Objective To investigate the effect of scalp acupuncture combined with repeated transcranial magnetic stimulation (rTMS) on neurological function in stroke rats and its effect on the protein kinase A (PKA) / cyclophosphine response element binding ( CREB) protein signaling pathway. Methods Fifty rats were randomly divided into a sham operation group, model group, scalp group, rTMS group, and combined group (10 rats per group). Nylon thread was not inserted in the sham operation group. The other groups were used to prepare a model of focal stroke rats using sutures. Nine rats in the model group, eight in the scalp group, eight in the rTMS group, and nine in the combined group were successfully modeled. Rats in the model group and sham operation group did not receive the intervention. The scalp group received scalp acupuncture, the rTMS group received rTMS, and the combined group received scalp acupuncture + rTMS. After the last intervention, neurological deficit score and Morris water maze test performance were assessed. After triphenyl tetrazolium chloride (TTC) and hematoxylin and eosin (HE) staining, the pathological cortical changes in and around the infarct of brain tissues were observed. The expressions of PKA, CREB protein, and phosphorylated CREB ( p-CREB) protein were measured. Results Compared with the sham operation group, the model group, scalp group, rTMS group, and combined group had higher neurological injury scores, longer escape latency, and shorter target quadrant stay duration (P< 0. 05). Compared with the model group, the scalp group, rTMS group, and combination group showed lower neurological injury scores, shorter escape latency, and longer target quadrant stay duration (P<0. 05). Compared with the scalp group and the rTMS group, the combined group showed decreased neurological injury score, shortened escape latency, and prolonged target quadrant stay duration (P<0. 05). TTC staining identified white infarcts in the cerebral cortex of the ischemic side of the model group and mild erosion. The infarct area of the cerebral cortex of the ischemic side of the scalp acupuncture group, rTMS group, and combination group reduced and the appearance of the brain tissue improved; the improvement effect was most obvious in the combination group. The brain tissue structure of the model group was loose and showed liquefying necrosis in the cribriform network, with a large number of neurons disappearing and a large number of microglia infiltrating. The brain tissue structure of the scalp acupuncture group and rTMS group showed less damage than that of the model group, with some remaining normal neurons. However, some cytoplasm showed vacuolation and edema. The brain tissue structure and neuron morphology of the combined group tended to be positive. Compared with the sham operation group, the model group, scalp acupuncture group, rTMS group, and combination group showed decreased relative expressions of PKA and p-CREB protein (P<0. 05). Compared with the model group, the scalp acupuncture group, rTMS group, and combination group showed increased relative expressions of PKA and p-CREB protein ( P < 0. 05 ); the expression was higher in the combined group than in the scalp acupuncture group and rTMS group (P<0. 05). Conclusions Scalp acupuncture combined with rTMS can effectively improve neurological function in stroke rats. Its mechanism may play a regulatory role by activating the PKA/ CREB signaling pathway.