Atrial rhythm disturbances, particularly atrial fibrillation (AF), are frequently encountered in trained athletes. P wave dispersion (PWD) is a recent electrocardiographic (ECG) marker that reflects velocity of atrial impulse propagation. However, it remains unknown whether the P wave duration and PWD are different between athletes and sedentary controls. In this study we therefore determined the P wave duration and PWD, markers for conduction abnormalities, in trained athletes and controls. Fifty athletes and sex and age-matched 40 healthy sedentary controls were included in the study. All of the athletes were the members of a local athletic college and they were regularly maintaining their sportive activities; the duration of athletic competition was 7.7 ± 3.3 years and the average athletic time was 10.1 ± 1.6 hours/week. The 12-lead surface ECG was obtained from each subject in the supine position. The P wave duration was measured, and the difference between the maximum and minimum P wave duration was defined as the PWD. Distribution of sex, age, body mass index, blood pressure was similar in athletic groups and controls. Heart rate was significantly lower in the athletes than in the controls (66 ± 7 vs 73 ± 9 beats/min, p < 0.05). Maximum and minimum P wave durations were not statistically different in athletic group and controls (115 ± 6 vs 114 ± 4 ms and 74 ± 8 vs 74 ± 7 ms, respectively). In addition, PWD did not differ significantly in both groups (41 ± 6 vs 40 ± 7 ms, respectively). Thus, athlete's heart is not associated with prolonged P wave duration and increased PWD, indicating that P wave duration or PWD could not be used as a predictor for AF developed in trained athletes.