To observe the clinical effect of moxibustion with deqi on Alzheimer's disease (AD) rats, and evaluate its effect on β-amyloid (Aβ) transport and enzymatic degradation proteins, to explore its molecular mechanism for improving cognitive function. Sixty SPF-grade male SD rats were randomly divided into a blank group (8 rats), a sham-operation group (8 rats) and a model establishment group (44 rats). The rats in the model establishment group were injected with Aβ1-42 at bilateral ventricles to establish AD model. Among the 38 rats with successful model establishment, 8 rats were randomly selected as the model group, and the remaining rats were treated with mild moxibustion at "Dazhui" (GV 14), once a day, 40 min each time, for 28 days. According to whether deqi appeared and the occurrence time of deqi, the rats were divided into a deqi group (12 rats), a delayed deqi group (10 rats) and a non-deqi group (8 rats). After the intervention, the Morris water maze test was applied to evaluate the cognitive function; the HE staining was applied to observe the brain morphology; the Western blot method was applied to measure the protein expression of Aβ and its receptor mediated transport [low-density lipoprotein receptor-related protein (LRP) 1, receptor for advanced glycation end products (RAGE), apolipoprotein E (ApoE)] and enzymatic degradation [neprilysin (NEP), insulin degrading enzyme (IDE), endothelin converting enzyme (ECE)-1 and angiotensin converting enzyme (ACE) 2]. Compared with the sham-operation group, in the model group, the escape latency was prolonged (P<0.01), and the times of platform crossing and the ratio of platform quadrant to total time were reduced (P<0.01); the brain tissue was seriously damaged; the expression of hippocampal Aβ and RAGE was increased (P<0.01), and the expression of hippocampal LRP1, ApoE, NEP, IDE, ECE-1 and ACE2 was decreased (P<0.01). Compared with the model group, the escape latency was shortened in the deqi group (P<0.05, P<0.01), and the escape latency in the delayed deqi group and the non-deqi group was shortened from Day 2 to Day 5 (P<0.05, P<0.01), and the times of platform crossing and the ratio of platform quadrant to total time were increased in the deqi group and the delayed deqi group (P<0.01, P<0.05); the brain damage in each moxibustion group was reduced, which was smallest in the deqi group, followed by the delayed deqi group and the non-deqi group; the expression of Aβ and RAGE was decreased (P<0.01, P<0.05) and the expression of LRP1 and IDE was increased in each moxibustion group (P<0.01, P<0.05); the expression of ApoE was increased in the deqi group and the delayed deqi group (P<0.01, P<0.05); the expression of NEP was increased in deqi group (P<0.05), and the expression of ECE-1 and ACE2 was increased in the deqi group and the delayed deqi group (P<0.05). Compared with the delayed deqi group and the non-deqi group, the escape latency in the deqi group was shortened from Day 3 to Day 5 (P<0.05), and the times of platform crossing and the ratio of platform quadrant to total time were increased (P<0.05, P<0.01). Compared with the non-deqi group, the expression of Aβ was reduced (P<0.05), the expression of LRP1 and ApoE was increased in the deqi group (P<0.05). The expression of NEP in the deqi group was higher than that in the delayed deqi group and the non-deqi group (P<0.05). Compared with non-deqi, moxibustion with deqi could promote Aβ transport and degradation, thereby reducing Aβ level in the brain and improving cognitive function for AD rats.