Study design: This was a randomized, blinded trial of the safety of the application of recombinant human bone morphogenetic protein (rhBMP)-2 or autologous bone graft onto a laminectomy defect of the dog in the presence or absence of a dural membrane puncture.
Objective: To test the safety of rhBMP-2 in an application in which direct contact of the material with neural tissue occurs.
Summary of background data: Application of rhBMP-2 in laboratory animals stimulates local bone formation to effect spinal fusion and healing of segmental bone defects. The use of rhBMP-2 as a bone graft substitute in spinal fusion would eliminate donor site morbidity and may augment the rate of successful fusion. Because rhBMP-2 may unintentionally come in contact with neural tissue, the consequences of such a safety issue must be addressed in an animal model before human trials.
Methods: Twenty skeletally mature beagles underwent spinal exposure followed by bilateral laminectomy at L5. In half of the dogs, a puncture wound was made to the dura with the expression of cerebrospinal fluid at the site of the puncture. In randomly selected animals, the exposed dural elements received either autologous bone graft with the bone removed from the laminectomy site or an implant of the rhBMP-2 device. The animals was observed for 12 weeks with periodic clinical examinations and monthly computed tomographic scans.
Results: There was no clinical, radiographic, or histologic evidence of neurologic abnormalities in these animals. The rhBMP-2 stimulated bone growth in the laminectomy defect and came into direct contact with the dural membrane. There was no evidence of abnormal mineralization within the thecal sac or in the spinal cord itself.
Conclusions: The rhBMP-2 implant stimulated bone formation in the laminectomy site. Neither autologous bone, rhBMP-2, nor the dural puncture had deleterious consequences for the animals.