BACKGROUND: Current guidelines recommend unfractionated heparin (UFH) or low-molecular-weight heparin plus an oral anticoagulant for the prevention of thromboembolism in patients undergoing electric cardioversion of atrial fibrillation (AF). Selective factor Xa inhibitors, such as fondaparinux, which has a favourable benefit-risk profile in the prevention and treatment of venous thromboembolism and the management of acute coronary syndromes, have not been systematically evaluated in this setting. AIM: To evaluate the efficacy and safety of fondaparinux versus standard treatment in patients undergoing echocardiographically-guided cardioversion of AF. METHODS: In this multicentre, randomized, open-label, controlled, two-parallel-group, phase II pilot study, patients with AF undergoing electric cardioversion following transoesophageal echocardiography (TEE) were randomized to fondaparinux or standard therapy (UFH plus vitamin K antagonist [VKA]). Patients showing an atrial thrombus in the first TEE (clot-positive) were randomized to treatment with fondaparinux or standard care for 4 weeks before cardioversion. RESULTS: The primary endpoint (combined rate of cerebral neurological events, systemic thromboembolism, all-cause death and major bleeding events) occurred in 3 of 174 (1.7%) patients on fondaparinux and 2 of 170 (1.2%) patients on UFH+VKA. The rate of thrombus disappearance among clot-positive patients was higher in the fondaparinux arm (11 of 14; 78.6%) than in the UFH+VKA arm (7 of 14; 50.0%). Incidences of adverse events were similar (45.4% with fondaparinux and 46.5% with UFH+VKA). CONCLUSION: In this pilot study in patients with TEE-guided cardioversion, the use of fondaparinux appeared to be well tolerated, with similar efficacy to UFH+VKA. Furthermore, a trend to greater thrombus resolution was observed.