Session I - Combined Session (International Society for Fracture Repair)

Growth Factor Expression in a New Atrophic Nonunion Model: Implications for the Treatment of Fractures

Anita Reed, DPhil; Clive Joyner, DPhil; Harry Brownlow, FRCS Orth; Hamish Simpson, Prof/FRCS Orth; Nuffield Department of Orthopaedic Surgery, University of Oxford, Oxford, United Kingdom

Purpose: The presence and distribution of important growth factors such as TGF-b, FGF, PDGF and BMP 2/4 has not previously been documented during abnormal bone healing at both early and late time points. Therefore, the aims of this study were to develop and validate a clinically relevant small animal atrophic nonunion model and to test the hypothesis that growth factor expression is diminished in atrophic nonunions.

Methods: Twenty-eight adult female Wistar rats underwent application of a novel circular-frame external fixator to the right tibia. The periosteum was removed from 14 of the 28 animals and the intramedullary canal was curetted. Both insults were performed proximally and distally for a distance equivalent to one diameter of the tibia. A -mm gap was introduced. Animals were sacrificed at 1, 3, 8, and 16 weeks. Paraffin sections were immunohistochemically stained for TGF-b, FGF basic, PDGF and BMP 2/4.

Results: At 8 and 16 weeks, all animals in which stripping was performed went on to form an atrophic nonunion. Those in which stripping was not performed united successfully. The immunohistochemical results were as follows. At 1 week, positive staining was observed in the hematoma in both groups and in the new bone matrix of the healing group. At 3 weeks, staining was similar in the two groups, with most tissues continuing to express growth factors. At 8 weeks, the periosteum in the healing group no longer expressed TGF-b and BMP 2/4 as it began to remodel. The nonunion group continued to express all four growth factors. At 16 weeks, as the healing group remodeled, staining of growth factors became weaker in all tissues and cells. The fibrous tissue in the nonunion group appeared to be more weakly stained than at earlier time points.

Conclusion: These results may have relevance for new therapies that can be aimed at delivering growth factors to treat fracture nonunions. At 1, 3, and 8 weeks, the nonunion group appeared to stain similarly to the healing group. At 16 weeks, however, consistent with our hypothesis, there was less staining of the fibrous tissue in the nonunion gap than at earlier time points. These results may suggest that, at 16 weeks, the fibrous tissue within the nonunion gap had reached a steady state. Therefore, this may be a suitable time for the delivery of growth factors to the nonunion site.