Improved Techniques to Determine Pedicle Screw Placement Using Computed Tomography

STUDY DESIGN: A radiologic study of pedicle screws to determine their exact position in relation to the pedicle.

OBJECTIVES: To improve on the present technique of using bone windows to assess the position of pedicle screws by altering the window width and window level.

SUMMARY OF BACKGROUND DATA: Clinically, computed tomography has become the gold standard of evaluating pedicle screw position. This method of accurately assessing position of the screw has the highest sensitivity and specificity when compared to other imaging techniques. In a review of the literature, there were no specific guidelines for obtaining the optimum image.

METHODS: A cadaver spine, which had been stripped of soft tissue, was instrumented at six vertebrae (T12-L5) with twelve titanium pedicle screws (Medtronic, Sofamor Danek). A wide laminectomy was performed to allow direct visualization of the screws as they were placed in the spine. The medial wall was intentionally perforated by some of the screws. These perforations ranged from 1mm to 5mm of canal encroachment. The specimen was then placed in a CT scanner (General Electric HiSpeed Dx/i) and had 2mm cuts performed. The standard bone window was formatted and a new “metal window” was created by manually altering the window width and window level. Four blinded fellowship trained orthopaedic surgeons and two blinded radiologists then reviewed the scans.

RESULTS: The “metal windows” proved to be more accurate in determining screw placement when compared to the traditional bone windows.

CONCLUSIONS: By manually altering the window width and window level, a more accurate assessment of pedicle screw position can be obtained when compared to that seen with standard bone windows. By decreasing the amount of diffraction caused by the screw, the definition seen at the screw/bone interface is improved. This can be done by altering the existing scan and does not require any increase in patient exposure or scanner time.