A paradigm of delayed union and nonunion in the lumbosacral joint : a study of motion and bone grafting of the lumbosacral spine in sheep

In a group of nine sheep (Group A), it was noted that when small, interlocking cancellous and cortical cancellous bone grafts are placed posteriorly on the lumbosacral spine, union always occurred in the interlumbar motion segments and almost never occurred at the lumbosacral joint. One of the main differences in these two areas is the amount of motion that occurs at each level with flexion and extension. Because nonunion following bone grafting for arthrodesis of the spine is a serious clinical problem, we have studied the amount of motion seen at the interlumbar and lumbosacral joints in sheep to ascertain how much motion is compatible with union and how much is associated with nonunion. In vivo studies were carried out in eight sheep (Group B), and five normal spine segments were studied in vitro to determine normal motion in this species. To simplify the complex in vivo motion that occurs at the lumbar motion segments, the simple linear displacement and strain of the fusion mass (consisting of fibrous tissue and bone grafts placed on and between the laminas posteriorly) was measured with the spine in flexion and extension. When the displacement and strain at the interlaminar level of the L6-S1 joint was measured, the linear displacement was found to be 5.2 mm and the associated linear strain 36%. The displacement at the L5-L6 interspace was 1.2 mm, and the strain 10%. The stiffness of the L5-L6 joint (which always fused) and the L6-S1 joint (which did not fuse, with one exception) were also studied. In a third group of four animals (Group C), internal fixation of the lumbosacral joint was attempted in addition to bone grafting. In this group bone grafts were put only across the lumbosacral joint. Two of these animals developed serious problems in the first 2 weeks after surgery (one fractured its ilium, and the other developed a deep infection) and were killed before the arthrodesis could become solid. The two other sheep in this group were killed at 6 months, and it was noted that the one fixed with distraction rods united its lumbosacral joint, whereas the one fixed with plates did not unite.