The optimal dose and efficacy of tranexamic acid (TXA) and epsilon‐aminocaproic acid (EACA) in total knee arthroplasty (TKA) were under controversial, and we aimed to make comparisons between different doses of TXA and EACA in intravenous (IV) or intra‐articular (IA) applications in patients undergoing TKA.
This network meta‐analysis was guided by the Priority Reporting Initiative for Systematic Assessment and Meta‐Analysis (PRISMA). According to the administrations of antifibrinolytic agents, patients in eligible studies were divided into three subgroups: (i) IA applications of TXA and EACA; (ii) IV applications (g) of TXA and EACA; (iii) IV applications (mg/kg) of TXA and EACA. Total blood loss (TBL), hemoglobin (HB) drops and transfusion rates were the primary outcomes, while drainage volume, pulmonary embolism (PE) or deep vein thrombosis (DVT) risk were the secondary outcomes. A multivariate Bayesian random‐effects model was adopted in the network analysis.
A total of 38 eligible trials with different regimens were assessed. Overall inconsistency and heterogeneity were acceptable. Taking all primary outcomes into account, 1.0–3.0 g TXA were most effective in IA applications, 1–6 g TXA and 10–14 g EACA were most effective in IV applications (g), while 30 mg/kg TXA and 150 mg/kg EACA were most effective in IV applications (mg/kg). None of the regimens showed increasing risk for pulmonary embolism (PE) or deep vein thrombosis (DVT) compared with placebo.