[Objective] To determine the optimal hemostatic drug in corrective surgery for severe spinal deformity, data were extracted from randomized controlled trials to performed a meta-analysis. [Methods] A search of tranexamic acid (TXA) infusion in spinal corrective surgery was conducted in PubMed, EMBASE, Web of Science,CNKI and Wanfang database.The results were limited to August 2019 and all the statistics were analyzed by Review Manager 5.3 software.[Results] A total of 306 relevant literatures were searched. According to the inclusion and exclusion criteria, 9 randomized controlled trials including 468 patients were eventually enrolled in this meta-analysis, which 292 patients were included in the TXA group with an age ranging from 11 to 65 years, while 289 patients were included in the placebo group ranging from 11 to 68 years old. Compared to the placebo group, the volume of blood loss in the TXA group decreased in 311ml (-311.47 [-385.72, -237.21]) for patients with severe spinal deformity. As for total blood loss, there was a significant decrease in TXA group (-159.41ml [-188.12, -130.70]).TXA significantly reduced the transfusion need (-186.44 ml [-294.59, -78.30]), which in TXA group the requirement of transfusion was lower in 186ml. However, there was no difference in the operation time between the TXA groups and placebo group (-1.46 [-15.92, 13.00]). The administration of TXA did not increase the rate of thrombosis-related complications (1.16 [0.44, 3.08]). The incidence of thrombotic events in the TXA group and placebo group were 5.1% and 4.2%, respectively (P=0.76). [Conclusion] In corrective surgery for severe spinal deformity, tranexamic acid can effectively reduce the the volume of perioperative blood loss and transfusion without increasing the risk of thrombotic events. There is a need for the larger sample size and multicenter trails in the future to provide higher levels of evidence.