Blood transfusions during major noncardiac surgeries reduced with tranexamic acid

The use of the antifibrinolytic agent tranexamic acid during major noncardiac surgeries results in fewer red blood cell (RBC) transfusions without an increase in venous thromboembolism, as shown in the phase IV TRACTION trial.

In a cohort of more than 8,000 patients, the primary effectiveness outcome of the proportion of patients transfused RBC was 7.4 percent with tranexamic acid vs 9.8 percent with placebo (odds ratio [OR], 0.61, 95 percent confidence interval [CI], 0.54–0.89). [ASH 2025, abstract LBA-5]

The primary safety outcome of venous thromboembolism within 90 days occurred in 2.1 percent of patients each in the tranexamic acid and placebo groups (OR, 1, 95 percent CI, 0.67–1.49).

Results for both outcomes were consistent across subgroups defined by age, surgical specialty (general, orthopaedic, spine, neurosurgery, thoracic, vascular, gynaecologic, and urologic), surgical urgency (elective and urgent/emergent), surgical indication (oncologic and nononcologic), and risk of RBC transfusion (<10 percent and ≥10 percent).

“Most notably, there was no increase in venous thromboembolism among patients undergoing cancer surgery,” which is a population at particularly high risk, according to lead study author Dr Brett Houston from the University of Manitoba in Winnipeg, Canada, who presented the trial at a late-breaking abstract session at ASH 2025.

Houston pointed out that the findings have important implications for patients and the sustainability of the blood supply, which is increasingly constrained due to rising demand and declining donor pools.

“Each year, approximately 100 million inpatient noncardiac surgeries are performed. If [tranexamic acid is] broadly implemented, 1–2 million people could potentially avoid perioperative RBC transfusion exposure, and 5–10 million units of RBCs could potentially be saved,” she said.

Meanwhile, tranexamic acid is cheap, costing approximately USD 3 for each gram, and has an established safety profile. Several studies provide evidence that the drug can reduce RBC transfusion in cardiac and hip and knee arthroplasty, as well as among trauma patients and women with postpartum haemorrhage. [Lancet 2010;376:23-32; N Engl J Med 2017;376:136-148; Lancet 2017;389:2105-2116]

“However, adoption of tranexamic has been variable, partially due to unresolved concerns of potential thrombotic complications, particularly amongst patients with cancer… We hope [the TRACTION] data will also set practitioners’ minds at rest that giving the drug is safe,” she said.

The TRACTION trial

TRACTION was a randomized placebo-controlled cluster crossover trial that 10 Canadian sites. Each trial site had to be performing at least 100 noncardiac surgeries each month, and its anaesthesia, surgery, and hospital leadership had to agree to manage patients according to the trial protocol.

Participating sites were randomly assigned to provide either tranexamic acid or matching placebo, with the treatment assignment switched every 4 weeks. Tranexamic acid was intravenously administered as a 1-g bolus at the start of surgery, followed by 1 g prior to skin closure at the discretion of the anaesthesiologist. Patients, healthcare providers, research staff investigators, and statisticians were all blinded.

“We evaluated the impact of a hospital policy of tranexamic administration rather than individual patient administration … [and] studied tranexamic acid effectiveness in the context of high-risk surgeries rather than high-risk patients,” Houston noted.

A total of 8,421 patients were included, with 4,156 receiving tranexamic acid and 4,117 receiving placebo. Baseline characteristics were similar in the two treatment groups. Median patient age was 64 years, and 46.2 percent were male. General surgeries (32.6 percent) were the most common procedures, followed by gynaecologic (18.3 percent) and urologic (17 percent) surgeries. Roughly 90 percent were elective surgeries, and 59 percent had an oncologic indication.

Patients in the tranexamic acid group received a mean of 0.25 RBC units, while those in the placebo group received 0.34 RBC units. There were no between-group differences in in-hospital diagnoses of myocardial infarction (29 vs 34 patients), stroke (9 patients in both), deep vein thrombosis (<6 patients in both), or pulmonary embolism (8 vs 6 patients).

Other outcomes were also similar in the tranexamic acid and placebo groups, including hospital length of stay (mean, 6.20 vs 6.26 days), ICU admission (16.7 percent vs 17.5 percent), hospital survival (98.9 percent vs 99 percent), survival at 90 days (97.8 percent vs 97.6 percent), and days alive and out of hospital to day 30 (median, 26 vs 26 days).

“Prespecified serious adverse events were rare. There was one instance of seizure in the tranexamic acid group and one instance of anaphylaxis in the placebo group,” Houston said.

Low trial cost

What’s unique about TRACTION is that it was integrated into routine clinical care. Houston stressed that this allowed them to conduct the trial at a fraction of typical costs.

“Our trial cost approximately USD 1.3 million. A typical trial of this size would usually cost somewhere in the 10–20-million range. This ultimately equates to a per-patient cost of about 150 dollars,” she said.

During the question-and-answer session, Dr Martin Dreyling from the Ludwig Maximilian University of Munich in Munich, Germany, congratulated Houston and her team for doing an “excellent job.”

“[TRACTION] is highly clinically relevant. It improves patient care, it’s cheap, and it really tells you that a lot of things we are doing in standard clinical trials are throwing money out of the window,” Dreyling commented.