Treatment of Atrial and Ventricular€Arrhythmias Through€Autonomic Modulation

13 Pages·2015·1.96 MB·English

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015
ª2015 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION ISSN 2405-500X/$36.00
PUBLISHED BY ELSEVIER INC. http://dx.doi.org/10.1016/j.jacep.2015.09.013
STATE-OF-THE-ART REVIEW
Treatment of Atrial and
Ventricular Arrhythmias
Through Autonomic Modulation
SébastienP.J.Krul,MD,*WouterR.Berger,MD,*MariekeW.Veldkamp,PHD,*AntoineH.G.Driessen,MD,*
ArthurA.M.Wilde,MD,PHD,*yThomasDeneke,MD,PHD,zJacquesM.T.deBakker,PHD,*x
RubenCoronel,MD,PHD,*kJorisR.deGroot,MD,PHD*
JACC:CLINICALELECTROPHYSIOLOGYCME
Thisarticlehasbeenselectedasthemonth’sJACC:ClinicalElectrophysiology CMEObjectiveforThisArticle:Atthecompletionofthisarticlethelearner
CMEactivity,availableonlineatwww.jacc-electrophysiology.orgby shouldbeabletodiscuss:1)theroleofnuclearimagingintheidentifi-
selectingtheCMEonthetopnavigationbar. cationofcardiacautonomicnervoussystem;2)thebenefitofintrinsic
autonomicmodulationthroughganglionicpleximodulationorablation
AccreditationandDesignationStatement inthetreatmentofatrialfibrillation;and3)thedevelopmentsinauto-
nomicmodulationofcardiacarrhythmiasthroughtargetingtheextrinsic
TheAmericanCollegeofCardiologyFoundation(ACCF)isaccreditedby nervoussystemsuchasrenaldenervationandvagalstimulation.
theAccreditationCouncilforContinuingMedicalEducation(ACCME)to
providecontinuingmedicaleducationforphysicians. CMEEditorDisclosure:CMEEditorSmitVasaiwala,MD,hasnothing
todeclare.
TheACCFdesignatesthisJournal-basedCMEactivityforamaximum
of1AMAPRACategory1Credit(s).Physiciansshouldonlyclaimcredit
AuthorDisclosures:Dr.DriessenhasservedasaconsultanttoAtricure
commensuratewiththeextentoftheirparticipationintheactivity.
Inc.Dr.deGroothasbeensupportedbypersonalgrantsfromthe
NetherlandsHeartFoundation2009T021andNWO/ZonMw016.146.310;
MethodofParticipationandReceiptofCMECertificate hasreceivedunrestrictedfundingfromandhasservedasaconsultantfor
Atricure;hasreceivedresearchfundingfromSt.JudeMedicalandMed-
ToobtaincreditforJACC:ClinicalElectrophysiologyCME,youmust:
tronic;andhasservedasaconsultantforDaiichiSankyo.Allotherau-
1. BeanACCmemberorJACC:ClinicalElectrophysiologysubscriber.
thorshavereportedthattheyhavenorelationshipsrelevanttothe
2. CarefullyreadtheCME-designatedarticleavailableonlineandinthis contentsofthispapertodisclose.
issueofthejournal.
3. Answerthepost-testquestions.Atleast2outofthe3questions MediumofParticipation:Print(articleonly);online(articleandquiz).
providedmustbeansweredcorrectlytoobtainCMEcredit.
CMETermofApproval
4. Completeabriefevaluation.
5. ClaimyourCMEcreditandreceiveyourcertificateelectronicallyby IssueDate:December2015
followingtheinstructionsgivenattheconclusionoftheactivity. ExpirationDate:November30,2016
Listentothismanuscript’s From the *Heart Center, Department of Clinical and Experimental Cardiology, Academic Medical Center, Amsterdam, the
audiosummarybyJACC: Netherlands;yPrincessAl-JawharaAl-BrahimCentreofExcellenceinResearchofHereditaryDisorders,Jeddah,KingdomofSaudi
ClinicalElectrophysiology Arabia;zHeartCenterBadNeustadt,BadNeustadta.d.Saale,Germany;xInteruniversityCardiologyInstituteoftheNetherlands,
Editor-in-Chief Utrecht,theNetherlands;andthekL’InstitutdeRYthmologieetdemodélisationCardiaque(LIRYC),UniversitéBordeauxSegalen,
Dr.DavidJ.Wilber. Bordeaux,France.Dr.DriessenhasservedasaconsultanttoAtricureInc.Dr.deGroothasbeensupportedbypersonalgrantsfrom
theNetherlandsHeartFoundation2009T021andNWO/ZonMw016.146.310;hasreceivedunrestrictedfundingfromandhas
servedasaconsultantforAtricure;hasreceivedresearchfundingfromSt.JudeMedicalandMedtronic;andhasservedasa
consultantforDaiichiSankyo.Allotherauthorshavereportedthattheyhavenorelationshipsrelevanttothecontentsofthis
papertodisclose.
ManuscriptreceivedJanuary5,2015;revisedmanuscriptreceivedAugust19,2015,acceptedSeptember24,2015.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015 Kruletal. 497
DECEMBER 2015:496–508 AutonomicModulationforTreatmentofArrhythmias
Treatment of Atrial and Ventricular Arrhythmias
Through Autonomic Modulation
ABSTRACT
Thispaperreviewsthecontributionofautonomicnervoussystem(ANS)modulationinthetreatmentofarrhythmias.
Boththeatriaandventriclesareinnervatedbyanextensivenetworkofnervefibersofparasympatheticandsympathetic
origin.Boththeparasympatheticandsympatheticnervoussystemexertarrhythmogenicelectrophysiologicaleffectson
atrialandpulmonaryveinmyocardium,whileintheventriclethesympatheticnervoussystemplaysamoredominantrole
inarrhythmogenesis.IdentificationofANSactivityispossiblewithnuclearimaging.Thistechniquemayprovidefurther
insightinmechanismsandtreatmenttargets.Additionally,themyocardialeffectsoftheintrinsicANScanbeidentified
throughstimulationoftheganglionicplexuses.Thesecanbeablatedforthetreatmentofatrialfibrillation.New(non-)
invasivetreatmentoptionstargetingtheextrinsiccardiacANS,suchaslow-leveltragusstimulationandrenaldenerva-
tion,provideinterestingfuturetreatmentpossibilitiesbothforatrialfibrillationandventriculararrhythmias.However,
thefirstrandomizedtrialshaveyettobeperformed.FutureclinicalstudiesonmodifyingtheANSmaynotonlyimprove
theoutcomeofablationtherapybutmayalsoadvanceourunderstandingofthemannerinwhichtheANSinteracts
withthemyocardiumtomodifyarrhythmogenictriggersandsubstrate. (JAmCollCardiolEP2015;1:496–508)
©2015bytheAmericanCollegeofCardiologyFoundation.
T
he atria and ventricles are innervated by an neural pathways of the ANS (1). Small nerve fibers
intricate network of autonomic nerves (1,2). form an extensive neural network of small inter-
Theroleofthesympatheticandparasympa- connecting efferent and afferent sympathetic, para-
theticnervoussysteminthepathophysiologyofcar- sympathetic,andmixednervefibers,thatcontainthe
diac arrhythmias is complex. Parasympathetic and neurotransmitters noradrenaline and acetylcholine,
sympathetic activationinfluence atrialandventricu- respectively, but some also contain neuropeptide Y,
lar electrophysiology and these changes can initiate, somatostatin, vasoactive intestinal polypeptide, and
facilitate,orcounteractcardiacarrhythmiasdepend- substance P (2,9–12). The density of small fibers and
ing on the presence of a suitable substrate (3,4). By ganglia is highest in the posterior part of the left
selectivelyablatingorstimulatingthedifferentcom- atriumandaroundtheantrumofthe(left)pulmonary
ponents of the autonomic nervous system (ANS), veins (PVs) (11,13). The atria are predominantly par-
suchasganglionicplexuses(GPs)orthevagalnerve, asympathetically innervated, while in the ventricles
thenetactivityoftheANScanbemodulatedandar- (where only 16% of total cardiac ganglia reside) pre-
rhythmiastreated(5,6). dominantly sympathetic nerve fibers are found
Here, we briefly review the role of the ANS as (1,14,15). GPs are conglomerates of ganglia from
trigger and modulator of cardiac arrhythmias. different subplexuses and function as an integration
Notably, we focus on novel methods to clinically center of the parasympathetic and sympathetic
identify the cardiac ANS and we discuss autonomic nerves and interconnect the intrinsic ANS (9,16,17).
modulationastreatmentforcardiacarrhythmias. TheatrialGPsarelocatednearthesinusnodeandPVs
andresideinepicardialfatpadsasshowninFigure1.
ANATOMY OF THE CARDIAC ANS Ventricular GPs are located near the interventricular
groove (9). The ligament of Marshall, the embryonic
Inthehumanheart,theextrinsicsympatheticinner- remnant of the left superior caval vein, near the left
vation is mediated via the cervical, stellate (cervico- superior PV is densely innervated with para-
thoracic), and thoracic ganglia. Parasympathetic sympatheticandsympatheticnerves(18,19).
extrinsic innervation is routed via the vagus nerve,
althoughsympatheticfibersarefoundinvagalnerves AUTONOMIC MODULATION IN
and parasympathetic fibers in sympathetic nerves as ATRIAL FIBRILLATION
well (7,8). The extrinsic nerves pass through the hi-
lum of the heart along the great cardiac vessels and The atrial susceptibility to autonomic nervous mod-
branch into 7 epicardial subplexuses, the intrinsic ulation relates to the high density of autonomic
498 Kruletal. JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015
AutonomicModulationforTreatmentofArrhythmias DECEMBER 2015:496–508
ABBREVIATIONS nerves and the presence of acetylcholine isoprenaline and acetylcholine. Isoprenaline facili-
AND ACRONYMS sensitive potassium channels. The influence tated initiation and maintenance of acetylcholine
of the intrinsic ANS on the atrium was induced AF through a decreased AF threshold
AF=atrialfibrillation
reviewedbyStavrakisetal.(20)recently. (lowering the acetylcholine dose required to induce
ANS=autonomicnervous
AF)(32).TheANSnotonlyaffectsrepolarization,but
system TRIGGERMODULATION. InthestudyofChoi
also conduction properties. In patients, GP stimula-
CBS=carotidbodystimulation et al. (21), atrial fibrillation (AF) and atrial
tionhasbeenshowntoaffectactivationtime(33,34).
GP=ganglionicplexus arrhythmias (induced by rapid atrial pacing)
GP stimulation activates all nerves present in the
I-123mIBG=iodine-123 were always preceded by activation of the GPandisthereforenotspecificfor1armoftheANS,
metaiodobenzylguanidine intrinsic ANS in dogs, but no intracardiac
but conduction slowing was particularly evident in
LLVS=low-levelvagal recordingsweremadetolocalizetheoriginof
patients using beta-blockers (33). Conduction slow-
stimulation AF. In patients, PV ectopy, most frequently
ing by acetylcholine could be attributed to a
PV=pulmonaryvein arising in the superior PVs, commonly trig-
reduced excitability, but a role of noncholinergic
PVI=pulmonaryveinisolation gersparoxysmalAF(22).Animalandhuman
neurotransmitters released upon vagal stimulation
RDN=renaldenervation studies have shown that stimulation of the
cannot be excluded (35). In dogs, vasoactive intes-
SCS=spinalcordstimulation GPsnearthePVtriggersPVectopy(23,24).In
tinal polypeptide induced conduction slowing in a
VT=ventriculartachycardia caninePVs,Pattersonetal.(3)demonstrated
dose-dependent manner (36). ANS-mediated con-
VF=ventricularfibrillation thatdecreasedactionpotentialdurationwas duction changes might cause or increase fraction-
mediated by the parasympathetic system,
ation in atrial electrograms with or without the
and that the sympathetic system increased myo-
presence of a suitable structural substrate such as
cardial cytoplasmatic [Ca2þ]. The combination of
fibrosis (37,38).
both components was required for early after-
depolarizations in the PVs, which in turn triggered AUTONOMIC REMODELING. Most of the data on
AF. In right atrial tissue a similar protocol did not autonomic modulation come from healthy tissue,
yield arrhythmias (3). A high degree of autonomic frequently from animal studies, with only short-
innervationnearthePVsandashorteractionpoten- term remodeling (3,32). In most cardiac patients,
tial duration in PV myocytes than in atrial myocytes the myocardium is electrically, structurally, and
might underlie these findings. Ectopic activity in autonomically remodeled for a long period. Auto-
other highly innervated structures, such as the liga- nomic remodeling may change the response to ANS
ment of Marshall, may also trigger AF (11,13,25). stimulation and (the balance of) the para- and
StimulationoftheligamentofMarshallcausedectopy sympathetic innervation (3,39–41). Increased sym-
and triggered AF in 8 AF patients during catheter pathetic nerve and beta-adrenergic receptor density
ablation (26). However, triggers of AF are not was observed in dogs with heart failure together
confined to highly innervated tissue. In 27% of 987 with increased acetylcholinesterase activity and less
patientswithrepeatablationafterafirstPVisolation shortening of effective refractory period on vagal
had been unsuccessful, with predominantly non- stimulation (41). In rapid pacing AF models in dogs,
paroxysmal AF, ectopic firing from the left atrial parasympathetic and sympathetic nerve density was
appendagewasobserved(27).Itisunknownwhether increased (42). An increase of sympathetic nerve
thistriggermechanismisdifferentfrompatientswith density has been described in human chronic AF
AFnotpreviouslytreatedorwhethernon-PVtriggers patients, but whether this is compensatory to
remainoncePVtriggersareexcludedafterablation. increased vagal stimulation or causal for AF in
unknown (12,43).
SUBSTRATE MODULATION. The perpetuation of AF
requires an arrhythmogenic substrate suitable for
reentry or the presence of a continuous trigger. A AUTONOMIC MODULATION IN
short effective refractory period, heterogeneity in VENTRICULAR ARRHYTHMIAS
repolarization, or slow conduction constitute the
main determinants for reentry. Vagal stimulation Inacutemyocardialinfarction,ahighheartrateisan
shortens the effective refractory period in rapid important risk factor for arrhythmias (44). Not only
pacing induced AF models in dogs and creates a is it a marker of a high sympathetic tone, but a high
functional substrate for AF (28–30). In addition, heart rate increases ventricular oxygen demand
repolarization becomes more heterogeneous, most and has important proarrhythmic effects such as a
likely due to the heterogeneous nature of para- decreased action potential duration and increased
sympathetic innervation (30,31). In canine atria, AF myocardial cytoplasmatic [Ca2þ] (45). Heart rate
wasinducedaftersuperfusionwithacombinationof variability has been used as an indirect measure of
JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015 Kruletal. 499
DECEMBER 2015:496–508 AutonomicModulationforTreatmentofArrhythmias
FIGURE1 AnatomyoftheIntrinsicAutonomicNervousSystem
Pauzaetal.(1)haveshownanextensivenetworkofepicardialnervesontheatriaandventriclesdividedin7subplexus(A).Leftisaposteriorandrightisan
anteriorviewoftheheart.Alongthesesubplexus,gangliaarelocalized,conglomeratedinganglionplexus(GP)markedinlightgray.Armouretal.(9)(B)
identifiedthemajoratrialandventricularGPfromaposteriorviewoftheheart.AisreprintedwithpermissionfromPauzaetal.(1);Bisreprintedwithpermission
fromArmouretal.(9).DRA¼dorsalrightatrialganglionatedsubplexus;IVC¼inferiorvenacava;LC¼leftcoronaryganglionatedsubplexus;LD¼left
dorsalganglionatedsubplexus;LV¼leftventricle;MD¼middledorsalganglionatedsubplexus;PA¼pulmonaryartery;RC¼rightcoronaryganglionated
subplexus;RV¼rightventricle;SVC¼superiorvenacava;VLA¼ventralleftatrialganglionatedsubplexus;VRA¼ventralrightatrialganglionatedsubplexus.
500 Kruletal. JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015
AutonomicModulationforTreatmentofArrhythmias DECEMBER 2015:496–508
autonomic balance and is significantly associated branch in particular remodels following myocardial
with the risk of life threatening arrhythmias (46). infarction(58).Thedensityofsynapsesandnervesand
Additionally, a high sympathetic tone during nerveactivityoftheleftstellateganglionincreasedin
myocardial infarction leads to vasoconstriction and dogs with myocardial infarction (59). These remote
increasedinfarctsize(47).However,thesympathetic changes preceded increased ventricular myocardial
nervoussystemalsoexertsanarrhythmogeniceffect sympatheticinnervation(60).Intransplantedhuman
independent of heart rate. Evidently, ventricular ar- heartsanincreaseddensityofsympatheticnerveswas
rhythmias in the setting of inherited arrhythmia found in patients with a history of ventricular ar-
syndromes such as catecholaminergic polymorphic rhythmias (61). Heterogeneity in sympathetic inner-
ventricular tachycardia (VT) are entirely dependent vationduetodenervationininfarctedareasinduced
on sympathetic activity (48). Furthermore, QT inter- supersensitivity,resultinginincreasedshorteningof
val and QT dispersion are markedly increased upon the effective refractory period during sympathetic
infusion of epinephrine (but not phenlyephrine) in stimulation (62). This may increase vulnerability to
patients with Long QT syndrome 1 compared to ventriculararrhythmiasduringischemia(63).
healthy controls, underscoring the arrhythmogenic
potentialofthesympatheticANS(49). IDENTIFICATION OF CARDIAC ANS ACTIVITY
TRIGGERMODULATION. Experimentaldatasuggesta
Noninvasive methods to identify autonomic activity
roleoftriggeredactivityinANS-mediatedventricular
are limited. The exception is heart rate variability,
arrhythmias,althoughthereislimiteddatafromclin-
which has been extensively studied as a marker for
ical studies. For example, in 20 dogs treated with
autonomic activity and balance but this parameter
cesium-chloridetoprolongQTtime,leftstellategan-
only supplies information on sinus node innerva-
glion stimulation caused early afterdepolarizations
tion and does not reflect the extrinsic and intrinsic
(50).Left stellate ganglion stimulation decreasedac-
nerve activity (39,64). The activity from the post-
tion potential duration and elicited delayed after-
ganglionic nerve fibers in skin and muscle appears
depolarizations in 10 of 14 cats (51). Early
to accurately reflect central sympathetic activity
afterdepolarizations were also elicited in isolated
more sensitively than heart rate variability in dogs,
ventricularcardiomyocytesfromfailinghumanhearts
but this option has to prove its clinical value
aftersuperfusionwithnoradrenaline(52).Thecontri-
(65,66).
butionoftriggeredactivitytoventriculararrhythmias
Computed tomography and magnetic resonance
inpatientswithmyocardialinfarctionisunknown(53).
imaginghavebeenofgreatvalueindemonstratingthe
SUBSTRATEMODULATION. Sympatheticstimulation structuralsubstrateofcardiacarrhythmias,butareof
either shortens or prolongs the ventricular effective limitedvalueintheidentificationoftheANS(67,68).
refractory period, depending on species and comor- Currently, only iodine-123 metaiodobenzylguanidine
bidities. For instance, in a study on dogs, the ven- (123I-mIBG) imaging provides information on the
tricular effective refractory period was prolonged sympathetic ANS and has been extensively studied
after vagal stimulation and shortened after sym- inpatientswithventriculararrhythmias.
pathetic stellate ganglion stimulation (4). Indeed,
123I-mIBG IMAGING. 123I-mIBG imaging assesses sym-
stellate ganglion stimulation reduced ventricular
pathetic nerve distribution and quantifies local
fibrillation(VF)intervalsindogsoncardiopulmonary
noradrenaline reuptake and sympathetic activity. A
bypass (54). Interestingly,these effectswere hetero-
lower heart/mediastinum 123I-mIBG ratio, represents
geneous, with differences upon right or the left
increased sympathetic neurotransmitter reuptake
stellate ganglion stimulation, and with marked indi-
activity.
vidual variation. Therefore, sympathetic stimulation
A high sympathetic activity on 123I-mIBG imaging
can facilitate reentry not only through shortening of
predicted progression to permanent AF during a
effectiverefractoryperiodbutalsothroughincreased
mean follow-up of 4 years or correlated with in-
repolarization heterogeneity (47,54–56). Although
creased recurrences of AF after ablation (69,70).
parasympathetic nerves are not abundantly present
However, 123I-mIBG imaging lacks the resolution for
on the ventricle, they may partially and regionally
assessment of local atrial innervation and denerva-
antagonize the effects of the sympathetic nervous
tion. Therefore, its application might be limited
system(4,57).
in unselected populations with atrial arrhythmias.
AUTONOMIC REMODELING. In addition to electrical There is an association between increased sympa-
andstructuralremodelingoftheANS,thesympathetic theticactivityon123I-mIBGimagingandtheincreased
JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015 Kruletal. 501
DECEMBER 2015:496–508 AutonomicModulationforTreatmentofArrhythmias
risk of recurrent ventricular arrhythmias in patients
FIGURE2 DifferencesBetweenAnatomicalandSelectiveFunctional
with structural heart disease, as in patients with
GanglionPlexusAblation
myocardialinfarction (71).Similar observations were
made in patients with VT and apparent normal
structural hearts (72,73). These patients present
eitherasubgroupofpatientsathighriskoragroupof
patients with a cardiac disease before it is clinically
detectable. Defects in sympathetic innervation can
be assessed and neural activity can be compared
with perfusion images from single-photon emission
computedtomographyscanning.Sympatheticdener-
vation on 123I-mIBG scans has a high accuracy in
predicting arrhythmic events in implantable car-
dioverter-defibrillator patients and patients with
previous ventricular arrhythmias (74–76). In the
ADMIRE-HF (AdreView Myocardial Imaging for Risk
Evaluation in Heart Failure) trial, a low 123I-mIBG
heart/mediastinum ratio (<1.6) was associated with
an increased risk for cardiac arrhythmias in patients
with heart failure and a left ventricular ejection
fraction of #35% (77). However, sympathetic dener- Freedomfromrecurrentatrialfibrillationoratrialflutterafterselectivefunctionalganglion
plexusablation(solidline)andanatomicablation(dashedline).Reprintedwithpermission
vation per se did not predict cardiac events, poten- fromPokushalovetal.(81).AGPA¼anatomicalganglionplexusablation;SGPA¼selective
tiallybecausethesepatientshadnopriorventricular ganglionplexusablation.
arrhythmias. Therefore,theheart/mediastinumratio
isusefulonlyinselectedpatientpopulationstoiden-
tifyriskforfuturearrhythmicevents.Kleinetal.(78)
investigated the use of 123I-mIBG acquired denerva- Therefore, absence of increase in R-R interval upon
tion maps in combination with voltage maps for the highfrequencystimulationisanunreliableendpoint
ablation of ventricular arrhythmias. In this study ofsuccessfulablationoftheGPs.Arandomizedstudy
thedenervationmapswerelargerthanthescarareas in 80 patients showed that an anatomical rather
and VT ablation sites were located in the areas of than a functional (eliciting a vagal response) ap-
abnormal innervation. However, the VT recurrence proach toward atrial GP localization and ablation
rateof43%after6monthswassimilartoconventional led to more favorable results (77.5% vs. 42.5%
treatment (not guided by 123I-mIBG- single-photon absence of AF, 13.1 (cid:2) 1.9 months follow-up)
emissioncomputedtomography).Thisstudyisofin- (Figure 2) (81). Despite high success rates for GP
terest because it looked beyond structural substrate ablationcomparedtootherstudies,thedatasuggest
ablation toward autonomic modulators in ablation that anatomical localization and ablation of the GPs
of ventricular arrhythmias. Thus, clinical validation might be superior (9). An anatomical epicardial
of123I-mIBGsingle-photonemissioncomputedtomo- approach of ablation, as with thoracoscopic surgery
graphy derived parameters in different patient pop- allows clear visualization of epicardial fat pad in
ulationswithdifferentriskprofilesisneeded. which the atrial GPs reside (82).
Areas of complex fractionated atrial electrograms,
CARDIAC ANS LOCALIZATION. GPs may be a target defined as low-voltage multiple potential bipolar
forablation,andthusitisimportanttoidentifytheir atrialelectrogramswithahighdegreeoffractionation
location. Endocardial stimulation of atrial GPs with arefoundaroundtheanatomicallocationsoftheGPs
highfrequency stimulation inducesavagal reaction, (38,83–85). They may represent a dynamic substrate
reflected in a R-R interval prolongation of >50% driven by both parasympathetic and sympathetic
(79). Right atrial GP ablation attenuates this vagal innervation,andareoftenidentifiedtoguideablation
response, as it is predominantly mediated by the procedures in patients with AF (84). However, com-
anteriorrightGPonthesinusnodeandbytheinferior plexfractionatedatrialelectrogramsarenotaspecific
rightGPontheAVnode(17,80).Thus,absenceofan markeroftheANSbutanelectrophysiologicalrepre-
increaseinR-Rintervalof50%,onlyprovesalossof sentation of structural, electrical and autonomic
innervation of the sinus node or AV-node and not remodeling resulting from a variety of pathophysio-
the dysfunction or destruction of the GPs per se. logicconditions(86).
502 Kruletal. JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015
AutonomicModulationforTreatmentofArrhythmias DECEMBER 2015:496–508
THERAPEUTIC MODULATION ischemiainducedVFafteratrialGPablationindogs,
OF THE INTRINSIC ANS potentially due to attenuated parasympathetic tone.
Few data exist of GP modulation of ventricular
PHARMACOLOGICAL GP MODULATION. Direct in- arrhythmias.StimulationoftheGPsindogshasbeen
jection of botulinum toxin, a cholinergic blocker, in reported to decrease the incidence of ventricular
epicardial fat pads containing the GPs, temporally arrhythmias, which may relate to parasympathetic
suppressed AF inducibility in dogs (87) (Central effects indirectly modifying sympathetic activity
Illustration).Inasmallpilotstudyinpatientsunder- (97). However, selective ablation of ventricular GPs
goingcoronaryarterybypasssurgerybotulinumtoxin hasnotbeenstudied.
injection resulted in less post-operative AF (7% vs.
30%) (88). Larger studies are needed to confirm this THERAPEUTIC MODULATION
finding. OF THE EXTRINSIC ANS
Yu et al. (89) delivered a neurotoxin N-isopropyl
acrylamide monomer to the GPs using iron-core Modulation of the extrinsic ANS, mainly stellate
nanoparticles, both after direct injection and intra- ganglion ablation, has been applied for VF (98).
coronary infusion. Nanoparticles injected into the Recently, there have been a number of studies
anterior right GP resulted in a reduced sinus node investigating the extrinsic ANS for the treatment of
response upon GP stimulation and increased AF AF(CentralIllustration)(99,100).However,mostdata
threshold. The nanoparticles were magnetically are derived from animal studies and await clinical
directedtotheinferiorrightGPusingalargemagnet confirmationofefficacyandsafety.
and resulted in a reduced ventricular rate response
STELLATE GANGLION ABLATION. Modification of
upon GP stimulation. This proof-of-concept study
thesympathetictonethroughablationofthestellate
demonstrates that it is possible to selectively target
ganglion reduces ventricular arrhythmias (101). In
the intricate neuralnetwork, althoughclinical appli-
patients with recurrent multiple VF episodes after
cationhastobeawaited.
myocardial infarction sympathetic blockade, either
withstellateganglionablationorwithbeta-blockers,
GP ABLATION. GPablationadditionaltoPVisolation
increased survival compared to standard antiar-
(PVI)(butnotasastand-aloneprocedure)appearsto
rhythmic drugs (102). Additionally, stellate ganglion
improve procedural efficacy (Central Illustration)
ablation has been applied in patients with inherited
(5,90).CombinedPVIandGPablationresultedina74%
arrhythmia disorders with drug resistant sympathet-
successrateinpatientswithparoxysmalAFcompared
ically induced arrhythmias, such as long QT
to56%and48%inGPablationandPVIalone,respec-
syndrome and catecholaminergic polymorphic VT
tively,after2yearsoffollow-up(5).Theeffectsmaybe
(103,104). Although highly successful, stellate gan-
due to prevention of PV ectopy through GP ablation
glion ablation surgery may lead to surgical compli-
andPVI,althoughitcannotbeexcludedthatGPabla-
cations and denervation is not always complete
tion involves more extensive left atrial myocardial
(103,104). There are no reports on stellate ganglion
ablation.Thisalone,evenwithouttheinvolvementof
ablation for the treatment of atrial arrhythmias in
ANSmodulation,mighthavecontributedtotheadded
humans. However, in a study with 6 dogs with
effectofGPablationandmakesthepotentialbeneficial
pacing-induced heart failure, left and right para-
effect of GP ablation uncertain (91). Compared to
vertebral T2 to T4 thoracic sympathetic ganglion
endocardialablation,wherethedeliveredenergymay
ablation reduced of the number of atrial tachycardia
notalwaysreachtheepicardiallylocatedfatpads,the
episodescomparedtoacontrolgroupwithoutstellate
epicardial approach through thoracoscopic surgery
gangliaablation(105).
allowsvisualizationofGPsandmoreeffectiveablation
ofalargerpartoftheepicardialneuralnetwork(82,92). SPINALCORDSTIMULATION. Spinalcordstimulation
However, regardless of the approach to GP ablation, (SCS)ofT1toT5withanexternalstimulatorappearsto
partial recovery of GPs function has been observed, modulateautonomicactivity,possiblyviainhibitionof
which may limit the long-term procedural effect the stellate ganglion, but a role for increased vagal
(93,94).Thismightbeduetoreinnervationbyefferent activity has been reported as well (106,107). This
nerves or through neural remodeling, leading to discrepancy may be related to the level (T1 to T5) of
increasedsensitivitytotheremainingneuralstimula- SCS.Antiarrhythmiceffectsinbothatrialandventric-
tionorboth(60,95). ular arrhythmias have been described in canine
AblationofGPmaynotbeharmless.Heetal.(96) studies.IncanineAFtachypacingmodelsSCSwasable
reported an increased incidence of myocardial toreduceAFburdenandinducibility(108).Noeffect
JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015 Kruletal. 503
DECEMBER 2015:496–508 AutonomicModulationforTreatmentofArrhythmias
CENTRALILLUSTRATION TargetsofAutonomicNervousSystemModulationforTreatmentofCardiacArrhythmias
Aschematicrepresentationoftheextrinsicandintrinsicautonomicnervoussystem.Thedifferenttargetsforautonomicnervoussystemmodulationinthetreatmentof
cardiacarrhythmiasareshown.AF¼atrialfibrillation;CP¼cardiacplexus;MI¼myocardialinfarction;PVI¼pulmonaryveinisolation;VF¼ventricularfibrillation;
VT¼ventriculartachycardia
504 Kruletal. JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015
AutonomicModulationforTreatmentofArrhythmias DECEMBER 2015:496–508
Although vagal stimulation has proarrhythmic
FIGURE3 IncidenceofAFRecurrencesinPatientsWithandWithout
effects in the atria because of the presence of
RenalArteryDenervation
acetylcholine sensitive ion channels, discrete modu-
lation of vagal tone with low-level vagal stimulation
(LLVS) prevents atrial arrhythmias (32). LLVS stimu-
lates the vagal nerve below the threshold at which
an effect on heart rate is elicited. One week of
LLVS reduced sympathetic nerve density in the
stellate ganglion, as observed in dogs with rapid
pacing-induced AF, which is associated with fewer
paroxysmsofAFanddecreasedAFinducibility(114).
Additionally, LLVS prevented and reversed atrial
remodeling induced by rapid atrial pacing(6).These
data have been confirmed in a proof-of-concept
study where patients with paroxysmal AF received
transcutaneous LLVS to the tracheal nerve, a branch
ofthevagusnerve.Thepatientshadshorterepisodes
of pacing induced AF (115). This study only applied
ThegroupthatunderwentbothPVIandrenalarteryablationhasasignificantlyreduced
atrialfibrillation(AF)recurrencerateovertimecomparedwiththecontrolpulmonaryvein short lasting vagal stimulation, during an ablation
isolation(PVI)–onlygroup.ReprintedwithpermissionfromPokushalovetal.(100). procedure. If, however, transcutaneous LLVS proves
AT¼atrialtachyarrhythmia. effective and safe in ambulatory patients with AF,
this may pave the way for noninvasive application
ofthistherapy.
was observed if SCS was initiated 8 weeks after CAROTID BODY STIMULATION. No large clinical
tachypacingwasstarted.SCSdecreasedtheoccurrence studies have been performed on electrical carotid
of ischemia-induced ventricular arrhythmias via an bodystimulation(CBS)inthetreatmentofatrialand
antisympatheticactionindogswithhealedmyocardial ventricular arrhythmias (116). However, high-output
infarctions (109,110). Interestingly, in animals using CBS induces a vagal response that can be proar-
beta-blockerstherewasanadditionalantiarrhythmic rhythmicintheatrium.CBSresultedinashortening
effectofSCS,whichsuggeststhatSCSisnotlimitedto of the effective refractory period and increased AF
the reported antisympathetic effects (110). Similarly, inducibilityinpigs(117).Low-levelCBSmightprevent
SCSledtofewerventriculararrhythmiasandreduced excessive vagal activation but still modulate auto-
heartratevariabilityandleftstellateganglionactivity nomic activity: it prolonged the effective refractory
inacutemyocardialinfarction(107).In2patientswith period and prevented rapid atrial pacing–induced
a cardiomyopathy and high burden of ventricular remodeling in rabbits (118). Low-level CBS also pro-
arrhythmias SCS reduced VT and VF episodes up to longed ventricular effective refractory period and
75% to 100% (111). The mechanism of SCS is not reduced ventricular arrhythmias after acute myocar-
completelyunderstood,andlongtermclinicaleffects dial infarction (119). However, in light of the recent
and safety have never been studied in patients with progress in LLVS, which appears to act in a similar
arrhythmias. manner,itremainsunclearwhetherCBSwilldevelop
asaviablealternative,asthecarotidbodyisadelicate
VAGAL NERVE STIMULATION. Stimulation of the
structure,locatedinacomplexanatomicalarea.
cervical vagosympathetic trunks can antagonize the
proarrhythmic sympathetic surge during acute RENAL DENERVATION. Renal denervation (RDN)
myocardial infarction. Indeed, high-intensity vagal caused decreased central sympathetic activity and
stimulation during myocardial infarction resulted in lowered blood pressure in patients with refractory
71% VF free survival versus 40% with low intensity hypertension in experimental and early clinical
and 10% with no vagal stimulation. During ventri- studies (120). However, the SYMPLICITY HTN-III
cular pacing to exclude heart rate effects, less VF (Renal Denervation in Patients With Uncontrolled
occurred, suggesting a protective effect of the vagal Hypertension)trial,alargemulticentertrial,showed
stimulation beyond heart rate reduction (112). no benefit of RDN on systolic blood pressure in 535
Vagal stimulation shortly after onset of myocardial patients with refractory hypertension compared to
ischemia also decreased VF incidence (from 92% to control. This could be explained by procedural char-
10%)indogswithpriormyocardialinfarction(113). acteristics,operatorexperience,placeboeffectofthe
JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 1, NO. 6, 2015 Kruletal. 505
DECEMBER 2015:496–508 AutonomicModulationforTreatmentofArrhythmias
sham procedure, or insufficient selection of patients acuteregionalischemia)inasimilarmannerasbeta-
(121).However,althoughnoeffectonbloodpressure blockade with atenolol (123). Additionally, in 4
wasobserved,andthereisevidencethattheeffecton patients with (non) ischemic cardiomyopathy and
bloodpressureandsympatheticactivityareindepen- frequentVTdespitemaximaldrugtherapyandabla-
dent(120).RDNaffectsatrialandventricularelectro- tion, RDN reduced the occurrence of VT episodes
physiology,independentofthebloodpressureeffect from 11.0 episodes during the month before to 0.3
asoutlinedsubsequently(117,122–124).Similarly,RDN episodes the month after ablation (129). However,
ledtodecreasedleftatrialvolumeindependentofthe in light of the SIMPLICITY HTN-III experience, blin-
change in blood pressure in 66 patients with hyper- ded clinical studies with sham interventions should
tension but without AF (125). Subjects with a high be performed to assess the antiarrhythmic effect of
burden of premature atrial complexes experienced a RDN.
reductionoftheseprematurecomplexes,butthiswas
not related to left atrial volume change. The lack of CONCLUSIONS
correlation between morphological and electric or
autonomicchangesmightbeduetoalowsamplesize Modulation of the ANS may be used to treat cardiac
and heterogeneity within the group. Nevertheless, arrhythmias.Animalandhumanstudieshaverevealed
thesefindingssuggestthatRDNhasaneffectbeyond anintricatenetworkofinterconnectednervesofboth
blood pressure reduction and appears to influence theparasympatheticandsympatheticnervoussystem
autonomicactivity. on the atria and ventricle. New functional imaging
In16dogswithahypersympatheticstate(induced techniquescanhelpusvisualizetheANSinindividual
by rapid atrial pacing and left stellate ganglion patients(70).AblationofGPs,theintegrationcenters
stimulation), RDN diminished AF inducibility and of the intrinsic ANS,showspromisingresults for the
reversed the effective refractory period shortening treatmentofAF(100).However,withcurrentimaging
followingstimulationoftheleftstellateganglionand techniques it is difficult to identify or localize the
rapidatrialpacing(126).RDNappliedbeforetheonset intrinsicANS.Newinvasivetreatmentoptionsaimed
ofAFreducedatrialremodelingafter5weeksofrapid at the extrinsic cardiac ANS, such as LLVS and RDN,
atrial pacing in dogs and after 6 weeks in goats in- provideinterestingfuturetreatmentpossibilitiesboth
dependentofthechangeinbloodpressure (122,127). forAFandventriculararrhythmias,butawaitthorough
Therefore, RDN may affect extensive remodeling clinicaltesting(114,128).Furthermore,targetingthese
duringlong-termfollow-up.Theseobservationssug- nerve structures is invasive and associated with the
gest that RDN is effective in patients with sympa- riskofcomplications(104).Onceweareabletoreliably
thetically driven AF. In a small clinical study, 69% identify the intrinsic ANS and define the optimal
of the 27 paroxysmal or persistent AF patients with treatmenttargetofANSmodification,clinicalstudies
refractoryhypertensionwhounderwentPVIandRDN onmodifyingtheANSmayopenanentirelynewstage
were free of AF versus 29% in the PVI alone group for treatment of arrhythmias. Such studies may not
(Figure 3) (100). A meta-analysis of 80 patients onlyimprovetheoutcomeofablationtherapybutmay
showed similar results, with a more profound effect also advance our understanding of how the ANS in-
in a subgroup analysis in patients with severe hy- teracts with the myocardium to cause cardiac
pertension (128). Although follow-up was performed arrhythmias.
without continuous monitoring, RDN appears bene-
ficial in patients with AF and moderate or severe REPRINT REQUESTS AND CORRESPONDENCE: Dr.
hypertension.However,therearenodataonRDNin Joris R. de Groot, Department of Cardiology, Aca-
patientswithAFwithouthypertension. demic Medical Center, Meibergdreef 9, 1100 DD
RDN also reduced the number of spontaneous Amsterdam, the Netherlands. E-mail: j.r.degroot@
ventricular extra systoles and VF in pigs (with amc.uva.nl.
REFERENCES
1.Pauza DH, Skripka V, Pauziene N, Stropus R. 3.Patterson E, Po SS, Scherlag BJ, Lazzara R. 5.KatritsisDG,PokushalovE,RomanovA,etal.
Morphology, distribution, and variability of the Triggered firing in pulmonary veins initiated by Autonomicdenervationaddedtopulmonaryvein
epicardiac neural ganglionated subplexuses in in vitro autonomic nerve stimulation. Heart isolationforparoxysmalatrialfibrillation:aran-
thehumanheart.AnatRec2000;259:353–82. Rhythm2005;2:624–31. domizedclinicaltrial.JAmCollCardiol2013;62:
2318–25.
2.MarronK,WhartonJ,SheppardMN,etal.Distri- 4.Martins JB, Zipes DP. Effects of sympathetic
bution,morphology,andneurochemistryofendo- andvagalnerveson recoverypropertiesof the 6.ShengX,ScherlagBJ,YuL,etal.Preventionand
cardialandepicardialnerveterminalarborizationsin endocardium and epicardium of the canine left reversalofatrialfibrillationinducibilityandauto-
thehumanheart.Circulation1995;92:2343–51. ventricle.CircRes1980;46:100–10. nomicremodelingbylow-levelvagosympathetic