FE-Analysis of Stabilization of Cervical Spine Fracture in Ankylosing Spondylitis.

Abstract: Ankylosing Spondylitis (AS), or Bechterew’s disease, is an inflammatory rheumaticdisease that through the formation of additional bone tissue in the spine eventuallyleads to the complete fusion of the vertebrae, in effect turning the spine into one longbone. Due to the reduced flexibility of the spine with the long lever arms, spinalfractures in AS-patients are relatively common even after minor trauma. The aim of this thesis was to use an existing finite element model of a healthy spineand adapt it to the conditions of AS, thus gaining some insight into the effects ofsurgical stabilization of cervical fractures, using posterior screws and rods. Althoughthis type of surgery is often performed, it has not been previously investigated in abiomechanical model. This thesis should be considered as a starting point for how afinite element model of the spine could be used to investigate the effect of spinalimplants in the case of a fracture in the ankylosed spine. An existing FE-model was modified to some of the conditions of AS: The vertebraewere fused by adding ossifications at the intervertebral discs (with the Head-C1 andC1-C2 joints left mobile). A fracture was simulated at the C6C7 disc level. Fourdifferent implant configurations were tested: Short instrumentation C6C7, mediuminstrumentation C5toT1, long instrumentation C3toT3, and a long instrumentationC3C6C7T3 with skipped intermediate levels. Three loads (1.5g, 3.0g, 4.5g) wereapplied according to a specific load curve. Kinematic data such as the gap distance inthe fracture site were obtained. Furthermore the stresses in the ossified parts of thediscs were evaluated. It was shown that the chosen methods of adapting the model to the AS conditions, andmodeling the fracture and implant, changed the kinematics so that less movementoccurred between the vertebra, which is typical for AS. Measured as fracture gap, alltested implant configurations were equally good at stabilizing the fracture, althoughthey all allowed more movement than the non-fractured AS-model did. All implantconfigurations were also able to stabilize the fracture in terms of the horizontal translation in the fracture. The disc ossifications were somewhat shielded from stress for those ossifications that were within the range of the implant. This was so for all implant configurations. No increased stress was observed in the ossifications immediately outside the range for the implants, relative the non-fractured AS-model. For the C6C7 and C5toT1 implant configurations as well as the non-fractured ASmodel,the stresses were highest at the T1T2 level. Stresses in the ossifications in the thoracic spine were generally low, apart from the T1T2 level. The results show that the chosen AS-adaptations and the modeled implant seem reasonable for testing some of the considerations of cervical fractures in the ankylosed spine as well as for some implant configurations. The results also make it possible to speculate about the optimal type of implant. The effects of screw placement and anchoring, osteoporosis, muscle activation and possible spinal deformity on the implant stability were not investigated, and should be a matter for further studies.