OBJECTIVE The aim of this prospective study was to assess the risks of electrical shock cardio-version in the treatment of supraventricular rhythm disorders when administered under effective-dose but short duration anticoagulation in patients with no intracavitary thrombus detectable by transesophageal echocardiography. METHODS One hundred nineteen patients, mean age 66 years, with permanent arrhythmia due to atrial fibrillation (n = 102), atrial flutter (n = 16) or atrial tachycardia (n = 1) and taking no long-term anticoagulant therapy were treated by electrical shock cardioversion. The patients were given heparin at an effective dose 72 hours prior to cardioversion. A transthoracic and a transesophageal echocardiography were performed less than 24 hours prior to cardioversion. RESULTS Twenty-one thrombi were evidenced in 16 patients (14.6%) including 18 in the left auricle, 1 in the left atrium and 2 in the right atrium. A spontaneous contrast was visualized in 38 patients (32%). Cardioversion was performed in 103 patients without thrombus and later in 9 of the 16 patients with thrombus after absorption under anticoagulant therapy as evidenced on the control transesophageal echocardiography. A sinus rhythm was obtained in 82% of the cases. All patients were given anti-vitamin K anticoagulants for one month. There were no clinical manifestation of ischemic vascular events during cardioversion nor during the one-month follow-up. CONCLUSIONS Early use of electrical shock cardioversion in patients with supraventricular rhythm disorders can be proposed without long-term anticoagulation therapy if the absence of thrombi is demonstrated by transesophageal echocardiography and short-term heparin is given followed by oral anticoagulants for at least 4 weeks. A large-scale randomized prospective study comparing the conventional strategy with the protocol used in this study would be required to definitively validate this approach and determine its possible advantages.