ObjectiveTo systematically evaluate the effectiveness of acupuncture combined with rehabilitation training compared with simple rehabilitation training or acupuncture treatment of dysphagia after stroke, and make clear whether the effect of acupuncture combined with rehabilitation training is better than simple rehabilitation training or acupuncture treatment.MethodsSix databases including China National Knowledge Infrastructure, Wanfang Data, Chongqing VIP, PubMed, Cochrance Library, and Embase were searched by computer for the randomized controlled trials on acupuncture combined with rehabilitation training treatment of post-stroke dysphagia, which were published from January 1st, 2010 to December 31st, 2018. After literature including, excluding, and screening, RevMan 5.3 software was used to conduct a meta-analysis.ResultsA total of 22 studies were included, including 1 987 patients. All the included studies took simple rehabilitation training or rehabilitation training combined with sham acupuncture as the control. Meta-analysis of efficiency and outcome measures for relevant studies showed that: compared with simple rehabilitation training, the effectiveness of acupuncture combined with rehabilitation training on post-stroke dysphagia was higher [17 studies included; odds ratio=3.66, 95% confidence interval (CI) (2.66, 5.05), P<0.000 01], the video fluoroscopy swallowing study score of acupuncture combined with rehabilitation training after treatment was higher [8 studies included; mean difference (MD)=2.31, 95%CI (1.75, 2.87), P<0.000 01], and the StandardizedSwallowing Assessment score of acupuncture combined with rehabilitation training after treatment was lower [6 studies included; MD=−3.20, 95%CI (−3.78, −2.61), P<0.000 01]; at the same time the Watian Drinking Water Test score of acupuncture combined with rehabilitation training after treatment was lower [6 studies included; MD=−0.65, 95%CI (−0.91, −0.39), P<0.000 01].ConclusionsAcupuncture combined with rehabilitation training is effective in dysphagia after stroke, and the combined effect is better than simple rehabilitation training. However, due to the limitations of quality of included literature and sample size, the above results and conclusions still require high quality and large sample studies to testify.