ObjectiveTo evaluate the effect of tissue engineered periosteum on the repair of large diaphysis defect in rabbit radius, and the effect of deproteinized bone (DPB) as supporting scaffolds of tissue engineering periosteum. MethodsBone marrow mesenchymal stem cells (BMSCs) were cultured from 1-month-old New Zealand Rabbit and osteogenetically induced into osteoblasts. Porcine small intestinal submucosa (SIS) scaffold was produced by decellular and a series mechanical and physiochemical procedures. Then tissue engineered periosteum was constructed by combining osteogenic BMSCs and SIS, and then the adhesion of cells to scaffolds was observed by scanning electron microscope (SEM). Fresh allogeneic bone was drilled and deproteinized as DPB scaffold. Tissue engineered periosteum/DPB complex was constructed by tissue engineered periosteum and DPB. Tissue engineered periosteum was "coat-like" package the DPB, and bundled with absorbable sutures. Forty-eight New Zealand white rabbits (4-month-old) were randomly divided into 4 groups (groups A, B, C, and D, n=12). The bone defect model of 3.5 cm in length in the left radius was created. Defect was repaired with tissue engineered periosteum in group A, with DPB in group B, with tissue engineered periosteum/DPB in group C; defect was untreated in group D. At 4, 8, and 12 weeks after operation, 4 rabbits in each group were observed by X-ray. At 8 weeks after operation, 4 rabbits of each group were randomly sacrificed for histological examination. ResultsSEM observation showed that abundant seeding cells adhered to tissue engineered periosteum. At 4, 8, and 12 weeks after operation, X-ray films showed the newly formed bone was much more in groups A and C than groups B and D. The X-ray film score were significantly higher in groups A and C than in groups B and D, in group A than in group C, and in group B than in group D (P<0.05). Histological staining indicated that there was a lot of newly formed bone in the defect space in group A, with abundant newly formed vessels and medullary cavity. While in group B, the defect space filled with the DPB, the degradation of DPB was not obvious. In group C, there was a lot of newly formed bone in the defect space, island-like DPB and obvious DPB degradation were seen in newly formed bone. In group D, the defect space only replaced by some connective tissue. ConclusionTissue engineered periosteum constructed by SIS and BMSCs has the feasibility to repair the large diaphysis defect in rabbit. DPB isn't an ideal support scaffold of tissue engineering periosteum, the supporting scaffolds of tissue engineered periosteum need further exploration.