Abstract
BACKGROUND: This study aimed to evaluate the efficacy of subtraction computed tomography arteriography (s-CTA) during preoperative embolization in spinal tumors. METHODS: The study analyzed 17 vertebrae in 13 patients who underwent preoperative embolization before spinal fixation surgery for malignant spinal tumors to decrease blood loss at our hospital from 2019 to 2021. Their ages ranged from 56 to 88 years (average, 73.5 years). Metastatic bone tumors were most common, including five cases originating as lung carcinomas and three as renal cancers. After digital subtraction angiography of selected tumor-feeding arteries and non-subtraction CTA (ns-CTA) were performed, s-CTA was conducted using data obtained from both procedures. A clarity score of the boundary between the normal bone and tumor was derived for each patient, which was then classified into four grades (good, 3 points; fair, 2 points; faint, 1 point; poor, 0 points) by two experienced radiologists, followed by a comparison between the s-CTA and ns-CTA groups using the Wilcoxon signed-rank test. RESULTS: Clarity scores were significantly higher in the s-CTA group than in the ns-CTA group (P < 0.001). The agreement of Cohen's coefficients between the two radiologists was κ = 0.724 in s-CTA scoring and κ = 0.622 in ns-CTA scoring, which were moderately matched. Seven arteries were not embolized due to insufficient tumor contrast enhancement and their poor relation to the surgical invasion zone. No complications were observed during or after embolization. CONCLUSION: S-CTA successfully distinguished between tumor and normal bone and may help avoid unnecessary embolization.