Accuracy and Safety in Screw Placement in the High Cervical Spines: Retrospective Analysis of O-arm® based Navigation-assisted C1 lateral Mass and C2 Pedicle Screw

Abstract

Listen

Translate article

Introduction High screw misplacement rates, various pedicle morphometry and vertebral body size variations have led to a search of image-guided systems to improve the surgical accuracy of screw insertion in high cervical spine. With the advances and improvement of computer-assisted surgery (CAS) devices, image-guided screw insertion technique has been applied to the fields of spine surgery. The use of O-arm® has been proposed for more accurate and efficient spinal instrumentation. The purpose of present study was to evaluate accuracy, efficiency and safety of intra-operative O-arm® based navigation system for the placement of C1 lateral mass and C2 pedicle screws in high cervical spine operations. To our knowledge, this is first report of O-arm® based navigation-assisted screw placement in high cervical spine. Material and Methods Between June 2009 and August 2013, a total of 24 patients with atlantoaxial instability were surgically treated using the image guidance system. To reconstruct atlantoaxial instability, we have been using Harm's technique of C1 lateral mass and C2 pedicle screw fixations. A frameless, stereotactic O-arm® based image-guidance system was used for correct screw placement. Postoperative CT scan with multi-planar reconstructions were used to determine the accuracy of the screw placement. Two independent observers evaluated the CT images with detailed descriptions of the accuracy of screw placement, number of misplaced screws, grade and type of screw perforation. The accuracy of screw placement was assessed using a grading system proposed by Gertzbein and Robbins. Results A total of 91 screws, including 45 C1 lateral mass and 46 C2 pedicle screws were inserted using image-guidance system respectively. 3 C1 lateral mass and 2 C2 pedicles were excluded due to anatomic anomalies. In 5 cases, perioperative reposition of screws was done due to wall perforation detected with intraoperative O-arm scan immediately after initial screw insertion. Accuracy of screw placement was as follows: C1 lateral mass screws showed 39 grade A (86.6%), six grade B (13.3%), two grade C (4.4%) and no grade D or E. C2 pedicle screws showed 31 grade A (67.4%), 10 grade B (21.8%), three grade C (6.5%), two grade D (4.3%) and no grade E. Mean time required for inserting a screw was 4.8 minutes and mean time required for preparation of screw placement was 4 minutes. Postoperative radiological evaluations revealed that 3 (3.3%) screws inserted had perforated the vertebral artery canal more than 25% (critical breach) and iatorgenic VA stenosis was proved with postoperative CT angiography. No patients underwent reoperation for screw reposition. Conclusion In this study, the authors demonstrated that image-guidance systems could be applied safely to the atlantoaxial instabilities. However, incidence of screw perforation does not disappear completely due to the close proximity to spinal canal and surrounding vessels. Therefore, because of theses potential risk of injury to the vertebral artery and neural elements, the use of image guidance system seems to be beneficial, especially for high cervical instrumentations which require much experience and steep learning curves.