Tranexamic acid for neurological outcomes and mortality in adults with hemorrhagic acute brain injury: an updated systematic review and meta-analysis of randomized controlled trials.

BACKGROUND: Hemorrhagic acute brain injury-encompassing traumatic brain injury, spontaneous intracerebral hemorrhage, and subarachnoid hemorrhage-carries substantial morbidity and mortality. Active hemorrhage and hematoma progression constitute shared pathological mechanisms across these conditions. As a lysine analog anti-fibrinolytic, tranexamic acid (TXA) has demonstrated mortality reduction in extracranial bleeding; however, its efficacy and safety in intracranial hemorrhagic lesions remain inconclusive. Prior systematic reviews predominantly targeted single disease categories, inadequately assessed core outcomes such as all-cause mortality and neurological functional recovery, and seldom applied the Risk of Bias 2 tool or GRADE framework for methodological appraisal. Therefore, an updated meta-analysis is warranted to re-evaluate the overall clinical value of TXA in hemorrhagic acute brain injury. OBJECTIVE: To evaluate the effect of TXA on neurological outcomes and all-cause mortality in adults with acute brain injury. METHODS: This meta-analysis was registered with PROSPERO (CRD420261400869). In accordance with the PRISMA 2020 statement, we systematically searched PubMed, EMBASE, Scopus, Web of Science, and the Cochrane Library from inception to April 2026. Eligible studies were randomized controlled trials comparing TXA with placebo in hospitalized adults with acute brain injury, including traumatic brain injury, spontaneous intracranial hemorrhage, and subarachnoid hemorrhage. Risk of bias was assessed using the RoB 2 tool, the certainty of evidence was evaluated using the GRADE framework, pooled effect estimates were generated using a random-effects model, heterogeneity was quantified using the I 2 statistic, and robustness was tested through leave-one-out sensitivity analysis. The primary outcomes were neurological outcomes and all-cause mortality; secondary outcomes were thromboembolic events and intracranial rebleeding. RESULTS: Seven high-quality randomized controlled trials comprising over 12,000 patients were included. Random-effects model analyses showed no statistically significant effect of TXA on short-term mortality (RR 0.93, 95% CI 0.86-1.00) or long-term mortality (RR 1.02, 95% CI 0.91-1.13). For neurological outcomes, TXA neither improved the proportion of patients with functional independence (RR 0.97, 95% CI 0.92-1.03) nor reduced the risk of functional dependence or death (RR 1.01, 95% CI 0.96-1.07). For rebleeding, the pooled estimate showed no statistically significant effect (RR 0.81, 95% CI 0.59-1.11), with substantial between-study heterogeneity (I 2 = 72%). The current evidence is inconclusive regarding a protective effect of TXA on rebleeding. The incidence of thromboembolic events did not differ significantly between groups (RR 1.17, 95% CI 0.90-1.52). Leave-one-out sensitivity analyses indicated that the primary outcomes were robust. CONCLUSION: TXA did not improve all-cause mortality or neurological outcomes in hemorrhagic acute brain injury, and the evidence for its effect on rebleeding remains inconclusive. SYSTEMATIC REVIEW REGISTRATION: PROSPERO, CRD420261400869.