-31.

19. Young AA, Cowan BR. Evaluation of left ventricular torsion by cardiovas-cular magnetic resonance. J Cardiovasc Magn Reson 2012;14:49-59.

20. Bertini M, Nucifora G, MarsanNA, Delgado V, van Bommel RJ, Boriani G,et al. Left ventricular rotational mechanics in acute myocardial infarctionand in chronic (ischemic and nonischemic) heart failure patients. Am JCardiol 2009;103:1506-12.

21. Nucifora G, Marsan NA, Bertini M, Delgado V, Siebelink HM, vanWerkhoven JM, et al. Reduced left ventricular torsion early after myocar-dial infarction is related to left ventricular remodeling. Circ Cardiovasc Im-aging 2010;3:433-42.

22. Van Dalen BM, Kauer F, Soliman OI, Vletter WB, Michels M, Ten Cate FJ,et al. Influence of the pattern of hypertrophy on left ventricular twist in hy-pertrophic cardiomyopathy. Heart 2009;95:657-61.

23. Shannon RP, Komamura K, Shen Y-T, Bishop SP, Vatner SF. Impairedregional subendocardial coronary flow reserve in conscious dogs withpacing-induced heart failure. Am J Physiol 1993;264:H801-9.

24. Prinzen FW, Cornelis HA, Arts T, Allessie MA, Reneman RS. Redistribu-tion of myocardial fiber strain and blood flow by asynchronous activation.Am J Physiol 1990;259:H300-8.

25. Delhaas T, Arts T, Prinzen FW, Reneman RS. Regional stress-fiber strainarea as an estimate of regional blood flow and oxygen demand in thecanine heart. J Physiol 1990;477:481-96.

26. Saito D, Takeda K, Hyodo T, Abe Y, Tani H, Nagahana H, et al. Effect ofpacemaker sites on contractile forces of the local myocardium and bloodflow in the major branches of the left coronary artery in anesthetizedopen-chest dogs. Jpn Circ J 1984;48:331-5.

27. Helmer GA, McKirnanMD, Shabetai R, Boss GR, Ross J Jr., Hammond HK.Regional deficits of myocardial blood flow and function in left ventricularpacing-induced heart failure. Circulation 1996;94:2260-7.

28. Kolettis TM, KremastinosDT, Kyriakides ZS, Tsirakos A, Toutouzas PK. Ef-fects of atrial, ventricular, and atrioventricular sequential pacing on coro-nary flow reserve. Pacing Clin Electrophysiol 1995;18:1628-35.

29. Takeuchi M, Nohtomi Y, Kuroiwa A. Effect of ventricular pacing on cor-onary blood flow in patients with normal coronary arteries. Pacing ClinElectrophysiol 1997;20:2463-9.

30. Nielsen JC, BøttcherM, Nielsen TT, Pedersen AK, Andersen HR. Regionalmyocardial blood flow in patients with sick sinus syndrome randomized tolong-term single chamber atrial or dual chamber pacing—effect of pacingmode and rate. J Am Coll Cardiol 2000;35:1453-61.

31. LiuWH, ChenMC, Chen YL, Guo BF, Pan KL, Yang CH, et al. Right ventric-ular apical pacing acutely impairs left ventricular function and induces me-chanical dyssynchrony in patients with sick sinus syndrome: a real-timethree-dimensional echocardiographic study. J Am Soc Echocardiogr 2008;21:224-9.

32. Tse HF, Lau CP. Long-term effect of right ventricular pacing onmyocardialperfusion and function. J Am Coll Cardiol 1997;29:744-9.

33. Sato-Iino T, Watanabe H, Kayama T, Iino K, Kosaka T, Ito H. The preva-lence of apical wall motion abnormalities in patients with long-term rightventricular apical pacing. J Am Soc Echocardiogr 2011;24:556-64.

34. Skalidis EI, Kochiadakis GE, Koukouraki SI, Chrysostomakis SI,Igoumenidis NE, Karkavitsas NS, et al. Myocardial perfusion in patientswith permanent ventricular pacing and normal coronary arteries. J AmColl Cardiol 2001;37:124-9.

35. Prinzen FW, Hunter WC, Wyman BT, McVeigh ER. Mapping of regionalmyocardial strain and work during ventricular pacing: experimental studyusing magnetic resonance imaging tagging. J Am Coll Cardiol 1999;33:1735-42.

36. Bassingthwaighte JB, Li Z. Heterogeneities in myocardial flow and meta-bolism: exacerbation with abnormal excitation. Am J Cardiol 1999;83:7H-12.

37. Ten Cate TJ, Van Hemel NM, Verzijlbergen JF. Myocardial perfusion de-fects in right ventricular apical pacing are caused by partial volume effectsbecause of wall motion abnormalities: a new model to study gatedmyocardial SPECT with the pacemaker on and off. Nucl Med Commun2009;30:480-4.

38. Westerhof N, Boer C, Lamberts RR, Sipkema P. Cross-talk between car-diac muscle and coronary vasculature. Physiol Rev 2006;86:1263-308.

39. Valzania C, Gadler F, Winter R, Braunschweig F, Brodin LA,Gudmundsson P, et al. Effects of cardiac resynchronization therapy oncoronary blood flow: evaluation by transthoracic Doppler echocardiogra-phy. Eur J Heart Fail 2008;10:514-20.

Effect of pacing mode and pacing site on torsional and strain parameters and on coronary flow

Documents