J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y VO L . 3 , N O . 1 , 2 0 1 7

ª 2 0 1 7 B Y T H E AM 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

P U B L I S H E D B Y E L S E V I E R

I S S N 2 4 0 5 - 5 0 0 X / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c e p . 2 0 1 6 . 0 4 . 0 0 4

Clinical Characteristics and Long-TermPrognosis of Senior Patients WithBrugada Syndrome

Takeshi Kitamura, MD,a Seiji Fukamizu, MD,a Iwanari Kawamura, MD,a Rintaro Hojo, MD,a Yuya Aoyama, MD, PHD,a

Mitsuhiro Nishizaki, MD, PHD,b Masayasu Hiraoka, MD, PHD,c Harumizu Sakurada, MD, PHDd

ABSTRACT

Fro

Mi

Me

ha

All

ins

vis

Ma

OBJECTIVES This study investigated clinical characteristics and prognosis of Brugada syndrome (BrS) in patients older

than 60 years of age during a long-term follow-up period.

BACKGROUND Clinical characteristics and prognosis of senior patients with BrS have not been clearly elucidated.

METHODS A total of 181 patients with BrS were divided into 2 groups by age at the time of diagnosis: the younger group

was <60 years of age (n ¼ 123), and the senior group was $60 years of age (n ¼ 58).

RESULTS Mean ages were 42.7 � 11 years and 68.6 � 7.1 years, respectively. Prevalence of spontaneous type 1

electrocardiogram (ECG) was lower in the senior group (22 of 58; 37.9%) than in the younger group (64 of 123; 51.9%)

(p ¼ 0.027). Among various ECG parameters, the senior group had a lower incidence of prolonged r-J intervals in

V2 $90 ms than the younger group (34 of 58; 58.6% vs. 90 of 123; 73.1%, p ¼ 0.049) and day-to-day variation of

Brugada ECG patterns (3 of 58; 5.2% vs. 23 of 123; 18.7%, p ¼ 0.032). During a mean follow-up period of 7.6 � 5.8 years,

no senior patients experienced documented fatal ventricular arrhythmias, but 11 younger patients did. Kaplan-Meier

analysis revealed a better prognosis in the senior group than in the younger group (log-rank, p ¼ 0.011).

CONCLUSIONS Senior BrS patients, $60 years of age, had a better prognosis than those <60 years of age.

Implantable cardioverter-defibrillator insertion for senior patients with BrS needs careful consideration.

(J Am Coll Cardiol EP 2017;3:57–67) © 2017 by the American College of Cardiology Foundation.

B rugada syndrome (BrS) is characterized byunique electrocardiogram (ECG) patterns onthe right precordial leads (V1 to V3) and

increased risk of sudden cardiac death (SCD) due toventricular fibrillation (VF) in the absence of majorstructural heart disease (1). Although various riskfactors for future development of VF have been pro-posed by large numbers of reports, no consensus hasbeen reached to predict fatal cardiac events, especiallyin BrS patients without a history of documented VF oraborted SCD. The only available therapeutic option isinsertion of an implantable cardioverter-defibrillator

m the aDepartment of Cardiology, Tokyo Metropolitan Hiroo Hospital, T

nami Kyosai Hospital, Yokohama, Japan; cTokyo Medical and Dental U

tropolitan Health and Medical Treatment Corporation, Ohkubo Hospital,

ve no relationships relevant to the contents of this paper to disclose.

authors attest they are in compliance with human studies committe

titutions and Food and Drug Administration guidelines, including patien

it the JACC: Clinical Electrophysiology author instructions page.

nuscript received January 8, 2016; revised manuscript received March 31

(ICD), but the role and indication for ICD implantationremain controversial, particularly in the elderly popu-lation. The mean age of patients with BrS has been re-ported to be in the fourth to fifth decade (2–4). Brugadatype 1 ECGs are observed less frequently in elderly pa-tients than in the younger patients (4). Recently, 2studies have indicated a benign prognosis of elderlyBrS patients (5,6). Those studies, however, do notclarify differences in clinical and ECG characteristicsunderlying prognostic factors between the seniorand younger patients. Moreover, the indication forICDs in senior patients, according to the latest

okyo, Japan; bDepartment of Cardiology, Yokohama

niversity, Yushima, Tokyo, Japan; and the dTokyo

Tokyo, Japan. The authors have reported that they

es and animal welfare regulations of the authors’

t consent where appropriate. For more information,

, 2016, accepted April 7, 2016.

FIGUR

Each c

shown

SEE PAGE 68

ABBR EV I A T I ON S

AND ACRONYMS

BrS = Brugada syndrome

ECG = electrocardiogram

ICD = implantable

cardioverter-defibrillator

SAECG = signal averaged

electrocardiogram

SCD = sudden cardiac death

VF = ventricular fibrillation

VT = ventricular tachycardia

Kitamura et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7

Prognosis and Features of Senior Brugada Patients J A N U A R Y 2 0 1 7 : 5 7 – 6 7

58

consensus statement, has not been verified.Because the United Nations classificationdefined people at $60 years of age as senior,we classified BrS patients with diagnosesat $60 years of age as the senior group andthose <60 years of age as the younger groupin our study and investigated their clinicalcharacteristics and prognostic variables. Inaddition, we reclassified all patients in classesof ICD indication according to the latestconsensus statement and then evaluated thedistribution and incidence of fatal ventricular

arrhythmia during a long-term follow-up period.

METHODS

STUDY POPULATION. The study included 181consecutive patients whose BrS was diagnosed andfollowed at the Tokyo Metropolitan Hiroo Hospitalfrom 1992 to 2014. Diagnosis of BrS was defined by the2 consensus reports in 2002 (7) and 2005 (8). Thisretrospective observational study protocol wasapproved by the Institutional Review Board of TokyoMetropolitan Hiroo Hospital, Tokyo, Japan. Patientswere divided into 2 groups according to their age atthe time of diagnosis: the younger group (<60 years ofage) and the senior group ($60 years of age). Clinicalhistories and 12-lead ECG findings, including those

E 1 Distribution of Patients’ Ages at Diagnosis of BrS

olumn is divided by 10 years, and numbers of patients are

on the top of each column. BrS ¼ Brugada syndrome.

from leads V1 to V3 placed at the 2nd, 3rd, or 4thintercostal space were accessed in all patients.

Organic heart diseases were excluded by exami-nations using ultrasound cardiography, coronaryangiography, right and left ventriculography, andcardiac magnetic resonance imaging. Acetylcholineprovocation test was performed to exclude vaso-spastic angina in 74 patients. ECGs for BrS wereclassified according to the 2002 and 2005 consensusreports (7,8), with or without provocation test bysodium-channel blockers. Patients with syncope ofunknown cause, family history of SCD or atrialfibrillation (AF) (to avoid a possible risk of unex-pected lethal ventricular arrhythmia to be provokedby sodium channel blockers) were screened by so-dium channel provocation test if their ECG revealedtype 2 or type 3 Brugada ECG. All baseline and drug-induced 12-lead ECG records were obtained at a pa-per speed of 25 mm/s and with amplitude of 10 mm/mV with the right precordial leads positioned at the2nd, 3rd, and 4th intercostal spaces. All ECGs wereanalyzed by 3 independent experienced electrophys-iologists. Early repolarization pattern was defined inthe presence of J-point elevation $1 mm in $2contiguous inferior and lateral leads of ECG, accord-ing to the 2013 consensus statement (9). FragmentedQRS was defined in the presence of abnormal frag-mentation within the QRS complex as 4 spikes in 1 or8 spikes in all of leads V1, V2, and V3 (10). Day-to-dayvariation in Brugada ECGs was positive if a type 1Brugada ECG was present on one day but spontane-ously disappeared or changed to type 2 or type 3 ECGon another day during the follow-up period. We ac-quired ECGs during the initial follow-up and at eachscheduled follow-up, and any unscheduled visits andduring any in-hospital stays. The alterations in theECG were evaluated at rest (commonly 2 h before orafter meal) and excluded the ECGs recorded with anystress (during exercise test, drug challenge test, fullstomach, and in febrile illness). Significant augmen-tation of ST-segment elevation during recoveryphase in treadmill exercise testing was defined asST-segment amplitude increase >0.05 mV in at least1 of leads V1 to V3 at early recovery (1 to 4 minat recovery) compared with the baseline level (pre-exercise) (11). The presence of late potentials (LPs)was evaluated with a signal-averaged ECG (noiselevel: 0.3 V, filtered with a high-pass filter by 40 Hz).Three parameters were assessed using a computeralgorithm: the filtered QRS duration (f-QRS); the root-mean-square voltage of the terminal 40 ms in thefiltered QRS complex (RMS40); and the duration of

TABLE 1 Comparison of Clinical Characteristic and Types of ST Elevation in the Younger

and Senior Groups

Younger Subjects(n ¼ 123)

Senior Subjects(n ¼ 58) p Value

Age, yrs 42.7 � 11 68.6 � 7.1

Males 113 (91.9%) 56 (96.5%) 0.237

Family history of SCD 26 (21.1%) 15 (25.7%) 0.479

Family history of type 1 ECG 5 (4.1%) 0 (0%) 0.141

Syncope 28 (22.8%) 14 (24.1%) 0.838

SCN5A mutation 2/6 0/4 -

Documented VF 19 (15.4%) 5 (8.6%) 0.206

Vasospastic angina 6/50 (12.0%) 3/21 (14.2%) 0.117

ICD implantation 47 (38.2%) 26 (44.8%) 0.397

Inappropriate therapy 12/47 (25.5%) 6/26 (23.1%) 0.854

Spontaneous type 1 ECG 64 (51.9%) 22 (37.9%) 0.027

Drug-induced type 1 ECG 59 (48.1%) 36 (62.1%) 0.027

Reasons for drug provocation test

Documented VF 7/59 (11.9%) 3/36 (8.3%) 0.842

Syncope of unknown cause withtype 2 or type 3 ECG

17/59 (28.8%) 9/36 (25.0%) 0.867

Family history of SCD withtype 2 or type 3 ECG

18/59 (30.5%) 9/36 (25.0%) 0.732

AF with type 2 or type 3 ECG 17/59 (28.8%) 15/36 (41.7%) 0.198

Newly developed IHD 3 (2.4%) 3 (5.2%) 0.292

Coronary angiography during follow-up 3 (2.4%) 7 (12.1%) 0.013

Values are mean � SD or n (%).

AF ¼ atrial fibrillation; ECG ¼ electrocardiogram; ICD ¼ implantable cardioverter-defibrillator; IHD ¼ ischemicheart disease; SCD ¼ sudden cardiac death; VF ¼ ventricular fibrillation.

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7 Kitamura et al.J A N U A R Y 2 0 1 7 : 5 7 – 6 7 Prognosis and Features of Senior Brugada Patients

59

low-amplitude signals, 40 mV in the terminal filteredQRS complex (LAS40). LPs were considered positivewhen 2 of 3 criteria (f-QRS >114 ms; RMS40 <20 mV;and LAS40 >38 ms) were met (12,13). Electrophysio-logical study (EPS) findings were evaluated among115 of 181 patients for diagnosis or risk stratificationor both. Those patients underwent programmedelectrical stimulation to assess ventricular tachy-cardia (VT) or VF inducibility. Our method for EPSand protocol of ventricular stimulation for inductionof VT/VF was described previously (14). In short, EPSwas performed using 3 multielectrode cathetersintroduced percutaneously through the femoral ves-sels. Programmed ventricular stimulation was per-formed with the use of a maximum of 3 ventricularextra-stimuli from the right ventricle apex andoutflow tract. Minimum coupling interval was effec-tive refractory period during single ventricular extra-stimulus, 180 ms during 2 ventricular extra-stimuli,200 ms during 3 ventricular extra-stimuli. Patientswith VF lasting for more than 30 s or who requiredelectrical cardioversion were classified as inducible.Informed consent was obtained from all patients.Indication for ICD implantation was determinedaccording to the 2002 or 2008 American Collegeof Cardiology/American Heart Association/HeartRhythm Society (ACC/AHA/HRS) guideline (15), theJapanese guideline (16), or 2013 HRS/European HeartRhythm Association/Asia Pacific Heart Rhythm Soci-ety (HRS/EHRA/APHRS) consensus statement (9).Because our study period spanned 1992 to 2014 andthe guidelines for ICD therapy were modified severaltimes during the entire period, we reclassified allpatients by the indication for ICD implantationaccording to the latest consensus statement of the2013 HRS/EHRA/APHRS (9). Fourteen patients wereregistered in the Japan idiopathic ventricular fibril-lation study (17).

FOLLOW-UP. Patients with ICD implantation under-went regular follow-up of the device and clinicalsymptoms, at least, every 3 to 4 months at ouroutpatient clinic. Patients without ICD implantationwere followed every 6 or 12 months with a visit to ouroutpatient clinic for checking clinical status and ex-aminations of resting ECG and signal averaged ECG(SAECG).

STATISTICS. Data are mean � SD or absolute valuesand percentages where appropriate. The chi-squaretest and Fisher exact test were used to compare cat-egorical variables. Continuous variables between the2 groups were analyzed using the unpaired Student ttest or Mann-Whitney U test as appropriate. Survivalcurves were constructed by using the Kaplan-Meier

method and compared using the log-rank test. Uni-variate Cox proportional hazards models were usedto assess the effect of each variable on VF duringfollow-up. A p value <0.05 was considered sta-tistically significant. Statistical analyses were con-ducted using SPSS version 19.0 software (SPSS Inc.,Chicago, Illinois).

RESULTS

CLINICALANDELECTROPHYSIOLOGICALCHARACTERISTICS

OF THE 2 GROUPS. The senior group, $60 years of age,consisted of 58 cases (32%), and the youngergroup, <60, was 123 cases (68%). Mean ages were 68.6� 7.1 years in the former and 42.7 � 11 years in thelatter group. The age distribution at diagnosis in allpatients is shown in Figure 1. The youngest patientwas 16 years of age, and the oldest was 91 years ofage. In the senior group, there were 35 patients60 years of age, 19 patients in their 70s, and 4 patientsin their 80s. In the younger group, there were38 patients in their 50s, 39 patients in their 40s (thehighest number of cases), 29 patients in their 30s, and17 patients less than 29 years of age.

Clinical characteristics of the 2 groups are shown inTable 1. The proportion of patients with spontaneoustype 1 ECG was lower, and that of drug-induced type 1

TABLE 2 Comparison of ECG Findings and Electrophysiological Parameters Between

Younger and Senior Groups

Younger Group(n ¼ 123)

Senior Group(n ¼ 58) p Value

ECG

P-wave duration, ms 107.2 � 16.0 103.1 � 14.3 0.096

PR interval, ms 171.3 � 28.7 172.4 � 24.6 0.794

QRS duration, ms 113.5 � 16.2 110.9 � 17.5 0.317

QTc interval, ms 413.3 � 27.6 416.5 � 25.7 0.455

r-J interval in V2 $90 ms 90 (73.1%) 34 (58.6 %) 0.049

r-J interval in V2, ms 100.0 � 14.5 96.9 � 18.1 0.212

Fragmented QRS 19 (15.4%) 6 (10.3%) 0.353

Inferior ERP 14 (11.4%) 11 (19.0%) 0.168

Lateral ERP 9 (7.3%) 3 (5.2%) 0.484

Infero-lateral ERP 4 (3.3%) 1 (1.7%) 0.481

Day-to-day variation 23 (18.7%) 3 (5.2%) 0.032

Documented atrial fibrillation 27 (21.9%) 16 (27.7%) 0.456

Positive LP by SAECG 90 (73.1%) 35 (60.3%) 0.082

Augmented ST-segment elevationduring recovery after exercise

3/111 (2.7%) 1/52 (1.9%) 0.542

EPS

Induced VF 56/74 (75.7%) 35/41 (85.4%) 0.057

VF induction by single or doubleextra-stimuli

37/74 (50.0%) 22/41 (53.7%) 0.943

VF induction by triple extra-stimuli 37/74 (50.0%) 19/41 (46.3%) 0.943

HV interval, ms 47.0 � 10.2 43.9 � 7.7 0.109

RVA ERP, ms 237.2 � 20.0 216.0 � 25.1 0.133

RVOT ERP, ms 240.0 � 20.6 245.3 � 18.7 0.349

Values are mean � SD, n (%), or n/N (%).

EPS¼ electrophysiological studies; ERP¼ effective refractory period; LP¼ late potential; RVA¼ right ventricularapex; RVOT¼ right ventricular outflow tract; SAECG¼ signal averaged electrocardiogram; TWA¼ T-wave alternans.

Kitamura et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7

Prognosis and Features of Senior Brugada Patients J A N U A R Y 2 0 1 7 : 5 7 – 6 7

60

was higher in the senior group than in the youngergroup. Other clinical parameters, except for theirages, were not significantly different between the 2groups. Male predominance was similar in the 2groups. Rates of patients with history of syncope,documented VF, and vasospastic angina were notdifferent. There were 5 patients in the younger groupwho had a family history of BrS, but all patients in thesenior group were probands without family history.No differences were noted between the 2 groups withregard to incidence of ICD implantation and inap-propriate ICD therapy. During follow-up, coronaryangiography was required to identify the cause ofchest pain in 10 patients. Six patients developed denovo coronary stenosis, which was treated by percu-taneous coronary intervention (mean: 60 � 11 years ofage; range: 42 to 74 years). The mean duration fromdiagnosis of BrS to diagnosis of newly developedcoronary stenosis was 10.5 � 6.2 years (range: 3.8 to16.5 years). Three patients underwent coronaryangiography to determine the cause of VF andappropriate therapy; however, there was no case ofnewly developed coronary stenosis. Other patientswith recurrence of VF during follow-up were also

screened for development of ischemic heart diseaseby using noninvasive examinations. No positivefindings of ischemic events by exercise test ormyocardial perfusion scintigraphy were detectedamong other patients.

Parameters of various ECG findings and electro-physiological studies of 2 groups are shown in Table 2.The number of ECGs recorded per patient per year was4.2 � 2.8/patient/year. P-wave duration, PQ interval,QRS duration, and QTc intervals were not differentamong them. Patients showing r-J interval $90 msand day-to-day variation of type 1 ECG had signifi-cantly higher ratios in the younger group than thosein the senior group. However, there were no differ-ences in percentages of patients with fragmentedQRS, inferolateral ER pattern, AF, positive LP bySAECG, and positive T-wave alternans in the 2 groups.

LONG-TERM FOLLOW-UP. After a mean follow-upperiod of 7.6 � 5.8 years (91 � 69 months), 11 pa-tients developed VT/VF episodes (cardiac events),and all of the events belonged to the younger groupbut none to the senior group. Figure 2 shows agedistributions of VF events at or before the age ofdiagnosis (VF history) (Figure 2A) and of VT/VFrecurrences during follow-up period (Figure 2B).Details of senior patients with VF events at diagnosisare shown in Table 3. The patient’s age at VF historyranged between 19 and 68 years of age, and the rangeof age at VT/VF recurrences during follow-up periodwas between 19 and 49 years of age. No patients over50 years of age had recurrences, despite history of VFin 7 patients. Kaplan-Meier analysis revealed a betterprognosis for the senior group than the youngergroup (log-rank p ¼ 0.011) (Figure 3). Univariate Coxregression analysis of each risk factor is shown inTable 4. Patients younger than 60 years of age(p ¼ 0.042; hazard ratio [HR]: 2.82; 95% confidenceinterval [CI]: 1.695 to 50.74), documented VF(p < 0.01; HR: 22.1; 95% CI: 10.22 to 54.33), ER pattern(p < 0.01; HR: 17.43; 95% CI: 4.597 to 66.1), frag-mented QRS (p < 0.01; HR: 8.574; 95% CI: 2.614to 28.12), and day-to-day variations of type 1 ECG(p < 0.01; HR: 9.617; 95% CI: 2.43 to 38.06) had acorrelation with VF recurrence. In addition, we alsoevaluated combinations of risk factors for predictionof VF recurrences (Table 4). Although several riskfactors could not singly predict VF events, combina-tions of age <60 years and risk factors (syncope,spontaneous type 1 ECG, induced VF, VF induction bysingle or double extra-stimuli) might predict a VFevent. Five combinations with highest HR amongall combinations are described at the bottomof Table 4. In addition, we evaluated alteration,

FIGURE 2 VF History and VF Events

(A) Presence and absence of VF history at each age group of diagnosis. (B) VF events during follow-up in each age group. Please note that no VF events occurred at ages

older than 50 years. VF ¼ ventricular fibrillation.

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7 Kitamura et al.J A N U A R Y 2 0 1 7 : 5 7 – 6 7 Prognosis and Features of Senior Brugada Patients

61

especially attenuation with age, in several parametersduring follow-up. There were no changes in r-Jinterval >90 and fragmented QRS during follow-up.On the other hand, 2 of 42 patients had day-to-dayvariations in ER pattern (ER was not documentedregularly). Six of 125 patients had a change frompositive to negative LP. According to the definitionof day-to-day variations in type 1 ECG, spontaneousalterations from type 1 to type 2 or 3 were observedin 14 of 26 cases. Among the 14 patients, 10 (of 26)had their ECGs reidentified as spontaneous type 1ECG. There were only 4 patients with type 1 ECGwho alternated to persistent type 2 or 3 ECG. Theprevalence of attenuating LP and day-to-day varia-tions was low (LP: 6 of 125 [4.8%]; day-to-dayvariation in type 1 ECG: 4 of 26 [15.3%]).

ICD INDICATION. Figure 4A shows the number of ICDindication category modified to the latest guidelineamong all patients and those with implanted ICD.There were 24 of 181 cases (13.2%) classified as Class 1;21 of 181 (11.6%) as Class 2a; 86 of 181 (47.5%) as Class2b; and 5 as Class 3 and 45 as others (e.g., sponta-neous type 1 ECG and family history of SCD) in allpatients (Figure 4A, left bar). Regarding the numbersof patients who received ICD implantation (Figure 4A,

right bar), 23 patients were in Class 1, 6 in Class 2a,and 44 in Class 2b. No patients in Class 3 or otherreceived ICD implantation. Figure 4B comparespatient numbers classified by ICD indication andpatients with implanted ICDs between the youngerand senior groups. In the senior group, there were5 patients in Class 1, 6 patients in Class 2a, and 32patients in Class 2b. Among senior patients with ICD,there were 5 patients in Class 1, 3 patients in Class 2a,and 19 patients in Class 2b.

COMPLICATIONS ASSOCIATED WITH THE ICD. Table 5presents details of complications such as inappro-priate shocks and other ICD complications. A total of18 patients with ICD implantation had inappropriatetherapy (18 of 73; 24.7%). The cause was AF in 10cases, paroxysmal supraventricular tachycardia in 4,lead fractures in 2, sinus tachycardia in 1, and T-waveoversensing in 1. The incidence of inappropriatetherapy due to AF was not statistically different be-tween the 2 groups. Other complications (14 events)were found in 14 patients who received ICD (14 of 73;19.2% [10 lead fractures, 3 ICD infections, and 1 he-matoma at ICD replacement]). Kaplan-Meier analysesof all complications and lead fracture are shownin Figure 5.

TABLE

3Pat

ient

Char

acte

risticsof

5Se

nior

Pat

ient

sWithBru

gad

aSy

ndro

meWho

Wer

eRes

uscita

tedFr

omVen

tricular

Fibrillat

ionat

Diagnos

is

Patient

#Age

,yrs

Sex

FHof

SCD

Sync

ope

VFSu

rvivor

Spon

tane

ous.

Type

1EC

GICD

App

ropr

iate

Therap

y*Inap

prop

riate

Therap

y*P-W

ave

Dur

ation,

ms

QRS

Dur

ation,

ms

QTc

Inte

rval,m

sr-JInte

rval

inV2>90

ms*

Frag

men

ted

QRS*

ERPattern

*

168

Males

þ�

þþ

þ�

�16

210

041

2þ

��

265

Males

��

þ�

þ�

�18

5123

428

��

þ3

64

Males

�þ

þ�

þ�

�172

99

407

��

�4

64

Males

��

þþ

þ�

þ(lea

dfracture)

155

106

403

��

�5

63

Males

��

þ�

þ�

�18

910

041

2�

��

64.8�

1.9

5/5(100%)

1/5(20%)

1/5(20%)

5/5(100%)

2/5(40%)

5/5(100%)

0/5

(0%)

1/5(20%)

172�

1510

2�

7.4

412�

9.5

1/5(20%)

0/5

(0%)

1/5(20%)

Day

-to-

Day

Variation

sof

Bru

gada

ECG

AF

Pos

itiveLP

bySA

ECG

VF

Indu

cibilit

yVFIndu

ctionby

Sing

leor

Dou

ble

VFIndu

ction

byTr

iple

HV

Inte

rval,m

sRVA

ERP,

ms

RVOTER

P,ms

New

lyDev

elop

edIH

DLe

adFr

actu

reICD

Infection

Hem

atom

aat

Rep

lace

men

t

1�

��

þþ

�48

400/230

400/220

��

��

2�

�þ

þ�

þ40

400/220

400/260

þ�

��

3�

þ�

þ�

þ50

400/210

400/220

��

��

4�

�þ

þþ

�35

400/210

400/240

�þ

��

5�

�þ

þ�

þ42

400/230

400/240

��

��

0/5

(0%)

1/5(20%)

3/5(60%)

5/5(100%)

2/5(40%)

3/5(60%)

42�

7.6

220

�8.9

236�

151/5(20%)

1/5(20%)

0/5

(0%)

0/5

(0%)

*Plus/minus

(�/þ

)sign

sindicate

nega

tive

orpo

sitive

inea

chva

riable.

AF¼

atria

lfibrillation;

ECG¼

(12-lead

)electrocardiogram

;ER¼

earlyrepo

lariz

ation;

ERP¼

effectiverefractory

perio

d;FH

¼family

history;

ICD¼

implan

tablecardiove

rter-defi

brillator;IHD¼

isch

emiche

artdisease;

LP¼

late

potential;RVA¼

right

ventric

ular

apex

;RVOT¼

right

ventric

ular

out;

SAEC

G¼

sign

al-ave

rage

dEC

G;SC

D¼

sudd

encardiacde

ath;

VF¼

ventric

ular

fibrillation.

Kitamura et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7

Prognosis and Features of Senior Brugada Patients J A N U A R Y 2 0 1 7 : 5 7 – 6 7

62

DISCUSSION

CLINICAL CHARACTERISTICS OF SENIOR PATIENTS

WITH BRUGADA SYNDROME IN COMPARISON TO

THOSE IN THE YOUNGER GROUP. BrS is generallycharacterized by male predominance with a middle-aged onset of cardiac events or diagnosis at 40 to50 years of age, or both. There are, however, certainnumbers of patients in whom BrS is diagnosed at $60years of age in the clinical setting. In the presentstudy, 32% (58 of 181 subjects) of our BrS cohortrepresented the senior group. Recent reports indi-cated that the senior BrS patients had benign prog-nosis compared with the younger age populations ingeneral BrS cohort as well as in high-risk patients(5,6). Our study also confirmed their results with abenign prognosis of the senior group $60 years ofage. Furthermore, the results indicated certain clin-ical characteristics suggestive of better prognosticsigns that were not indicated in the previous reports.

Many clinical variables have been proposed as riskfactors with which to predict cardiac events inpatients with BrS. Among these variables, symptom-atic patients including a history of VT/VF, syncope ofunknown origin, and spontaneous type 1 ECG wereassumed to be important predictors (2). Furthermore,whether other symptoms or clinical signs could beapplicable to predict cardiac events in senior BrS pa-tients is not known. Conte et al. (5) reported a betterprognosis of the senior BrS patients with significantlyless frequent prevalence in family history of SCD thanin the younger patients. Our study did not show adifference in prevalence of family history of SCD be-tween the senior and younger groups. The discrepantresults may be due to differences in the subjects’backgrounds, as their study population included arelatively high prevalence of family members of BrS(49% in 58 cases), and our cases represented 21.1% to25.7% of the family history in both groups. The reportby Conte et al. (5) also indicated fewer numbers ofpatients with induced VT/VF during programmedventricular stimulation in the senior group than theyounger group, but we could not observe a differencein the ratio of induced VT/VF between the 2 groups.As to the induced VT/VF by EPS, its positive predic-tive value for cardiac events has been in intensedispute and a consensus has been reached (2,3,9,18–20). Different outcomes by EPS might be attributedto differences in stimulation protocol, but a pro-spective study by a fixed stimulation protocol couldnot demonstrate a predictive value of VT/VF induc-tion for cardiac events (20).

In the present study, spontaneous type 1 ECG andday-to-day variations in Brugada ECG patterns were

FIGURE 3 Kaplan-Meier Curves of Ventricular Fibrillation in

the Senior and Younger Groups During Follow-Up Period

TABLE 4 Results of Univariate Cox Regression Analysis of Risk Factors and Combination

of Risk Factors

Hazard Ratio 95% Confidence Interval p Value

Documented VF 22.1 10.22–54.33 <0.01

Syncope 2.89 0.884–9.5 0.079

Spontaneous type 1 ECG 2.94 0.78–11.01 0.111

Family history of SCD 0.32 0.001–11.18 0.249

Male 22.1 0.022–102.5 0.579

Age <60 yrs 2.82 1.695–50.74 0.042

ER pattern 17.43 4.597–66.1 <0.01

Fragmented QRS 8.574 2.614–28.12 <0.01

Day-to-day variation of type 1 ECG 9.617 2.43–38.06 <0.01

r-J interval $90 ms 4.626 0.592–36.141 0.144

Positive LP by SAECG 0.771 0.226–2.636 0.679

Induced VF during EPS 2.599 0.333–20.31 0.362

VF induction by single ordouble extra-stimuli

4.235 0.51–35.19 0.181

RVA ERP <200 ms 0.833 0.162–4.295 0.827

RVOT ERP <200 ms 2.837 0.572–14.07 0.202

Combined risk factors

VF and age <60 yrs 22.18 12.13–82.4 <0.01

Syncope and age <60 yrs 4.918 1.548–15.623 0.007

Spontaneous type 1 ECG and age <60 yrs 4.09 1.229–13.62 0.022

ER pattern and age <60 yrs 20.88 5.602–77.833 <0.01

Fragmented QRS and age <60 yrs 10.74 3.457–33.37 <0.01

Day-to-day variation of type 1ECG and age <60 yrs

9.347 2.614–33.42 <0.01

Positive LP and age <60 yrs 1.511 0.479–4.762 0.481

Induced VF and age <60 yrs 10.7 1.364–83.32 0.024

VF induction by single or doubleextra-stimuli and age <60 yrs

11.64 1.4–96.8 0.023

RVOT ERP <200 ms and age <60 yrs 0.955 0.193–4.733 0.955

Five combined risk factors with the highest hazard ratios

VF and fragmented QRS 55.2 16.25–98.2 <0.01

VF and day-to-day variation of type 1 ECG 26.5 7.954–88.34 <0.01

VF and ER pattern 25.0 7.556–82.7 <0.01

VF and age <60 yrs 22.18 12.13–82.4 <0.01

ER pattern and age <60 yrs 20.88 5.602–77.833 <0.01

Abbreviations as in Tables 1 and 2.

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7 Kitamura et al.J A N U A R Y 2 0 1 7 : 5 7 – 6 7 Prognosis and Features of Senior Brugada Patients

63

significantly less prevalent in the senior group thanin the younger group. Type 1 ECG and day-to dayvariations of Brugada ECG pattern are thought tobe risks factor for fatal ventricular arrhythmiadepending on different ethnic groups and popula-tion of the study subjects, values, and timing of ECGrecordings, and selections of ECG lead placements atdifferent intercostal spaces (21–23). Therefore, thelower frequency of those findings might contributeto a better prognosis in the senior group. Amongother conditions supposed to predict cardiac events,there were fewer patients with r-J interval >90 ms(17) in the senior group than in the younger group.However, other parameters including positive LPby SAECG (12,24), fragmented QRS (10), history ofAF (17,25,26), ER patterns (22,27,28), augmented STelevation during recovery phase after exercise (11),and ventricular effective refractory period <200 ms(20) were not different between the 2 groups.

POSSIBLE REASONS FOR BETTER

PROGNOSIS IN SENIOR PATIENTS

WITH BRUGADA SYNDROME

Our results indicated that BrS patients $60 years ofage had a better prognosis than those <60 years ofage over 7-year follow-up, confirming the resultsof previous reports (5,6). Five patients with a historyof VF at the age of diagnosis older than 60 years of agedid not experience any recurrence of VF events dur-ing the follow-up (11.7 � 3.2 years). Furthermore,patients including the younger group older than 50

years of age had no recurrences of VF. As to themechanism of ST-segment elevation and develop-ment of VT, repolarization theory (29) and depolari-zation theory (30) have been proposed withoutreaching a firm consensus. The results that sponta-neous type 1 ECG were less frequently observed inthe senior group with better prognosis mightbe explained by decreased hormonal influence oftestosterone with age (31,32). In addition, a prolongedr-J interval of $90 ms, which might indicate con-duction abnormality and day-to-day variations ofBrugada ECG patterns, which might indicate auto-nomic tone disorder as markers of the proposed risk(13,29), were less frequent in the senior group.Moreover, attenuation or disappearance of severalrisk factors with age during follow-up were observedin a small number of patients. Therefore, the true

FIGURE 4 Number of Patients in Each Class of ICD Indication and Actual Numbers of Patients With ICD

(A) Graphs show numbers of each class of ICD indication in all cases (left) and numbers of patients with ICD (ICDþ) (right). (B) Graphs present

numbers of each class of ICD indication in all cases and numbers of ICD(þ) in the younger group (left) and those in the senior group (right).

ICD ¼ implantable cardioverter-defibrillator.

Kitamura et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7

Prognosis and Features of Senior Brugada Patients J A N U A R Y 2 0 1 7 : 5 7 – 6 7

64

reason why the senior patients showed a betterprognosis than the younger group <60 years of agewas not clarified by the present study, and furtherexploration should be continued. In addition, in ourcase series, there were no VF events after 5-yearfollow-up, contrary to our expectations. We antici-pated that there would be events in the late period offollow-up, as reported in other studies (18,20).Therefore, we also expected that there would be pa-tients with VF events during later follow-up periodin future. One possible reason could be that thepatients had more time to receive education atoutpatient clinics to avoid being exposed to cir-cumstances such as high-grade fever, taking medi-cation worsening BrS ECG (Na channel blockers,calcium channel blockers, and so forth) (33,34), or

hyperactivated parasympathetic tone as the follow-up duration became longer. However, we do nothave evidence or data in support of this.

AGE AS A PROGNOSTIC FACTOR. A number of clin-ical prognostic factors have been shown to predict VFin patients with BrS, as mentioned above. In our caseseries, younger patients, <60 years of age, had worseprognosis than senior patients based on the log-ranktest and univariate Cox regression analysis. In theCox regression model, documented VF, fragmentedQRS, ER pattern, and day-to-day variation in type 1ECG could also predict VF events. Moreover, althoughseveral risk factors (syncope, spontaneous type 1ECG, induced VF, VF induction mode single or dou-ble) do not alone statistically predict VF events,

TABLE 5 Inappropriate Therapy and Complication During Follow-Up

in 2 Groups

Younger Group(n ¼ 47)

Senior Group(n ¼ 26) p Value

Inappropriate therapy 12/47 (25.5%) 6/26 (23.1%) 0.854

AF 6/12 (50.0%) 4/6 (66.7%) 0.86

Other SVTs 3/12 (25.0%) 1/6 (3.9%) 0.536

Lead fracture (as cause ofinappropriate therapy)

2/12 (4.3%) 0/6 (0%) -

Sinus tachycardia 1/12 (2.1%) 0/6 (0%) -

T-wave oversensing 0/47 (0%) 1/26 (3.9%) -

Other complications 10/47 (21.3%) 4/26 (15.4%) 0.224

Lead fracture 7/10 (70.0%) 3/4 (75.0%) 0.64

ICD infection 2/10 (20.0%) 1/4 (25.0%) 0.595

Hematoma at implantation orreplacement of ICD

1/10 (10.0%) 0/6 (0%) -

Values are n/N (%).

SVT ¼ supraventricular tachycardia; other abbreviations as in Tables 1 and 2.

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7 Kitamura et al.J A N U A R Y 2 0 1 7 : 5 7 – 6 7 Prognosis and Features of Senior Brugada Patients

65

combinations of age <60 years and those factors(syncope, spontaneous type 1 ECG, induced VF, VFinduction mode single or double) may predict VFevents. Recently, a combination of risk factors hasbeen proposed to predict VF in patients with BrS(35,36). Age <60 years might be helpful not only as asimple prognostic factor but also as one of thereasonable risk factors in combination with other riskfactors to stratify VF recurrence risk. However, wecould not conduct a multivariate analysis because ofan insufficient number of VF events. Therefore, astudy with a larger sample size with enough VFevents to conduct multivariate analysis is warrantedto confirm the results.

ICD INDICATION FOR THE SENIOR PATIENTS WITH

BRUGADA SYNDROME. In the present study, allpatients were reclassified by ICD indication accord-ing to the latest consensus statement (9). Figure 4shows the distribution of reclassified patients. Pa-tients with ICD insertion were reclassified into class1, 2a, or 2b, but none in class 3 or other. There were5 patients reclassified into class 1, and 6 patientsinto class 2a in senior group. There was no VT/VF inthose patients during the long-term follow-upperiod. The indication for ICD implantation hasnot been clarified in senior BrS patients. In theconsensus statements for BrS (9), a specific con-sideration of ICD implantation for senior BrS pa-tients was not described. Although ICD implantation

FIGURE 5 ICD Complications

Kaplan-Meier curves of all ICD complications (A) and lead fractures (B) d

and lead fracture per 73 cases increased over time. Abbreviations as in F

is supposed to be the only therapeutic means toprotect against SCD at the present time, there arefactors associated with device complications afterthe implantation (6). Therefore, the risk stratifica-tion of cardiac events and a better prognosis of thesenior patients must be carefully considered for theselection of ICD implantation. However, it shouldbe emphasized that 5 senior patients in their 60salready had VF episodes at the time of diagnosis inthe present study. Results further indicate that pa-tients with BrS over 60 years of age have the

uring follow-up (8.2 � 6.8 years). The incidence of ICD complication

igure 4.

PERSPECTIVES

COMPETENCY IN MEDICAL KNOWLEDGE:

Senior BrS patients $60 years of age at diagnosis

have a better prognosis than the younger age group.

ICD implantations for BrS patients $60 years of age

should be carefully evaluated.

TRANSLATIONAL OUTLOOK: A study with a

larger population and a longer follow-up period,

specifically a span lasting until the end of life in senior

patients, is needed.

Kitamura et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7

Prognosis and Features of Senior Brugada Patients J A N U A R Y 2 0 1 7 : 5 7 – 6 7

66

potential to develop VF and that ICD indicationshould not be avoided simply because of advancedage at the time of diagnosis; the decision should bemade by weighing the risks and benefits of im-plantation in each individual.

COMPLICATIONS ASSOCIATED WITH THE ICD. Inthe present study, there were no statistical differ-ences between the 2 groups in the prevalence of totalnumber of inappropriate therapies or inappropriatetherapy due to AF and other supraventricular tachy-cardias (SVT). However, during the follow-up period,complications or lead fracture incidence graduallyincreased, which is a finding in line with that ofa previous study (6). Kamakura et al. (6) reportedthat numbers of SVT and inappropriate shocksincreased with age. In our study, this did not achievestatistical significance; however, the prevalence ofinappropriate therapy due to AF was slightly higherin the senior group than in the younger group.Therefore, considering the increasing accumulationof complications and inappropriate therapy due toSVT including AF in senior patients, careful decisionmaking for ICD implantation may be necessary insenior patients with BrS, particularly in patients withClass 2b indication.

STUDY LIMITATIONS. Our study has several limita-tions: it is a single-center study with retrospectiveanalysis with heterogeneous clinical characteristics.The numbers of the study subjects were rathersmall. In addition, a direct correlation between car-diac events during follow-up and the positive find-ings (i.e., type 1 ECG, day-to-day variation, orprolonged R-J interval) was not confirmed. Further-more, there were only 11 events during the follow-upperiod in the present study, which contributed toseveral risk factors with large hazard ratios and large

confidence intervals in Cox regression analysis.Therefore, we might have overestimated the impactof those risk factors, and the results in Table 4should be interpreted carefully. Moreover, becausea multivariate analysis needs at least 10 events perrisk factor, we could not conduct a multivariate Coxregression analysis to confirm the independence ofeach risk factor. A large-scale prospective study isneeded to confirm these results and to clarify themechanism.

CONCLUSIONS

Senior BrS patients $60 years of age at diagnosisseem to have a better prognosis than younger pa-tients. ICD implantations for BrS patients $60 yearsof age should be carefully evaluated.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Takeshi Kitamura, Department of Cardiology, TokyoMetropolitan Hiroo Hospital, 2-34-10 Ebisu, Shibuya-ku, Tokyo, Japan. E-mail: take1124@hotmail.co.jp.

RE F E RENCE S

1. Brugada P, Brugada J. Right bundle branchblock, persistent ST segment elevation and sud-den cardiac death: a distinct clinical and electro-cardiographic syndrome. A multicenter report.J Am Coll Cardiol 1992;20:1391–6.

2. Mizusawa Y, Wilde A. Brugada syndrome. CircArrhythm Electrophysiol 2012;5:606–16.

3. Berne B, Brugada J. Brugada syndrome. Circ J2012;76:1563–71.

4. Matsuo K, Akahoshi M, Hayano M, et al. Theprevalence, incidence and prognostic value of theBrugada-type electrocardiogram: a population-based study of four decades. J Am Coll Cardiol2001;38:765–70.

5. Conte G, De Asmundis C, Sieira J, et al. Clinicalcharacteristics, management, and prognosis of

elderly patients with Brugada syndrome.J Cardiovasc Electrophysiol 2014;25:514–9.

6. Kamakura T, Kamakura S, Kusano K, et al.Evaluation of the necessity for cardioverter-defibrillator implantation in elderly patients withBrugada syndrome. Circ Arrhythm Electrophysiol2015;8:785–91.

7. Wilde AA, Antzelevitch C, Borggrefe M, et al.Proposed diagnostic criteria for the Brugada syn-drome: consensus report. Eur Heart J 2002;23:1648–54.

8. Antzelevitch C, Brugada P, Wilde A, et al.Brugada syndrome: report of the secondconsensus conference: endorsed by the HeartRhythm Society and the European HeartRhythm Association. Circulation 2005;111:659–70.

9. Priori SG, Wilde AA, Horie M, et al. HRS/EHRA/APHRS expert consensus statement on the diag-nosis and management of patients with inheritedprimary arrhythmia syndromes. Heart Rhythm2013;10:1932–63.

10. Morita H, Zipes DP, Wu J, et al. FragmentedQRS as a marker of conduction abnormality and apredictor of prognosis of Brugada syndrome. Cir-culation 2008;118:1697–704.

11. Makimoto H, Kamakura S, Shimizu W, et al.Augmented ST-segment elevation during recoveryfrom exercise predicts cardiac events in patientswith Brugada syndrome. J Am Coll Cardiol 2010;56:1576–84.

12. Ikeda T, Sakurada H, Yamaguchi T, et al.Assessment of noninvasive markers in identifyingpatients at risk in the Brugada syndrome: insight

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 3 , N O . 1 , 2 0 1 7 Kitamura et al.J A N U A R Y 2 0 1 7 : 5 7 – 6 7 Prognosis and Features of Senior Brugada Patients

67

into risk stratification. J Am Coll Cardiol 2001;37:1628–34.

13. Breithardt G, Simson MB, Steinbeck G, et al.Standards for analysis of ventricular late poten-tials using high-resolution or signal-averagedelectrocardiography: a statement by a task forcecommittee of the European Society of Cardiology,the American Heart Association, and the AmericanCollege of Cardiology. J Am Coll Cardiol 1991;17:999–1006.

14. Mizusawa Y, Sakurada H, Nishizaki M,Hiraoka M. Effects of low-dose quinidine on ven-tricular tachyarrhythmias in patients with Brugadasyndrome. Low-dose quinidine therapy as anadjunctive treatment. J Cardiovasc Pharmacol2006;47:359–64.

15. Epstein AE, Dimarco JP, Ellenbogen KA, et al.ACC/AHA/HRS2008 guidelines for device-basedtherapy of cardiac rhythm abnormalities: a reportof the American College of Cardiology/AmericanHeart Association task force on practice guidelines(Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline update for implantation ofcardiac pacemakers and antiarrhythmia devices).J Am Coll Cardiol 2008;51:2085–105.

16. Ohe T, Aizawa F, Atarashi H, et al. Guidelinesfor diagnosis and management of patients withlong QT syndrome and Brugada syndrome. Circ J2007;71:1207–51.

17. Takagi M, Yokoyama Y, Aonuma K, Aihara N,Hiraoka M. Japan Idiopathic Ventricular FibrillationStudy (J-IVFS) Investigators. Clinical characteris-tics and risk stratification in symptomatic andasymptomatic patients with Brugada syndrome:multicenter study in Japan. J Cardiovasc Electro-physiol 2007;18:1244–51.

18. Wilde AA, Viskin S. EP testing does not predictcardiac events in Brugada syndrome. HeartRhythm 2011;8:1598–600.

19. Brugada J, Brugada R, Brugada P. Electro-physiologic testing predicts events in Brugadasyndrome. Heart Rhythm 2011;8:1595–7.

20. Priori SG, Gasparini M, Napolitano C, et al. Riskstratification in Brugada syndrome: results of thePRELUDE (PRogrammed ELectrical stimUlation pre-Dictive valuE) registry. J Am Coll Cardiol 2012;59:37–45.

21. Gehi AK, Duong TD, Metz LD, Gomes JA,Mehta D. Risk stratification of individuals with theBrugada electrocardiogram: a meta-analysis.J Cardiovasc Electrophysiol 2006;17:577–83.

22. Kamakura S, Ohe T, Tomoike H, et al. BrugadaSyndrome Investigators in Japan. Long-termprognosis of probands with Brugada-pattern ST-elevation in leads V1-V3. Circ Arrhythm Electro-physiol 2009;2:495–503.

23. Probst V, Veltmann C, Wilde AA, et al.Longterm prognosis of patients diagnosed withBrugada syndrome: results from the FINGER Bru-gada Syndrome Registry. Circulation 2010;121:635–43.

24. Huang Z, Patel C, Yan G-X, et al. Role ofsignal-averaged electrocardiograms in arrhythmicrisk stratification of patients with Brugada syn-drome: a prospective study. Heart Rhythm 2009;6:1156–62.

25. Morita H, Kusano-Fukushima K, Ohe T, et al.Atrial fibrillation and atrial vulnerability in patientswith Brugada syndrome. J Am Coll Cardiol 2002;40:1437.

26. Bordachar P, Reuter S, Garrigue S, et al. Inci-dence, clinical implications and prognosis of atrialarrhythmias in Brugada syndrome. Eur Heart J2004;25:879–84.

27. Sarkozy A, Chierchia GB, Paparella G, et al.Inferior and lateral electrocardiographic repolari-zation abnormalities in Brugada syndrome. CircArrhythm Electrophysiol 2009;2:154–61.

28. Takagi M, Aonuma K, Sekiguchi Y,Yokoyama Y, Aihara N, Hiraoka M. The prognosticvalue of early repolarization (J wave) and ST-segment morphology after J wave in Brugadasyndrome: multi-center study in Japan. HeartRhythm 2013;10:533–9.

29. Yan GX, Antzelevitch C. Cellular basis for theBrugada syndrome and other mechanisms ofarrhythmogenesis associated with ST-segmentelevation. Circulation 1999;100:1660–6.

30. Wilde AA, Postema PG, Antzelevitch C, et al.The pathophysiological mechanism underlyingBrugada syndrome depolarization versus repolar-ization. J Mol Cell Cardiol 2010;49:543–53.

31. Shimizu W, Matsuo K, Kokubo Y, et al. Sexhormone and gender difference—role of testos-terone on male predominance in Brugada syn-drome. J Cardiovasc Electrophysiol 2007;18:415–21.

32. Ezaki K, Nakagawa M, Taniguchi Y, et al.Gender differences in the ST segment: effect ofandrogen-deprivation therapy and possible role oftestosterone. Circ J 2010;74:2448–54.

33. Postema PG, Brugada J, Wilde AA, et al. Drugsand Brugada syndrome patients: review of theliterature, recommendations, and an up-to-datewebsite (www.brugadadrugs.org). Heart Rhythm2009;6:1335–41.

34. Postema PG, Wilde AA, Woosley RL, et al. Safedrug use in long QT syndrome and Brugada syn-drome: comparison of website statistics. Europace2013;15:1042–9.

35. Okamura H, Kamakura T, Morita H, et al.Risk stratification in patients with Brugada syn-drome without previous cardiac arrest—prognosticvalue of combined risk factors. Circ J 2015;79:310–7.

36. Tokioka K, Kusano K, Morita H, et al. Elec-trocardiographic parameters and fatalarrhythmic events in patients with Brugada syn-drome: combination of depolarization and repo-larization abnormalities. J Am Coll Cardiol 2014;63:2131–8.

KEY WORDS Brugada syndrome,implantable cardioverter-defibrillator,ventricular fibrillation