The use of 3D computer graphics in the diagnosis and treatment of spinal vascular malformations

Abstract

Object Digital subtraction (DS) angiography is the gold standard for diagnosing spinal vascular malformations. Recently, multidetectorrow spiral CT and contrast-enhanced MR angiography have been introduced as screening examinations before DS angiography. These methods, however, do not always determine the accurate location of an arteriovenous shunt because the resulting images lack information about the spinal cord or the dura mater. Methods Between April 2009 and December 2010, 13 patients underwent imaging evaluations for spinal vascular malformations at the authors' university hospital. This group included 8 patients with spinal dural arteriovenous fistulas (AVFs), 3 with perimedullary AVFs, and 2 with intramedullary arteriovenous malformations. Using data from these patients, the authors attempted to develop 3D computer graphics (CG) based upon the fusion of 3D rotational angiography and postmyelographic CT. They subsequently verified the accuracy of this imaging method. Ten of these 13 patients underwent surgical treatment for their lesions (11 AVFs), and for these 11 lesions the authors compared the diagnoses obtained using 3D CG with those obtained using conventional DS angiography. Results In all 13 cases, 3D CG images of the spinal lesions were successfully developed using the patients' actual data. Four (36%) of 11 AVFs were correctly identified using DS angiography, whereas 10 (91%) were correctly identified using 3D CG. Results from 3D CG of spinal AVFs corresponded well with operative findings, and 3D CG was significantly better than conventional DS angiography at predicting AVF location (p = 0.024, Fisher exact test). Conclusions To the authors' knowledge, this is the first reported case series in which 3D CG of spinal vascular malformations was used to provide simultaneous, stereoscopic visualization of the spinal vascular system, spinal cord, dura mater, and bone. The 3D CG method provides precise visual images for the diagnosis and treatment of these lesions.