The prognostic value of the seasonal variations of T-wave alternans (TWA) and heart rate variability (HRV), and the seasonal distribution of ventricular fibrillation (VF) in Brugada syndrome (Br-S) is unknown. We assessed the utility of seasonal variations in TWA and HRV for risk stratification in Br-S using a 24-h multichannel Holter electrocardiogram (24-M-ECG). We enrolled 81 patients with Br-S (grouped according to their history of VF, n = 12; syncope, n = 8; no symptoms, n = 61) who underwent 24-M-ECG in all four seasons. Precordial electrodes were attached to the third (3L-V2) and fourth (4L-V2, 4L-V5) intercostal spaces. We determined the maximum TWA (max-TWA) values and calculated HRV during night and morning time periods for all seasons. During a follow-up period of 5.8 ± 2.8 years, 11 patients experienced new VF episodes and there was a peak in new VF episodes in the summer. The VF group had the greatest 3L-V2 max-TWA value during morning time in the summer among the three groups and showed higher 3L-V2 max-TWA value than in the other seasons. The cutoff value for the 3L-V2 max-TWA during morning time in the summer was determined to be 42 µV using ROC analysis (82 % sensitivity, 74 % specificity; p = 0.0006). Multivariate analysis revealed that a 3L-V2 max-TWA value ≥42 µV during morning time in the summer and previous VF episodes were predictors of future VF episodes. The 3L-V2 max-TWA value during morning time in the summer may be a useful predictor of future VF episodes in Br-S.