Objective: To determine the effectiveness of the unsubtracted roadmap technique (USRT) for decreasing cement leakage (CL) during percutaneous vertebroplasty (PVP). Methods: In this retrospective cohort study, patients who underwent PVP between January and November 2018 were included. Patients were divided into the unsubtracted roadmap (UR) (n = 20) and conventional venography (CV) (n = 22) groups. USRT was performed before cement injection in the UR group. Computed tomography was performed on the first day after PVP in both groups to assess the CL. The vertebral height restoration rate was confirmed by radiography before and after the procedure. The CL location was classified as basivertebral vein (type B), segmental vein (type S), or cortical defect (type C). Visual analog scale and Oswestry Disability Index scores before and after surgery were calculated to confirm clinical outcomes. Results: Based on the computed tomography images, CL occurred in 7 (34.78%) and 15 (63.64%) cases in the UR and CV groups, respectively. Number of CL type B, S, and C in the UR and CV groups were 2 and 6, 3 and 6, and 2 and 3, respectively. The injected cement volume was not significantly different between the UR (4.86 mL) and the CV (4.75 mL) groups (P = 0.450). Regarding vertebral body height restoration (P = 0.856) and clinical outcomes, there were no significant differences between both groups. Radiation exposure was significantly less in the UR group (P = 0.019). Conclusions: USRT is an effective method for reducing CL and radiation exposure during PVP.