Native bovine bone morphogenetic protein in the healing of segmental long bone defects

Tapio Tuominen

Abstract

A new animal model was developed to evaluate the effect of bovine native bone morphogenetic protein (BMP) on the healing of segmental, critical-sized bone defects. Laboratory-bred adult beagle dogs were used in the study. A 2 cm corticoperiosteal defect was created using an oscillating saw in mid-ulna, and the defect was treated with bone grafts and implants fixed by an intramedullary Kirschner wire through predrilled holes in the middle of the implant. Plate and screw fixation was also used in some groups. Coral, hydroxyapatite and demineralized xenograft bone were placed in the defects with or without BMP. Autografts and allografts were used as controls. The BMP was extracted from bovine diaphyseal bone.

The follow-up period was 36 weeks. Radiographs were taken at regular intervals during the follow-up period, and bone formation and bone union were evaluated. The radiographs were digitized, and callus was measured and CT scans obtained to define bone density. At the end of the study, the bones were harvested and tested mechanically in a torsion machine until failure. After mechanical testing, the bones were reconstructed and histological sections were made.

With autograft and allograft bone grafts, healing was nearly complete. Hydroxyapatite and demineralized xenograft bone did not result in healing of the bone defect, while coral enhanced bone formation, but the healing was not comparable to autografts or allografts. Hydroxyapatite implants did not resorb during the 36 weeks of follow-up to enhance bone healing, and there was a fibrous capsule around the hydroxyapatite implants in histology. Xenograft bone was resorbed, and very little bone formation and extensive fibrosis were seen at the implant site. Coral was resorbed and gradually replaced by new bone, but did not heal the defect completely. With every implant, added BMP had a positive effect on healing as evaluated either radiographically, mechanically or histologically. Coral was the most optimal carrier material for BMP among the materials tested in this study.

The animal model seems to be suitable for studying the healing of bone defects, as all the animals were physically active from the first postoperative day and did not seem to have problems with motion during the follow-up period. Intramedullary fixation lacks rotational stability, which may have a negative effect on healing. The bones fixed with a plate and screws showed better scores in radiographs and were mechanically stronger, although the study groups were too small to allow definitive conclusions. As a conclusion, none of the transplants or implants were equally efficient as cortical autograft in healing segmental ulnar defects. BMP did not enhance the poor capacity of hydroxyapatite and xenograft bone to heal the bone defect. According to the present findings, the composite implant consisting of coral and BMP seemed to be the best of the composite implants tested

Table of Contents
Acknowledgements
Glossary
List of original publications
1. Introduction
2. Review of the literature
2.1. Bone grafting and bone substitutes
2.2. Osteoinduction
2.3. Bone morphogenetic protein
2.3.1. TGF-β superfamily
2.3.2. Extracted BMPs
2.3.3. Recombinant BMPs
2.3.4. Functions of BMPs
2.3.5. BMP carriers
2.4. Canine ulnar segmental defect
2.5. Treatment of a segmental bone defect with BMP
2.5.1. Naturally occurring BMP preparations in segmental bone defects
2.5.2. Naturally occurring BMP preparations in canine ulnar defect
2.5.3. Recombinant BMP in segmental bone defects
2.5.4. Recombinant BMP in canine ulnar defect
2.6. Gene therapy in a bone defect model
2.7. Other animal models with BMP
2.8. Clinical use of BMP
2.9. Other applications under development
3. Aims of the present study
4. Materials and methods
4.1. Animals
4.2. Patient
4.3. BMP
4.4. Bone grafts and implants
4.5. Surgical procedures
4.5.1. Clinical case
4.6. Methods of analysis
4.6.1. Radiography
4.6.2. Mechanical testing
4.6.3. Densitometry
4.6.4. Histology
4.6.5. Statistical analysis
5. Results
5.1. Coral and tricalcium phosphate healing sheep tibial defects
5.2. Healing of canine ulnar defects
5.2.1. Auto- and allografts
5.2.2. Xenografts and xenograft composite implants
5.2.3. Coral implants and coral composite implants
5.2.4. Hydroxyapatite implants and hydroxyapatite composite implants
5.3. Coral composite implant in the treatment of ulnar pseudoarthrosis (A clinical case)
6. Discussion
6.1. Methodological considerations
6.1.1. Animal models and fixation methods
6.1.2. Healing evaluated from radiographs
6.1.3. Mechanical testing
6.1.4. Densitometry
6.1.5. Histology
6.2. Bone grafts and implants
6.3. Effect of BMP
6.4. Future prospects of clinical use of BMP
7. Conclusions
References
List of Tables
1. Bone Morphogenetic Protein Superfamily in Mammals (Reddi 1998).
2. Summary of the methods of analysis used in segmental bone defect models treated with BMP.
3. Summary of scores for bone union and bone formation.
List of Figures
1. A radiograph showing the study model. An autograft transplant in the defect fixed with an intramedullary Kirschner wire.
2. Torsion testing machine with a bone attached ready to be tested.
3. A series of radiographs of an ulnar defect treated with an autograft transplant leading to solid bone union.
4. A histological section of an autograft transplant showing new bone filling the gap and remodellation.
5. A series of radiographs showing an ulnar defect treated with coral-BMP composite implant. The implant was resorbed gradually and replaced by bone. The union was not complete despite the good new bone formation. The Kirschner wire came out in this case and was removed after 3 weeks.
6. Area of callus measured in digitized roentgenograms of the autograft, coral and coral composite implants (o = autograft, x = coral implant, ¤ = coral+BMP composite implant).
7. A. A histological section of a coral implant. Some remnants of the coral material can be seen with a small amount of new bone in the middle. B. A histological section of a coral+BMP composite implant showing more bone formation in the implant area, which was here fixed with screws and plate (Original magnification x4).
8. A series of radiographs of an ulnar defect treated with a hydroxyapatite-BMP composite implant. The implant did not resorb and there is non-union. There is some callus formation, which is not, however, in contact with the hydroxyapatite implant.
9. A histological section of a hydroxyapatite implant showing unresorbable material and some callus, but there is clearly non-union. (Original magnification x4).
  Next
  Acknowledgements

Homepage of this publication

Library Units | Collections | Databases | Library News | Library Services | Electronic Collection | Links elsewhere | Alphabetical Index

© 2000 Oulu University Library