Electrophysiologic investigation in Brugada syndrome

European Heart Journal (2002) 23, 1394 1401 doi:10.1053/euhj.2002.3256, available online at http://www.idealibrary.com on Electrophysiologic investigation in Brugada syndrome Yield of programmed ventricular stimulation at two ventricular sites with up to three premature beats L. Eckardt 1, P. Kirchhof 1, E. Schulze-Bahr 1, S. Rolf 1, M. Ribbing 1,P.Loh 1, H.-J. Bruns 1, A. Witte 1, P. Milberg 1, M. Borggrefe 2, G. Breithardt 1, T. Wichter 1 and W. Haverkamp 1 1 Hospital of the Westfälische Wilhelms-University, Department of Cardiology and Angiology and Institute for Arteriosclerosis Research, Münster, Germany; 2 University Hospital Mannheim, Mannheim, Germany Introduction Numerous reports on the inducibility of ventricular tachyarrhythmias (VT) in patients with atypical right bundle branch block and right precordial ST-elevation (Brugada syndrome) are based on multicentre studies that have used different stimulation protocols. Therefore, we prospectively investigated the inducibility of VT in these patients using a uniform protocol. Methods In 41 consecutive patients (29 males) showing a pattern of right bundle branch block and ST-elevation, programmed ventricular stimulation was performed in the right ventricular apex with up to three premature stimuli at sinus rhythm and at four different paced cycle lengths (500, 430, 370, and 330 ms) until refractoriness was reached or reproducible induction of a sustained (>30 s) VT occurred. If a VT was not reproducibly inducible, the same protocol was repeated in the right ventricular outflow tract. Results A history of life-threatening events defined as syncope (n=17) or aborted sudden cardiac death (n=13) was present in 30 patients (73%); 11 individuals were asymptomatic. Inducibility (68%) was similar between symptomatic (n=21, 70%) and asymptomatic patients (n=7, 64%). In 16 (39%) patients, VT were reproducibly inducible. If patients were only stimulated in the right ventricular apex, inducibility rate decreased to 39%. If only two premature beats at two sites were used it was as low as 32%. The mean coupling intervals of the second and third premature stimuli inducing sustained VT were short: 189 21 ms vs 186 22 ms, respectively. Forty-four percent of all patients (i.e. 64% of the inducible patients) had inducible VT only with coupling intervals shorter than 200 ms. Conclusions The stimulation protocol markedly influences the extent of inducibility of VT in patients with right bundle branch block and ST-segment elevation. These findings question the significance of previous multicentre studies using different stimulation protocols and should have implications for further studies. (Eur Heart J, 2002; 23: 1394 1401, doi:10.1053/euhj.2002. 3256) 2002 The European Society of Cardiology. Published by Elsevier Science Ltd. All rights reserved. Key Words: Brugada syndrome, programmed ventricular stimulation Revision submitted 4 March 2002, and accepted 13 March 2002. Partly supported by: (1) Deutsche Forschungsgemeinschaft, Bonn, Germany (Schu 1082/2-2, later Sonderforschungsbereich 556; projects A1, C4), and (2)Alfried-Krupp-von-Bohlen-Halbach Foundation, Essen, Germany. (3) Grants from Interdisciplinary Center for Clinical Research Münster, Germany (BMBF, AZ 01 KS 9604) Correspondence: Dr med. Lars Eckardt, Medizinische Klinik und Poliklinik C, Kardiologie Angiologie, Universitätsklinikum Münster, Germany. 0195-668X/02/$35.00 2002 The European Society of Cardiology. Published by Elsevier Science Ltd. All rights reserved.

Programmed electrical stimulation in Brugada syndrome 1395 Introduction In 1992, Brugada and Brugada [1] identified a syndrome characterized by atypical right bundle branch block and right precordial ST-elevation associated with episodes of syncope or sudden cardiac death caused by rapid polymorphic ventricular tachyarrhythmias (VT). This has now become known as a new entity called Brugada syndrome. It is a genetically transmitted disease that may cause sudden cardiac death in individuals without structural heart disease. Available multicentre data by Brugada et al. [2,3] have confirmed the malignant character of this syndrome showing that it is associated with a recurrence rate of arrhythmic events of about 40% in 3 years after the initial event of resuscitation or syncope and a 5% event rate in asymptomatic individuals. Programmed ventricular stimulation has been suggested as a means to identify patients at risk [4]. Brugada et al. [2] reported that the inducibility of VT was a good predictor of poor outcome in asymptomatic patients with an abnormal ECG. However, the results of programmed ventricular stimulation especially in asymptomatic individuals with the characteristic ECG findings have been controversial. In contrast to Brugada and coworkers, Priori et al. [5] published multicentre data that questioned the diagnostic predictive accuracy of programmed ventricular stimulation and the risk of cardiac events in asymptomatic patients. Thus, although programmed ventricular stimulation is largely used to define risk in Brugada syndrome caution about its prognostic value may be justified. As both reports were based on multicentre data, criticism has been raised that the differences in the data may in part be related to different stimulation protocols. The coupling intervals, the number of premature beats, and the stimulation sites have also not been systematically differentiated. Therefore, we prospectively investigated the yield of programmed ventricular stimulation using a uniform stimulation protocol with stimulation at two ventricular sites and at five different cycle lengths of basic drive with up to three premature stimuli in a large group of patients with an ECG pattern compatible with the Brugada syndrome. Patients and methods A total of 41 consecutive patients (29 males, 71%) were diagnosed to have an ECG compatible with the Brugada syndrome within the last 6 years in our institution (Münster, Germany). Structural heart disease was excluded after extensive evaluation, including echocardiography (n=41), coronary angiography (n=34), right (n=30) and left ventricular angiography (n=34), right ventricular biopsies (n=25), and nuclear MRI (n=19). Patients were divided into two groups depending on their clinical presentation at the time of diagnosis. The first group (Table 1) consisted of 30 symptomatic patients. Thirteen patients (12 male) had a history of documented ventricular fibrillation (VF) and aborted sudden cardiac death and 17 patients (nine male) had one or more unexplained episodes of syncope (group I). In addition, there were 11 (eight male) individuals with the characteristic ECG pattern but without a history of resuscitation or syncope (group II, Table 2). They were seen because of the characteristic ECG pattern diagnosed by their general practitioner (n=3; patients 31, 37, 40), a positive family history (n=8; patients 32, 34 38, 40, 41) and/or palpitations (n=2; patients 37, 38). The age of all patients at diagnosis ranged between 24 and 66 years (median 44 13 years). The medical histories of all patients had been unremarkable before their first clinical presentation. Six patients (patients 24, 25, 36 39) had palpitations. In four patients (patients 24, 25, 37, 38), a regular supraventricular tachyarrhythmia had been documented [6]. All patients were unrelated except patients 14 and 25 (father/daughter), patients 5 and 36 (brother/sister), and patients 20 and 37 (sisters). Patients 36 and 37 presented to our department because of palpitations and relatives with known Brugada syndrome (patients 20 and 5, respectively). In 18 (44%) patients (patients 8, 13 16, 23 26, 29, 32, 34 38, 40, 41), there was a family history of sudden cardiac death and/or syncope. All except five patients (patients 2, 3, 13, 18, 19) were investigated for mutations in the sodium channel gene SCN5A. Three patients (8%, two were related) with a SCN5A mutation were identified. Patient 25 had a heterogeneous two-basepair deletion in SCN5A. This mutation was also found in her father (patient 14) but not in any other family member nor in a control sample of 100 unrelated, healthy individuals. Noteworthy, the father of patient 25 (patient 14) had no inducible VT despite having the same genotype as his daughter in whom VT was inducible. Electrocardiographic findings and electrophysiologic testing All patients had ECG documentation with RBBB and ST-segment elevation ( 0 2 mv) in the right precordial leads. The ECG was spontaneously abnormal in 28 patients and became abnormal after intravenous administration of ajmaline (1 mg. kg 1 body weight, 10 mg-. min 1 ) in the remaining 13 individuals. Ten (24%) patients (eight symptomatic patients) had a HV-interval 60 ms (Tables 1 and 2). Electrophysiologic testing was performed in the fasting, drug free, non-sedated state, after written informed consent had been obtained. At first, stimulation was performed in the right ventricular apex with one and two premature stimuli at sinus rhythm and at four different cycle lengths (500, 430, 370, and 330 ms). Premature beats were started in late diastole; coupling intervals were then reduced in decrements of 10 ms until refractoriness was reached. The endpoint of programmed ventricular stimulation was the

1396 L. Eckardt et al. Table 1 Clinical and electrophysiologic characteristics of Brugada patients presenting with cardiac arrest or syncope Pt. no. Sex/Age (years) Clinic Family history SCN5A Doc.arrhy. HV (ms) Induction Mode of VT/VF induction (ms) Therapy/Follow up 1 m/59 cardiac arrest 0 0 VF 50 VF RVA (500/230/170/230) RVA (500/230/180/180) ICD/64 2 m/36 cardiac arrest 0 VF 55 VF RVO (500/220/190) 0 ICD/53 3 m/30 cardiac arrest 0 VF 50 VF RVO (500/210/160/160) RVO (500/220/200/200) ICD/49 4 m/31 cardiac arrest 0 0 VF 40 VF RVA (430/220/240) RVA (500/230/170/220) ICD/75 5 m/37 cardiac arrest 0 0 VF 35 VF RVA (500/240/190/170) RVO (500/210/160) ICD/25 6 m/49 cardiac arrest 0 0 VF 60 VF RVA (500/230/190/180) 0 ICD/67 7 m/52 cardiac arrest 0 0 VF 35 0 0 0 CD/64 8 m/30 cardiac arrest SCD 0 ns VT 60 VF RVA (330/200/200) RVA (500/230/210/200) ICD/75 9 m/59 cardiac arrest 0 0 VF 50 nsvt 0 0 Death/3 10 f/66 cardiac arrest 0 0 VF 50 VF RVA (500/230/200/170) 0 ICD/30 11 m/30 cardiac arrest 0 0 VF 40 VF RVO (370/210/170) RVO (370/210/190) ICD/50 12 m/35 cardiac arrest 0 0 VF 55 VF RVA(430/230/180) 0 ICD/9 13 m/48 cardiac arrest SCD VF 60 VF RVA (430/210/190) RVA (430/210/180) ICD/38 14 m/64 syncope syncope + nsvt 70 nsvt 0 0 refused ICD/26 15 m/58 syncope syncope 0 nsvt 60 VF RVO (500/220/180/180) RVO (500/240/210/190) ICD/15 16 m/32 syncope syncope 0 0 85 VF RVO (500/220/180/180) 0 ICD/27 17 m/33 syncope 0 + 0 40 VF RVA (500/220/200) 0 ICD/34 18 m/59 syncope 0 0 ns VT 35 VF RVO (330/200/170) 0 ICD/56 19 m/31 syncope 0 0 70 VF RVO (370/180/200) RVO (500/210/170/220) refused ICD/27 20 f/43 syncope 0 0 50 ns-vt 0 0 ICD/46 21 m/54 syncope 0 0 0 55 VF RVA (500/260/210/160) 0 ICD/44 22 f/59 syncope 0 0 ns VT 40 0 0 0 refused ICD/27 23 f/48 syncope SCD, syncope 0 0 50 0 0 0 ICD/40 24 f/52 syncope syncope 0 SVT 45 VF RVO (500/190/170) 0 ICD/28 25 f/28 syncope syncope + SVT 60 VF RVO (300/280/160) RVO (500/230/200/200) ICD/26 26 f/58 syncope syncope 0 0 50 0 0 0 refused ICD/23 27 f/57 syncope 0 0 ns VT 50 0 0 0 ICD/50 28 m/55 syncope 0 0 0 45 VT RVO (330/220) 0 ICD/6 29 m/24 syncope syncope 0 0 35 VF RVO (500/220/160/170) 0 ICD/7 30 f/22 syncope 0 0 0 45 0 0 0 refused ICD/4 Pt. no.=patient number; Doc.arrhy.=documented arrhythmia; ns VT/VF=non-sustained ventricular tachyarrhythmia/fibrillation; RVO/A=right ventricular outflow tract/apex; =not performed; + =positive; 0=negative; ICD=implantable cardioverter defibrillator (in bold pts with adequate ICD discharge on follow-up), Follow-up in months after ICD implantation.

Programmed electrical stimulation in Brugada syndrome 1397 Table 2 Clinical and electrophysiologic characteristics of asymptomatic individuals with an ECG compatible with Brugada syndrome Pt. no. Sex/Age (years) Clinic Family history SCN5A Doc. arrhy. HV (ms) Induction VT/VF Mode of VT/VF induction (ms) Therapy/Follow-up 31 m/50 asymptom. 0 0 0 50 VF RVA (370/220/230) RVA (370/200/170) ICD/31 32 m/38 asymptom. SCD 0 0 50 0 0 0 no ICD/31 33 m/54 asymptom. 0 0 0 60 VF RVA (500/230/170/170) RVA (500/230/170/170) ICD/27 34 m/34 asymptom. syncope 0 0 40 0 0 0 no ICD/25 35 m/46 asymptom. SCD 0 0 30 VF RVO (500/190/140/150) 0 refused ICD/n.a 36 f/30 asymptom. SCD 0 0 45 0 0 0 no ICD/22 37 f/48 asymptom. syncope 0 SVT 40 VF RVA (500/200/170/180) RVO (500/210/180) ICD/29 38 m/57 asymptom. SCD 0 SVT 50 VF RVA (370/210/180) RVA (370/210/160) ICD/16 39 m/61 asymptom. 0 0 ns VT 60 VF RVA (330/210/180) RVA 500/260/220/200 ICD/23 40 m/22 asymptom. syncope 0 0 45 VF RVA (500/240/210/220) RVO (500/210/220) ICD/5 41 f/45 asymptom. syncope 0 0 45 0 0 0 no ICD/4 Pt. no.=patient number; Doc.arrhy.=documented arrhythmia; ns VT/VF=non-sustained ventricular tachyarrhythmia/fibrillation; RVO/A=right ventricular outflow tract/apex; =not performed; + =positive; 0=negative; ICD=implantable cardioverter defibrillator; n.a.=follow-up data not available.

1398 L. Eckardt et al. Table 3 Yield of PVS for stimulation stopped at a coupling interval of 200 ms or continued to refractoriness Coupling interval Symptomatic Asymptomtic 200 msec Refractoriness 200 msec Refractoriness (I) Right ventricular apex n=30 n=11 With 1 premature beat 0 0 0 0 2 premature beats 3 (10 3%) 5 (17 2%) 1 (9 1%) 3 (27 3%) 3 premature beats 1 (3 4%) 5 (17 2%) 1 (9 1%) 3 (27 3%) N (%) inducible 4 (13 8%) 10 (33 3%) 2 (18 2%) 6 (54 5%) (II) Right ventricular outflow tract n=26* With 1 premature beat 1 (3 4%) 1 (3 4%) 0 0 2 premature beats 1 (3 4%) 5 (17 0%) 0 0 3 permature beats 2 (6 8%) 5 (17 2%) 0 1 (9 1%) N (%) inducible 4 (13 8%) 11 (37 9%) 0 1 (3 4%) (I)+(II) With 1 premature beat 1 (3 4%) 1 (3 4%) 0 0 2 premature beats 4 (13 8%) 10 (34 5%) 1 (9 1%) 3 (27 3%) 3 premature beats 3 (10 33%) 10 (34 5%) 1 (9 1%) 4 (36 4%) N (%) inducible 8 (30 8%) 21 (70 0%) 2 (18 2%) 7 (63 6%) *Only those patients who were not reproducible inducible in the right ventricular apex were stimulated in the ouflow tract. reproducible induction of a sustained ventricular arrhythmia lasting for >30 s, causing syncope, or requiring intervention to be terminated. Reproducibility was tested by repeating stimulation at the same cycle length at which the VT/VF was inducible and by continuation of the protocol. Reproducibility was not tested if the patient refused further stimulation (n=3). Patients with inducible non-sustained ventricular arrhythmias (>6 beats that terminated spontaneously within 30 s) were classified as noninducible. If the endpoint was not reached with two premature beats at one site, three premature beats were administered at a cycle length of 500 ms at the same site. Thereafter, this stimulation protocol was repeated at the right ventricular outflow tract with one, two, and finally three premature beats. Results Programmed ventricular stimulation Sustained ventricular arrhythmias were inducible in the majority of patients (n=28, 68%). Almost all induced arrhythmias were rapid (mean cycle length 210 25 ms), polymorphic VT that could not be differentiated from VF. A monomorphic VT was inducible in only one patient (patient 5). The highest rate of inducibility occurred in the group of patients who had been resuscitated from cardiac arrest (85%), followed by those who were asymptomatic (64%), and the group of patients with syncopal episodes (59%). Thus, a comparable number of symptomatic (70%) and asymptomatic (64%) individuals were inducible. In those with a positive family history (n=18; 44% of all patients), seven patients (three symptomatic patients) were not inducible. Of all inducible patients (n=28), 16 (57%) patients were already inducible from the right ventricular apex (Table 3). The remaining patients were only inducible from the right ventricular outflow tract. Only one patient (patient 27) was inducible with one extrastimulus. In 17 (61%) of the inducible patients, ventricular arrhythmias were inducible by two premature stimuli whereas in the remaining 11 patients (39%) sustained arrhythmias were only inducible by the third premature stimulus. In 16 of the inducible patients (55%), ventricular arrhythmias were reproducibly inducible. In 29 (66%) of all 44 induced episodes of ventricular arrhythmias, coupling intervals shorter than 200 ms were required to induce sustained arrhythmias. The mean values for two and three premature beats that induced ventricular tachyarrhythmias were 189 21 ms and 186 22 ms, respectively. If programmed ventricular stimulation was stopped at a coupling interval of 200 ms, 62% and 71% of the otherwise inducible symptomatic and asymptomatic patients would not have been inducible, respectively, and only 14% would have been inducible twice. In four of the inducible patients, programmed ventricular stimulation was not continued after defibrillation because the patients declined treatment (patients 2, 10, 12) or because of unsterile catheters (patient 6). In the remaining nine inducible patients (32% of all inducible patients; patients 16 18, 21, 24, 28, 29, 35, 40), the inducibility of sustained ventricular arrhythmias was not reproducible. Treatment and clinical follow-up All patients with a history of cardiac arrest, and those with syncope, as well as those with no clinical event, but

Programmed electrical stimulation in Brugada syndrome 1399 inducible arrhythmias and/or a family history of sudden cardiac arrest were offered an ICD. The ICD was refused by six patients (patients 14, 19, 22, 26, 30, 35). During a mean follow-up of 29 18 months (symptomatic patients 39 23 months vs asymptomatic patients 21 10 months), four patients (10%) (patients 5, 6, 13, 17) experienced adequate ICD discharges due to spontaneous ventricular arrhythmias. The first episode of VT requiring ICD discharges occurred 4 months (patient 5), 3 months (patient 13), 16 months (patient 6), and 3 months (patient 18) after ICD implantation. All had previously been symptomatic with a history of cardiac arrest (n=3) or syncope (n=1, patient 18) and were inducible during programmed ventricular stimulation (except for patient 9). Patient 5 died suddenly 3 months after ICD implantation. Unfortunately, the ICD was not interrogated. None of the asymptomatic patients became symptomatic (syncope, sustained ventricular arrhythmia) within the mean follow-up of 21 10 months in this group. Interrogation of the ICD did not reveal non-sustained VT. The shorter follow-up is the result of a later recognition of the asymptomatic individuals during the course of this study. Discussion At present, the role of invasive electrophysiologic testing in patients with Brugada syndrome as well as in asymptomatic individuals with a similar ECG pattern is a matter of some controversy. In contrast to previous studies [2,4,5], the present study is the first to use a uniform stimulation protocol in a large group of individuals with an ECG pattern of atypical right bundle branch block and right precordial ST-segment elevation. The main findings are the following: (1) Of all individuals with an ECG compatible with Brugada syndrome, 68% had inducible sustained VT with the use of three premature beats at five different basic cycle lengths and two ventricular sites. If patients were only stimulated in the right ventricular apex inducibility rate decreased to 39%. If only two premature stimuli at two sites were used it also fell to 32%. (2) The mean coupling intervals of the premature stimuli inducing sustained ventricular arrhythmias were short, with 64% of individuals having inducible sustained tachyarrhythmias with coupling intervals shorter than 200 ms. Thus, 39% of all symptomatic and 45% of the asymptomatic patients would not have been inducible when the stimulation protocol was stopped at a coupling interval of 200 ms. All patients who experienced adequate ICD discharges during follow-up were only inducible with coupling intervals shorter than 200 ms. These data should be considered when programmed ventricular stimulation is used as a diagnostic criterion in patients with right bundle branch block and ST-elevation syndrome. Programmed ventricular stimulation in Brugada syndrome There is general agreement on the overall high rate of inducibility in patients with Brugada syndrome. Most previous studies [7] reported a high degree of inducibility ranging from 30% [8] to 93% [7] and even 100% [9]. Brugada et al. [2] and Priori et al. [5] recently reported multicentre data on 252 (116 symptomatic patients) and 65 patients (31 symptomatic patients) with right bundle branch block and ST-elevation, respectively. Because patients were included at different centres in both studies, the stimulation protocols varied from one centre to another. In the large series reported by Brugada et al. [2], the protocol included a minimum of two basic pacing cycle lengths and a minimum of two premature beats per basic cycle length up to refractoriness or induction of a sustained VT. Priori and coworkers [5] did not further specify the different protocols but a maximum of three premature beats was used. No coupling intervals or data on reproducibility were reported in either study. Brugada et al. [2] reported that 130 of 252 patients (52%) had inducible ventricular arrhythmias whereas Priori et al [5] demonstrated that in 26 (67%) of 39 investigated patients, sustained polymorphic ventricular arrhythmias were inducible (35% with three premature beats). Despite using an identical stimulation protocol with decremental stimulation to the point of refractoriness, nine (30%) out of 30 symptomatic patients (two patients with a history of cardiac arrest, seven with syncope, one with a SCN5A mutation) were not inducible in the present study. In addition, we found that 43% of the inducible patients were only inducible from the right ventricular outflow tract after initial stimulation from the apex had been unremarkable. Thus, the stimulation protocol markedly influences the degree of inducibility in patients with right bundle branch block and ST-elevation in the right precordial leads. Stimulation at other sites i.e. an epicardial site [10] was not investigated in the present study but may increase inducibility. At present, we do not know whether different stimulation protocols will also affect the practical recommendations for patients but our findings question the significance of previous multicentre studies and should have implications for further studies. In the initial study by Brugada and Brugada [1], four of eight patients (two with three extrastimuli) had inducible sustained polymorphic VT degenerating into VF. The initiating ventricular premature beat of the documented VT had a short coupling interval. The present data demonstrated that in the majority of inducible patients, coupling intervals shorter than 200 ms were required to induce sustained ventricular arrhythmias. This observation is in marked contrast to the worldwide experience with programmed ventricular stimulation in patients with structural heart disease. As these patients have an anatomical substrate for arrhythmogenesis, it is more likely to induce sustained arrhythmias with premature stimuli at longer coupling intervals. Thus, although it has previously not been shown, it is not surprising

1400 L. Eckardt et al. that short coupling intervals are required in Brugada syndrome patients with normal hearts. Spontaneous VT occurrence in Brugada syndrome The induction of sustained ventricular arrhythmias with short coupling intervals during the remote phase after an arrhythmic event in Brugada syndrome resembles the clinical situation. Viskin et al. [11] reported that spontaneous VF among patients with idiopathic VF has a very distinctive mode of onset. In 32 patients studied, a single premature ventricular complex with a short coupling interval initiated a polymorphic VT, which rapidly deteriorated into VF. Kakishita et al. [9] also demonstrated that premature, short coupled complexes occurred more frequently before the onset of VF in Brugada syndrome. However, the mean coupling interval of the first premature complex of VF reported by Kakishita et al. [9] was 388 ms which is, therefore, much longer than the coupling intervals required during programmed ventricular stimulation. This may indicate the importance of additional environmental factors, a site specificity of the premature ventricular complex and/or the increased vulnerability of the ventricle before spontaneous occurrence of VT as compared to the situation during the electrophysiologic investigation. The aetiology of premature complexes and the mechanism by which they precipitate arrhythmias in patients with Brugada syndrome has not yet been definitely clarified. The short coupled premature beat may meet the heart at a particular vulnerable phase, at a time when there is maximal dispersion of early repolarization. This may create unidirectional block and allow multiple reentrant waves. Antzelevitch et al. [12,13] reported that voltage gradients between sites of action potential dome and those where it is lost may give rise to phase-2- reentry and the generation of a very early ventricular complexes, which is then able to precipitate circus movement reentrant VT. Prognosis and treatment At present, there is much debate on the definition and management strategies in patients with ECG pattern compatible with Brugada syndrome. Brugada et al. [2] reported that the inducibility of ventricular tachyarrhythmias was an excellent predictor of arrhythmia recurrence during a mean follow-up of 34 40 months. In spite of not having any previous symptoms, about 5% of the individuals with a Brugada typical ECG presented with an arrhythmic event. Only three (one asymptomatic) non-inducible patients became symptomatic during follow-up. However, the asymptomatic patients had only a mean follow-up of 25 months with a large standard deviation of 27 months, and these patients were on average 5 years younger than the symptomatic patients (41 15 years vs 46 14 years, P<0 007). This may overestimate the negative predictive value of programmed ventricular stimulation in this study. In contrast, Priori et al. [5] found that electrophysiologic testing did not correctly identify patients at high or low risk for cardiac arrest during a comparable follow-up period (positive predictive value 50%, negative predictive value 46%). Because no arrhythmic event was observed in the group of asymptomatic patients it was concluded that in particular these asymptomatic individuals are at very low risk of sudden cardiac death [5]. Our limited experience with a follow-up period of 29 18 months supports data reported by Priori et al. [5]. We also identified six of 17 (35%) patients with syncope and two of 13 patients (15%) with previous resuscitation who were not inducible. During follow-up, there were only four patients with VF episodes. The majority of all patients (87%) and all asymptomatic patients were free of symptoms during follow-up. However, due to the later recognition of the asymptomatic individuals in the course of this study, which is similar to previous studies [2], the follow-up period of 21 10 months was shorter as compared to the symptomatic patients and is too short to reassure all asymptomatic patients. The lack of correlation between the inducibility of sustained ventricular arrhythmias during programmed ventricular stimulation and the clinical presentation at diagnosis does not necessarily exclude the possibility that the inducibility may correlate to future arrhythmic events. At present, we recommend ICD in all symptomatic patients for secondary prophylaxis of cardiac arrest irrespective of the results of programmed ventricular stimulation. For those who are asymptomatic, it is always an individualized decision that takes into account multiple parameters, such as sex, family history, ECG pattern (persistent or transient), late potentials [14], body surface potential mapping [15], genetic testing, but also the results of programmed ventricular stimulation. There is, of course, urgent need for practical guidelines but extension of the follow-up period to at least 10 years as suggested by Priori et al. [5] is probably required before drawing definitive conclusions about the risk of death and the prognostic significance of these parameters in patients with Brugada syndrome. References [1] Brugada P, Brugada J. Right bundle branch block, persistent ST segment elevation and sudden cardiac death: A distinct clinical and electrocardiographic syndrome. A multicenter report. J Am Coll Cardiol 1992; 20: 1391 6. [2] Brugada P, Geelen P, Brugada R, Mont L, Brugada J. Prognostic value of electrophysiologic investigations in Brugada syndrome. J Cardiovasc Electrophysiol 2001; 12: 1004 7. [3] Brugada J, Brugada R, Antzelevitch C, Towbin J, Nademanee K, Brugada P. Long-term follow-up of individuals with the electrocardiographic pattern of right bundle branch block and ST-segment elevation in precordial leads V1 to V3. Circulation 2002; 105: 73 8. [4] Brugada J, Brugada R, Brugada P. Right bundle-branch block and ST-segment elevation in leads V-1 through V-3: A

Programmed electrical stimulation in Brugada syndrome 1401 marker for sudden death in patients without demonstrable structural heart disease structural heart disease. Circulation 1998; 97: 457 60. [5] Priori SG, Napolitano C, Gasparini M et al. Clinical and genetic heterogeneity of right bundle branch block and ST- Segment elevation syndrome A prospective evaluation of 52 families. Circulation 2000; 102: 2509 15. [6] Eckardt L, Kirchhof P, Loh P et al. Brugada syndrome and supraventricular tachyarrhythmias: a novel association? J Cardiovasc Electrophysiol 2001; 12: 680 5. [7] Nademanee K, Veerakul G, Nimmannit S et al. Arrhythmogenic marker for the sudden unexplained death syndrome in Thai men. Circulation 1997; 96: 2595 600. [8] Kasanuki H, Ohnishi S, Ohtuka M et al. Idiopathic ventricular fibrillation induced with vagal activity in patients without obvious heart disease. Circulation 1997; 95: 2277 85. [9] Kakishita M, Kurita T, Matsuo K et al. Mode of onset of ventricular fibrillation in patients with Brugada syndrome detected by implantable cardioverter defibrillator therapy. J Am Coll Cardiol 2000; 36: 1646 53. [10] Carlsson J, Erdogan A, Schulte B, Neuzner J, Pitschner HF. Possible role of epicardial left ventricular programmed stimulation in Brugada syndrome. Pacing Clin Electrophysiol 2001; 24: 247 9. [11] Viskin S, Lesh MD, Eldar M et al. Mode of onset of malignant ventricular arrhythmias in idiopathic ventricular fibrillation. J Cardiovasc Electrophysiol 1997; 8: 1115 20. [12] Antzelevitch C. The Brugada Syndrome. J Cardiovasc Electrophysiol 1998; 9: 513 16. [13] Yan GX, Antzelevitch C. Cellular basis for the Brugada syndrome and other mechanisms of arrhythmogenesis associated with ST-segment elevation. Circulation 1999; 100: 1660 6. [14] Eckardt L, Bruns H, Kirchhof P et al. Body surface area of ST-elevation and presence of late potentials correlate to the inducibility of ventricular tachyarrhythmias in Brugada syndrome. J Cardiovasc Electrophysiol, (in press). [15] Bruns J, Eckardt L, Vahlhaus C et al. Body surface potential mapping in patients with Brugada syndrome: right precordial ST segment variations and reverse changes in left precordial leads. Cardiovasc Res 2002; 54: 58 66.