Massive intraoperative blood loss remains a major complication in surgery for metastatic spinal tumors, and accurate preoperative assessment of tumor vascularity is essential for bleeding risk prediction and surgical planning. Although digital subtraction angiography (DSA) is considered the gold standard for evaluating tumor vascularity, it is invasive and logistically demanding. In routine clinical practice, vascularity is often inferred from primary tumor type; however, this approach is frequently unreliable because of substantial tumor heterogeneity. The purpose of this study was to evaluate the diagnostic performance of intratumoral flow void (IFV) diameter measured on standard MRI for predicting tumor vascularity, using DSA as the reference standard, and to compare its performance with primary tumor type–based assessment.

We retrospectively analyzed 134 patients who underwent spondylectomy for isolated spinal metastasis between 2010 and 2023 and had both preoperative MRI and DSA. Tumor hypervascularity was defined as an angiographic grade of 3 or higher on DSA. IFV diameter was measured on standard MRI sequences and correlated with angiographic grade using Spearman’s rank correlation. Receiver operating characteristic analysis was performed to identify the optimal IFV diameter cutoff for predicting hypervascular tumors, and diagnostic performance was compared with primary tumor type–based vascularity assessment.

IFV diameter demonstrated a strong correlation with angiographic grade (r = 0.73, p < 0.05), whereas primary tumor type–based assessment showed only a moderate correlation (r = 0.39, p < 0.05). An IFV diameter cutoff of 2.5 mm provided optimal discrimination for hypervascular tumors, yielding a sensitivity of 78.0%, specificity of 97.6%, and an area under the curve of 0.91. In contrast, primary tumor type–based assessment showed inferior diagnostic performance, with a sensitivity of 70.0% and specificity of 60.7%.

IFV diameter measured on standard MRI represents a reliable, noninvasive imaging marker for evaluating tumor vascularity in metastatic spinal tumors. A cutoff value of 2.5 mm enables effective preoperative bleeding risk stratification and may support decision- making regarding surgical strategy and adjunctive interventions, particularly in clinical settings where DSA is not routinely available.