J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7

ª 2 0 1 7 P U B L I S H E D B Y E L S E V I E R O N B E H A L F O F 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 1 9 3 6 - 8 7 9 8 / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j c i n . 2 0 1 7 . 0 4 . 0 3 6

Low-Level Tragus Stimulation for theTreatment of Ischemia and ReperfusionInjury in Patients With ST-SegmentElevation Myocardial InfarctionA Proof-of-Concept Study

Lilei Yu, MD, PHD,a Bing Huang, MD, PHD,a Sunny S. Po, MD, PHD,b Tuantuan Tan, MD, PHD,c Menglong Wang, MD,a

Liping Zhou, MD,a Guannan Meng, MD,a Shenxu Yuan, MD,a Xiaoya Zhou, MD, PHD,a Xuefei Li, MD,a

Zhuo Wang, MD,a Songyun Wang, MD,a Hong Jiang, MDa

ABSTRACT

Fro

sit

of

Wu

Na

inc

au

con

Ma

OBJECTIVES The aim of this study was to investigate whether low-level tragus stimulation (LL-TS) treatment could

reduce myocardial ischemia-reperfusion injury in patients with ST-segment elevation myocardial infarction (STEMI).

BACKGROUND The authors’ previous studies suggested that LL-TS could reduce the size of myocardial injury induced

by ischemia.

METHODS Patients who presented with STEMI within 12 h of symptom onset, treated with primary percutaneous

coronary intervention, were randomized to the LL-TS group (n ¼ 47) or the control group (with sham stimulation

[n ¼ 48]). LL-TS, 50% lower than the electric current that slowed the sinus rate, was delivered to the right tragus once

the patients arrived in the catheterization room and lasted for 2 h after balloon dilatation (reperfusion). All patients

were followed for 7 days. The occurrence of reperfusion-related arrhythmia, blood levels of creatine kinase-MB,

myoglobin, N-terminal pro–B-type natriuretic peptide and inflammatory markers, and echocardiographic characteristics

were evaluated.

RESULTS The incidence of reperfusion-related ventricular arrhythmia during the first 24 h was significantly attenuated

by LL-TS. In addition, the area under the curve for creatine kinase-MB and myoglobin over 72 h was smaller in the LL-TS

group than the control group. Furthermore, blood levels of inflammatory markers were decreased by LL-TS. Cardiac

function, as demonstrated by the level of N-terminal pro–B-type natriuretic peptide, the left ventricular ejection fraction,

and the wall motion index, was markedly improved by LL-TS.

CONCLUSIONS LL-TS reduces myocardial ischemia-reperfusion injury in patients with STEMI. This proof-of-concept

study raises the possibility that this noninvasive strategy may be used to treat patients with STEMI undergoing primary

percutaneous coronary intervention. (J Am Coll Cardiol Intv 2017;10:1511–20) © 2017 Published by Elsevier on behalf of

the American College of Cardiology Foundation.

m the aDepartment of Cardiology, Renmin Hospital of Wuhan University; Cardiovascular Research Institute, Wuhan Univer-

y; Hubei Key Laboratory of Cardiology, Wuhan, Hubei, China; bHeart Rhythm Institute and Department of Medicine, University

Oklahoma Health Sciences Center, Oklahoma City, Oklahoma; and the cDepartment of Ultrasound Imaging, Renmin Hospital of

han University, Wuhan, Hubei, China. This work was supported by grants 81530011, 81570463, and 81600395 from the National

ture Science Foundation of China, grants 2016CFA065 and 2016CFA048 from the Natural Science Foundation of Hubei Prov-

e, and grant WJ2017C0005 from the Foundation of Health and Family Planning Commission of Hubei Province. All other

thors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Yu and Huang

tributed equally to this work.

nuscript received January 23, 2017; revised manuscript received April 11, 2017, accepted April 19, 2017.

ABBR EV I A T I ON S

AND ACRONYMS

AUC = area under the curve

cVNS = cervical vagus nerve

stimulation

LL-TS = low-level tragus

stimulation

LV = left ventricular

MIRI = myocardial ischemia-

reperfusion injury

NT-proBNP = N-terminal

pro–B-type-natriuretic peptide

PCI = percutaneous coronary

intervention

STEMI = ST-segment elevation

myocardial infarction

VA = ventricular arrhythmia

VPB = ventricular premature

beat

Yu et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7

Noninvasive Neuromodulation in Patients With STEMI A U G U S T 1 4 , 2 0 1 7 : 1 5 1 1 – 2 0

1512

S T-segment elevation myocardial infar-ction (STEMI) accounts for approxi-mately 25% to 40% of acute myocardial

infarction and remains an important cause ofdisability and mortality throughout the world(1–3). Although reopening the culprit coronaryartery via mechanical or pharmacologicalreperfusion intervention is necessary torescue the ischemic myocardium and toreduce infarct size in patients with STEMI,paradoxically, reperfusion itself also triggersfurther injury, which is known as myocardialischemia-reperfusion injury (MIRI) (4,5).Growing evidence from experimental studiesand small-sized proof-of-concept clinical trialsshows thatMIRI contributes greatly to the finalinfarct size and cardiac function (6,7). Howev-er, there is currently no specific treatmentthat targets MIRI in patients with STEMI.

Thus, new therapies that can decrease MIRI in thesepatients are needed.

SEE PAGE 1521

Cervical vagus nerve stimulation (cVNS), whichinduces significant heart rate reduction (8–10), orwith a stimulus strength 50% to 80% below thethreshold needed to reduce heart rate (11–13), hasbeen shown to inhibit inflammatory responses,decrease the release of reactive oxygen species, sup-press cellular apoptosis, and attenuate MIRI inseveral myocardial ischemia-reperfusion models.Transcutaneous electric stimulation of the auricularbranch of the vagus nerve, located at the tragus, is anoninvasive approach to stimulating the afferentvagal nerve fibers (14). We recently found thatlow-level tragus stimulation (LL-TS) could decreaseinflammatory responses, attenuate cardiac structuraland autonomic remodeling, and improve ventricularfunction in a canine model of chronic myocardialinfarction (15,16). LL-TS also has been shown tosuppress atrial fibrillation and to decrease levels ofinflammatory cytokines in patients with paroxysmalatrial fibrillation (17). In the present study, weinvestigated whether LL-TS could attenuate MIRI inpatients with STEMI undergoing primary percuta-neous coronary intervention (PCI).

METHODS

ETHICS STATEMENT. This was a prospective, single-center, randomized, open-labeled study. The studyprotocol was approved by the ethics committee ofRenmin Hospital of Wuhan University. All patientsprovided informed consent.

STUDY POPULATION. Patients 18 to 80 years of agewho were diagnosed with STEMI, presented within12 h of symptom onset, and underwent primary PCIfrom November 2015 until September 2016 wereenrolled. Major exclusion criteria included a historyof prior myocardial infarction, severe heart failure(left ventricular [LV] ejection fraction <30%),cardiogenic shock, ventricular fibrillation or cardiacarrest, previous malignant hematological disease,previous known renal failure (estimated glomerularfiltration rate <30 ml/min), and inability or unwill-ingness to provide informed consent (Figure 1A). Inaddition, patients with left main or multiple coronaryartery disease were also excluded from this study.

STUDY DESIGN. The study design is presented inFigure 1B. All patients were randomized to receiveeither LL-TS and PCI (LL-TS group) or sham LL-TS andPCI treatment (control group). Aspirin (300 mg) andticagrelor (180 mg) were used once patients werediagnosed with STEMI. Patients underwent PCIaccording to standard guidelines (1). The use ofthrombus aspiration, glycoprotein IIb/IIIa inhibition,and drug-eluting stents during PCI was left tothe discretion of the treating physician. Standardmedical therapy, including statins, aspirin and tica-grelor or clopidogrel, beta-blockers, and angiotensin-converting enzyme inhibitors or angiotensin receptorblockers at recommended daily doses, was adminis-tered for post–myocardial infarction secondary pre-vention (1) (Table 1). Blood sampling, 24 h Holtermonitoring, and echocardiography were conducted atpre-specified time points in the 2 groups.

LL-TS. All patients were conscious and were notgiven any sedatives during the operation. LL-TS wasperformed on the tragus in the right ear (Figure 2A).Incremental electric currents were applied to thetragus (20 Hz, 1-ms duration) using a stimulator (S20,Jinjiang, Chengdu City, China) until slowing of thesinus rate was achieved (Figures 2B and 2C). Thelowest electric current necessary to slow the sinusrate was defined as the stimulation threshold. LL-TSwas set at 50% below the threshold with a dutycycle of 5 s on and 5 s off. LL-TS or sham LL-TS wasinitiated once the patient arrived in the catheteriza-tion room and lasted for 2 h after balloon dilatation(reperfusion) (Figure 2D). The sham stimulationduration in the control group and stimulationduration in the LL-TS group were 156 � 8 min and155 � 6 min, respectively (p ¼ 0.49) (Table 1).

ISCHEMIA-REPERFUSION RELATED VENTRICULAR

ARRHYTHMIAS. In both groups, continuous, digital12-lead electrocardiographic Holter monitoring

FIGURE 1 Diagram of Patient Flow and Study Flowchart in This Study

(A) Patient flow; (B) study flowchart. BD ¼ balloon dilatation (reperfusion); CR ¼ catheterization room; LL-TS ¼ low-level tragus stimulation;

MI ¼ myocardial infarction; PCI ¼ percutaneous coronary intervention; STEMI ¼ ST-segment elevation myocardial infarction.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7 Yu et al.A U G U S T 1 4 , 2 0 1 7 : 1 5 1 1 – 2 0 Noninvasive Neuromodulation in Patients With STEMI

1513

(CT-86, Baihui, Hangzhou City, China) was initiatedafter reperfusion and lasted for 24 h. Holter softwarewas used to analyze the severity of ventriculararrhythmias (VAs), which were classified as isolatedventricular premature beats (VPB), coupled VPBs(2 consecutive VPBs), or ventricular tachycardia (3 ormore consecutive VPBs).

MEASUREMENTS OF BLOOD LEVELS OF CREATINE

KINASE-MB, MYOGLOBIN, AND N-TERMINAL

PRO-B-TYPE-NATRIURETIC PEPTIDE. Venous bloodsamples at baseline (on admission) and at 0, 6, 12, 24,48, and 72 h, as well as 7 days, after reperfusionwere obtained from patients in a supine position.For measurements of creatine kinase-MB andmyoglobin, samples from 0, 6, 12, 24, 48, and 72 hafter reperfusion were evaluated using an ADVIACentaur XP automatic chemiluminescence immuno-assay analyzer (Siemens Healthcare Diagnostics,

Munich, Germany). For measurement of N-terminalpro–B-type natriuretic peptide (NT-pro BNP), samplesthat were obtained at baseline (on admission) and24 h and 7 days after reperfusion were analyzed usinga Dimension EXL with an LM automatic biochemicalanalyzer (Siemens Healthcare Diagnostics).

MEASUREMENT OF INFLAMMATORY MARKERS. Bloodsamples that were obtained at baseline (on admis-sion) and 24 h after reperfusion were usedto evaluate interleukin-6, interleukin-1b, high-mobility group-box 1 protein, and tumor necrosisfactor–a levels with commercially available enzyme-linked immunosorbent assay kits according to themanufacturer’s instructions (Human ELISA kit,CUSABIO Science, Wuhan, China). All samples wereanalyzed in duplicate, and the analysis wasrepeated if the difference between duplicates was>15%.

TABLE 1 Characteristics of the Patients

Control(n ¼ 48)

LL-TS(n ¼ 47) p Value

Age (yrs) 58 � 9 59 � 11 0.63

Male 34 (70.8) 37 (78.7) 0.38

Smoking 28 (58.3) 30 (63.8) 0.58

Hypertension 37 (77.1) 32 (68.1) 0.33

Diabetes 16 (33.3) 14 (29.8) 0.71

Dyslipidemia 10 (20.8) 8 (17.0) 0.64

Total ischemia time (h) 5.8 � 3.1 6.3 � 3.2 0.51

Ischemia time distribution 0.84

Ischemia time <3 h 11 (22.9) 9 (19.1) 0.65

Ischemia time 3–8 h 23 (47.9) 22 (46.9) 0.91

Ischemia time >8 h 14 (29.2) 16 (34.0) 0.61

Baseline sinus rate (beats/min) 67 � 10 65 � 9 0.31

Discomfort threshold (mA) 4.0 � 1.8 4.1 � 1.6 0.78

Threshold for slowing sinus rate (mA) 5.2 � 2.7 5.8 � 3.1 0.32

Average heart rate during LL-TS/sham LL-TS (beats/min)

81 � 13 79 � 10 0.40

Door-to-balloon time (min) 67 � 9 65 � 8 0.26

Time from admission to stimulation (min) 31 � 8 30 � 7 0.52

Time from stimulation to reperfusion (min) 36 � 8 35 � 6 0.49

LL-TS or sham LL-TS duration (min) 156 � 8 155 � 6 0.49

Culprit vessel 0.80

RCA 14 (29.2) 11 (23.4) 0.52

LAD 31 (64.6) 32 (68.1) 0.72

LCx 3 (6.3) 4 (8.5) 0.71

Thrombus aspiration 5 (10.4) 4 (8.5) 1.00

Glycoprotein IIb/IIIa inhibitor during PCI 33 (68.8) 32 (68.1) 0.94

Stenting 45 (93.8) 43 (91.5) 0.71

TIMI flow grade 3 after PCI 44 (91.7) 43 (91.5) 1.00

Oral medicines after PCI

Aspirin 48 (100) 47 (100) NS

Clopidogrel or ticagrelor 48 (100) 47 (100) NS

Statins 48 (100) 47 (100) NS

Beta-blockers 44 (91.7) 42 (89.4) 0.74

ACE inhibitors or ARBs 39 (81.3) 39 (83.0) 0.83

Calcium blockers 14 (29.2) 11 (23.4) 0.52

Diuretic agents 5 (10.4) 3 (6.4) 0.71

Values are mean � SD or n (%).

ACE ¼ angiotensin-converting enzyme; ARB¼ angiotensin receptor blocker; LAD ¼ left anteriordescending coronary artery; LCx ¼ left circumflex coronary artery; LL-TS ¼ low-level tragusstimulation; PCI ¼ percutaneous coronary intervention; RCA ¼ right coronary artery;TIMI ¼ Thrombolysis In Myocardial Infarction.

Yu et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7

Noninvasive Neuromodulation in Patients With STEMI A U G U S T 1 4 , 2 0 1 7 : 1 5 1 1 – 2 0

1514

ECHOCARDIOGRAPHY. All patients underwentconventional 2-dimensional echocardiography ondays 5 to 7 after revascularization. Echocardiographywas performed using a Philips iE33 equipped withS5-1 transducers (Philips Medical Systems, Andover,Massachusetts) or a GE Vivid 7 model equipped withan M4S transducer (GE Healthcare, Little Chalfont,United Kingdom). All measurements included at least3 consecutive beats for patients in sinus rhythm. Theechocardiographic data were interpreted by 2 expertswho were blinded to group information. Ventriculardiameters (normalized to body surface area) and

interventricular septal and posterior wall thicknesswere measured. The modified biplane Simpson rulewas applied to calculate the LV ejection fraction. The16-segment model was applied to evaluate theseverity of LV regional wall motion abnormality (18).The percentage that was obtained by dividing thenumber of akinetic and dyskinetic segments by thetotal number of segments evaluated was used toindicate the extent of LV damage. Early transmitralflow velocity (E), late atrial contraction (A) velocity,and the E/A ratio were used to reflect LV diastolicfunction.

STATISTICAL ANALYSIS. Preliminary measurementsin patients with STEMI at our hospital had an areaunder the curve (AUC) for creatine kinase-MB in thefirst 72 h of approximately 6,000 � 1,000 ng h/ml. Itwas calculated that to detect a 15% reduction in theAUC for creatine kinase-MB with 95% probability(type II error probability of 0.05), at an alpha level(type I error probability) of 0.05 using a 2-sided test,40 subjects in each group should be included.Continuous and categorical variables are presentedas mean � SD and as count (percentage), respec-tively. Independent-sample Student t tests or chi-square tests were used for comparisons of contin-uous and categorical variables, respectively, forbaseline characteristics between the 2 groups. TheVAs, the AUCs for creatine kinase-MB and myoglobinduring the 72 h after reperfusion, and the echocar-diographic characteristics were compared betweenthe 2 groups using independent-sample Studentt tests. Levels of NT-proBNP and inflammatorymarkers were compared between the groups using2-way analysis of variance followed by the Bonfer-roni post hoc test. SPSS version 18.0 (SPSS, Chicago,Illinois) was used for data analysis. The differenceswere considered statistically significant at a 2-sidedp value of <0.05.

RESULTS

BASELINE CHARACTERISTICS. As shown in Figure 1A,128 consecutive patients with STEMI were included,and 106 patients were randomized to the LL-TS group(n ¼ 53) or control group (n ¼ 53). The results of cor-onary angiography showed that 11 higher risk patientsfor PCI (e.g., left main or multiple coronary arterydisease) were excluded from this study, 6 patients inthe LL-TS group and 5 patients in the control group.Ultimately, 47 patients in the LL-TS group and48 patients in the control group were followed inthis study (Figure 1A). There were no differenceswith respect to baseline parameters in the 2 groups,

FIGURE 2 Tragus Stimulation

(A) The positioning of the electrode for stimulation of the right tragus. (B) Before stimulation, sinus cycle length is 783 ms. (C) During

stimulation at 5 mA, there is an increase in the sinus cycle length to 813 ms. (D) Low-level tragus stimulation (LL-TS) or sham LL-TS was

started once the patient arrived in the catheterization room and lasted for 2 h after balloon dilatation (reperfusion).

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7 Yu et al.A U G U S T 1 4 , 2 0 1 7 : 1 5 1 1 – 2 0 Noninvasive Neuromodulation in Patients With STEMI

1515

such as the sex ratio, ischemic time, culprit vessel,door-to-balloon time, stimulation threshold, andstimulation or sham stimulation duration (Table 1).

EFFECT OF LL-TS ON ISCHEMIA-REPERFUSION-RELATED

VAs. The results of Holter analysis showed that thetotal number of VBPs (154 � 115 vs. 551 � 214;p < 0.05), isolated VPBs (133 � 109 vs. 362 � 104;p < 0.05), coupled VPBs (3 � 3 vs. 22 � 16, p < 0.05),and ventricular tachycardia (2 � 3 vs. 18 � 12;p < 0.05) values in the LL-TS group were all signifi-cantly lower than the values in the control group(Figure 3).

EFFECTS OF LL-TS ON CREATINE KINASE-MB AND

MYOGLOBIN LEVELS. The AUCs for creatine kinase-MB and myoglobin during the 72 h after reperfusionwere calculated to reflect the myocardial infarctionsize (Figure 4). The AUC (0 to 72 h) for creatine kinase-MB level was significantly lower in the LL-TS groupthan in the control group (5,156 � 782 ng h/ml vs.7,646 � 742 ng h/ml; p < 0.05). In addition, the AUC (0to 72 h) for myoglobin level was also significantlylower in the LL-TS group (8,632 � 551 mg h/l vs. 10,361� 624 mg h/l; p < 0.05).

EFFECTS OF LL-TS ON NT-proBNP LEVELS. Theblood levels of NT-proBNP in both groups are shownin Figure 5. There was no significant difference inNT-proBNP level between the 2 groups at baseline(317 � 132 ng/l vs. 365 � 111 ng/l; p > 0.05). However,LL-TS significantly decreased the blood level ofNT-proBNP at the time points of 24 h (905� 213 ng/l vs.1,595� 432 ng/l; p<0.05) and 7 days (548� 201 ng/l vs.1,127 � 302 ng/l; p < 0.05) after reperfusion.

EFFECTS OF LL-TS ON INFLAMMATORY MARKER

LEVELS. The blood levels of interleukin-6, inter-leukin-1b, high-mobility group-box 1 protein 1, andtumor necrosis factor-a were evaluated at baselineand 24 h after reperfusion in both groups (Figure 6).There were no significant differences in theseinflammatory markers between the 2 groups at base-line. However, a significant improvement in inflam-mation was observed in the LL-TS group at the timepoint of 24 h after reperfusion.

ECHOCARDIOGRAPHIC CHARACTERISTICS. As shown inTable 2, there were no differences in LV end-diastolicvolume and LV end-systolic volume between the2 groups. However, a significant improvement in LV

FIGURE 3 Reperfusion-Related Ventricular Arrhythmias

The values of total ventricular premature beats (VPBs) (A), isolated VPBs (B), coupled VPBs (C-VPB) (C), and ventricular tachycardia (D) were

all significantly lower in the low-level tragus stimulation (LL-TS) group than the values in the control group. *p < 0.05 versus control group.

Yu et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7

Noninvasive Neuromodulation in Patients With STEMI A U G U S T 1 4 , 2 0 1 7 : 1 5 1 1 – 2 0

1516

ejection fraction was observed in the LL-TS groupcompared with the control group. The wall motionindex, which was calculated using the 16-segmentmodel, was used to evaluate the extent of LVdysfunction. In agreement with the LV ejection frac-tion, the percentage of LV akinetic and dyskineticsegments was significantly lower in the LL-TS group.

DISCUSSION

MAJOR FINDINGS. To the best of our knowledge, thisstudy provides the first clinical evidence that LL-TSmarkedly reduces MIRI as reflected by ischemia-reperfusion-associated VAs, myocardial injury bio-markers, and cardiac function in patients with STEMIundergoing primary PCI. This study also showed thatthe MIRI reduction is accompanied by an improve-ment in blood inflammatory cytokine levels. Thisproof-of-concept study raises the possibility that thisnoninvasive strategy may be used to treat patientswith STEMI undergoing primary PCI.

LL-TS FOR THE TREATMENT OF MIRI. Considerableevidence has shown that cVNS with a stimulus

strength that was sufficient to reduce heart rate by20% to 50% is capable of relieving MIRI in differentanimal species (8,9). However, a reduction in heartrate via cVNS of >20% is not tolerated by mostpatients, which limits its clinical application. Tocircumvent this problem, we recently found thatlow-level cVNS that did not cause heart ratereduction significantly decreased the incidence ofVAs and reduced myocardial infarct size in a caninemodel of acute myocardial ischemia and reperfusion(11). Similar conclusions were reached by Shinlapa-wittayatorn et al. (12) and by Zhang et al. (13), whoalso showed that low-level cVNS could attenuateMIRI and improve ventricular function during theischemia-reperfusion process. During the past fewyears, stimulation of the auricular branch of thevagus nerve has been developed to overcome thepotential barriers of conventional cVNS (19).Anatomic studies suggest that the auricular branchof the vagus nerve is the only peripheral branch ofthe vagus nerve distribution on the surface of theear (20). Stimulation of this region on the ear re-duces seizure frequency in patients with intractableepilepsy, similar to cVNS (21). Preclinical and

FIGURE 4 Myocardial Injury Biomarkers

(A, B) Area under the curve (AUC) for creatine kinase-MB (CK-MB) (A) and myoglobin (MYO) (B) over the initial 72 h post–percutaneous

coronary intervention. There was a significant decrease in the AUC (0 to 72 h) for both CK-MB (C) and MYO (D) in the low-level tragus

stimulation (LL-TS) group compared with control group. *p < 0.05 versus control group.

FIGURE 5 Blood Levels of N-Terminal Pro–B-Type

Natriuretic Peptide

There was no significant difference in blood levels of

N-terminal pro–B-type natriuretic peptide (NT-proBNP)

between the 2 groups at baseline (BS). However, the blood

level of NT-proBNP was significantly lower in the low-level

tragus stimulation (LL-TS) group at 24 h and 7 days post–

percutaneous coronary intervention. *p < 0.05 versus control

group at the same time point.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7 Yu et al.A U G U S T 1 4 , 2 0 1 7 : 1 5 1 1 – 2 0 Noninvasive Neuromodulation in Patients With STEMI

1517

clinical studies have also confirmed that LL-TS is anoninvasive alternative to low-level cVNS to treatatrial fibrillation (17,22). Our recent studies showedthat LL-TS could attenuate cardiac structural andautonomic remodeling, improve ventricular func-tion, and decrease VA induction in conscious dogswith healed myocardial infarction (15,16). In thepresent study, we applied LL-TS in patients withSTEMI to evaluate whether LL-TS could attenuateMIRI in these patients. MIRI was evaluated via 2different methods: myocardial injury biomarkers(creatine kinase-MB and myoglobin) andreperfusion-related VAs. The results of both of themethodologies provided evidence of a significantreduction in MIRI that resulted from LL-TS. Inaddition, LL-TS also significantly decreased theblood level of NT-proBNP and the wall motionindex but increased the LV ejection fraction,suggesting a significant improvement in the recov-ery of cardiac function after reperfusion. Takentogether, our data suggest that LL-TS may be anovel, invasive therapy for the treatment ofpatients with STEMI.

FIGURE 6 Blood Levels of Inflammatory Cytokines

There were no significant differences in blood levels of interleukin (IL)-6 (A), IL-1b (B), high-mobility group-box 1 protein (HMGB1) (C), and

tumor necrosis factor (TNF)–a (D) between the 2 groups at baseline. However, the blood levels of these inflammatory cytokines were all

significantly decreased in the low-level tragus stimulation (LL-TS) group at 24 h. *p < 0.05 versus control group at the same time point.

TABLE 2

LV end-di

LV end-sy

LV ejectio

Wall moti

Mitral infl

Mitral infl

Ventricula

Posterior

Mitral reg

Values are m

LL-TS ¼ l

Yu et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7

Noninvasive Neuromodulation in Patients With STEMI A U G U S T 1 4 , 2 0 1 7 : 1 5 1 1 – 2 0

1518

POSSIBLE MECHANISMS UNDERLYING THE BENEFICIAL

EFFECTS OF LL-TS. It is well established that inflam-mation is critically involved in the pathogenesis ofSTEMI. After the initial ischemic event, an intenseinflammatory response is triggered by myocardialischemia, characterized mainly by leukocyte infiltra-tion and proinflammatory cytokine production,resulting in cardiomyocyte damage and affectingventricular remodeling (23). The clinical studies have

Echocardiographic Characteristics

Control LL-TS p Value

astolic volume (ml) 92.4 � 22.9 95.3 � 24.1 0.54

stolic volume (ml) 48.0 � 16.5 45.6 � 14.4 0.45

n fraction (%) 49 � 8 52 � 6 0.01

on index 1.4 � 0.3 1.2 � 0.2 <0.01

ow E wave (cm/s) 71.1 � 15.4 69.7 � 14.9 0.65

ow E/A ratio 1.1 � 0.3 1.0 � 0.3 0.11

r septal thickness (mm) 10.0 � 1.2 10.2 � 0.9 0.36

LV wall thickness (mm) 9.8 � 0.9 10.0 � 0.4 0.17

urgitation (any) 30 (62.5) 32 (68.1) 0.57

ean � SD or n (%).

ow-level tragus stimulation; LV ¼ left ventricular.

shown that the levels of inflammation markers canpredict ventricular remodeling, cardiac dysfunction,and death in patients with STEMI undergoing pri-mary angioplasty (24). In view of this evidence, thesignificant inverse correlation of LL-TS treatmentwith the rise in post-reperfusion inflammatory cyto-kines suggests a potential mechanism of the observedbeneficial effect of LL-TS. Recent studies also haveshown that LL-TS could decrease the levels of in-flammatory cytokines in post–myocardial infarctioncanine models (16) and in patients with paroxysmalatrial fibrillation (15). Given that the functionalthreshold for activation of vagal afferent fibers islower than that for activation of efferent fibers (25),LL-TS used in the present study probably mainlyactivated the afferent fibers, and therefore mightachieve its anti-inflammatory effects through activa-tion of the hypothalamic-pituitary-adrenal axispathway (26).

Reperfusion can result in the generation andrelease of reactive oxygen species, which inducesmyocardial oxidative stress and ultimately exacer-bates myocardial cell apoptosis. A prior experimentalstudy from our group showed that low-level cVNS

PERSPECTIVES

WHAT IS KNOWN? Our previous studies suggested that LL-TS

could reduce the size of myocardial injury induced by ischemia.

WHAT IS NEW? In this proof-of-concept study, we have

demonstrated for the first time that LL-TS markedly reduces

myocardial ischemia and reperfusion injury in patients with

STEMI undergoing primary PCI, indicating that this novel neu-

romodulatory approach may be an important step toward a

noninvasive and nonpharmacological therapy for the treatment

of STEMI.

WHAT IS NEXT? Further adequately powered prospective

multicenter clinical trials are warranted.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7 Yu et al.A U G U S T 1 4 , 2 0 1 7 : 1 5 1 1 – 2 0 Noninvasive Neuromodulation in Patients With STEMI

1519

could attenuate MIRI by suppressing oxidative stressactivity and regulating myocardial apoptosis (11). Inaddition, sympathetic activation has been shown toincrease the susceptibility of the heart to ischemia- orreperfusion-related VAs (27). Interventions thatdecrease sympathetic activity are associated with alower incidence of VAs in both patients and animalmodels (27). Low-level cVNS has been shown tosuppress the activity of the left stellate ganglion,the gateway of sympathetic innervation to theheart, in ambulatory dogs (28). Taken together, theseobservations indicate that antioxidative stress, anti-apoptosis, and antisympathetic effects produced bylow-level cVNS may all contribute to its beneficialeffects on MIRI. In the present study, similar mech-anisms are likely to underlie the beneficial effects ofLL-TS.

CLINICAL IMPLICATIONS. Over the past fewdecades, several advances have been made in thetreatment of patients with STEMI; however, there iscurrently no specific treatment that decreases MIRI,which inevitably occurs as a result of the restorationof coronary patency. The present study indicates thatLL-TS is well tolerated and significantly improvesinflammation responses, reperfusion-related VAs,blood levels of myocardial injury biomarkers andNT-proBNP, and LV ejection fraction and wall motionindex in patients with STEMI. LL-TS may become anovel, noninvasive, and nonpharmacological therapyfor the treatment of patients with STEMI undergoingprimary PCI if these observations are confirmed inlarger randomized trials.

STUDY LIMITATIONS. First, cardiac magnetic reso-nance imaging was not performed to evaluate infarctsize and cardiac function during the first 7 days offollow-up. However, echocardiographic methods forthe determination of cardiac function have also beenwidely used (29,30).

Second, we evaluated only the acute (7-day) effectsof LL-TS in patients with STEMI. Further studies arenecessary to identify whether LL-TS could improvethe long-term clinical outcome in these patients.

Third, higher risk patients with left main ormultiple coronary artery disease were excluded from

this study to avoid introducing confounding factorssuch as complications caused by high-risk PCI.Therefore, it is unclear whether LL-TS would havesimilar benefits in these patients.

Finally, the parameters of LL-TS, such as fre-quency, intensity, and duration, in the present studywere based on previous experience with basic andclinical findings, and the optimal parameters of LL-TSwere not determined.

CONCLUSIONS

LL-TS significantly improves inflammatory responses,reperfusion-related VAs, blood levels of myocardialinjury biomarkers and NT-proBNP, and LV ejectionfraction andwall motion index in patients with STEMI.This proof-of-concept study raises the possibility thatthis noninvasive therapy can be used to treat patientswith STEMI who are undergoing primary PCI. Futureadequately powered prospective multicenter clinicaltrials are warranted.

ADDRESS FOR CORRESPONDENCE: Dr. Hong Jiang,Department of Cardiology, Renmin Hospital ofWuhan University, No. 238 Jiefang Road, WuchangDistrict, Wuhan City, Hubei Province 430060, China.E-mail: hong-jiang@whu.edu.cn.

RE F E RENCE S

1. Steg PG, James SK, Atar D, et al. ESC guidelinesfor the management of acute myocardial infarc-tion in patients presenting with ST-segmentelevation. Eur Heart J 2012;33:2569–619.

2. Reed GW, Rossi JE, Cannon CP. Acute myocar-dial infarction. Lancet 2016;389:197–210.

3. Heusch G, Rassaf T. Time to give up oncardioprotection? A critical appraisal of clinicalstudies on ischemic pre-, post-, and remoteconditioning. Circ Res 2016;119:676–95.

4. Yellon DM, Hausenloy DJ. Myocardial reperfu-sion injury. N Engl J Med 2007;357:1121–35.

5. Heusch G, Gersh BJ. The pathophysiology ofacute myocardial infarction and strategies of pro-tection beyond reperfusion: a continual challenge.Eur Heart J 2017;38:774–84.

6. Ibanez B, Heusch G, Ovize M, Van deWerf F. Evolving therapies for myocardial

Yu et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 0 , N O . 1 5 , 2 0 1 7

Noninvasive Neuromodulation in Patients With STEMI A U G U S T 1 4 , 2 0 1 7 : 1 5 1 1 – 2 0

1520

ischemia/reperfusion injury. J Am Coll Cardiol2015;65:1454–71.

7. Ibanez B, Macaya C, Sanchez-Brunete V, et al.Effect of early metoprolol on infarct size inST-segment-elevation myocardial infarctionpatients undergoing primary percutaneous coro-nary intervention: the Effect of Metoprolol inCardioprotection During an Acute MyocardialInfarction (METOCARD-CNIC) trial. Circulation2013;128:1495–503.

8. Zhao M, He X, Bi XY, Yu XJ, Gil WW, Zang WJ.Vagal stimulation triggers peripheral vascularprotection through the cholinergic anti-inflammatory pathway in a rat model of myocar-dial ischemia/reperfusion. Basic Res Cardiol 2013;108:345.

9. Katare RG, Ando M, Kakinuma Y, et al. Vagalnerve stimulation prevents reperfusion injurythrough inhibition of opening of mitochondrialpermeability transition pore independent of thebradycardiac effect. J Thorac Cardiovasc Surg2009;137:223–31.

10. Uitterdijk A, Yetgin T, Te LHM, et al. Vagalnerve stimulation started just prior to reperfusionlimits infarct size and no-reflow. Basic Res Cardiol2015;110:508.

11. Chen M, Zhou X, Yu L, et al. Low-level vagusnerve stimulation attenuates myocardial ischemicreperfusion injury by antioxidative stress andantiapoptosis reactions in canines. J CardiovascElectrophysiol 2016;27:224–31.

12. Shinlapawittayatorn K, Chinda K, Palee S,et al. Low-amplitude, left vagus nerve stimula-tion significantly attenuates ventricular dysfunc-tion and infarct size through prevention ofmitochondrial dysfunction during acute ischemia-reperfusion injury. Heart Rhythm 2013;10:1700–7.

13. Zhang L, Lu Y, Sun J, Zhou X, Tang B.Subthreshold vagal stimulation suppresses ven-tricular arrhythmia and inflammatory response in a

canine model of acute cardiac ischaemia andreperfusion. Exp Physiol 2016;101:41–9.

14. Fallgatter AJ, Neuhauser B, Herrmann MJ,et al. Far field potentials from the brain stem aftertranscutaneous vagus nerve stimulation. J NeuralTransm (Vienna) 2003;110:1437–43.

15. Yu L, Wang S, Zhou X, et al. Chronic intermit-tent low-level stimulation of tragus reducescardiac autonomic remodeling and ventriculararrhythmia inducibility in a post-infarction caninemodel. J Am Coll Cardiol EP 2016;2:330–9.

16. Wang Z, Yu L, Wang S, et al. Chronicintermittent 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.

17. Stavrakis S, Humphrey MB, Scherlag BJ, et al.Low-level transcutaneous electrical vagus nervestimulation suppresses atrial fibrillation. J Am CollCardiol 2015;65:867–75.

18. Lang RM, Badano LP, Mor-Avi V, et al. Rec-ommendations for cardiac chamber quantificationby echocardiography in adults: an update from theAmerican Society of Echocardiography and theEuropean Association of Cardiovascular Imaging.J Am Soc Echocardiogr 2015;28:1–39.

19. Howland RH. New developments with vagusnerve stimulation therapy. J Psychosoc Nurs MentHealth Serv 2014;52:11–4.

20. Peuker ET, Filler TJ. The nerve supply of thehuman auricle. Clin Anat 2002;15:35–7.

21. He W, Jing X, Wang X, et al. Transcutaneousauricular vagus nerve stimulation as a comple-mentary therapy for pediatric epilepsy: a pilottrial. Epilepsy Behav 2013;28:343–6.

22. 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.

23. Seropian IM, Toldo S, Van Tassell BW,Abbate A. Anti-inflammatory strategies for ven-tricular remodeling following ST-segment eleva-tion acute myocardial infarction. J Am Coll Cardiol2014;63:1593–603.

24. van Diepen S, Roe MT, Lopes RD, et al. Base-line NT-proBNP and biomarkers of inflammationand necrosis in patients with ST-segment eleva-tion myocardial infarction: insights from theAPEX-AMI trial. J Thromb Thrombolysis 2012;34:106–13.

25. Ardell JL, Rajendran PS, Nier HA,KenKnight BH, Armour JA. Central-peripheralneural network interactions evoked by vagusnerve stimulation: functional consequences oncontrol of cardiac function. Am J Physiol Heart CircPhysiol 2015;309:H1740–52.

26. Bonaz B, Sinniger V, Pellissier S. Anti-inflam-matory properties of the vagus nerve: potentialtherapeutic implications of vagus nerve stimula-tion. J Physiol 2016;594:5781–90.

27. Shen MJ, Zipes DP. Role of the autonomicnervous system in modulating cardiac arrhythmias.Circ Res 2014;114:1004–21.

28. Shen MJ, Shinohara T, Park HW, et al.Continuous low-level vagus nerve stimulation re-duces stellate ganglion nerve activity and parox-ysmal atrial tachyarrhythmias in ambulatorycanines. Circulation 2011;123:2204–12.

29. Cung TT, Morel O, Cayla G, et al. Cyclosporinebefore PCI in patients with acute myocardialinfarction. N Engl J Med 2015;373:1021–31.

30. Lunde K, Solheim S, Aakhus S, et al. Intra-coronary injection of mononuclear bone marrowcells in acute myocardial infarction. N Engl J Med2006;355:1199–209.

KEY WORDS acute myocardial infarction,inflammation, ischemia-reperfusion injury,low-level vagal stimulation, ventriculararrhythmia