Radiological evaluation of atlantoaxial fusion using C2 translaminar screws and C2 pedicle screws: Does the screw halo sign imply fusion failure?

The purpose of this study was to identify the criteria for atlantoaxial (AA) fusion by comparing follow-up lateral radiographs and computed tomography (CT) images. We retrospectively analyzed data from 161 consecutive patients undergoing AA fusion. Patients with a minimum of 1 year of CT follow-up after AA fusion surgery using C2 pedicle screws or translaminar screws (C2TLS) were included. Patients were followed up radiographically at 3, 6, and 12 months after surgery, and dynamic lateral radiographs were also evaluated. A total of 49 patients were analyzed, with a mean CT image follow-up of 41.6 ± 37.6 months. Thirty eight patients had C2 pedicle screw placement, and 11 patients underwent planned C2TLS. AA fusion with bridging bone mass formation was achieved in 45/49 (91.8%) patients. Screw halos were observed in 14/49 (28.6%) patients. Among them, final fusion failure occurred in 2 (14.3%) patients. The last follow-up CT showed no difference in the fusion failure rate according to the presence or absence of a screw halo (no halo, 5.7%; halo, 14.3%; P = .33). The differences in C1-2 segmental angles (SA) in flexion-extension dynamic lateral radiographs were 1.99 ± 1.62° in the fusion group and 4.37 ± 2.13° in the non-fusion group (P = .01). The likelihood of fusion failure increased when the SA gap was greater than 2.62° (P = .05). C2TLS placement had a significantly higher incidence of screw halos. However, the halo sign was not significantly related to final bone fusion. Bone fusion could be predicted when the SA gap of C1-2 was less than 2.62° on the dynamic radiograph.