Surgical Outcomes after Preoperative Embolization in Hypervascular Spinal Metastases: A Propensity Scored Study of 191 Patients

Study Design. Retrospective propensity-score matched, case-control study at 2 academic tertiary care centers. Objective. To assess the effect of preoperative embolization (PE) on (1) intraoperative blood loss, defined as conventional estimates of blood loss (EBL) and hemoglobin mass loss; and (2) secondary outcomes in patients with spinal metastases from hypervascular histologies. Background Context. PE intends to reduce blood loss during surgery for spinal metastases of hypervascular tumors such as renal cell carcinoma. However, studies investigating the effect of PE in hypervascular tumors often consist of small cohorts, do not correct for confounding factors, and have conflicting results. Materials and Methods. After propensity score matching, 46 PE patients were matched to 46 non-PE patients without baseline differences. The constraints of propensity score matching did not allow analysis of patients with tumor volumes >9 cm³. Multiple linear regression models were fitted for EBL and hemoglobin mass loss. Poisson regression models were fitted for both intraoperative and postoperative transfusions. Results. There was no difference in EBL [948 mL (IQR: 500-1750) vs. 1100 mL (IQR: 388-1925), P=0.68] and hemoglobin mass loss [201 g (IQR: 119-307) vs. 232 g (IQR: 173-373), P=0.18] between PE and non-PE patients. Other than higher 1-year survival rates (65% vs. 43%, P=0.05) in PE patients, there were no differences in secondary outcomes. In multiple regression analyses, PE was not associated with decreased intraoperative blood loss, hemoglobin mass loss, or perioperative blood transfusions. Conclusions. Our study demonstrated that, for tumors <9 cm³, PE did not reduce EBL, hemoglobin mass loss, or perioperative blood transfusions in patients undergoing spine surgery for metastases from hypervascular histologies. These findings suggest that urgent spine surgeries indicated for hypervascular histologies should not be delayed based on the availability of PE and accurate detection of preoperative hypervascularity, beyond histology, will likely be an important determination of future PE utilization for spinal metastases.