• OBJECTIVES
    • To compare radiographic union of tibia fractures with bone defects treated with recombinant bone morphogenetic protein-2 (rhBMP-2) with allograft to autogenous iliac crest bone graft (ICBG).
  • DESIGN
    • FDA-regulated multicenter randomized trial.
  • SETTING
    • Sixteen US trauma centers.
  • PATIENTS/PARTICIPANTS
    • Thirty patients (18-65 years of age) with Type II, IIIA, or IIIB open tibia fracture and bone defect treated with an intramedullary nail.
  • INTERVENTION
    • rhBMP-2 (n = 16) versus ICBG (n = 14).
  • MAIN OUTCOME MEASUREMENTS
    • Radiographic union within 52 weeks. Secondary outcomes included clinical healing, patient-reported function, major complications, and treatment cost. Equivalence was evaluated by testing whether a 90% two-sided confidence interval for the difference in the probability of radiographic union between rhBMP-2 or ICBG is contained with the interval [220% to +20%]. A post hoc Bayesian analysis, using data from a previous trial, was also conducted.
  • RESULTS
    • Twenty-three patients had union data at 52 weeks: 7/12 (58.3%) rhBMP-2 were radiographically united compared with 9/11 (81.8%) ICBG, resulting in a treatment difference of -0.23 (90% CI: -0.55 to 0.10). Patients treated with rhBMP-2 had lower rates of clinical healing at 52 weeks (27% vs. 54%), higher mean Short Musculoskeletal Function Assessment scores (dysfunction: 33.3 vs. 23.7; bother score: 32.8 vs. 21.4) and experienced more complications (5 vs. 3). Mean treatment cost for rhBMP-2 was estimated at $14,155 versus $9086 for ICBG.
  • CONCLUSIONS
    • These data do not provide sufficient evidence to conclude that ICBG and rhBMP-2 are equivalent regarding radiographic union.
  • LEVEL OF EVIDENCE
    • Therapeutic Level II. See Instructions for Authors for a complete description of levels of evidence.