Introduction. Surgical treatment of spinal metastasis belongs to the standards of oncology. The risk of spinal cord compression represents an operative indication. Intraoperative bleeding may vary, depending on the extent of the surgical technique. Some primary tumors, such as the renal cell carcinoma, present a major risk for hemorrhage and preoperative embolisation is mandatory. The purpose of this study is to evaluate the possible benefit of embolisation in different types of primary tumors. Material and Methods. The charts of 93 patients (42 women, 51 men, mean age 60.5 years) who were operated for spinal metastasis, 30 cases with multiple levels, were reviewed. Surgical procedures were classified as: (1) thoracolumbar laminectomy and instrumentation, (2) thoracolumbar corpectomy or vertebrectomy, (3) cervical corpectomy. A preoperative microsphere embolisation was performed in 35 patients. The following parameters, describing blood loss, were evaluated: hemoglobin variation from beginning to end of surgery, blood volume in suction during the intervention, number transfused packed red blood cells units until day 5 after surgery. A Poisson model was used for statistical evaluation. Results. The origins of spinal metastasis were: 28 breast cancer (30.1%), 19 pulmonary carcinoma (20.4%), 16 renal cell carcinoma (17.2%), 30 other cancers (32.3%). An embolisation was always performed in metastasis of renal cell carcinoma. An embolisation was performed in 8 cases in breast, 3 in pulmonary and 9 in other cancers. In the breast cancer group, there was no difference between embolisation versus non-embolisation concerning intraoperative blood loss and transfusion (P=0.404). In the pulmonary group, no difference was found either, but the number of embolisation cases was limited. For other metastasis types, the embolisation had no significant influence (P=0.697). The type of surgical intervention (2) increased intraoperative bleeding significantly in all groups: breast (P=0.002), pulmonary (P=0007), others (P=0.001). The average intraoperative hemoglobin decrease was: 2.3 in renal, 2.5 in breast, 3.0 in pulmonary, 1.9 in other metastasis (P=0.692). Conclusion. Several studies have clearly shown that the preoperative embolisation of renal cell carcinoma is recommended because of their risk of hemorrhage. However, the benefits of this procedure have been less described for other metastatic vertebral lesions. For breast cancer and other carcinoma (mainly digestive and prostate), the results of this study do not indicate a clear benefit for patients who received an embolisation. The same tendency was observed for pulmonary metastasis. The extent of the operation (corpectomy or vertebrectomy) represents the main factor that influences intraoperative bleeding