BMP-Induced Anterior Spine Fusion in a Thoracoscopically Instrumented Animal Model

· (a - Medtronic Sofamor Danek)

Texas Scottish Rite Hospital, Dallas, TX, USA

PURPOSE: Thoracoscopically-assisted anterior spinal fusion and instrumentation for idiopathic scoliosis has taken on greater clinical applications. However, harvesting autologous bone adds operative time and morbidity. The purpose of this study was to compare autologous iliac crest and rib graft with recombinant human BMP-2 (rhBMP-2) in a thoracoscopically-instrumented animal spinal fusion model.

METHODS: 16 pigs were anesthetized intubated and underwent thoracoscopically-assisted instrumentation and fusion. Five contiguous thoracic discectomies were performed followed by anterior instrumentation using a 4.5 stainless steel rod and 5.5 and 6.5 mm screws. The animals were randomly assigned to four groups: Group I: Control- no graft material; Group II: Rib autologous graft; Group III: Iliac crest autologous graft; Group IV: rhBMP-2. The animals were euthanized at 4 months and the spines were harvested. Fusion mass was assessed using CT scan sections in the sagittal and axial planes. The spines were biomechanically tested with incremental loads applied in the frontal and axial planes to achieve bending moments up to 6.0 N-m. Angular motion at each segment was recorded using a 3-dimensional motion analysis system. The spines were then fixed/embedded and 30µm sections were made in the mid-sagittal plane. Histomorphometric analysis of each disc segment was performed.

RESULTS: All animals were successfully instrumented using the thoracoscopic approach. Operative time averaged 214 minutes with an average blood loss of 112 cc. The operative times and blood loss were less in the control group when compared to the remaining 3 groups. The CT fusion grade using a 4-point system was I: 0.6, II: 2.1, III: 2.3, IV: 3.8. (p ‹ 0.001). In axial rotation, the rhBMP-2 group was stiffer than all three groups (p ‹: 0.05); the rib and iliac crest group were similar, and both were stiffer than the control group. In lateral bending, rhBMP-2 › › Iliac Crest › Rib › Control.(p ‹ .001) Histologic analysis demonstrated bone volumes, expressed as a percentage of the total disc space area as- group I: 47.8%; II: 62.6%; III: 59.6% and IV: 95.2% (p ‹ 0.001).

CONCLUSIONS: rhBMP-2 significantly enhanced anterior spine fusion in a thoracoscopically-assisted instrumented animal model at 4 months when compared to autologous iliac crest and rib graft. The fusion mass was confirmed by histologic analysis and provided significantly greater torsional and coronal plane bending stiffness. The use of rhBMP-2 in the thoracoscopic approach to idiopathic scoliosis would save operative time, decrease morbidity, enhance fusion mass and may obviate the need for post-operative bracing in the clinical setting.