To explore the influence of electroacupuncture (EA) on the expression of AMPA receptor subunit GluR1 in the rats with acute spinal cord injury (SCI) and explore the potential effect mechanism of EA in treatment of acute SCI. A total of 80 SD rats were randomly divided into five groups, i.e. a sham-operation group, a model group, an AMPA antagonist (DNQX) group, an EA group and a DNQX+EA group, 16 rats in each group. The modified Allen's impacting method was adopted to prepare the rat model of acute SCI at T10. In the DNQX group, the intrathecal injection of 10 μL DNQX solution with a concentration of 1 nmol/μL was administered in 0.5 h after modeling success. In the EA group, EA (disperse-dense wave, 2 Hz/100 Hz in frequency, 0.5 mA in output current) was given at "Dazhui" (GV 14) and "Mingmen" (GV 4) in 0.5 h, 12 h and 24 h after modeling success for 30 min and totally 3 times. In the DNQX + EA group, the interventions in the above two groups were managed. The Basso, Beattie and Bresnahan locomotor rating score (BBB) was applied to evaluate the changes of locomotor function in the rats before modeling and in 6 h, 24 h and 48 h after modeling successively. Using the hematoxylin-eosin (HE) staining, the histopathological changes in the spinal anterior horn were observed in the spinal injured area. The immunofluorescence method was adopted to determine the number of GluR1 positive neuron of the spinal anterior horn. The Western blot method was used to determine the protein expression of GluR1 in the injured area. Compared to the sham-operation group in 6 h, 24 h and 48 h after modeling, the BBB scores were all significantly decreased in the model group (P<0.001) at the corresponding points. The BBB score was increased in each of intervention groups, but without statistical difference as compared with the model group (P>0.05). In the model group, it was found that the boundary between gray matter and white matter in the spinal anterior horn was blurred, the interstitial space enlarged, the neuron volume obviously shrunken, the cytoplasm decreased, the red stain deepened and some neuron nuclei fixed and shrunk. In the EA group, the morphology of the spinal anterior horn in the injured area was improved obviously, which was similar in the DNQX group and the DNQX + EA group. Compared with the sham-operation group, the GluR1 protein expression in the spinal injury area was increased (P<0.001) and the number of GluR1 positive neurons elevated (P<0.001) in the spinal anterior horn in the model group. Compared with the model group, in the EA group, the DNQX group and the DNQX + EA group, GluR1 protein expression was decreased (P<0.05, P<0.01) and the number of GluR1 positive neurons in the spinal anterior horn reduced (P<0.001). The intervention with EA at "Dazhui" and "Mingmen" promotes the repair of the injured nerve in the spinal anterior horn probably through inhibiting GluR1 expression in the spinal injured area in the rats with acute SCI.