EDITORIAL

TRACTION for Greater Surgical Use of Tranexamic Acid

Michael F. Murphy, Ian Roberts

N Engl J Med Published June 10, 2026

DOI: 10.1056/NEJMe2603105

Every year, many millions of surgical patients worldwide are unnecessarily exposed to a higher-than-necessary risk of blood transfusion because they do not receive a single dose of tranexamic acid just before the surgery. This failing is due not to a lack of high-quality scientific evidence on the overall safety, efficacy, or cost-effectiveness of tranexamic acid but to poor implementation into clinical practice and lingering concerns regarding whether previous studies had ruled out the risk of thrombosis with sufficient confidence.

Perioperative bleeding is common and an important cause of surgical deaths. Approximately 1 in 65 patients undergoing major noncardiac surgery have perioperative bleeding, with the highest risk in transplantation and noncardiac vascular surgeries.¹ In a retrospective study, 1 in 17 patients with documented perioperative bleeding died during hospitalization and 1 in every 3 survivors were readmitted within 6 months.¹ Transfusion can be lifesaving after major blood loss, but donor blood is a scarce resource and there are well-known transfusion-related risks. For these reasons, attention is increasingly focused on strategies to reduce perioperative bleeding.

Tranexamic acid was invented in the early 1960s.² Since then, the evidence for the safety and effectiveness of a single preoperative dose in reducing surgical bleeding and the need for transfusion has steadily accumulated. A 2012 systematic review including 129 randomized, controlled trials involving 10,488 patients concluded that tranexamic acid reduced the risk of perioperative blood transfusion by up to one third, with fewer surgical deaths.³ The substantial reduction in surgical bleeding with tranexamic acid was confirmed by the POISE-3 (Perioperative Ischemic Evaluation–3) trial.⁴ The trialists randomly assigned 9535 adults undergoing noncardiac surgery who were at risk for bleeding and cardiovascular complications to receive tranexamic acid or placebo. Tranexamic acid resulted in an approximately 25% lower incidence of major bleeding and of blood transfusion than placebo. The effects did not vary according to type of surgery. However, questions remain about the safety profile of tranexamic acid in surgery. Systematic reviews and meta-analyses of randomized trials of tranexamic acid including close to 100,000 patients have shown no evidence of any increased risk of thrombosis.^5,6 However, for statistical reasons, it is difficult to rule out a small increase or decrease in the risk of thrombosis.

This concern about thrombosis is one of the main barriers to the use of tranexamic acid and is the reason that the data from the TRACTION (Tranexamic Acid to Reduce Transfusion in Major Noncardiac Surgery) trial now reported by Houston et al. in the Journal⁷ are so welcome. More than 8000 patients undergoing noncardiac surgeries who were at high risk for transfusion were enrolled in a double-blind, placebo-controlled, cluster-crossover design trial in 10 Canadian hospitals that were randomly assigned at 4-week intervals to a hospital-wide policy. Excluded were patients with active thromboembolic disease; those undergoing cardiac, hip, or knee surgery (in which the use of tranexamic acid is routine practice); and pregnant persons. Venous thromboembolism within 90 days after surgery occurred in 2.1% of the patients in both the tranexamic acid group and the placebo group, which met criteria for noninferiority. And the risk of red-cell transfusion was approximately 25% lower with tranexamic acid than with placebo. Of note, the trial included a high proportion of patients undergoing oncologic surgery, a group often excluded or underrepresented in tranexamic acid trials, but only a small proportion of patients undergoing vascular surgery, in which use of tranexamic acid is low. The TRACTION trial was well conducted and is an exemplar for conducting large, multicenter, randomized trials with the use of institutional policy randomization and integration of electronic health data.

This further evidence of the safety and effectiveness of tranexamic acid in surgery should encourage its more widespread use. It may also stimulate consideration of following the lead of the United Kingdom in extending the use of tranexamic acid to surgeries even with a low likelihood of major blood loss or red-cell transfusion.⁸ However, implementation of evidence into practice, as in other areas of clinical medicine, can be a problem. In the United Kingdom, the National Institute for Health and Care Excellence made the use of tranexamic acid in adults who were undergoing surgery and were expected to have “moderate (>500 ml) blood loss” a quality standard in 2016, but nearly 10 years later approximately one third of eligible surgical patients did not receive this treatment.⁹ There is also a wide variation in tranexamic acid use among surgical specialties, being most commonly used in cardiac, trauma, and orthopedic surgery and least in genitourinary, colorectal, and noncardiac vascular surgery.⁹Strategies based on an understanding of clinical behavior are needed to increase its use and improve patient outcomes. A recently proposed implementation strategy for tranexamic acid in surgery included opinion leadership, training, clinical decision support, and performance feedback.¹⁰ Effective strategies to increase the use of tranexamic acid in surgery in low- and middle-income countries, where blood stocks are lower and transfusion risks are higher, are urgently needed. The strong evidence that tranexamic acid improves surgical safety supports its inclusion in the World Health Organization Safe Surgical Checklist.