(Carlos Chagas, 1916) (1) Since the fi rst descriptions of Chagas disease (ChD), sudden death is considered one of the characteristics and responsible for >50% of the mortality in chagasic patients with heart failure. In this issue of RSBMT, Barbosa et al (2) . highlight several important aspects of the pathogenesis, prognosis, and treatment of ventricular arrhythmias in ChD (2) . Chagas disease has an essentially arrhythmogenic nature characterized by dense and complex ventricular arrhythmias; therefore, ventricular fi brillation likely constitutes the terminal event in most cases of sudden death, which accounts for the majority of deaths in ChD. Chagas cardiomyopathy gives rise to the most intense fi brous reaction amongst all causes of myocarditis, leading to severe disorganization of the myocardial architecture and structure. In addition to the fi brotic substrate, electrical dispersion plays an important role in arrhythmogenesis, and it is caused by areas of slow conduction leading to electrical instability. Myocardial fi brosis and areas of slow electrical conduction result in mechanical alterations in both timing and function. With the development of therapeutic strategies to prevent death from ventricular fi brillation, particularly the use of implantable cardioverter-defi brillators (ICDs), accurate patient stratifi cation based on the future risk of arrhythmic events has become very relevant in ChD (3) .A major therapeutic goal in the management of Chagas cardiomyopathy is prevention of sudden cardiac death. While ICD therapy is the standard treatment for secondary prevention of cardiac arrest, the strategy for primary prevention of cardiac arrest in Chagas cardiomyopathy is much more controversial. Although several clinical features have been assessed as possible risk factors for the development of fatal arrhythmias, the sensitivity and specifi city of any single test to predict fatal arrhythmias are limited. For example, left ventricular dysfunction is recognized as a strong general risk predictor, particularly for arrhythmic death in ChD (4) ; however, the mere presence of a depressed left ventricular ejection fraction (LVEF) is not sensitive or specifi c enough to stratify the risk of sudden death. Most patients with ChD that have an ICD for secondary prevention and thus have experienced a potentially fatal arrhythmic event have an LVEF >30%.
Electrocardiographic techniquesIn the last few years, several new electrocardiographic methods have been used to predict mortality in ChD including measurement of the length of the QRS and QT intervals from a standard 12-lead ECG, which is a simple and inexpensive method of risk assessment in ChD. QT dispersion is associated with left ventricular (LV) systolic dysfunction, and both increased QT dispersion and QT interval are related to a higher risk of death in ChD patients (5) . Moreover, in another study performed in the same study sample, T-wave axis deviation was a strong and independent predictor of mortality risk. Furthermore,