Erick M Westbroek1, A Karim Ahmed2, Zach Pennington2, Matthew L Goodwin2, Yuanxuan Xia2, Christine Boone3, Philippe Gailloud4, Daniel M Sciubba2. 1. Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA. Electronic address: erickw@jhmi.edu. 2. Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA. 3. Division of Interventional Radiology, University of California, San Diego School of Medicine, La Jolla, California, USA. 4. Division of Interventional Neuroradiology, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.
Abstract
BACKGROUND: Preoperative endovascular embolization of hypervascular metastatic spine tumors can reduce intraoperative blood loss. One frequent objection to embolizing these tumors is the concern for associated arteries feeding the spinal cord, such as the artery of Adamkiewicz. This study aimed to elucidate a relationship between spinal levels affected by hypervascular spine metastases and associated radiculomedullary arteries (RMAs). METHODS: A retrospective review of 46 patients who underwent preoperative embolization of hypervascular metastatic spine tumors was conducted. A total of 484 spinal levels were evaluated by diagnostic spinal angiography during embolization procedures. Each spinal level was categorized based on the presence or absence of tumor and RMA. RESULTS: No statistically significant associations were found. The relative risk of affected spinal levels having an associated RMA was 1.10 (95% confidence interval, 0.66-1.85). The attributable risk was 0.01 (-0.01 to 0.02). Chi-squared statistic was 0.13, with a P value of 0.7. Subgroup analysis in renal cell patients was also statistically insignificant, with a relative risk of 0.97 (95% confidence interval, 0.43-2.16). Chi-square statistic was 0.01, with a P value 0.94. A total of 32 patients (69.6%) had an RMA identified on their regional angiogram. Seventeen patients (37%) had an RMA at an embolized level. CONCLUSIONS: In this study, no association was found between spinal levels affected by hypervascular metastatic spine tumors and RMAs feeding the spinal cord by diagnostic spinal angiography. This suggests that these tumors have little intrinsic affinity for spinal levels with an RMA. However, given that tumor embolization frequently requires accessing multiple levels, the risk of encountering an RMA during embolization remains significant.
BACKGROUND: Preoperative endovascular embolization of hypervascular metastatic spine tumors can reduce intraoperative blood loss. One frequent objection to embolizing these tumors is the concern for associated arteries feeding the spinal cord, such as the artery of Adamkiewicz. This study aimed to elucidate a relationship between spinal levels affected by hypervascular spine metastases and associated radiculomedullary arteries (RMAs). METHODS: A retrospective review of 46 patients who underwent preoperative embolization of hypervascular metastatic spine tumors was conducted. A total of 484 spinal levels were evaluated by diagnostic spinal angiography during embolization procedures. Each spinal level was categorized based on the presence or absence of tumor and RMA. RESULTS: No statistically significant associations were found. The relative risk of affected spinal levels having an associated RMA was 1.10 (95% confidence interval, 0.66-1.85). The attributable risk was 0.01 (-0.01 to 0.02). Chi-squared statistic was 0.13, with a P value of 0.7. Subgroup analysis in renal cell patients was also statistically insignificant, with a relative risk of 0.97 (95% confidence interval, 0.43-2.16). Chi-square statistic was 0.01, with a P value 0.94. A total of 32 patients (69.6%) had an RMA identified on their regional angiogram. Seventeen patients (37%) had an RMA at an embolized level. CONCLUSIONS: In this study, no association was found between spinal levels affected by hypervascular metastatic spine tumors and RMAs feeding the spinal cord by diagnostic spinal angiography. This suggests that these tumors have little intrinsic affinity for spinal levels with an RMA. However, given that tumor embolization frequently requires accessing multiple levels, the risk of encountering an RMA during embolization remains significant.