Anticoagulation is recommended for thromboprophylaxis after lower-limb orthopedic surgery. The suggested dosage is based on creatinine clearance (CCr) in the labels. However, most facilities only provide estimated glomerular filtration rate (eGFR) as laboratory data. Because the eGFR equation adjusts for a body surface area (BSA) of 1.73 m2, it may overestimate renal function in patients with a small BSA. This retrospective study aimed to determine whether different renal function estimation formulas affect the incidences of venous thromboembolism (VTE) and bleeding when determining anticoagulant dosages. This study included patients who underwent lower-limb orthopedic surgery and received anticoagulants (edoxaban, enoxaparin, and fondaparinux) between 2017 and 2020 at Yaizu City Hospital. Anticoagulant dosing was evaluated using CCr, eGFR, and de-indexed eGFR (without correction for BSA), and the incidences of VTE and bleeding were compared among these formulas. The median values for BSA, CCr, eGFR, and de-indexed eGFR were 1.40 m2, 56.0mL/min, 73.0mL/min/1.73m2, and 60.9mL/min, respectively. There was no significant difference in the VTE incidence among these formulas. However, when dose reduction or contraindication threshold was determined by eGFR vs. CCr, the bleeding incidence was significantly higher in the group that was overdosed by CCr (6.0% vs. 25.7%, p < 0.05). Similarly, using de-indexed eGFR vs. CCr, the bleeding incidence was significantly higher in the group that was overdosed by CCr (7.5% vs. 28.6%, p < 0.05). In orthopedic surgery, anticoagulant dosages should be based on CCr for patients with a small BSA to avoid bleeding risks.