Image-guided spine surgery: a cadaver study comparing conventional open laminoforaminotomy and two image-guided techniques for pedicle screw placement in posterolateral fusion and nonfusion models.

STUDY DESIGN: A randomized comparison of conventional and image-guided technology techniques for pedicle screw placement was performed.

OBJECTIVE: To evaluate the accuracy of thoracolumbosacral pedicle screw placement in simulated posterior fusion and nonfusion models via conventional and image-guided surgical techniques.

SUMMARY OF BACKGROUND DATA: Computer-assisted image-guided technology has been promoted as a means for theoretically improving the accuracy of spinal instrumentation placement, especially when visual landmarks are obscured.

METHODS: Seven embalmed cadaveric spines were cleared of all posterior soft tissue and mounted. The posterior elements of four spines were obscured so as to simulate a fusion mass using a synthetic bone cement. Three nonobscured spines also were instrumented. Pedicle screws were placed from T6 to S1 in two obscured specimens (24 screws) using a computer-assisted image-guided system, in one obscured specimen from T6 to S1 (12 screws) using a fluoroscopically assisted system, and in one obscured specimen from T6 to S1 (14 screws) using a conventional open laminoforaminotomy technique. In addition, pedicle screws were placed from T6 to S1 using a fluoroscopically assisted technique in two unobscured specimens (36 screws), and from from T6 to S1 (14 screws) via a laminoforaminotomy technique in one unobscured specimen. Pedicle violation was assessed by computed tomography scanning and direct visual inspection. The degree of screw misplacement noted visually was quantified with an electronic caliper.

RESULTS: Pedicle screws placed via open laminoforaminotomy resulted in a pedicle cortex breach rate of 21.43% in fused specimens and 14.29% in nonfused specimens. Screws placed in the nonfused model (two cadavers) via fluoroscopically assisted methods had pedicle cortical breaches, respectively, in 6.25% and 10% of the specimens, whereas the same method was noted to have a 8.33% violation rate in the fusion model. Finally, computed tomography-based image-guided placement through a simulated fusion mass resulted in no pedicle wall violations.

CONCLUSIONS: Accuracy of pedicle screw placement in the thoracolumbosacral spine is improved with the use of image-guided methods, particularly guidance by computed tomography. This is especially relevant clinically when the anatomy is obscured or altered as a result of inflammatory spondyloarthropathy (e.g., ankylosing spondylitis in which spontaneous fusions obscure surgical landmarks for pedicle access), or when used postsurgically in the setting of a posterolateral fusion.