Posterolateral lumbar spine fusion with INFUSE bone graft

BACKGROUND CONTEXT: INFUSE has been proven effective in conjunction with threaded cages and bone dowels for single-level anterior lumbar interbody fusion (ALIF). The published experience with posterolateral fusion, although encouraging, utilizes a significantly higher dose and concentration of recombinant human bone morphogenic protein-2 (rhBMP-2) and a different carrier than the commercially available INFUSE. PURPOSE: To present an assessment of fusion rate for posterolateral spine fusion with INFUSE Bone Graft. STUDY DESIGN/SETTING: Retrospective review of patients treated using INFUSE in posterolateral spine fusion in a single institution. PATIENT SAMPLE: 91 patients with minimum 2-year follow-up who underwent posterolateral spine fusion using INFUSE as an iliac crest bone graft (ICBG) substitute. OUTCOME MEASURES: Fusion rate based on fine-cut computed tomographic (CT) scans with sagittal and coronal reconstructions. METHODS: Fusion was performed using one large INFUSE kit (12 mg rhBMP-2, 1.5 mg/mL), which was prepared according to the manufacturer's instructions. The INFUSE sponge was wrapped around the local bone or graft extender and placed over the decorticated surfaces in the lateral gutters. Postoperative CT scans with reconstructions were reviewed by two independent orthopedic spine surgeons. CT scans of a comparison group of 35 patients who underwent primary single-level posterolateral fusion with ICBG were also reviewed. RESULTS: The overall group had a mean 4.38 CT fusion grade and a 6.6% nonunion rate. Primary one-level fusion cases (n=48) had a mean 4.42 fusion grade a 4.2% nonunion rate. Primary multilevel fusions (n=27) had a mean 4.65 CT grade and no nonunions detected. Assessment of the 35 primary one-level ICBG control cases demonstrated a mean CT grade of 4.35 and a nonunion rate of 11.4%. In the 16 cases of revision for prior nonunion, mean CT grade was 3.81 and 4 subjects had nonunions. Additional subgroup analysis showed that smokers (n=14) had a mean 4.32 CT grade with no nonunions. Men had a mean 4.04 CT grade and an 11.1% nonunion rate compared with a mean 4.61 CT grade and 3.6% nonunion rate in women. This difference was statistically significant (p=.036). No significant differences in fusion rate were observed based upon the specific graft extender used (p=.200). CONCLUSIONS: Posterolateral spine fusion involves a more difficult healing environment with a limited surface for healing, a gap between transverse processes and the milieu of distractive forces. Historically, only ICBG has been able to overcome these challenges and reliably generate a successful posterolateral lumbar spine fusion. In contrast to prior studies, clinically available INFUSE delivers only 12 mg rhBMP-2 at a concentration of 1.5 mg/mL. Despite the lower dose and concentration of rhBMP-2, this study suggests that fusion success with INFUSE is equivalent to ICBG for posterolateral spine fusion. As with ICBG, development of solid fusion or nonunion is a multifactorial process. The use of INFUSE is not a substitute for proper surgical technique or optimization of patient-related risk factors. Additional studies are needed to determine the incremental benefit of a greater rhBMP-2 dose or use of alternative carriers for posterolateral fusion. Finally, correlation between radiographic findings and clinical outcomes, and a cost-benefit analysis are needed. Despite these issues, this study presents compelling evidence that commercially available INFUSE is an effective ICBG substitute for one- and two-level posterolateral instrumented spine fusion. (C) 2007 Elsevier Inc. All rights reserved.