JBJS | Article

The Journal of Bone & Joint Surgery, Volume 81, Issue 12

Articles | December 01, 1999

Bone Morphogenetic Protein But Not Transforming Growth Factor-β Enchances Bone Formation in Canine Diaphyseal Nonunions Implanted with a Biodegradable Composite Polymer*†

JAMES D. HECKMAN, M.D.‡, SAN ANTONIO; WILLIAM EHLER, D.V.M.§, LACKLAND AIR FORCE BASE; BRYAN P. BROOKS, B.A.#; THOMAS B. AUFDEMORTE, D.D.S.#; CHRISTOPH H. LOHMANN, M.D.#; THANE MORGAN, M.D.#; BARBARA D. BOYAN, PH.D.#, SAN ANTONIO, TEXAS

View Disclosures and Other Information

Investigation performed at the Departments of Orthopaedics and Pathology, The University of Texas Health Science Center at San Antonio, San Antonio; the Audie L. Murphy Memorial Veterans Administration Hospital, San Antonio; and Wilford Hall Medical Center, Lackland Air Force Base

The Journal of Bone & Joint Surgery. 1999; 81:1717-29

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

Background: The purpose of the present study was to create an effective bone-graft substitute for the treatment of a diaphyseal nonunion.Methods: A standardized nonunion was established in the midportion of the radial diaphysis in thirty mongrel dogs by creating a three-millimeter segmental bone defect (at least 2 percent of the total length of the bone). The nonunion was treated with implantation of a carrier comprised of poly(DL-lactic acid) and polyglycolic acid copolymer (50:50 polylactic acid-polyglycolic acid [PLG50]) containing canine purified bone morphogenetic protein (BMP) or recombinant human transforming growth factor-ß (TGF-ß1), or both, or the carrier without BMP or TGF-ß1. Five groups, consisting of six dogs each, were treated with implantation of the carrier alone, implantation of the carrier with fifteen milligrams of BMP, implantation of the carrier with 1.5 milligrams of BMP, implantation of the carrier with fifteen milligrams of BMP and ten nanograms of TGF-ß1, or implantation of the carrier with ten nanograms of TGF-ß1. At twelve weeks after implantation, the radii were examined radiographically and the sites of nonunion were examined histomorphometrically.Results: We found that implantation of the polylactic acid-polyglycolic acid carrier alone or in combination with ten nanograms of TGF-ß1 failed to induce significant radiographic or histomorphometric evidence of healing at the site of the nonunion. The radii treated with the carrier enriched with either 1.5 or fifteen milligrams of BMP showed significantly increased periosteal and endosteal bone formation on histomorphometric (p < 0.05) and radiographic (p < 0.02) analysis.Conclusions: Bone formation in a persistent osseous defect that is similar to an ununited diaphyseal fracture is increased when species-specific BMP incorporated into a polylactic acid-polyglycolic acid carrier is implanted at the site of the nonunion. TGF-ß1 at a dose of ten nanograms per implant did not induce a similar degree of bone formation or potentiate the effect of BMP in this model.Clinical Relevance: The biodegradable implant containing BMP that was used in the present study to treat diaphyseal nonunion is an effective bone-graft substitute.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

Numerous recent studies have indicated that acute segmental bone defects can be treated effectively with a variety of materials, including demineralized bone^25,26,29,33, a combination of collagen and calcium phosphate^{17,35,37,53,57}, synthetic biodegradable materials such as polylactic acid²⁷, and copolymers of polylactic acid and polyglycolic acid^10,34. Supplementation of these bone-graft materials with osteogenic materials such as autogenous marrow^13,35,42 facilitates the healing of large acute osseous defects, presumably because the implant is enriched with pluripotent osteogenic cells and growth factors, such as the bone morphogenetic proteins (BMPs).

One advantage of the use of synthetic bone-graft materials is that the need to obtain autogenous bone graft or marrow is eliminated. Growth factors that promote bone formation are attractive for use with osteoconductive, biodegradable scaffolds because they have the potential to enhance the proliferation and differentiation of osteogenic cells that are present at the site of the defect. BMP has been tested in animal models^30,40,52,55 and has demonstrated a positive effect on bone formation^4,20,31. Bovine, canine, and human BMPs have been shown, in rats and mice, to induce heterotopic formation of bone in the subcutaneous fascia of the thorax and in the muscles of the thigh^{14,24,27,44-47,54}. Osteoinductive protein isolated from bovine bone has been used effectively to promote fusion in a rabbit model of intertransverse process arthrodesis of the lumbar spine^5,6. Recently, Zegzula et al.⁵⁹ demonstrated that a critical-size defect in a rabbit radius can be induced to heal with a poly(DL-lactide) implant containing recombinant human bone morphogenetic protein-2 (rhBMP-2).

BMPs are also being examined for their ability to induce the healing of nonunions. Recombinant human BMP-2 is being tested in humans to assess its ability to promote the healing of ununited fractures³¹. However, there has been only limited study of the response of nonunions to BMP in animal models, as most research has focused on the treatment of acute segmental defects^12,51,60. In acute defects, the species specificity of BMP does not appear to be critical⁵⁹. However, the importance of species specificity was demonstrated in a canine nonunion model, in which the nonunion failed to heal when exposed to bovine BMP but showed significant bone formation when exposed to canine BMP (p < 0.03)²⁷. In the canine nonunion study, BMP that had been isolated from canine bone, rather than recombinant canine or bovine BMP, was used. When the response of the cells at the site of the nonunion to the canine BMP was examined in vitro, the cells differentiated into chondrocytes¹⁰ and, when implanted in nude mice, they formed cartilage nodules⁹.

Transforming growth factor-ß (TGF-ß) has been studied extensively with regard to its influence on orthotopic bone formation^32,41 and the response of osteogenic cells and chondrocytes to this growth factor in vitro^7,48. TGF-ß enhances bone formation when implanted orthotopically, but its effects are not osteoinductive and it does not induce bone formation when implanted heterotopically⁴³. TGF-ß1 causes mesenchymal cells to differentiate into chondrocytes⁴⁹, and it stimulates committed chondrocytes to enter the endochondral developmental pathway¹⁰, suggesting that TGF-ß1 could potentiate the effects of BMP in the treatment of nonunions.

The BMPs are members of the TGF-ß family of proteins. As a group, the proteins regulate recruitment, proliferation, and differentiation of a variety of cells. It is not yet clear how TGF-ß and the BMPs interact to promote endochondral bone formation, but recent studies have indicated that they work synergistically to modulate the differentiation of cartilage and bone cells¹⁰. Because TGF-ß1 is one of the growth factors that is initially released by platelets at a wound site and is involved in recruitment of mesenchymal cells, we hypothesized that it might enhance the effect of BMP in the repair of nonunions.

We recently developed a biodegradable polylactic acid-polyglycolic acid scaffold¹⁹ to facilitate migration of mesenchymal cells into a site of nonunion and osteogenesis at that site while at the same time providing controlled release of a growth factor during healing. Polylactic acid and polyglycolic acid, alone and in combination, have been studied extensively; they are biodegradable^8,21,22; and they elicit only a minimum foreign-body or allergic reaction^22,27. Studies in our laboratory indicated that protein is released from these scaffolds with an initial burst in the first forty-eight hours, followed by continuous release over the twelve-week test period¹.

The experiments in the present report were done to continue previous work performed with a standard, consistent, and reproducible model of chronic nonunion in the radius of dogs^10,27. The biodegradable delivery system made with a polylactic acid-polyglycolic acid copolymer was combined with various doses of purified canine BMP or recombinant human TGF-ß1, or both, and was implanted into the site of the nonunion in an attempt to stimulate healing. The purposes of the present study were to determine whether such a biodegradable composite would facilitate the healing of an established diaphyseal nonunion, whether BMP or TGF-ß1 could enhance this effect, and whether TGF-ß1 would potentiate the effect of BMP.

*Although none of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article, benefits have been or will be received but are directed solely to a research fund, foundation, educational institution, or other nonprofit organization with which one or more of the authors is associated. Funds were received in total or partial support of the research or clinical study presented in this article. The funding sources were the United States Department of Veterans Affairs Grant 88-038, United States Public Health Service Grant DE08603, and the Center for the Enhancement of the Biology/Biomaterials Interface. Support was also provided by the United States Department of Defense through the clinical investigation facilities at Wilford Hall Medical Center at Lackland Air Force Base, Texas.

†The views expressed in this article are those of the authors and do not reflect the official policy of the Department of Defense or other Departments of the United States Government.

‡Audie L. Murphy Memorial Veterans Administration Hospital, 7400 Merton Minter Boulevard, San Antonio, Texas 78284.

§Wilford Hall Medical Center, Lackland Air Force Base, Texas 78326-5300.

#Departments of Orthopaedics (J. D. H., B. P. B., C. H. L., T. M., and B. D. B.) and Pathology (T. B. A.), The University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, Texas 78284-7774. E-mail address for Dr. Heckman: heckman@uthscsa.edu.

†The views expressed in this article are those of the authors and do not reflect the official policy of the Department of Defense or other Departments of the United States Government.

‡Audie L. Murphy Memorial Veterans Administration Hospital, 7400 Merton Minter Boulevard, San Antonio, Texas 78284.

§Wilford Hall Medical Center, Lackland Air Force Base, Texas 78326-5300.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1 Photograph of a three-millimeter-thick microporous-macroporous polylactic acid-polyglycolic acid carrier.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 2-A through 2-D: Radiographs of the radius of a fourteen-kilogram dog that was treated with implantation of a polylactic acid-polyglycolic acid carrier without growth factors (group A). Fig. 2-A: Immediately after the ostectomy.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-B Three months after the ostectomy, moderately hypertrophic nonunion was evident.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-C Immediately after implantation of the carrier.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-D Three months after implantation of the carrier, no bone-healing was evident and the plate was broken, indicative of persistent motion at the site of the nonunion.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 3-A through 3-D: Radiographs of the radius of a thirteen-kilogram dog that was treated with implantation of a polylactic acid-polyglycolic acid carrier with fifteen milligrams of BMP (group D). Fig. 3-A: Immediately after the ostectomy.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-B Three months after the ostectomy, hypertrophic nonunion was evident.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-C Immediately after implantation of the carrier.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-D Twelve weeks after implantation of the carrier, there was clear evidence of complete endosteal bridging and some evidence of periosteal bridging.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 4-A through 4-E: Photomicrographs of paraffin-embedded sections of specimens from the five groups (hematoxylin and eosin; original magnification, x 16). Fig. 4-A: Specimens treated with PLG50 alone (group A), showing no appreciable healing of the site of the ostectomy, which was filled largely with fibrovascular collagenous connective tissue and cartilage.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-B Specimen treated with fifteen milligrams of BMP (group B), showing substantial healing, particularly of the upper 50 percent of the specimen. Bone has completely bridged the ostectomy defect, connecting its cut margins. Only a small central zone of hyaline cartilage, which appears to be converting into new bone, and some collagenous connective tissue at the bottom persist.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-C Specimen treated with 1.5 milligrams of BMP (group C), showing a moderate amount of residual matrix persisting in the upper 50 percent of the nonunion site. This material, which is amorphous in appearance, is actually finely degraded particles that are not visible at this magnification. The healing in this specimen is essentially equivalent to that in the specimen shown in Fig. 4-B. Bridging trabecular bone is evident in the middle zone of the defect site.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-D Specimen treated with ten nanograms of TGF-ß1 combined with fifteen milligrams of BMP (group D), showing appreciable healing. Little or no fibrocollagenous connective tissue or residual carrier is present.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-E Specimen treated with ten nanograms of TGF-ß1 alone (group E), showing fibrous connective tissue with inflammation and focal residual carrier and no bridging of the defect site.

View Larger

TABLE I RADIOGRAPHIC ANALYSIS OF PERIOSTEAL AND ENDOSTEAL HEALING OF CHRONIC NONUNIONS OF CANINE RADII TWELVE WEEKS AFTER RESECTION AND IMPLANTATION OF BIODEGRADABLE SCAFFOLDS WITH OR WITHOUT GROWTH FACTOR

*The values are expressed as the least square mean and the standard error of the least square mean. The grading system is described in the text. †Compared with group A, the difference was significant (p < 0.05). ‡Compared with groups B, C, and D, the difference was significant (p < 0.05).

Group	Treatment	Periosteal Healing* (grade)	Endosteal Healing* (grade)
A (n = 6)	PLG50	1.63 ± 0.26	1.63 ± 0.24
B (n = 6)	15.0 mg BMP	2.48 ± 0.26†	2.08 ± 0.24
C (n = 6)	1.5 mg BMP	2.56 ± 0.26†	2.37 ± 0.24†
D (n = 6)	15.0 mg BMP and 10.0 ng TGF-ß1	2.48 ± 0.26†	2.25 ± 0.24†
E (n = 5)	10.0 ng TGF-ß1	1.35 ± 0.29‡	1.30 ± 0.27‡

Add to My JBJS

Group	Treatment	Periosteal Healing* (grade)	Endosteal Healing* (grade)
A (n = 6)	PLG50	1.63 ± 0.26	1.63 ± 0.24
B (n = 6)	15.0 mg BMP	2.48 ± 0.26†	2.08 ± 0.24
C (n = 6)	1.5 mg BMP	2.56 ± 0.26†	2.37 ± 0.24†
D (n = 6)	15.0 mg BMP and 10.0 ng TGF-ß1	2.48 ± 0.26†	2.25 ± 0.24†
E (n = 5)	10.0 ng TGF-ß1	1.35 ± 0.29‡	1.30 ± 0.27‡

View Larger

TABLE II HISTOMORPHOMETRIC ANALYSIS OF THE SITES OF CHRONIC NONUNIONS OF CANINE RADII TWELVE WEEKS AFTER RESECTION AND IMPLANTATION OF BIODEGRADABLE SCAFFOLDS WITH OR WITHOUT GROWTH FACTOR

*The values are expressed as the least square mean and the standard error of the least square mean. †Compared with group A, the difference was significant (p < 0.05). ‡Compared with group B, the difference was significant (p < 0.05).

Group	Treatment	Percent Total Bone Volume*	Percent Lamellar Bone*	Percent Active Osteoblasts*	Percent Residual Matrix*
A (n = 6)	PLG50	11.28 ± 7.27	10.80 ± 6.70	45.16 ± 9.87	19.54 ± 9.60
B (n = 6)	15.0 mg BMP	22.17 ± 7.27	18.66 ± 6.70	37.14 ± 9.87	33.63 ± 9.60
C (n = 6)	1.5 mg BMP	33.62 ± 7.27†	30.29 ± 6.70†	33.23 ± 9.87	5.42 ± 9.60‡
D (n = 6)	15.0 mg BMP and 10.0 ng TGF-ß1	36.58 ± 7.27†	32.63 ± 6.70†	40.43 ± 9.87	27.92 ± 9.60
E (n = 5)	10.0 ng TGF-ß1	18.80 ± 7.26	15.55 ± 7.34	18.92 ± 10.81	25.30 ± 10.51

Add to My JBJS

TABLE II HISTOMORPHOMETRIC ANALYSIS OF THE SITES OF CHRONIC NONUNIONS OF CANINE RADII TWELVE WEEKS AFTER RESECTION AND IMPLANTATION OF BIODEGRADABLE SCAFFOLDS WITH OR WITHOUT GROWTH FACTOR

Group	Treatment	Percent Total Bone Volume*	Percent Lamellar Bone*	Percent Active Osteoblasts*	Percent Residual Matrix*
A (n = 6)	PLG50	11.28 ± 7.27	10.80 ± 6.70	45.16 ± 9.87	19.54 ± 9.60
B (n = 6)	15.0 mg BMP	22.17 ± 7.27	18.66 ± 6.70	37.14 ± 9.87	33.63 ± 9.60
C (n = 6)	1.5 mg BMP	33.62 ± 7.27†	30.29 ± 6.70†	33.23 ± 9.87	5.42 ± 9.60‡
D (n = 6)	15.0 mg BMP and 10.0 ng TGF-ß1	36.58 ± 7.27†	32.63 ± 6.70†	40.43 ± 9.87	27.92 ± 9.60
E (n = 5)	10.0 ng TGF-ß1	18.80 ± 7.26	15.55 ± 7.34	18.92 ± 10.81	25.30 ± 10.51

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

Animals

Thirty mature, healthy mongrel dogs with a mean weight of 12.5 kilograms (range, ten to fifteen kilograms) were obtained from Brink Kennels (Little Rock, Arkansas). All of the animals were parasite-free and had normal hematological profiles. They all were skeletally mature, as confirmed by the presence of closed growth plates in the distal part of the radius and ulna on preoperative radiographs. All animals were acclimated to the environment for two weeks before the first operation. The study protocol was approved by the Institutional Animal Care and Use Committees at The University of Texas Health Science Center at San Antonio, San Antonio; the Audie L. Murphy Memorial Veterans Administration Hospital, San Antonio; and Wilford Hall Medical Center, Lackland Air Force Base, Texas. The experiments were conducted according to the principles set forth in Guide for the Care and Use of Laboratory Animals (National Institutes of Health Publication Number 86-23)¹⁸ and the Animal Welfare Act of 1966, as amended.

A nonunion was created in the midportion of the radial diaphysis in all of the animals with use of a technique adapted from that described by Müller et al.^27,39. With the animal under general anesthesia, the diaphysis of the radius was exposed through a longitudinal extensile incision. The periosteum was elevated circumferentially, and the midportion of the radius was identified. At this point, two parallel cuts were made with use of a power-driven oscillating saw. The cuts were perpendicular to the shaft of the bone and were approximately three millimeters apart. The full-thickness wafer of bone from the midportion of the radial diaphysis, which represented at least 2 percent of the total length of the bone, was then removed, creating a consistent three-millimeter defect that was held distracted by the intact ulna. The operative wound was then closed in layers.

The size of the defect created during the original operation is critical. In our first series of larger dogs²⁷, the defect had healed in most of the animals at three months. When smaller adult dogs (ten to fifteen kilograms) were used, an operative defect that measured approximately three millimeters (at least 2 percent of the length of the radius) was found to result in a persistent gap in more than 90 percent of the animals at three months after the osteotomy.

A second cause of early failure of the model, besides healing of the gap, was fracture of the ulna²⁷. Particular care must be taken at the time of the osteotomy to avoid scoring the radial cortex of the adjacent ulna, as this creates a stress-riser, predisposing the ulna to fracture. Early in our studies, the animals were allowed to roam free immediately after the osteotomy, and they usually were able to bear full weight within forty-eight hours after the operation. One dog in a previous study group sustained an ulnar fracture and was excluded from that study. Because of this fracture, we became concerned that unprotected weight-bearing produced excessive stress on the ulna. Therefore, all thirty animals in the present study were managed with a protective plaster splint for two weeks after the initial osteotomy. Postoperative radiographs were made every two weeks for twelve weeks. The radiographs made at twelve weeks consistently demonstrated a hypertrophic mid-diaphyseal nonunion of the radius.

Between twelve and sixteen weeks after creation of the original nonunion, the animals were again anesthetized, the nonunion site was exposed, and the tissue in the defect was sharply excised. An implant (described later) was placed in the defect, completely filling it, and then a five-hole, one-third tubular stainless-steel plate was applied to the radius with four screws to stabilize, but not to compress, the bone and to act as a cap to hold the implant in place²⁷. The wound was then closed in layers, and the animals were returned to their cages. The involved limb was placed in a splint for three to five days. Radiographs of randomly selected dogs were made every two weeks for a period of twelve weeks to monitor healing. At the end of twelve weeks, the animals were killed humanely and the entire radius was removed for radiographic and histological evaluation.

Implants

Polylactic acid-polyglycolic acid implants were fabricated in our laboratory from a copolymer of 50:50 poly(DL-lactide-coglycolide) (PLG50) (Medisorb; E. I. DuPont, Wilmington, Delaware)^1,2,19. A 15 percent (weight per volume) solution of PLG50 in reagent-grade acetone was prepared at 52 degrees Celsius and then cooled to room temperature. Each implant was produced in an individual mold. The desired dose of BMP, TGF-ß1, or BMP and TGF-ß1 was suspended in fifty microliters of physiological saline solution. This suspension was mixed with the dissolved polymer by aspiration of the mixture back and forth in the pipette used to deliver the protein suspension. A glass pipette was then used to deliver the mixture into the mold from the bottom up to prevent air bubbles from forming. All procedures were performed at room temperature. The implants were cured under vacuum during a period of two weeks, resulting in a combined microporous-macroporous structure (Fig. 1). The implants were fabricated under sterile conditions. Randomly selected implants were cultured in brain-heart infusion agar to verify sterility.

The size of the micropores was determined by OsteoBiologics (San Antonio, Texas). The implants contained random micropores of less than 150 nanometers in diameter and macropores with a mean diameter of 500 micrometers along the entire length of the implant. The implants had a length of three millimeters and a diameter of one centimeter.

Growth Factors

BMP was isolated from canine bone powder as described previously^9,56 and was stored as a lyophilized powder at -70 degrees Celsius. To verify the osteoinductive activity of each batch of canine BMP, five milligrams of the lyophilized powder was implanted in a gelatin capsule in the calf muscles of nude mice. All batches that demonstrated osteoinduction both radiographically and histologically were pooled together and used in the study. Recombinant human TGF-ß1 was purchased from R and D Systems (Minneapolis, Minnesota). The TGF-ß1 was shown to be active in in vitro assays. Cells isolated from explanted tissue from the nonunion site also responded to TGF-ß1 in vitro¹¹. Both growth factors were used as powders, as already described.

Study Groups

Five groups consisting of six dogs each were used. All animals had an established nonunion that was treated with excision of the fibrocartilage in the defect and implantation of a PLG50 molded carrier with or without a growth factor. The excised cartilage was used as a source of cells from the nonunion site in a related study¹¹. In group A, the defect was filled with PLG50 without growth factor (the control group); in group B, with PLG50 with fifteen milligrams of BMP; in group C, with PLG50 with 1.5 milligrams of BMP; in group D, with PLG50 with fifteen milligrams of BMP and ten nanograms of TGF-ß1; and in group E, with PLG50 with ten nanograms of TGF-ß1. One dog in group E sustained an acute fracture of the ipsilateral ulna while roaming free in the eighth week of the study. The dog was killed at that time and is not included in the data analysis; thus, group E included only five dogs. We did not include controls that were managed with only a plate, as we had previously demonstrated that osseous union was not achieved under those conditions²⁷ and we believed that it would be inhumane to repeat this type of treatment. Moreover, our goal was to compare the carrier with and without growth factors, rather than to make a comparison with a nontreated control. We did not include a control group without plate fixation, as a plate is necessary to maintain both the stability of the bone ends and the dimensions of the critical-size defect after the tissue has been excised from the nonunion site.

Radiographic Analysis

After the animals were killed, the entire radius was excised and standardized anteroposterior and lateral radiographs of the specimen were made. A blinded evaluation of these radiographs was performed by two independent examiners who graded the degree of healing of the nonunion. The degrees of endosteal and periosteal healing were graded separately with use of a semiquantitative scale. Grade 1 indicated no callus; grade 2, callus but no bridging of the defect; grade 3, callus with incomplete bridging of the defect; grade 4, mature callus with complete bridging across the entire defect; and grade 5, mature callus with trabecular bone crossing the defect. The score for periosteal healing was determined by averaging the values obtained for each cortical surface as seen on the anteroposterior radiograph (top and bottom) and the lateral radiograph (top and bottom) (four sites). The score for endosteal healing was determined by averaging the amounts of healing in the central region of the bone as seen on the anteroposterior and lateral radiographs (two sites).

Histological Analysis

The tissue in the previous nonunion site as well as the bone at either end of the site was processed for histological analysis. After fixation and decalcification, this tissue was processed with use of graded concentrations of alcohol and then it was embedded in paraffin, sectioned, and stained with hematoxylin and eosin. Gross macroscopic sequential sections (five sections per specimen) were made with use of an EXAKT diamond saw (Exakt Medical Instruments, Oklahoma City, Oklahoma) to identify the most representative mid-longitudinal area of the specimen that best characterized the histological response. To do this, multiple serial sections of each specimen were prepared from medial to lateral at one-millimeter intervals, and those sections most closely approximating the midcentral aspect were used. The most representative section was chosen for histomorphometric analysis after examination by two pathologists who were blinded to the identity of the specimen.

Histomorphometric analyses were performed with use of an image-analysis system (Bioquant IV; R and M Biometrics, Nashville, Tennessee). The parameters that were measured included the area of the defect, the percentage of new bone that had formed in the defect, and the percentages of trabecular bone volume, lamellar bone, osteoblasts, and residual PLG50 per defect site. When visualized under polarized light, lamellar bone was seen to have organized birefringent collagen fibers. In contrast, woven bone had haphazard collagen fibers. Active osteoblasts were defined as cells containing plump Golgi apparatuses and osteoid.

Statistical Analysis

The data were analyzed with one-way analysis of variance with contrasts for pairwise comparison of the group means³⁵. A residual analysis of skewness, kurtosis, and movability with the Shapiro and Wilk test⁵⁰ verified that the assumptions for a valid analysis had been met. For the semiquantitative assessments of healing based on radiographic findings, the data are presented as the combined scores of the two examiners (the least square means and the standard error of the mean). Variance components analysis was used to determine intraclass correlation coefficients for the radiographic assessments. This analysis indicated an intraclass correlation coefficient of 0.60 for endosteal healing and 0.62 for periosteal healing for the two examiners. For all other parameters, the data are presented as the least square means and the standard error of the mean for each treatment group.

Results

Introduction | Materials and Methods | Results | Discussion | References

Radiographic Analysis

Semiquantitative analysis of the radiographs made at twelve weeks indicated that both periosteal and endosteal bone formation were enhanced when BMP was included in the implants (p < 0.02) (Figs. 2-A, 2-B, 2-C, 2-D, 3-A, 3-B, 3-C, 3-D and Table I). The effect on periosteal bone formation was independent of the concentration of BMP. However, there appeared to be a dose-dependent effect of BMP on endosteal bone formation, as only the lower concentration resulted in significant (p < 0.05) differences in endosteal bone formation compared with that after implantation of the carrier alone. No significant differences were found between the effects of BMP at concentrations of 1.5 and fifteen milligrams on either periosteal or endosteal bone formation. Moreover, TGF-ß1 did not potentiate the effect of BMP and healing in the presence of TGF-ß1 alone was reduced in comparison with that following all other treatments.

Histological and Histomorphometric Analysis

In the group treated with PLG50 implants alone (group A) (Fig. 4-A), material forming in the site of the ostectomy consisted largely of a fibrovascular collagenous connective tissue with residual islands of fibrocartilage as well as hyaline cartilage. The ends of the ostectomy site abutted the implant, blending with bone and fibrous tissue proliferating across perforated segments of the implant material. Although the amount of new bone was relatively small, the bone that had formed had surfaces lined with proliferating osteoblasts and osteoblasts actively synthesizing osteoid. Residual PLG50 matrix was present in small amounts.

When fifteen milligrams of BMP had been included in the implants (group B) (Fig. 4-B), bone formation in the defect was marked and osteoblastic activity was more brisk than was noted in the absence of this growth factor, resulting in active synthesis of osteoid. Residual PLG50 matrix, which was mostly degraded into particles, was present in the cytoplasm of occasional multinucleated foreign-body giant cells. The bone at the ends of the ostectomy site bluntly approximated the implant with zones of ingrowth from peripheral to central. Osteoblasts in large aggregates were seen adjacent to the implant. These osteoblasts had prominent Golgi apparatuses indicative of a high rate of bone protein synthesis.

When implants containing 1.5 milligrams of BMP had been used (group C) (Fig. 4-C), the defect sites also had a high bone volume with high levels of osteoblastic activity. A variable (small-to-moderate) amount of residual PLG50, mostly degraded into fine particles, was observed in the cytoplasm of macrophages and multinucleated foreign-body giant cells.

When fifteen milligrams of BMP combined with ten nanograms of TGF-ß1 had been implanted (group D) (Fig. 4-D), the effect was comparable with that seen in the implants containing 1.5 milligrams of BMP. In some specimens, bone formation was nearly complete across the defect, with osseous trabeculae bridging and connecting most of the ostectomy site. The amount of residual PLG50 was comparable with that observed in the control specimens.

In the defect sites that had been treated with ten nanograms of TGF-ß1 alone (group E) (Fig. 4-E), the bone volume was reduced compared with that in the groups treated with BMP. The surfaces of the bone were lined with active osteoblasts, although these were reduced in number. These implants also had a larger volume of residual PLG50 than the BMP groups, and this was associated with multinucleated foreign-body giant cells and a nonspecific fibroproliferative reaction.

Histomorphometric measurements confirmed these observations (Table II). Compared with the defects that had been treated with PLG50 alone (group A), those treated with fifteen milligrams of BMP (group B) had approximately a twofold increase in the total bone volume and those treated with 1.5 milligrams of BMP (group C) had a threefold increase (p < 0.05). The addition of ten nanograms of TGF-ß1 to fifteen milligrams of BMP resulted in a total bone volume similar to that seen in sites treated with 1.5 milligrams of BMP. However, TGF-ß1 alone had no effect on total bone volume. The amount of lamellar bone was greatest in sites treated with 1.5 milligrams of BMP and those treated with fifteen milligrams of BMP and ten nanograms of TGF-ß1. In general, the osteoblasts on the bone surfaces were active. However, in sites treated with ten nanograms of TGF-ß1 alone, there was an apparent but not significant reduction in the number of active osteoblasts even when compared with sites treated only with PLG50. Some residual PLG50 was present in all specimens, although the amount (as determined on histomorphometric analysis) and the morphology (histological analysis) varied with the treatment. The least amount of residual PLG50 matrix was found in sites treated with 1.5 milligrams of BMP.

Discussion

Introduction | Materials and Methods | Results | Discussion | References

Our results confirmed our previous observations that bone morphogenetic protein incorporated into a biodegradable carrier can be used to treat chronic nonunions. This property of BMP may be independent of the type of carrier used. In our previous study, we used polylactic acid scaffolds that were highly porous fiber meshes²⁷. The findings of the present study showed that polylactic acid-polyglycolic acid implants containing purified BMP can also be used effectively as a bioactive material for treating hypertrophic diaphyseal nonunions.

Comparable implants containing ten nanograms of TGF-ß1 alone were not effective, indicating that components of the purified bone morphogenetic protein are responsible for the new-bone formation. Moreover, TGF-ß1 did not seem to enhance the effects of BMP on the amount or quality of new-bone formation. The failure of TGF-ß1 to elicit a response may have been due to the concentration. Studies in which TGF-ß had been used in carriers made of other materials²⁸ indicated that substantially higher doses of the growth factor (120 nanograms per implant) were needed to elicit large amounts of bone formation. Whether the results in the present study differed as a consequence of the concentration of TGF-ß1 or the type of carrier is not known.

The results suggest that the maturation of bone within the defect may depend on the dose of BMP, as the greatest amount of lamellar bone was found in sites treated with 1.5 milligrams of BMP. This finding may indicate that components in the BMP preparation inhibit bone-remodeling at higher concentrations.

Interestingly, the addition of TGF-ß1 to fifteen milligrams of BMP resulted in both total bone volume and lamellar bone formation that were similar to those seen at the lower dose of BMP. TGF-ß1 may exert this effect through a number of mechanisms. This growth factor promotes proliferation of mesenchymal cells¹⁶, potentially increasing the pool of osteochondral progenitor cells. TGF-ß1 also enhances the early stages of differentiation of both osteogenic cells¹⁵ and cartilage cells⁴⁸. Thus, it may increase new-bone formation through increased osteogenesis by committed osteoblasts or increased endochondral ossification by committed chondrocytes, or both. Finally, TGF-ß1 may work in concert with other regulatory factors to enhance the differentiation of chondrocytes in the endochondral pathway⁴⁸. More rapid osteogenesis would result in earlier bone-remodeling and, ultimately, in more lamellar bone. Because we did not use a single recombinant BMP isoform for these studies, it is not possible to know if TGF-ß1 worked synergistically with BMP itself or with another component in the purified preparation.

The polylactic acid-polyglycolic acid implants in the present study were tolerated well. There were no signs of infection and only a minimum foreign-body giant-cell reaction secondary to the implant. These findings are commensurate with the observations of others^34,58, indicating that polylactic acid-polyglycolic acid composites are biocompatible. Although reports in the literature have suggested that polylactic acid-polyglycolic acid materials can cause inflammation and toxicity because of the release of acid during dissolution³, our results failed to demonstrate an untoward tissue response. Macrophages and multinucleated giant cells were present in low numbers, a response that is typically associated with degrading suture material manufactured from polylactic acid-polyglycolic acid copolymers. It is possible that in larger segmental defects, the persistence of the PLG50 might be a problem; however, the porosity of the design, which included both micropores and macropores, should provide sufficient fluid flow to ensure adequate buffering of any local build-up of acidic byproducts resulting from degradation of the implant.

Histological analysis demonstrated substantial but incomplete dissolution of the carrier during the three months in vivo. The amount of residual polymer was smaller than that noted in our previous study²⁷, in which implants fabricated from polylactic acid only were used. The more rapid dissolution of the PLG50 copolymer implants may have been due to the physical structure of the material, but it is more likely that incorporation of polyglycolic acid, which degrades at a faster rate than polylactic acid, was the factor responsible³⁸. Our data also suggest that stimulation of new-bone formation driven by the BMP seems to accelerate the resorption of the carrier, as reflected by the trend toward an inverse relationship between the percentage of total bone volume and the percentage of residual matrix in the defects.

The results indicate that both BMP and TGF-ß1 are released from the implants and that they are active. In a previous study¹, we showed that a model protein, soybean trypsin inhibitor, was released from implants similar to those used in the present study in a burst during the first forty-eight hours followed by a slow, continuous release as the implant degraded. Canine BMP exhibited similar release kinetics from the implants. This finding supports the hypothesis that one or more of the growth factors that are present in the implants could affect recruitment and proliferation of mesenchymal cells during the initial forty-eight hours and then affect differentiation of committed cells at later times. The macroporous design of the implants facilitated guided tissue regeneration⁶⁰ because, histologically, ingrowth of bone and connective tissue was seen to have occurred from the ends of the ostectomy site into the implant along the perforations. It is certainly possible, however, that the design of the implant impeded periosteal bone formation somewhat, as there were no macropores on the outer surface of the cylinder. This finding may account, in part, for our failure to detect a difference in periosteal healing as a function of the dose of BMP.

We concluded that it is possible to create a biodegradable carrier consisting of polylactic acid-polyglycolic acid that degrades substantially but incompletely during a three-month period and that can release bioactive factors. The implants seem to be tolerated well, with limited foreign-body reaction as seen histologically. We believe that the carrier acts as an effective system for the delivery of BMP, as the addition of purified BMP to the carrier induced a significant increase in the degree of new-bone formation (p < 0.05). The addition of TGF-ß1 alone to the carrier did not seem to enhance this repair process. This lack of response may have been due to the low dose of TGF-ß1, release of the TGF-ß1 at a time when the cellular environment was not prepared to respond, inactivation of the TGF-ß1 by local environmental factors such as altered pH, or rapid systemic dissipation of the substance through the circulatory system. Our results with use of a canine model, which mimics the clinical situation of chronic nonunion in humans, suggest that BMP may be used to treat chronic nonunions just as others have shown it to be effective in the treatment of acute segmental defects^{23,25-27,30,40,47,52}.

NOTE: The authors thank Dr. Thomas J. Prihoda, Department of Pathology, The University of Texas Health Science Center at San Antonio, San Antonio, Texas, for his careful analysis of the data. They also thank OsteoBiologics, San Antonio, for the assessment of the pore size of the implants.

References

Introduction | Materials and Methods | Results | Discussion | References

Agrawal, C. M.; Best, J.; Heckman, J. D.; and Boyan, B. D.: Protein release kinetics of a biodegradable implant for fracture non-unions. Biomaterials,16: 1255-1260, 1995.161255 1995 [PubMed]

Agrawal, C. M.; Boyan, B. D.; and Heckman, J. D.: Biodegradable implant for fracture nonunions. United States patent number 5,492,697, issued in 1995.

Athanasiou, K. A.; Niederauer, G. G.; and Agrawal, C. M.: Sterilization, toxicity, biocompatibility and clinical applications of polylactic acid/polyglycolic acid copolymers. Biomaterials,17: 93-102, 1996.1793 1996 [PubMed]

Beck, L. S.; Amento, E. P.; Xu, Y.; Degsuzman, L.; Lee, W. P.; Nguyen, T.; and Gillett, N. A.: TGF-ß1 induces bone closure of skull defects: temporal dynamics of bone formation in defects exposed to rhTGF-ß1. J. Bone and Min. Res.,8: 753-761, 1993.8753 1993

Boden, S. D.; Schimandle, J. H.; and Hutton, W. C.: Lumbar intertransverse-process spinal arthrodesis with use of a bovine bone-derived osteoinductive protein. A preliminary report. J. Bone and Joint Surg.,77-A: 1404-1417, Sept. 1995.77-A1404 1995

Boden, S. D.; Schimandle, J. H.; Hutton, W. C.; Damien, C. J.; Benedict, J. J.; Baranowski, C.; and Collier, S.: In vivo evaluation of a resorbable osteoinductive composite as a graft substitute for lumbar spinal fusion. J. Spinal Dis.,10: 1-11, 1997.101 1997

Bonewald, L. F., and Munday, G. R.: Role of transforming growth factor-beta in bone remodeling. Clin. Orthop.,250: 261-276, 1990.250261 1990 [PubMed]

Bos, R. R. M.; Rozema, F. R.; Boering, G.; Nijenhuis, A. J.; and Pennings, A. J.: Bio-absorbable plates and screws for internal fixation of mandibular fractures. A study in six dogs. Internat. J. Oral and Maxillofac. Surg.,18: 365-369, 1989.18365 1989

Boyan, B. D.; Schwartz, Z.; Swain, L. D.; Khare, A. G.; Heckman, J. D.; Ramirez, V.; Peters, P.; and Carnes, D. L., Jr.: Initial effects of partially purified bone morphogenetic protein on the expression of glycosaminoglycan, collagen, and alkaline phosphatase in nonunion cell cultures. Clin. Orthop.,278: 286-304, 1992.278286 1992 [PubMed]

Boyan, B. D.; Sylvia, V. L.; Piorkowski, G.; Kieswetter, K.; Liu, Y.; Wozney, J.; Harris, S. E.; Dean, D. D.; and Schwartz, Z.: BMP-2 and TGF-ß induce chondrocytes in the endochondral pathway to become 1,25-(OH)2D3-responsive [abstract]. . J. Bone and Min. Res.,12 (Supplement 1): 308, 1997.12 (Supplement 1)308 1997

Boyan, B. D.; Caplan, A. I.; Heckman, J. D.; Lennon, D. P.; Ehler, W.; and Schwartz, Z.: Osteochondral progenitor cells in acute and chronic canine nonunions. J. Orthop. Res.,17: 246-255, 1999.17246 1999 [PubMed]

Boyne, P. J.: Animal studies of application of rhBMP-2 in maxillofacial reconstruction. Bone,19 (Supplement 1): 83S-92S, 1996.19 (Supplement 1)83 1996

Burwell, R. G.: The function of bone marrow in the incorporation of a bone graft. Clin. Orthop.,200: 125-141, 1985.200125 1985 [PubMed]

Canalis, E.; Centrella, M.; and Urist, M. R.: Effect of partially purified bone morphogenetic protein on DNA synthesis and cell replication in calvarial and fibroblast cultures. Clin. Orthop.,198: 289-296, 1985.198289 1985 [PubMed]

Caplan, A. I., and Boyan, B. D.: Endochondral bone formation: the lineage cascade. In Bone 8: Mechanisms of Bone Development and Growth, pp. 1-46. Edited by B. K. Hall. Boca Raton, CRC Press, 1994.

Centrella, M.; McCarthy, T. L.; and Canalis, E.: Skeletal tissue and transforming growth factor beta. FASEB J.,2: 3066-3073, 1988.23066 1988 [PubMed]

Chapman, M. W.; Bucholz, R.; and Cornell, C.: Treatment of acute fractures with a collagen-calcium phosphate graft material. A randomized clinical trial. J. Bone and Joint Surg.,79-A: 495-502, April 1997.79-A495 1997

Committee on Care and Use of Laboratory Animals of the Institute for Laboratory Animal Resources, Commission on Life Sciences, National Research Council: Guide for the Care and Use of Laboratory Animals. Publication 86-23, revised 1985. Washington, D.C., United States Department of Health and Human Resources, Public Health Service, National Institutes of Health, 1986.

Coombs, A. G. A.; Heckman, J. D.; and Boyan, B. D.: Unpublished data, 1995.

Cornell, C. N., and Lane, J. M.: Newest factors in fracture healing. Clin. Orthop.,277: 297-311, 1992.277297 1992 [PubMed]

Craig, P. H.; Williams, J. A.; Davis, K. W.; Magoun, A. D.; Levy, A. J.; Bogdansky, S.; and Jones, J. P., Jr.: A biologic comparison of polyglactin 910 and polyglycolic acid synthetic absorbable sutures. Surg., Gynec. and Obstet.,141: 1-10, 1975.1411 1975

Cutright, D. E., and Hunsuck, E. E.: Tissue reaction to the biodegradable polylactic acid suture. Oral Surg., Oral Med., and Oral Pathol.,31: 134-139, 1971.31134 1971

dos Santos Neto, F. L., and Volpon, J. B.: Experimental nonunion in dogs. Clin. Orthop.,187: 260-271, 1984.187260 1984 [PubMed]

Drivdahl, R. H.; Howard, G. A.; and Baylink, D. J.: Extracts of bone contain a potent regulator of bone formation. Biochim. Biophys. Acta,714: 26-33, 1982.71426 1982 [PubMed]

Glowacki, J.; Kaban, L. B.; Murray, J. E.; Folkman, J.; and Mulliken, J. B.: Application of the biological principle of induced osteogenesis for craniofacial defects. Lancet,1: 959-962, 1981.1959 1981 [PubMed]

Glowacki, J.; Kaban, L. B.; Murray, J. E.; Folkman, J.; and Mulliken, J. B.: Bone implants and induced osteogenesis [letter]. Lancet,1: 452, 1982.1452 1982

Heckman, J. D.; Boyan, B. D.; Aufdemorte, T. B.; and Abbott, J. T.: The use of bone morphogenetic protein in the treatment of non-union in a canine model. J. Bone and Joint Surg.,73-A: 750-764, June 1991.73-A750 1991

Heckman, J. D.; Aufdemorte, T. B.; Athanasiou, K. A.; Schwartz, Z.; Korvick, D. L.; Brooks, B. P.; Nguyen, T.; Wong, R.; Orgpipattanakul, B.; and Boyan, B. D.: Treatment of acute osteotomy defects in the dog radius with rhTGF-ß1. Trans. Orthop. Res. Soc.,20: 590, 1995.20590 1995

Hollinger, J.; Mark, D. E.; Bach, D. E.; Reddi, A. H.; and Seyfer, A. E.: Calvarial bone regeneration using osteogenin. J. Oral and Maxillofac. Surg.,47: 1182-1187, 1989.471182 1989

Johnson, E. E.; Urist, M. R.; and Finerman, G. A. M.: Bone morphogenetic protein augmentation grafting of resistant femoral nonunions. A preliminary report. Clin. Orthop.,230: 257-265, 1988.230257 1988 [PubMed]

Joyce, M. E.; Jingushi, S.; and Bolander, M. E.: Transforming growth factor-beta in the regulation of fracture repair. Orthop. Clin. North America,21: 199-209, 1990.21199 1990

Joyce, M. E.; Roberts, A. B.; Sporn, M. B.; and Bolander, M. E.: Transforming growth factor-beta and the initiation of chondrogenesis and osteogenesis in the rat femur. J. Cell Biol.,110: 2195-2207, 1990.1102195 1990 [PubMed]

Kaban, L. B., and Glowacki, J.: Augmentation of rat mandibular ridge with demineralized bone implants. J. Dental Res.,63: 998-1002, 1984.63998 1984

Kieswetter, K.; Schwartz, Z.; Niederauer, M. O.; Leatherbury, N. C.; Korvick, D.; Xiong, L.; Wirth, M.; Diaz, J. L.; Niederauer, G. G.; and Boyan, B. D.: Segmental defect repair with biodegradable tissue scaffolds. Trans. Orthop. Res. Soc.,22: 900, 1997.22900 1997

Lane, J. M., and Sandhu, H. S.: Current approaches to experimental bone grafting. Orthop. Clin. North America,18: 213-225, 1987.18213 1987

Maxwell, S. E., and Delaney, H. D.: Designing Experiments and Analyzing Data: A Model Comparison Perspective. Pacific Grove, California, Brooks/Cole Publishing, 1990.

Miller, E. J., and Martin, G. R.: The collagen of bone. Clin. Orthop.,59: 195-232, 1968.59195 1968 [PubMed]

Miller, R. A.; Brady, J. M.; and Cutright, D. E.: Degradation rates of oral resorbable implants (polylactates and polyglycolates): rate modification with changes in PLA/PGA copolymer ratios. J. Biomed. Mater. Res.,11: 711-719, 1977.11711 1977 [PubMed]

Müller, J.; Schenk, R.; and Willenegger, H.: Experimentelle Untersuchungen uber die Entstehung reaktiver Pseudarthrosen am Hunderadius. Helvetica Chir. Acta,35: 301-308, 1968.35301 1968

Nilsson, O. S.; Urist, M. R.; Dawson, E. G.; Schmalzried, T. P.; and Finerman, G. A. M.: Bone repair induced by bone morphogenetic protein in ulnar defects in dogs. J. Bone and Joint Surg.,68-B(4): 635-642, 1986.68-B(4)635 1986

Noda, M., and Camilliere, J. J.: In vivo stimulation of bone formation by transforming growth factor-beta. Endocrinology,124: 2991-2994, 1989.1242991 1989 [PubMed]

Rosen, D. M.; Stempien, S. A.; Thompson, A. Y.; and Seyedin, S. M.: Transforming growth factor-beta modulates the expression of osteoblast and chondroblast phenotypes in vitro. J. Cell. Physiol.,134: 337-346, 1988.134337 1988 [PubMed]

Rosen, D. M.; Nathan, R.; Armstrong, R.; Bentz, H.; Thompson, A.; DeLeon, E.; Buckman, E.; Fiedler, L.; and Seyedin, S.: Bone induction and transforming growth factor-beta. Ann. New York Acad. Sci.,593: 98-106, 1990.59398 1990

Sampath, T. K.; DeSimone, D. P.; and Reddi, A. H.: Extracellular bone matrix-derived growth factor. Exper. Cell Res.,142: 460-464, 1982.142460 1982

Sampath, T. K.; Nathanson, M. A.; and Reddi, A. H.: In vitro transformation of mesenchymal cells derived from embryonic muscle into cartilage in response to extracellular matrix components of bone. Proc. Nat. Acad. Sci.,81: 3419-3423, 1984.813419 1984 [PubMed]

Sampath, T. K.; Muthukumaran, N.; and Reddi, A. H.: Isolation of osteogenin, an extracellular matrix-associated, bone-inductive protein, by heparin affinity chromatography. Proc. Nat. Acad. Sci.,84: 7109-7113, 1987.847109 1987 [PubMed]

Sato, K., and Urist, M. R.: Induced regeneration of calvaria by bone morphogenetic protein (BMP) in dogs. Clin. Orthop.,197: 301-311, 1985.197301 1985 [PubMed]

Schwartz, Z.; Bonewald, L. F.; Caulfield, K.; Brooks, B.; and Boyan, B. D.: Direct effects of transforming growth factor-beta on chondrocytes are modulated by vitamin D metabolites in a cell maturation-specific manner. Endocrinology,132: 1544-1552, 1993.1321544 1993 [PubMed]

Seyedin, S. M.; Thompson, A. Y.; Bentz, H.; Rosen, D. M.; McPherson, J. M.; Conti, A.; Siegel, N. R.; Galluppi, G. R.; and Piez, K. A.: Cartilage-inducing factor-A. Apparent identity to transforming growth factor-beta. J. Biol. Chem.,261: 5693-5695, 1986.2615693 1986 [PubMed]

Shapiro, S. S., and Wilk, M. B.: An analysis of variance test for normality (complete samples). Biometrika,52: 591-611, 1965.52591 1965

Stevenson, S.; Cunningham, N.; Toth, J.; Davy, D.; and Reddi, A. H.: The effect of osteogenin (a bone morphogenetic protein) on the formation of bone in orthotopic segmental defects in rats. J. Bone and Joint Surg.,76-A: 1676-1687, Nov. 1994.76-A1676 1994

Takagi, K., and Urist, M. R.: The reaction of the dura to bone morphogenetic protein (BMP) in repair of skull defects. Ann. Surg.,196: 100-109, 1982.196100 1982 [PubMed]

Tuli, S. M.: Bridging of bone defects by massive bone grafts in tumorous conditions and in osteomyelitis. Clin. Orthop.,87: 60-73, 1972.8760 1972 [PubMed]

Urist, M. R.; Terashima, Y.; Nakagawa, M.; and Stamos, C.: Cartilage tissue differentiation from mesenchymal cells derived from mature muscle in tissue culture. In Vitro,14: 697-706, 1978.14697 1978 [PubMed]

Urist, M. R.; Lietze, A.; Mizutani, H.; Takagi, K.; Triffitt, J. T.; Amstutz, J.; DeLange, R.; Termine, J.; and Finerman, G. A. M.: A bovine low molecular weight bone morphogenetic protein (BMP) fraction. Clin. Orthop.,162: 219-232, 1982.162219 1982 [PubMed]

Urist, M. R.; Huo, Y. K.; Brownell, A. G.; Hohl, W. M.; Buyske, J.; Lietze, A.; Tempst, P.; Hunkapiller, M.; and DeLange, R. J.: Purification of bovine bone morphogenetic protein by hydroxyapatite chromatography. Proc. Nat. Acad. Sci.,81: 371-375, 1984.81371 1984 [PubMed]

Werntz, J.; Lane, J.; Piez, K.; Seyedin, S.; Burstein, A.; and Gebhardt, M.: The repair of segmental bone defects with collagen and marrow. Orthop. Trans.,10: 262-263, 1986.10262 1986

Winet, H., and Hollinger, J. O.: Incorporation of polylactide-polyglycolide in a cortical defect: neoosteogenesis in a bone chamber. J. Biomed. Mater. Res.,27: 667-676, 1993.27667 1993 [PubMed]

Zegzula, H. D.; Buck, D. C.; Brekke, J.; Wozney, J. M.; and Hollinger, J. O.: Bone formation with use of rhBMP-2 (recombinant human bone morphogenetic protein-2). J. Bone and Joint Surg.,79-A: 1778-1790, Dec. 1997.79-A1778 1997

Zellin, G., and Linde, A.: Treatment of segmental defects in long bones using osteopromotive membranes and recombinant human bone morphogenetic protein-2. An experimental study in rabbits. Scandinavian J. Plast. and Reconstr. Surg. and Hand Surg.,31: 97-104, 1997.3197 1997

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1 Photograph of a three-millimeter-thick microporous-macroporous polylactic acid-polyglycolic acid carrier.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-B Three months after the ostectomy, moderately hypertrophic nonunion was evident.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-C Immediately after implantation of the carrier.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-D Three months after implantation of the carrier, no bone-healing was evident and the plate was broken, indicative of persistent motion at the site of the nonunion.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-B Three months after the ostectomy, hypertrophic nonunion was evident.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-C Immediately after implantation of the carrier.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-D Twelve weeks after implantation of the carrier, there was clear evidence of complete endosteal bridging and some evidence of periosteal bridging.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-E Specimen treated with ten nanograms of TGF-ß1 alone (group E), showing fibrous connective tissue with inflammation and focal residual carrier and no bridging of the defect site.

View Larger

Group	Treatment	Periosteal Healing* (grade)	Endosteal Healing* (grade)
A (n = 6)	PLG50	1.63 ± 0.26	1.63 ± 0.24
B (n = 6)	15.0 mg BMP	2.48 ± 0.26†	2.08 ± 0.24
C (n = 6)	1.5 mg BMP	2.56 ± 0.26†	2.37 ± 0.24†
D (n = 6)	15.0 mg BMP and 10.0 ng TGF-ß1	2.48 ± 0.26†	2.25 ± 0.24†
E (n = 5)	10.0 ng TGF-ß1	1.35 ± 0.29‡	1.30 ± 0.27‡

Add to My JBJS

Group	Treatment	Periosteal Healing* (grade)	Endosteal Healing* (grade)
A (n = 6)	PLG50	1.63 ± 0.26	1.63 ± 0.24
B (n = 6)	15.0 mg BMP	2.48 ± 0.26†	2.08 ± 0.24
C (n = 6)	1.5 mg BMP	2.56 ± 0.26†	2.37 ± 0.24†
D (n = 6)	15.0 mg BMP and 10.0 ng TGF-ß1	2.48 ± 0.26†	2.25 ± 0.24†
E (n = 5)	10.0 ng TGF-ß1	1.35 ± 0.29‡	1.30 ± 0.27‡

View Larger

TABLE II HISTOMORPHOMETRIC ANALYSIS OF THE SITES OF CHRONIC NONUNIONS OF CANINE RADII TWELVE WEEKS AFTER RESECTION AND IMPLANTATION OF BIODEGRADABLE SCAFFOLDS WITH OR WITHOUT GROWTH FACTOR

Group	Treatment	Percent Total Bone Volume*	Percent Lamellar Bone*	Percent Active Osteoblasts*	Percent Residual Matrix*
A (n = 6)	PLG50	11.28 ± 7.27	10.80 ± 6.70	45.16 ± 9.87	19.54 ± 9.60
B (n = 6)	15.0 mg BMP	22.17 ± 7.27	18.66 ± 6.70	37.14 ± 9.87	33.63 ± 9.60
C (n = 6)	1.5 mg BMP	33.62 ± 7.27†	30.29 ± 6.70†	33.23 ± 9.87	5.42 ± 9.60‡
D (n = 6)	15.0 mg BMP and 10.0 ng TGF-ß1	36.58 ± 7.27†	32.63 ± 6.70†	40.43 ± 9.87	27.92 ± 9.60
E (n = 5)	10.0 ng TGF-ß1	18.80 ± 7.26	15.55 ± 7.34	18.92 ± 10.81	25.30 ± 10.51

Add to My JBJS

TABLE II HISTOMORPHOMETRIC ANALYSIS OF THE SITES OF CHRONIC NONUNIONS OF CANINE RADII TWELVE WEEKS AFTER RESECTION AND IMPLANTATION OF BIODEGRADABLE SCAFFOLDS WITH OR WITHOUT GROWTH FACTOR

Group	Treatment	Percent Total Bone Volume*	Percent Lamellar Bone*	Percent Active Osteoblasts*	Percent Residual Matrix*
A (n = 6)	PLG50	11.28 ± 7.27	10.80 ± 6.70	45.16 ± 9.87	19.54 ± 9.60
B (n = 6)	15.0 mg BMP	22.17 ± 7.27	18.66 ± 6.70	37.14 ± 9.87	33.63 ± 9.60
C (n = 6)	1.5 mg BMP	33.62 ± 7.27†	30.29 ± 6.70†	33.23 ± 9.87	5.42 ± 9.60‡
D (n = 6)	15.0 mg BMP and 10.0 ng TGF-ß1	36.58 ± 7.27†	32.63 ± 6.70†	40.43 ± 9.87	27.92 ± 9.60
E (n = 5)	10.0 ng TGF-ß1	18.80 ± 7.26	15.55 ± 7.34	18.92 ± 10.81	25.30 ± 10.51