Repairing peripheral nerve defects with tissue engineered artificial nerves in rats

OBJECTIVE: To observe the effect of tissue engineered nerves in repairing peripheral nerve defects (about 1.5 cm in length) in rats to provide data for clinical application. METHODS: Glycerinated sciatic nerves (2 cm in length) from 10 Sprague Dawley (SD) rats (aged 4 months) were used to prepare homologous dermal acellular matrix. Other 10 neonate SD rats (aged 5-7 days) were killed by neck dislocation. After removing the epineurium, the separated sciatic nerve tracts were cut into small pieces, then digested by 2.5 g/L trypsin and 625 U/ml collagenase and cultured in Dulbecco's modified Eagle's medium (DMEM) for 3 weeks. After proliferation, the Schwann cells (SCs) were identified and prepared for use. And other 40 female adult SD rats (weighing 200 g and aged 3 months) with sciatic nerve defects of 1.5 cm in length were randomly divided into four groups: the defects of 10 rats bridged with proliferated SCs and homologous dermal acellular matrix (the tissue engineered nerve group, Group A), 10 rats with no SCs but homologous dermal acellular matrix with internal scaffolds (Group B), 10 with autologous nerves (Group C), and the other 10 with nothing (the blank control group, Group D). The general status of the rats was observed, the wet weight of triceps muscle of calf was monitored, and the histological observation of the regenerated nerves were made at 12 weeks after operation. RESULTS: The wounds of all 40 rats healed after operation and no death was found. No foot ulceration was found in Groups A, B and C, but 7 rats suffered from foot ulceration in Group D. The triceps muscles of calf were depauperated in the experimental sides in all the groups compared with the uninjured sides, which was much more obvious in Group D. The wet weight of triceps muscle of calf and nerve electrophysiologic monitoring showed no statistical difference between Group A and Group C, but statistical difference was found between Groups A and B and Groups B and D. And significant statistical difference was found between Group B and Group D. Obvious compound muscle (or motor) action potential (CMAP) could be evoked in Group A and Group C, but the evoked amplitude was very low in Group B and Group D. The axons of regenerated nerves penetrated through the whole graft in Group A and Group C, and partly penetrated through the graft in Group B, but did not penetrate in Group D. The two tips of the separated sciatic nerves of Groups A , B , and C were connected together, without formation of neuroma. But those of Group D were not connected together and neuroma formed in 6 rats. CONCLUSIONS: Tissue engineered nerves can be used for repairing long defects of the peripheral nerves of rats and ideal repairing effects can be obtained.