[Effects of bone morphogenetic protein-2 gene therapy on the bone-implant interface: an experimental study with dogs].

OBJECTIVE: To investigate the effects of bone morphogenetic protein-2 (BMP-2) gene therapy on the bone-implant interface in the reconstruction of periprosthetic bone defect. METHODS: Transverse defects were caused in the external condylae of both femurs of 14 adult Beagle dogs. Titanium alloy implants were inserted and a bone defect 3 mm wide around the titanium alloy implant was preserved. Then the total 28 defects were divided into 4 groups: 8 bone defects remained untreated (blank control group); 8 bone defects were implanted with heterogeneous freeze-dried bone by impaction grafting technique (non-cell group); 8 bone defects were implanted with heterogeneous freeze-dried bone loaded with autogenous bone marrow stromal cells (BMSCs) from the greater trochanter of the same dog (cell group); and 10 bone defects were implanted with freeze-dried allograft loaded with autogenous BMSCs from the greater trochanter of the same dog which were transfected by Adv-BMP-2 gene (gene group). Three, 6, and 12 weeks after implantation X-ray examination was carried out to observe the place of the implant and the absorption of the implants. Six and 12 weeks after the dogs were killed and their bone defects were taken out to undergo histological, histomorphometric and biomechanical examination to observe the healing and oseeointegration of the bone-implant interface. RESULTS: Histological examination showed that 6 weeks after implantation new bone formation was found on the implant surface and there was point contact between the bone and implant in the gene group with the bone-to-impact contact (BIC) of about 10%; and continuous soft tissue was found at bone-implant interface in all other groups. Twelve weeks after, there was thick soft tissue membrane between the new bone and implant in the blank control group; most of the interface was connective fibrous tissue in the non-cell group and cell group with point contact between the bone and implant and a BIC lower than 10%; and in the gene group the interface consisted mainly of bone tissue and continuous bone-implant contact was found with the BIC of 50%, significantly higher than those of the other 2 groups (both P < 0.01). The mechanical strength of interface increased time-dependently in all groups, that of the gene group being significantly higher than those of the other 2 groups at any time-points (both P < 0.01). CONCLUSION: BMP-2 gene therapy can improve the osseointegration of bone-implant interface.