Spontaneous atlantoaxial facet joint autofusion after posterior C1-2 fusion

Posterior C1-2 fusion using the Goel-Harms construct is widely performed for atlantoaxial instability; however, computed tomography (CT) frequently demonstrates incomplete posterior graft union despite stable radiographic alignment. Spontaneous facet joint autofusion (SFA) across the C1-2 facet joints may represent an underrecognized mechanism contributing to postoperative stability.

To determine the proportion of patients with SFA following posterior C1-2 fusion and to identify patient- and surgery-related factors associated with its development, with particular emphasis on preoperative C1-2 facet osteoarthritis (OA).

Retrospective cohort study at a single tertiary referral center.

Fifty-three consecutive patients (median age: 56.0 years) who underwent posterior C1-2 fusion with the Goel-Harms technique and had at least 24 months of clinical and radiologic follow-up.

Radiological outcomes included posterior graft union, SFA on CT, and segmental stability on flexion-extension radiographs. The clinical outcomes included the Neck Disability Index, Japanese Orthopaedic Association score, EuroQol-5D, improvement rate, and pain visual analog scale.

All patients underwent CT and dynamic radiography at 12 months postoperatively, and those without definite graft union were re-evaluated at 24 months. Posterior graft union was defined as continuous trabecular bridging between the C1 posterior arch and the C2 lamina on CT. SFA was defined as continuous osseous bridging across the C1-2 facet joint. Stable fibrous union was defined as absence of osseous union on CT, with no implant failure and no segmental instability on dynamic radiographs (C1-2 segmental angle gap ≤2.62°). Preoperative C1-2 facet OA was graded on CT using the Shimizu et al classification (Grades 0-3). Radiological and clinical outcomes were compared between the patients with and without SFA.

Posterior graft union at 12 months was observed in 35 of 53 patients (66.0%), and 18 patients (34.0%) showed no definite graft union; however, none demonstrated segmental instability at 12 or 24 months. SFA was identified in 12 (22.6%) patients at 12 months. All cases of SFA occurred in patients with Grade 3 facet OA (p<.001), whereas posterior graft union did not differ significantly by facet OA grade at 12 or 24 months. When posterior graft union and SFA were combined as a composite "fusion" variable, 40 patients (75.5%) had at least one form of osseous fusion at 12 months, increasing to 44 patients (83.0%) at 24 months. Both the SFA and non-SFA groups showed substantial and sustained improvements in Neck Disability Index, Japanese Orthopaedic Association, EuroQol-5D, and pain visual analog scale scores, with only small and inconsistent differences in improvement rate, favoring the SFA group at selected time points.

SFA was identified in approximately one-quarter of patients undergoing posterior C1-2 fusion and was observed only in those with Grade 3 preoperative facet OA. SFA represents an important "hidden fusion" mechanism that, together with posterior graft union, contributes to overall radiographic stability and excellent clinical outcomes. CT assessment of the C1-2 facet joints should be considered an essential component of postoperative fusion evaluation after atlantoaxial instrumentation.