Objective To assess the clinical efficacy and safety of Shuxuetong in the treatment of cerebral infarction. Methods Randomized controlled trials (RCTs) were identified from MEDLINE (1996 to Feb. 2006), EMBASE (1984 to Dec. 2005), Cochrane Controlled Trials Register (Issue 4, 2005), Chinese Cochrane Centre Database, CBMdisc (1978 to Dec. 2005). We handsearched the related published and unpublished data and their references. All trials about Shuxuetong injection for cerebral infarction were included. Data were extracted and evaluated by two reviewers independently with designed extraction form. RevMan 4.2.8 software was used for data analysis. Results Eleven RCTs involving 1 122 patients were included. The results of meta-analysis were listed as follows: ① Total effective rate: Compared with Danshen, three studies showed that Shuxuetong were more effective with OR 4.01, 95%CI 2.00 to 8.04; ② Adverse effect: The number of adverse drug reaction was small and the symptoms were moderate; ③ Neurologic impairment score: Compared with safflower, one study showed that Shuxuetong had better improvement with WMD -2.60, 95%CI -3.23 to -1.97. Conclusions Shuxuetong may increase the total effective rate of cerebral infarction. More high quality trials are required.