novel interventional therapies to modulate the autonomic...

J A C C : H E A R T F A I L U R E V O L . 3 , N O . 1 0 , 2 0 1 5

ª 2 0 1 5 B Y T H E A M 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 U N D A T I O N I S S N 2 2 1 3 - 1 7 7 9 / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j c h f . 2 0 1 5 . 0 5 . 0 0 8

STATE-OF-THE-ART PAPER

Novel Interventional Therapies toModulate the Autonomic Tone inHeart Failure
Neal A. Chatterjee, MD,* Jagmeet P. Singh, MD, DPHIL*y
JACC: HEART FAILURE CME

This article has been selected as the month’s JACC: Heart Failure CME

activity, available online at http://www.acc.org/jacc-journals-cme by

selecting the CME tab on the top navigation bar.

Accreditation and Designation Statement

The American College of Cardiology Foundation (ACCF) is accredited by

the Accreditation Council for Continuing Medical Education (ACCME) to

provide continuing medical education for physicians.

The ACCF designates this Journal-based CME activity for a maximum

of 1 AMA PRA Category 1 Credit(s). Physicians should only claim credit

commensurate with the extent of their participation in the activity.

Method of Participation and Receipt of CME Certificate

To obtain credit for JACC: Heart Failure CME, you must:

1. Be an ACC member or JACC subscriber.

2. Carefully read the CME-designated article available online and in this

issue of the journal.

3. Answer the post-test questions. At least 2 out of the 3 questions

provided must be answered correctly to obtain CME credit.

4. Complete a brief evaluation.

5. Claim your CME credit and receive your certificate electronically by

following the instructions given at the conclusion of the activity.

CME Objective for This Article: After reading this article, the reader

should be able to discuss: 1) the role of the autonomic nervous system

From the *Cardiology Division, Department of Medicine, Massachusetts G

Massachusetts; and the yCardiac Arrhythmia Service, Department of Medicin

School, Boston, Massachusetts. Dr. Singh has received research grants fro

and Sorin Group; and has been a consultant for Boston Scientific, St.

sight, and Respicardia. Dr. Chatterjee has reported that he has no relat

disclose.

Manuscript received March 18, 2015; revised manuscript received April 17, 2

(ANS) in the pathophysiology of heart failure; 2) the current state of

knowledge related to ANS modulation for the treatment of heart fail-

ure; and 3) the implications of these data related to clinical practice

and future research.

CME Editor Disclosure: Deputy Managing Editor Mona Fiuzat, PharmD,

FACC, has received research support from ResMed, Gilead, Critical

Diagnostics, Otsuka, and Roche Diagnostics. Tariq Ahmad, MD, MPH, has

received a travel scholarship from Thoratec. Robert Mentz, MD, has

received a travel scholarship from Thoratec; research grants from

Gilead; research support from ResMed, Otsuka, Bristol-Myers Squibb,

AstraZeneca, Novartis, and GlaxoSmithKline; and travel related to

investigator meetings from ResMed, Bristol-Myers Squibb, AstraZeneca,

Novartis, and GlaxoSmithKline. Adam DeVore, MD, has received research

support from the American Heart Association, Novartis Pharmaceuticals,

Thoratec, and Amgen.

Author Disclosures: Dr. Singh has received research grants from Boston

Scientific, St. Jude Medical, Medtronic, and Sorin Group; and has been

a consultant for Boston Scientific, St. Jude Medical, Medtronic, Sorin

Group, CardioInsight, and Respicardia. Dr. Chatterjee has reported that

he has no relationships relevant to the contents of this paper to

disclose.

Medium of Participation: Print (article only); online (article and quiz).

CME Term of Approval

Issue date: October 2015

Expiration date: September 30, 2016

eneral Hospital, Harvard Medical School, Boston,

e, Massachusetts General Hospital, Harvard Medical

m Boston Scientific, St. Jude Medical, Medtronic,

Jude Medical, Medtronic, Sorin Group, CardioIn-

ionships relevant to the contents of this paper to

015, accepted May 1, 2015.

http://www.acc.org/jacc-journals-cme

http://crossmark.crossref.org/dialog/?doi=10.1016/j.jchf.2015.05.008&domain=pdf

http://dx.doi.org/10.1016/j.jchf.2015.05.008

J A C C : H E A R T F A I L U R E V O L . 3 , N O . 1 0 , 2 0 1 5 Chatterjee and SinghO C T O B E R 2 0 1 5 : 7 8 6 – 8 0 2 Autonomics in Heart Failure

787

Novel Interventional Therap
ies to Modulate theAutonomic Tone in Heart Failure
ABSTRACT

Heart failure (HF) represents a significant and expanding public health burden associated with increasing prevalence and

exponential growth in related health care costs. Contemporary advances in both pharmacological and nonpharmaco-

logical therapies have often been restricted in application and benefit. Given the critical role of the autonomic nervous

system (ANS) in maintaining cardiovascular homeostasis in the failing heart, there has been increasing interest in the role

of ANS modulation as a therapeutic modality in HF. In this review, we highlight the anatomy of the ANS and its role in the

pathophysiology of HF, as well as metrics of its assessment. Given the limitations associated with pharmacological ANS

modulation, including lack of specificity and medication intolerance, we focus in this review on contemporary non-

pharmacological ANS modulation therapies. For each therapy—vagal nerve stimulation, carotid baroreceptor stimulation,

spinal cord stimulation, and renal denervation—we review the rationale for modulation, pre-clinical and clinical assess-

ments, as well as procedural considerations and limitations. We conclude by commenting on novel technologies and

strategies for ANS modulation on the horizon. (J Am Coll Cardiol HF 2015;3:786–802) © 2015 by the American College

of Cardiology Foundation.

H eart failure (HF) represents a significantand expanding public health burden aff-ecting nearly 25 million patients globally

(1,2). In the United States alone, the prevalence ofHF is nearly 6 million and is estimated to doubleby the year 2030 (2). Although contemporary strate-gies in the management of HF have improved sur-vival after diagnosis, overall mortality remains highbecause nearly one-half of patients die within 5years (3). The expanding prevalence of HF, coupledwith improved survival after diagnosis, has framedan exponential growth in HF-related costs, estimatedto range between $30 and $60 billion, and are ex-pected to more than double in the next 20 years(2,4).

In the face of rising HF morbidity, mortality,and costs, there have been important contempo-rary advances in both pharmacological (5) andnonpharmacological therapies (6), although theirapplication and benefit are often restricted to asubset of patients (7). In this context, given thelong-recognized relationship between autonomicnervous system (ANS) function and HF, there isincreasing interest in ANS modulation as a thera-peutic modality. In this review, we highlight theanatomy of the ANS and its role in the pathophys-iology of HF, as well as metrics of its assessment.We then review contemporary nonpharmacologicalANS modulation therapies in HF before commentingon novel technologies and strategies on thehorizon.

ANATOMY, REFLEXES, AND REGULATION OF

THE ANS

In its most reductive form, the ANS primarily com-prises 2 systems: the sympathetic and parasym-pathetic. The sympathetic nervous system (SNS)serves a predominant cardioacceleratory function,and its activation is associated with augmentation ofheart rate (HR), increased ventricular contraction, andenhanced atrioventricular conductivity. In counter-balance, the parasympathetic nervous system (PNS)serves a predominant cardioinhibitory function as-sociated with attenuation of HR and ventricularcontraction, reduced arterial stiffness, and increasedvenous capacitance. The dynamic interaction of these2 limbs of the ANS, modulated by physiological inputsand reflexes, ultimately regulates the hemodynamicand electrical functions of the heart and vascularsystem (Central Illustration).

Understanding the anatomy of the ANS is critical todefining the scope of its therapeutic targeting andmodulation (Figure 1A). The SNS output is located inthe spinal column (first thoracic to fourth lumbarsegments), extending rostrally and caudally to theadjacent paravertebral ganglia forming the sympa-thetic trunk. SNS signals are carried via pre-ganglionicneurons to post-ganglionic ganglia, which are eitherlocated directly on viscera (adrenal cortex, smoothmuscle cells of blood vessels) or, in the caseof myocardial input, organized into 2 major ganglia(superior cervical and stellate) (9). For myocardial SNS

ABBR EV I A T I ON S

AND ACRONYMS

ANS = autonomic nervous

system

CBS = carotid body stimulation

HF = heart failure

HRR = heart rate recovery

HRV = heart rate variability

NE = norepinephrine

NO = nitric oxide

NYHA = New York Heart

Association

PET = positron emission

tomography

PNS = parasympathetic

nervous system

RAAS = renin-angiotensin-

aldosterone system

SNS = sympathetic nervous

system

VNS = vagal nerve stimulation

VTA = ventricular

tachyarrhythmia

Chatterjee and Singh J A C C : H E A R T F A I L U R E V O L . 3 , N O . 1 0 , 2 0 1 5

Autonomics in Heart Failure O C T O B E R 2 0 1 5 : 7 8 6 – 8 0 2

788

input, post-ganglionic fibers coalesce intocardiac nerves (superior, middle, inferior)whose sympathetic fibers extend to the heartand travel subepicardially. For peripheralSNS inputs, stimulation of post-ganglionicganglia leads to secretion of epinephrinefrom the adrenal cortex or local release ofepinephrine and norepinephrine (NE) nearperipheral vessels. Secretion of these sym-pathetic transmitters bind adrenergic re-ceptors in the myocardium (predominantly b-receptors with chronotropic, lusitropic,inotropic, and dromotropic effects) and pe-riphery (a-receptors with vasoconstrictiveeffects) to exert their end-organ influence(10).

By contrast, PNS visceral innervation isreflected by regional nerve inputs (oculomo-tor, facial, glossopharyngeal, vagal, sacral).Myocardial PNS innervation is via the vagusnerve with pre-ganglionic origins in thecentral nervous system (medulla, nucleusambiguus). The vagus nerve and its branches

synapse on myocardial ganglionated plexuses locatedwithin the epicardial fat pads of the atria and ven-tricles, with additional concentration of inputsinvolving the vena cava, coronary sinus, and ostia ofthe pulmonary veins. The concentration of PNSinnervation is generally greater in the atria comparedwith the ventricles, where PNS fibers extend sub-endocardially into the ventricular muscle (11).Peripherally, PNS inputs on blood vessels can lead tovasorelaxation (via nitric oxide [NO] pathways) orvasoconstriction (via activation of smooth muscle)(12).

The extracardiac inputs of both the SNS and PNSfurther interact with a complex network of intrinsiccardiac neurons (approximately 43,000 in the adultheart) known as the epicardial neural plexus (9)(Figure 1B). Through organization into ganglionatedsubplexuses on the surface of atria and ventricles(with proximity to both the sinoatrial and atrioven-tricular nodes) (8), the epicardial neural plexus addsan additional layer of cardio-cardiac regulation ofautonomic function (13).

The homeostatic functions of the ANS are modu-lated by autonomic reflexes as well as integration withneurohormonal axes. Stress-sensitive baroreceptors(mechanoreceptors) are present in both high-pressurearterial (carotid sinus, aortic arch) and low-pressurevenous systems (atria/pulmonary arterial interface,systemic veins), whereas chemoreceptors responsiveto changes in arterial oxygen and carbon dioxideconcentration are present both peripherally (carotid

sinus, aortic arch) and centrally (brainstem). There areadditional low-threshold polymodal receptors (sensi-tive to both mechanical and chemical stimuli) presentin the walls of all cardiac chambers that stimulate SNSoutput in the setting of reduced receptor activity (10).The ANS is further modulated through bidirectionalfeedback from the renin-angiotensin-aldosteronesystem (RAAS). For example, reduced renal perfu-sion (reflective of reduced cardiac output) is associ-ated with release of renin and downstream synthesisof angiotensin II (ATII), which functions centrally toincrease SNS activity (14,15) while also inhibitingbaroreflex-mediated suppression of SNS tone (16,17).Inversely, increased SNS output leads to increasedrenin secretion (14).AUTONOMIC FUNCTION AND HF. The ANS plays acritical compensatory role in maintaining cardiovas-cular homeostasis in the failing heart. Reduction incardiac output leads to stimulation of multiple car-diovascular reflexes (low-flow stimulation of arterialbaroreceptors, increased blood volume stimula-tion of venous baroreceptors, and reduced activityof intramyocardial receptors) and activation ofneurohormonal axes (reduced renal perfusion andRAAS stimulation) culminating in stimulation of theSNS. Increased SNS activity is well described in pa-tients with systolic HF (18) and has recently beenimplicated in the pathogenesis of HF with preservedejection fraction (19,20). Increased sympathetic in-puts maintain cardiac output initially via positiveinotropic and chronotropic effects. Over time, how-ever, chronic activation of the SNS and withdrawal ofPNS input lead to progressive myocardial dysfunc-tion, neurohormonal activation, and increased sus-ceptibility to malignant arrhythmias. Indeed, morethan 3 decades ago, Cohn et al. (21) demonstrated thatplasma NE levels were independently predictive ofmortality in patients with systolic HF. Chronic acti-vation of SNS inputs leads to reduced cardiacneuronal density and responsiveness (22), dysfunc-tion of SNS reflexes including augmentation ofexcitatory arterial baro- and chemoreceptor inputs(23), as well as subcellular myocardial dysfunction(increased apoptosis, abnormal calcium handling,increased interstitial fibrosis) (23–25). In addition toabnormalities of the SNS, withdrawal and attenuationof PNS input has also been implicated in the patho-genesis of HF (12). PNS alterations include reducedvagal ganglionic activity as well as loss of PNS re-ceptor density and neurotransmitter activity (12,26).Sympathovagal imbalance in HF, including loss ofPNS inhibition of SNS reflex arcs (27–29), has beenassociated with increased resting HR and worse clin-ical outcomes in HF (30). Reduced PNS activity may


789

further contribute to HF pathogenesis through dys-regulation of NO signaling (12,31,32) as well as loss ofPNS inhibition of inflammatory cytokine release(33,34) and RAAS activation (35).

Electrophysiologically, HF is associated withneuronal and ionic channel remodeling (36,37) as wellas abnormal SNS and PNS discharges (38). SNS-mediated effects on the electrical refractory period,spatial heterogeneity of electrical remodeling, ionicchannel remodeling, and aberrancy of intracellularcalcium handling may underlie the increased risk ofventricular tachyarrhythmias and sudden cardiacdeath in patients with HF (39,40). PNS influencesin HF are more complex as vagal stimulation has beenshown to reduce ventricular ectopy and ventriculartachyarrhythmias (41), whereas increased atrial PNStone has been associated with atrial fibrillation (37).MEASURING AUTONOMIC FUNCTION. Objectiveassessment and quantification of ANS activity isessential to defining ANS pathology as well as target-ing and assessing the efficacy of ANS interven-tions. Conceptually, ANS activity can be measureddirectly or indirectly and reflect general or regionalassessment (Table 1). Historically, SNS activity wasmeasured via plasma NE levels utilizing radioimmu-noassay of regional venous or arterial blood (21,42,43).Unfortunately, plasma NE levels are at best a crudeassessment of SNS function subject to the heteroge-neity of synthesis, reuptake and clearance of NE indifferent tissue beds (44,45). More direct quantifica-tion of SNS activity includesmicroneurography, whichmeasures post-ganglionic SNS neuronal firing withincutaneous or muscular vasculature (46,47), althoughthis is resource intensive and subject to measurementvariability (48).

There has also been significant efforts to utilizeneuronal imaging of ANS activity. Positron emissiontomography coupled with radiolabeled analogs ofNE (49–51) can assess myocardial NE concentration aswell as adrenergic receptor density, although theprognostic implications of these measures in pa-tients with HF have been inconsistent (52). Oneparticular neuronal imaging surrogate that has gainedincreasing traction has been use of the NE analogue123I-metaiodobenzylguanidine (MIBG) (53). Semi-quantitative measures of MIBG imaging such asreduced heart-to-mediastinal (H/M) ratio or reducedwashout rate of radiotracer have been predictive ofarrhythmia and adverse cardiovascular events, evenafter accounting for standard predictors as demon-strated in the ADMIRE-HF (AdreView Myocardial Im-aging for Risk Evaluation in Heart Failure) trial (54).

Most practical in the noninvasive assessment ofANS activity are electrocardiographic surrogates

including resting HR, frequency and time domainanalysis of HR variability (HRV), provocative ma-neuvers assessing baroreflex sensitivity, and mea-sures of heart recovery following exercise (55–59).Resting HR has been viewed simplistically as a mea-sure of the net effect of SNS and PNS input to thesinus node, although this obviously has its limita-tions as a static assessment (59). Beat-to-beat vari-ability in HR (HRV) has long been recognized to beunder the control of ANS inputs (60) and is assessedusing either time-domain (variation in R-R intervals)or frequency domain (spectral analysis assessingdistribution of R-R intervals) methods (59). Fre-quency domain measures of HRV (e.g., the ratio oflow- and high-frequency variance, which are underSNS and PNS control, respectively) may reflect sym-pathovagal balance (61), although the relationshipbetween discrete measures of HRV and limbs of theANS is not always linear and is further subject to in-fluence from non-ANS inputs (respiration, thermo-regulation) (58,60). Provocation (including exercise)has also been utilized to assess ANS functionincluding, for example, post-exercise HR recovery,which is thought to reflect parasympathetic reac-tivation and SNS withdrawal (59). Each of theseelectrical variables—resting HR, time and frequencydomain measures of HRV, and HR recovery—hasdemonstrated prognostic utility in patients with andwithout cardiovascular disease (30,55,59,62). Ofparticular practical interest, intracardiac deviceshave the capacity to measure multiple noninvasiveelectrical surrogates of ANS function (e.g., restingHR, HRV) (63). Device-based measures of ANSsurrogates have been shown to predict HF hos-pitalization and mortality (64,65) and may offerfuture opportunities to dynamically regulate stimu-lation and pacing rates in response to theseparameters.

NONPHARMACOLOGICAL MODULATION OF ANS

IN HF. Given the implicate role of the ANS in thepathogenesis of HF, there has been expanding interestin the role of ANS modulation as HF therapy. Thesalutary effects of established pharmacotherapies forHF, including b-blockade and RAAS inhibitors, arethought, in part, to be related to modulation of ANStone and related reductions in arrhythmia and favor-able ventricular remodeling (10). Indeed, themortalitybenefit associated with b-blockade in HF is closelyrelated to the degree of HR reduction (30), and thera-pies targeted at elevated resting HR in HF have beenassociated with improved clinical outcome as shownwith use of ivabradine in the SHIFT (Systolic HeartFailure Treatment with the If Inhibitor Ivabradine)

CENTRAL ILLUSTRATION Autonomic Dysfunction and Heart Failure Pathogenesis

(Lower left) Schematic demonstrating the synergistic relationships between autonomic imbalance, neurohormonal activation, and the pathogenesis of heart failure.

(Lower right) Schematic highlighting the interactions between the central nervous system, heart, and kidneys. Anatomic sites of autonomic modulation are highlighted.

ATII ¼ angiotensin II; AV ¼ atrioventricular; HR ¼ heart rate; LV ¼ left ventricular; NO ¼ nitric oxide; PNS ¼ parasympathetic nervous system; RAAS ¼ renin-angiotensin-

aldosterone system; SA ¼ sinoatrial; SNS ¼ sympathetic nervous system.



790


791

trial (66). There are important limitations, however, ofpharmacological strategies to modulate ANS tone,including lack of specificity in modulating discretelimbs of the ANS and frequent side effects associatedwith medication intolerance. For example, althoughthe centrally acting agent moxonidine was shownto significantly reduce plasma NE in select patientswith HF (67), it was associated with increased mor-tality in randomized controlled assessment (68),highlighting the potential dangers associated withnonspecific sympatholysis in HF (69). These limita-tions have spurred the development of non-pharmacological approaches of ANS modulation,including vagal nerve stimulation (VNS), spinal cordstimulation (SCS), baroreceptor stimulation, andrenal denervation.

VAGAL NERVE STIMULATION

Myocardial input of the PNS is organized viathe vagus nerve—the left and right vagus nervesresponsible for regulation of cardiac contractility andsinoatrial function (9,70). More than 4 decades ago,Braunwald et al. (71) demonstrated dysfunction of thePNS in HF. Subsequent elegant animal models of HFsuggested that VNS was associated with normaliza-tion of abnormal ANS surrogates (HRV, baroreflexsensitivity, plasma NE) and improved survival (72,73).Importantly, the salutary influence of vagal nervemodulation in HF almost certainly extends beyondinfluence on autonomic measures (resting HR, HRV)and includes its impact on ventricular NO signaling,inflammatory cytokine activation, and neurohor-monal activation (12). For example, NO is known tomodulate parasympathetic influences on ventricularcontractility (74), and VNS-associated improvementin left ventricular (LV) systolic function is closelycorrelated with normalization of NO expression (75).Similarly, vagal nerve stimulation has been shown toreduce expression of multiple inflammatory cyto-kines (e.g., tumor necrosis factor [TNF]-a, interleukin[IL]-6, IL-1), which have been implicated in thepathogenesis of HF, including aberrant b-adrenergicsignaling, increased myocyte apoptosis, increasedmyocardial fibrosis, and ultimately, adverse LVremodeling (76,77).

Implantation of a vagal nerve stimulator involvesthe coordinated expertise of a surgeon and electro-physiologist. Stimulation of the vagal nerve isaccomplished via an implantable stimulator systemthat delivers electrical impulses via a bipolar cuffelectrode around the vagus nerve in the neckapproximately 3 cm below the carotid bifurcation(Figure 2A). At this level, the vagus nerve comprises

both afferent and efferent fibers organized into 1 of 3types (A–C), each with distinct stimulation thresh-olds, conductive properties, and organ targets. Theefficacy of VNS turns on the ability of the device toselectively stimulate and inhibit appropriate vagalnerve fibers. Strategies for selective targeting in-clude electrode design as well as modulation ofstimulation intensities and waveforms. For example,the CardioFit VNS system (BioControl Medical,Yehudi, Israel) utilizes a multicontact electrodedesigned to preferentially activate vagal efferentfibers in the right cervical vagus nerve. The stimula-tion lead was designed to recruit myocardial-specificefferent vagal B-fibers with minimal recruitment ofvagal A-fibers that could lead to unwanted centralside effects (78). The system additionally incor-porates a right ventricular lead to provide backuppacing in the event of VNS-mediated bradycardia.Following implantation of the vagal cuff, the stimu-lation lead is subsequently tunneled to an infracla-vicular pocket (Figure 2B). The efficacy of myocardialvagal nerve targeting is confirmed by up-titration ofthe amplitude of stimulation (target usually w5.5 mA)associated with a reduction in HR of 5 to 10 beats andwithout side effects (neck pain, coughing, swallowingdifficulty, nausea) (79,80). Of note, alternative pro-cedural approaches to VNS that have shown earlypromise include endovascular stimulation at thecoronary sinus ostium and/or superior vena cava (81).

The first human study of VNS employed the afore-mentioned CardioFit system in 8 patients with HF,demonstrating safety as well as significant improve-ments in HF symptoms and LV end-systolic volumes(82). De Ferrari et al. (79) subsequently led an open-label multicenter trial of the CardioFit device in 32patients with symptomatic HF and reduced LV ejec-tion (left ventricular ejection fraction [LVEF] <35%)showing significant improvements in functional sta-tus and favorable LV remodeling (improved LVEF,reduced LV systolic volumes). These initial favorablestudies led to 3 randomized controlled assessments ofVNS, including NECTAR-HF (Neural Cardiac Therapyfor Heart Failure), ANTHEM-HF (Autonomic NeuralRegulation Therapy of Enhance Myocardial Functionin Heart Failure), and INOVATE-HF (Increase of VagalTone in Congestive Heart Failure) (83–85) trials(Table 2). The NECTAR-HF study enrolled 96 patientswith symptomatic HF, depressed LVEF, and dilatedLV cavity and randomized patients in a 2:1 fashionusing a sham-controlled design (83). VNS wasachieved via a vagal cuff lead (NCT lead, BostonScientific Corporation, Marlborough, Massachusetts)without use of a sensing right ventricular lead,and thus activation and inactivation were unrelated

FIGURE 1 Anatomy of the Autonomic Nervous System

(A) Anatomic organization of the extra-cardiac limbs of the autonomic nervous system

(ANS)—the cardio-acceleratory sympathetic and cardio-inhibitory parasympathetic sys-

tems. Reproduced with permission from Martini FH, Nath JL. Fundamentals of Anatomy and

Physiology. 8th ed. San Francisco, CA: Pearson, 2008. (B) Anatomic representation of the

epicardial neural plexus. Adapted with permission from Pauza et al. (8).

Continued on the next page



792

to cardiac cycle. Preliminary 6-month follow-upidentified no significant difference in the primaryefficacy endpoint of change in LV end-systolicdimension, although patients undergoing VNS diddemonstrate significant improvements in subjectivequality-of-life scores and improvement in New YorkHeart Association (NYHA) functional class (62% vs.45% in the VNS vs. control groups, p ¼ 0.032). Thelack of VNS efficacy in the NECTAR-HF study may berelated to multiple factors, including enrollment of arelatively “stable” HF population (baseline N-termi-nal pro–B-type natriuretic peptide [NT-proBNP] was

879 to 882 pg/ml across study groups), suboptimaldelivery of stimulation current (average of 1.4 mA,which was lower than w4 mA current in trialsdemonstrating VNS efficacy), use of a helicalbipolar electrode stimulating both afferent andefferent vagal fibers (in contrast to the asymmetric,preferential stimulation of efferent fibers utilizedpreviously), and suboptimal modulation of vagalmyocardial inputs (nonsignificant changes inresting HR and variable improvement in indices ofHRV).

The ANTHEM-HF study was an open-label assess-ment of 60 patients with symptomatic HF andreduced LVEF (<40%) randomized to right versus leftvagal nerve stimulation (84). VNS was accomplishedvia the VNS Therapy System (Cyberonics, Houston,Texas) using a pulse frequency lower than that usedin the NECTAR-HF study (10 vs. 20 Hz). Similar to theNECTAR-HF study, and unlike the CardioFit system,VNS was performed without intracardiac sensing andwas thus untimed to the cardiac cycle. In the entirestudy population, there were modest, but significant,improvements in the primary efficacy endpointsof change in LVEF (þ4.5%) and reduction in LVend-systolic volume (�4.1 ml) at 6 months. NYHAfunctional class improved in 77% of patients. ANSsurrogates were variably affected with improvementin time-domain measures of HRV but no significantchange in resting HR or plasma NE and ATII. Theincidence of serious therapy-related adverse eventswas very low (w2%). The ongoing INOVATE-HF studyis a multicenter phase III trial that aims to enroll 650patients with symptomatic HF, dilated LV cavity, anddepressed LVEF (#40%) who will be randomized in a3:2 fashion to VNS or standard therapy (no implant)(85). The study will utilize the CardioFit VNS system,delivering 1 to 2 pulses per cardiac cycle (timed viaintracardiac sensing lead).

Taken together, the results and design of VNSstudies to date highlight several unanswered chal-lenges in utilizing VNS in HF. Optimal patient selec-tion is important to consider because patients withlimited neurohormonal derangement (as seen in theNECTAR-HF study) may not experience incrementalbenefit when compared with goal-directed medicaltherapy. Furthermore, whether patients with baselineincreased SNS activity are most likely to benefit re-mains an open question. The range of stimulationsites (right vs. left), stimulation protocols (timed vs.untimed to cardiac cycle, variability of current de-livery), and lead design (asymmetric vs. symmetricactivation of afferent and efferent vagal fibers) frameopen questions and challenges in improving theefficacy of VNS in HF. Finally, given the protean

FIGURE 1 Continued


793

influences of VNS on intersecting pathway of HFpathogenesis—from innate immunity and proin-flammatory cytokines to NO signaling—additionalwork elucidating the impact of VNS on these path-wayswill likely enhance patient selection and improvethe ability to assess the efficacy of different VNStargeting strategies. For example, the ongoing NoSIRSpilot study (Effects of Transvenous Vagus NerveStimulation on Immune Response, NCT01944228) willexamine the impact of VNS on TNF-a expressionbefore and after exposure to endotoxin in healthyvolunteers (86).

CAROTID BARORECEPTOR STIMULATION

The carotid baroreflex arc plays a significant role insympathovagal balance and is implicated in theregulation of blood pressure and HR (87,88). Thecarotid body and carotid sinus are innervated byboth the PNS (vagus and glossopharyngeal fibers) andSNS (via nearby cervical sympathetic ganglia), andcontain both chemoreceptors (predominantly in thecarotid body; responsive to oxygen and carbon di-oxide tension, blood pH, hypoglycemia) as well asstretch-sensitive mechanoreceptors (predominantlyin the carotid sinus; responsive to increase in bloodpressure) (89). Stimulation of the carotid sinusmechanoreceptors results in afferent signals to thedorsal medulla of the brainstem and, ultimately,attenuation of SNS and augmentation of vagaloutflow reflected by reduction in system blood pres-sure and HR (89–91). Stimulation of carotid bodychemoreceptors is associated with augmentation ofSNS tone (17). Given the close relationship betweencarotid baroreceptor activation and blood pressuremodulation, there has been long-standing historicalinterest in the efficacy of carotid body stimulation(CBS) in patients with hypertension (92). Initialenthusiasm nearly a half century ago was temperedby technological limitations as well as the advent ofeffective pharmacotherapy.

In patients with HF, there is reduced sensi-tivity of the normal inhibitory function of carotidbaroreceptors, related to primary alterations withinthe carotid body as well as dysfunction of centralnervous system processing, ultimately resulting inexcessive sympathetic tone (93–96). Additionally, theneurohormonal (RAAS) activation associated withHF may further exacerbate SNS activity via ATII-mediated augmentation of carotid chemoreceptorsensitivity (96,97). Therapeutically, stimulation of thecarotid body in canine models of HF is associated withnormalization of ANS surrogates (reduced plasma NE,normalized expression of cardiac b-receptors) as well

as favorable LV reverse remodeling (improved LVEF,reduced LV volumes) (94). CBS may further amelio-rate HF via reduction in plasma ATII levels (91,95)with related improvements in extracellular volumeregulation, endothelial function, and favorable ven-tricular remodeling (17,98).

Renewed contemporary interest in CBS as a ther-apy in HF has been catalyzed by technological ad-vances and improvement in implant techniques.Implantation is performed under the direction of amulti-disciplinary team of vascular surgeons, cardio-logists, and anesthesiologists. The most investigatedRheos system (CVRx, Minneapolis, Minnesota) has 3components: an implantable pulse generator, carotidsinus leads, and an external programmer (Figure 3A).Stimulation can be targeted to either or both carotidsinuses (e.g., a newer-generation system [Barostimneo, CVRx] consists of a single carotid sinus elec-trode). Stimulation leads are typically targeted to thecarotid sinus of the common carotid artery approxi-mately 5 mm from the carotid bifurcation. Leadsare connected to an implantable pulse generatorthat is placed within an infraclavicular pocket. Theelectrode is standardly tested for hemodynamic

https://clinicaltrials.gov/ct2/show/NCT01944228

FIGURE 2 Vagal Nerve Stimulation

(A) Examples of leads associated with available vagal nerve stimulation (VNS) systems. (B) Chest radiograph of patients with a vagal nerve stimulator (VNS) and

previously implanted implantable cardioverter-defibrillator. Highlighted is the sensing lead of the VNS system located in the right ventricle. CRT ¼ cardiac resynch-

ronization therapy device.

TABLE 1 Cardiac Autonomic Tests

Tests Measurement Units Description Additional Information

Norepinephrine levels Norepinephrine spillover mol/min/m2 Plasma NE levels are a sensitive guideto sympathetic nervous function

Limited availability and utilityConsiderable variability in release and

uptake of catecholamines in varioustissues

Microneurography Sympathetic nerve activityin cutaneous or muscularvasculature

burst/100 beats orbursts/min

Commonly used to measure of nerveactivity using microelectrode nervessuch as in common peroneal nerve

Can be used to measure efferent multifibertraffic in sympathetic nerves

Limited use. Low reliability andlogistically challenging

Scintigraphic imaging 123I-mIBG imaging Heart-to-mediastinumratio of cardiacMIBG activity

Myocardial sympathetic imaging Limited availability and standardizationMay be useful in risk stratification

Resting heart rate beats/min Net influence of sympathetic and vagalinfluence on sinus node

Predictive of morbidity and mortality

Heart rate variability

Frequency domain Total power ms2 Total variance in heart rate pattern Using for measuring sympathovagalbalance and in risk stratification

Low-frequency power ms2 Sympathetic pattern

High-frequency power ms2 Parasympathetic pattern

Time domain SDNN ms2 SD of average RR interval Useful for risk stratificationEasily measured via implantable cardiac

devices (HRV footprint)RMSs ms2 RMSD of difference between adjacent

intervals

DPNN 50 Percentage Number of pairs of adjacent RR intervalsdiffering by >50 ms/total RR intervals

Heart rate recovery DHR post-exercise(reference is eitherpeak or end-exercise)

Reflects parasympathetic reactivation andsympathetic withdrawal post-exercise

Predictive of arrhythmic mortality;possibly modifiable by exercisetraining

Baroreflex sensitivity Cardiovagal baroreflexsensitivity

ms/mm Hg Index of baroreflex control of autonomicoutflow. Close relationship with cardiacvagal tone

Estimated by changes in systolic arterialpressure

Limited availability but useful in riskstratification and post-myocardialinfarction prognostication

DPNN ¼ differing proportion of normal to normal; HR ¼ heart rate; HRV ¼ heart rate variability; MIBG ¼ 123I-metaiodobenzylguanidine; NE ¼ norepinephrine; RMSD ¼ root mean square of successivedifferences; SDNN ¼ standard deviation of normal to normal intervals.



794

TABLE 2 Comparison of Randomized Controlled Comparisons of Vagal Nerve Stimulation in HF

INOVATE-HF(Ongoing) ANTHEM-HF NECTAR-HF

n 650 60 96 (87 with paired data)

NYHA functional class III II–IV II–IV

Ejection fraction #40% #40% #35%

LVEDD 50–80 mm 50–80 mm $55 mm

QRS width <120 ms <130 ms #150 ms

Trial design Randomized 3:2 (implant vs.no implant)

Open label, randomized right vs.left VNS

Randomized 2:1 (VNS on vs. off)

Control Optimal medical treatment Right vs. left Stimulation off

Device used CardioFit System, BioControlMedical, Israel

Demipulse model 103, Cyberonics,United States

Precision, Boston Scientific,United States

Side stimulated Right Right and left Right

Stimulation protocol Target output: 3.3–5.5 mA,titrated on/off times tomaximum of 10 s on/30 s off

Target output 1.5–3.0 mA (averageachieved 2.0 mA), frequency10 Hz, pulse width 130 ms,14 s on/66 s off

Target output: maximal 4 mA(average achieved 1.4 mA),frequency 20 Hz, pulsewidth300ms, 10 son/50s off

Intracardiac lead Yes No No

Study duration 18 months 6 months 6 months

Primary efficacy endpoints Death or HF hospitalization(up to 5.5 years)

Change in LVESV and LVEF at6 months

Change in LVESD from baselineat 6 months

Primary efficacy endpoint met Trial ongoing Yes, LVEF improvement of 4.5%(95% CI: 2.4–6.6)

No

CI ¼ confidence interval; HF ¼ heart failure; LVEDD ¼ left ventricular end-diastolic dimension; LVEF ¼ left ventricular ejection fraction; LVESD ¼ left ventricular end-systolicdimension; LVESV ¼ left ventricular end-systolic volume; NYHA ¼ New York Heart Association; VNS ¼ vagal nerve stimulation.


795

response utilizing an incremental voltage protocol.Similar to VNS, there are myriad stimulation param-eters including sequence (continuous, burst) as wellas impulse duration and amplitude. Despite thesetechnological advances, contemporary trials utilizingthese systems have been associated with adverseevents (surgical complications, hypoglossal nerveinjury, and respiratory complications), suggestingneed for further technological refinement (102).

The evaluation of efficacy of CBS in HF is in itsnascent stages. Recently, Gronda et al. (103) reporteda single-center, open-label evaluation of CBS in 11patients with symptomatic HF and depressed LVEF(<40%). Utilizing the Barostim neo system targeting asingle carotid sinus, the study assessed a primaryefficacy endpoint of muscle sympathetic nerve ac-tivity at 6 months. CBS was associated with signifi-cant reduction in SNS activity in parallel with modestimprovements in functional status, quality of life,and LVEF. Extending these findings, the recently re-ported Barostim HOPE4HF study randomized 146patients to goal-directed medical therapy aloneversus the addition of CBS via the Barostim neo sys-tem (104). There was no standard dosing protocol,with device stimulation intensity incremented over a3-month period unless associated with excessive re-ductions in HR or blood pressure. At 6-month follow-up, compared with patients with medical therapy,

CBS was associated with modest, but significant,improvement in functional endpoints (e.g., 6-minwalk distance, quality-of-life score, NYHA functionalclass), reduction in natriuretic peptide concentration,and a trend to reduction in days of HF hospitalization.There was no significant difference in LV reverseremodeling between the 2 groups. Ongoing studiesinvolving CBS in HF include the Rheos DiastolicHeart Failure trial (CVRx, NCT00718939) and theongoing Rheos HOPE4HF (Hope for Heart Failure,CVRx, NCT00957073) trial, both of which are random-ized controlled assessments of CBS in patients withrelatively preserved LVEF ($40%) (105,106). End-points include major adverse cardiac events, changesin LV mass index, and safety. Alternative strategies tocarotid sinus stimulation including endovascularstimulation are being tested, for example, in the ACESII study (Acute Carotid Sinus Endovascular Stimula-tion II Study, Medtronic, Minneapolis, Minnesota,NCT01458483) (107).

Although early efficacy assessments of CBS in HFare promising, and technological evolution of thedevice and implant system has improved safety, thereremain significant questions regarding the role ofunilateral versus bilateral stimulation, selection ofpatients with carotid baroreceptor dysfunction, andoptimization of stimulation protocols (impulse dura-tion, frequency, sequence).






796

SPINAL CORD STIMULATION

SCS is a well-established therapy for chronic painsyndromes and refractory angina (108). Implantationtypically occurs under conscious sedation withpercutaneous insertion of a multipolar 8-electrodelead into the epidural space at the mid-thoraciclevel (109) (Figure 3B). Stimulation is applied typi-cally at 90% of the motor threshold with a setof frequency (usually w50 Hz) and pulse width(usually w200 ms) for a prescribed duration oftime (either cyclic or continuously). For patientswith implantable-cardioverter defibrillators, the po-tential presence of SCS-induced myopotentials aswell as interference with ventricular fibrillationdetection warrant interrogation and, if needed,reprogramming of the stimulator at the time ofimplant.

SCS is thought to mediate cardioprotective effectsvia augmentation of parasympathetic tone throughan unknown mechanism postulated to includemodulation of higher-order PNS processing (110).Given the demonstrated influence of SCS on auto-nomic function, there has been increasing interest inits role as a therapy for HF (111). Consistent with itspostulated effects on PNS tone, SCS has been shownto increase atrial refractory period and decrease theinducibility of atrial fibrillation in a tachy-pacingcanine model (112). In addition, SCS has beenshown to reduce the incidence of ventricular tachy-arrhythmias in several post-infarction animal models(113,114). The most robust evidence of the efficacy ofSCS in HF includes a randomized assessment ofthoracic SCS in a post-infarction canine model of HF(115). Over 10 weeks, SCS was performed at the T4level, 3 times a day for 2 h each, at 90% of the motorthreshold. SCS was associated with significantdecrement in ANS surrogates (plasma NE) andneurohormonal activation (reduced NT-proBNP), aswell as significant improvement in LVEF. Stimula-tion at 60% of the motor threshold and at a higherthoracic level (i.e., T1) was also shown to signifi-cantly reduce ambulatory HR in a similar caninemodel (116).

The efficacy of SCS as HF therapy in humans hasbeen mixed (117). The SCS HEART (Spinal CordStimulation for Heart Failure) study was an open-label, nonrandomized assessment of safety and ef-ficacy of SCS in 22 patients (17 patients and 5 con-trols who did not consent to implant) withsymptomatic HF, reduced LVEF (20% to 35%), anddilated LV cavity (118). The SCS device (Eon MiniNeurostimulation System, St. Jude Medical, Plano,Texas) included 2 percutaneous leads covering the

high-thoracic space (T1 to T3), set at 90% to 110% ofthe paresthesia threshold, and delivering therapycontinuously (24 h/day). At 6-month follow-up,SCS was associated with statistically significantimprovement in functional class, quality-of-lifescores, metrics of LV reverse remodeling (improvedLVEF, reduced LV volumes), and peak oxygen con-sumption. The procedure was well tolerated, with noacute complications. In contrast to the favorableresults of the SCS HEART study, preliminary resultsof the single-blind, phase II efficacy study DEFEAT-HF (Determining the Feasibility of Spinal CordNeuromodulation for the Treatment of Chronic HeartFailure, NCT01112579) were generally null (118). TheDEFEAT-HF study randomized 66 patients withsymptomatic HF, depressed LVEF (<35%), anddilated LV cavity in a 3:2 fashion to SCS for 12 h perday or no activation for 6 months. A single, 8-electrode lead (Medtronic lead model 3777) was uti-lized at 90% of the motor threshold. There was nosignificant improvement in the primary efficacyendpoint of LV end-systolic volume index, and nodifference in secondary endpoints, which includedpeak oxygen consumption, change in NT-proBNP,change in functional status, or major adverse cardiacevents (death, HF hospitalization).

The neutral results of the DEFEAT-HF trial raiseseveral questions regarding the implementation ofSCS in HF. Whether more continuous SCS exposure(24 h in the SCS HEART study vs. 12 h in theDEFEAT-HF study) or multiple electrode poles areassociated with greater benefit is unknown. Thereremains no standardized “dose” of SCS in HF, anddose studies of electrode output in humans may bewarranted. Finally, whether SCS benefits select pa-tients with HF (e.g., ischemic cardiomyopathy, aswas utilized in animal models) or whether efficacyis related to duration of HF and degree of patho-logical LV remodeling remain questions for futureinvestigation.

RENAL DENERVATION

In patients with HF, there is bidirectional feedbackbetween the ANS and RAAS. The reduced renalperfusion associated with HF leads to stimulation ofthe RAAS with ATII-mediated increases in centralSNS output (14,15) as well as ATII-mediated modu-lation of carotid body chemoreceptor sensitivity(16,17). Inversely, efferent sympathetic fibers in-nervating the renal cortex and terminating withinthe glomerular arteriole lead to increased reninsecretion and maladaptive handling of sodiumand water (14,119). Additionally, renal afferents


FIGURE 3 Carotid Baroreceptor Stimulation in Heart Failure

(A) Schematic diagram of patient with carotid body stimulation system. The top panel shows the location of the baroreflex activation lead and

implantable pulse generator. The lower panel shows the first-generation CVRx Rheos device (left) and the single-lead second-generation

Rheos device. Adapted, with permission, from Gassler et al. (99) and Karunaratne et al. (100). (B) Chest radiograph of patient with spinal cord

stimulator device (red arrow) and previously implanted biventricular pacemaker-implantable cardioverter-defibrillator. Adapted with

permission from Kang et al. (101).


797

responsive to renal hypoxemia and ischemia maydirectly trigger increased central SNS output (120).The synergistic relationship between autonomic andneurohormonal dysfunction in HF is thought to bean important mediator of progressive pathological

ventricular remodeling, increased peripheral vas-cular resistance, and abnormal sodium and waterhomeostasis.

Although initially assessed in patients withresistant hypertension, there has been growing



798

interest in the use of renal denervation as therapyin HF (121). Pre-procedural evaluation includesassessment of renal arterial anatomy and renalfunction testing (122). The proceduralist must beexperienced in renal angiography as well as man-agement of acute complications such as renal dis-sections or perforation. Denervation is accomplishedvia circumferential, low-energy radiofrequency ap-plications within the renal artery, typically posi-tioned just proximal to the origin of a second-orderrenal artery branch. In general, 4 to 8 applicationsare applied along the length of each renal artery,although individual anatomy and catheter technol-ogy affect the total number of lesions delivered.Given the recent demonstration of the lack ofbenefit associated with renal denervation in pa-tients with resistant hypertension (123), there issignificant interest in the use of novel proceduralapproaches (e.g., more distal renal artery ablation;increase in number of ablative lesions) as wellas consideration of optimal patient selection andtrial design. Procedural innovation in renal dener-vation includes the use of different ablation cath-eter designs (unipolar, bipolar, multipolar), balloon-based application technology, use of alternativeenergy sources (cryo, laser, ultrasound), and alter-native ablative agents (ethanol) (123,124).

Pre-clinical models of HF demonstrated im-provements in sodium handling and cardiac output,as well as decrease in angiotensin receptor densityfollowing renal denervation (121). The first humandemonstration of renal denervation safety and ef-ficacy in HF was in the REACH (Renal ArteryDenervation in Chronic Heart Failure) pilotstudy (125). In 7 patients with symptomatic systolicHF, depressed LVEF and normotension at baseline,renal denervation was well tolerated as reflectedby nonsignificant reduction in blood pressure, noepisodes of syncope or hypotension, and no sig-nificant change in renal function at 6 months. Inthis limited study, patients experienced significantimprovement in subjective quality-of-life measuresand 6-min walk distance. These results in patientswith systolic HF remain to be validated in largerrandomized studies with longer follow-up,including the SYMPLICITY-HF study (Renal Dener-vation in Patients with Chronic Heart Failureand Renal Impairment, NCT01392196) and theRSD4CHF trial (Renal Sympathetic Denervation forPatients with Chronic Heart Failure, NCT01790906)(126,127).

Additionally, given the demonstrated relationshipbetween renal denervation and regression of LVhypertrophy (128,129), there are ongoing trials

assessing the efficacy of renal denervation in pa-tients with HF and preserved ejection fraction. Ofinterest, in previous trials of patients with resistanthypertension, regression of myocyte hypertrophywas only partly dependent on decrement in bloodpressure, suggesting a more direct relationship be-tween autonomic modulation and HF pathogenesis(129). The ongoing DIASTOLE trial (Denervation ofthe Renal Sympathetic Nerves in Heart Failure withNormal LV Ejection Fraction, NCT01583881) willrandomize 60 patients with HF and LVEF $50% torenal denervation or continued optimal medicaltherapy. Efficacy endpoints include measures of LVremodeling (change in LVEF, LV mass) as well asmultiple ANS surrogates (MIBG washout, HRV)(130). Similarly, the ongoing SWAN-HF study (RenalSympathetic Modification in Patients with HeartFailure, NCT01402726) will randomize 200 patientswith symptomatic HF (both reduced and preservedejection fraction) to renal denervation or continuedmedical therapy with a primary endpoint of com-posite cardiovascular events (131). Finally, given thepotential benefits of renal denervation on ventric-ular and atrial arrhythmias (132,133), glucose meta-bolism (134), and sleep apnea (135), there may be anadditional role for this strategy in managing rele-vant comorbidities implicated in the pathogenesisand exacerbation of HF.

In summary, preliminary work suggests that renaldenervation is safe and tolerated in patients withHF, although its ultimate efficacy and tolerabilityremains to be tested in larger studies. Sympathetictone plays an important compensatory role in HF,and historic trials of nonspecific pharmacologicalsympatholysis have been associated with increasedmortality (68,69). Heterogeneity of sympatheticrenal innervation and the presence of comorbidrenal dysfunction in patients with HF will also needto be considered in the application of renal dener-vation for this population.

FUTURE STRATEGIES OF

NEUROMODULATION IN HF

There are several technologies on the horizon thatmay further shape the landscape of neuro-modulation in HF. First, there is expanding interestin the role of tragus stimulation as an alternative,less-invasive method of vagal nerve stimulation(136). In a pre-clinical model of post-infarction car-diomyopathy, tragus stimulation was associatedwith attenuation of ANS surrogates (plasma NE) andneurohormonal activation (NT-proBNP), as well asimprovement in LV function (137). Similar efforts to






799

stimulate the vagus nerve through less invasivemeans include endovascular stimulation (e.g.,within the superior vena cava) (138). Second,although the strategies discussed in the previoustext have focused exclusively on extracardiacautonomic modulation, another area of intenseinvestigation involves modulation of the intrinsiccardiac neural plexus that lies on the epicardialsurface of the heart as well as within the adventitiaof the great vessels between the ascending aortaand pulmonary artery (9,12). Endovascular cardiacplexus stimulation in preliminary pre-clinical workwas associated with augmented LV contractility andcardiac output without influence on vascular resis-tances, central venous pressure, or HR (139). Giventhe anatomic organization of the intrinsic neuralplexus, the ability to selectively modulate compo-nents of the intrinsic cardiac nervous system offersthe speculative possibility of truly targeted auto-nomic modulation (e.g., sinoatrial node in sick sinussyndrome vs. ventricular myocardium in cardio-myopathy). Third, given the ability of implantabledevices to measure surrogates of ANS function, weanticipate the development of pacing and stimula-tion technologies dynamically responsive to ANStone. Finally, there is increasing evidence that HF isoften accompanied by central sleep apnea (CSA),which in turn increases morbidity and mortality(140). Mechanistically, CSA is mediated via therespiratory control center, which is sensitive to theincreased sympathetic drive and overactive chemo-receptors observed in HF. The respiratory controlcenter regulates the blood CO2 and also sends sig-nals to the diaphragm via the phrenic nerves tocontrol the pattern of breathing. Notably, everyepisode of CSA also activates the SNS, perpetuatingthe vicious cycle. Treating the CSA accompanyingHF may in fact offset some of the accompanying

autonomic imbalance and novel transvenous pacingof the phrenic nerve (via axillary or subclavian veinaccess of the left pericardiophrenic or right bra-chiocephalic vein) has been associated with apromising efficacy in mitigating apneic episodes andimproving sleep apnea indices (141). The long-termclinical impact of phrenic nerve pacing in HFpatients remains to be proven.

CONCLUSIONS

The ANS plays a critical role in the pathogenesis ofHF. As acknowledged at the outset, the reductive“yin-yang” formulation of the SNS and PNS com-ponents of the ANS does not reflect the complexand integrative relationships among autonomicfunction (extracardiac and intracardiac), neurohor-monal activation, innate immune system activation,and subcellular functions (e.g., NO signaling).Continued translational efforts focused on refining ourunderstanding of the complex interactions betweenneurohormonal and ANS dysfunction, as well asinsight into regional autonomic regulation, will onlyaugment the development of ANS targeting tech-nologies in HF. Experience to date with non-pharmacological ANS therapy highlights ongoingchallenges for improving patient selection, optimiza-tion of implant or denervation strategies, and refine-ment of stimulation protocols. As technologicalcapabilities and scientific understanding converge,effective modulation of the ANS may transform thelandscape of HF.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Jagmeet P. Singh, Cardiac Arrhythmia Service, Mass-achusetts General Hospital, Harvard Medical School,55 Fruit Street, Boston, Massachusetts 02114. E-mail:[email protected].

RE F E RENCE S

1. Ambrosy AP, Gheorghiade M, Chioncel O,Mentz RJ, Butler J. Global perspectives inhospitalized heart failure: regional andethnic variation in patient characteristics, man-agement, and outcomes. Curr Heart Fail Rep2014;11:416–27.

2. Mozaffarian D, Benjamin EJ, Go AS, et al. Heartdisease and stroke statistics—2015 update: areport from the American Heart Association.Circulation 2015;131:e29–322.

3. Roger VL, Weston SA, Redfield MM, et al.Trends in heart failure incidence and survival in acommunity-based population. JAMA 2004;292:344–50.

4. Voigt J, Sasha John M, Taylor A, Krucoff M,Reynolds MR, Gibson MC. A reevaluation of the

costs of heart failure and its implications forallocation of health resources in the United States.Clin Cardiol 2014;37:312–21.

5. McMurray JJ, Packer M, Desai AS, et al.Angiotensin-neprilysin inhibition versus enal-april in heart failure. N Engl J Med 2014;371:993–1004.

6. Leyva F, Nisam S, Auricchio A. 20 years ofcardiac resynchronization therapy. J Am Coll Car-diol 2014;64:1047–58.

7. Jessup M. Neprilysin inhibition—a noveltherapy for heart failure. N Engl J Med 2014;371:1062–4.

8. Pauza DH, Skripka V, Pauziene N, Stropus R.Morphology, distribution, and variability of the

epicardiac neural ganglionated subplexuses in thehuman heart. Anat Rec 2000;259:353–82.

9. Armour JA. Cardiac neuronal hierarchy in healthand disease. Am J Physiol Regul Integr Comp Physiol2004;287:R262–71.

10. Triposkiadis F, Karayannis G, Giamouzis G,Skoularigis J, Louridas G, Butler J. The sympa-thetic nervous system in heart failure physiology,pathophysiology, and clinical implications. J AmColl Cardiol 2009;54:1747–62.

11. Zipes DP. Heart-brain interactions in cardiacarrhythmias: role of the autonomic nervous sys-tem. Cleve Clin J Med 2008;75 Suppl 2:S94–6.

12. Olshansky B, Sabbah HN, Hauptman PJ,Colucci WS. Parasympathetic nervous system and

mailto:[email protected]

http://refhub.elsevier.com/S2213-1779(15)00416-3/sref1
















































800

heart failure: pathophysiology and potentialimplications for therapy. Circulation 2008;118:863–71.

13. Wickramasinghe SR, Patel VV. Local innerva-tion and atrial fibrillation. Circulation 2013;128:1566–75.

14. Goldsmith SR. Interactions between the sym-pathetic nervous system and the RAAS in heartfailure. Curr Heart Fail Rep 2004;1:45–50.

15. Li YF, Wang W, Mayhan WG, Patel KP. Angio-tensin-mediated increase in renal sympatheticnerve discharge within the PVN: role of nitric ox-ide. Am J Physiol Regul Integr Comp Physiol2006;290:R1035–43.

16. Murakami H, Liu JL, Zucker IH. Angiotensin IIblockade [corrected] enhances baroreflex controlof sympathetic outflow in heart failure. Hyper-tension 1997;29:564–9.

17. Paton JF, Sobotka PA, Fudim M, et al. Thecarotid body as a therapeutic target for thetreatment of sympathetically mediated diseases.Hypertension 2013;61:5–13.

18. Pepper GS, Lee RW. Sympathetic activation inheart failure and its treatment with beta-blockade.Arch Intern Med 1999;159:225–34.

19. Grassi G, Seravalle G, Quarti-Trevano F, et al.Sympathetic and baroreflex cardiovascular controlin hypertension-related left ventricular dysfunc-tion. Hypertension 2009;53:205–9.

20. Hogg K, McMurray J. Neurohumoral pathwaysin heart failure with preserved systolic function.Prog Cardiovasc Dis 2005;47:357–66.

21. Cohn JN, Levine TB, Olivari MT, et al. Plasmanorepinephrine as a guide to prognosis in patientswith chronic congestive heart failure. N Engl J Med1984;311:819–23.

22. Regitz V, Leuchs B, Bossaller C, Sehested J,Rappolder M, Fleck E. Myocardial catecholamineconcentrations in dilated cardiomyopathy andheart failure of different origins. Eur Heart J 1991;12 Suppl D:171–4.

23. Watson AM, Hood SG, May CN. Mechanisms ofsympathetic activation in heart failure. Clin ExpPharmacol Physiol 2006;33:1269–74.

24. Piacentino V 3rd, Weber CR, Chen X, et al.Cellular basis of abnormal calcium transients offailing human ventricular myocytes. Circ Res2003;92:651–8.

25. Brouri F, Hanoun N, Mediani O, et al. Blockadeof beta 1- and desensitization of beta 2-adrenoceptors reduce isoprenaline-induced car-diac fibrosis. Eur J Pharmacol 2004;485:227–34.

26. Dunlap ME, Bibevski S, Rosenberry TL,Ernsberger P. Mechanisms of altered vagal controlin heart failure: influence of muscarinic receptorsand acetylcholinesterase activity. Am J PhysiolHeart Circ Physiol 2003;285:H1632–40.

27. Dibner-Dunlap ME, Thames MD. Control ofsympathetic nerve activity by vagal mechanore-flexes is blunted in heart failure. Circulation 1992;86:1929–34.

28. Kinugawa T, Dibner-Dunlap ME. Altered vagaland sympathetic control of heart rate in left ven-tricular dysfunction and heart failure. Am J Physiol1995;268:R310–6.

29. Newton GE, Parker AB, Landzberg JS,Colucci WS, Parker JD. Muscarinic receptor mod-ulation of basal and beta-adrenergic stimulatedfunction of the failing human left ventricle. J ClinInvest 1996;98:2756–63.

30. Fox K, Borer JS, Camm AJ, et al. Resting heartrate in cardiovascular disease. J Am Coll Cardiol2007;50:823–30.

31. Champion HC, Skaf MW, Hare JM. Role of nitricoxide in the pathophysiology of heart failure.Heart Fail Rev 2003;8:35–46.

32. Binkley PF, Nunziatta E, Liu-Stratton Y,Cooke G. A polymorphism of the endothelial nitricoxide synthase promoter is associated with an in-crease in autonomic imbalance in patients withcongestive heart failure. Am Heart J 2005;149:342–8.

33. Aukrust P, Gullestad L, Ueland T, Damas JK,Yndestad A. Inflammatory and anti-inflammatorycytokines in chronic heart failure: potentialtherapeutic implications. Ann Med 2005;37:74–85.

34. Sloan RP, McCreath H, Tracey KJ, Sidney S,Liu K, Seeman T. RR interval variability is inverselyrelated to inflammatory markers: the CARDIAstudy. Mol Med 2007;13:178–84.

35. Elsner D, Kromer EP, Riegger GA. Effects ofvagal blockade on neurohumoral systems inconscious dogs with heart failure. J CardiovascPharmacol 1990;15:586–91.

36. Tomaselli GF, Marban E. Electrophysiologicalremodeling in hypertrophy and heart failure. Car-diovasc Res 1999;42:270–83.

37. Ng J, Villuendas R, Cokic I, et al. Autonomicremodeling in the left atrium and pulmonary veinsin heart failure: creation of a dynamic substrate foratrial fibrillation. Circ Arrhythm Electrophysiol2011;4:388–96.

38. Ogawa M, Zhou S, Tan AY, et al. Left stellateganglion and vagal nerve activity and cardiac ar-rhythmias in ambulatory dogs with pacing-inducedcongestive heart failure. J Am Coll Cardiol 2007;50:335–43.

39. Cao JM, Fishbein MC, Han JB, et al. Relation-ship between regional cardiac hyperinnervationand ventricular arrhythmia. Circulation 2000;101:1960–9.

40. Zipes DP, Rubart M. Neural modulation ofcardiac arrhythmias and sudden cardiac death.Heart Rhythm 2006;3:108–13.

41. Brack KE, Winter J, Ng GA. Mechanismsunderlying the autonomic modulation of ventric-ular fibrillation initiation—tentative prophylacticproperties of vagus nerve stimulation on malig-nant arrhythmias in heart failure. Heart Fail Rev2013;18:389–408.

42. Liang C, Rounds NK, Dong E, Stevens SY,Shite J, Qin F. Alterations by norepinephrine ofcardiac sympathetic nerve terminal function andmyocardial beta-adrenergic receptor sensitivity inthe ferret: normalization by antioxidant vitamins.Circulation 2000;102:96–103.

43. Esler MD, Jennings GL, Johns J, Burke F,Little PJ, Leonard P. Estimation of ’total’ renal,cardiac and splanchnic sympathetic nervous tonein essential hypertension from measurements of

noradrenaline release. J Hypertens Suppl 1984;2:S123–5.

44. Esler M, Jennings G, Lambert G, Meredith I,Horne M, Eisenhofer G. Overflow of catecholamineneurotransmitters to the circulation: source, fate,and functions. Physiol Rev 1990;70:963–85.

45. Eisenhofer G, Rundquist B, Aneman A, et al.Regional release and removal of catecholaminesand extraneuronal metabolism to metanephrines.J Clin Endocrinol Metab 1995;80:3009–17.

46. Hogarth AJ, Graham LN, Mary DA,Greenwood JP. Gender differences in sympatheticneural activation following uncomplicated acutemyocardial infarction. Eur Heart J 2009;30:1764–70.

47. Lambert E, Straznicky N, Schlaich M, et al.Differing pattern of sympathoexcitation in normal-weight and obesity-related hypertension. Hyper-tension 2007;50:862–8.

48. Floras JS, Mak S. Muscle sympathetic nerveactivity in women and men following acutemyocardial infarction: a meaningful difference?Eur Heart J 2009;30:1692–4.

49. Schwaiger M, Kalff V, Rosenspire K, et al.Noninvasive evaluation of sympathetic nervoussystem in human heart by positron emission to-mography. Circulation 1990;82:457–64.

50. Wieland DM, Brown LE, Rogers WL, et al.Myocardial imaging with a radioiodinated norepi-nephrine storage analog. J Nucl Med 1981;22:22–31.

51. Bengel FM, Schwaiger M. Assessment of car-diac sympathetic neuronal function using PETimaging. J Nucl Cardiol 2004;11:603–16.

52. Link JM, Caldwell JH. Diagnostic and prog-nostic imaging of the cardiac sympathetic nervoussystem. Nat Clin Pract Cardiovasc Med 2008;5Suppl 2:S79–86.

53. Perrone-Filardi P, Paolillo S, Dellegrottaglie S,et al. Assessment of cardiac sympathetic activityby MIBG imaging in patients with heart failure: aclinical appraisal. Heart 2011;97:1828–33.

54. Jacobson AF, Senior R, Cerqueira MD, et al.Myocardial iodine-123 meta-iodobenzylguanidineimaging and cardiac events in heart failure. Re-sults of the prospective ADMIRE-HF (AdreViewMyocardial Imaging for Risk Evaluation in HeartFailure) study. J Am Coll Cardiol 2010;55:2212–21.

55. La Rovere MT, Pinna GD, Maestri R, et al.Autonomic markers and cardiovascular andarrhythmic events in heart failure patients: still aplace in prognostication? Data from the GISSI-HFtrial. Eur J Heart Fail 2012;14:1410–9.

56. Piotrowicz E, Baranowski R, Piotrowska M,Zielinski T, Piotrowicz R. Variable effects ofphysical training of heart rate variability, heartrate recovery, and heart rate turbulence in chronicheart failure. Pacing Clin Electrophysiol 2009;32Suppl 1:S113–5.

57. La Rovere MT, Pinna GD, Hohnloser SH,et al. Baroreflex sensitivity and heart ratevariability in the identification of patients atrisk for life-threatening arrhythmias: implica-tions for clinical trials. Circulation 2001;103:2072–7.







































































































































































































801

58. Adamopoulos S, Piepoli M, McCance A, et al.Comparison of different methods for assessingsympathovagal balance in chronic congestiveheart failure secondary to coronary artery disease.Am J Cardiol 1992;70:1576–82.

59. Lahiri MK, Kannankeril PJ, Goldberger JJ.Assessment of autonomic function in cardio-vascular disease: physiological basis and prog-nostic implications. J Am Coll Cardiol 2008;51:1725–33.

60. Furlan R, Guzzetti S, Crivellaro W, et al.Continuous 24-hour assessment of the neuralregulation of systemic arterial pressure and RRvariabilities in ambulant subjects. Circulation1990;81:537–47.

61. Malliani A, Pagani M, Furlan R, et al. Individualrecognition by heart rate variability of twodifferent autonomic profiles related to posture.Circulation 1997;96:4143–5.

62. Bauer A, Kantelhardt JW, Barthel P, et al.Deceleration capacity of heart rate as a predictorof mortality after myocardial infarction: cohortstudy. Lancet 2006;367:1674–81.

63. Merchant FM, Dec GW, Singh JP. Implantablesensors for heart failure. Circ Arrhythm Electro-physiol 2010;3:657–67.

64. Singh JP, Rosenthal LS, Hranitzky PM, et al.Device diagnostics and long-term clinical outcomein patients receiving cardiac resynchronizationtherapy. Europace 2009;11:1647–53.

65. Whellan DJ, Ousdigian KT, Al-Khatib SM, et al.Combined heart failure device diagnostics identifypatients at higher risk of subsequent heart failurehospitalizations: results from PARTNERS HF (Pro-gram to Access and Review Trending Informationand Evaluate Correlation to Symptoms in PatientsWith Heart Failure) study. J Am Coll Cardiol 2010;55:1803–10.

66. Swedberg K, Komajda M, Bohm M, et al.Ivabradine and outcomes in chronic heart failure(SHIFT): a randomised placebo-controlled study.Lancet 2010;376:875–85.

67. Lang CC, Rayos GH, Chomsky DB, Wood AJ,Wilson JR. Effect of sympathoinhibition on exer-cise performance in patients with heart failure.Circulation 1997;96:238–45.

68. Cohn JN, Pfeffer MA, Rouleau J, et al. Adversemortality effect of central sympathetic inhibitionwith sustained-release moxonidine in patientswith heart failure (MOXCON). Eur J Heart Fail2003;5:659–67.

69. Bristow M. Antiadrenergic therapy of chronicheart failure: surprises and new opportunities.Circulation 2003;107:1100–2.

70. Van Stee EW. Autonomic innervation of theheart. Environ Health Perspect 1978;26:151–8.

71. Eckberg DL, Drabinsky M, Braunwald E.Defective cardiac parasympathetic control in pa-tients with heart disease. N Engl J Med 1971;285:877–83.

72. Li M, Zheng C, Sato T, Kawada T,Sugimachi M, Sunagawa K. Vagal nerve stimula-tion markedly improves long-term survival afterchronic heart failure in rats. Circulation 2004;109:120–4.

73. Zhang Y, Popovic ZB, Bibevski S, et al.Chronic vagus nerve stimulation improves auto-nomic control and attenuates systemic inflam-mation and heart failure progression in a caninehigh-rate pacing model. Circ Heart Fail 2009;2:692–9.

74. Hare JM, Keaney JF Jr., Balligand JL,Loscalzo J, Smith TW, Colucci WS. Role of nitricoxide in parasympathetic modulation of beta-adrenergic myocardial contractility in normaldogs. J Clin Invest 1995;95:360–6.

75. Sabbah HN, Gupta RC, Rastogi S, Mishra S,Mika Y, Burkhoff D. Treating heart failure withcardiac contractility modulation electrical signals.Curr Heart Fail Rep 2006;3:21–4.

76. Mann DL. Innate immunity and the failingheart: the cytokine hypothesis revisited. Circ Res2015;116:1254–68.

77. Springer J, Okonko DO, Anker SD. Vagal nervestimulation in chronic heart failure: an antiin-flammatory intervention? Circulation 2004;110:e34; author reply e34.

78. Anholt TA, Ayal S, Goldberg JA. Recruitmentand blocking properties of the CardioFit stimula-tion lead. J Neural Eng 2011;8:034004.

79. De Ferrari GM, Crijns HJ, Borggrefe M, et al.Chronic vagus nerve stimulation: a new andpromising therapeutic approach for chronic heartfailure. Eur Heart J 2011;32:847–55.

80. Milby AH, Halpern CH, Baltuch GH. Vagusnerve stimulation for epilepsy and depression.Neurotherapeutics 2008;5:75–85.

81. Schauerte P, Mischke K, Plisiene J, et al.Catheter stimulation of cardiac parasympatheticnerves in humans: a novel approach to the cardiacautonomic nervous system. Circulation 2001;104:2430–5.

82. Schwartz PJ, De Ferrari GM, Sanzo A, et al.Long term vagal stimulation in patients withadvanced heart failure: first experience in man. EurJ Heart Fail 2008;10:884–91.

83. Zannad F, De Ferrari GM, Tuinenburg AE, et al.Chronic vagal stimulation for the treatment of lowejection fraction heart failure: results of the NEu-ral Cardiac TherApy foR Heart Failure (NECTAR-HF) randomized controlled trial. Eur Heart J 2015;36:425–33.

84. Premchand RK, Sharma K, Mittal S, et al.autonomic regulation therapy via left or rightcervical vagus nerve stimulation in patients withchronic heart failure: results of the ANTHEM-HFtrial. J Card Fail 2014;20:808–16.

85. Hauptman PJ, Schwartz PJ, Gold MR, et al.Rationale and study design of the increase of vagaltone in heart failure study: INOVATE-HF. Am HeartJ 2012;163:954–62.e1.

86. Effects of Transvenous Vagus Nerve Stimula-tion on Immune Response. Available at: https://clinicaltrials.gov/ct2/show/NCT01944228. AccessedApril 12, 2015.

87. Chapleau MW, Hajduczok G, Abboud FM.Mechanisms of resetting of arterial barorecep-tors: an overview. Am J Med Sci 1988;295:327–34.

88. Thrasher TN. Unloading arterial barorecep-tors causes neurogenic hypertension. Am J

Physiol Regul Integr Comp Physiol 2002;282:R1044–53.

89. Heusser K, Tank J, Engeli S, et al. Carotidbaroreceptor stimulation, sympathetic activity,baroreflex function, and blood pressure in hy-pertensive patients. Hypertension 2010;55:619–26.

90. Biaggioni I, Whetsell WO, Jobe J, Nadeau JH.Baroreflex failure in a patient with central nervoussystem lesions involving the nucleus tractus sol-itarii. Hypertension 1994;23:491–5.

91. Robertson D, Hollister AS, Biaggioni I,Netterville JL, Mosqueda-Garcia R, Robertson RM.The diagnosis and treatment of baroreflex failure.N Engl J Med 1993;329:1449–55.

92. Epstein SE, Beiser GD, Goldstein RE, et al.Circulatory effects of electrical stimulation of thecarotid sinus nerves in man. Circulation 1969;40:269–76.

93. Wang W, Brandle M, Zucker IH. Influence ofvagotomy on the baroreflex sensitivity in anes-thetized dogs with experimental heart failure. AmJ Physiol 1993;265:H1310–7.

94. Sabbah HN, Gupta RC, Imai M, et al. Chronicelectrical stimulation of the carotid sinus barore-flex improves left ventricular function and pro-motes reversal of ventricular remodeling in dogswith advanced heart failure. Circ Heart Fail 2011;4:65–70.

95. Creager MA. Baroreceptor reflex function incongestive heart failure. Am J Cardiol 1992;69:10G–5G; discussion 15G–6G.

96. Schultz HD, Li YL. Carotid body function inheart failure. Respir Physiol Neurobiol 2007;157:171–85.

97. Sun SY, Wang W, Zucker IH, Schultz HD.Enhanced peripheral chemoreflex function inconscious rabbits with pacing-induced heart fail-ure. J Appl Physiol (1985) 1999;86:1264–72.

98. Higashi T, Kobayashi N, Hara K, Shirataki H,Matsuoka H. Effects of angiotensin II type 1receptor antagonist on nitric oxide synthaseexpression and myocardial remodeling in Gold-blatt hypertensive rats. J Cardiovasc Pharmacol2000;35:564–71.

99. Gassler JP, Bisognano JD. Baroreflex activa-tion therapy in hypertension. J Hum Hypertens2014;28:469–74.

100. Karunaratne H, Muluk S, Papademetriou V,Park WM, Sample R, Irwin E. Implantation of acarotid baroreceptor stimulator in patients withpacemakers and hypertension. Pacing Clin Elec-trophysiol 2011;34:354–6.

101. Kang S, Singh J, Vidal Melo M, Gulur P. Spinalcord stimulation for intercostal neuralgia in a pa-tient with implantable cardiac defibrillator andbiventricular pacing. Neuromodulation 2014;17:386–8.

102. Bisognano JD, Bakris G, Nadim MK, et al.Baroreflex activation therapy lowers bloodpressure in patients with resistant hypertension:results from the double-blind, randomized,placebo-controlled Rheos pivotal trial. J Am CollCardiol 2011;58:765–73.

103. Gronda E, Seravalle G, Brambilla G, et al.Chronic baroreflex activation effects on sympathetic






































































































































































































802

nerve traffic, baroreflex function, and cardiac hae-modynamics in heart failure: a proof-of-conceptstudy. Eur J Heart Fail 2014;16:977–83.

104. Abraham WT, Zile MR, Weaver FA, et al.Baroreflex activation therapy for the treatment ofheart failure with a reduced ejection fraction. J AmColl Cardiol HF 2015;3:487–96.

105. Rheos Diastolic Heart Failure Trial. Availableat: https://clinicaltrials.gov/ct2/show/NCT00718939.Accessed March 1, 2015.

106. Rheos Hope For Heart Failure (HOPE4HF)Trial. Available at: https://clinicaltrials.gov/ct2/show/NCT00957073. Accessed March 1, 2015.

107. Acute Carotid Sinus Endovascular StimulationII Study (ACES II). Available at: https://clinicaltrials.gov/ct2/show/NCT01458483. Accessed March 1,2015.

108. Song JJ, Popescu A, Bell RL. Present andpotential use of spinal cord stimulation to controlchronic pain. Pain Physician 2014;17:235–46.

109. Falowski S, Celii A, Sharan A. Spinal cordstimulation: an update. Neurotherapeutics 2008;5:86–99.

110. Olgin JE, Takahashi T, Wilson E, Vereckei A,Steinberg H, Zipes DP. Effects of thoracic spinalcord stimulation on cardiac autonomic regulation ofthe sinus and atrioventricular nodes. J CardiovascElectrophysiol 2002;13:475–81.

111. Huynh K. Heart failure: thoracic spinal cordstimulation for the treatment of HF. Nat RevCardiol 2015;12:66.

112. Bernstein SA, Wong B, Vasquez C, et al. Spinalcord stimulation protects against atrial fibrillationinduced by tachypacing. Heart Rhythm 2012;9:1426–33.e3.

113. Issa ZF, Zhou X, Ujhelyi MR, et al. Thoracicspinal cord stimulation reduces the risk ofischemic ventricular arrhythmias in a post-infarction heart failure canine model. Circulation2005;111:3217–20.

114. Odenstedt J, Linderoth B, Bergfeldt L, et al.Spinal cord stimulation effects on myocardialischemia, infarct size, ventricular arrhythmia, andnoninvasive electrophysiology in a porcine ischemia-reperfusion model. Heart Rhythm 2011;8:892–8.

115. Lopshire JC, Zhou X, Dusa C, et al. Spinal cordstimulation improves ventricular function and re-duces ventricular arrhythmias in a canine post-infarction heart failure model. Circulation 2009;120:286–94.

116. Lopshire JC, Zipes DP. Spinal cord stimulationfor heart failure: preclinical studies to deter-mine optimal stimulation parameters for clinicalefficacy. J Cardiovasc Transl Res 2014;7:321–9.

117. Tse HF, Turner S, Sanders P, et al. ThoracicSpinal Cord Stimulation for HEArt Failure as aRestorative Treatment (SCS HEART Study): first-in-man experience. Heart Rhythm 2015;12:588–95.

118. Determining the Feasibility of Spinal CordNeuromodulation for the Treatment of ChronicHeart Failure (DEFEAT-HF). Available at: https://clinicaltrials.gov/ct2/show/NCT01112579. AccessedMarch 1, 2015.

119. Ljungqvist A, Wagermark J. The adrenergicinnervation of intrarenal glomerular and extra-glomerular circulatory routes. Nephron 1970;7:218–29.

120. Sobotka PA, Mahfoud F, Schlaich MP,Hoppe UC, Bohm M, Krum H. Sympatho-renal axisin chronic disease. Clin Res Cardiol 2011;100:1049–57.

121. Bohm M, Ewen S, Kindermann I, Linz D,Ukena C, Mahfoud F. Renal denervation and heartfailure. Eur J Heart Fail 2014;16:608–13.

122. Mahfoud F, Luscher TF, Andersson B, et al.Expert consensus document from the EuropeanSociety of Cardiology on catheter-based renaldenervation. Eur Heart J 2013;34:2149–57.

123. Epstein M, de Marchena E. Is the failureof SYMPLICITY HTN-3 trial to meet itsefficacy endpoint the “end of the road” forrenal denervation? J Am Soc Hypertens 2015;9:140–9.

124. Wilson GJ, Winsor-Hines D, Tunstall RR, et al.Bipolar multi-electrode balloon catheter radio-frequency renal denervation with the Vessix sys-tem: preclinical safety evaluation. EuroIntervention2015;10:1239–46.

125. Davies JE, Manisty CH, Petraco R, et al. First-in-man safety evaluation of renal denervation forchronic systolic heart failure: primary outcomefrom REACH-Pilot study. Int J Cardiol 2013;162:189–92.

126. Renal Denervation in Patients With ChronicHeart Failure & Renal Impairment Clinical Trial(SymplicityHF). Available at: https://clinicaltrials.gov/ct2/show/NCT01392196. Accessed March 1,2015.

127. Renal Sympathetic Denervation for PatientsWith Chronic Heart Failure (RSD4CHF). Availableat: https://www.clinicaltrials.gov/ct2/show/NCT01790906. Accessed March 1, 2015.

128. Brandt MC, Mahfoud F, Reda S, et al. Renalsympathetic denervation reduces left ventricularhypertrophy and improves cardiac function in pa-tients with resistant hypertension. J Am Coll Car-diol 2012;59:901–9.

129. Zile MR, Little WC. Effects of autonomicmodulation: more than just blood pressure. J AmColl Cardiol 2012;59:910–2.

130. Verloop WL, Beeftink MM, Nap A, et al. Renaldenervation in heart failure with normal left ven-tricular ejection fraction. Rationale and design ofthe DIASTOLE (DenervatIon of the renAl Sympa-thetic nerves in hearT failure with nOrmal LvEjection fraction) trial. Eur J Heart Fail 2013;15:1429–37.

131. Renal Sympathetic Modification in Patients withHeart Failure (SWAN-HF). Available at: https://www.clinicaltrials.gov/ct2/show/study/NCT01402726.Accessed April 12, 2015.

132. Bradfield JS, Vaseghi M, Shivkumar K. Renaldenervation for refractory ventricular arrhythmias.Trends Cardiovasc Med 2014;24:206–13.

133. Linz D, Hunnik A, Ukena C, et al. Effects ofrenal denervation on atrial arrhythmogenesis.Future Cardiol 2014;10:813–22.

134. Mahfoud F, Schlaich M, Kindermann I, et al.Effect of renal sympathetic denervation onglucose metabolism in patients with resistant hy-pertension: a pilot study. Circulation 2011;123:1940–6.

135. Witkowski A, Prejbisz A, Florczak E, et al.Effects of renal sympathetic denervation on bloodpressure, sleep apnea course, and glycemic con-trol in patients with resistant hypertension andsleep apnea. Hypertension 2011;58:559–65.

136. Yu L, Scherlag BJ, Li S, et al. Low-leveltranscutaneous electrical stimulation of theauricular branch of the vagus nerve: a noninvasiveapproach to treat the initial phase of atrial fibril-lation. Heart Rhythm 2013;10:428–35.

137. Wang Z, Yu L, Wang S, et al. Chronic in-termittent low-level transcutaneous electricalstimulation of auricular branch of vagus nerveimproves left ventricular remodeling in consciousdogs with healed myocardial infarction. Circ HeartFail 2014;7:1014–21.

138. Nabutovsky Y, Florio J, Morgan K, Grill WM,Farazi TG. Lead design and initial applications of anew lead for long-term endovascular vagal stim-ulation. Pacing Clin Electrophysiol 2007;30 Suppl1:S215–8.

139. Kobayashi M, Sakurai S, Takaseya T, et al.Effects of percutaneous stimulation of both sym-pathetic and parasympathetic cardiac autonomicnerves on cardiac function in dogs. Innovations(Phila) 2012;7:282–9.

140. Costanzo MR, Khayat R, Ponikowski P,et al. Mechanisms and clinical consequences ofuntreated central sleep apnea in heart failure.J Am Coll Cardiol 2015;65:72–84.

141. Ponikowski P, Javaheri S, Michalkiewicz D,et al. Transvenous phrenic nerve stimulation forthe treatment of central sleep apnoea in heartfailure. Eur Heart J 2012;33:889–94.

KEY WORDS carotid baroreceptor,heart failure, renal denervation,spinal cord stimulation, vagal nervestimulation

Go to http://www.acc.org/jacc-journals-cme to take the CMEquiz for this article.
























































































https://www.clinicaltrials.gov/ct2/show/NCT01790906

https://www.clinicaltrials.gov/ct2/show/NCT01790906
















https://www.clinicaltrials.gov/ct2/show/study/NCT01402726

https://www.clinicaltrials.gov/ct2/show/study/NCT01402726
















































novel interventional therapies to modulate the autonomic...

Documents