The Use of Allograft and Recombinant Human Bone Morphogenetic Protein for Instrumented Atlantoaxial Fusions

Abstract

Iliac crest autograft is the historic gold standard for bone grafting, but is associated with a significant patient morbidity. Fusion rates of C1-C2 up to 88.9% using allograft and 96.7% using autologous iliac crest bone graft can be achieved when combined with rigid screw fixation. We sought to determine our fusion rate when combining allograft with recombinant human bone morphogenetic protein-2 (rh-BMP2) and rigid screw fixation. We reviewed our experience using allograft, bone morphogenetic protein (rh-BMP2) and screw fixation of C1-C2 in 52 patients and examined indications, surgical technique, fusion rates, and complications. In 28 patients, corticocancellous allograft pieces were laid along decorticated bone after a C2 neurectomy was performed. In 24 patients, unicortical iliac crest allograft was precision-cut to fit between the C1 lamina and C2 spinous processes. Fifty-two C1-C2 fusions were performed with allograft, rh-BMP2, and rigid screw fixation. There were 25 female and 27 male patients ranging in age from 6 to 92 years (mean, 65.8 years). Operative indications included trauma (56%), degenerative disease (21%), rheumatoid arthritis (15%), congenital anomalies (6%), and synovial cyst (2%). The mean follow-up was 23.9 ± 2.1 months (range, 2-55 months). The mean dose of rh-BMP2 used for all patients was 4.5 mg (range, 2.2-12 mg). In patients who achieved sufficient follow-up, 100% achieved solid fusion: 45/50 Lenke A, 5/50 Lenke B. There were no known complications attributable to the use of rh-BMP2. The use of small doses of rh-BMP2 added to allograft in addition to rigid screw fixation is a safe and highly effective means of promoting a solid fusion of the atlantoaxial complex and spares the patient the morbidity of iliac crest harvest.