*Recombinant Human Bone Morphogenetic Protein-2 (rhBMP-2; DiboterminAlfa) in the Management of Open Tibial Fractures: A Prospective, Randomized, Controlled Study in 450 Patients

Session I - Tibia

Thurs., 10/18/01 Tibia, Paper #6, 9:00 AM

*Recombinant Human Bone Morphogenetic Protein-2 (rhBMP-2; Dibotermin Alfa) in the Management of Open Tibial Fractures: A Prospective, Randomized, Controlled Study in 450 Patients

Robert J. Feibel, MD, The Ottawa Hospital-General Campus, Ottawa, ON, Canada; Cristina I. Csimma, RPh, MHP; Alexandre Valentin-Opran, MD, BESTT Study Group, Cambridge, MA (all authors a-Genetics Institute; b-Genetics Institute Wyeth Research)

Purpose: In this prospective clinical study, we evaluated the osteoinductive potential of rhBMP-2 (recombinant human bone morphogenetic protein-2 [dibotermin alfa]) to improve the outcome of patients with open tibial fractures by assessing the need for secondary interventions to achieve fracture healing.

Methods: A total of 450 patients (364 men; 86 women; mean age, 35.7 years; 49% with a recent smoking history) with OTA class 42A to 42C open tibia fractures were enrolled in this one-year, prospective, controlled, randomized, multinational trial. Randomization was stratified by Gustilo-Anderson grade (I, II, IIIA = stratum A; IIIB = stratum B). Patients were randomly assigned to three groups: the control group received standard care (SC; intramedullary nail fracture fixation and routine soft-tissue management). The two other treatment groups received SC plus rhBMP-2 at either 0.75 mg/mL or 1.50 mg/mL implanted on an absorbable collagen sponge (ACS), at the time of definitive wound closure. The primary efficacy endpoint was the proportion of patients requiring secondary interventions for delayed union or nonunion (autogenous bone grafting, IM nail exchange, dynamization and ultrasound) within 12 months after the operation. Additional outcome measures included fracture and soft tissue healing rates, number and invasiveness of procedures to promote fracture union, and safety assessments.

Results: Ninety-four percent of patients completed the 12-month follow-up. The three groups were comparable except for the proportions receiving reamed or unreamed IM nails. Patients treated with rhBMP-2/ACS experienced significantly fewer secondary interventions (dose-dependent overall rate reduction; P = 0.0004). Patients receiving rhBMP-2/ACS (1.50 mg/mL) had a 44% reduced risk of secondary intervention compared to the SC control patients (RR = 0.56; 95% CI = 0.40 to 0.78; pairwise P = 0.0005). The rhBMP-2/ACS 1.50 mg/mL group showed a 59% reduction in the number of invasive interventions (bone grafting and exchange nailing) compared with the SC controls (P = 0.0264). Significantly fewer secondary interventions were required for Gustilo IIIB injuries among patients who received rhBMP-2 1.50 mg/mL (52% reduction; P = 0.0074) or rhBMP-2 0.75 mg/mL (49% reduction; P = 0.0157) compared with SC controls. An independent radiology panel corroborated the primary endpoint findings, as demonstrated by a combined clinical/radiologic endpoint that required both radiographic fracture union and avoidance of a secondary intervention (P = 0.0042; rhBMP-2 1.50 mg/mL vs. SC controls). A significantly greater proportion of patients in the rhBMP-2/ACS 1.50 mg/mL group was healed compared with SC controls at all visits, from 10 to 52 weeks postoperatively. At 6 months, 58% of patients receiving rhBMP-2/ACS (1.50 mg/mL) were healed, compared to 38% of SC controls. Fracture healing time was significantly reduced among rhBMP-2/ACS 1.50 mg/mL patients compared to SC controls. Infection rates were comparable among the groups. Furthermore, a significantly decreased incidence of infections in the limb under study was observed among Gustilo IIIA and IIIB fractures in the rhBMP-2/ACS 1.50 mg/mL group compared with SC (P = 0.0219). The rhBMP-2/ACS 1.50 mg/mL group also experienced significant reductions in hardware failure (P = 0.0174) and pain after the first month of follow-up (P = 0.0343), and a significantly higher rate of wound healing at 6 weeks (83% versus 65%; P = 0.001) compared with SC controls.

Discussion: The effect observed with the addition of rhBMP-2 to standard tibial trauma care shows great potential in the management of high-energy fractures. Despite recent advances in adjuvant fracture management, open tibial fractures remain a challenge. The results of this study demonstrated that implantation of rhBMP-2/ACS (1.50 mg/mL) at the time of definitive wound closure significantly improves the probability and rate of bone and soft tissue healing. The observed reduction in infection rate, acceleration in soft-tissue healing, and reduction in pain may relate to an increased vascular supply in newly induced bone as suggested in preclinical studies.

Conclusions: The most effective dose of rhBMP-2 was 1.50mg/mL in patients with open tibia shaft fractures; it was safe, well tolerated, and demonstrated significant advantages over SC alone by reducing the need for secondary interventions to promote fracture union, reducing the invasiveness of any needed interventions, accelerating fracture and wound healing, and reducing infection rates in severe fractures.