Outcome of Surgery for Congenital Craniovertebral Junction Anomalies with Atlantoaxial Dislocation/Basilar Invagination: A Retrospective Study of 94 Patients.

Objective: To evaluate the results of surgery for congenital craniovertebral junction (CVJ) anomalies with atlantoaxial dislocation (AAD)/basilar invagination (BI) and compare the results of transoral odontoidectomy and posterior fusion (TOO+PF) with only posterior fusion (PF) in patients with irreducible AAD/BI.

Methods And Results: All 94 patients with congenital CVJ anomalies with AAD/BI operated on during the 3-year study period (June 2013-May 2016) were included. Of these patients, 55 had irreducible AAD/BI and the remaining 39 had reducible AAD/BI. TOO+PF was restricted to patients (34/94; 36.2%) with irreducible AAD/BI when reduction and realignment by intraoperative C1-C2 facet joint manipulation were considered technically difficult and risky. The remaining patients with irreducible AAD/BI and all the patients with reducible AAD/BI (60/94; 63.8%) were managed with only posterior fusion. Poor preoperative Nurick grade, preoperative dyspnea/lower cranial nerve deficits, and syringomyelia were associated with significantly higher incidence of postoperative pulmonary complications. Follow-up ≥3 months (final follow-up) was available for 87 patients. Good outcome (Nurick grade 0-3) at final follow-up was noted in 90% (45/50) of the patients with irreducible AAD/BI and 91.9% (34/37) of the patients with reducible AAD/BI. Preoperative poor Nurick grade (4-5) was the only factor associated with poor outcome. No significant difference in perioperative complications, outcome, and fusion was noted between patients who underwent TOO+PF or only PF for irreducible AAD/BI.

Conclusions: Many of the patients with congenital AAD/BI showed remarkable recovery after surgery. Preoperative poor Nurick grade (4-5) is associated with poor outcome. TOO+PF is a safe alternative treatment option for irreducible AAD/BI when only PF techniques are technically difficult/risky.