Variables affecting pedicle screw plate fixation of an unstable L3-L4 defect.

Fresh frozen human cadaveric spinal specimens (T8-S1) were subjected to pure flexion extension bending moment and pure axial torque loadings while intervertebral rotations were recorded at the L3-L4, L2-L3, and Ll-L2 discs. A standardized unstable defect was created at the L3-L4 disc, and loading tests were repeated after application of bilateral Steffee plates in 2 configurations: a short plate with 2 pedicle screws (spanning the defect) and a longer plate with 3 pedicle screws (spanning the defect and 1 disc above). Each plating configuration was tested in the unlocked state (nuts compressing the plate down onto the spine) and locked state (nuts above and below the plate tightened against each other to clamp the plate to the screws). Locking the plates to the screws had no effect on any intervertebral rotation at any disc level. Use of a longer plate that also spanned the disc above the defect offered no advantage in controlling flexion extension rotations at the defect site. However, mean torsional rotation at the defect site with the 3-screw plate was approximately 50% of the mean for a 2-screw plate. Extension and torsional rotations at the L2-L3 disc (1 level above the defect site) were unaffected by application of a 2-screw plate; flexion rotation at this level increased slightly after plating. All motions at the L2-L3 disc were reduced (as would be expected) when the 3-screw plate spanned this uninjured disc. Plating the defect had no effect on disc rotations at the L1-L2 disc (2 levels above the fracture site).