Brugada syndrome in Hong Kong: long term outcome prediction through machine learning
Abstract Funding Acknowledgements Type of funding sources: None. Introduction Brugada syndrome (BrS) is an ion channelopathy that predisposes affected patients to spontaneous ventricular tachycardia/fibrillation (VT/VF) and sudden cardiac death (SCD). Despite its greater prevalence in Asia and epidemiological heterogeneity in disease manifestation, the majority of the conducted cohort studies available in current literature are based in Western countries. Purpose The aim of this study is to examine the clinical and electrocardiographic predictive factors of spontaneous VT/VF for Asian BrS patients. Methods This was a territory-wide retrospective cohort study of patients diagnosed with BrS between 1997 and 2019. The primary outcome was spontaneous VT/VF detected either during hospital admission or by implantable-cardioverter defibrillator (ICD) data. Cox regression was used to identify significant clinical and electrocardiographic risk predictors. Non-linear interactions between variables (latent patterns) were extracted using non-negative matrix factorization (NMF) and used as inputs into the random survival forest (RSF) model. Results This study included 516 consecutive BrS patients (mean age of initial presentation= 50 ± 16 years, male= 92%) with a median follow-up of 86 (interquartile range: 45-118) months. The cohort was divided into subgroups based on initial disease manifestation: asymptomatic (n = 314), syncope (n = 159) or VT/VF (n = 41). Annualized event rates per person-year were 1.70%, 0.05% and 0.01% for the VT/VF, syncope and asymptomatic subgroups, respectively. Multivariate Cox regression analysis revealed initial presentation of VT/VF (HR = 24.0, 95% CI = [1.21, 479] , P= 0.037) and standard deviation of P-wave duration (HR = 1.07, 95% CI = [1.00, 1.13], P = 0.044) were significant predictors. The NMF-RSF showed the best predictive performance compared to RSF and Cox regression models (precision: 0.87 v.s. 0.83 v.s. 0.76, recall: 0.89 v.s. 0.85 v.s. 0.73, F1-score: 0.88 v.s. 0.84 v.s. 0.74). Conclusions This is one of the largest territory-wide cohort studies on BrS and the largest study in Asia published to date, with an extensive median follow-up duration of 7 years. Clinical history, electrocardiographic markers and investigation results provide important information for risk stratification. Machine learning techniques using NMF and RSF significantly improves overall risk stratification performance.