100 patients scheduled for posterolateral fusion due to severe spinal stenosis, grade 1 spondylolisthesis, or spondylolysis were randomized to graft material of either Escherichia coli-derived recombinant human bone morphogenetic protein-2 (rhBMP-2) in a hydroxyapatite carrier or an autologous iliac crest bone graft. Patients were assessed primarily for fusion success on CT at 12 and 24 weeks. Secondary outcomes included clinical scores and rates of adverse events. Results demonstrated no significant difference in successful fusion rate between groups at 12 or 24 weeks, as well as no significant differences between groups in improvement in clinical scores or in the rate of adverse events.