Placement of thoracic transvertebral pedicle screws using 3D image guidance.

OBJECT: Transvertebral pedicle screws have successfully been used in the treatment of high-grade L5-S1 spondylolisthesis. An advantage of transvertebral pedicle screws is the purchase of multiple cortical layers across 2 vertebrae, thereby increasing the stability of the construct. At the lumbosacral junction, transvertebral pedicle screws have been shown to be biomechanically superior to pedicle screws placed in the standard fashion. The use of transvertebral pedicle screws at spinal levels other than L5-S1 has not been reported in the literature. The authors describe their technique of transvertebral pedicle screw placement in the thoracic spine using 3D image guidance.

METHODS: Twelve patients undergoing cervicothoracic or thoracolumbar fusion had 41 thoracic transvertebral pedicle screws placed across 26 spinal levels using this technique. Indications for placement of thoracic transvertebral pedicle screws in earlier cases included osteoporosis and pedicle screw salvage. However, in subsequent cases screws were placed in patients undergoing multilevel thoracolumbar fusion without osteoporosis, particularly near the top of the construct. Image guidance in this study was accomplished using the Medtronic StealthStation S7 image guidance system used in conjunction with the O-arm. All patients were slated to undergo postoperative CT scanning at approximately 4-6 months for fusion assessment, which also allowed for grading of the transvertebral pedicle screws.

RESULTS: No thoracic transvertebral pedicle screw placed in this study had to be replaced or repositioned after intraoperative review of the cone beam CT scans. Review of the postoperative CT scans revealed all transvertebral screws to be across the superior disc space with the tips in the superior vertebral body. Six pedicle screws were placed using the in-out-in technique in patients with narrow pedicles, leaving 35 screws that underwent breach analysis. No pedicle breach was noted in 34 of 35 screws. A Grade 1 (< 2 mm) medial breach was noted in 1 screw without clinical consequence. Solid fusion was observed across 25 of 26 spinal levels that underwent transvertebral screw placement including 7 spinal levels located at the top of a multilevel construct.

CONCLUSIONS: This report describes the authors' initial in vivo experience with the 3D image-guided placement of 41 thoracic transvertebral pedicle screws. Advantages of thoracic transvertebral screws include the purchase of 2 vertebral segments across multiple cortical layers. A high fusion rate was observed across spinal levels in which transvertebral screws were placed. A formal biomechanical study is needed to assess the biomechanical advantages of this technique and is currently being planned.