Programmed Ventricular Stimulation for Risk Stratification in the Brugada Syndrome: A Pooled Analysis

Abstract

Background: The role of programmed ventricular stimulation in identifying patients with Brugada syndrome at the highest risk for sudden death is uncertain.

Methods and results: We performed a systematic review and pooled analysis of prospective, observational studies of patients with Brugada syndrome without a history of sudden cardiac arrest who underwent programmed ventricular stimulation. We estimated incidence rates and relative hazards of cardiac arrest or implantable cardioverter-defibrillator shock. We analyzed individual-level data from 8 studies comprising 1312 patients who experienced 65 cardiac events (median follow-up, 38.3 months). A total of 527 patients were induced into arrhythmias with up to triple extrastimuli. Induction was associated with cardiac events during follow-up (hazard ratio, 2.66; 95% confidence interval [CI], 1.44-4.92, P<0.001), with the greatest risk observed among those induced with single or double extrastimuli. Annual event rates varied substantially by syncope history, presence of spontaneous type 1 ECG pattern, and arrhythmia induction. The lowest risk occurred in individuals without syncope and with drug-induced type 1 patterns (0.23%, 95% CI, 0.05-0.68 for no induced arrhythmia with up to double extrastimuli; 0.45%, 95% CI, 0.01-2.49 for induced arrhythmia), and the highest risk occurred in individuals with syncope and spontaneous type 1 patterns (2.55%, 95% CI, 1.58-3.89 for no induced arrhythmia; 5.60%, 95% CI, 2.98-9.58 for induced arrhythmia).

Conclusions: In patients with Brugada syndrome, arrhythmias induced with programmed ventricular stimulation are associated with future ventricular arrhythmia risk. Induction with fewer extrastimuli is associated with higher risk. However, clinical risk factors are important determinants of arrhythmia risk, and lack of induction does not necessarily portend low ventricular arrhythmia risk, particularly in patients with high-risk clinical features.

Keywords: Brugada syndrome; arrhythmias, cardiac; death, sudden, cardiac; electrophysiology; risk assessment.

Figure 1

Unadjusted survival curves demonstrating freedom from sudden cardiac arrest or ventricular tachycardia resulting in defibrillator shock are displayed for patients according to presence or absence of syncope and electrocardiographic type 1 pattern on presentation.

Figure 2

Risk of sudden cardiac arrest or ventricular tachycardia resulting in defibrillator shock relative to individuals that were not induced is displayed according to the number of extrastimuli in the protocol associated with induction. The hazard ratios for each extrastimulus protocol (black points and 95% confidence bars) relative to those that were not induced are displayed. Superimposed is the association and 95% confidence interval (gray) derived from a model comprised of both an indicator for induction and linear term for the number of extrastimuli causing induction, based on the effect estimates and covariance between both terms. Estimates were derived from models including the 1247 individuals in whom ventricular arrhythmias were induced or testing was carried forward to triple extrastimuli. Models were adjusted for age at electrophysiology study, sex, syncope, and spontaneous type 1 Brugada pattern.

Figure 3

Survival curves demonstrating freedom from sudden cardiac arrest or ventricular tachycardia resulting in defibrillator shock are displayed for patients according to induction with single or double extrastimuli in the following clinical subgroups: A) syncope and spontaneous type 1 electrocardiographic pattern, B) syncope and drug-induced type 1 electrocardiographic pattern, C) asymptomatic and spontaneous type 1 electrocardiographic pattern, D) asymptomatic and drug-induced type 1 electrocardiographic pattern. Survival estimates were derived from Cox models adjusted for age at electrophysiology study, sex, cohort, syncope status, and presence of type 1 electrocardiographic pattern. Mean values of each covariate were used for probability estimation and plotting. Shading represents the 95% confidence intervals for survival estimates. Induction was significantly associated with an increased risk of events (P=0.002).