Background and ObjectiveTranscranial magnetic stimulation (TMS) is considered as a promising treatment option for post-stroke cognitive impairment (PSCI).Some meta-analyses have indicated that TMS can be effective in treating cognitive decline in stroke patients, but the quality of the studies included and the methodologies employed were less than satisfactory. Thus, this meta-analysis aimed to evaluate the efficacy and safety of TMS for treating post-stroke cognitive impairment.MethodsWe searched online databases like PubMed, Embase, Cochrane Library, and Web of Science to retrieve randomized controlled trials (RCTs) of TMS for the treatment of patients with PSCI. Two independent reviewers identified relevant literature, extracted purpose-specific data, and the Cochrane Risk of Bias Assessment Scale was utilized to assess the potential for bias in the literature included in this study. Stata 17.0 software was used for data analysis.ResultsA total of 10 studies involving 414 patients were included. The results of the meta-analysis showed that TMS was significantly superior to the control group for improving the overall cognitive function of stroke patients (SMD = 1.17, 95% CI [0.59, 1.75], I2 = 86.1%, P < 0.001). Subgroup analyses revealed that high-frequency rTMS (HF-rTMS), low-frequency rTMS (LF-rTMS), and intermittent theta burst stimulation (iTBS) all have a beneficial effect on the overall cognitive function of stroke patients. However, another subgroup analysis failed to demonstrate any significant advantage of TMS over the control group in terms of enhancing scores on the Loewenstein Occupational Therapy Cognitive Assessment (LOTCA) and Rivermead Behavioral Memory Test (RBMT) scales. Nonetheless, TMS demonstrated the potential to enhance the recovery of activities of daily living in stroke patients, as indicated by the Modified Barthel Index (MBI) (SMD = 0.76; 95% CI [0.22, 1.30], I2 = 52.6%, P = 0.121).ConclusionThis meta-analysis presents evidence supporting the safety and efficacy of TMS as a non-invasive neural modulation tool for improving global cognitive abilities and activities of daily living in stroke patients. However, given the limited number of included studies, further validation of these findings is warranted through large-scale, multi-center, double-blind, high-quality randomized controlled trials.PROSPERO registration numberCRD42022381034.