Abstract: OBJECTIVE  To observe the effect of acupuncture on the safety of hypertime window thrombolysis in rats with acute cerebral infarction, and to explore its mechanism from the way of astrocytes.METHODS  Cerebral infarction model was prepared by improved autoembolism method in rats. "Xingnao Kaiqiao" acupuncture was used as interventional method, and rt-PA was injected through caudal vein as thrombolysis method. 66 SD rats were randomly divided into sham operation group, model group, 4.5 h thrombolysis group, 6 h thrombolysis group, acupuncture+4.5 h thrombolysis group and acupuncture+6 h thrombolysis group, with 11 rats in each group. Bederson method was used to evaluate the neurobehavioral scores of rats at the 2nd and 24th hour after modeling. TTC staining method was used to compare the cerebral infarct volume percentage of rats in each group, EB osmotic method was used to compare the blood-brain barrier (BBB) permeability of rats in each group, and dry-wet weight method was used to measure the brain water content of rats in each group. 48 SD rats were randomly divided into sham operation group, model group, 6 h thrombolysis group and acupuncture+6 h thrombolysis group, with 12 rats in each group. The qPCR and Western blot were used to detect the mRNA and protein expressions of astrocyte-related indexes glial fibrillary acidic protein (GFAP) and aquaporin-4 (AQP-4) in the cerebral cortex of rats in each group.RESULTS  Compared with the sham operation group, the neurobehavioral score, percentage of cerebral infarct volume and brain water content in the model group significantly increased (P < 0.01). The neurobehavioral scores, percentage of cerebral infarct volume and brain water content in the 4.5 h thrombolysis group and the acupuncture+4.5 h thrombolysis group were lower than those in the model group (P < 0.01). Compared with the model group and the 6 h thrombolytic group, the neurobehavioral score, the percentage of cerebral infarct volume and brain water content in the acupuncture+6 h thrombolytic group significantly decreased (P < 0.05, P < 0.01). The mRNA and protein expression levels of GFAP and AQP-4 in model group were significantly higher than those in sham group (P < 0.05, P < 0.01). The mRNA and protein expression levels of GFAP and AQP-4 in acupuncture+6 h thrombolysis group were significantly lower than those in model group and 6 h thrombolysis group (P < 0.05, P < 0.01).CONCLUSION  Acupuncture can inhibit the activation of astrocytes by down-regulating the expression of GFAP and AQP-4, thus improve the safety of hypertime window thrombolysis in cerebral infarction.