BACKGROUND: For some patients with atrial fibrillation, direct current cardioversion (DCCV) is one strategy that can be used to establish sinus rhythm but appropriate anticoagulation is mandatory to prevent thromboembolic events. Historically, patients were anticoagulated with warfarin with bridging with unfractionated or low molecular weight heparin, however, recently novel oral anticoagulants (NOACs), apixaban, dabigatran and rivaroxaban have become more popular. Despite the increase in use, real world data on safety and efficacy is limited. METHODS: We retrospectively analysed patients that underwent DCCV at Wollongong Hospital from 1 January 2014 to 30 June 2016 and compared peri-procedural anticoagulation with warfarin and the three NOACs. Patients were treated with at least 24hours of anticoagulation before and at least four weeks after the procedure unless contraindication developed. All patients underwent transoesophageal echocardiography prior to cardioversion regardless of anticoagulation type or duration. Patients with left atrial or left atrial appendage thrombus did not undergo cardioversion. We analysed the utilisation rates of NOACs and compared the incidence of post procedural ischaemic strokes and major bleeding events at eight weeks follow-up. RESULTS: Over the study period, 284 patients underwent DCCV; 109 (38.4%) patients were anticoagulated with warfarin and 175 (61.6%) with one of the three NOACs; 77 (27.1%) with apixaban, 60 (21.1%) with rivaroxaban and 38 (13.4%) with dabigatran. Patients treated with warfarin were on average older (71.3±9.7 vs. 65.2±12.9; p value, 0.0005) with more cardiac risk factors including documented heart failure with reduced ejection fraction (39.4% vs. 22.9%; p value, 0.0032), medically treated hypertension (76.1% vs. 48.6%; p value, 0.0001) and peripheral vascular disease (31.2% vs. 12.1%; p value, 0.0004). The NOACs were more frequently used in patients with lower CHADS-VASc scores; 179 patients had a score≤3 with 52 (29.1%) patients treated with warfarin and 127 (70.9%) treated with a NOAC (p value, 0.0001). In our cohort, the use of NOACs increased over the study period from 45.6% in 2014 to 82.8% in 2016. There was a low incidence of ischaemic stroke and bleeding events in both groups, 1.8% versus 0.6% (p value, 0.5607) and 3.6% versus 1.7% (p value, 0.4343) respectively. In the NOAC group, 95 of the 174 patients were anticoagulation-naïve and anticoagulated for less than five days; in comparison to longer duration therapy there was no difference in ischaemic stroke and bleeding events. CONCLUSION: In our institution, the use of NOACs in electrical cardioversion increased significantly over the study period and in our experience, they appear to be as safe as warfarin with low rates of ischaemic stroke and major bleeding. In addition, a short duration NOAC strategy was comparable to longer duration therapy.