Objectives: To observe the effects of electroacupuncture (EA) at "Jiaji" (EX-B 2) combined with neurodynamic mobilization (NM) on the cross-sectional area of the gastrocnemius muscle fibers after sciatic nerve injury in rabbits, and the expression of nuclear factor κB (NF-κB) and muscle-specific ring-finger protein 1 (MuRF1)., Methods: A total of 180 common-grade New Zealand rabbits (half male and half female) were randomly divided into five groups, i.e. a normal control group, a model control group, a NM group, an EA group and a combined intervention group, 36 rabbits in each group. Except in the normal control group, clipping method was used to prepare the model of sciatic nerve injury in the rest groups. On the 3rd day of successful modeling, NM was delivered in the NM group. In the EA group, EA was exerted at bilateral "Jiaji" (EX-B 2) of L 4 to L 6 , stimulated with disperse-dense wave and the frequency of 2 Hz/100 Hz. In the combined intervention group, after EA delivered at bilateral "Jiaji" (EX-B 2) of L 4 to L 6 , NM was operated. The intervention in each group was delivered once daily, for 6 days a week, and lasted 1, 2 or 4 weeks according to the collection time of sample tissue. After 1, 2 and 4 weeks of intervention, in each group, the toe tension reflex score and the modified Tarlov test score were observed; the morphology of the gastrocnemius muscle was observed by HE staining and the cross-sectional area of muscular fiber was measured; using Western blot method, the expression of NF-κB and MuRF1 of the gastrocnemius muscle was detected., Results: After 1, 2 and 4 weeks of intervention, the toe tension reflex scores and the modified Tarlov scores in the model control group were lower than those of the normal control group ( P< 0.05), and these two scores in the NM group, the EA group and the combined intervention group were all higher than those of the model control group ( P <0.05); the scores in the combined intervention group were higher than those in the EA group and the NM group ( P< 0.05). The gastrocnemius fibers were well arranged and the myocyte morphology was normal in the normal control group. In the model control group, the gastrocnemius fibers were disarranged, the myocytes were irregular in morphology and the inflammatory cells were infiltrated in the local. In the NM group, the EA group and the combined intervention group, the muscle fibers were regularly arranged when compared with the model control group. After 1, 2 and 4 weeks of intervention, the cross-sectional areas of the gastrocnemius muscle fibers in the model control group were smaller than those of the normal control group ( P< 0.05). The cross-sectional areas in the NM group, the EA group and the combined intervention group were larger than those of the model control group ( P< 0.05), and the cross-sectional areas in the combined intervention group were larger than those in the NM group and the EA group ( P< 0.05). After intervention for 1, 2 and 4 weeks, the protein expressions of NF-κB and MuRF1 in the gastrocnemius muscle were higher in the model control group in comparison of those in the normal control group ( P <0.05). In the NM group, the EA group and the combined intervention group, the expressions of NF-κB after intervention for 1, 2 and 4 weeks and the expressions of MuRF1 after 2 and 4 weeks of intervention were lower when compared with those in the model control group ( P< 0.05). In the combined intervention group, the protein expressions of NF-κB after intervention for 1, 2 and 4 weeks and the expressions of MuRF1 after 2 and 4 weeks of intervention were decreased when compared with those in the NM group and the EA group ( P< 0.05)., Conclusions: Electroacupuncture at "Jiaji" (EX-B 2) combined with NM may increase the muscle strength and sciatic function and alleviate gastrocnemius muscle atrophy in the rabbits with sciatic nerve injury. The underlying mechanism is related to the inhibition of NF-κB and MuRF1 expression.