Bone Morphogenetic Protein-2 Compared to Autologous Iliac Crest Bone Graft in the Treatment of Long Bone Nonunion

Transcription

1 Bone Morphogenetic Protein-2 Compared to Autologous Iliac Crest Bone Graft in the Treatment of Long Bone Nonunion MARC A. TRESSLER, DO; JUSTIN E. RICHARDS, MD; D MITRI SOFIANOS, MD; F. KYLE COMRIE, MD; PHILIP J. KREGOR, MD; WILLIAM T. OBREMSKEY, MD, MPH abstract Full article available online at ORTHOSuperSite.com. Search: This retrospective study investigated the effect of recombinant human bone morphogenetic protein-2 (rhbmp-2) mixed with cancellous allograft on fracture healing compared to iliac crest autograft in the treatment of long bone nonunion. Eighty-nine patients with 93 established long bone nonunions treated between January 2002 and June 2004 at a single academic Level I trauma center were evaluated. Patients with clinical and radiographic evidence of failed fracture union underwent nonunion debridement, revision of fixation, and implantation at the nonunion site of either rhbmp-2 or the standard treatment autologous iliac crest bone graft. Union rate, operative time, estimated intraoperative blood loss, hospital length of stay, and postoperative infections were recorded. Nineteen nonunions received rhbmp-2 on a specialized carrier matrix (an absorbable collagen sponge) mixed with cancellous allograft, and 74 nonunions were treated with autologous iliac crest bone graft. There was no statistical difference in the rate of healing between treatment groups (68.4% vs 85.1%, respectively; P.09). Incidence of postoperative infection was 16.2% after autologous iliac crest bone graft and 5.3% after rhbmp-2/absorbable collagen sponge (P.22). Iliac crest autograft was associated with longer operative procedures ( vs minutes; P.0007) and greater intraoperative blood loss ( vs ml; P.01). These outcomes suggest that rhbmp-2 may provide a suitable alternative to autologous iliac bone graft, with the possible advantages of shorter operative time and reduced intraoperative blood loss, and may be considered as part of the orthopedic surgeon s treatment options. Dr Tressler is from Premier Orthopaedics, Hendersonville, and Drs Richards, Kregor, and Obremskey are from Vanderbilt University, Nashville, Tennessee; Dr Sofi anos is from the University of Utah, Salt Lake City, Utah; and Dr Comrie is from the University of Washington, Seattle, Washington. Drs Tressler, Richards, Sofi anos, Comrie, Kregor, and Obremskey have no relevant fi nancial relationships to disclose. No external funding was received for this study. Recombinant human bone morphogenetic protein-2/absorbable collagen sponge is not approved by the US Food and Drug Administration for use in nonunion revision. Correspondence should be addressed to: William T. Obremskey, MD, MPH, Vanderbilt University, 2100 Pierce Ave, 131 MCS, Nashville, TN (william.obremskey@vanderbilt.edu). doi: / DECEMBER 2011 Volume 34 Number 12 e877

2 Fracture complications, particularly long bone fracture nonunions, represent a significant challenge to the orthopedic surgeon and carry an inherent risk of morbidity to the patient. 1-3 Although approximately 90% of osseous injuries heal without impairment, the prevalence of nonunion has been recorded in as many as 10% of fractures and is increasingly associated with open and extensive soft tissue injuries. 4-6 The incidence of long bone fracture nonunion varies, with rates between 1% and 20% reported for humerus, femur, and tibia fractures. 1 However, up to 23.1% of all fracture patients require reoperation for failed or inappropriate fracture union. 4 Despite the success of recent advancements in fracture management, the incidence and associated complications of nonunion procedures are disconcerting and remain concerns for the treating surgeon. Although not all nonunions are the same, the general principal behind an optimal outcome is restoration of the limb to as complete of function as possible. 7 The gold standards of long bone fracture nonunion management are fibrous debridement, stable fracture fixation, augmentation of the nonunion site with autologous iliac crest bone graft when necessary, and adequate soft tissue coverage. 2,5,8 Mechanical stability is provided through internal fixation, whereas biological enhancement of the osseous environment via autologous graft affords in-growth of bone material. The reported morbidity associated with such a procedure is a significant consideration for both the patient and the surgeon. The incidence of major complications associated with bone graft harvesting is reported in approximately 10% of patients, with minor complications occurring in 40%. 5,9,10 Chronic pain at the donor site is documented in nearly 20% of iliac crest harvest procedures. 11 Investigations to determine an equally suitable and less morbid alternative to autologous iliac crest bone graft to stimulate fracture healing have resulted in the development of several bone graft substitutes Certain orthobiologics, in particular recombinant human bone morphogenetic protein-2 (rhbmp-2), are an appealing option to autologous bone graft and provide a nonunion environment with molecular properties that are necessary for appropriate bone union. 4,15 The use of rhbmp-2 as an adjunct in the treatment of fracture nonunions is becoming a significant area of interest to the orthopedic trauma community. 4,16 To our knowledge, little documentation exists in the literature evaluating rhbmp-2 to autologous bone graft for use in nonunion revision. The purpose of this clinical study was to assess the efficacy of rhbmp-2 as compared to harvested autologous iliac crest bone graft in the treatment of established long bone fracture nonunions. Table 1 Inclusion and Exclusion Criteria Inclusion Age 18 years Established nonunion of humerus, femur, or tibia Nonunion treated with rhbmp-2 or autologous iliac crest bone graft Exclusion Active infection at nonunion Inadequate fixation of nonunion Nonunion treated with combination rhbmp-2 and autologous iliac crest bone graft Abbreviation: rhbmp-2, recombinant human bone morphogenetic protein-2. MATERIALS AND METHODS Following Institutional Review Board approval, data from patients older than 18 years who were treated at a universitybased Level I trauma center from January 2002 to June 2004 with the diagnosis of a long bone nonunion were retrospectively reviewed. Long bone fractures included for evaluation were femur, tibia, and humerus fractures, because these represent the most common nonunion following long bone fracture (Table 1). 1 Patient demographic, medical history, initial fracture treatment, subsequent nonunion treatment, and operative and follow-up data were entered into a database. A diagnosis of nonunion was identified in the patient medical record and crossreferenced with radiographic data. Clinical evidence of a nonunion was determined by documented pain and motion at the fracture site. Images were reviewed by the 3 senior authors (M.A.T., P.J.K., W.T.O.), all boardcertified and fellowship-trained orthopedic trauma surgeons, to determine study inclusion. Nonunion was determined by lack of radiographic evidence of bone bridging on 3 of 4 cortices 6 months after injury or a fracture that had not displayed progression of healing over a 3-month period When there was disagreement between clinical and radiographic data, a computed tomography (CT) scan was reviewed to confirm the diagnosis. All nonunions included in the final analysis achieved simultaneous clinical and radiographic criteria of a nonunion. No patients were included in analysis if they were actively being treated for an infection at the fracture site at the time of nonunion revision. SURGICAL TECHNIQUE Patients were selected to receive either iliac crest autograft or an orthobiologic synthetic bone substitute based on the attending surgeon s preference. Surgeons tended to use only 1 method of grafting in their practice, and the basic principles of nonunion treatment were used by all involved surgeons. These include (1) determining whether an infectious process is present with appropriate preoperative laboratories (ie, white blood cell count [WBC], erythrocyte sedimentation rate [ESR], and C-reactive protein [CRP]) and intraoperative cultures; (2) debridement of fibrous tissue that may impede osseous bridging between bone ends; (3) provision of appropriate mechanical stability for osseous bridging and remodeling; (4) addi- e878 ORTHOPEDICS ORTHOSuperSite.com

3 RHBMP-2 IN LONG BONE NONUNION TRESSLER ET AL tion of osteogenic material to the area of the nonunion; and (5) assuring adequate soft tissue coverage. 7,19 Patients were excluded from data analysis if purulence was visible at the nonunion site and microbiologic data demonstrated an infection because this represented an active infectious process. Those patients who had been previously treated for an infected nonunion with operative debridement and completion of antibiotic medication and had demonstrated resolution of infection by laboratory markers (ie, normal CRP, ESR, WBC, and intraoperative gram stain) prior to nonunion treatment with bone graft were included in study analysis. These patients were documented as having a previous nonunion infection. Immediately following the operative procedure to correct the nonunion, all patients included in the study received an appropriate dosage and duration of postoperative antibiotic prophylaxis. Autologous Bone Harvesting Harvesting of autologous iliac crest bone graft was conducted in the standard surgical fashion. 20 The choice of an anterior or posterior approach to the ilium was determined by the location of the fracture nonunion and the subsequent operative plan that most appropriately accessed the nonunion site. Defects in the ilium were filled with Gelfoam (Pfizer, New York, New York) containing bupivicaine, thrombin, and epinephrine. The subcutaneous tissue was approximated in layers and additional bupivicaine was injected subcutaneously. The amount of harvested graft was not uniformly recorded and therefore was not obtainable from either the official operative record or the attending surgeon s operative dictation. Therefore, total amount of graft harvested for each patient in the autograft cohort was not available for study analysis. Preparation of rhbmp-2 and Allograft Patients who were treated with a synthetic orthobiologic graft received a combination of rhbmp-2 on an absorbable collagen sponge (Medtronic Sofamor Danek, Memphis, Tennessee) that was applied with allograft cancellous bone chips. The device was prepared with 12 mg of rhbmp-2 on a cm absorbable collagen sponge. The administration of an inert carrier substance in the form of an absorbable collagen sponge is necessary to (1) maintain a critical threshold concentration of rh- BMP-2 at the nonunion site; (2) provide a matrix for osteoprogenitor cell infiltration; and (3) contain the rhbmp-2 at a localized site and prevent heterotopic bone formation. 21 The rhbmp-2 was reconstituted under appropriate conditions in the operating room with 8 ml of sterile water, resulting in the concentration of 1.50 mg/ml, and was applied to the carrier sponge. Freeze-dried, cancellous allograft bone chips were provided in pre-packaged containers. An amount necessary to completely fill the bone defect was applied to the fracture gap, as described in further detail below. Cancellous allograft provides a scaffold on which bone regeneration consolidates a nonunion defect. 4,16,22 The product of this mixture of allograft and rhbmp-2/absorbable collagen sponge has been demonstrated to be efficacious in the treatment of open tibial fractures and tibial fractures with cortical defects 14 ; however, its application in nonunion procedures is undetermined, and current use for this indication is considered off-label. Nonunion Revision Following a standard operative approach and exposure down to osseous structures, fibrous tissue at the fracture site was debrided with a burr until healthy and viable bone was visualized. Stabilization was provided by either open reduction internal fixation (ORIF) with a plate or with an intramedullary nail. The choice of fixation method was at the discretion of the treating surgeon. In general, metaphyseal nonunions were treated with plate fixation, and diaphyseal nonunions were treated with intramedullary nail stabilization. Either the synthetic orthobiologic device combined with allograft cancellous bone chips or autologous bone graft was applied to the nonunion site. Each graft was used exclusively, and no patient received a combination of treatment methods. For patients who received autologous iliac crest graft, the harvested bone was implanted within the fracture defect to completely fill the void and cover the exposed bone edges both proximally and distally. Patients in the synthetic bone graft group received cancellous bone chips in a similar manner, circumferentially filling the fracture nonunion defect and covering the proximal and distal bone surfaces. The rh- BMP-2/absorbable collagen sponge device was then applied to the allograft implant. When plate fixation was used, the grafting technique was applied prior to stabilization. For patients who received an intramedullary nail, application of the graft was performed after final seating of the nail. Appropriate soft tissue coverage was provided prior to complete wound closure, and any additional procedures to afford a soft tissue muscular flap were recorded from the operative record for final analysis. Operative time and estimated blood loss were recorded from the official intraoperative record, and hospital length of stay was calculated from the medical record. DECEMBER 2011 Volume 34 Number 12 Outcome Assessment Standard postoperative procedures were followed. Immediate postoperative visits occurred at 1 to 2 weeks for assessment of wound healing. Further follow-up evaluations were documented, and data relevant to the physical examination of the affected extremity, including infectious complications, radiographic images, and need for further procedures to address the nonunion site, were noted. Postoperative infection was considered if gross drainage was noted from the nonunion surgical site wound and treated with oral antibiotics or when the wound necessitated operative debridement. Positive microbiologic confirmation by culture from the nonunion site was treated with species-specific intravenous antibiote879

4 Table 2 Nonunion Characteristics by Treatment Group ics administration for 6 weeks in conjunction with infectious disease consultation. All nonunion operations were determined to have adequate stable fixation. Adequacy of internal fixation was judged by the 3 senior authors, who retrospectively reviewed the immediate postoperative radiographs. Union was established by concomitant clinical and radiographic evidence. Anteroposterior (AP) and lateral radiographs were examined by the senior authors, and union was determined by majority agreement of osseous bridging or disappearance of fracture lines on at least 3 of 4 cortical surfaces. These radiographs were cross-referenced with the physical examination record to demonstrate that the patient was weight bearing as tolerated with no pain at the fracture site. Union was established when both clinical and radiographic evidence were in agreement. 23 Fractures that did not demonstrate consolidation on both clinical and radiographic examination at the most recent clinic visit were documented as a failed union. No. (%) Characteristic rhbmp-2/acs (n 19) Autograft (n 74) P Initial fracture fixation.13 External fixation 2 (10.5) 3 (4.1) ORIF 4 (21.1) 29 (39.2) IMN 13 (68.4) 35 (47.3) Nonoperative 0 (0) 7 (9.5) Nonunion type.6 Hypertrophic 1 (5.3) 9 (12.2) Atrophic 13 (68.4) 43 (58.1) Oligotrophic 5 (26.3) 22 (29.7) Nonunion location.15 Diaphyseal 14 (73.7) 41 (55.4) Metaphyseal diaphyseal 5 (26.3) 33 (44.6) Previous infection 5 (26.3) 19 (25.7).9 Abbreviations: ACS, absorbable collagen sponge; IMN, intramedullary nail; ORIF, open reduction internal fi xation; rhbmp-2, recombinant human bone morphogenetic protein-2. Statistical Analysis Descriptive statistics were used to evaluate the study population demographics and the perioperative outcomes of estimated blood loss, operative time, and hospital length of stay relative to the bone graft treatment group. Bivariate statistical analysis was conducted to determine the associations of patient demographic factors, initial fracture treatment method, fracture nonunion characteristics, perioperative infections, and type of bone graft on the outcome of fracture nonunion healing. Chi-square test and Fisher s exact test were used for dichotomous variables, such as sex, diabetes mellitus, nicotine use, osteoporosis, nonunion type, nonunion location, nonunion fixation method, bone defect 2 cm, perioperative infections, and bone graft type. Nonparametric continuous variables were evaluated by the Kruskal-Wallis test. Variables are reported as mean standard deviation (SD) or as raw percentages where applicable. The primary outcome of nonunion healing is presented as an odds ratio (OR) with 95% confidence interval (CI) for ultimate fracture consolidation when evaluating the 2 graft types. Statistically significant variables, in addition to the clinically significant variables of age and female sex, 7,19,24 were then inserted into a multivariable logistic regression model to determine the effect of confounding variables on bone graft type for the primary outcome. Results are reported as OR with 95% CIs. Statistical significance was considered for a P value.05. Analysis was conducted with STATA Statistics/Data Analysis 11.0 (College Station, Texas) software. RESULTS Complete data were available for 89 patients with 93 established long bone nonunions. Mean follow-up time was months. Autologous iliac crest bone graft was harvested from 74 patients and applied to the nonunion site. Recombinant human BMP-2/absorbable collagen sponge with allograft cancellous bone chips was administered to 19 patients. There was no difference among autologous iliac crest bone graft and rhbmp-2/ absorbable collagen sponge treatment groups with respect to age ( vs years, respectively; P.31), female sex (41.9% [31/74] vs 26.32% [5/19], respectively; P.2), nicotine use (35.1% [26/74] vs 36.8% [7/19], respectively; P.9), diabetes mellitus (13.5% [10/74] vs 15.8% [3/19], respectively; P.8), or history of osteoporosis (6.8% [5/74] vs 5.3% [1/19], respectively; P.8). Initial fracture treatment method was similar between patients who received rhbmp-2/absorbable collagen sponge and iliac crest autograft. No difference was identified between groups and the type of nonunion or nonunion location (ie, diaphyseal or metaphyseal diaphyseal). There was no difference in the frequency of previous infection between the treatment groups (Table 2). For the entire population, 76/93 (81.7%) long bone nonunions achieved fracture e880 ORTHOPEDICS ORTHOSuperSite.com

5 RHBMP-2 IN LONG BONE NONUNION TRESSLER ET AL consolidation. Bivariate analysis revealed no difference in the rate of healing between patients treated with autologous iliac crest bone graft (63/74; 85.1%) and those treated with rhbmp-2/absorbable collagen sponge (13/19; 68.4%) (P.09). There was no difference in the demographic variables of age ( vs years, respectively; P.44), female sex (34.2% [26/76] vs 58.8% [10/17], respectively; P.06), nicotine use (34.2% [26/76] vs 41.2% [7/17], respectively; P.59), or diabetes mellitus (14.5% [11/76] vs 11.8% [2/17], respectively; P.77) with respect to nonunion resolution. There was no difference in healing in patients with a history of osteoporosis (6.6% [5/76] vs 5.9% [1/17], respectively; P.92). Further breakdown of the treatment groups revealed no differences in regard to initial fracture fixation, nonunion fixation, nonunion location, nonunion type, previous infection, defects 2 cm, and postoperative infections among healed unions (Table 3). Initial fracture treatment was nonoperative in 7 (7.5%) of 93 patients, external fixation in 5 (5.4%) of 93, ORIF in 33 (35.5%) of 93, and intramedullary nail in 48 (51.6%) of 93. There was no difference in initial fracture treatment and nonunion healing. Nonunion location was diaphyseal in 55 (59.1%) of 93 patients and metaphyseal diaphyseal in 38 (40.9%) of 93. There was no difference in osseous consolidation with regard to diaphyseal and metaphyseal diaphyseal nonunion location and the type of nonunion had no association with ultimate resolution. The average osseous defect size was cm (range, 0-7 cm). There was no difference in healing for bone defects 2 cm. Previous infection at the nonunion site demonstrated no association with fracture consolidation (Table 4). Internal fixation during the nonunion revision procedure consisted of ORIF in 51 (54.8%) of 93 patients and intramedullary nail in 42 (45.2%) of 93 patients. There was a significant difference in whether a nonunion healed when compared by fixation methods (Table 4). There was no difference in the incidence of postoperative infections between autologous iliac crest bone graft (12/74; 16.2%) and rhbmp-2/absorbable collagen sponge (1/19; 5.3%) (P.22). Only 1 of the 12 postoperative wound infections in the iliac crest bone graft group was localized to the graft harvest site. A muscular flap was required for soft tissue coverage in 4 (5.4%) of 76 patients treated with autologous graft and 1 (5.3%) of 19 patients who received rhbmp-2/absorbable collagen sponge (P.98). Iliac crest autograft was associated with longer operative procedures ( vs minutes, respectively; P.0007) and greater intraoperative blood loss ( vs ml, respectively; P.01). Table 3 Characteristics of Healed Nonunions No. (%) Characteristic rhbmp-2/acs (n 13) Autograft (n 63) P Initial fracture fixation.15 External fixation 2 (15.3) 2 (3.2) ORIF 3 (23.1) 24 (38.1) IMN 8 (61.5) 31 (49.2) Nonoperative 0 (0.0) 6 (9.5) Nonunion fixation.93 ORIF 8 (61.5) 38 (60.3) IMN 5 (38.5) 25 (39.7) Nonunion location.69 Diaphyseal 8 (61.5) 35 (55.6) Metaphyseal diaphyseal 5 (38.5) 28 (44.4) Nonunion type.33 Hypertrophic 0 (0.0) 9 (14.3) Atrophic 9 (69.2) 35 (55.6) Oligotrophic 4 (30.8) 19 (30.2) Previous infection 3 (23.1) 14 (22.2).95 Defect 2 cm 2 (15.4) 15 (23.8).50 Postoperative infection 1 (7.7) 9 (14.3).52 Abbreviations: ACS, absorbable collagen sponge; IMN, intramedullary nail; ORIF, open reduction internal fi xation; rhbmp-2, recombinant human bone morphogenetic protein-2. There was no difference in mean hospital length of stay between the autologous bone graft ( days) and rhbmp-2/ absorbable collagen sponge ( days) (P.16) treatment groups. Logistic regression revealed no significant difference in autologous iliac crest bone graft (OR, 2.64; 95% CI, ) or rhbmp-2/absorbable collagen sponge (OR, 0.38; 95% CI, ) on the effect of long bone union. After adjusting for potential confounding variables of age, female sex, and nonunion fracture stabilization with ORIF and plate fixation, multivariable logistic regression demonstrated no significant difference in the effect of bone graft type on the primary outcome (Table 5). Further analysis revealed that after inclusion in the multivariable regression DECEMBER 2011 Volume 34 Number 12 e881

6 Table 4 Proportion of Healed Nonunions Table 5 Multivariable Logistic Regression Characteristic No. (%) P Graft.09 rhbmp-2/acs 13/19 (68.4) Autograft 63/74 (85.1) Initial fracture fixation.99 External fixation 4/5 (80.0) ORIF 27/33 (81.8) IMN 39/48 (81.3) Nonoperative 6/7 (85.7) Nonunion fixation.02 ORIF 46/51 (90.2) IMN 30/42 (71.4) Nonunion location.29 Diaphyseal 43/76 (56.6) Diaphyseal metadiaphyseal 33/76 (43.4) Nonunion type.90 Hypertrophic 9/76 (11.8) Atrophic 44/76 (57.9) Oligotrophic 23/76 (30.3) Previous infection 17/76 (22.4).26 Defect 2 cm 17/76 (22.4).26 Postoperative infection 10/76 (13.2).60 Abbreviations: ACS, absorbable collagen sponge; IMN, intramedullary nail; ORIF, open reduction internal fi xation; rhbmp-2, recombinant human bone morphogenetic protein-2. model, female sex had a significant effect on nonunion resolution (OR, 0.23; 95% CI, ). Plate fixation of the nonunion site did not have a significant effect when included in the multivariable model (OR, 3.12; 95% CI, ). A post-hoc power analysis was performed to determine the power of our study. Regarding the primary objective of determining a significant difference in the rate of nonunion resolution between autologous bone graft and rhbmp/absorbable collagen sponge, our results are reported with a 40% power. A sample size of 124 would have been required to detect a true difference. Characteristic Odds Ratio 95% CI P rhbmp-2/acs Age, y Female sex Nonunion fixation with ORIF Abbreviations: ACS, absorbable collagen sponge; CI, confi dence interval; ORIF, open reduction internal fi xation; rhbmp-2, recombinant human bone morphogenetic protein-2. DISCUSSION Complications of fracture healing, specifically fracture nonunion, are significant concerns for the orthopedic surgeon. 2,6,25 Despite the improvements in surgical technique and orthopedic implants, recent literature has noted that standard treatment of certain long bone fractures results in higher rates of nonunion than what was previously believed. 26 Increased rates of complex, high-energy traumatic fractures have been reported, and subsequently the degree to which these complex injuries are aggressively treated has also increased. 3,19 Furthermore, the resources available to stimulate consolidation of a nonunion are valuable tools in fracture management. Potential causes of nonunion are excess motion, inadequate vascularity, significant fracture gap, and infection. 2,7 Treatment of a fracture nonunion requires attention to the principles of bone regeneration and healing in conjunction with an understanding of the local pathologic process. Multiple factors are involved in the sequence of events that results in normal fracture union. 19,27,28 Maintaining or restoring these factors is essential to successful fracture management. 7 Osteogenesis, osteoinduction, and osteoconduction are the principal elements involved in bone graft design and selection for nonunion management. 29 Osteogenesis describes the ability of living cells to survive transplantation, proliferate, and differentiate into viable osteocytes. 30 Osteoinduction is characterized by the stimulation and activation of host cells from surrounding tissue to differentiate into bone-forming osteoblasts. 31 Osteoconduction is the ability to create and orient the surrounding environment into a stable bone scaffold or matrix capable of supporting the mature osteocytes. 16 Each type of bone graft and bone graft substitute maintains a portion of these essential elements, and the ideal graft incorporates all properties of osteogenesis, osteoinduction, and osteoconduction into a healing osseous environment. 16,22,29 Autologous cancellous bone is the gold standard by which bone graft is measured and is the only graft capable of providing all 3 elements of bone regeneration. 8,22,29 The historical success of autologous grafting is not without complications and morbidity because approximately 25% of patients treated with autologous graft are noted to have prolonged pain at the harvest site, seroma, hematoma, infection, or neurologic symptoms (eg, meralgia paresthetica) The amount of graft is also limited in quantity and quality, especially in certain patient populations (eg, patients of advanced age). 5,22,29 Therefore, e882 ORTHOPEDICS ORTHOSuperSite.com

7 RHBMP-2 IN LONG BONE NONUNION TRESSLER ET AL investigations to provide a less morbid yet equally effective alternative have focused on a group of proteins, normally found in bone, that have been synthetically derived and are easily and practically applied to the nonunion site. 31 Since their discovery by Urist 32 in 1965, bone morphogenetic proteins have been described by the inherent property of osteoinductivity that affects surrounding bone cells and tissues to actively participate in the fracture repair process. 15,16,29,31 The ability to attract and stimulate osteoprogenitor cells to differentiate into chondrocytes and osteoblasts and participate in bone formation is a defining characteristic of this synthetic bone substitute. Although BMP lacks the element of osteoconductivity present in autologous bone graft, certain inert, allographic materials have been used in conjunction with BMP to provide the necessary biologic scaffold on which bone formation is integrated. 14,32 The application of this device has been shown to be a safe and effective alternative to autologous graft in certain clinical settings, and numerous studies have documented the efficacy of this bone substitute for these indications. 4,33-36 Our study is unique in that we compare an alternative to the gold standard treatment of autologous iliac crest bone graft for long bone fracture nonunions. Of further clinical importance is that patients who received rhbmp-2/absorbable collagen sponge had significantly shorter operative times and less intraoperative blood loss. These findings are in agreement with a prior study that noted a reduction in certain perioperative outcomes in patients who receive synthetic bone graft. 37 Previous literature in acute fracture care describes the osteoinductive properties of rhbmp-2 but that the efficacy is not superior to that of autologous bone graft. 14,31 Our data revealed that autologous graft yielded a higher rate of bone union when compared to the synthetic graft; however, this failed to achieve statistical significance. The context in which our evaluation of rhbmp-2 and autologous bone graft was performed should not be misunderstood. The lack of a statistically significance difference between rhbmp-2 and autologous bone graft on the primary outcome must be interpreted with caution. This was a retrospective study with all the inherent weaknesses. Limitations include the use of surgeon s preference in grafting method which could potentially introduce a source of bias. Also, the amount of autologous graft harvested was not uniformly recorded. Whereas the rhbmp dose was identical, the iliac crest graft group most likely did not receive uniform amounts. This could have served as a potential confounder to our results. Furthermore, this is a small sample size of a heterogenous population of long bones with the potential for -error. These fracture nonunions represent challenging complications from relatively common injuries; however, it was impossible to control for subtle differences that are inherent to the treatment of each fracture type. Future studies would require multi-center organization and randomization to appropriately control for preexisting factors and achieve clinically significant power. CONCLUSION This study suggests that rhbmp-2 may provide a suitable alternative to autologous iliac bone graft and may be considered as part of the orthopedic surgeon s treatment options. The ability to treat highly complex fractures will continue to evolve and subsequently provide an opportunity to manage potential complications. The morbidity associated with failure of fracture union extends beyond surgeon and patient and is a serious concern for all parties involved in fracture care. We acknowledge that autologous iliac crest bone graft remains the gold standard against which all bone graft materials are measured. However, evidence from our study suggests rhbmp might offer potential advantages over the gold standard, such as reduced operative time and intraoperative blood loss. Clinical application of rhbmp is a rapidly expanding focus of orthopedic research and requires appropriate scientific investigation. Definitive judgment cannot be extrapolated until more rigorous, large, randomized clinical studies provide critical evaluation and evidence for the efficacy and safety of orthobiologic materials in the management of long bone nonunion. DECEMBER 2011 Volume 34 Number 12 REFERENCES 1. Tzioupis C, Giannoudis PV. Prevalence of long-bone non-unions. Injury. 2007; 38(Suppl 2): S3-S9. 2. Kwong FN, Harris MB. Recent developments in the biology of fracture repair. J Am Acad Orthop Surg. 2008; 16(11): Crowley DJ, Kanakaris NK, Giannoudis PV. Femoral diaphyseal aseptic non-unions: is there an ideal method of treatment? Injury. 2007; 38(Suppl 2):S55-S Schmidmaier G, Schwabe P, Wildemann B, Haas NP. Use of bone morphogenetic proteins for treatment of non-unions and future perspectives. Injury. 2007; 38(Suppl 4):S35-S Sen MK, Miclau T. Autologous iliac crest bone graft: should it still be the gold standard for treating nonunions? Injury. 2007; 38(Suppl 1):S75-S Morshed S, Corrales L, Genant H, Miclau T III. Outcome assessment in clinical trials of fracture-healing. J Bone Joint Surg Am. 2008; 90(Suppl 1): Rodriguez-Merchan EC, Forriol F. Nonunion: general principles and experimental data. Clin Orthop Relat Res. 2004; (419): Jones CB, Mayo KA. Nonunion treatment: iliac crest bone graft techniques. J Orthop Trauma. 2005; 19(10 Suppl):S11-S Arrington ED, Smith WJ, Chambers HG, Bucknell AL, Davino NA. Complications of iliac crest bone graft harvesting. Clin Orthop Relat Res. 1999; (329): Morgan SJ, Jeray KJ, Saliman LH, et al. Continuous infusion of local anesthetic at iliac crest bone-graft sites for postoperative pain relief. A randomized, double-blind study. J Bone Joint Surg Am. 2006; 88(12): Goulet JA, Senunas LE, DeSilva GL, Greenfield ML. Autogenous iliac crest bone graft. Complications and functional assessment. Clin Orthop Relat Res. 1997; (339): Bouxsein ML, Turek TJ, Blake CA, et al. Recombinant human bone morphogenetic protein-2 accelerates healing in a rabbit ulnar osteotomy model. J Bone Joint Surg Am. 2001; 83(8): Einhorn TA, Majeska RJ, Mohaideen A, et al. A single percutaneous injection of recombie883

8 nant human bone morphogenetic protein-2 accelerates fracture repair. J Bone Joint Surg Am. 2003; 85(8): Friedlaender GE, Perry CR, Cole JD, et al. Osteogenic protein-1 (bone morphogenetic protein-7) in the treatment of tibial nonunions. J Bone Joint Surg Am. 2001; 83(Suppl 1 Pt 2): S151-S Tsiridis E, Upadhyay N, Giannoudis P. Molecular aspects of fracture healing: which are the important molecules? Injury. 2007; 38(Suppl 1):S11-S DeLong WG Jr, Einhorn TA, Koval K, et al. Bone grafts and bone graft substitutes in orthopaedic trauma surgery. A critical analysis. J Bone Joint Surg Am. 2007; 89(3): Mechrefe AP, Koh EY, Trafton PG, DiGiovanni CW. Tibial nonunion. Foot Ankle Clin. 2006; 11(1): Hierholzer C, Sama D, Toro JB, Peterson M, Helfet DL. Plate fixation of ununited humeral shaft fractures: effect of type of bone graft on healing. J Bone Joint Surg Am. 2006; 88(7): Lynch JR, Taitsman LA, Barei DP, Nork SE. Femoral nonunion: risk factors and treatment options. J Am Acad Orthop Surg. 2008; 16(2): Stannard JP, Schmidt AH, Kregor PJ. Surgical Treatment of Orthopaedic Trauma New York, NY: Thieme Medical Publishers, Inc. 21. Vaibhav B, Nilesh P, Vikram S, Anshul C. Bone morphogenic protein and its application in trauma cases: current concept update [published online ahead of print February 20, 2007]. Injury. 2007; 38(11): Giannoudis PV, Dinopoulos H, Tsiridis E. Bone substitutes: an update. Injury. 2005; 36(Suppl 3):S20-S McKay WF, Peckham SM, Badura JM. A comprehensive clinical review of recombinant human bone morphogenetic protein-2 (INFUSE Bone Graft) [published online ahead of print July 17, 2007]. Int Orthop. 2007; 31(6): Dijkman BG, Sprague S, Schemitsch EH, Bhandari M. When is a fracture healed? Radiographic and clinical criteria revisited. J Orthop Trauma. 2010; 24(Suppl 1):S76-S Brinker MR, O Connor D. Nonunions: evaluation and treatment. In: Browner BD, Jupiter JB, Levine AM, Trafton PG, Krettek C, eds. Skeletal Trauma. 2008; Philadelphia, PA: WB Saunders: Marsell R, Einhorn TA. Emerging bone healing therapies. J Orthop Trauma. 2010; 24(Suppl 1):S4-S Mahendra A, Maclean AD. Available biological treatments for complex non-unions. Injury. 2007; 38(Suppl 4):S7-S Kakar S, Einhorn TA. Biology and enhancement of skeletal repair. In: Browner BD, Jupiter JB, Levine AM, Trafton PG, Krettek C, eds. Skeletal Trauma. 2008; Philadelphia, PA: WB Saunders: Starr AJ. Fracture repair: successful advances, persistent problems, and the psychological burden of trauma. J Bone Joint Surg Am. 2008; 90(Suppl 1): Müller ME. Treatment of nonunions by compression. Clin Orthop Relat Res. 1965; 43: Pape HC, Evans A, Kobbe P. Autologous bone graft: properties and techniques. J Orthop Trauma. 2010; 24(Suppl 1):S36-S Urist MR. Bone: formation by autoinduction. Science. 1965; 150(698): Boyce AS, Reveal G, Scheid DK, et al. Canine investigation of rhbmp-2, autogenous bone graft, and rhbmp-2 with autogenous bone graft for the healing of a large segmental tibial defect. J Orthop Trauma. 2009; 23(10): Baskin DS, Ryan P, Sonntag V, Westmark R, Widmayer MA. A prospective, randomized, controlled cervical fusion study using recombinant human bone morphogenetic protein-2 with the CORNERSTONE-SR allograft ring and the ATLANTIS anterior cervical plate. Spine (Phila Pa 1976). 2003; 28(12): Boyne PJ, Lilly LC, Marx RE, et al. De novo bone induction by recombinant human bone morphogenetic protein-2 (rhbmp-2) in maxillary sinus floor augmentation. J Oral Maxillofac Surg. 2005; 63(12): Govender S, Csimma C, Genant HK, et al. Recombinant human bone morphogenetic protein-2 for treatment of open tibial fractures: a prospective, controlled, randomized study of four hundred and fifty patients. J Bone Joint Surg Am. 2002; 84(12): Jones AL, Bucholz RW, Bosse MJ, et al. Recombinant human BMP-2 and allograft compared with autogenous bone graft for reconstruction of diaphyseal tibial fractures with cortical defects. A randomized, controlled trial. J Bone Joint Surg Am. 2006; 88(7): e884 ORTHOPEDICS ORTHOSuperSite.com