LAUSR

BACKGROUND Posterior atlantoaxial fusion is an important armamentarium for neurosurgeons to treat several pathologies involving the craniovertebral junction. Although the potential advantages of recombinant human bone morphogenetic protein-2 (rhBMP-2) are well documented in the lumbar spine, its indication for C1-C2 fusion has not been well characterized. In our institution, we apply rhBMP-2 to the C1-C2 joint either alone or with hydroxyapatite, locally harvested autograft chips, and/or morselized allogenic bone graft for selected cases-without conventional posterior structural bone graft. We report the clinical outcomes of the surgical technique to elucidate its feasibility. METHODS We performed a single-center, retrospective review of data from 2008 to 2016 and identified 69 patients who had undergone posterior atlantoaxial fusion with rhBMP-2. The clinical records of these patients were reviewed, and the baseline characteristics, operative data, and postoperative complications were collected and statistically analyzed. RESULTS The average age of the 69 patients was 60.8 ± 4.5 years, and 55.1% were women. With an average follow-up period of 21.1 ± 4.2 months, the C1-C2 fusion rate was 94.3% (65 of 69), and the average time to fusion was 11.4 ± 2.6 months (range, 5-23). The overall reoperation rate was 10.1% (7 of 69), with instrumentation failure in 7 patients (10.1%), adjacent segment disease in 2 (2.9%), and postoperative dysphagia and dyspnea in 2 patients (2.9%). No ectopic bone formation or soft tissue edema developed. CONCLUSIONS Although retrospective and from a single center, our study has shown that rhBMP-2 usage at the C1-C2 joint without posterior structural bone grafting is a safe and reasonable surgical option.