Invited Commentary 

January 15, 2025

Tranexamic Acid in General Surgery—Who Benefits the Most?

Marjorie Liggett, Hasan Alam

JAMA Surg. Published online January 15, 2025. doi:10.1001/jamasurg.2024.6060

Tranexamic acid (TXA) is not a new drug. First discovered by Utako Okamoto in 1962,¹ it regained attention decades later after the CRASH-2 (Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage) trial,² where it was tested in trauma patients. It works primarily by stabilizing the fibrin matrix in the clotted blood (decreased lysis) through competitive inhibition of plasminogen’s lysine receptor. While the only US Food and Drug Administration–approved usage for TXA is for heavy menstrual bleeding and short-term prevention in patients with hemophilia, it has recently been tested for a variety of clinical problems.

In this substudy of the POISE-3 (Perioperative Ischemic Evaluation-3) trial,³ Park et al⁴ explored the utility of prophylactic TXA to prevent major bleeding complications in general surgery. They analyzed 9535 patients from the POISE-3 cohort, who were classified as high risk for bleeding and cardiovascular complications.⁵ The authors compared TXA with placebo control in patients undergoing general surgery and nongeneral, noncardiac surgery procedures. Primary effect outcome was 30-day composite life-threatening bleeding, major bleeding, and critical organ bleeding. The primary safety outcome was a 30-day composite of myocardial injury after noncardiac surgery, nonhemorrhagic stroke, peripheral arterial thrombosis, and symptomatic proximal venous thromboembolism. Patients randomized to the TXA group received 1 g of TXA at the start of and the end of surgery.

While this was a subgroup analysis of the POISE-3 trial, it is the largest population of general surgery patients where safety and efficacy of TXA have been investigated. Importantly, Park and colleagues were able to demonstrate noninferiority to placebo for the composite safety outcome, which was not achieved in the original POISE-3 trial.³ Their results demonstrated reduced composite bleeding and major bleeding in the TXA group compared with placebo among patients undergoing general surgery, particularly those undergoing hepatopancreaticobiliary and colorectal procedures.⁴ But the question continues, who benefits the most from prophylactic TXA?

As was acknowledged by the authors,⁴ bleeding risk varies tremendously across general surgery procedures. The current study, unfortunately, was not powered to explore this further. Additionally, how do we define major bleeding? As we discovered in the CRASH-2 trial, not all bleeding patients respond the same to TXA.² In this study,⁴ there was no benefit of TXA in groups receiving 2 or more units of transfusion. Although there was a significant reduction in major bleeding using International Society of Thrombosis and Hemostasias criteria and those of the authors.^3,5,6And lastly, how do we stratify patients into high risk of bleeding? The criteria used in this study was generated from the VISION (Vascular Events in Noncardiac Surgery Patients Cohort Evaluation) cohort,⁵ with enrollment up to 2011 and in which transfusion practices differed from those practiced today. Additionally, transfusion triggers vary widely across different sites and even among surgeons at the same hospital. The dosage and timing of TXA administration are also somewhat dated, and the lack of thromboelastographic data makes it difficult to determine whether the observed benefits were due to decreased thrombolysis or a completely different mechanism. In conclusion, while this study⁴ shows a reduction in bleeding with prophylactic TXA administration for specific patients undergoing general surgical operations, more information is needed before we can adopt this approach broadly for all types of procedures.