Study design: A retrospective study of the accuracy of the pedicle screw placement in the thoracic spine using the Funnel technique in idiopathic scoliosis was conducted by using CT-scan analysis.

Objective: To evaluate the accuracy of thoracic pedicle screw placement using the Funnel technique in the surgical management of idiopathic scoliosis.

Background data: Placement of thoracic pedicle screw especially in idiopathic scoliosis is technically challenging. CT scan navigator is time-consuming and very expensive. Funnel technique was developed to locate the position of the pedicle and without using the CT guided navigator. There are no reports on the accuracy of pedicle screw instrumentation of the thoracic spine using the funnel technique in scoliosis surgery.

Material and Method: 117 screws in 14 patients with idiopathic scoliosis were investigated by computed tomography. Screw positioning was analyzed based on each spinal level, side of deformities (convex or concave side) and direction of cortical penetration.

Result: Forty-one screws (35%) were placed within the pedicle. Forty-five screws (38%) breeched the medial cortex of pedicle and thirty-one screws (27%) breeched the lateral cortex of pedicle. The percentage of screws totally contained within the pedicle also varied by the thoracic regions: 43% was in between T1-T4, 29% was in between T5-T8, and 37% was in between T9-T12 level. However, there was no statistical difference between the spinal regions and the accuracy rate. Fifty-four screws were placed on the convex side of the spine and sixty-three screws were placed on the concave side. The percentage of totally contained within the pedicle in the convex and the concave side were 20% and 48% respectively. It had statistical difference (p = 0.004). Although medial perforation of the pedicle wall occurred in 38%, there were only 6.7% (3/45) of these that had canal encroachment of more than 4 mm. All of these occurred on the convex side. 62% and 31% of screws with medial perforation were less than 2 mm and 2.0-4.0 mm of canal intrusion respectively. Among the lateral penetration, 42% of these screws penetrated <2 mm., 48% penetrated 2.0-4.0 mm., and 10% penetrated more than 4 mm. There was 3.4% (4/117 screws) that did not purchase the anterior portion of vertebral body. Although the percentage of totally contained screws was low, there was an 82.1% acceptable rate of screw position. These are screws that were fully contained within the pedicle plus medial perforation less than 2 mm. plus screws that had lateral penetration but purchased into the vertebral body). No screws perforated the anterior cortex of the vertebral body. There were no neurovascular complications.

Conclusion: Placement of the thoracic pedicle screws using the Funnel technique in idiopathic scoliosis had an accuracy of 82.1%(screws that were in acceptable position). There were no neurovascular related-complications by using this technique in the present study.

Keywords: Accuracy, Pedicle screw, Thoracic spine, Scoliosis