The atlantoaxial joint is the most mobile joint of the spine because of the horizontally oriented facet joints, the lax ligaments and the robust synovium that allow hypermobility (1). But these anatomic features also predispose this complex to a disorder called atlantoaxial rotatory fixation (AARF). In children, the larger head-to-body mass ratio and the undeveloped bone structures exacerbate the incidence of AARF (2). Children with AARF typically present with torticollis, facial asymmetry with oral malocclusion, and limited neck motion (3). The primary treatment of AARF in children is conservative, while surgery is recommended when conservative treatment fails (4).

With the development of modern techniques and instruments, the atlantoaxial posterior screw-rod technique has become the gold standard of the posterior fixation technique for AARF (5). This has been widely recognized by spine surgeons because of its advantages in long-term stability, high fusion rates, and excellent correction (6, 7). However, the placement of the pedicle screw is technically challenging for pediatric patients due to the small size of the pedicle, high variation, and high risk of neurovascular injury (8). Different intraoperative image-guided systems have been developed to reduce the risk of screw malposition and achieve safe fixation in cervical spine surgery (9). Traditionally, intraoperative C-arm fluoroscopy has been used during surgery. However, the pitfalls of C-arm fluoroscopy include single-plane imaging, insufficiently clear imaging, and high occupational radiation exposure (10). Therefore, just as surgical techniques have advanced over the years, imaging modalities have also.

The O-arm navigation system (Medtronic® Sofamor Danek, Minneapolis, MN, USA), a three-dimensional intraoperative imaging device, has gained worldwide popularity since its first introduction in 2008 (11, 12). However, there are no reports on the feasibility and efficacy of O-arm navigation-guided pedicle screw placement in the treatment of children with atlantoaxial rotatory fixation. In this study, we compared O-arm navigation with C-arm fluoroscopic guidance in surgical outcomes of children with atlantoaxial rotatory fixation.

We retrospectively reviewed the medical records of consecutive patients who underwent C1–C2 fusion between April 2014 and December 2020. The study was reviewed and approved by the Institutional Review Board of Second Affiliated Hospital of Naval Medical University. All patients were subjected to instrumentation utilizing the Resnick and Benzel's procedure (6). for C1–C2 fixation performed under fluoroscopic guidance of the C-arm or navigation of the O-arm.

Eighty-five patients who underwent C1-2 fusion were included in the present study: 46 in the C-arm and 39 in the O-arm groups. Mean age at surgery was 9.67 years (range, 6–15). The mean follow-up was 12 months (range 13–16) (Table 1).

A total of 340 screws were inserted. On postoperative CT-scan, the screw accuracy rate of the O-arm group was 97.4%, which was significantly higher than that of the C-arm group (91.8%, p = 0.025), details shown in Table 2.

EBL showed a statistical difference between C-arm and O-arm groups with 230.0 and 150.6 ml respectively, p < 0.05. No statistically significant difference (122.0 ± 16.5 min vs. 110.0 ± 14.4 min, p = 0.604) was observed regarding the median operative time between the two groups, details shown in Table 3.

No serious complications such as vascular complications, spinal cord injury, or cerebrospinal leak occurred and, the postoperative course was uneventful for all patients. The average length of stay was 3.5 days (range 3–6). Both groups had satisfied bony fusion (100%) at 12 months of follow-up.

In this study, we showed that the accuracy of pedicle screw placement in C1-2 fusion surgeries using the O-arm guidance was higher than those using C-arm. Ling et al. reported that 98% of their screws were placed in ideal positions in 21 patients treated with a Harm's construct or occipital cervical fusion using the O-arm (16). We achieved similar results in the present study. With the use of intraoperative navigation, the trajectory of each screw could be easily visualized to ensure that nerve roots, spinal cord or vertebral arteries were not damaged.

Another benefit of O-arm navigation was that it could reduce intraoperative blood loss. Hitti et al. (17) reported a significant reduction in EBL, by more than 50% using O-arm navigation in C1-2 posterior cervical fixation, which was consistent with our results. This finding was likely explained by the minimal disturbance of the cervical venous plexus. Typically, surgeries performed with navigation required a limited exposure due to image guidance of the hardware placement and, therefore, less need for extensive dissection for direct visualization.

More importantly, as the pedicle size in children is small, repeated adjustment during screw placement could lead to screw loosening, which would seriously affect the pullout strength. Navigation with the O-arm provided the precision to insert all C1 and C2 screws in a single path, which technically also translated into greater bony purchase and biomechanical stability. Additionally, because of interference from the mandible and teeth that overlap, conventional fluoroscopic techniques cannot clearly assess the degree of reduction due to the relationship between the odontoid and the lateral mass. The advent of O-arm navigation has overcome this shortcoming by providing axial CT images during navigation, which clearly shows the position of the odontoid in relation to the lateral mass, as well as the foramen transversarium and spinal canal in relation to the pedicle screws. Another point, when navigated with fluoroscopy, interpretation of the screw trajectory requires constant switching between the lateral and antero–posterior views. O-arm navigation based on simultaneous guidance by axial, sagittal, and coronal views allowed the positions of the screws to be updated in real time as they advanced into the vertebrae. Therefore, we could check the trajectory and position of the screw to reduce the possibility of reoperation due to mispositioned screws.

Prolonged operative time was another potential problem with O-arm navigation due to the time required to set up and scan. Hitti et al. reported that posterior fixation of C1–C2 with O-arm navigation (198.4 ± 13.56 min) takes longer than with the C arm fluoroscopy (156.9 ± 10.03 min) (17). However, they also reported that with increased experience and familiarity with navigation equipment, there was a decrease in the duration of the procedure and eventually, similar operative times were spent between the surgery with O-arm navigation and that with the C arm fluoroscopy. In this study, there were also no statistically significant differences in median operative time between the groups. The reason for this was likely that our surgeons minimized the exposure of bony landmarks for screw placement and insert the screws with confidence using the O-arm navigation. Fixation assisted by C-arm fluoroscopy required meticulous dissection of bony landmarks such as the lateral portion of the posterior arch of C1 and the isthmus of C2. O-arm-assisted fixation could cut these steps short and reduce operative time.

This study was limited by its retrospective design and the non-randomization of patients. Since the introduction of the O-arm in our institution, only a few C-arm surgeries have been performed. Furthermore, surgery for pediatric C1-2 fusion was relatively rare and required a steep learning curve, which was a limitation for large-scale studies. Further studies will be required to reach a more definitive conclusion.

O-arm-assisted navigation allowed more accurate screw placement and less intraoperative blood loss in pediatric C1-2 fixation, compared to a conventional fluoroscopic guided device. Using O-arm navigation did not prolong the operative time despite the time required for setting and scanning.