CT Fluoroscopy–guided Percutaneous Vertebroplasty for the Treatment of Osteolytic Breast Cancer Metastases: Results in 62 Sessions with 86 Vertebrae Treated

Abstract

This retrospective study aimed to assess the results of computed tomographic (CT) fluoroscopy-guided percutaneous vertebroplasty (PV) of painful osteolytic spinal metastases from breast cancer, focusing on the frequency and clinical impact of polymethylmethacrylate (PMMA) leaks. Within 48 months, 53 patients (52 women; mean age of 62 y +/- 13) with painful osteolytic breast cancer metastases underwent vertebroplasty. Eighty-six vertebrae were treated in 62 sessions under CT fluoroscopy guidance (single slice and four- and 16-row CT). In the planning CT scan, osteolytic destruction (ie, none, < or =25%, < or =50%, < or =75%, or < or =100%) was assessed regarding the vertebral cross-sectional area, the cortical border of the spinal canal, and the outer circumference. CT performed after vertebroplasty was used to detect local PMMA leaks (intradiscal, intraspinal, paravertebral, intercostovertebral/posterolateral, and vascular). Patient charts were reviewed with respect to adverse events. Clinical outcome was assessed on a visual analog scale (VAS) 24 hours before, immediately after, and 6 months after PV. Overall, 37.2%, 12.8%, and 1.2% of vertebrae (N = 86) showed at least a 50% osteolytic involvement of the cross-sectional area, spinal canal, and outer vertebral cortex, respectively. Intradiscal, intraspinal, paravertebral, and intercostovertebral/posterolateral leaks were seen in 31.3%, 26.9%, 26.9%, and 14.9% of vertebrae, respectively. The ratio of basivertebral to segmental venous leaks was 22.4%/23.9%. No major complications (eg, radiculopathy or pulmonary embolism) occurred. VAS scores decreased significantly (P < .05) from 6.4 at 24 hours before PV to 3.4 at a mean follow-up of 9.2 months. PV of osteolytic breast cancer metastases can be performed safely under CT fluoroscopic guidance even with substantial involvement of the vertebral cross-sectional area or cortical bone. A high clinical success rate was achieved and cortical and vascular PMMA leaks had no impact.