Abstract
Aims
To investigate the prognostic significance of heterogeneity in the refractoriness of right ventricular (RV) outflow tract (RVOT) and RV apex at the electrophysiological study (EPS) in Brugada syndrome (BrS).
Methods and results
A cohort of BrS patients (primary prevention) from five Italian centres was retrospectively analysed. Patients with spontaneous or drug-induced Type-1 electrocardiogram (ECG) + symptoms were offered an EPS for prognostic stratification. The primary endpoint was a composite of sudden cardiac death (SCD), resuscitated cardiac arrest, or appropriate intervention by the implantable cardioverter-defibrillator (ICD). Three hundred and seventy-two patients with BrS were evaluated (44 ± 15 years, 69% males, 23% with ICD): 4 SCDs and 17 ICD interventions occurred at follow-up (median 48, interquartile range: 36–60 months). Family history of SCD, syncope, and a spontaneous Type-1 ECG pattern were univariate predictors of the primary endpoint in the whole population. In patients undergoing EPS (n = 198, 53%, 44 ± 12 years, 71% males, 39% with ICD), 3 SCD and 15 ICD interventions occurred at follow-up. In this subset, the primary endpoint was not only predicted by ventricular tachycardia/fibrillation inducibility but also by a difference in the refractory period between RVOT and RV apex (ΔRPRVOT-apex) >60 ms. ΔRPRVOT-apex > 60 ms remained an independent predictor of SCD/ICD shock at bivariate analysis, even when adjusted for the other univariate predictors, showing the highest predictive power at C-statistic analysis (0.75, 95% confidence interval 0.63–0.86).
Conclusions
Heterogeneity of RV refractory periods is a strong, independent predictor of life-threatening arrhythmias in BrS patients, beyond VT/VF inducibility at EPS and common clinical predictors.
Keywords: Brugada syndrome, Sudden cardiac death, Electrophysiological study, Refractory period
Graphical Abstract
Graphical Abstract.
In patients with Type-1 ECG Brugada pattern, the presence of an increased difference in the refractoriness between the right ventricular outflow tract (RVOT) and the apex, mainly driven by an increased refractory period in the RVOT, increases the prognostic discrimination power of the electrophysiological study, currently based only on inducibility of life-threatening arrhythmias. BrS, Brugada syndrome; EPS, electrophysiological study; RP, refractory period; RV, right ventricle; RVOT, right ventricular outflow tract; VF, ventricular fibrillation; VT, ventricular tachycardia.
What’s new?
The heterogeneity of right ventricular (RV) refractory periods was assessed at the electrophysiological study (EPS) in a large multicentric Italian cohort of patients with Brugada syndrome (BrS) followed-up for a median of 42 months.
A difference in the refractory period between RV outflow tract (RVOT) and RV apex (ΔRPRVOT-apex) >60 ms improved risk stratification of patients with BrS beyond ventricular tachycardia/ventricular fibrillation inducibility at EPS and common clinical predictors.
Prospective trials in larger cohorts should evaluate the value of heterogeneity of RV refractoriness at EPS to test in the future the potential of this novel measure in the decision-making of patients with BrS.
Introduction
Indications for implantable cardioverter-defibrillator (ICD) implantation in primary prevention are still debated in patients with Brugada syndrome (BrS).1–3 Several models combining clinical factors and ventricular arrhythmias inducibility at the electrophysiological study (EPS) have been proposed to improve prognostic stratification in BrS patients.4,5
As recently reviewed, slow discontinuous conduction and dispersion of repolarization in the right ventricle (RV) have been described as specific electrophysiological features and potential arrhythmic substrates of BrS.6
Beyond the inducibility of sustained ventricular arrhythmias, we hypothesized that EPS may increase its prognostic predictive power in BrS by providing information related to the heterogeneity of RV refractoriness.
The present study aims at assessing the clinical usefulness of a combined approach including clinical data and EPS in patients with either spontaneous or drug-induced Brugada Type-1 electrocardiogram (ECG) pattern.
Methods
This was a retrospective multicentric study (Pisa, Arezzo, Lido di Camaiore, Siena, and Naples) on patients with a diagnosis of BrS, without a history of cardiac arrest or sustained ventricular tachycardia (VT) or ventricular fibrillation (VF). Any underlying structural or functional cardiac abnormality was excluded. The study was approved by Institutional Ethics Committee of each participant centre.
Brugada Type-1 ECG pattern was defined as the presence of coved ST-segment elevation ≥0.2 mV in ≥1 lead from standard/modified V1–V3 assessed in basal conditions, during drug challenge, or at 12-lead 24-h Holter monitoring.1
Genetic testing limited to sodium voltage-gated channel alpha subunit 5 (SCN5A) variants was recommended to all patients and only pathognomonic variants were collected.
Family history of BrS (first- or second-degree relatives)7 or juvenile sudden cardiac death (SCD) before the age of 40 years,5 and history of syncope, palpitations, or nocturnal agonic breathing were collected.8 Syncope was anamnestically defined by referring cardiologists after exclusion of signs suggesting a vasovagal origin. ‘Symptoms’ were defined as the presence of syncope and/or palpitations, and/or nocturnal agonic breathing.
Electrophysiological study was proposed to all individuals with spontaneous Brugada Type-1 ECG pattern, regardless of symptoms, and to symptomatic subjects with drug-induced Brugada Type-1 ECG pattern. As for ‘symptomatic patients’, we considered those with a high suspicion of arrhythmic events based on anamnestic evaluation.
An ICD was implanted in patients with syncope and spontaneous Brugada Type-1 ECG pattern or with a positive EPS study according to recommendations of the second consensus BrS conference.1,9
Electrophysiological study
Electrophysiological study included basal measurements of conduction intervals, ventricular programmed stimulation (VPS), and collection of ventricular effective refractory periods (VERP) at two sites in the RV, namely the apex and the RV outflow tract (RVOT).
In all centres, VPS was performed with a drive of 8–10 beats (basic cycle lengths of 600 and 400 ms) and up to two extrastimuli anticipated in 10 ms decrements up to the shortest coupling interval that resulted in ventricular capture from the RV apex and the RVOT.8,10
Electrophysiological study was assumed as ‘positive’ in case of sustained VT or VF induction (duration up to 30 s), with clinical symptoms or collapse, or requiring direct current-shock. Ventricular effective refractory periods were defined as the longest interval between two stimulated beats that failed to achieve ventricular capture. Ventricular effective refractory period values were collected using a drive of 600 ms and a single extrastimulus from the RV apex and RVOT. In all centres, programmed electrical stimulation from the RV apex and RVOT was performed at twice the diastolic threshold strength and a pulse of 2 ms duration.
Study follow-up and clinical endpoints
All patients were followed through in-office or remote follow-up visits every 12 months or in case of symptoms or ICD interventions until March 2021. The primary endpoint was a composite of SCD, resuscitated cardiac arrest, or appropriate ICD therapies for sustained VT or VF (antytachicardia algorithms or endocardial shocks). A secondary endpoint including inappropriate ICD therapies or related complications was also evaluated.
Statistical analysis
Statistical analysis was performed using SPSS (version 25.0, IBM Statistics) and R statistical software (version 3.4.0). A two-tailed P-value ≤0.05 was considered significant.
Quantitative data were expressed as mean ± standard deviation (SD) or median and interquartile range (IR) according to distribution (Shapiro–Wilk test) and compared by independent t-test or Mann–Whitney U test. Discrete variables were expressed as numbers and/or percentages and compared by χ2 or Fisher exact tests.
Kaplan–Meier method was performed to build the survival curves with log-rank statistic used for comparison, while Cox proportional hazard regression model (univariate and bivariate analysis) was used for event analysis over time, with the risk quantified as a hazard ratio (HR) with 95% confidence interval (CI). The list of risk factors included in survival analysis, in line with the literature, includes: age at diagnosis, male sex, family history of SCD, presence of spontaneous Type-1 ECG, symptoms (syncope/agonic nocturnal breathing), presence of a pathogenic SCN5A mutation, VT/VF at EPS, and a metric of heterogeneity in endocardial VERP between the apex and the RVOT at EPS (as explained below). Considering the number of events in our population, we limited multivariate analysis to a bivariate analysis to reduce the risk of overfitting models.
Cubic spline interpolation was used for an exploratory assessment of the prognostic relationship between a covariate and the hazard risk. The optimal prognostic threshold was then chosen by using the maximally selected rank statistics.
For event prediction, C-statistics analysis was used (ranges from 0.5 as no discrimination to 1.0 as maximum discrimination ability). The Integrated Brier score (IBS) was employed to evaluate the accuracy of a predicted survival function (ranges between 0.0 and 1.0, with lower values being better and 0.0 being the best possible value).
Results
A cohort of 372 patients with Type-1 ECG pattern was finally evaluated (Pisa n = 111, Arezzo n = 52, Lido di Camaiore n = 165, Siena n = 21, and Naples n = 23). The clinical characteristics of the population are reported in Table 1.
Table 1.
Clinical characteristics of patients with Brugada syndrome
| All patients (n = 372) | No EPS (n = 174) | EPS (n = 198) | P-value | |
|---|---|---|---|---|
| Age at diagnosis, years | 44 ± 15 | 44 ± 17 | 43 ± 13 | 0.999 |
| Males | 257 (69) | 117 (67) | 140 (71) | 0.501 |
| Family history of Brugada syndrome | 98 (26) | 59 (34) | 39 (20) | 0.002 |
| Family history of SCD | 105 (28) | 44 (25) | 61 (31) | 0.250 |
| ȃSpontaneous Type-1 ECG | 185 (49) | 66 (38) | 119 (60) | <0.001 |
| Symptoms | 170 (46) | 68 (39) | 102 (52) | 0.017 |
| ȃSyncope | 94 (25) | 36 (21) | 58 (29) | 0.057 |
| ȃPresyncope | 51 (14) | 24 (14) | 27 (14) | 0.965 |
| ȃPalpitations | 87 (23) | 31 (18) | 56 (28) | 0.017 |
| ȃAgonic nocturnal breathing | 10 (3) | 3 (1.7) | 7 (3.5) | 0.310 |
| Genetic testing performed | 175 (47) | 72 (41) | 103 (52) | 0.048 |
| ȃSCN5A mutationa | 55 (31) | 24 (33) | 31 (30) | 0.502 |
| Cardiovascular risk factors | ||||
| ȃHypertension | 64 (17) | 31 (18) | 33 (17) | 0.785 |
| ȃDiabetes | 12 (3) | 7 (4) | 5 (3) | 0.559 |
| ȃDyslipidaemia | 35 (9) | 11 (6) | 24 (12) | 0.074 |
| ȃSmoke | 59 (16) | 20 (12) | 39 (20) | 0.033 |
Data are given as mean ± standard deviation or number (percentage). ECG, electrocardiogram; SCD, sudden cardiac death; SCN5A, sodium voltage-gated channel alpha subunit 5.
% are related to patients with a genetic sampling.
Patients (age at diagnosis 44 ± 15 years) were mainly males (69%). A family history of BrS and SCD was present in 26 and 28% of patients, respectively; symptoms and syncope were present in 46 and 25% of cases, respectively. Patients had spontaneous and drug-induced Type-1 ECG pattern in 49 and 51% of cases, respectively. A total of 167 genetic samples (45%) were collected, finding 55 (33% of those evaluated) SCN5A pathognomonic variants.
Electrophysiological study
Electrophysiological study was proposed to 212 (57% of the whole population) BrS subjects; of them, seven patients were excluded for lack of consent and seven for incompleteness of the VERP assessment due to programmed ventricular stimulation performed only at the RV apex in four patients, unavailability of VERP for both sites of pacing for VF induction with single extrastimulus either at the apex (two patients) or the RVOT (one patient). Data from EPS were finally analysed in 198 patients (53% of the whole population) with VERP available for both RV pacing sites.
Patients studied with EPS compared with those who did not (Table 1) had more frequently a spontaneous Type-1 ECG pattern (73 vs. 56%, P < 0.001), a family history of SCD (42 vs. 27%, P = 0.036) and a history of syncope (42 vs. 25%, P = 0.041). As expected, patients who underwent EPS were more likely to be implanted with an ICD (39 vs. 6%, P < 0.001).
Out of 198 patients, 52 (27%) had a positive EPS (Table 2): all patients but one (regular VT) had VF inducible with two extrastimuli. Overall, VERP assessed at the RV apex and RVOT were 220 ± 39 and 244 ± 44, respectively. Ventricular effective refractory periods assessed at the RV apex was shorter, while that measured at the RVOT was longer in VT/VF inducible patients compared with non-inducible patients (VERP RV apex 210 ± 17 vs. 223 ± 43 ms, P = 0.034; VERP RVOT 259 ± 36 vs. 239 ± 45 ms, P = 0.005, respectively) (Table 2). As for the difference between RVOT and RV apex (ΔRPRVOT-apex), a longer ΔRPRVOT-apex mean values were observed in inducible compared with non-inducible patients [40 (10–80) vs. 10 (0–20) ms, P < 0.001]. Furthermore, a ΔRPRVOT-apex > 60 ms was more likely in patients with positive EPS (45 vs. 8%, P < 0.001). Patients with a positive EPS also showed a higher prevalence of appropriate ICD shocks (15 vs. 5%, P = 0.025). Out of 53 inducible patients, 20 (38%) had inducibility at the apex and 33 (62%) at the RVOT (P = 0.002). With regard to the induction site, no differences were found between patients with ΔRPRVOT-apex > 60 ms or ≤60 ms (P = 0.31).
Table 2.
Patients’ characteristics according to a negative or positive electrophysiological study
| EPS− (n = 146) | EPS+ (n = 52) | P-value | |
|---|---|---|---|
| Age, years | 43 ± 13 | 45 ± 12 | 0.360 |
| Males | 99 (68) | 41 (79) | 0.133 |
| Family history of Brugada syndrome | 31 (21) | 8 (15) | 0.363 |
| Family history of SCD | 39 (27) | 22 (42) | 0.036 |
| Spontaneous Type-1 ECG | 81 (56) | 38 (73) | 0.039 |
| Symptoms | 75 (51) | 27 (52) | 0.945 |
| ȃSyncope | 37 (25) | 21 (42) | 0.041 |
| ȃPresyncope | 22 (15) | 5 (10) | 0.325 |
| ȃPalpitations | 46 (32) | 10 (19) | 0.091 |
| ȃAgonic nocturnal breathing | 4 (3) | 3 (6) | 0.382 |
| Genetic testing performeda | 75 (51) | 28 (54) | 0.759 |
| ȃSCN5A mutation | 22 (30) | 9 (32) | 0.845 |
| VERP RV apex, ms | 223 ± 43 | 210 ± 17 | 0.034 |
| VERP RVOT, ms | 239 ± 45 | 259 ± 36 | 0.005 |
| ȃΔRPRVOT-RVapex, ms | 10 (0–20) | 40 (10–80) | <0.001 |
| ȃΔRPRVOT-RVapex > 60 ms | 11 (8) | 23 (45) | <0.001 |
| SCD | 0 (0) | 3 (6) | 0.017 |
| ICD implanted | 36 (24) | 42 (81) | <0.001 |
| ȃAppropriate ICD shock | 7 (5) | 8 (15) | 0.025 |
| ȃInappropriate ICD shock | 3 (9) | 4 (10) | 0.865 |
Data are given as mean ± standard deviation or number (percentage). ECG, electrocardiogram; EPS, electrophysiological study; ICD, implantable cardioverter-defibrillator; RV, right ventricle; RVOT, right ventricular outflow tract; SCD, sudden cardiac death; SCN5A, sodium voltage-gated channel alpha subunit 5; VERP, ventricular effective refractory period.
% are related to patients with a genetic sampling.
At univariate logistic regression analysis, a positive EPS was predicted by the family history of SCD (odds ratio, OR 1.99, 95% CI 1.03–3.82, P = 0.039), the presence of spontaneous Type-1 ECG pattern (OR 2.21, 95% CI 1.11–4.39, P = 0.024), and a ΔRPRVOT-apex > 60 ms (OR 9.78, 95% CI 4.28–22.35, P < 0.001). At multivariate logistic regression analysis, only ΔRPRVOT-apex > 60 ms (OR 8.01, 95% CI 3.41–18.78, P < 0.001) remained an independent predictor of a positive EPS.
Implantable cardioverter-defibrillators
As a primary prevention strategy, 89 patients (24%) received an ICD (64 transvenous ICD, 23 subcutaneous ICD) that was programmed according to current medical practice.
Patients with ICD had more frequently a family history of SCD (44 vs. 23%, P < 0.001), a history of syncope (51 vs. 17%, P < 0.001), a positive EPS (53 vs. 2%, P < 0.001), and a ΔRPRVOT-apex > 60 ms (34 vs. 7%, P < 0.001).
Out of total 89 ICD implants, 52 patients were implanted due to VT/VF inducibility during EPS, 35 patients due to syncope and spontaneous Type-1 electrocardiographic pattern, and 2 patients due to a family history of SCD, non-sustained VT during EPS, and symptoms (presyncope and palpitations).
Device-related complications occurred in 13 ICD recipients, with two infections, four lead fractures, and eight inappropriate shocks.
Survival analysis
In the whole population, 21 patients experienced the primary endpoint, with 4 SCD, no case of resuscitated cardiac arrest, and 17 VT/VF appropriately treated by the ICD during a median follow-up of 48 months (IR 36–60 months) (Table 3). The cumulative incidence of SCD and ICD appropriate interventions was 0.27 and 1.9%, respectively. Patients experiencing the primary endpoint had more frequently a family history of SCD (P = 0.002), a history of syncope (P = 0.007) and a spontaneous Type-1 ECG pattern (P = 0.003).
Table 3.
Patients’ characteristics according to the absence or presence of sudden cardiac death/appropriate implantable cardioverter-defibrillator shock
| All patients (n = 372) | SCD/ICD intervention− (n = 351) | SCD/ICD intervention+ (n = 21) | P-value |
|---|---|---|---|
| Age at diagnosis, years | 43 ± 15 | 45 ± 12 | 0.583 |
| Males | 241 (69) | 16 (76) | 0.682 |
| Family history of SCD | 92 (26) | 13 (62) | 0.002 |
| Syncope | 83 (24) | 11 (52) | 0.007 |
| Spontaneous Type-1 ECG | 168 (48) | 17 (81) | 0.003 |
| Patients with genetic testing (n = 167) | n = 154 | n = 13 | |
| ȃSCN5A mutationa | 48 (31) | 7 (54) | 0.124 |
| Patients undergoing EPS (n = 198) | n = 180 | n = 18 | |
| ȃVT/VF at EPSa | 43 (24) | 11 (61) | 0.002 |
| ȃVERP RV apex, ms | 221 ± 40 | 207 ± 20 | 0.146 |
| ȃVERP RVOT, ms | 241 ± 44 | 271 ± 32 | 0.007 |
| ȃΔRPRVOT-RVapex, ms | 10 (0–40) | 70 (35–90) | <0.001 |
| ȃΔRPRVOT-RVapex >60 msa | 23 (13) | 11 (61) | <0.001 |
Data are given as mean ± standard deviation or number (percentage). ECG, electrocardiogram; EPS, electrophysiological study; ICD, implantable cardioverter-defibrillator; RV, right ventricle; RVOT, right ventricle outflow tract; SCD, sudden cardiac death; SCN5A, sodium voltage-gated channel alpha subunit 5; VERP, ventricular effective refractory period; VF, ventricular fibrillation; VT, ventricular tachycardia.
% to be referred to patients with genetic testing or who underwent EPS.
Considering the subset who underwent EPS, 18 patients experienced the primary endpoint, with 3 SCD, no case of resuscitated cardiac arrest, and 15 appropriate ICD intervention during follow-up (48 IR 27–84 months). Comparing patients experiencing or not the primary endpoint, no difference was observed regarding VERP at the apex (207 ± 20 vs. 221 ± 40 ms, P = 0.15), while a longer VERP at the RVOT (271 ± 32 vs. 241 ± 44, P = 0.007), and a longer ΔRPRVOT-apex were observed in patients with events [70 (35–90) vs. 10 (0–40), P < 0.001] (Table 3). A higher risk of events was observed in patients with a positive EPS (61 vs. 24%, P = 0.004), or with a ΔRPRVOT-apex > 60 ms (61 vs. 13%, P < 0.001). The distribution of ΔRPRVOT-apex in BrS patients and the relationship (P-Spline HR analysis) between ΔRPRVOT-apex and the risk of events is displayed in Figure 1. The risk of events start to rise above a ΔRPRVOT-apex of 20 ms, but the optimal prognostic cutpoint was identified as 60 ms by using the maximally selected rank statistics (sensitivity 61%, specificity 87%, positive predictive value 32%, negative predictive value 96%, accuracy 85%). Notably, patients with a ΔRPRVOT-apex > 60 ms compared with those with a ΔRPRVOT-apex ≤ 60 ms had a both a longer VERP at RVOT (298 ± 19 vs. 234 ± 42 ms, P < 0.001) and a shorter VERP at the RV apex (202 ± 16 vs. 224 ± 41, P < 0.001). A ΔRPRVOT-apex > 60 ms was more frequently observed in patients with a spontaneous rather than drug-induced Type-1 ECG (25 vs. 6%, P < 0.001).
Figure 1.
P-spline HR curve of SCD/appropriate ICD shock related to ΔRPRVOT-RVapex. The continuous curve represents the P-spline describing hazard ratio (HR) for sudden cardiac death (SCD) or aborted appropriate ICD shocks as a function of ΔRPRVOT-RVapex, while the dashed lines represent the 95% CIs. While a higher risk of events is already evident with a ΔRPRVOT-RVapex > 20 ms, above a ΔRPRVOT-RVapex of 60 ms the HR for SCD/appropriate ICD shocks steeply increases. This cutpoint corresponds to the optimal cutpoint identified by the maximal log-rank statistic analysis. In the upper-left panel the distribution of different ΔRPRVOT-RVapex intervals is also displayed.
At Kaplan–Meier analysis, the family history of SCD (Figure 2A), the presence of syncope (Figure 2B), and spontaneous Type-1 ECG pattern (Figure 2C) were all associated with a higher risk of experiencing the primary endpoint. As for symptoms, the presence of presyncope or palpitation was not associated with events (all P > 0.05).
Figure 2.
Kaplan–Meier curves of traditional risk factors for Brugada syndrome. Family history of sudden cardiac death (SCD, A) and syncope (B) and spontaneous Type-1 ECG pattern (C) were all adverse prognostic factors associated with SCD and ICD appropriate shock in patients with Brugada syndrome.
A higher risk of events was also observed in patients with a pathogenic SCN5A mutation, but without a family history of BrS (to avoid any referral bias, Supplementary material online, Figure S1).
In patients undergoing EPS, inducibility of VF/VT at EPS (Figure 3A) and ΔRPRVOT-apex > 60 ms (Figure 3B) stratified the risk of events. The prognostic power of EPS was improved by using the ΔRPRVOT-apex > 60 ms. Indeed, the primary endpoint occurred more frequently in patients in whom a positive EPS was accompanied by ΔRPRVOT-apex > 60 ms (Figure 3C).
Figure 3.
Prognostic significance of VT/VF inducibility and the difference in the refractory period between the right ventricle outflow tract and apex (ΔRPRVOT-RVapex). In patients undergoing EPS (n = 198), VT/VF inducibility (A) and a difference in the refractoriness between the right ventricular outflow tract and the apex (B) were predictors of SCD and ICD shocks. Notably, patients with a positive EPS (VT/VF inducibility) but a ΔRPRVOT-RVapex ≤ 60 ms showed a similar risk of events compared with patients with a negative EPS, while those with a positive EPS and a ΔRPRVOT-RVapex > 60 ms were found to be at a higher risk (C).
At Cox proportional hazard univariate analysis (Table 4), family history of SCD, history of syncope/agonic nocturnal breathing, a spontaneous Type-1 ECG pattern, a positive EPS (VT/VF inducibility), and a ΔRPRVOT-apex > 60 ms at EPS were all univariate predictors of the primary endpoint (all P < 0.05). At bivariate analysis (Table 5), ΔRPRVOT-apex > 60 ms remained an independent predictor of events, even when adjusted for the family history of SCD, history of syncope, and a positive EPS.
Table 4.
Univariate predictors of sudden cardiac death/appropriate implantable cardioverter-defibrillator shock
| Univariate analysis | |||
|---|---|---|---|
| Hazard ratio | 95% CI | P-value | |
| Age at diagnosis, years | 1.01 | 0.98–1.04 | 0.604 |
| Males | 1.38 | 0.50–3.76 | 0.533 |
| Family history of SCD | 4.25 | 1.76–10.28 | 0.001 |
| Spontaneous Type-1 ECG | 4.18 | 1.39–12.41 | 0.010 |
| Symptoms (syncope/agonic nocturnal breathing) | 4.36 | 1.42–13.35 | 0.009 |
| SCN5A mutation | 2.39 | 0.80–7.10 | 0.112 |
| VT/VF at EPS | 3.65 | 1.45–9.07 | 0.005 |
| ΔRPRVOT-RVapex > 60 ms | 7.72 | 2.98–19.95 | <0.001 |
CI, confidence interval; ECG, electrocardiogram; EPS, electrophysiological study; ICD, implantable cardioverter-defibrillator; RP, refractory period; RV, right ventricle; RVOT, right ventricle outflow tract; SCD, sudden cardiac death; SCN5A, sodium voltage-gated channel alpha subunit 5; VF, ventricular fibrillation; VT, ventricular tachycardia.
Table 5.
Predictive value of ΔRPRVOT-RVapex at bivariate analysis for sudden cardiac death and appropriate implantable cardioverter-defibrillator shock
| Bivariate analysis | |||
|---|---|---|---|
| Hazard ratio | 95% CI | P-value | |
| ΔRPRVOT-RVapex > 60 ms | 6.66 | 2.55–17.41 | <0.001 |
| Family history of SCD | 2.68 | 1.02–7.01 | 0.045 |
| ΔRPRVOT-RVapex > 60 ms | 7.68 | 2.93–20.08 | <0.001 |
| Symptoms (syncope/agonic nocturnal breathing) | 1.04 | 0.39–2.72 | 0.944 |
| ΔRPRVOT-RVapex > 60 ms | 6.17 | 2.31–16.48 | <0.001 |
| Spontaneous Type-1 ECG pattern | 2.16 | 0.60–7.78 | 0.239 |
| ΔRPRVOT-RVapex > 60 ms | 5.84 | 1.87–18.24 | 0.002 |
| VT/VF at EPS | 1.64 | 0.53–5.12 | 0.392 |
CI, confidence interval; ECG: electrocardiogram; EPS, electrophysiological study; ICD, implantable cardioverter-defibrillator; RP, refractory period; RV, right ventricle; RVOT, right ventricle outflow tract; SCD, sudden cardiac death; VF, ventricular fibrillation; VT, ventricular tachycardia.
At discrimination analysis by C-statistic, the strongest model was the one using ΔRPRVOT-apex > 60 ms as a covariate, with a C-statistics (95% CI) of 0.75 (0.63–0.86), followed by those using VT/VF inducibility at EPS, 0.67 (0.54–0.79), family history of SCD, 0.66 (0.54–0.78), a spontaneous Type-1 ECG pattern, 0.65 (0.57–0.73), and a history of syncope, 0.64 (0.51–0.76). Overall, a model including clinical variables (family history of SCD, syncope, and spontaneous Type-1 ECG pattern) and EPS variables (VT/VF inducibility and ΔRPRVOT-apex > 60 ms) provide a very high prognostic precision with C-statistic 0.88 (0.81–0.92).
The accuracy of the ΔRPRVOT-apex > 60 ms model was also confirmed by IBS analysis: indeed, models based on ΔRPRVOT-apex > 60 ms and family history of SCD showed the lowest IBS (0.15), followed by VT/VF inducibility at EPS (0.25), family history of SCD, history syncope (0.29), and a spontaneous Type-1 ECG pattern (0.54).
Discussion
The heterogeneity of right ventricular (RV) refractory periods was assessed at EPS in a large multicentric Italian cohort of patients with BrS. The difference in the refractory period between RVOT and RV apex (ΔRPRVOT-apex) > 60 ms improved risk stratification beyond VT/VF inducibility at EPS and common clinical predictors. If confirmed in larger cohorts, the assessment of the heterogeneity of RV refractory periods may help clinical decision-making on whether or not to prescribe more aggressive therapeutic options (e.g. an ICD) to BrS patients in primary prevention.
In our cohort, the family history of SCD, the presence of syncope, or nocturnal agonic breathing, a spontaneous Type-1 ECG pattern, and a positive EPS were confirmed to be predictors of life-threatening arrhythmic events.5,11–14 Delise et al.5 proposed a multiparametric approach, in which the addition of VT/VS inducibility at EPS to accepted clinical predictors increased the prognostic discriminative capability of the model. Indeed, a VT/VS inducibility at EPS predicted arrhythmic events also in our population, but the assessment of RV refractoriness heterogeneity during EPS added further precision to the EPS model. In fact, patients with VT/VF inducibility but with a short ΔRP RVOT-apex (≤60 ms) at EPS showed a similar risk of events than patients without VT/VF inducibility. Furthermore, at bivariate analysis, VT/VF inducibility was no longer predictive of SCD/appropriate ICD shocks when adding to the model ΔRPRVOT-apex > 60 ms, which instead remained an independent predictor of survival. Of note, mean ΔRPRVOT-apex values were significantly higher in patients with inducible ventricular arrhythmias compared with non-inducible ones.
The longer ΔRPRVOT-apex in inducible patients seems mainly driven by longer VERP at the RVOT than a shorter VERP at the apex in our population. Lambiase et al.15 reported a significant conduction delay in the RVOT compared with the RV body and apex in BrS subjects. The same region of delayed conduction gave rise to wavefront fragmentation and functional block that led to initiation of polymorphic VT/VF. These findings may explain our observation that a greater ΔRPRVOT-apex, mainly driven by a longer VERP assessed at the RVOT, was observed particularly in inducible patients. A reduction of VERP was also observed, especially in patients with an increased ΔRPRVOT-apex > 60 ms. Data from PRELUDE study12 showed that a VERP <200 ms, assessed only at the RV apex, represents a significant risk factor of arrhythmic events during follow-up.
The association between an increased VERP in the RVOT and the risk of arrhythmias is in line with what was observed in Scn5a+/− mouse model presented by Martin et al.16 Indeed, VERP is influenced also by the recovery of a critical number of activatable sodium channels to favour resumption of excitability, and they may be critically decreased in the RVOT (increasing VERP), thus reverting the physiological gradient between the apex and RVOT in high risk individuals. Overall, our findings suggest that the heterogeneity of refractoriness rather than the evaluation of refractoriness in a single site could represent an additional parameter to improve predictivity of EPS in BrS and further refine risk stratification. Noteworthy, no correlation was found regarding the induction site between patients with ΔRPRVOT-apex > 60 ms and those without.
Concerning the mechanisms and the substrate underlying the observed heterogeneity of refractoriness, several authors reported the presence of delayed and discontinuous conduction in the RVOT in patients with BrS,6,17 likely caused by abnormal active membrane processes and electric coupling.4,6 Although originally reported as a primary electrical disorder in the absence of overt structural disease, there is growing evidence that various degrees of structural alterations may be found by imaging,18 endomyocardial biopsy,19 or autopsy studies.20 Based on this evidence, BrS was suggested to be reclassified as a combination of structural and electrical defects, paving the way to a new risk stratification approach. It is plausible that the presence of a structurally abnormal myocardium at the RVOT would explain the increased refractoriness observed in our patients, with longer values being associated with increased arrhythmic susceptibility, especially when refractoriness is concomitantly reduced at the RV apex.
The arrhythmogenesis in BrS patients implies the occurrence of early re-excitation phenomena that would require early recovery for refractoriness. It is plausible that the presence of a structurally abnormal myocardium at the RVOT would explain the increased endocardial refractoriness observed in our patients, with longer values being associated with increased arrhythmic susceptibility, especially when refractoriness is concomitantly reduced at the RV apex, as previously highlighted in the PRELUDE.12 Not surprisingly, arrhytmogenic effects of increased VERP dispersion have been observed also in experimental models of long QT syndrome and ischaemic cardiomyopathy.21,22
Beyond the endocardial gradient of VERP across the RV found in our study, a transmural dispersion of repolarization has been described in BrS as well as in other ion channelopathies.23 This transmural heterogeneinty may further increase the risk of reentry tachyarrhytmias. Whether this gradient might be higher in patients with prolonged endocardial VERP or ΔRPRVOT-apex is a topic of great interest that should be explored in future studies. Finally, it is noteworthy to consider the combination of shortened action potential duration and lengthened VERP, especially in the RVOT, as key mechanism of electric vulnerability.16 Overall, we believe that VERP heterogeneity could then favour arrhythmogenicity by altering action potentials recovery or affecting conduction characteristics, conditions critical to the initiation of reentrant arrhythmias.
Study limitations
The first limitation of the study is its retrospective design, with potential heterogeneities among the five Italian centres. However, EPS indication, VPS protocol with two extrastimuli at the RV apex and RVOT, and management of ICD implantation were the same between centres and complied with international guidelines.
The measurement of the VERP heterogeneity was possible only when VERP was assessed at both the RV apex and the RVOT. Indeed, seven patients were excluded from the study for VERP being measured only at one site. As previously reported by Sroubek et al.10 in a large pooled analysis, patients induced by one extrastimulus represent a small part of the inducible population (2%), and should be considered at higher risk according to the presence of other accepted risk factors.
As previously reported, we also defined appropriate ICD shocks as fast VT/VF terminated by ICD intervention. However, ‘appropriate ICD shock’ should only be assumed as a surrogate of SCD (especially VT), considering that some of those arrhythmic events may spontaneously terminate without a risk of death. Furthermore, having the cases of SCD in our series occurred out of the hospital the ECG was not available and thus we cannot exclude that non arrhythmic conditions may have caused SCD. However, patients’ age, lack of relevant comorbidities, and Brs diagnosis make the probability of ventricular arrhythmias highly likely. Of note, in two cases of SCD an ICD had been implanted (after a positive EPS) and multiple shocks were delivered without resuscitation.
Finally, the relatively low number of events in our population is a limitation in evaluating the prognostic value of risk factors. However, our study reflects a ‘real-world’ situation including a large population of ‘primary prevention’ BrS patients and the low incidence of events is in line with previous published cohorts.5,12 Given that BrS patients present a lifelong arrhythmic risk, a longer follow-up is notwithstanding advisable.
Conclusions
Risk stratification in BrS remains a significant challenge. Our study showed for the first time that a ΔRPRVOT-apex > 60 ms, easily achievable during EPS, is independently associated with an increased risk of SCD/appropriate ICD shocks at follow-up. These findings, to be confirmed in larger studies with different cohorts and a prospective design, may increase the prognostic value of EPS and improve risk stratification of BrS patients deserving more aggressive therapeutic options.
Supplementary Material
Contributor Information
Andrea Rossi, Fondazione Toscana Gabriele Monasterio, Pisa, Italy.
Alberto Giannoni, Fondazione Toscana Gabriele Monasterio, Pisa, Italy; Health Science Interdisciplinary Center, Scuola Superiore Sant'Anna, Pisa, Italy.
Martina Nesti, Cardiovascular and Neurological Department, San Donato Hospital, Arezzo, Italy.
Pasquale Notarstefano, Cardiovascular and Neurological Department, San Donato Hospital, Arezzo, Italy.
Vincenzo Castiglione, Health Science Interdisciplinary Center, Scuola Superiore Sant'Anna, Pisa, Italy.
Gianluca Solarino, Cardiology Division, Versilia Hospital, Lido di Camaiore, Italy.
Lucio Teresi, Health Science Interdisciplinary Center, Scuola Superiore Sant'Anna, Pisa, Italy.
Gianluca Mirizzi, Fondazione Toscana Gabriele Monasterio, Pisa, Italy.
Vincenzo Russo, Department of Translational Medical Sciences, University of Campania Luigi Vanvitelli, Monaldi Hospital, Naples, Italy.
Luca Panchetti, Fondazione Toscana Gabriele Monasterio, Pisa, Italy.
Umberto Startari, Fondazione Toscana Gabriele Monasterio, Pisa, Italy.
Andrea Ripoli, Fondazione Toscana Gabriele Monasterio, Pisa, Italy.
Francesco Gentile, Fondazione Toscana Gabriele Monasterio, Pisa, Italy.
Amato Santoro, Cardiology Division, Azienda Ospedaliera Universitaria Senese, Siena, Italy.
Nicoletta Botto, Fondazione Toscana Gabriele Monasterio, Pisa, Italy.
Giancarlo Casolo, Cardiology Division, Versilia Hospital, Lido di Camaiore, Italy.
Juan Sieira, Heart Rhythm Management Centre, Universitair Ziekenhuis Brussel, Brussels, Belgium.
Maurizio Pieroni, Cardiovascular and Neurological Department, San Donato Hospital, Arezzo, Italy.
Pasquale Santangeli, Electrophysiology Section, Cardiovascular Division, Hospital of the University of Pennsylvania, Philadelphia, PA, USA.
Michele Emdin, Fondazione Toscana Gabriele Monasterio, Pisa, Italy; Health Science Interdisciplinary Center, Scuola Superiore Sant'Anna, Pisa, Italy.
Marcello Piacenti, Fondazione Toscana Gabriele Monasterio, Pisa, Italy.
Supplementary material
Supplementary material is available at Europace online.
Data Availability
Data are available from the corresponding author upon reasonable request.
References
- 1. Antzelevitch C, Yan GX, Ackerman MJ, Borggrefe M, Corrado D, Guo Jet al. J-Wave syndromes expert consensus conference report: emerging concepts and gaps in knowledge. J Arrhythm 2016;32:315–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Steven D, Roberts-Thomson KC, Inada K, Seiler J, Koplan BA, Tedrow UBet al. Long-Term follow-up in patients with presumptive Brugada syndrome treated with implanted defibrillators. J Cardiovasc Electrophysiol 2011;22:1115–9. [DOI] [PubMed] [Google Scholar]
- 3. Milman A, Andorin A, Gourraud JB, Postema PG, Sacher F, Mabo Pet al. Profile of patients with Brugada syndrome presenting with their first documented arrhythmic event: data from the survey on arrhythmic events in BRUgada syndrome (SABRUS). Heart Rhythm 2018;15:716–24. [DOI] [PubMed] [Google Scholar]
- 4. Zhang J, Sacher F, Hoffmayer K, O’Hara T, Strom M, Cuculich Pet al. Cardiac electrophysiological substrate underlying the ECG phenotype and electrogram abnormalities in Brugada syndrome patients. Circulation 2015;131:1950–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Delise P, Allocca G, Marras E, Giustetto C, Gaita F, Sciarra Let al. Risk stratification in individuals with the Brugada type 1 ECG pattern without previous cardiac arrest: usefulness of a combined clinical and electrophysiologic approach. Eur Heart J 2011;32:169–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Behr ER, Ben-Haim Y, Ackerman MJ, Krahn AD, Wilde AAM. Brugada syndrome and reduced right ventricular outflow tract conduction reserve: a final common pathway? Eur Heart J 2021;00:1–12. [DOI] [PubMed] [Google Scholar]
- 7. Kawada S, Morita H, Antzelevitch C, Morimoto Y, Nakagawa K, Watanabe Aet al. Shanghai Score system for diagnosis of Brugada syndrome: validation of the score system and system and reclassification of the patients. JACC Clin Electrophysiol 2018;4:724–30. [DOI] [PubMed] [Google Scholar]
- 8. Wilde AAM, Antzelevitch C, Borggrefe M, Brugada J, Brugada R, Brugada Pet al. Proposed diagnostic criteria for the Brugada syndrome. Eur Heart J 2002;106:2514–9. [DOI] [PubMed] [Google Scholar]
- 9. Antzelevitch C, Brugada P, Borggrefe M, Brugada J, Brugada R, Corrado Det al. Brugada syndrome: report of the second consensus conference. Circulation 2005;111:659–70. [DOI] [PubMed] [Google Scholar]
- 10. Sroubek J, Probst V, Mazzanti A, Delise P, Hevia JC, Ohkubo Ket al. Programmed ventricular stimulation for risk stratification in the Brugada syndrome: a pooled analysis. Circulation 2016;133:622–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Macedo PG, Brugada J, Leinveber P, Benito B, Molina I, Sert-Kuniyoshi Fet al. Sleep-disordered breathing in patients with the Brugada syndrome. Am J Cardiol 2011;107:709–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Priori SG, Gasparini M, Napolitano C, Della Bella P, Ottonelli AG, Sassone Bet al. Risk stratification in Brugada syndrome: results of the PRELUDE (PRogrammed ELectrical stimUlation preDictive valuE) registry. J Am Coll Cardiol 2012;59:37–45. [DOI] [PubMed] [Google Scholar]
- 13. Sieira J, Conte G, Ciconte G, Chierchia G-B, Casado-Arroyo R, Baltogiannis Get al. A score model to predict risk of events in patients with Brugada syndrome. Eur Heart J 2017;38:1756–63. [DOI] [PubMed] [Google Scholar]
- 14. Honarbakhsh S, Providencia R, Garcia-Hernandez J, Martin CA, Hunter RJ, Lim WYet al. A primary prevention clinical risk score model for patients with brugada syndrome (BRUGADA-RISK). JACC Clin Electrophysiol 2020;7:210–22. [DOI] [PubMed] [Google Scholar]
- 15. Lambiase PD, Ahmed AK, Ciaccio EJ, Brugada R, Lizotte E, Chaubey Set al. High-density substrate mapping in Brugada syndrome. Circulation 2009;120:106–17. [DOI] [PubMed] [Google Scholar]
- 16. Martin CA, Grace AA, Huang CL. Refractory dispersion promotes conduction disturbance and arrhythmias in a Scn5a (+/−) mouse model. Pflugers Arch 2011;462:495–504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Guillem MS, Climent AM, Millet J, Berne P, Ramos R, Brugada Jet al. Conduction abnormalities in the right ventricular outflow tract in Brugada syndrome detected body surface potential mapping. Ann Int Conf IEEE Eng Med Biol Soc 2010;2010:2537–40. [DOI] [PubMed]
- 18. Rudic B, Schimpf R, Veltmann C, Doesch C, Tülümen E, Schoenberg SOet al. Brugada syndrome: clinical presentation and genotype-correlation with magnetic resonance imaging parameters. Europace 2016;18:1411–9. [DOI] [PubMed] [Google Scholar]
- 19. Pieroni M, Notarstefano P, Oliva A, Campuzano O, Santangeli P, Coll Met al. Electroanatomic and pathologic right ventricular outflow tract abnormalities in patients with Brugada syndrome. J Am Coll Cardiol 2018;72:2747–57. [DOI] [PubMed] [Google Scholar]
- 20. Nademanee K, Raju H, De Noronha VS, Papadakis M, Robinson L, Rothery Set al. Fibrosis, connexin-43, and conduction abnormalities in the Brugada syndrome. J Am Coll Cardiol 2015;66:1976–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Inoue H, Toda I, Nozaki A, Matsuo H, Mashima S, Sugimoto T. Inhomogeneity of ventricular refractory period in canineheart with quinidine-induced long QT interval: a comparative study on effects of heart rate, isoprenaline, and lignocaine. Cardiovasc Res 1985;19:623–30. [DOI] [PubMed] [Google Scholar]
- 22. Isomura S, Ando S, Kobe T, Ohta T, Toyama J, Yamada K. Dispersion of effective refractory period in the ischemic borderzone in canine heart. Jpn Circ J 1981;45:905–6. [Google Scholar]
- 23. Antzelevitch C. Heterogeneity and cardiac arrhythmias: an overview. Heart Rhythm 2007;4:964–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available from the corresponding author upon reasonable request.