Increasing loads and diminishing returns: a biomechanical study of direct vertebral rotation.

Abstract

Biomechanical simulation of DVR and pure-moment testing on thoracic spines. Characterize load-deformation response of thoracic spines under DVR maneuvers until failure, and compare to pure-moment testing of same spines. Despite reports of surgical complications, few studies exist on increase in ROM under DVR torque. Biomechanical models predicting increases from surgical releases have consistently used "pure-moments", a standard established for non-destructive measurement of ROM. Yet, DVR torque is not accurately modeled using pure moments and, moreover, magnitudes of torque applied during DVR maneuvers may be substantially higher than pure-moment testing. Cadaveric thoracic spines (N = 11) were imaged, then prepared. Polyaxial pedicle screws were implanted at T7-T10 after surgical releases. Bilateral facetectomies and Ponte osteotomies were completed at T10-T11. A custom apparatus, mounted into an 8-dof MTS load frame, was used to attach to pedicle screws, allowing simulation of surgical DVR maneuvers. Motions of vertebrae were measured using optical motion tracking. Torque was increased until rupture of the T10-T11 disc or fracture at the pedicle screw sites at any level. The torque-rotation behavior was compared to its behavior under pure-moment testing performed prior to the DVR maneuver. Under DVR maneuvers, failure of the T10-T11 discs accompanied in most cases by pedicle screw loosening, occurred at 13.7-54.7Nm torque, increasing axial rotation by 1.4°-8.9°. In contrast, pure-moment testing (4Nm) increased axial rotation by only 0.0°-0.9°. DVR resulted in substantially greater correction potential increases compared to pure-moment testing even at the same torque. These results suggest increased flexibility obtained by osteotomies and facetectomies is underestimated using pure-moment testing, misrepresenting clinical expectations. The present study is an important and necessary step toward the establishment of a more accurate and ultimately surgically applied model. III.