15 Stereotactic radiosurgery has emerged over the past three decades as an alternative to microsurgical resection and embolization for cerebral arteriovenous malformations (AVMs). Since this technique was frst described by Steiner et al in 1972,1 over 5000 patients worldwide were successfully treated with this modality through December 2006. Radiosurgery causes gradual hyperplasia of the endothelial tissue within the arteries of the nidus of the AVM, which leads to vessel occlusion.2 Numerous reports show that cerebral AVMs have an 80 to 85% obliteration rate for vascular malformations < 2.5 cm.3–9 With the success of radiosurgery in the management of brain AVMs, the treatment of spinal cord AVMs with radiosurgery would be a logical next step. Given the various subtypes of spinal cord AVMs, those with a relatively compact nidus would represent the optimal targets. Spinal cord AVMs are typically classified into four distinct pathologic groups based on the location of the arteriovenous connections. Types I and IV are dural and perimedullary arteriovenous fistulas, and are often optimally treated with endovascular embolization and/or microsurgical resection. Type III, also called juvenile-type spinal AVMs, are characterized by large and diffuse intramedullary nidi, which can also extend into the extramedullary space. They are less well defined and do not represent optimal radiosurgical targets. Type II, also called glomus AVMs, represent a compact vascular nidus and are often suitable radiosurgery targets. These are also difficult if not impossible to treat with embolization and microsurgery alone and often were not treated prior to the development of spine radiosurgery. The ability to treat spinal cord AVMs with radiosurgery was not feasible prior to the development of image-guided spine radiosurgery.10 As with brain radiosurgery, spine radiosurgery relies on the delivery of a large number of cross-fired radiation beams to deliver an effective dose of radiation to a specifc target within the spine. Much of radiosurgery to spinal AVMs was based on prior experience of radiosurgery for spinal tumors.10–12 Despite the success with tumors, there remained a concern that high doses of radiation to focal areas within the spinal cord itself would represent a potentially signifcant risk to patients. The ability of frameless imageguided radiosurgery to deliver multiple sessions of radiosurgery to intramedullary vascular malformations played a role in reducing this risk. Our experience to date in the treatment of patients with spinal cord radiosurgery is based on a series of 23 patients treated between 1997 and 2006 with the CyberKnife. The following chapter represents the details of this 10-year experience. It has been our practice to offer CyberKnife (Accuray Inc., Sunnyvale, California) radiosurgery to those patients with type II spinal cord AVMs who are not candidates for either microsurgical or endovascular embolization. In addition, a small subset of type III spinal cord AVMs are considered candidates for radiosurgery if the vascular nidus is a reasonable target based on size. Currently, we do not feel that type I and type IV spinal cord AVMs are optimally treated with radiosurgery. At Stanford, these types of spinal cord AVMs typically are treated successfully with either open microsurgery or embolization. Low-flow spinal cord and vascular malformations, such as cavernous malformations, are not considered candidates for spinal cord radiosurgery. All potential candidates for radiosurgery of spinal cord AVMs undergo formal review by our multidisciplinary cerebrovascular and radiosurgery team. Team members include neurosurgeons and interventional radiologists experienced in spinal cord AVMs, as well as radiation oncologists. All patients who are considered candidates for radiosurgery undergo formal radiographic evaluation. This consists of spine magnetic resonance imaging (MRI) scans and conventional spinal angiography to identify the size, shape, and location of the vascular malformation.13 Feeding arteries and draining veins are identified to exclude them from the target volume. More recently, the use of three-dimensional (3D) angiography has improved our ability to visualize these vascular lesions.13–16 Although we have not utilized fusions between 3D angiography and computed tomography (CT) or MRI, we typically use 3D angiography to assist with our treatment planning. In some cases, patients with spinal cord AVMs undergo preradiosurgery embolization if it is felt that the AVM volume could be significantly reduced before radiosurgery. We have found, however, that the embolization glue can cause significant artifact on treatment planning CT and MRI scans; thus, we typically do not perform preradiosurgical embolization for smaller spinal cord AVMs. Once a candidate is approved for radiosurgery, informed consent is obtained. Our early experience in treating spinal cord AVMs relied on the placement of percutaneous fiducials into the bony anatomy of the spine to assist with target localization. More recently, advances in the CyberKnife treatment planning software have allowed the treatment of these spinal cord AVMs without fiducials, thereby eliminating one step in the treatment process.
Spinal Cord Arteriovenous Malformation Radiosurgery
Patient Selection
Pretreatment Preparation
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
