Background: There is no ideal treatment for end-stage degenerative wrist disorders and subsequent carpal collapse. The purpose of this study was to investigate whether autologous cartilage constructs tissue-engineered from bone-marrow-derived mesenchymal stem cells can be effective for carpal bone reconstruction.
Methods: Total lunate excision was performed in twenty-seven adult New Zealand White rabbits. Mesenchymal stem cells were isolated from marrow and then were culture-expanded. Group-1 rabbits underwent excision only. Group-2 animals underwent excision followed by implantation of a scaffold consisting of gelatin and hyaluronan. Group-3 animals underwent excision followed by implantation of a mesenchymal stem cell-seeded scaffold that had been preincubated in chondrogenic medium. The group-1 animals were killed at six weeks, whereas the group-2 and group-3 animals were killed at six or twelve weeks. Tissues were harvested for radiographic and histologic analysis.
Results: Significant carpal collapse (a 5.4% +/- 2.8% reduction in the carpometacarpal index, p < 0.05) was observed in the group-1 animals by six weeks. In contrast, the carpal height was maintained in the group-2 and 3 animals. There was no radiographic evidence of ossification in the group-1 or 2 animals, whereas there was radiographic evidence of ossification in all six group-3 rabbits killed at the twelve-week time-point. Histologic sections from the group-3 animals showed filling of the lunate space with islands of cartilage with interspersed bone ossicles at six weeks. At twelve weeks, there was abundant bone formation as well as evidence of neovascularization. Osseous tissue was present in the central portions of the constructs while the periphery was lined with cartilage. In groups 1 and 2, the lunate space was filled with poorly organized fibrous tissue.
Conclusions: Cartilaginous implants preformed from autologous mesenchymal stem cells seeded onto biodegradable scaffold can prevent carpal collapse. The newly formed osteochondral tissue appears to function as an adequate biologic lunate spacer for at least twelve weeks in this animal model.