The reported causes of sudden death (SD) in young people at autopsy include a variety of structural, often genetic, heart diseases. In young competitive athletes, arrhythmogenic cardiomyopathy (AC) in Italy (1) and hypertrophic cardiomyopathy (HCM) in the United States (2) are the most common causes of SD. However, recent postmortem studies suggest that most SDs are associated with a normal heart, implying a major role of ion channelopathies (3–5). We believe that it will be useful to offer a measure of clarification by reporting our long-lasting prospective experience at Padua University using a standardized autopsy protocol over 30 years, which suggests that structural cardiac diseases are the primary cause of SD in young athletes. According to the Regional Health System, every case of SD occurring in people aged 40 years or younger in northeast Italy should undergo autopsy, and the heart is referred to our Cardiovascular Pathology Unit. Between 1980 and 2015, we investigated 75 consecutive SD cases in young people (male: n= 73; 97.5%) who were engaged in competitive sports programs; 91% of SDs occurred during exercise. The most frequent causes (90%) were structural: AC (n= 20; 27%), atherosclerotic coronary artery disease (n= 17; 24%), congenital coronary artery anomalies (n= 12; 16%), HCM (n= 4; 5%), and myocarditis (n= 3; 4%). Only 8 (10%) hearts turned out morphologically normal. The subjects had undergone evaluation for competitive sports participation, including electrocardiography and echocardiography (when indicated), according to Italian legislation (1). During the same period, hearts were examined from 575 other young people not engaged in sport activities. Of these, 102 (18%) were judged to be structurally normal at autopsy. In contrast to our data, retrospective autopsy studies reported structurally normal hearts (so-called primary sudden arrhythmic death syndrome) in up to

42% of SDs during competitive and recreational sports activities and in 25% to 40% of young people not engaged in sport activities (Table 1)(3–5). We believe these discrepancies can be explained, at least in part, by differences in case referral patterns, selection bias, and methodological shortcomings and inequalities. Most of the data that support a high prevalence of normal hearts are taken retrospectively from coroner registries and surveys of individual pathology, or forensic pathology centers. Coronary atherosclerosis and HCM can be identified reliably at autopsy and are not likely to be referred to a tertiary center, which results in their underestimation in study cohorts and potentially leads to overestimation of hearts without structural abnormalities. Furthermore, differences in expertise and criteria used by individual autopsy laboratories to report abnormal anatomic structure can account for these discrepancies. It is possible that even our reported prevalence of 10% for normal hearts in SD in athletes represented an overestimation. This prevalence is consistent with, but somewhat higher than that reported in the large 35-year US Sudden Death Registry of 2,406 young competitive athletes (2). Being a single region experience, geographic and/or ethnic differences of SD in athletes could account in part for our findings. Among 52,130 athletes who underwent assessment for sport eligibility at the Center for Sport Medicine of Padua from 2005 to 2016, 199 (0.4%) were disqualified for cardiac reasons, and only 7 of these (3.5%) for channelopathies (2 long QT, 5 catecholaminergic polymorphic ventricular tachycardia, none for Brugada and short QT syndromes). In conclusion, structurally normal hearts occurred in 10% of a consecutive group of …