Paravertebral foramen screw fixation for posterior cervical spine fusion: biomechanical study and description of a novel technique.

OBJECTIVE Fusion surgery with instrumentation is a widely accepted treatment for cervical spine pathologies. The authors propose a novel technique for subaxial cervical fusion surgery using paravertebral foramen screws (PVFS). The authors consider that PVFS have equal or greater biomechanical strength than lateral mass screws (LMS). The authors' goals of this study were to conduct a biomechanical study of PVFS, to investigate the suitability of PVFS as salvage fixation for failed LMS, and to describe this novel technique. METHODS The authors harvested 24 human cervical spine vertebrae (C3-6) from 6 fresh-frozen cadaver specimens from donors whose mean age was 84.3 ± 10.4 years at death. For each vertebra, one side was chosen randomly for PVFS and the other for LMS. For PVFS, a 3.2-mm drill with a stopper was advanced under lateral fluoroscopic imaging. The drill stopper was set to 12 mm, which was considered sufficiently short not to breach the transverse foramen. The drill was directed from 20° to 25° medially so that the screw could purchase the relatively hard cancellous bone around the entry zone of the pedicle. The hole was tapped and a 4.5-mm-diameter × 12-mm screw was inserted. For LMS, 3.5-mm-diameter × 14-mm screws were inserted into the lateral mass of C3-6. The pullout strength of each screw was measured. After pullout testing of LMS, a drill was inserted into the screw hole and the superior cortex of the lateral mass was pried to cause a fracture through the screw hole, simulating intraoperative fracture of the lateral mass. After the procedure, PVFS for salvage (sPVFS) were inserted on the same side and pullout strength was measured. RESULTS The CT scans obtained after screw insertion revealed no sign of pedicle breaching, violation of the transverse foramen, or fracture of the lateral mass. A total of 69 screws were tested (23 PVFS, 23 LMS, and 23 sPVFS). One vertebra was not used because of a fracture that occurred while the specimen was prepared. The mean bone mineral density of the specimens was 0.29 ± 0.10 g/cm3. The mean pullout strength was 234 ± 114 N for PVFS, 158 ± 91 N for LMS, and 195 ± 125 N for sPVFS. The pullout strength for PVFS tended to be greater than that for LMS. However, the difference was not quite significant (p = 0.06). CONCLUSIONS The authors introduce a novel fixation technique for the subaxial cervical spine. This study suggests that PVFS tend to provide stronger fixation than LMS for initial applications and fixation equal to LMS for salvage applications. If placement of LMS fails, PVFS can serve as a salvage fixation technique.