Abstract
Background:
Recombinant human bone morphogenetic protein-2 (rhBMP-2) improves healing of open tibial fractures treated with unreamed intramedullary nail fixation. We evaluated the use of rhBMP-2 in the treatment of acute open tibial fractures treated with reamed intramedullary nail fixation.
Methods:
Patients were randomly assigned (1:1) to receive the standard of care consisting of intramedullary nail fixation and routine soft-tissue management (the SOC group) or the standard of care plus an absorbable collagen sponge implant containing 1.5 mg/mL of rhBMP-2 (total, 12.0 mg) (the rhBMP-2/ACS group). Randomization was stratified by fracture severity. The absorbable collagen sponge was placed over the fracture at wound closure. The primary efficacy end point was the proportion of subjects with a healed fracture as demonstrated by radiographic and clinical assessment thirteen and twenty weeks after definitive wound closure.
Results:
Two hundred and seventy-seven patients were randomized and were the subjects of the intent-to-treat analysis. Thirteen percent of the fractures were Gustilo-Anderson Type IIIB. The proportions of patients with fracture-healing were 60% and 48% at week 13 (p = 0.0541) and 68% and 67% at week 20 in the rhBMP-2/ACS and SOC groups, respectively. Twelve percent of the subjects underwent secondary procedures in each group; more invasive procedures (e.g., exchange nailing) accounted for 30% of the procedures in the rhBMP-2/ACS group and 57% in the SOC group (p = 0.1271). Infection was seen in twenty-seven (19%) of the patients in the rhBMP-2/ACS group and fifteen (11%) in the SOC group (p = 0.0645; difference in infection risk = 0.09 [95% confidence interval, 0.0 to 0.17]). The adverse event incidence was otherwise similar between the treatment groups.
Conclusions:
The healing of open tibial fractures treated with reamed intramedullary nail fixation was not significantly accelerated by the addition of an absorbable collagen sponge containing rhBMP-2.
Level of Evidence:
Therapeutic Level I. See Instructions to Authors for a complete description of levels of evidence.
Open tibial fractures are associated with a high prevalence of delayed union and nonunion. Secondary interventions are commonly required to achieve fracture-healing; these include reaming of the medullary canal, which may stimulate fracture-healing1. Clinical studies and meta-analyses comparing reamed nailing with unreamed nailing in the treatment of closed tibial fractures have demonstrated a modest beneficial effect of intramedullary reaming on outcomes2-6; however, the benefit of reaming in the treatment of open tibial fractures remains unclear6-8.
Recombinant human bone morphogenetic protein-2 (rhBMP-2) is an osteoinductive protein. Implanted surgically on an absorbable matrix, rhBMP-2 binds to mesenchymal cell receptors, inducing differentiation into cartilage-forming and bone-forming cells9-11. Differentiated cells form trabecular bone, and vascular invasion occurs. In preclinical and human studies, rhBMP-2 has been shown to induce bone formation12-16, which is integrated with surrounding bone via bone remodeling17,18.
In a trial of 450 patients with an acute open tibial fracture treated with reamed or unreamed intramedullary nail fixation, the addition of rhBMP-2 on an absorbable collagen sponge was associated with a 44% decrease in the risk of secondary intervention, a higher prevalence of fracture-healing at ten weeks to twelve months, fewer failures of the intramedullary nail, fewer infections, and faster wound-healing than intramedullary nail fixation without rhBMP-2 on an absorbable collagen sponge15. However, the subset of patients with reamed intramedullary nail fixation in that study was too small to determine conclusively whether rhBMP-2 on an absorbable collagen sponge improves outcomes in those patients. We conducted this randomized controlled trial of treatment of acute open tibial fractures with reamed intramedullary nail fixation with and without rhBMP-2 on an absorbable collagen sponge to further clarify this point.
This was a multicenter single-blinded randomized study conducted at twenty-eight European and South African sites. The study was designed and performed in compliance with Good Clinical Practice guidelines19 and was approved by an institutional review board. All subjects provided written informed consent prior to study enrollment. The study was registered with ClinicalTrials.gov (NCT00161616).
Patients and Treatment
All patients were adults with an open tibial shaft fracture (Orthopaedic Trauma Association [OTA] Type 42-A, 42-B, or 42-C) requiring statically locked intramedullary nail fixation with reaming. Within fourteen days after the injury, the patients were randomly assigned (1:1) to receive the standard of care consisting of intramedullary nail fixation and routine soft-tissue management (the SOC group) or the standard of care plus treatment, at the time of definitive fracture coverage, with an absorbable collagen sponge implant containing 12.0 mg of rhBMP-2 (the rhBMP-2/ACS group). Randomization was stratified by the Gustilo-Anderson classifications20 of the open wounds at the time of presentation in the emergency room, with stratum A including Types I, II, and IIIA and stratum B consisting of Type IIIB; classification was reassessed intraoperatively at the time of definitive fracture coverage.
The study was performed with use of a randomized block method generated by Wyeth employees not involved in the conduct of the study and accessible by central telephone.
At the time of fracture fixation, rhBMP-2 protein was reconstituted with 8 mL of sterile water, resulting in a 1.5-mg/mL solution. The rhBMP-2 solution was soaked onto a 7.5 × 10-cm absorbable collagen sponge (Type-I collagen derived from bovine tendon) for a minimum of fifteen minutes prior to implantation. A single implant was placed through the soft-tissue wound as an onlay bridging the fracture site.
Secondary interventions to promote fracture-healing performed after sixteen weeks on the basis of the surgeon's judgment were classified as most invasive (e.g., autogenous bone graft), less invasive (e.g., dynamization of the intramedullary nail), or noninvasive (e.g., adjustment of the external fixator).
Assessment
The fractured tibia and the surrounding tissues were assessed at six, ten, thirteen, sixteen, twenty, twenty-four, thirty-two, forty-one, and fifty-two weeks. Evaluations at these time points included recording of concomitant treatments, physical examination, laboratory evaluations, and tests for antibody formation against BMP-2 and Type-I collagen. Delayed union and nonunion were recorded on the basis of the surgeons’ assessment of radiographs and according to prespecified definitions. A delayed union was defined as a tibial fracture that had not healed by sixteen weeks after the injury and had had no radiographic signs of progression toward union for one month or longer, and nonunion was defined as an unhealed fracture site, seen nine months or more after the injury, that had showed no visibly progressive signs of healing for three months or more.
End Points and Statistical Analysis
The primary efficacy end point was the proportion of subjects with a healed fracture at thirteen and twenty weeks.
A fracture was considered healed if, on clinical assessment, there was no fracture site tenderness on manual palpation or no pain at the fracture site with full weight-bearing, radiographic fracture union demonstrated by the presence of bridging callus or the disappearance of the fracture lines on at least one diaphyseal aspect of each orthogonal radiograph, no hardware failure resulting in intramedullary nail dynamization or dislocation, and no secondary procedure recommended or performed to promote fracture-healing or any other procedure that would interfere with the process of fracture-healing. The reason for the use of two of four cortices to define fracture union was to detect the earliest signs of fracture union and therefore optimize the detection of any acceleration of fracture-healing. Treatment failure was defined as a diagnosis of delayed union or nonunion, failure of the intramedullary nail resulting in self-dynamization, or a recommendation for or performance of a secondary intervention procedure that could interfere with fracture-healing. For interim data points, fractures that did not meet criteria for either healing or treatment failure were classified as no outcome.
The prevalence and incidence of anti-BMP-2, anti-bovine Type-I collagen, and anti-human Type-I collagen antibody formation were assessed at six, twenty, and fifty-two weeks with the use of enzyme-linked immunosorbent assay (ELISA; Alta Analytical Laboratory, San Diego, California). Samples with detectable antibodies to BMP-2 were further tested for neutralizing activity (Pfizer, Pearl River, New York).
All randomized subjects were included in the intent-to-treat population. A two-sided Fisher exact test of the hypothesis of equality of treatment efficacy was conducted to assess the primary efficacy end points at the thirteen and twenty-week visits. Statistical tests were conducted at a = 0.025 to adjust for testing at two visits. Breslow-Day tests of homogeneity of treatment effects across study subgroups were conducted to identify modifications of treatment effect by baseline characteristics.
A sample size of 137 subjects offered 90% power to observe a difference of 40% versus 20% in the healing rates in the rhBMP-2/ACS versus the SOC group and 85% power to observe a difference of 50% versus 30% with a Type-I error rate of 0.025 via the Fisher exact test. A Type-I error rate of 0.025 is an acknowledgment of the intention to compare treatments at two time points. The 20% difference in healing rates was considered to be clinically meaningful.
Source of Funding
This research was funded by Wyeth Research, Collegeville, Pennsylvania. Wyeth was acquired in October 2009 by Pfizer, Inc.
Patients
The number of subjects that we planned to include in this study was 300 (150 in each treatment group). However, study enrollment was suspended because of an observed imbalance in infection rates between the treatment groups. Between September 2003 and May 2007, 277 subjects were randomized: 139 to the rhBMP-2/ACS group and 136 to the SOC group. Four subjects randomized to the rhBMP-2/ACS group did not receive the implant, but were included in the intent-to-treat analysis. Twenty-six subjects randomized to the rhBMP-2/ACS group and nineteen randomized to the SOC group did not complete the study. Clinical outcomes were available for eleven of these patients and were missing for seven subjects randomized to the rhBMP-2/ACS group and eight randomized to the SOC group (Fig. 1).
Patient enrollment in the trial.
Demographic characteristics were similar between the treatment groups (Table I). Two fractures in the rhBMP-2/ACS group that were classified as Gustilo-Anderson Type IIIB at the time of screening were reclassified intraoperatively as Type IIIA. The distribution by strata remained similar between the treatment groups.
Treatment
Wound care procedures were similar in type and frequency between the treatment groups (see Appendix). Preliminary wound care procedures were performed in 249 (90%) of the 277 subjects. Definitive fracture coverage was most commonly achieved by delayed primary closure, performed at a median of four days (range, zero to fourteen days) and three days (range, zero to eighteen days) after injury in the rhBMP-2/ACS and SOC groups, respectively.
No differences between the treatment groups were noted with regard to fracture treatment or the type of intramedullary nail that was used (see Appendix). Most patients (111 [80%] of 139 and 107 [78%] of 138 in the rhBMP-2/ACS and SOC groups, respectively) had no preliminary fracture fixation before definitive treatment; the most common procedure among the remaining fifty-nine patients was placement of an external fixator. The mean time (and standard deviation) between injury and definitive fixation was 2.0 ± 2.8 days and 2.1 ± 3.0 days in the rhBMP-2/ACS and SOC groups, respectively, with a range of zero to eighteen days. Most of the patients with an ipsilateral fibular fracture did not require surgical fixation of that fracture. When surgical fixation was performed, it was done concurrently with nailing of the tibial fracture.
Power reaming was used in 127 (91%) of the 139 patients in the rhBMP-2/ACS group and 128 (93%) of the 138 in the SOC group. (Manual reaming was used in the remainder of the patients.) Titanium or stainless-steel intramedullary nails (produced by several leading implant manufacturers) with a median diameter of 10 mm (range, 8 to 13 mm) were used in all but one subject (see Appendix).
The rhBMP-2/ACS implant was placed adjacent to the posterior, medial, or lateral aspect of the tibia just prior to wound closure in 134 (96%) of the 139 patients and covered at least one-third of the tibial circumference in all but one patient. One hundred percent of the sponge was implanted in 125 (90%) the 139 patients; all patients except one had implantation of =50% of the sponge.
Fracture-Healing
The percentage of subjects with a healed fracture at thirteen and twenty weeks (Fig. 2) did not differ significantly between the treatment groups. At week 13, the rate of fracture-healing was higher in the rhBMP-2/ACS group (eighty-three [60%] of 139) than in the SOC group (sixty-six [48%] of 138); however, this difference failed to reach significance (p = 0.0541). At week 20, there was no difference between the groups with regard to the proportion of subjects with a healed fracture (ninety-five [68%] of 139 in the rhBMP-2/ACS group and ninety-two [67%] of 138 in the SOC group). The median time to the recommendation for full weight-bearing on the fractured limb was seventy days (range, thirty-two to 356 days) and was comparable between the treatment groups.
Percentage of patients with fractures classified as healed according to the week after the fracture fixation in the rhBMP-2 and SOC groups. rhBMP-2/ACS = recombinant human bone morphogenetic protein-2 in an absorbable collagen sponge implant, and SOC = standard of care.
There were no significant between-group differences at thirteen or twenty weeks when healing rates were analyzed according to the Gustilo-Anderson-type stratification.
The overall number of secondary interventions (twenty-three and twenty-one in the rhBMP-2/ACS and SOC groups, respectively) and the number and percentage of patients undergoing secondary interventions (sixteen [12%] and seventeen [12%] in the rhBMP-2/ACS and SOC groups, respectively) did not differ significantly between the treatment groups. Fewer secondary procedures were classified as "most invasive" in the rhBMP-2/ACS group than in the SOC group (see Appendix). This difference did not reach significance (p = 0.1271).
Adverse Events
Adverse events were reported by 127 (91%) of the subjects in the rhBMP-2/ACS group and 123 (89%) in the SOC group.
One control subject died from multiple trauma with septic shock, multiple organ failure, and acute respiratory distress syndrome. An HIV-positive patient in the rhBMP-2/ACS group died from tuberculous peritonitis.
There was a higher rate of serious adverse events (e.g., those requiring hospitalization) in the rhBMP-2/ACS group than in the SOC group (four [3%] versus zero), and this was primarily due to the difference in the number of infections.
One patient in each treatment group had a prevalent elevated anti-BMP-2 antibody level (titer, =50). The proportion of patients with an immune response to BMP-2 increased six weeks after definitive treatment to twelve (9%) in the rhBMP-2/ACS group and three (2%) in the SOC group. Two percent of the patients in the rhBMP-2/ACS group and none in the SOC group had elevated titers at the last follow-up visit. No neutralizing antibodies to BMP-2 were found in the patients presenting elevated anti-BMP-2 titers, and no clear association between observed immune responses and reported adverse events or lack of efficacy was observed.
Two subjects in the SOC group had prevalent elevated anti-bovine collagen antibody levels. In the rhBMP-2/ACS group, there was a small increase from a rate of 0% preoperatively to a rate of 5% (seven of 139) at week 6, whereas the rate at week 6 was 2% (three of 138) in the SOC group. At the time of the last follow-up, titers remained elevated in one patient in the rhBMP-2/ACS group and two in the SOC group. No anti-human Type-I collagen antibodies were found in the subjects presenting with elevated levels of anti-bovine Type-I collagen antibodies.
The prevalences of hardware failures, peripheral edema, heterotopic ossification/soft-tissue calcification, and pain affecting the region of the tibial fracture and surrounding soft tissue were similar between the treatment groups (Table II). The incidence of infections affecting the limb under study in the rhBMP-2/ACS group was higher than that in the SOC group, but the difference did not reach significance (p = 0.0645; difference in risk = 0.09 [95% confidence interval, 0.00 to 0.17]).
No confounding factors among patients with infection, such as geographical distribution, demographic characteristics, or treatment (type, number, and/or timing of preliminary or definitive fracture fixation, wound coverage procedures, and reported antibiotic prophylaxis), could explain why infections occurred more often in one group than the other.
Adequate prophylactic antibiotic therapy, defined as the administration of at least one oral or intravenous antibiotic starting within one day after the injury and lasting for three consecutive days or more, was documented for 114 (82%) of the subjects in the rhBMP-2/ACS group and for 117 (85%) in the SOC group. The types of antibiotics used were comparable between the treatment groups, and patients who were identified as having had inadequate antibiotic prophylaxis in both groups did not have higher rates of infection than those who had adequate prophylaxis. There was a slightly longer time from the fracture to definitive wound coverage in the rhBMP-2/ACS group than in the SOC group, but there was no association between delayed coverage and the risk of infection.
The numbers of infections reported in the first two weeks following fracture fixation were comparable between the treatment groups (four in each). Most infections (73% and 60% in the rhBMP-2/ACS and SOC groups, respectively) were reported during the first quarter following fracture fixation. There was no difference between groups in terms of the bacterial species when these were identified. Higher rates of infection were observed in men and in tobacco users, but this association did not differ between the treatment groups. More subjects with an OTA Type-42-A fracture developed an infection in the rhBMP-2/ACS group (eighteen [26%] of seventy) than in the SOC group (three [5%] of sixty-five).
Deep infections involving bone were more common overall in the rhBMP-2/ACS group (twelve [9%]) than in the SOC group (three [2%]), whereas the numbers of superficial infections were similar between the treatment groups (fifteen [11%] and twelve [9%] in the rhBMP-2/ACS and SOC groups, respectively). When analysis of deep infections according to the Gustilo-Anderson classification was performed, deep infections in Type-I and Type-III wounds were found to be more common in the rhBMP-2/ACS group than in the SOC group (two [5%] of forty-one versus zero of forty-six and seven [16%] of forty-four versus one [2%] of forty-five for Types I and III, respectively).
The primary treatment of infection was with antibiotics; fourteen patients in the rhBMP-2/ACS group and four in the SOC group also had surgical interventions, including removal of the intramedullary nail, to treat infection. Nineteen of the twenty-seven infections in the rhBMP-2/ACS group and twelve of the fifteen in the SOC group were deemed to have resolved without sequelae.
Among the subjects with infection, there was no difference in the fracture-healing rates between the treatment groups. The healing rates at twenty weeks were fifteen (56%) of twenty-seven in the rhBMP-2/ACS group and seven (47%) of fifteen in the SOC group.
In this trial, the use of rhBMP-2 in an absorbable collagen sponge did not significantly accelerate the healing of open tibial fractures treated with reamed intramedullary nail fixation. A higher rate of fracture-healing in the rhBMP-2/ACS group as compared with that in the SOC group at thirteen weeks did not reach significance (p = 0.0541). Sensitivity analyses also did not show increased efficacy in subjects at higher risk for treatment failure because of a history of tobacco use or greater baseline wound severity (a Gustilo-Anderson Type-IIIB fracture). Although the use of rhBMP-2 was generally safe, the patients who received such treatment had a trend toward a higher prevalence of infection compared with that observed in the SOC group (p = 0.0645).
These findings clarify the conclusion of a similarly designed study (BMP-2 Evaluation in Surgery for Tibial Trauma; BESTT) of 450 patients with an open tibial fracture, in which rhBMP-2 was associated with improvement in multiple measures of healing compared with the standard of care in a population of patients treated with either reamed or unreamed intramedullary nailing15. In the BESTT study, rhBMP-2/ACS was effective in the subgroup treated with unreamed nailing but did not appear to be effective among patients treated with reamed nailing (n = 146). In the BESTT subgroup treated with reamed nailing, fifty-nine patients received 1.5 mg/mL of rhBMP-2 and thirty-nine patients received only the standard of care. With the exception of the rate at the fourteen-week visit, the rates of fracture-healing in the rhBMP-2/ACS group did not differ significantly from those in the SOC group in the period from ten weeks through six months. The rate of secondary interventions in the reamed-nailing subpopulation was 24% in both the rhBMP-2/ACS group and the SOC group. However, the BESTT study was inadequately powered to demonstrate a between-group difference in the reamed-nailing subpopulation. The present study was designed to remedy this shortcoming.
The lack of efficacy of rhBMP-2 noted in the present study may have been caused in part by the inclusion of patients with fractures that were less severe than those in the BESTT study. The proportion of Gustilo-Anderson Type-IIIA and IIIB fractures, known to be associated with a higher incidence of reoperations21, was lower in our study (32% versus 43% in the BESTT study) whereas the proportion of Type-I fractures was higher (29% versus 22%) (Table III). The percentage of secondary procedures in the current study was 12% in both groups compared with 23% in the rhBMP-2/ACS group and 42% in the SOC group in the BESTT study. The requirement in the current study that repeat interventions not be performed for sixteen weeks or the improvement in standard treatment methods introduced between 1996 (the year in which the BESTT study was initiated) and 2001 (the year in which the current study was initiated) may also explain the decreased rate of secondary procedures. In either case, the relatively small number of secondary procedures overall in the current study may have obscured any between-group difference.
In keeping with the small number of secondary procedures reported in both treatment groups in the current study, healing rates overall were higher than those in the BESTT study. The healing rates in the SOC group in the current study were 30% and 67% at ten and twenty weeks, respectively, compared with 13% and 40% in the subgroup receiving SOC with reamed nailing in the BESTT trial. The difference in the time to healing may also be due to the fact that radiographic bridging of only two cortices of the tibia was required to determine healing in the current study, whereas three were required in the BESTT study. This methodological approach may have led to the unintended result of erasing differences between groups in our study. The time window for the possibility of demonstrating a shift in healing rates between the treatment groups was narrower and the trend toward a difference was observed only at thirteen weeks, at which time 60% of the fractures in the rhBMP-2/ACS group and 48% in the SOC group were healed. It was noted, however, that in most cases the healing of the second and third cortices occurred almost at the same time point.
Our results confirm that rhBMP-2 in an absorbable collagen sponge does not improve the efficacy of reamed nailing in the treatment of open tibial fractures. Regarding the different performances of reamed and unreamed nailing, it is noteworthy that, in a subgroup with an open tibial fracture in the SPRINT (Study to Prospectively Evaluate Reamed Intramedullary Nails in Patients with Tibial Fractures), a negative trend with respect to study outcomes was noted at twelve months postoperatively in subjects treated with reamed nailing: sixty of 206 subjects treated with reamed nailing versus forty-six of 194 treated with unreamed nailing experienced a primary event (bone-grafting, implant exchange, dynamization, infection, or fasciotomy)6. The BESTT study did not suggest such benefits of unreamed nailing. In fact, the benefit of rhBMP-2 in the BESTT study may have been related to the higher rate of secondary interventions in the unreamed group, while the effect remained undetermined in the subset of patients who were treated with reamed nailing in that study.
The reason for, and significance of, the small increase in the rate of deep infection of the limb under study that was observed in our rhBMP-2/ACS group remain unclear and contradict previously published results. The results for the overall population that received reamed or unreamed nailing in the BESTT study indicated a lower infection rate of 21% in the two rhBMP-2/ACS groups (the group that received 0.75 mg/mL of rhBMP-2 and the group that received 1.5 mg/mL) compared with 27% in the SOC group, with no significant between-group differences. The infection rate in the SOC group in the current study (11%) was considerably lower than that reported in the BESTT study (36% in the SOC group treated with reamed nailing), whereas the rates in the rhBMP-2 groups were similar across the two studies. This lower infection rate in the SOC group, rather than an increased infection rate in the rhBMP-2/ACS group, may be the source of the difference observed between groups. The lower infection rate that was observed may reflect interim improvements in surgical techniques, although it is not apparent why these would have lowered the rates only in the SOC group.
The higher rate of infections in the rhBMP-2/ACS group as compared with the SOC group among subjects with a Gustilo-Anderson Type-I fracture contributed to the observed between-group difference and may be an alternative explanation for at least a portion of the between-group difference observed. This higher rate of infections in association with the least severe fractures may be due to the surgical procedures used to implant the absorbable collagen sponge containing the rhBMP-2. The size of the sponge may have compressed soft tissue and compromised blood supply, or the sponge may have been contaminated by contact with surrounding skin during insertion. Although the cause of the observed increase in the infection rate in the rhBMP-2/ACS group compared with that in the SOC group remains unclear, it should be noted that this increase had no observed effect on the overall rate of fracture-healing.
There were few detectable antibody responses to the bovine collagen, and those that did occur had no effect on healing and most resolved by six weeks. No anti-human collagen antibodies developed. These findings are consistent with those of other published studies of bovine collagen and of rhBMP-2 for orthopaedic use15,16,22-24.
This study was limited by its single-blind design. Given the nature of the intervention under study, it was not possible to blind the investigators to the study group. It was considered to be unethical to include an absorbable collagen sponge without rhBMP-2 as a placebo in an open-fracture setting associated with a high risk of infection.
In conclusion, the findings of this study do not confirm that rhBMP-2 in an absorbable collagen sponge improves fracture-healing when used with reamed intramedullary nail fixation for the treatment of open tibial fractures. The issue of increased infection in the patients treated with this implant in our study remains unresolved.
Tables showing the wound procedures before and at definitive wound closure, characteristics of the definitive fracture fixation, and secondary procedures promoting or interfering with fracture-healing are available with the online version of this article on our web site at jbjs.org.
Note: Medical writing support was provided by Naomi Pliskow at On Assignment.
Anglen
JO;
Blue
JM. A comparison of reamed and unreamed nailing of the tibia. J Trauma.
1995;39:351-5.[CrossRef][PubMed]
Court-Brown
CM;
Will
E;
Christie
J;
McQueen
MM. Reamed or unreamed nailing for closed tibial fractures. A prospective study in Tscherne C1 fractures. J Bone Joint Surg Br.
1996;78:580-3.[PubMed]
Forster
MC;
Bruce
AS;
Aster
AS. Should the tibia be reamed when nailing?Injury.
2005;36:439-44.[CrossRef][PubMed]
Bhandari
M;
Guyatt
GH;
Tong
D;
Adili
A;
Shaughnessy
SG. Reamed versus nonreamed intramedullary nailing of lower extremity long bone fractures: a systematic overview and meta-analysis. J Orthop Trauma.
2000;14:2-9.[CrossRef][PubMed]
Gómez-Benito
MJ;
Fornells
P;
García-Aznar
JM;
Seral
B;
Seral-Iñnigo
F;
Doblaré
M. Computational comparison of reamed versus unreamed intramedullary tibial nails. J Orthop Res.
2007;25:191-200.[CrossRef][PubMed]
Study to Prospectively Evaluate Reamed Intramedullary Nails in Patients with Tibial Fractures Investigators, Bhandari
M;
Guyatt
G;
Tornetta
P
3rd;
Schemitsch
EH;
Swiontkowski
M;
Sanders
D;
Walter
SD. Randomized trial of reamed and unreamed intramedullary nailing of tibial shaft fractures. J Bone Joint Surg Am.
2008;90:2567-78.[CrossRef][PubMed]
Bhandari
M;
Guyatt
GH;
Swiontkowski
MF;
Schemitsch
EH. Treatment of open fractures of the shaft of the tibia. J Bone Joint Surg Br.
2001;83:62-8.[CrossRef][PubMed]
Finkemeier
CG;
Schmidt
AH;
Kyle
RF;
Templeman
DC;
Varecka
TF. A prospective, randomized study of intramedullary nails inserted with and without reaming for the treatment of open and closed fractures of the tibial shaft. J Orthop Trauma.
2000;14:187-93.[CrossRef][PubMed]
Shen
B;
Wei
A;
Tao
H;
Diwan
AD;
Ma
DD. BMP-2 enhances TGF-beta3-mediated chondrogenic differentiation of human bone marrow multipotent mesenchymal stromal cells in alginate bead culture. Tissue Eng Part A.
2008;15:1311-20.[CrossRef]
Gori
F;
Thomas
T;
Hicok
KC;
Spelsberg
TC;
Riggs
BL. Differentiation of human marrow stromal precursor cells: bone morphogenetic protein-2 increases OSF2/CBFA1, enhances osteoblast commitment, and inhibits late adipocyte maturation. J Bone Miner Res.
1999;14:1522-35.[CrossRef][PubMed]
Ji
X;
Chen
D;
Xu
C;
Harris
SE;
Mundy
GR;
Yoneda
T. Patterns of gene expression associated with BMP-2-induced osteoblast and adipocyte differentiation of mesenchymal progenitor cell 3T3-F442A. J Bone Miner Metab.
2000;18:132-9.[CrossRef][PubMed]
Boden
SD;
Zdeblick
TA;
Sandhu
HS;
Heim
SE. The use of rhBMP-2 in interbody fusion cages. Definitive evidence of osteoinduction in humans: a preliminary report. Spine (Phila Pa 1976).
2000;25:376-81.[CrossRef][PubMed]
Boyne
PJ;
Marx
RE;
Nevins
M;
Triplett
G;
Lazaro
E;
Lilly
LC;
Alder
M;
Nummikoski
P. A feasibility study evaluating rhBMP-2/absorbable collagen sponge for maxillary sinus floor augmentation. Int J Periodontics Restorative Dent.
1997;17:11-25.[PubMed]
Burkus
JK;
Dorchak
JD;
Sanders
DL. Radiographic assessment of interbody fusion using recombinant human bone morphogenetic protein type 2. Spine (Phila Pa 1976).
2003;28:372-7.[PubMed]
Govender
S;
Csimma
C;
Genant
HK;
Valentin-Opran
A;
Amit
Y;
Arbel
R;
Aro
H;
Atar
D;
Bishay
M;
Börner
MG;
Chiron
P;
Choong
P;
Cinats
J;
Courtenay
B;
Feibel
R;
Geulette
B;
Gravel
C;
Haas
N;
Raschke
M;
Hammacher
E;
van der Velde
D;
Hardy
P;
Holt
M;
Josten
C;
Ketterl
RL;
Lindeque
B;
Lob
G;
Mathevon
H;
McCoy
G;
Marsh
D;
Miller
R;
Munting
E;
Oevre
S;
Nordsletten
L;
Patel
A;
Pohl
A;
Rennie
W;
Reynders
P;
Rommens
PM;
Rondia
J;
Rossouw
WC;
Daneel
PJ;
Ruff
S;
Rüter
A;
Santavirta
S;
Schildhauer
TA;
Gekle
C;
Schnettler
R;
Segal
D;
Seiler
H;
Snowdowne
RB;
Stapert
J;
Taglang
G;
Verdonk
R;
Vogels
L;
Weckbach
A;
Wentzensen
A;
Wisniewski
T; BMP-2 Evaluation in Surgery for Tibial Trauma (BESTT) Study Group. 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;84:2123-34.[CrossRef][PubMed]
Jones
AL;
Bucholz
RW;
Bosse
MJ;
Mirza
SK;
Lyon
TR;
Webb
LX;
Pollak
AN;
Golden
JD;
Valentin-Opran
A; BMP-2 Evaluation in Surgery for Tibial Trauma-Allgraft (BESTT-ALL) Study Group. 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;88:1431-41.[CrossRef][PubMed]
Wozney
JM. Biology and clinical applications of rhBMP-2. : Lynch
SE;
Genco
RJ;
Marx
RE, . Tissue engineering: applications in maxillofacial surgery and periodontics. Chicago: Quintessence; 1999. p 103-23.
Schmidmaier
G;
Wildemann
B;
Cromme
F;
Kandziora
F;
Haas
NP;
Raschke
M. Bone morphogenetic protein-2 coating of titanium implants increases biomechanical strength and accelerates bone remodeling in fracture treatment: a biomechanical and histological study in rats. Bone.
2002;30:816-22.[CrossRef][PubMed]
Food and Drug Administration, Center for Drug Evaluation and Research; Center for Biologics Evaluation and Research. Guidance for Industry.
.
Gustilo
RB;
Anderson
JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am.
1976;58:453-8.[PubMed]
Harris
I;
Lyons
M. Reoperation rate in diaphyseal tibia fractures. ANZ J Surg.
2005;75:1041-4.[CrossRef][PubMed]
Haid
RW
Jr;
Branch
CL
Jr;
Alexander
JT;
Burkus
JK. Posterior lumbar interbody fusion using recombinant human bone morphogenetic protein type 2 with cylindrical interbody cages. Spine J.
2004;4:527-39.[CrossRef][PubMed]
Ellingsworth
LR;
DeLustro
F;
Brennan
JE;
Sawamura
S;
McPherson
J. The human immune response to reconstituted bovine collagen. J Immunol.
1986;136:877-82.[PubMed]
Frank
DH;
Vakassian
L;
Fisher
JC;
Ozkan
N. Human antibody response following multiple injections of bovine collagen. Plast Reconstr Surg.
1991;87:1080-8.[CrossRef] [PubMed]