Independent predictors of vertebral compression fracture following radiation for metastatic spine disease

Anthony L. Mikula, Zach Pennington, Nikita Lakomkin, Michelle J. Clarke, Peter S. Rose, Mohamad Bydon, Brett Freedman, Arjun S. Sebastian, Lichun Lu, Roman O. Kowalchuk, Kenneth W. Merrell, Jeremy L. Fogelson, Benjamin D. Elder

Research output: Contribution to journalArticlepeer-review


OBJECTIVE The goal of this study was to determine independent risk factors for vertebral compression fracture (VCF) following radiation for metastatic spine disease, including low bone mineral density as estimated by Hounsfield units (HU). METHODS A retrospective chart review identified patients with a single vertebral column metastasis treated with radiation therapy, a pretreatment CT scan, and a follow-up CT scan at least 6 weeks after treatment. Patients with primary spine tumors, preradiation vertebroplasty, preradiation spine surgery, prior radiation to the treatment field, and proton beam treatment modality were excluded. The HU were measured in the vertebral bodies at the level superior to the metastasis, within the tumor and medullary bone of the metastatic level, and at the level inferior to the metastasis. Variables collected included basic demographics, Spine Instability Neoplastic Score (SINS), presenting symptoms, bone density treatment, primary tumor pathology, Weinstein-Boriani-Biagini (WBB) classification, Enneking stage, radiation treatment details, chemotherapy regimen, and prophylactic vertebroplasty. RESULTS One hundred patients with an average age of 63 years and average follow-up of 18 months with radiation treatment dates ranging from 2017 to 2020 were included. Fifty-nine patients were treated with external-beam radiation therapy, with a median total dose of 20 Gy (range 8–40 Gy). Forty-one patients were treated with stereotactic body radiation therapy, with a median total dose of 24 Gy (range 18–39 Gy). The most common primary pathologies included lung (n = 22), prostate (n = 21), and breast (n = 14). Multivariable logistic regression analysis (area under the curve 0.89) demonstrated pretreatment HU (p < 0.01), SINS (p = 0.02), involvement of ≥ 3 WBB sectors (p < 0.01), primary pathology other than prostate (p = 0.04), and ongoing chemotherapy treatment (p = 0.04) to be independent predictors of postradiation VCF. Patients with pretreatment HU < 145 (n = 32), 145–220 (n = 31), and > 220 (n = 37) had a fracture rate of 59%, 39%, and 11%, respectively. An HU cutoff of 157 was found to maximize sensitivity (71%) and specificity (75%) in predicting postradiation VCF. CONCLUSIONS Low preradiation HU, higher SINS, involvement of ≥ 3 WBB sectors, ongoing chemotherapy, and nonprostate primary pathology were independent predictors of postradiation VCF in patients with metastatic spine disease. Low bone mineral density, as estimated by HU, is a novel and potentially modifiable risk factor for VCF.

Original languageEnglish (US)
Pages (from-to)617-623
Number of pages7
JournalJournal of Neurosurgery: Spine
Issue number4
StatePublished - Oct 2022


  • Hounsfield units
  • bone mineral density
  • oncology
  • radiotherapy
  • spinal metastasis
  • stereotactic body radiation therapy
  • vertebral compression fracture

ASJC Scopus subject areas

  • Surgery
  • Neurology
  • Clinical Neurology


Dive into the research topics of 'Independent predictors of vertebral compression fracture following radiation for metastatic spine disease'. Together they form a unique fingerprint.

Cite this