Search

Currently, there is no unified consensus on short-term anticoagulation after cardioversion of atrial fibrillation lasting less than 48 h in low-cardioembolic-risk patients. The aim of this study is to evaluate the rate of transient ischemic attacks, stroke and death in this subset of patients after cardioversion without post-procedural anticoagulation. In a prospective observational study, patients with recent-onset AF undergoing cardioversion attempts in the Emergency Department were evaluated over the past 3 years. Inclusion criteria were conversion to sinus rhythm, low thromboembolic risk defined by a CHA2DS2VASc score of 0-1 points for males (0-2 points for females aged over 65 years), and hospital discharge without anticoagulant treatment. Patients with severe valvular heart disease, underlying systemic causes of AF, and those discharged with anticoagulant therapy were excluded. The main outcomes measured were TIA, stroke and death at thirty days' follow-up after discharge. During the study period, 218 successful cardioversions, obtained both electrically and pharmacologically, were performed on 157 patients. One hundred and eleven patients were males (71%), the mean age was 55.2 years (±standard deviation 10.7), 99 patients (63%) reported a CHA2DS2VASc score of 0, and the remaining 58 (37%) had a risk profile of 1 point. Of these, latter 8 were females (5%) older than 65 years (risk score 2 points). At the thirty days outcome, none of the 150 enrolled patients who completed a follow-up visit has reported TIA or stroke, nor died, in the overall 211 successful cardioversions evaluated. In our study, the rate of thromboembolic events after cardioversion of recent-onset AF of less than 48 h duration, in patients with a 0-1 CHA2DS2VASc risk profile (females 0-2), appeared to be extremely low even in absence of post-procedural anticoagulation. These findings seem to confirm data from previous studies, and suggest that routine post-procedural short-term anticoagulation may be considered as an overtreatment in this very low-risk subset of patients.

The members of the PERNO Study Group were not individually captured in the metadata of the original publication. They are included in the metadata of this publication.