Janet D. Conway, MD; Lior Shabtai, MD; Alexandra Bauernschub, BS; Stacy C. Specht, MPA
The authors compared the results of treating nonunion with either bone morphogenetic protein (BMP)-7 or BMP-2. Between 2001 and 2009, 214 limb segments were treated for nonunion with either BMP-7 or BMP-2 at the authors’ institution. Sixty-three subjects received 76 units of BMP-7. Radiographic healing occurred in 70% of limb segments at an average of 30 weeks after surgery, and 75% of limb segments were weight bearing at an average of 23 weeks after surgery. In 15.8%, persistent nonunion necessitated additional surgery. Average follow-up was 32 months. One hundred twelve subjects received 138 units of BMP-2. Radiographic healing occurred in 93% of limb segments at an average of 19 weeks after surgery, and 88% were weight bearing at an average of 15 weeks after surgery. In 6.25%, persistent nonunion necessitated additional surgery. Average follow-up was 17 months. Several significant differences were observed. Patients in both BMP-7 and BMP-2 groups were able to fully weight bear at an average of 23 and 15 weeks, respectively (P<.001). Further, both BMP-7 and BMP-2 groups healed at an average of 30 and 19 weeks, respectively (P<.001). Additionally, healing occurred in more limb segments in the BMP-2 group (93%) than in the BMP-7 group (70%) (P<.001). No difference in the complication rate was seen between groups. Patients who received BMP-2 for the treatment of nonunion had a higher rate of radiographic healing, achieved radiographic healing more quickly, and were able to bear weight sooner than those who underwent treatment with BMP-7. [Orthopedics. 2014; 37(12):e1049–e1057.]
The authors are from the Rubin Institute for Advanced Orthopedics, Sinai Hospital of Baltimore, Baltimore, Maryland.
Dr Shabtai, Ms Bauernschub, and Ms Specht have no relevant financial relationships to disclose. Dr Conway is a paid consultant for Biomet and has received research support from Medtronic Sofamor Danek.
This study was supported by funding from Medtronic.
The authors thank John E. Herzenberg, MD, Shawn C. Standard, MD, Dror Paley, MD, and Amanda Chase, MA, for their invaluable assistance with the manuscript.
Correspondence should be addressed to: Lior Shabtai, MD, Rubin Institute for Advanced Orthopedics, Sinai Hospital of Baltimore, 2401 Belvedere Ave, Baltimore, MD 21215 ( firstname.lastname@example.org).
Received: January 06, 2014
Accepted: March 04, 2014
In the United States, approximately 5% to 10% of fractures result in non-union or delayed healing.1 Nonunion leads to a considerable economic burden, results in a large expenditure of health care dollars, and reduces the patient’s ability to be a contributing member of the work force.2 Song et al3 compared the medical costs of tibial fractures in patients who healed normally and those who experienced nonunion and estimated that the annual incremental medical cost associated with fracture nonunion was $20,364 (P<.001).
Nonunion occurs most commonly in the tibia,4,5 but it also occurs in other long bones, such as the fibula, humerus, and femur. Many factors contribute to the development of nonunion, including nutritional or smoking status,6–10 diabetes,11infection,12,13 inadequate fixation, vitamin D or calcium deficiency,14 the degree of initial bone defect and soft tissue destruction,15,16 and the use of steroids17 or nonsteroidal anti-inflammatory drugs (NSAIDs).18,19
Because of the complex nature of nonunion, it is difficult to compare the modalities of effective treatment. The 2 governing principles in the treatment of nonunion are (1) achieving adequate stability with fixation and (2) improving bone biology. To improve bone biology, the surgeon should debride the bone ends until reaching clean, healthy, bleeding bone. The surgeon should also promote proper signaling with the use of bone morphogenetic protein (BMP) and create scaffolding for bone ingrowth with autograft or allograft.20–25 Autograft has been the gold standard for supplementing bone biology. It contains BMPs and the scaffolding needed for bone ingrowth.
Growth factors that are part of the transforming growth factor beta family, BMPs play a critical role during many processes, such as embryogenesis and skeletal formation.26 Kloen et al27 documented the relationship between BMPs and BMP inhibitors in bone callus and compared it with nonunion sites. The expression of BMP inhibitors was comparable between nonunions and fracture callus, but BMP expression was notably lower in the cartilaginous component of nonunions compared with fracture callus.
Several technologic advances in the isolation and production of 2 BMPs (BMP-2 and BMP-7) made it possible to augment the nonunion site and increase the local signals needed to initiate the cascade of bone healing. Both BMP-2 and BMP-7 have been shown to induce ectopic bone formation.28
Several clinical studies showed the efficacy of BMP-7 alone in supplementing bone union compared with only autologous iliac crest bone graft.20–23 Jones et al24 reported the efficacy of using BMP-2 plus allograft compared with autograft alone for open tibial shaft fractures. Swiontkowski et al25 also showed the efficacy of BMP-2 compared with fixation alone in treating open tibial shaft fractures. In the current study, the BMP-2 group had statistically significantly fewer secondary bone grafting procedures as well as a lower rate of infection.25
Giannoudis and Dinopoulos29 recognized the clinical implications and adverse economic effect of nonunions and recommended the proactive use of BMPs in difficult cases (eg, persistent nonunion, bone loss, infection, unavailable or insufficient autograft). Dahabreh et al2 showed a 47% decrease in the cost of nonunion treatment using BMP-7 compared with all previously unsuccessful procedures (P=.001).
However, no previously published clinical reports compared BMP-2 with BMP-7 in the treatment of long bone non-unions. The current study compares the results of the use of BMP-2 and BMP-7 at a single center to treat complex nonunions caused by trauma, limb lengthening, and failed arthrodesis.