Objective To preliminarily evaluate the safety and efficacy of percutaneous vertebroplasty (PVP) with 131I-loaded bone cement in treating vertebral tumor in rabbit models. Methods Twelve New Zealand white rabbits with lumbar vertebral tumor, which was established by puncturing transplant of VX2 carcinoma, were randomly and equally divided into the study group and the control group with 6 rabbits in each group. PVP with injection of 131I-loaded bone cement was carried out in the rabbits of the study group, while PVP with injection of pure bone cement was employed in the rabbits of the control group. The blood cell count was determined in all the animals one day before PVP as well as on the 4th day after PVP. PET-CT examination was performed one day before PVP as well as on the 4th day after PVP to check the stand uptake value (SUV) of each vertebral tumor. SPECT was performed in all rabbits of the study group at one, 4 and 8 days after PVP respectively to estimate the distribution of 131I in the animals’ bodies. Eight days after PVP, blood cell counts, which were determined both before and after PVP, existed between the study group and the control group. SPECT that was performed after PVP indicated that 131I was mainly accumulated within PVP-treated vertebrae, and the distribution of 131I showed no obvious changes at different points of time after the procedure. Before PVP, the difference in SUV between the two groups was of no statistical significance (F = 0.765, P > 0.05). In the study group, the postoperative SUV was significantly lower than the preoperative SUV (F = 423.792, P < 0.05). Pathological examination showed that the extent of tumor cell necrosis around the bone cement in the study group was remarkably bigger than that in the control group. Conclusion In treating vertebral tumors with PVP, the use of 131I-loaded bone cement is clinically feasible, and short-term follow-up indicates that this technique is safe and effective.