OBJECTIVE: To systematically evaluate the clinical efficacy of acupuncture for dysarthria, and to explore the rules of acupoints selection for dysarthria. METHODS: The clinical randomized control trial literature regarding acupuncture for dysarthria published before January of 2018 were searched in databases, including CNKI, Wanfang, VIP, CBM, PubMed, Ebsco, Science Direct and Cochrane Library. The information of included studies was extract and the quality was assessed. The Meta analysis was performed by using RevMan 5.3 software. The frequency of acupoints was calculated by using Excel software to analyzed the rules of acupoints selection. RESULTS: Totally 21 papers were included, involving 1651 patients. The pooled effects of clinical efficacy: heterogeneity test P =0.74, I 2=0%, OR =6.36, 95% CI: 4.55, 8.88, Z =10.84 (P<0.01), indicating the efficacy in the treatment group was significantly higher than that in the control group. The pooled effects of the symptom score in Frenchay scale: heterogeneity test P =0.56, I 2=0%, WMD =3.20, 95% CI: 1.38, 5.02, Z =3.45 (P<0.01), indicating the efficacy in the treatment group was significantly higher than that in the control group. The acupoints with frequency of more than 5 times were Fengchi (GB 20), Yuye (EX-HN 13), Jinjin (EX-HN 12), Lianquan (CV 23), Baihui (GV 20), tongue-three needles and Yamen (GV15). The meridians with frequency of more than 5 times were the extra channels, governor vessel, gallbladder channel, conception vessel and stomach channel. CONCLUSION: The clinical efficacy of acupuncture combined with speech training/regular treatment is significantly superior to that of control group (speech training, medication, regular treatment); acupuncture is safe and effective for dysarthria; the majority of selected acupoint is local acupoints around tongue, throat and neck, as well as extra points and empirical points. However, high-quality randomized controlled trials with large sample sizes are still needed to provide further evidence.