for the Brugada Syndrome Investigators in Japan Background-The prognosis of patients with saddleback or noncoved type (non-type 1) ST-elevation in Brugada syndrome is unknown. The purpose of this study was to clarify the long-term prognosis of probands with non-type 1 ECG and those with coved (type 1) Brugada-pattern ECG. Methods and Results-A total of 330 (123 symptomatic, 207 asymptomatic) probands with a coved or saddleback ST-elevation Ն1 mm in leads V 1 -V 3 were divided into 2 ECG groups-type 1 (245 probands) and non-type 1 (85 probands)-and were prospectively followed for 48.7Ϯ15.0 months. The absence of type 1 ECG was confirmed by drug provocation test and multiple recordings. The ratio of individuals with a family history of sudden cardiac death (14%) was lower than previous studies. Clinical profiles and outcomes were not notably different between the 2 groups (annual arrhythmic event rate of probands with ventricular fibrillation; type 1: 10.2%, non-type 1: 10.6%, probands with syncope; type 1: 0.6%, non-type 1: 1.2%, and asymptomatic probands; type 1: 0.5%, non-type 1: 0%). Family history of sudden cardiac death at age Ͻ45 years and coexistence of inferolateral early repolarization with Brugada-pattern ECG were independent predictors of fatal arrhythmic events (hazard ratio, 3.28; 95% confidence interval, 1.42 to 7.60; Pϭ0.005; hazard ratio, 2.66; 95% confidence interval, 1.06 to 6.71; Pϭ0.03, respectively, by multivariate analysis), although spontaneous type 1 ECG and ventricular fibrillation inducibility by electrophysiological study were not reliable parameters. Conclusions-The long-term prognosis of probands in non-type 1 group was similar to that of type 1 group. Family history of sudden cardiac death and the presence of early repolarization were predictors of poor outcome in this study, which included only probands with Brugada-pattern ST-elevation. (Circ Arrhythmia Electrophysiol. 2009;2:495-503.)