Exposure–Response Relationship of Tranexamic Acid in Cardiac Surgery: A Model-based Meta-analysis

Background

It is unclear whether high-dose regimens of tranexamic acid in cardiac surgery (total dose, 80 to 100 mg/kg) confer a clinical advantage over low-dose regimens (total dose, approximately 20 mg/kg), particularly as tranexamic acid–associated seizure may be dose-related. The authors’ aim was to characterize the exposure–response relationship of this drug.

Methods

Databases were searched for randomized controlled trials of intravenous tranexamic acid in adult patients undergoing cardiopulmonary bypass surgery. Observational studies were added for seizure assessment. Tranexamic acid concentrations were predicted in each arm of each study using a population pharmacokinetic model. The exposure–response relationship was evaluated by performing a model-based meta-analysis using nonlinear mixed-effect models.

Results

Sixty-four randomized controlled trials and 18 observational studies (49,817 patients) were included. Seventy-three different regimens of tranexamic acid were identified, with the total dose administered ranging from 5.5 mg/kg to 20 g. The maximum effect of tranexamic acid for postoperative blood loss reduction was 40% (95% credible interval, 34 to 47%), and the EC₅₀ was 5.6 mg/l (95% credible interval, 0.7 to 11 mg/l). Exposure values with low-dose regimens approached the 80% effective concentration, whereas with high-dose regimens, they exceeded the 90% effective concentration. The predicted cumulative blood loss up to 48 h postsurgery differed by 58 ml between the two regimens, and the absolute difference in erythrocyte transfusion rate was 2%. Compared to no tranexamic acid, low-dose and high-dose regimens increased the risk of seizure by 1.2-fold and 2-fold, respectively. However, the absolute risk increase was only clinically meaningful in the context of prolonged open-chamber surgery.

Conclusions

In cardiopulmonary bypass surgery, low-dose tranexamic acid seems to be an appropriate regimen for reducing bleeding outcomes. This meta-analysis has to be interpreted with caution because the results are observational and dependent on the lack of bias of the predicted tranexamic acid exposures and the quality of the included studies.