Training-induced electrocardiographic changes are common in adult athletes. However, a few data are available on electrocardiogram (ECG) in preadolescent athletes and little is known about the potential changes induced by training on 12-lead electrocardiogram (ECG) at rest. Twelve-lead ECGs at rest and complete echocardiographic examinations were performed in 94 children (57 endurance athletes, 37 sedentary controls; mean age 10.8 ± 0.2 and 10.2 ± 0.2 years, respectively) at baseline and after 5 months of growth and training in athletes and of natural growth in controls. At baseline, athletes had lower heart rate at rest compared with controls (p = 0.046) and a further decrease was observed after training (p <0.0001). An incomplete right bundle branch block was found in 19% of athletes and 15% of controls (p = 0.69) with no changes after training. Although none of the athletes showed negative T waves from V1 to V3, 6% of controls at baseline had T-wave inversion V1 to V3 with a decrease to 3% after 5 months (p = 0.16). The early repolarization pattern did not differ between athletes and controls and was correlated with Tanner's scale score in the overall population both at first and second evaluation (R = 0.30, R = 0.27, p = 0.005, p = 0.012, respectively). No correlations were found between ECG and echocardiographic data. In conclusion, 12-lead ECG at rest is not substantially affected by training in children, despite a physiological increase in cavity size. Thus, in preadolescent athletes, 12-lead ECG at rest does not reflect exercise-induced morphologic remodeling and seems to be influenced more by sexual maturation than by training.