#359 Biomechanics of a New Dynamic Lumbar Posterior Stabilizer

Oral Posters: Lumbar

Presented by: N.R. Crawford


J.F. Zucherman (1)
K. Hsu (1)
L. Perez-Orribo (2)
P.M. Reyes (2)
N.R. Crawford (2)

(1) St. Mary's Spine Center, San Francisco, CA, USA
(2) Barrow Neurological Institute, Spinal Biomechanics, Phoenix, AZ, USA


Introduction: Dynamic instrumentation for stabilizing the lumbar spine may prevent fusion-related adjacent segment disease (ASD); it may replace fusion in some situations; it may diminish hardware failure. Two specific scenarios of benefit are 1-level moderate instability where the stabilizing tissues are relatively incompetent, and multi-level fusion where the last instrumented level requires intermediate stiffness (“topping off”) to prevent transfer of high stresses from the fusion mass to the intact adjacent levels. Both scenarios were evaluated in vitro to determine the biomechanical response of a new pedicle screw-based dynamic stabilizer (Staflex, Spartek, Inc., Concord, CA).

Materials and Methods: Seven human cadaveric L2-S1 segments (47-59 years) were tested (1) intact, (2) after medial facetectomy and discectomy to achieve 50% increase in range of motion (ROM), (3) after 2-level hybrid posterior fixation, consisting of bilateral dynamic pedicle screws at L4 interconnected with rigid rods to standard pedicle screws at L5 and S1, (4) after 2-level rigid fixation, achieved by applying a locking collar to the dynamic pedicle screw to simulate a rigid screw, (5) after 1-level dynamic fixation, achieved by resizing the interconnecting rods to span only L4-L5, and (6) after 1-level rigid fixation, achieved by reapplying the locking collar at the L4 screw. In each condition, pure moments (7.5 Nm maximum) were applied to induce flexion, extension, lateral bending, and axial rotation. Angular motion was tracked optoelectronically. Sagittal instantaneous axis of rotation (IAR) was also calculated.

Results: Both the dynamic and rigid 1-level constructs significantly reduced ROM relative to the destabilized condition (Fig. 1). In 1-level constructs, the percentage of intact motion allowed by dynamic vs. rigid fixation was 86% vs. 46% during flexion, 94% vs. 49% during extension, 95% vs. 35% during lateral bending, and 128% vs. 64% during axial rotation (all p< 0.001). Relative to the intact IAR location at L4-L5, the IAR was shifted significantly farther posterior by rigid 1-level instrumentation than dynamic 1-level instrumentation (p=0.005). In 2-level constructs, the dynamic level (L4-L5) allowed significantly greater normalized ROM than the rigid level (L5-S1) in all directions (p< 0.02) but allowed significantly less ROM than the intact level (L3-L4) in all directions except axial rotation (p=0.13, Fig. 2).

Discussion: In 1-level constructs, dynamic instrumentation caused significantly less of a posterior shift in IAR position than fusion, indicating less impact on kinematics. Compared to other reported 1-level dynamic devices in the literature, this dynamic stabilizer provided 1-level ROM closer to intact during all loading modes. In the “topping off” construct, the dynamic stabilizer allowed intermediate ROM to give well balanced transitional flexibility.

Fig 1. Mean normalized L4-5 ROM (1-Lvl constructs)

Fig 2. Mean normalized ROM (2-level constructs)
D2 = Contained disc herniation
D3 = Free fragment disc herniation

F= Facet joint