Modeling the Spine, Cord and All

When the bones and discs of the spinal column are broken, crushed, or displaced, the spinal cord itself may be devastatingly damaged. Now, a new computer model suggests that the manner in which the injury occurred may affect the spinal cord in distinct and significant ways.

This work could have a wide-reaching impact on spinal treatment, says Thomas Oxland, PhD, professor of orthopaedics and mechanical engineering at the International Collaboration on Repair Discoveries (ICORD) Centre at the University of British Columbia. If cord injuries could be subclassified by type, it is possible that physicians may be able to treat them differently. Oxland was lead author of the work, published in Annals of Biomedical Engineering in March 2008.

Before modeling the human spine, Oxland’s team, which included his master’s student Carolyn Greaves, and Mohamed Gadala, PhD, a professor of mechanical engineering, had already begun animal studies to examine the relationship between the type of spinal column injury and the strain on the cord. But they wanted to compare their animal data to the human spine. Because it’s impossible to use human experimental models, the group simulated the spine and the spinal cord using data from the Visible Human project.

Like others who have modeled the spine, Oxland and his colleagues created a finite element model of the human cervical (neck) spine. They then simulated injury to it by applying engineering torques, not unlike those used to study the strain on a bridge. What’s new here is that they observed the effect of different types of injuries on the spinal cord itself. The result: distinct patterns of strain and deformation depending on whether the spine suffered a burst fracture, a dislocation, or a stretching injury. The work stopped short of examining actual cord damage but, Oxland says, “one would expect that [these mechanisms] would produce very different patterns of damage in the cord.”

Oxland acknowledges that their now-static model cannot yet capture the dynamic forces at work when a real-life injury happens, often in a fraction of a second. His team is working on introducing more variables and lifelike properties now. He also plans to match up the simulation results with his lab’s animal experiment data to better understand cord damage.

David Shreiber, PhD, an assistant professor of biomedical engineering at Rutgers University, thinks this model will help advance the field—one that still lags behind brain injury research. “It’s significant because it’s the foundation of more work on injury to the cord,” says Shreiber. The model is flexible enough that it can be used to understand many types of injuries. “The nice thing about this computational system is that you can apply the loading conditions however you want—you can look at twisting, at pressure applied internally, and other cases of spinal injury,” he adds.