ObjectiveTo study the effect of the loaded concentration gradient of nerve growth factor (NGF) immobilized conduit on rat peripheral nerve defect repair. MethodsThe peripheral nerve conduits made of poly (ε-caprolactone)-block-poly (L-lactide-co-ε-caprolactone) were prepared with uniform loads or concentration gradient loads by combining differential absorption of NGF/silk fibroin (SF) coating, and the gradient of NGF was immobilized in the nerve conduits. ELISA method was used to exam the NGF release for 12 weeks in vitro. Twenty-four male Sprague Dawley rats (weighing, 220-250 g) were selected to establish the right sciatic nerve defect model (14 mm in length) and randomly divided into 4 groups according to repair methods. The transected nerve was bridged by a blank conduit without NGF in group A, by a conduit containing uniform loads of NGF in group B, by a conduit concentration gradient loads of NGF in group C, and by the autogenous nerve segment in group D. The gross observation, electrophysiological examination, histological observation, and transmission electron microscope observation were carried out to assess the nerve regeneration at 12 weeks after surgery. ResultsThe cumulative release amount of NGF was (14.2±1.4) ng/mg and (13.7±1.3) ng/mg in gradient of NGF loaded conduits and uniform NGF loaded conduits respectively at 12 weeks, showing no significant difference (t=0.564, P=0.570). All the animals survived to completion of the experiment; plantar ulcers occurred at 4 days, which healed at 12 weeks; groups C and D were better than groups A and B in ulcerative healing. At 12 weeks after surgery, the compound muscle action potential of group A was significantly lower than that of groups B, C, and D (P<0.05), and group B was significantly lower than groups C and D (P<0.05), but no significant difference was found between groups C and D (P>0.05). The axon density of group C was significantly higher that of groups A, B, and D (P<0.05); group D was significantly higher than groups A, B, and C, and group C was significantly higher than groups A and B in the axon number, axon diameter, and area of muscle fiber (P<0.05); the thickness of myelin sheath of groups C and D was significantly larger than that of groups A and B (P<0.05), but no significant difference was found between groups C and D (P>0.05). ConclusionGradient of NGF loaded nerve condnits for rat sciatic nerve defect has similar results to autogenous nerve, with a good bridge, which can promote the sciatic nerve regeneration, improve the myelinization of the regenerating nerve, and accelerate the function reconstruction of the regenerating nerve.