The burden of atrial fibrillation (AF) and the risk of stroke are high in dialysis patients. The decision to use anticoagulation rests heavily on effective risk stratification. Because both the pathophysiology of the disease and the response to therapy differ in dialysis, data from the general population cannot be extrapolated. The effect of vitamin K antagonists (VKAs) on the risk of stroke in dialysis patients with AF has not been studied in randomized trials. The available observational data provide contradictory results, reflecting differences in the degree of residual confounding, quality of international normalized ratio control, and stroke characterization. Dialysis patients have a high baseline bleeding risk. It remains unclear to what extent VKAs affect the overall bleeding propensity, but they may significantly increase the risk of intracerebral hemorrhage. Vascular calcifications are extremely prevalent in dialysis patients and independently associated with an adverse outcome. Vitamin K antagonists inhibit the activity of key anticalcifying proteins and may thus compound the risk of vascular calcification progression in dialysis. In the absence of evidence-based guidelines for anticoagulation in dialysis patients with AF, we provide recommendations to assist clinicians in individualized risk stratification. We further propose that new oral anticoagulants may have a better benefit-risk profile in dialysis patients than VKA, provided appropriate dose reductions are made. New oral anticoagulant may yield more on-target anticoagulation, reduce the risk of intracerebral bleeding, and not interfere with vascular calcification biology. Clinical trials with new oral anticoagulant in dialysis patients are eagerly awaited, to reveal whether these assumptions can be confirmed.