the diabetic rat model was established by intraperitoneal injection of streptozotocin, and 8 of the rat models were randomly selected as the diabetic group. Meanwhile, the remaining 24 diabetic rats were treated with rose bengal combined with 532 nm laser to construct the AION rat model. Afterwards, the above-mentioned AION rat models were randomly divided into three groups with 8 in each group, including the AION rat group (received no treatment), the electrical stimulation group and the sham electrical stimulation group. The electrical stimulation group was given corneal electrical stimulation with the followingparameters: Current of 1 mA, frequency of 20 Hz, wave width of 1 ms/phase, and stimulation time of 1 hour once every other day for 2 weeks. Thesham electrical stimulation group had an electrode placed in the same position as that of the electrical stimulation group, but the power was notconnected. After 2 weeks, rats in the 5 above-mentioned groups were analyzed with fundus color photography, optical coherence tomography (OCT) and visual evoked potentials (VEP). Moreover, after the rats were sacrificed, the retina and optic nerve were frozen and sectioned, and hematoxylin and eosin stains were applied to the tissues. LSD-t tests and one-way analysis of variance were used to analyze the data. Results: The average retinal thickness of the superior optic disc was 211±13 μm in the normal rat group, 206±16 μm in the diabetic rat group, 240±54 μm in the AION rat group, 216±11 μm in the sham stimulation group, and 198±4 μm in the electrical stimulation group. There were significant differences in the average retinal thickness of the superior optic disc among the five groups (F=2.854, P=0.038). Of them, the average retinal thickness of the superior optic disc in the AION group was greater than the  

However, no significant difference was found between the normal group and the diabetic group (P>0.05), or between the AION group and the sham stimulation group (P>0.05). VEP revealed that the N1 latency of the AION group was longer than that of the electrical stimulation group, and the difference was statistically significant (t=4.1, P<0.001). The P1 latency of the AION group was extended compared to latencies in the normal group, the diabetic group, the sham electrical stimulation group and the electrical stimulation group, and the differences were statistically significant (t=4.1, 2.5, 2.6, 3.2, P<0.05). In addition, the amplitude of N1-P1 in the electrical stimulation group increased relative to that in the sham electrical stimulation group, and the difference was statistically significant (t=4.0, P<0.001). Conclusions: Corneal electrical stimulation can promote optic disc thinning in AION rats under hyperglycemia status, accelerate the regression of optic disc edema, and partially improve visual function.