Lessons from aprotinin: is the routine use and inconsistent dosing of tranexamic acid prudent? Meta-analysis of randomised and large matched observational studies.

In view of the safety concerns that led to the withdrawal of aprotinin, should antifibrinolytics be used indiscriminately in cardiac surgery? This meta-analysis examines the efficacy and safety profile of tranexamic acid, and in comparison to aprotinin. We identified randomised trials and large observational studies investigating the use tranexamic acid from January 1995 to January 2009 using Pubmed/Cochrane search engine and included them in a two-tier meta-analysis. There were 25 randomised trials and four matched studies with a total of 5411 and 5977 patients, respectively, reporting tranexamic acid use in varying dosages. Tranexamic acid is administered intravenously either as single dose, infusion or both, sometimes added to pump prime or applied topically. Total intravenous dose of tranexamic acid varies from 1g to 20 g, administered over a period of 20 min to 12h. Compared with placebo, tranexamic acid is associated with a lower mean difference in blood loss (random effect -298 ml, 95% confidence [CI] -367 to -229, p<0.001) and decease in rates of re-operation for bleeding by 48%, transfusion of packed red cell by 47% and use of haemostatic blood products by 67%. A non-significant tendency for postoperative neurological events but a decrease in operative mortality was observed in patients treated with tranexamic acid compared with non-treatment group. Compared to aprotinin, tranexamic acid has less effective blood-conserving effect and mortality risk. Given the potential to increase neurological complications, the current trend towards indiscriminate use of tranexamic acid for all cardiac patients needs to be re-evaluated. Further studies are needed to clarify the neurological risk, appropriate indications and dosing of tranexamic acid. Copyright 2010 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.