The most significant complication in patients with atrial fibrillation (AF) is stroke, and oral anticoagulants should be used to prevent stroke in patients at high risk of thromboembolism in AF, and anticoagulation is the only way to reduce mortality in patients with AF. Current anticoagulants are mainly vitamin K antagonists (warfarin) and newer oral anticoagulants (dabigatran, rivaroxaban, etc.). The most feared complication of oral anticoagulation therapy is intracranial hemorrhage. Can one still anticoagulate once intracranial bleeding has occurred? This has been a clinical dilemma, and the best treatment for cerebral hemorrhage in patients with atrial fibrillation is uncertain. If not anticoagulated, the patient may have an ischemic stroke (cerebral infarction) or other systemic embolism (mesenteric artery embolism, limb artery embolism, etc.); if anticoagulated, another intracranial hemorrhage may occur. Both outcomes are dangerous. The trade-off between the pros and cons of anticoagulation needs to be confirmed by clinical trials. Recently, a Danish study was published in Circulation, the professional journal of the American Heart Federation. The investigators analyzed data on oral anticoagulation for atrial fibrillation complicated by intracranial hemorrhage in the Danish National Registry between 1997 and 2013. Patients were stratified according to their treatment regimen after cerebral hemorrhage (no treatment, oral anticoagulation therapy, or antiplatelet therapy). In 1752 patients (1 year follow-up) ischemic stroke/systemic embolism and all-cause mortality (per 100 person-years) were 13.6 in the oral anticoagulation group, 27.3 in the non-treatment group, and 25.7 in the antiplatelet therapy group. recurrent cerebral hemorrhage: 8.0 in the oral anticoagulation group, 8.6 in the non-treatment group, and 5.3 in the antiplatelet therapy group. oral anticoagulation compared with no treatment for ischemic stroke/systemic embolism and all-cause mortality. The adjusted risk ratio for systemic embolism and all-cause mortality was 0.55, which means that the risk reduction rate was nearly half. Thus, the significant reduction in ischemic stroke/all-cause mortality with oral anticoagulation supports the feasibility of resuming oral anticoagulation after intracranial hemorrhage. That is, even if intracranial hemorrhage occurs, it is likely that the benefits of continuing anticoagulation still outweigh the harms. Of course, more clinical trials are needed to confirm the clinical management options for this group of patients.