Long-Segment Fusion for Adult Spinal Deformity Correction Using Low-Dose Recombinant Human Bone Morphogenetic Protein-2

Abstract

Abstract BACKGROUND: Although use of very high-dose recombinant human bone morphogenetic protein-2 (rhBMP-2) has been reported to markedly improve fusion rates in adult spinal deformity (ASD) surgery, most centers use much lower doses due to cost constraints. How effective these lower doses are for fusion enhancement remains unclear. OBJECTIVE: To assess fusion rates using relatively low-dose rhBMP-2 for ASD surgery. METHODS: This was a retrospective review of consecutive ASD patients that underwent thoracic to sacral fusion. Patients that achieved 2-year follow-up were analyzed. Impact of patient and surgical factors on fusion rate was assessed, and fusion rates were compared with historical cohorts. RESULTS: Of 219 patients, 172 (78.5%) achieved 2-year follow-up and were analyzed. Using an average rhBMP-2 dose of 3.1 mg/level (average total dose = 35.9 mg/case), the 2-year fusion rate was 73.8%. Cancellous allograft, local autograft, and very limited iliac crest bone graft (<20 mL, obtained during iliac bolt placement) were also used. On multivariate analysis, female sex was associated with a higher fusion rate, whereas age, comorbidity score, deformity type, and 3-column osteotomy were not. There were no complications directly attributable to rhBMP-2. CONCLUSION: Fusion rates for ASD using low-dose rhBMP-2 were comparable to those reported for iliac crest bone graft but lower than for high-dose rhBMP-2. Importantly, there were substantial differences between patients in the present series and those in the historical comparison groups that could not be fully adjusted for based on available data. Prospective evaluation of rhBMP-2 dosing for ASD surgery is warranted to define the most appropriate dose that balances benefits, risks, and costs.