This paper investigated the effect of rhBMP-2 use for spinal fusion in a rat model of spinal cord injury. The authors selected the worst-case scenario, in which the dura is lacerated irreparably and a hemisection injury of the spinal cord is performed. A posterolateral fusion was then performed with 43 μg of rhBMP-2 impregnated within an absorbable collagen sponge, and the rats were analyzed for improvement in neurological function. The authors found that the mean BBB score (Basso, Beattie, and Bresnahan scoring system, a neurological outcome measure) was initially worse for the rhBMP-2 group than for the control animals but became similar at later time points. However, there was increased histological evidence of scarring of the spinal cord, and there were subtle deficits in fine motor function in the rats treated with rhBMP-2, even at later time points. On the basis of these findings, the authors raised the question of whether similar negative effects may occur in humans. However, the answer to that question is not so clear.
When trying to translate this rat study to human clinical outcomes, several limitations of this animal model must be recognized. Rats have much better spontaneous recovery rates after spinal cord injury than humans, and therefore mechanisms that inhibit this more “rat-specific” recovery mechanism may not be directly relevant to humans. Rats and humans have very different sensitivities to rhBMP-2, with humans requiring much higher doses of rhBMP-2 than rats to get reliable spinal fusion. Furthermore, the carrier that is used with rhBMP-2 may alter the complication profile. It is known that the rhBMP-2 elution profile over time is highly affected by the carrier, and the current rat study uses an absorbable collagen sponge as opposed to the ceramic carrier used for posterolateral fusions with rhBMP-2 in Food and Drug Administration trials. Finally, this animal study is unable to replicate issues that could have an impact in trauma surgery such as increased length of time or blood loss in obtaining autograft.
Despite the limitations of this rat study, the authors raise important concerns for surgeons treating patients with spinal cord injury, especially in clinical scenarios that are very similar to those used in this study, namely, patients with the dura open and a lacerated spinal cord who have posterior spinal fusion in which rhBMP-2 on an absorbable collagen sponge is used. However, the answer to these concerns cannot be determined without further study. There are just too many unknowns and factors that are hard to simulate in animal studies. Ultimately, clinical studies need to be performed to address the concern raised by this investigation.