Biomechanical evaluation of Caspar cervical screws: comparative stability under cyclical loading.

Anterior cervical instrumentation is used as an adjunct to bone fusion; however, definitive biomechanical data to support some applications and techniques are lacking. In the absence of supportive experimental data, posterior cortical penetration has been recommended with the Caspar system. Previously, we compared the axial pull-out strength of Caspar screws with and without posterior cortical penetration. This study compares the stability of unicortical versus bicortical screw penetration groups under cyclical loading simulating physiological flexion-extension. Caspar screws were placed in human cadaveric vertebrae with or without posterior cortical purchase. Each screw was separately tested, simulating flexion-extension to 200 cycles. Deformation time data allowed a direct comparison of screw "wobble" with and without posterior cortical purchase. The mean deformation differences between subcortical and bicortical groups were statistically significant and increased over time within both groups. Enhanced stability was noted with bicortical purchase throughout most of the examined range, becoming more pronounced over longer periods of cyclical loading. Significant (P < 0.05) increases in deformation over time were noted for both groups, suggesting potentially significant deterioration at the screw-bone interface, despite bicortical purchase. Such deterioration with repeated flexion-extension loading may be of concern in the use of Caspar plates in the presence of multicolumn instability.