Introduction: Factor Xa (FXa) inhibitors are superior to vitamin K antagonists (VKAs) in terms of avoiding hemorrhagic complications. However, no robust data are available to date as to whether this also applies to the early phase after stroke. In this prospective registry study, we aimed to investigate whether anticoagulation with FXa inhibitors in the early phase after acute ischemic stroke or transient ischemic attack (TIA) is associated with a lower risk of major bleeding events compared with VKAs. Materials and methods: The Prospective Record of the Use of Dabigatran in Patients with Acute Stroke or TIA (PRODAST) study is a prospective, multicenter, observational, post-authorization safety study at 86 German stroke units between July 2015 and November 2020. Primary outcome was a major bleeding event during hospital stay. Secondary endpoints were recurrent strokes, recurrent ischemic strokes, TIA, systemic/pulmonary embolism, myocardial infarction, death and the composite endpoint of stroke, systemic embolism, life-threatening bleeding and death. Results: In total, 10,039 patients have been recruited. 5,874 patients were treated with FXa inhibitors and 1,050 patients received VKAs and were eligible for this analysis. Overall, event rates were low. We observed 49 major bleeding complications during 33,297 treatment days with FXa-inhibitors (rate of 14.7 cases per 10,000 treatment days) and 16 cases during 7,714 treatment days with VKAs (rate of 20.7 events per 10,000 treatment days), translating into an adjusted hazard ratio (aHR) of 0.70 (95% confidence interval (95% CI): 0.37–1.32) in favor of FXa inhibitors. Hazards for ischemic endpoints (63 vs 17 strokes, aHR: 0.96 (95% CI: 0.53–1.74), mortality (33 vs 6 deaths, aHR: 0.87 (95% CI: 0.33–2.34)) and the combined endpoint (154 vs 39 events, aHR: 0.99 (95% CI: 0.65–1.41) were not substantially different. Discussion and conclusion: This large real-world study shows that FXa inhibitors appear to be similarly effective in terms of bleeding events and ischemic endpoints compared to VKAs in the early post-stroke phase of hospitalization. However, the results need to be interpreted with caution due to the low precision of the estimates.