P6588Impact of age and sex on the long-term prognosis associated with early repolarization in the general population

Abstract Background Early repolarization (ER) in the electrocardiogram (ECG) has been associated with increased sudden cardiac death (SCD) risk in the general population. However, controversy remains about the risk associated with ER in different population subgroups. Purpose We evaluated whether age and sex affect the prognostic significance of ER in general population subjects. Methods The study population consisted of Finnish general population subjects aged over 30 years participating in the Mini-Finland Health Survey in 1978–80. Subjects underwent extensive baseline health examinations including a resting 12-lead ECG and were followed for 24±10 years. After excluding ECGs with QRS duration >110ms, the presence of ER (J-point elevation ≥0.1 mV in ≥2 inferior/lateral leads) was assessed from the remaining 6,631 subjects. The association between ER and SCD risk was analysed in subgroups according to age (<50 years or ≥50 years) and sex by using the Cox proportional hazards model. Effect modification of ER by age group and by sex was tested using the Wald test. Results ER was present in 250 (15.7%) of the 1,592 male subjects aged <50 years, 117 (6.8%) of the 1,713 female subjects aged <50 years, 182 (13.4%) of the 1,362 male subjects aged ≥50 years, and 244 (12.4%) of the 1,964 female subjects aged ≥50 years. There was a significant interaction between age group and ER (p=0.011). Among subjects aged <50 years, ER was associated with SCD risk after adjusting for age, sex, blood pressure, body mass index, serum cholesterol, coronary artery disease, diabetes, and active smoking (hazard ratio [HR] 1.79; 95% confidence interval [CI] 1.09–2.92; p=0.021), whereas in subjects aged ≥50 years ER was not associated with increased risk of SCD. Particularly female subjects aged <50 years with ER had a high risk of SCD (HR 4.85; 95% CI 1.65–14.29; p=0.004) in the multivariate analysis, whereas among <50-year-old male subjects ER was not associated with SCD (p=0.024 for interaction between sex and ER). Conclusion Our results suggest that among adults <50 years old, and especially among <50-year-old women, ER is associated with increased SCD risk, whereas among older subjects ER is not associated with SCD. Future research should focus on identifying the factors accounting for the difference between the age groups and to improve the risk stratification in the younger female patient populations with ER. Acknowledgement/Funding Aarne Koskelo foundation and Paavo Ilmari Ahvenainen Foundation