Although transpedicular fixation is a biomechanically superior technique, it is not routinely used in the cervical spine. The risk of neurovascular injury in this region is considered high because the diameter of cervical pedicles is very small and their angle of insertion into the vertebral body varies. This study was conducted to analyze the clinical accuracy of stereotactically guided transpedicular screw insertion into the cervical spine.

Twenty-seven patients underwent posterior stabilization of the cervical spine for degenerative instability resulting from myelopathy, fracture/dislocation, tumor, rheumatoid arthritis, and pyogenic spondylitis. Fixation included 1–6 motion segments (mean 2.2 segments). Transpedicular screws (3.5-mm diameter) were placed using 1 of 2 computer-assisted guidance systems and lateral fluoroscopic control. The intraoperative mean deviation of frameless stereotaxy was < 1.9 mm for all procedures.

No neurovascular complications resulted from screw insertion. Postoperative computed tomography (CT) scans revealed satisfactory positioning in 104 (90%) of 116 cervical pedicles and in all 12 thoracic pedicles. A noncritical lateral or inferior cortical breach was seen with 7 screws (6%). Critical malplacement (4%) was always lateral: 5 screws encroached into the vertebral artery foramen by 40–60% of its diameter; Doppler sonographic controls revealed no vascular compromise. Screw malplacement was mostly due to a small pedicle diameter that required a steep trajectory angle, which could not be achieved because of anatomical limitation in the exposure of the surgical field.

Despite the use of frameless stereotaxy, there remains some risk of critical transpedicular screw malpositioning in the subaxial cervical spine. Results may be improved by the use of intraoperative CT scanning and navigated percutaneous screw insertion, which allow optimization of the transpedicular trajectory.