An association of atrial septal aneurysm (ASA) with cardiac arrhythmias has been described, and it has been suggested that undulating movements of the aneurysm initiate these arrhythmias, thereby causing arterial embolism. In this prospective study, all available electrocardiograms were reviewed and Holter monitoring was performed in 50 consecutive patients with echocardiographic diagnosis of ASA in order to assess the relationship between the occurrence of arrhythmias, morphologic characteristics of ASA and arterial embolism. Significant arrhythmias were identified in 26 (52%) patients (supraventricular n = 15, ventricular n = 6, both n= 5). Patients with arrhythmias were older (65 +/- 12 vs. 54 +/- 13 years, p = 0.005), frequently had palpitations (21/26 vs. 1/24, p < 0.0001) and an abnormal resting electrocardiogram (18/26 vs. 5/24, p < 0.001). By echocardiography, patients with arrhythmias had a larger left atrial (42.8 +/- 7.4 vs. 35.3 +/- 4.6 mm, p < 0.0001) and left ventricular enddiastolic diameter (53.8 +/- 5.6 vs. 49.7 +/- 4.1, p < 0.01) and a higher prevalence of associated mitral valve prolapse (12/26 vs. 4/24, p = 0.05). Potential cardiovascular causes for arrhythmia other than ASA were present in the great majority of patients with documented arrhythmias (24/26 vs. 7/24, p < 0.0001). The base diameter of ASA was larger in patients with arrhythmias (25.5 +/- 6.2 vs. 21.4 +/- 3.4, p < 0.01) and correlated with a larger left atrial diameter (r = 0.72, p < 0.0001). Concerning the mobility of ASA (maximal protrusion or phasic excursion), there was no significant difference between the two patient groups. Arterial embolism, however, predominantly occurred in ASA patients without arrhythmias (16/24 vs. 9/26, p < 0.05). In conclusion, the majority of patients with ASA and arrhythmias has underlying structural heart disease other than ASA which may be responsible for the arrhythmias observed. Arrhythmias in association with ASA do not play a major role as a mechanism for arterial embolism.