Edoxaban is used to prevent and treat stroke or systemic embolism such as venous thromboembolism. Although bleeding is the most common complication of anticoagulants, only a few studies have addressed the safety of direct oral anticoagulants in East Asian patients. In this study, we investigated the risk factors for bleeding in Japanese patients receiving edoxaban. A retrospective review of the records of 198 patients who received 30 mg/d edoxaban in our hospital between April 2015 and March 2017 was performed. Subsequently, these patients were followed up to 1 year. Seven (3.5%) and 22 (11.1%) patients developed major bleeding and clinically relevant bleeding, respectively. In the univariate Cox regression analyses, low baseline hemoglobin levels (p = 0.002) and low baseline creatinine clearance (p = 0.020) were significantly associated with major bleeding. Multivariate Cox regression analysis revealed that a low baseline hemoglobin level was a significant risk factor for major bleeding and clinically relevant bleeding [hazard ratio 1.67 per 1 g/dL decrease (95% confidence interval 1.14–2.56, p = 0.008) and hazard ratio 1.31 per 1 g/dL decrease (95% confidence interval 1.06–1.62, p = 0.013), respectively]. Baseline hemoglobin level in quartiles also showed a quartile-dependent decrease in major bleeding and clinically relevant bleeding event. These results suggest that low baseline hemoglobin level is a significant risk factor for both major bleeding and clinically relevant bleeding in Japanese patients receiving edoxaban. Thus, these patients should be carefully monitored.