C-arm Cone-beam CT: Applications for Spinal Cement Augmentation Demonstrated by Three Cases Jesse R. Knight, MD, Manraj Heran, MD, Peter L. Munk, MD, FRCPC, Rodney Raabe, MD, and David M. Liu, MD Spinal canal narrowing as a result of retropulsion of spinal structures before or during cement augmentation has been considered a contraindication to therapy. The authors describe three cases of compression fractures safely treated with cement augmentation with the novel application of C-arm cone-beam computed tomography (CT). All cases involved small amounts of posterior cement extrusion, after which osteotomy needles were left in place during C-arm cone-beam CT. Rapidly reformatted images were viewed with the use of bone windows, yielding three-dimensional visualization of pertinent anatomy to confirm endpoints of posterior extrusion and adequate bone filling. J Vasc Interv Radiol 2008; 19:1118 –1122 Abbreviation: 3D = three-dimensional CEMENT augmentation has been es- tablished as a method to primarily relieve pain caused by vertebral com- pression fracture. Although applica- tions vary, osteoporosis and tumoral infiltration remain the most common indications. Vertebroplasty has re- mained a popular choice as a result of its proven efficacy, relatively smaller cannula (vs that used for kypho- plasty), speed, and cost. Kyphoplasty has been gaining popularity as a method in which kyphotic wedging of the compression fracture can be re- versed, and it allows injection of ce- ment into a preformed cavity through displacement of cancellous bone with a balloon tamp. The purpose of this report is not to address nor discuss the controversy surrounding the applica- tions of vertebroplasty, kyphoplasty, or other cement augmentation tech- niques (1,2), but rather to describe a novel application of C-arm cone-beam computed tomography (CT) during cement augmentation to further in- crease the safety and efficacy of ther- apy. Feared complications of these ce- ment augmentation procedures in- clude posterior extrusion of cement and end-plate breakthrough. These complications may occur as a result of cortical disruption by tumor infiltra- tion. There have been suggestions of a considerable increase in complication rate when performing vertebroplasty or kyphoplasty in the setting of patho- logic fractures with posterior break- through (3). A published study of the incidence and consequences of poste- rior wall breakthrough (4) demon- strated that cord compression requir- ing surgical decompression and death has occurred as a result of uncon- trolled extravasation of cement out- side the vertebral body. In earlier ex- periences, the posterior extrusion of contrast medium as a result of tumoral disruption of the posterior cortical margin has been problematic, espe- cially with the injection of thin bone cement (ie, polymerized methyl- methacrylate) through small needle systems. In this setting, rapid influx of cement into the posterior column of the vertebral body can occur very sud- denly and unpredictably. End-plate extrusion of bone cement may also re- sult suddenly because of the compro- mised cortical bone. A number of techniques have been developed to minimize the chances of these two complications occurring. The use of biplane fluoroscopic guid- ance during the procedure has been used to allow continuous two-plane monitoring and mental construction of a three-dimensional (3D) model of the vertebral body. In a similar fashion, use of CT-guided fluoroscopy has also been touted as potentially useful for real-time observation of bone cement distribution, as well as identification of tumor infiltration and bone density (5). The advent of rotational angiogra- phy has revolutionized cerebral an- giography in allowing 3D reconstruc- tions of cerebral vasculature, thereby allowing for better characterization of From the Angiography and Interventional Radiol- ogy Section (J.R.K., R.R., D.M.L.), Inland Imaging, 801 South Stevens Street, Spokane, WA 99204; and Department of Radiology (M.H., P.LM.), Vancouver General Hospital, Vancouver, British Columbia, Canada. Received September 12, 2007; final revision received February 27, 2008; accepted April 7, 2008. Address correspondence to D.M.L.; E-mail: dliu@inlandimaging.com None of the authors have identified a conflict of interest. © SIR, 2008 DOI: 10.1016/j.jvir.2008.04.001 1118