Study design: A prospective study of the accuracy of thoracic pedicle screw placement in patients with idiopathic scoliosis.
Objectives: To evaluate the accuracy of thoracic pedicle screw placement in the surgical management of idiopathic scoliosis and to establish its risks and benefits.
Summary of background data: Lumbar pedicle screw instrumentation has proven to be reliable and effective in the surgical management of scoliosis. No reports exist on the accuracy and benefits of pedicle screw instrumentation of the thoracic spine in scoliosis surgery.
Methods: One hundred and twenty thoracic pedicle screws in 32 consecutively treated patients with idiopathic scoliosis were investigated immediately after surgery by computed tomography scans that were analyzed by three examiners.
Results: Thirty (25%) of the screws penetrated the pedicle cortex or the vertebral body anterior cortex. Ten screws (8.3%) penetrated the medial cortex of the pedicle by an average of 1.5 mm and a maximum of 3.0 mm. Seventeen screws (14.2%) penetrated laterally by an average of 2.1 mm. There were two cases of caudad penetration. Three screws penetrated the anterior vertebral cortex, of which two also penetrated the pedicle cortex. Also, one of these three screws was replaced because of its direct proximity to the thoracic aorta. There were no neurologic complications. The correlation between the pedicle cortical penetration rate and the preoperative Cobb angle, vertebral rotation or level, or site of screw insertion was statistically insignificant (P > 0.05). Curve correction in the cases of mainly hook instrumentation averaged 52.5% versus 59.2% in the cases of mainly screw instrumentation. This difference was statistically insignificant (P > 0.05).
Conclusions: Pedicle or vertebral body cortical penetration occurred with 25% of the screws but with no neurologic compromise. Curve correction was slightly greater than with hooks, but not to a statistically significant extent.