Augmentation of an anterior solid rod construct with threaded cortical bone dowels. A biomechanical study.

Abstract

This static, nondestructive, in vitro biomechanical study examines anterior solid rod construct stiffness following the addition of multilevel, threaded cortical bone dowels in a bovine model. A comparison is made with a clinically relevant posterior construct with and without an anterior release. To determine if the addition of solid, multilevel disc space implants will increase construct rigidity, while maintaining or enhancing anterior column length. Anterior instrumentation for thoracolumbar and lumbar scoliosis has achieved greater correction and preserved distal motion segments; however, kyphosis over the instrumented segments and nonunion have been observed more frequently than with posterior segmental spinal instrumentation. Fifteen calf spines underwent mechanical testing. Group A (n = 7) included anterior constructs: 1) intact, 2) anterior release/rod/rib graft (L2-L5), and 3) anterior release/rod/dowels (L2-L5). Group B (n = 8) included posterior constructs: 1) intact, 2) posterior rod without anterior release (T13-L5), 3) posterior rod (T13-L5)/anterior release/rib graft (L2-L5). The protocol included axial compression (-600 N), axial rotation (+7 Nm), flexion/extension (+7.5 Nm), and lateral bending (+7.5 Nm). An anterior extensometer measured segmental displacements to calculate construct stiffness. Lateral radiographs evaluated alignment for the anterior constructs. Statistical analysis involved a one way analysis of variance (ANOVA) and a Student-Newman-Keuls post hoc test. All reconstructions restored stiffness to intact values with the exception of the dowels alone in axial rotation. The rod/dowel construct was stiffer than all other groups in axial compression, flexion/extension, and lateral bending, with the exception of the posterior rod without discectomy, which was superior in flexion and statistically similar in extension, lateral bending, and axial rotation. The anterior construct with rib graft was equivalent to the posterior construct with rib graft in all modes of testing. The dowels created greater lordosis than the bicortical rib grafts. Disc space augmentation increased stiffness except in axial rotation, in which values were restored to the intact level. Stiffness was superior to a clinically relevant posterior instrumentation comparison group following anterior release, and was equivalent to a posterior construct without anterior release except in anterior flexion. In addition, the implants enhanced lordosis. Increased rigidity should improve rates of arthrodesis, while maintenance of sagittal alignment may prevent pathologic compensatory curves in adjacent spinal segments. Further research is required to determine the optimal method of achieving structural interspace support.