Author + information
- Received September 28, 2015
- Revision received April 17, 2016
- Accepted May 3, 2016
- Published online August 9, 2016.
- Ruben Casado-Arroyo, MD, PhDa,b,∗ (, )
- Paola Berne, MDc,
- Jayakeerthi Yoganarasimha Rao, MDa,
- Moisés Rodriguez-Mañero, MDa,
- Moisés Levinstein, MDa,
- Giulio Conte, MD, PhDa,
- Juan Sieira, MD, PhDa,b,
- Mehdi Namdar, MD, PhDa,
- Danilo Ricciardi, MDa,
- Gian-Battista Chierchia, MD, PhDa,
- Carlo de Asmundis, MD, PhDa,
- Gudrun Pappaert, RNa,
- Mark La Meir, MD, PhDa,
- Francis Wellens, MD, PhDa,
- Josep Brugada, MD, PhDc and
- Pedro Brugada, MD, PhDa
- aHeart Rhythm Management Center, Cardiovascular Division, UZ Brussel–Vrije Universiteit Brussel, Brussels, Belgium
- bCardiology Department, Arrhythmia Section, Erasmus Hospital, Université Libre de Bruxelles, Brussels, Belgium
- cCardiology Department, Arrhythmia Section, Thorax Institute, Hospital Clinic, University of Barcelona, Barcelona, Catalonia, Spain
- ↵∗Reprint requests and correspondence:
Dr. Ruben Casado-Arroyo, Electrophysiology, Cardiology Department, Erasmus Hospital, Université Libre de Bruxelles, Route de Lennik 808, 1070 Brussels, Belgium.
Background A proband of Brugada syndrome (BrS) is the first patient diagnosed in a family. There are no data regarding this specific, high-risk population.
Objectives This study sought to investigate the Brugada probands diagnosed from 1986 through the next 28 years.
Methods We included 447 probands belonging to families with a diagnostic type 1 electrocardiogram Brugada pattern. The database was divided into 2 periods: the first period identified patients who were part of the initial cohort that became the consensus document on BrS in 2002 (early group); the second period reflected patients first diagnosed from 2003 to January 2014 (latter group).
Results There were 165 probands in the early group and 282 in the latter group. Aborted sudden death as the first manifestation of the disease occurred in 12.1% of the early group versus 4.6% of the latter group (p = 0.005). Inducibility during programmed electrical stimulation was achieved in 34.4% and 19.2% of patients, respectively (p < 0.001). A spontaneous type 1 electrocardiogram pattern at diagnosis was present in 50.3% early versus 26.2% latter patients (p = 0.0002). Early group patients had a higher probability of a recurrent arrhythmia during follow-up (19%) than those of the latter group (5%) (p = 0.007). The clinical suspicion and use of a sodium-channel blocker to unmask BrS has allowed earlier diagnoses in many patients.
Conclusions Since being first described, the presentation of BrS has changed. There has been a decrease in aborted sudden cardiac death as the first manifestation of the disease among patients who were more recently diagnosed. These variations in initial presentation have important clinical consequences. In this setting, the value of inducibility to stratify individuals with BrS has changed.
Brugada syndrome (BrS) is an inherited cardiac arrhythmic disorder characterized by atypical electrocardiographic (ECG) patterns and an elevated risk of sudden cardiac death (SCD) (1,2). The disease is associated with mutations in the cardiac sodium channel gene SCN5A in 21% of patients (3,4). Patients with BrS can experience SCD due to rapid ventricular arrhythmias (generally polymorphic) and can endure other arrhythmias, such as atrial fibrillation (AF), atrial flutter, and other supraventricular arrhythmias (1–7). Likewise, syncope can be the first and only manifestation of the disease. The disease is characterized by an ECG with a coved-type ST-segment elevation in the right precordial leads (V1 to V3) (1), but variations of this classic pattern have been described by our group (8).
Several clinical variables have been demonstrated to predict a worse outcome in patients with BrS. The occurrence of syncope, a spontaneous type 1 ECG, inducibility of ventricular arrhythmias during programmed electrical stimulation, and male sex have been shown in some studies to be related to the occurrence of cardiac events during follow-up (5–11), although debate regarding these predictors continues.
No study has assessed in detail the characteristics of the first member (proband) diagnosed from each family with BrS. We sought to evaluate the change of their characteristics over time, and we hypothesized that the presentation of BrS has experienced an important change since the first consensus criteria.
A “proband” was defined as the first patient diagnosed with BrS in a family on the basis of a type 1 Brugada ECG pattern. Since the first proband identified with BrS in 1986 by our group, all probands and family members have been prospectively included in a dual-center registry at the Free University of Brussels (UZ Brussel-VUB, Belgium) and at the University of Barcelona (Hospital Clinic, Spain) and followed in a prospective fashion. All patients included gave informed consent to participate in the registry. The ethics committee of the 2 centers approved the study protocol. For the purpose of this study, the database was assessed in January 2014.
A proband was included only if spontaneous or class I antiarrhythmic drug (ADD)-induced coved type I ≥ 2 mm ST-segment elevation in ≥1 lead from V1 to V3 was documented. A detailed family history was taken from each proband. Arrhythmic events, AF, and device implantation and complications related to this therapy during follow-up were recorded in the database.
The database was divided into 2 equivalent periods in accordance with the criteria for BrS in 2002 (2). The first period identified patients who were part of the initial study that became the consensus document on BrS in 2002 (early group). The second period reflected patients from 2003 until January 2014 (latter group).
The data recorded in the database were: 1) documentation of spontaneous or drug-induced ≥2-mm coved (type 1) Brugada ECG pattern; 2) personal and family history of aborted SCD (12); 3) history of syncope (at any age) or other symptoms; 4) sustained ventricular arrhythmia induced during the electrophysiological study (EPS); 5) arrhythmic event during follow-up and/or SCD; 6) implantable cardioverter-defibrillator (ICD) implantation and/or follow up; and 7) arrhythmic episodes, syncope, or appropriate or inappropriate shocks during the follow-up period.
The primary endpoint of the study was to identify the probability of first arrhythmic event at follow-up (defined as appropriate ICD shock for ventricular fibrillation or ventricular tachycardia or SCD) for the entire population of probands and for different subgroups.
Definitions and testing
A type 1 Brugada ECG pattern was diagnosed if a coved-type ST-segment elevation of ≥2 mm followed by a negative T-wave was documented in >1 lead from V1 to V3 either spontaneously or after class I AAD administration (Figure 1) (2). If the ST-segment elevation was coved type, patients with negative/isoelectric or positive T waves were included and were classified as having a type 1 ECG pattern (7).
A class I AAD test was performed to unmask the diagnostic ECG pattern. Most frequently, 1 mg/kg of intravenous ajmaline administered within 5 min was used; less often, 2 mg/kg flecainide or 10 mg/kg procainamide given over a 10-min period were used for this purpose. The test was performed only in case of clinical suspicion of BrS: nocturnal agonal respiration, seizures, palpitations, pre-syncope, or chest discomfort, in addition to an ECG that was suggestive of BrS but does not fulfil the diagnostic criteria (e.g., type 2 Brugada syndrome). The test was considered positive only if a coved type 1 ECG was documented.
An EPS was performed only from the right ventricular apex and included 3 basic cycle lengths (600, 500, and 430 ms), with a minimum coupling interval limited to 200 ms and a maximum of 3 extra stimuli. The EPS result was only considered positive if a sustained ventricular arrhythmia (lasting ≥30 s, accompanied by syncope or requiring intervention for termination) was induced.
Continuous variables are expressed as mean ± SD or median (interquartile range). Statistical differences were calculated using the chi-square test for discrete variables and the Student t test for continuous variables (expressed as mean values ± SD).
The cumulative probability of first arrhythmic event at follow-up (defined as appropriate ICD shock for ventricular fibrillation or ventricular tachycardia or SCD) was determined using the Kaplan-Meier method for the entire population and for different subgroups with and without adjusting for longer follow-up. Cox multivariate survival analysis was applied to evaluate the possible survival predictors (SCD, inducibility, syncope, and spontaneous type 1 ECG) of an arrhythmic event. Given the difference in exposure time, the statistical analysis was performed by restricting the analysis to the first 4 years in both arms. A variable time has been created in the Cox model to explore if the differences were due to baseline characteristics or were autonomous.
A value of p ≤ 0.05 was considered statistically significant. All tests were 2-sided. All statistics were performed with the use of the SPSS Statistics, version 22.0 (IBM Corp., Armonk, New York) and STATA software, version 11.1 (StataCorp, College Station, Texas).
The study included all 447 probands in our database, all of whom fulfilled the diagnosis consensus criteria. The diagnosis of BrS has increased significantly over time since its first description in 1992 (Table 1). Mean follow-up was 50 ± 51 months (range 3 to 377 months).
Population characteristics have changed over time. In the early group, probands were significantly more symptomatic and more likely to present with an episode of SCD before a diagnosis of BrS than in the latter group. A family history of SCD was similar in both groups. Syncope prevalence was also similar between both groups. Significantly more patients had a history of AF before diagnosis in the earlier group than the latter group (Table 1). A spontaneous Brugada pattern on ECG at diagnosis occurred more frequently in the earlier group. Sodium-channel blockers were used less frequently in asymptomatic patients in the earlier period (50% [51 of 102]) vs. 74% (136 of 183) in the latter group (p <0.001).
Inducibility during EPS was different in the 2 groups. In the earlier group, 34.8% (54 of 155) of the patients were inducible compared with only 19.2% (50 of 260) in the latter group (p = 0.001). The rate of ICD implantation was higher in the earlier group, but the difference was not statistically significant.
The percentage of patients presenting with cardiac arrest in the earlier group who had an EPS performed was 90% (18 of 20) versus 54% (7 of 13) in the latter group (p = 0.07). Two patients with aborted cardiac arrest were implanted with an ICD without EPS in the earlier group; the total rose to 8 patients in the latter group.
There were 42 events (9.4%) in total: 30 of 165 (18.2%) in the earlier group and 12 of 287 (4.1%) in the latter group (Table 2). Patients in the earlier group had a significantly higher probability of presenting with an event during follow-up compared with those in the latter group, even after adjusting for the longer follow-up in the earlier group (log-rank p = 0.007) (Figure 2). The annual event rate was higher in the earlier group (2.5 vs. 1.8; p < 0.001) (Table 2). When the events were analyzed in relation to inducibility, the number of events during follow up was ∼5× higher in the inducible patients: 21.1% (22 of 104) compared with 4.2% (13 of 311) for the noninducible group.
By univariate and multivariate analysis, a previous episode of SCD and inducibility of a sustained ventricular arrhythmia during EPS were predictors of events across the entire population (Tables 3 and 4). The hazard rate (close to 1) of the variable time shows that the differences between both groups are explained by differences in baseline characteristics. When analyzing only the earlier group, previous SCD and inducibility were predictors of events, whereas previous SCD was the only predictor of events in the latter group. If we analyzed patients without SCD, inducibility was a consistent predictor of events in the whole population, as well as in the earlier group; however, inducibility no longer predicted spontaneous arrhythmic events in the latter group (Tables 3 and 4).
When only the first group was considered, 35% (19 of 54) of the inducible patients presented with events versus 8% (8 of 101) of the noninducible patients of the same group. In the latter group, 6% (3 of 50) of the inducible patients versus 2.4% (5 of 210) of the noninducible group experienced events. The number of events of asymptomatic patients at the time of diagnosis and drug-induced type I ECG was 3 out of 51 patients (5.7%) in the earlier group and 2 out of 136 (1.5%) in the latter group, for a follow- up of 7.6 and 2.93 years, respectively. Figure 5 presents the analysis of the asymptomatic probands (spontaneous versus induced-type 1 Brugada pattern) and also the predictive value of the EPS of this subpopulation.
AF during follow-up was documented more often in patients in the earlier group, 9.1% (15 of 165) versus 3.2% (9 of 282) in the latter group, although the difference was not statistically significant (Table 2).
This study evaluated the evolution of the characteristics of BS probands over time. The study confirmed the hypothesis of a changing clinical presentation of patients with BrS. As a result, there is a clear shift toward milder forms of the syndrome due to a better identification of patients.
The probands in the earlier group were significantly more symptomatic than those in the latter group at presentation, with a higher prevalence of SCD as the first presenting symptom, spontaneous type I ECG pattern at baseline, and more often inducible during EPS (Central Illustration). It should be noted that the rate of annual events of the latter group is only moderately lower than the earlier group.
These variations in clinical presentation have important consequences. Newly diagnosed probands and family members have a more benign profile than patients diagnosed more than 25 years ago, probably due to the fact that we are starting a new era of the disease, with patients who are greater in number but who are less symptomatic. Also, an earlier diagnosis (clinical suspicion) and risk stratification can help the clinician prevent events during follow-up. More studies are needed to confirm these data.
Spontaneous type 1 ECG at diagnosis was not a predictor of arrhythmic events at follow-up in this series. This result can be explained due to the fact that the Brugada pattern is dynamic over time. It is variable and sensitive to emotions and other external influences (e.g., a meal, fever, changes in heart rate from any cause, drugs, exercise, and even body position). Fever aggravates the coved-type ST-segment elevation in patients with BrS and, in parallel, the risk for ventricular arrhythmias (8).
EPS was a predictor of overall events in the early but not in the latter group. A factor that has likely affected the predictive value of the EPS was the population tested: 54% of the patients who presented with SCD in the latter group decided to undergo an ICD implantation without EPS versus 10% in the early group. With the low incidence of events published in other series (9–11,13–16), tests like EPS may no longer have a predictive value because the pre-test probability of a severe outcome (SCD) might not be sufficiently high to allow recognition of significant differences. Regarding the value of EPS in asymptomatic patients, Figure 5 showed that if we segregated the database in different subgroups, the result was a loss of statistical power. With such a low rate of events, we would need longer follow-up to find any significant difference. A recently published meta-analysis (of 3,536 asymptomatic patients with BrS from 14 studies) demonstrated that inducibility during EPS predicted events during follow up (odds ratio: 3.51; 95% confidence interval: 1.60 to 7.67; p = 0.002) (16). Another publication from our group has shown that EPS also was predictive of events in a multivariate model in asymptomatic patients (17). The study included 363 asymptomatic patients. Meanwhile, the present study had 100 fewer asymptomatic patients. Other groups used EPS-guided class 1A AADs for all types of patients with BrS to stratify risk due to the high negative predictive value of EPS (18).
As shown in Table 5, with longer follow-up, the number of arrhythmic events increased independent of the study’s design. It should be noted that even with the different risk stratification approaches (with or without EPS), the percentage of ICDs implanted was almost the same in the largest registries published. The rate of ICD implantation was 47.9% in the earlier and 44% in the latter group. A similar percentage (44.5%) was shown in the PRELUDE (Programmed Electrical Stimulation Predictive Value) study, with an arrhythmic event rate of 4.5% (10), as well as in the FINGER (France, Italy, the Netherlands, Germany) registry (11), where the ICD implantation rate was 42.1% (arrhythmic event rate: 5%) (Table 5). It should be noted that the percentage of arrhythmic events in our study was higher: 19% in the earlier and 5% in the latter group (9.4% overall).
Most of the studies have shown that when a patient is inducible, the number of events during follow-up is much higher than those of noninducible patients: 5.7% versus 2.5% (overall population), 3% versus 1% (asymptomatic patients) in the FINGER registry (11), 15.2% versus 0% in Giustetto et al. (9), 35.2% (19 of 54) compared with 8% (8 of 101) in the earlier group in our study, and 6% (3 of 50) versus 2.4% (5 of 210) in the latter group in the present study.
In this setting, the indication for ICD implantation should be made on the basis of all clinical information about the patient and not solely on 1 isolated clinical variable or test. Due to the relatively young age in patients with BrS and the absence of macrostructural abnormalities, the added years of quality life gained by preventing SCD in patients with BrS are greater than in patients with structural heart disease. Nowadays, improvements in device programming (such as higher detect heart rates or longer episode durations) have been shown to be effective in reducing inappropriate shocks as confirmed by our results. The ventricular fibrillation detection rate has changed from 180 beats/min in the earlier group to >200 beats/min in the latter period. A monitor zone and long-detection intervals (30 of 40 intervals) have been adopted recently, which was not the case in the early periods.
The follow-up of a patient with BrS is lifelong and is not limited to a period of time or to the publication of a study. In this study, we have shown that the number of arrhythmic events has increased over time and has occurred in up to 9.4% of the population.
The risk of an arrhythmic event will be life-long, too, and for that reason we will need studies with even longer follow-up. Currently, the only proven therapy to prevent SCD in patients with BrS is ICD implantation, although this intervention is not without complications.
The discrepancies between the results obtained by different groups might be related to differences in inclusion criteria and hence, different groups of patients to analyze, and even differences in the test performed to unmask BrS. There were also potential conflicts in the interpretation of the test, differences in the protocol of EPS used, and a selection bias toward patients who were more severely affected in some registries or less affected in others. However, many different mutations that alter ion channel functions have been identified in patients with BrS, but in most cases the mutation remains unknown, as has been reported by our group and others (13,15).
This study included all probands of BrS since the first diagnosed patient more than 28 years ago. However, the follow-up period was still too short to make definite conclusions regarding the management of individuals with Brugada ECG. These individuals carry the disease all of their life. It is conceivable that prognosis might worsen as follow-up progresses. This may explain the increased number of arrhythmic events (9.4%) in the cohort. Also, the differences in SCD as first manifestation could be due to different methods used for diagnosis in both periods. This was particularly significant if we consider that earlier diagnostic regimens did not include the use of sodium-channel blockers.
The presentation of BrS has witnessed an intriguing shift since its first description. These variations in the initial presentation have important clinical consequences. More recently diagnosed patients seem to have a more favorable profile. However, follow-up of all patients until death is necessary to draw correct conclusions. That may explain why in other series, some tests (like EPS) may not have had predictive value because the pre-test probability of a severe outcome (like SCD) may not have been sufficiently high as to allow recognition of significant differences.
COMPETENCY IN MEDICAL KNOWLEDGE: BrS is a genetically heterogeneous disease, and its clinical presentation has changed over the last 30 years. Compared to an earlier cohort, patients identified in the more recent period have less inducible ventricular arrhythmias during programmed electrical stimulation and a lower probability of recurrent arrhythmia during clinical follow-up.
TRANSLATIONAL OUTLOOK: Further research is needed to reassess predictors of arrhythmic events in patients with BrS.
This study was funded by a post-graduate grant for international research, Spanish Society of Cardiology and Horlait-Dapsens Foundation (to Dr. Casado-Arroyo). Dr. Brugada has received research grants from Biotronik, Boston Scientific, Medtronic, St. Jude Medical, and Sorin; and has served as a consultant for Biotronik. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- antiarrhythmia drug
- atrial fibrillation
- electrophysiological study
- implantable cardioverter-defibrillator
- sudden cardiac death
- Received September 28, 2015.
- Revision received April 17, 2016.
- Accepted May 3, 2016.
- American College of Cardiology Foundation
- Brugada P.,
- Brugada J.
- Wilde A.A.,
- Antzelevitch C.,
- Borggrefe M.,
- et al.,
- for the Study Group on the Molecular Basis of Arrhythmias of the European Society of Cardiology
- Brugada J.,
- Brugada R.,
- Antzelevitch C.,
- et al.
- Brugada J.,
- Brugada R.,
- Brugada P.
- Antzelevitch C.,
- Brugada P.,
- Borggrefe M.,
- et al.
- Sarkozy A.,
- Boussy T.,
- Kourgiannides G.,
- et al.
- Richter S.,
- Sarkozy A.,
- Paparella G.,
- et al.
- Giustetto C.,
- Drago S.,
- Demarchi P.G.,
- et al.
- Priori S.G.,
- Gasparini M.,
- Napolitano C.,
- et al.
- Probst V.,
- Veltmann C.,
- Eckardt L.,
- et al.
- Myerburg R.J.,
- Castellanos A.
- Delise P.,
- Allocca G.,
- Marras E.,
- et al.
- Eckardt L.,
- Probst V.,
- Smits J.P.,
- et al.
- Benito B.,
- Sarkozy A.,
- Mont L.,
- et al.
- Sieira J.,
- Ciconte G.,
- Conte G.,
- et al.
- Belhassen B.,
- Rahkovich M.,
- Michowitz Y.,
- et al.