Atlantoaxial fixation using the polyaxial screw-rod system.

The aim of this study is to evaluate the first results of the atlantoaxial fixation using polyaxial screw-rod system. Twenty-eight patients followed-up 12-29 months (average 17.1 months) were included in this study. The average age was 59.5 years (range 23-89 years). The atlantoaxial fusion was employed in 20 patients for an acute injury to the upper cervical spine, in 1 patient with rheumatoid arthritis for atlantoaxial vertical instability, in 1 patient for C1-C2 osteoarthritis, in 2 patients for malunion of the fractured dens. Temporary fixation was applied in two patients for type III displaced fractures of the dens and in two patients for the atlantoaxial rotatory dislocation. Retrospectively, we evaluated operative time, intraoperative bleeding and the interval of X-ray exposure. The resulting condition was subjectively evaluated by patients. We evaluated also the placement, direction and length of the screws. Fusion or stability in the temporary fixation was evaluated on radiographs taken at 3, 6, 12 weeks and 6 and 12 months after the surgery. As concerns complications, intraoperatively we monitored injury of the nerve structures and the vertebral artery. Monitoring of postoperative complications was focused on delayed healing of the wound, breaking or loosening of screws and development of malunion. Operative time ranged from 35 to 155 min, (average 83 min). Intraoperative blood loss ranged from 50 to 1,500 ml (average 540 ml). The image intensifier was used for a period of 24 s to 2 min 36 s (average 1 min 6 s). Within the postoperative evaluation, four patients complained of paresthesia in the region innervated by the greater occipital nerve. A total of 56 screws were inserted into C1, their length ranged from 26 to 34 mm (average, 30.8 mm). All screws were positioned correctly in the C1 lateral mass. Another 56 screws were inserted into C2. Their length ranged from 28 to 36 mm (average 31.4 mm). Three screws were malpositioned: one screw perforated the spinal canal and two screws protruded into the vertebral artery canal. C1-C2 stability was achieved in all patients 12 weeks after the surgery. No clinically manifested injury of the vertebral artery or nerve structures was observed in any of these cases. As for postoperative complications, we recorded wound dehiscence in one patient. The Harms C1-C2 fixation is a very effective method of stabilizing the atlantoaxial complex. The possibility of a temporary fixation without damage to the atlantoaxial joints and of reduction after the screws and rods had been inserted is quite unique.