Biomechanical properties of threaded inserts for lumbar interbody spinal fusion

Calf and human cadaveric spines were used to determine motion segment stiffness and laxity after implantation of threaded inserts (the Ray Threaded Fusion Cage, Surgical Dynamics, Inc., Concord, CA), comparing direction of placement, number of implants, shape of the device, and integrity of anterior spine structures. Stiffness and laxity of spines with inserts were compared with those with bone grafts, with and without posterior fixation plates.

To determine the mechanical stabilizing properties of a threaded insert used for lumbar and lumbosacral fusion and the factors affecting stability.

Limited biomechanical information has shown that implantation of these devices adds stiffness to the lumbar spine, but little information is available concerning stiffness in loading directions other than flexion and extension, the effect on stiffness of position and number of implants, and the effect of this device on motion segment laxity.

Mechanical properties were determined by testing lumbar vertebral motion segments in flexion, extension, lateral bending, and torsion combined with axial compressive loading. Stiffness (slope of the load/deflection curve) and neutral zone angle or laxity (angular displacement of the vertebra from no load to 1.0 Nm moment) were determined. Initial tests were performed on calf lumbar vertebrae to determine the effects of placement and number of inserts. Comparisons of bone grafts and inserts with and without supplemental plates were made using human lumbar spines. Cylindrical- and conical-shaped inserts, when placed from anterior, were tested in calf spines. The load-bearing capacity of the insert supported in calf vertebral body bone was determined.

There was no significant effect of placement of inserts in different orientations (lateral, posterolateral, or posterior) on stiffness, except in torsion where posterior placement damaged facets or lamina, reducing stiffness. Placement of two inserts from posterior decreased flexion and lateral bending laxity compared with the intact motion segment. Compared with intact, bone grafts produced more stiffness only in lateral bending and had no effect on laxity. Supplemental posterior plates fixed by pedicle screws across the fusion segment increased flexion and lateral bending stiffness and reduced laxity in flexion, extension, and lateral bending. Conical-shaped inserts placed from anterior into cylindrical holes distracted soft tissue structures, decreasing laxity. Cutting the anterior structures increased laxity by relieving some tissue tension caused by distraction. The mean maximum compressive load that could be supported by the insert was 2998 N (standard deviation = 980 N). Structural failure occurred in the supporting bone.

Threaded inserts increase vertebral motion segment stiffness and decrease laxity by distracting intervertebral structures. They are not sensitive to placement, except if vertebral structures are injured during insertion and produce constructs with more consistent mechanical properties than bone grafts.