Scholarly article on topic 'Autonomic regulation therapy suppresses quantitative T-wave alternans and improves baroreflex sensitivity in patients with heart failure enrolled in the ANTHEM-HF study'

Autonomic regulation therapy suppresses quantitative T-wave alternans and improves baroreflex sensitivity in patients with heart failure enrolled in the ANTHEM-HF study Academic research paper on "Basic medicine"

CC BY-NC-ND
0
0
Share paper
Academic journal
Heart Rhythm
OECD Field of science
Keywords
{"Heart failure" / "T-wave alternans" / "Baroreceptor sensitivity" / "Vagus nerve stimulation" / "Autonomic regulation therapy" / "Heart rate turbulence" / "Ventricular tachycardia" / "Heart rate variability" / "Autonomic tone" / "Autonomic reflexes"}

Abstract of research paper on Basic medicine, author of scientific article — Imad Libbus, Bruce D. Nearing, Badri Amurthur, Bruce H. KenKnight, Richard L. Verrier

Background Autonomic regulation therapy (ART) with chronic vagus nerve stimulation improves ventricular function in patients with chronic heart failure, but its effects on quantitative T-wave alternans (TWA), ventricular tachycardia (VT), baroreflex sensitivity, and autonomic tone remained unknown. Objective Effects on TWA, a marker of risk of life-threatening arrhythmias; heart rate turbulence (HRT), an indicator of baroreflex sensitivity; heart rate variability; and VT incidence were studied in 25 patients with chronic symptomatic heart failure and reduced ejection fraction enrolled in the ANTHEM-HF study (NCT01823887). Methods Twenty-four-hour ambulatory electrocardiographic recordings made before ART system (Cyberonics, Inc., Houston, TX) implantation involving the left or right vagus nerve and after 6 and 12 months of chronic therapy (10-Hz frequency, 250-μs pulse width, maximum tolerable current amplitude after 10 weeks of titration) at low-intensity (<2 mA; n = 10, 40%) or high-intensity (≥2 mA; n = 15, 60%) stimulation levels were analyzed. Results At 12 months, peak TWA levels were reduced by 29% from 71.0 ± 4.6 to 50.5 ± 1.8 μV (P < .0001). The number of patients with severely abnormal TWA (≥60 μV) was reduced by 76% from 17 to 4 (P < .0005), and the number of patients with nonsustained VT decreased by 73% from 11 to 3 (P < .025). HRT slope (P < .025), high frequency heart rate variability (HRV) (P = .05), and square root of the mean squared differences of successive normal-to-normal interval HRV (P = .013) increased. The mean heart rate derived from 24-hour Holter electrocardiograms decreased by 10% from 77 ± 2 to 69 ± 2 beats/min (P = .0002). HRT onset was unchanged. Conclusion Chronic ART in patients with symptomatic heart failure improves cardiac electrical stability, as reflected by reduced TWA levels and heart rate, suppresses VT, and increases baroreceptor sensitivity. These observations deserve study in a larger population.

Academic research paper on topic "Autonomic regulation therapy suppresses quantitative T-wave alternans and improves baroreflex sensitivity in patients with heart failure enrolled in the ANTHEM-HF study"

Autonomic regulation therapy suppresses quantitative T-wave alternans and improves baroreflex sensitivity in patients with heart failure enrolled in the ANTHEM-HF study ©

Imad Libbus, PhD,* Bruce D. Nearing, PhD,f Badri Amurthur, MS,* Bruce H. KenKnight, PhD,* Richard L. Verrier, PhD, FHRSt

From * Cyberonics, Inc., Houston, Texas, and ^Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts.

CrossMark

BACKGROUND Autonomic regulation therapy (ART) with chronic vagus nerve stimulation improves ventricular function in patients with chronic heart failure, but its effects on quantitative T-wave alternans (TWA), ventricular tachycardia (VT), baroreflex sensitivity, and autonomic tone remained unknown.

OBJECTIVE Effects on TWA, a marker of risk of life-threatening arrhythmias;heart rate turbulence (HRT), an indicator of baroreflex sensitivity; heart rate variability; and VT incidence were studied in 25 patients with chronic symptomatic heart failure and reduced ejection fraction enrolled in the ANTHEM-HF study (NCT01823887).

METHODS Twenty-four-hour ambulatory electrocardiographic recordings made before ART system (Cyberonics, Inc., Houston, TX) implantation involving the left or right vagus nerve and after 6 and 12 months of chronic therapy (10-Hz frequency, 250-|s pulse width, maximum tolerable current amplitude after 10 weeks of titration) at low-intensity (<2 mA; n = 10, 40%) or high-intensity (> 2 mA; n = 15, 60%) stimulation levels were analyzed.

RESULTS At 12 months, peak TWA levels were reduced by 29% from 71.0 ± 4.6 to 50.5 ± 1.8 |V (P < .0001). The number of patients with severely abnormal TWA (>60 |V) was reduced by 76% from 17 to 4 (P < .0005), and the number of patients with nonsustained VT decreased by 73% from 11 to 3 (P < .025). HRT slope (P < .025), high frequency heart rate variability (HRV) (P = .05), and square root of the mean squared differences of successive normal-to-normal interval HRV (P = .013) increased. The mean heart rate derived from 24-hour Holter electrocardiograms

decreased by 10% from 77 ± 2 to 69 ± 2 beats/min (P = .0002). HRT onset was unchanged.

CONCLUSION Chronic ART in patients with symptomatic heart failure improves cardiac electrical stability, as reflected by reduced TWA levels and heart rate, suppresses VT, and increases baroreceptor sensitivity. These observations deserve study in a larger population.

KEYWORDS Heart failure;T-wave alternans;Baroreceptor sensitivity;Vagus nerve stimulation;Autonomic regulation therapy;Heart rate turbulence;Ventricular tachycardia;Heart rate variability;Autonomic tone;Autonomic reflexes

ABBREVIATIONS ANTHEM-HF = Autonomic Neural Regulation Therapy to Enhance Myocardial Function in Heart Failure;ART = autonomic regulation therapy; ECG = electrocardiogram/ electrocardiographic;^ = high-frequency;HRT = heart rate turbulence;HRV = heart rate variability; LF = low-frequency; LV = left ventricular; LVEF = left ventricular ejection fraction; LVESV = left ventricular end-systolic volume;MMA = Modified Moving Average;NECTAR-HF = NEural Cardiac TherApy foR Heart Failure;NYHA = New York Heart Association;rMSSD = square root of the mean squared differences of successive normal-to-normal intervals;SCD = sudden cardiac death;SDNN = standard deviation of normal-to-normal intervals;TWA = T-wave alternans;VNS = vagus nerve stimulation;VT = ventricular tachycardia

(Heart Rhythm 2016;13:721-728)1 2016 Heart Rhythm Society. All rights reserved.

This study was funded by a grant from Cyberonics to Beth Israel Deaconess Medical Center, RL Verrier, Principal Investigator. Dr Libbus, Mr Amurthur, and Dr KenKnight are employed by Cyberonics. Dr Nearing and Dr Verrier receive royalty from Georgetown University and BIDMC for the Modified Moving Average method for T-wave alternans analysis, which is licensed by GE Healthcare. Address reprint requests and correspondence: Dr Richard L. Verrier, Division of Cardiovascular Medicine, Harvard Medical School, Beth Israel Deaconess Medical Center, Harvard-Thorndike Electrophysiology Institute, 99 Brookline Avenue, RN-301, Boston, MA 02215. E-mail address: rverrier@ bidmc.harvard.edu.

Introduction

The prognosis of patients with advanced heart failure remains poor despite extensive use of pharmacologic therapy and devices.1 Because autonomic dysfunction characterized by excessive sympathetic nerve activity and concomitant withdrawal of parasympathetic activation2 are critically implicated in heart failure morbidity and mortality, there has been strong interest in chronic nerve stimulation strategies to counteract these abnormalities.3 Autonomic regulation therapy (ART) via chronic vagus nerve stimulation

1547-5271/$-see front matter © 2016 Heart Rhythm Society. All rights reserved.

http://dx.doi.org/10.1016/j.hrthm.2015.11.030

(VNS) has been a focus of considerable research because of its multifactorial cardioprotective mechanisms4-7 and the excellent safety profile established in managing > 100,000 patients with epilepsy or depression for > 2 decades.8

In a pioneering study, Schwartz and De Ferrari9 conducted a "first-in-man" study of the effects of chronic VNS in a series of 8 patients with advanced heart failure. They demonstrated a significant improvement in left ventricular end-systolic volume (LVESV) and New York Heart Association (NYHA) class, supporting feasibility and safety. This experience was then extended to the multicenter international CardioFit study (NCT00461019), which enrolled a total of 32 patients and confirmed favorable effects on left ventricular ejection fraction (LVEF), LVESV, NYHA class, and quality of life measures.10

However, there is limited information on the effects of VNS on cardiac electrophysiological properties and susceptibility to ventricular arrhythmias in patients with heart failure. Recently, it was demonstrated that VNS reduced T-wave alternans (TWA) and improved heart rate variability (HRV) in patients with drug refractory epilepsy in a dose-dependent manner.11

In the present study, we hypothesized that concomitant with the previously demonstrated improvement in left ventricular (LV) function,10,12,13 VNS would improve autonomic reflexes, reduce VT incidence, and decrease TWA, an established marker of risk of sudden cardiac death (SCD) in patients with diverse cardiovascular diseases including heart failure.14,15 To address this issue, we investigated the effects of ART using previously established quantitative methods in ambulatory electrocardiographic (ECG) recordings (24-hour duration) from patients with chronic symptomatic heart failure and reduced LVEF enrolled in the Autonomic Neural Regulation Therapy to Enhance Myocardial Function in Heart Failure (ANTHEM-HF) study (NCT01823887). The autonomic measures included HRV16 and heart rate turbulence (HRT),17 which reflect autonomic tone and baroreflex sensitivity, respectively.

Methods

Study design and patient selection

The design and patient selection criteria of the ANTHEM-HF study design have been previously described.12,13 The study complied with the Declaration of Helsinki. The study protocol was approved by local ethics committees at all sites, and all patients gave written informed consent translated into local languages.

Briefly, individuals in NYHA functional class II/III heart failure, aged >18 years, were enrolled at 10 sites. Inclusion criteria included LVEF <40%, LV end-diastolic dimension >50 and <80 mm, QRS width <150 ms, and receiving optimal medical management, including stable p-blocker therapy for heart failure as indicated and tolerated for > 3 months and all other oral pharmacologic therapies for heart failure, including angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, loop diuretics, and spiro-nolactone, for > 1 month. Patients were also required to be

capable of performing the 6-minute walk test with a baseline distance of 150-425 meters, as limited by heart failure symptoms. All 25 patients enrolled in the ANTHEM-HF study whose Holter ECG files could be extracted and were available for analysis at all 3 time points (baseline, 6 months, and 12 months) were included in the present analysis.

System implantation and VNS stimulation

Individuals fulfilling the inclusion and exclusion criteria underwent VNS Therapy System implantation (Demipulse Model 103 pulse generator and PerenniaFLEX Model 304 lead; Cyberonics, Inc., Houston, TX) with 1:1 randomized lead placement on either the left (n = 12) or the right (n = 13) cervical vagus nerve. The pulse generator was activated at 15 ± 3 days after implantation. All patients were initially stimulated at a pulse width of 130 ms, a pulse frequency of 10 Hz, and continuous cyclic (14-second active [on] and 66-second inactive [off]; 1080 cycles/day) stimulation.

Stimulation parameters were systematically adjusted during periodic clinic visits over a 10-week titration period to a pulse width of 250 ms, a pulse frequency of 10 Hz, and a target output current amplitude of 1.5-3.0 mA. VNS activation and inactivation periods were unrelated to the cardiac cycle (i.e., open loop), so that no intracardiac sensing lead was used. During titration sessions, VNS intensity was gradually increased in 0.25-mA steps with the use of a radiofrequency programmer (Model 250 programming system, Cyberonics) to levels that produced acute VNS-related adverse effects (tolerance zone boundary), such as activation of the expiratory reflex (mild cough) or moderate heart rate reduction during the VNS-active phase. When the VNS tolerance zone boundary was established by evidence of expiratory reflex activation or heart rate reduction, the output current was reduced by >1 output current step (0.25 mA) to ensure that the therapy was well tolerated. During the 10-week titration period, VNS intensity was progressively increased to an average output current of 2.0 ± 0.1 mA (left side: 2.2 ± 0.2 mA; right side: 1.8 ± 0.2 mA). Continuous cyclic stimulation at low-intensity (<2 mA; n = 10) or high-intensity (>2 mA; n = 15) levels was maintained as tolerated throughout the titration period and the 12-month follow-up period.

Ambulatory ECG recordings and analysis

Twenty-four-hour ambulatory ECG recordings were made (DigiTrak XT, Philips Medical Systems, Best, The Netherlands) at baseline and after 6 and 12 months of chronic therapy. Measurements of TWA, HRT, HRV, ventricular tachycardia (VT) incidence, and heart rate were performed by an investigator (B.D.N.) blinded to clinical status and recording sequence using Food and Drug Administration-cleared commercial software running on the MARS Ambulatory ECG Analysis System (GE Healthcare, Milwaukee, WI).

The peak TWA level was quantified from standard precordial leads V1 and V5 and aVF using the Modified Moving

Average (MMA) method.15 These configurations provide satisfactory reliability from one recording to another from the same individual. The MMA method uses the noise-rejection principle of recursive averaging. The maximum TWA level throughout the recording is reported as the TWA value for that patient. As established in patients with heart disease, a TWA cut point of > 47 mV was defined in this study as an abnormal TWA test and >60 mV as markedly abnormal.15 MMA-based TWA has been analyzed from ambulatory ECG recordings in studies enrolling ~ 1600 patients and generated odds ratios ranging from 2.94 to 17.1 for cardiovascular death and from 4.8 to 22.6 for SCD,18 comparable or superior to those obtained with exercise-based TWA.15

HRT parameters turbulence onset, which calculates the initial brief acceleration of sinus rate after a premature ventricular contraction, and turbulence slope, which characterizes the subsequent heart rate deceleration, were determined as continuous variables as described by Bauer et al.17 Normal levels of HRT onset are < 0%, while normal values of HRT slope are > 2.5 ms per RR interval.

HRV was analyzed in the frequency domain using the fast Fourier spectral transform. Accordingly, the beat stream of the RR interval series was transformed to compute high-frequency (HF) power within the frequency band 0.1500.400 Hz and low-frequency (LF) power within the frequency band 0.040-0.150 Hz and reported in milliseconds squared. The LF/HF ratio was calculated as LF divided by HF and is unitless. HF HRV is a general indicator of parasympathetic tone, while LF HRV and the LF/HF HRV ratio are indicators of autonomic tone and balance.16 Mean ± S.D. normal values of HF and LF HRV are 975 ± 203 and 1170 ± 416 ms2, respectively, while an LF/HF HRV ratio of 1.5-2.0 is considered normal. In the time domain, HRV was determined on the basis of standard deviation of normal-to-normal intervals (SDNN) and the square root of the mean squared differences of successive normal-to-normal intervals (rMSSD). Normal levels are 141 ± 39 ms2 for SDNN and 27 ± 12 ms2 for rMSSD (mean ± S.D.).16

Statistical methods

The effects of VNS on TWA, HRT, HRV, and VT were analyzed with paired t tests using a SAS statistical package (version 9.2, SAS Institute, Cary, NC). Correlation of TWA magnitude with VNS intensity was analyzed using linear regression in quartiles from lowest to highest current levels. Data are reported as mean ± SEM. P < .05 was considered statistically significant.

Results

Baseline characteristics

All 25 patients (Table 1) were classified as NYHA functional class II (68%) or III (32%), with an average LVEF of 33% ± 1%.

Table 1 Patient characteristics at enrollment

Left Right Overall

Characteristic (n = 12) (n = 13) (n = 25)

Demographic

Age (y) 46.8 ± 2.4 46.8 ± 2.1 46.8 ± 2.2

Sex: male 10 (83) 10 (77) 20 (80)

Medical history

Heart failure 4.0 ± 0.8 4.0 ± 0.9 4.0 ± 0.8

duration (y) Heart failure etiology

Ischemic 8 (67) 10 (77) 18 (72)

Nonischemic 4 (33) 3 (23) 7 (28)

Clinical examination

NYHA class II/III 8/4 (67/33) 9/4 (69/31) 17/8 (68/32)

MLHFQ score 39.6 ± 2.7 42.1 ± 2.0 40.9 ± 2.4

Body mass index 23.3 ± 0.6 26.0 ± 0.9 24.7 ± 0.8

(kg/m2)

LVEF (%) 33.8 ± 1.4 32.0 ± 1.2 32.9 ± 1.3

LVESV (mL) 100.3 ± 7.1 105.2 ± 5.9 102.9 ± 6.4

LVESD (mm) 50.4 ± 1.3 52.8 ± 1.0 51.7 ± 1.1

LVEDV (mL) 149.8 ± 8.5 154.1 ± 8.1 152.0 ± 8.1

LVEDD (mm) 60.8 ± 1.1 62.5 ± 0.9 61.6 ± 1.0

Heart rate (beats/min) 75 ±1 78 ±2 77 ± 2

Systolic BP (mm Hg) 110 ±3 105 ±3 107 ± 3

Diastolic BP (mm Hg) 75 ±2 70 ±2 72 ± 2

6MWT distance (meters) 312 ±7 285 ± 15 298 ± 12

QRS width (ms) 109 ±5 107 ±4 108 ± 4

Values are presented as mean ± SEM or as n (%). 6MWT — 6-minute walk test; BP — blood pressure;LVEDD — left ventricular end-diastolic dimension;LVEDV — left ventricular end-diastolic volume; LVEF — left ventricular ejection fraction; LVESD — left ventricular end-systolic dimension;LVESV — left ventricular end-systolic volume;MLHFQ — Minnesota Living with Heart Failure Questionnaire; NYHA — New York Heart Association.

The average left ventricular end-systolic dimension was 52 ± 1.1 mm, and the average LVESV was 103 ± 6 mL. Heart failure etiology was ischemic in 72% of patients. None of the patients had an implantable cardioverter-defibrillator or cardiac resynchronization therapy device at enrollment.

Heart failure assessments

Echocardiographic measures and measurements of patient functional status and quality of life are presented in Table 2. In the analysis population, all heart failure assessment measures improved significantly over the 12-month study period. ART increased 2 HRV markers of vagus nerve activity, namely, rMSSD (from 26.3 ± 1.9 to 33.6 ± 2.6 ms2 P — .013, to above the normal range) and HF (from 96.3 ± 13.5 to 185.0 ± 42.7 ms2; P — .05, which remained below the normal range). It did not alter SDNN (from 92.4 ± 4.6 to 99.8 ± 4.8 ms2; P — .16) or LF (from 195.9 ± 29.2 to 31.3 ± 62.7 ms2; P — .08), which were reduced compared to normal individuals, nor did it change the LF/HF ratio, which remained at 2.3 ± 0.3 (P — .80), and which was elevated compared to normal individuals.

Fewer patients experienced nonsustained VT >3 beats during 24-hour Holter recordings after ART at both 6 and 12 months (from 11 of 25 to 9 of 25 to 3 of 25; P < .025).

Table 2 Cardiac efficacy measures

Measure Baseline 6 mo P* 12 mo P*

LVEF (%) 32.9 ± 1.3 37.7 ± 1.6 <.00025 37.3 ± 1.4 < .0005

LVESV (mL) 102.9 ± 6.4 95.2 ± 6.8 <.025 94.0 ± 6.9 <.025

LVESD (mm) 51.7 ± 1.1 49.3 ± 1.0 <.005 49.5 ± 1.0 <.025

6MWT distance (meters) 298 ± 12 353 ± 12 <.00025 358 ± 13 < .0001

MLHFQ score 41 ± 2 21 ± 2 <.00001 21 ± 2 < .00001

Heart rate (beats/min) 76.9 ± 1.7 71.6 ± 1.7 <.01 69.0 ± 1.9 <.00025

NYHA class (I/II/III) 0/17/8 (0/68/32) 17/8/0 (68/32/0) <.0001 21/4/0 (84/16/0) < .0001

rMSSD HRV (ms2) 26.3 ± 1.9 29.8 ± 2.5 = 0.19 33.6 ± 2.6 .013

High-frequency HRV(ms2) 96.3 ± 13.5 120.5 ± 25.6 = 0.35 185.0 ± 42.7 .05

Patients with VT >3 beats 11 (44) 9(36) = 0.77 3(12) <.025

Values are presented as mean ± SEM or as n (%).

Abbreviations as in Table 1. HRV = heart rate variability;rMSSD = square root of the mean squared differences of successive normal-to-normal intervals; VT = ventricular tachycardia. *Compared to baseline.

The mean of peak TWA levels was 71 ± 4.6 mV at baseline (left side implant: 74 ± 6.3 p.V; right side implant: 68 ± 6.7 mV). Changes in TWA are presented in Table 3 and Figures 1 and 2. TWA in the pooled cohort (n = 25) did not improve after 6 months of continuous cyclic stimulation at either high (>2.0 mA; n = 15) or low (<2.0 mA; n = 10) intensity. After 12 months of ART, TWA at both levels of intensity was reduced by 29% (by 20.5 ± 4.2 mV, P = .0001). Patients who were titrated to a low- vs high-intensity stimulation achieved TWA reductions of 24% (by 15.5 ± 5.3 mV, P < .05) and 31% (by 23.8 ± 5.6 mV, P < .001), respectively. A strong correlation was found between VNS intensity and reduction in TWA level (r2 = 0.99; P = .011) (Figure 3). At baseline, 68% of patients (17 of 25) had TWA >60 pV, the cut point for strongly abnormal levels15; at 6 months, 56% (14 of 25) had TWA >60 pV (P = 0.56); and at 12 months, 16% (4 of 25) had TWA >60 pV (P < .0005). Eight patients achieved TWA <47 pV by the end of the 12-month follow-up period, an increase of 4 patients. Improvement in TWA at 12 months was similar in patients with left-sided vs right-sided stimulation (Table 3).

At baseline, the mean HRT onset value was —1.0% ± 0.3% and the mean HRT slope value was 4.6 ± 0.8 ms per RR interval. Thus, the study patients registered normal levels of HRT onset and slope.17 Changes in HRT parameters are

Table 3 T-wave alternans (mV)

Measure Baseline 6 mo

Left VNS 74.3 ± 6.3 61.9 ± 4.1 = 0.18 53.8 ± 1.9 <.005

Right VNS 67.9 ± 6.7 79.6 ± 11.0 = 0.45 47.4 ± 2.8 <.01

Low 64.0 ± 0.5 76.2 ± 0.7 = 0.46 48.5 ± 2.5 <.05 current

High 75.6 ± 5.7 67.7 ± 6.9 = 0.47 51.8 ± 2.5 <.001 current

Overall 71.0 ± 4.6 71.1 ± 6.2 =0.99 50.5 ± 1.8 <.0001

Values are presented as mean ± SEM. VNS = vagus nerve stimulation. "Compared to baseline.

presented in Figures 4 and 5. An improvement in HRT slope occurred in patients who received high-intensity stimulation (by 2.6 ± 0.7 ms per RR interval at 6 months, P < .005; by 4.5 ± 1.7 ms per RR interval at 12 months, P < .025), but the HRT slope was not altered in patients who received low-intensity stimulation at either period. In the combined cohort, HRT slope improved by 2.6 ± 0.7 ms per RR interval at 6 months (P < .001) and by 3.2 ± 1.3 ms per RR interval

Figure 1 Assessment of T-wave alternans (TWA) using the Modified Moving Average method in 3 representative patients with heart failure enrolled in the Autonomic Neural Regulation Therapy to Enhance Myo-cardial Function in Heart Failure (ANTHEM-HF) study at baseline and after 12 months of chronic autonomic regulation therapy. Templates of QRS-aligned superimposed beats reveal a separation in the morphology of A and B beats, reflecting the ABABAB pattern between the J point and the T wave, which is the hallmark of TWA and its electrophysiologic basis for estimating risk of arrhythmic death. In all these patients, tracings from before (left) and after vagus nerve stimulation (VNS) implantation (right) indicate substantial reductions in TWA from above to below the >60-mV cut point for severe abnormality.15

Overall Low Intensity High Intensity

(n=25) (n=10) (n=15)

Figure 2 Vagus nerve stimulation reduced the magnitude of T-wave alternans (TWA) in ambulatory electrocardiographic recordings at 12 months in the combined cohort (*P < .0001) and in patients receiving high-intensity (^P < .001) and low-intensity (*P < .05) stimulation. Results were also significant and similar for right- and left-sided VNS at 12 months.

at 12 months (P < .025). Compared to baseline, HRT onset was not statistically different at either 6 months or 12 months in either the left-sided or right-sided stimulation groups or combined cohort (Figure 5).

Discussion

This is the first report that chronic ART reduces TWA in patients with advanced heart failure, an established marker of risk of lethal arrhythmias in patients with cardiac dysfunc-tion.14,15 There was a corresponding reduction in VT incidence. HRT slope, a measure of baroreceptor sensitivity, was improved. As the HRT slope changes preceded the reduction in TWA, improvement in autonomic function may in part underlie the cardioprotective effect of VNS in patients with heart failure. Two HRV measures of vagus nerve activity improved, similar to findings in the parent study.13

Previous studies with chronic VNS in patients with heart failure

De Ferrari and colleagues9,10 were the first investigators to demonstrate proof of principle that VNS could be cardio-protective in patients with heart failure, presumably by enhancing parasympathetic tone to counteract the adverse effects of excessive adrenergic activity. This study represented a direct translation of experimental studies, in particular from a canine model of postinfarction heart failure, in which VNS decreased the incidence of sudden cardiac arrest.19 VNS has also been shown dramatically to reduce circulating cytokines, myocytes, and hypertrophy and to normalize baroreceptor control.4,5 Sabbah et al5 made the intriguing observation in canines that the combination of VNS and p-adrenergic blockade is particularly effective in improving LVEF.

De Ferrari et al10 reported the first clinical experience with ART in a phase II trial that enrolled 32 patients with heart failure. In the CardioFit study, a VNS system incorporating a right ventricular endocardial sensing electrode (CardioFit, BioControl Medical, Yehud, Israel) was used to titrate VNS

to achieve a targeted heart rate reduction. This was an open-label study involving right cervical VNS synchronized to the cardiac cycle. Therapy was well-tolerated with only minor adverse effects, such as cough, dysphonia, and stimulation-related pain, which were observed early in the study but which regressed with stimulation titration and resolved over time. Overall, there was an improvement in LVEF (from 22% to 29%; P < .001) and reduction in LV systolic volumes at 6 months and 1 year of treatment. Most recently, Zannad et al20 reported in the NEural Cardiac TherApy foR Heart Failure (NECTAR-HF) study that at 6 months of stimulation at 1.2 mA and 20 Hz, VNS failed to demonstrate significant effects on cardiac remodeling and functional capacity in patients with symptomatic heart failure. As the NECTAR-HF investigators suggested, HF stimulation used in the NECTAR-HF study may have provoked adverse effects (e.g., coughing) that prevented titration to output current levels needed to produce therapeutic effects.

The only other published report of the effects of VNS in patients with chronic heart failure is the ANTHEM-HF study, a multicenter, open-label, feasibility study designed to evaluate the safety, tolerability, and efficacy of a new approach to VNS in patients with chronic, stable, symptomatic heart failure and reduced LVEF.12,13 Continuous cyclic (14-seconds on, 66-seconds off) stimulation at an amplitude of 1.5-3.0 mA and at a frequency of 10 Hz remained fixed throughout the study. After 6 months of treatment, there were significant improvements in several measures of LV function, including LVEF, left ventricular end-systolic dimension, HRV (SDNN, rMSSD, LF, HF, and LF/HF ratio), 6-minute walk distance, and NYHA class. Although some outcome measures exhibited a trend toward higher efficacy with right-sided stimulation, implant side did not appear to be a statistically significant factor. Despite the use of lower-intensity VNS stimulation in the ANTHEM-HF study than in the CardioFit study (mean output current levels 2.0 mA vs 4.1 mA), the magnitude of improvement in efficacy measures appears to be comparable. The near equivalence may be due to the use of a higher 10-Hz stimulation frequency in the ANTHEM-HF study as compared with a stimulation frequency of 1 -2 pulses per heart beat in the CardioFit study.

"Hi (11

Figure 3 Reduction in T-wave alternans (TWA) was strongly correlated with vagus nerve stimulation intensity.

Figure 4 Heart rate turbulence slope (Ts) response to autonomic regulation therapy over a 12-month period in a representative patient, indicative of significantly improved baroreceptor sensitivity. PVC — premature ventricular contraction; RR — R-wave to R-wave; VNS — vagus nerve stimulation.

Present study

The present investigation involved an analysis of TWA, HRT, HRV, and ventricular arrhythmias from ambulatory ECG recordings of 25 patients enrolled in the ANTHEM-HF study. The study points included baseline, 6 months, and 12 months. Improvements in efficacy measures (cardiac function and heart failure symptoms) at 6 and 12 months were comparable with those in the full cohort (Table 2). Two HRV measures of vagus nerve tone—rMSSD and HF—improved as in the full cohort, although SDNN remained depressed.

u 12.00 ■ Baseline ■ 6 Months ■ 12 Months

1 ill || |

Figure 5 High- but not low-intensity autonomic regulation therapy significantly increased heart rate turbulence (HRT) slope at 6 months (fP < .005) with a further increase at 12 months (*P < .025). In the overall cohort, HRT slope increased at 6 months (*P < .001) and at 12 months (*P < .025).

TWA magnitude, the primary measure in the present analysis, was significantly reduced at 12 months in both the low- and high-intensity VNS stimulation groups, whether VNS implantation was left- or right-sided (Table 3 and Figure 2). A strong correlation was found between VNS intensity and reduction in TWA level (Figure 3). This observation concurs with our previous report in patients with drug refractory epilepsy.11 The mean TWA levels observed at baseline were in the severely abnormal range (>60 mV),15 an observation consistent with other studies on patients with LV dysfunction and cardiac mortality using the MMA method.14 Sakaki et al14 reported that in patients with reduced LVEF and both ischemic and nonischemic cardio-myopathy, TWA levels signaled an increased risk of cardiac mortality and witnessed SCD with an odds ratio of 17.1 (P < .0001) and 22.6 (P < .005), respectively. VNS substantially reduced the magnitude of TWA in the overall cohort by > 20 mV to a near-normal range. A differential of 20 mV has been shown to indicate differences of >55% in risk of cardiac mortality and of >58% in risk of SCD.15 Consistent with the decrease in TWA, we observed that VNS reduced the number of patients with VT > 3 beats.

Establishing the precise mechanisms whereby VNS reduces TWA is particularly challenging in light of the multifactorial effects of chronic VNS, which include reduction in inflammation, cytokine expression, apoptosis, heart rate, and sympathetic nerve effects and improvement in

baroreceptor sensitivity.4-7,9,10 Because heart rate and sympathetic nerve activity have been shown to affect TWA magnitude directly,21 they are strong putative mechanisms. Excessive sympathetic nerve activity independent of heart rate can increase TWA, and its antagonism protects against TWA. VNS is antisympathetic through classical mechanisms involving accentuated antagonism22 and inhibition of stellate ganglion activity.23 In this regard, it is relevant that left-sided VNS is effective in reducing TWA, as Chen and colleagues23 have shown that stimulation of the left vagus nerve results in ipsilateral inhibition of the left stellate ganglion. This structure exerts a major influence on susceptibility to ventricular fibrillation, as evidenced by the effectiveness of left stellectomy in reducing SCD. Our finding that VNS improves HRT slope, a measure of baroreceptor sensitiv-ity,17 also implicates effects on autonomic reflexes in VNS-mediated reduction in TWA. It is of interest that HRT changes occurred at 6 months, before the 12-month reduction in TWA, suggesting that autonomic mediators exert salutary influences that translate into myocardial substrate changes, which are manifest in improved cardiac electrical stability, as reflected in the reduction in TWA. While heart rate has been implicated as a factor in TWA level, it is not likely to have played a major role in the present study. This inference is based on the fact that the heart rate reduction between 6 and 12 months was only 2.6 beats/min (Table 2), yet the overall 29% decrease in TWA from 71.1 ± 6.2 to 50.5 ± 1.8 mV was highly significant (P < .0001) (Table 3).

Autonomic alterations with VNS

In the present study, we observed improvements in rMSSD and HF, HRV parameters indicating vagus nerve activity,16 as was the case for SDNN HRV in the full ANTHEM-HF study cohort.13 We also observed substantial increases in HRT slope, as illustrated in a representative patient comparing baseline screening to 12-month levels. In response to high-intensity stimulation, a significant increase in HRT slope was evident at 6 months and continued at 12 months. The basis whereby VNS improves HRT slope is unclear. Presumably, it reflects in part a central nervous system action through activation of vagal afferent fibers. The fact that VNS is effective in suppressing epileptic seizures24 supports the potent action of afferent activity on central nervous system function. However, changes in cardiac substrate are also known to be associated with improvement in HRT, as has been shown in patients recovering from myocardial infarc-tion.17 The fact that VNS affects both TWA and HRT slope favorably reinforces the view that this form of ART reduces risk of arrhythmic events, as the combination of these 2 measures yields a high level of prediction.15 The rationale appears to be that HRT reflects the existence of autonomic triggers and TWA indicates cardiac substrate vulnerability.

Study limitations

The sample size of this subanalysis, which has an enrollment of 25 patients, is relatively small. However, it is more than

half of the 46-patient enrollment of the second 6-month study of the ANTHEM-HF and represents a close matching with respect to measures of LV function, NYHA class, and heart failure etiology. Thus, the present findings merit further exploration in larger cohorts.

Conclusions and implications

Chronic high-intensity ART in patients with symptomatic heart failure can decrease cardiac electrical instability, as reflected in reduced TWA levels and suppression of VT, and can improve baroreflex sensitivity, as reflected in increased HRT slope, both indicators of cardiovascular mortality risk. These significant effects are comparable for left- and right-sided VNS. There appears to be an intensity-related effect indicative of a dose-response relationship of VNS, underscoring the importance of appropriate VNS parameter selection to optimize the potential benefits of ART. This method of chronic VNS may provide a safe and effective means not only to improve cardiac mechanical function in patients with reduced LVEF but also to protect against life-threatening ventricular arrhythmias and to improve cardio-protective autonomic reflexes.

References

1. Joffe SW, Webster K, McManus DD, Kiernan MS, Lessard D, Yarzebski J, Darling C, Gore JM, Goldberg RJ. Improved survival after heart failure: a community-based perspective. J Am Heart Assoc 2013;2:1-9.

2. Brunner-la Rocca HP, Esler MD, Jennings GL, Kaye DM. Effects of cardiac sympathetic nervous activity on mode of death in congestive heart failure. Eur Heart J 2001;22:1136-1143.

3. Lopshire JC, Zipes DP. Device therapy to modulate the autonomic nervous system to treat heart failure. Curr Cardiol Rep 2012;14:593-600.

4. Li W, Olshansky B. Inflammatory cytokines and nitric oxide in heart failure and potential modulation by vagus nerve stimulation. Heart Fail Rev 2011;16: 137-145.

5. Sabbah HN, Ilsar I, Zaretsky A, Rastogi S, Wang M, Gupta RC. Vagus nerve stimulation in experimental heart failure. Heart Fail Rev 2011;16:171-178.

6. Shen MK, Zipes DP. Role of the autonomic nervous system in modulating cardiac arrhythmias. Circ Res 2014;114:1004-1021.

7. Buckley U, Shivkumar K, Ardell JL. Autonomic regulation therapy in heart failure. Curr Heart Fail Rep 2015;12:284-293.

8. Conway CR, Colijn MA, Schachter SC. Vagus nerve stimulation for epilepsy and depression. In: Reti IM, ed. Brain Stimulation: Methodologies and Interventions. West Sussex, UK: John Wiley & Sons, Ltd; 2015:305-336.

9. Schwartz PJ, De Ferrari GM. Vagal stimulation for heart failure: background and first in-man study. Heart Rhythm 2009;6:S76-S81.

10. De Ferrari GM, HJGM Crijns, Borggrefe M, et al. Chronic vagus nerve stimulation: a new and promising therapeutic approach for chronic heart failure. Eur Heart J 2011;32:847-855.

11. Schomer AC, Nearing BD, Schachter SC, Verrier RL. Vagus nerve stimulation reduces cardiac electrical instability assessed by quantitative T-wave alternans analysis in patients with drug-resistant focal epilepsy. Epilepsia 2014;55: 1996-2002.

12. DiCarlo L, Libbus I, Amurthur B, KenKnight BH, Anand IS. Autonomic regulation therapy for the improvement of left ventricular function and heart failure symptoms: the ANTHEM-HF study. J Card Fail 2013;19:655-660.

13. Premchand RK, Sharma K, Mittal S, et al. Extended follow-up of patients with heart failure receiving autonomic regulation therapy in the ANTHEM-HF study. J Cardiac Fail, in press.

14. Sakaki K, Ikeda T, Miwa Y, Miyakoshi M, Abe A, Tsukada T, Ishiguro H, Mera H, Yusu S, Yoshino H. Time-domain T-wave alternans measured from Holter electrocardiograms predicts cardiac mortality in patients with left ventricular dysfunction: a prospective study. Heart Rhythm 2009;6:332-337.

15. Verrier RL, Klingenheben T, Malik M, El-Sherif N, Exner D, Hohnloser S, Ikeda T, Martinez JP, Narayan S, Nieminen T, Rosenbaum DS. Microvolt T-wave alternans: physiologic basis, methods of measurement, and clinical utility. Consensus guideline. J Am Coll Cardiol 2011;44:1309-1324.

16. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation 1996;93:1043-1065.

17. Bauer A, Malik M, Schmidt G, et al. Heart rate turbulence: standards of measurement, physiological interpretation, and clinical use. J Am Coll Cardiol 2008;52:1353-1365.

18. Verrier RL, Ikeda T. Ambulatory ECG-based T-wave alternans monitoring for risk assessment and guiding medical therapy: mechanisms and clinical applications. Prog Cardiovasc Dis 2013;56:172-185.

19. Vanoli E, De Ferrari GM, Stramba-Badiale M, Hull SS Jr, Foreman RD, Schwartz PJ. Vagal stimulation and prevention of sudden death in conscious dogs with a healed myocardial infarction. Circ Res 1991;68:1471-1481.

20. Zannad F, De Ferrari GM, Tuinenburg AE, et al. Chronic vagal stimulation for the treatment of low ejection fraction heart failure: results of the NEural Cardiac

TherApy foR Heart Failure (NECTAR-HF) randomized controlled trial. Eur Heart J 2015;36:425-433.

21. Verrier RL, Kumar K, Nearing BD. Basis for sudden cardiac death prediction by T-wave alternans from an integrative physiology perspective. Heart Rhythm 2009;6:416-422.

22. Kolman BS, Verrier RL, Lown B. The effect of vagus nerve stimulation upon vulnerability of the canine ventricle: role of sympathetic-parasympathetic interactions. Circulation 1975;52:578-585.

23. Shen MJ, Shinohara T, Park HW, et al. Continuous low-level vagus nerve stimulation reduces stellate ganglion nerve activity and paroxysmal atrial tachyarrhythmias in ambulatory canines. Circulation 2011;123:2204-2212.

24. Annegers JF, Coan SP, Hauser WA, Leestma J. Epilepsy, vagal nerve stimulation by the NCP system, all-cause mortality, and sudden, unexpected, unexplained death. Epilepsia 2000;41:549-553.

CLINICAL PERSPECTIVES

The present study demonstrates that chronic high-intensity autonomic regulation therapy in individuals with advanced heart failure enrolled in the Autonomic Neural Regulation Therapy to Enhance Myocardial Function in Heart Failure (ANTHEM-HF) study improved cardiac electrical stability, as indicated in a decrease in T-wave alternans magnitude and in ventricular tachycardia incidence, and improvement in baroreflex sensitivity, as manifest by an increase in heart rate turbulence slope. These protective effects were conferred by either left- or right-sided vagus nerve stimulation (VNS). The improvement in T-wave alternans exhibited a dose-response relationship with respect to stimulation intensity. The current method of chronic VNS may provide a safe and effective approach to improving cardiac mechanical function in patients with significantly reduced left ventricular ejection fraction and concurrently reduce risk of malignant ventricular arrhythmias. More than 2 decades of experience with VNS in > 100,000 patients with drug refractory epilepsy or depression supports the potential clinical utility of this technology. These new developments may help to fill an important void, given the fact that the prognosis of patients with advanced heart failure remains poor, despite extensive use of pharmacologic therapy and contemporary devices. As the tools for assessment of the benefits of VNS as well as the stimulation procedures have been developed, the main steps required for translation of these discoveries to clinical practice will involve expanded enrollment in multicenter trials and further optimization of stimulation parameters.