Abstract: Objective: The aim of this study was to prepare a new model of cognitive impairment induced by cerebral ischemia in diabetic rats. Method: A total of 70 female Wistar rats，weighing 150-200g (around 6 weeks of age), were randomly divided into 5 groups: a normal control group (n=15), an ischemic group(n=18), a diabetic sham-operation group (n=19) and a diabetic ischemic group (n=18). To induce diabetes, streptozotocin (STZ) was injected i.p. at a dose of 60 mg/kg, 4 days before the preparation of experimental ischemia. The method used to produce cerebral ischemia was the two-vessel occlusion followed by re-perfusion. Under anesthesia with sodium pentobarbital, both common carotid arteries were exposed through a ventral midline cervical incision. The common carotid arteries were occluded with an artery forceps for 15min which was followed by a re-perfusion for 5min. The operation of occlusion and re-perfusion was performed twice. The rats for sham-operation controls (operated at the same time points) were treated in the same manner except the occlusion of common carotid arteries. The blood sugar level was determined 7 days after STZ administration. The animals with the blood glucose concentration higher than 15.5mM were selected as diabetic rats. After 1-month recovery the animal model with cognitive impairment was confirmed by behavioral detection. The animals Learning and memorial behaviors were investigated using a passive and active avoidance response tests and a spatial version of the Morris water maze test. The situation of neuron loss in the area of hippocampus was determined by the pathological H-E staining method. Result: The performance of passive avoidance in diabetic rats with cerebral ischemia was significantly impaired. The step-down latency of the passive avoidance response of the model rats was shorter, especially at 5th min after training (P<0.01), than that in other groups 30 days after production of cerebral ischemia. During 5 days-training of active avoidance, the value of AR in the model group was less than that of the control group, especially at the 1st and 3rd day, there was statistical significance between the two groups (P<0.01). The training times of the model group were significantly more than that of the other three groups (P<0.001). The swimming time of the rats in Morris water maze in model group was significantly shorter than that in other groups (P<0.05 as compared with DM+Sham; P<0.01 as compared with other 2 groups). The swimming distance of rats in model group was the shortest among all 4 groups (P<0.05). The result indicated that the memory ability in the model rats was deficient. Conclusion: The cognitive ability in diabetic rats with cerebral ischemia 30 days after being induced was significantly declined compared with the other three groups and diabetes can worsen neuronal damage of the brain via cerebral ischemia.

Key words: diabetes, cerebral ischemia, cognitive impairment, animal model