The use of tranexamic acid (TXA) in patients undergoing oncology and other major noncardiac surgeries reduces the administration of red blood cell (RBC) transfusions across patient and surgical subgroups, new research shows.

This effect of perioperative TXA administration “has implications for both patients as well as the sustainability of the blood supply,” said first author Brett Houston, MD, PhD, of the University of Manitoba in Winnipeg, Manitoba, Canada, while presenting findings at the American Society of Hematology (ASH) 2025 Annual Meeting.

Perioperative bleeding is a leading reason for the utilization of RBC transfusions; however, donor supply and ever-increasing demand for transfusions are ongoing challenges, she said. The use of antifibrinolytic medication TXA is an inexpensive option to reduce the need for transfusions, with a long-established safety and efficacy record, particularly in the context of cardiac and orthopedic surgeries.

However, with concerns of the potential for thrombotic complications, particularly among patients with cancer, there is some apprehension about its use, Houston added.

Studies supporting the safety and efficacy of TXA in noncardiac surgery have included the POISE-3 trial, of nearly 10,000 patients at a risk for bleeding, showing significant reductions in the composite risk for bleeding (9.1% vs 11.7%) and reductions in RBC transfusion at 30 days (9.4% vs 12%).

Study Methods and Results

To build on that and other smaller studies, Houston and colleagues conducted a multicenter, registry-based, cluster-crossover trial, enrolling 8273 patients at 10 hospitals across Canada who were undergoing noncardiac surgeries and had a high risk for RBC transfusion.

The hospitals were randomized in 4-week intervals to implement hospital-wide policies of using intraoperative TXA or placebo. The majority of surgeries (5002; 60%) were for oncologic indications, with the most common surgical specialties including general surgery (n = 2742; 33.1%), gynecology (n = 1540; 18.6%), urology (n = 1434; 17.3%), vascular surgery (n = 582; 7.0%), and spine surgery (n = 461; 5.6%). The patients had a median age of 64 years, about 47% were male, and their median weight was 80 kg.

Overall, the proportion of patients transfused with RBCs during hospital admission was 7.4% (306 of 4156) in the TXA group compared with 9.8% (403 of 4117) in the placebo group for an absolute reduction of 2.4 percentage points (odds ratio [OR], 0.61; 95% CI, 0.54-0.89).

Importantly, TXA was not associated with an increased risk for venous thromboembolism (VTE) at 90 days, with a confidence interval (2.1% vs 2.1%; OR, 1.0; 95% CI, 0.67-1.49) that “narrowly exceeded the prespecified noninferiority margin of 1.47,” Houston noted.

She added that TXA was also not associated with an increase in VTE in prespecified subgroups, and notably, no increase in VTE was observed among patients undergoing cancer surgery.

Patients in the TXA group received fewer RBC units than those in the placebo group (mean, 0.25 units [± 1.43 units] vs 0.34 units [± 1.86 units]; P < .01).

Meanwhile, there were no differences between the groups in terms of in-hospital diagnoses of myocardial infarction (0.7% and 0.8% in the TXA and placebo groups, respectively), stroke (0.2% in each group), deep vein thrombosis (below 6% in each group), or pulmonary embolism (0.2% in each group).

The mean hospital length of stay was about 6.2 days in each group — ICU admission was 16.7% and 17.5%, hospital survival was 98.9% and 99.0%, and survival at 90 days was 97.8% and 97.6% in the TXA and placebo groups, respectively.

Ultimately, the findings overall show that “TXA did not increase venous thromboembolism, even among patients with cancer, which is a population at particularly high risk [for VTE],” Houston said.

She underscored the potential cost implications of the results, noting that approximately 100 million inpatient noncardiac surgeries are performed each year, and 1 g of TXA costs approximately $3.

“If broadly implemented, 1-2 million people could potentially avoid perioperative red blood cell transfusion exposure, and 5-10 million units of red cells could potentially be saved, ultimately resulting in significant cost savings.”

Asked by an audience member about the number that might be needed to treat for one patient to benefit, Houston did not have the data but acknowledged that the findings indicate “there are modest individual patient benefits (of use of TXA).”

Pros and Cons to Approach

Commenting to Medscape Medical News on the study, Charles S. Abrams, MD, who is chief of non-malignant hematology at the University of Pennsylvania School of Medicine and director of the Penn-CHOP Blood Center, both in Philadelphia, noted “there remains a substantial need for safe, inexpensive blood-sparing strategies in major noncardiac surgery, particularly in oncology, and vascular procedures where transfusion remains common and blood is a constrained resource.”

“Tranexamic acid is attractive because it is low cost and easy to administer,” he said.

The reductions observed in the study, though modest, suggest important benefits of TXA use, Abrams said.

“A 2.4-point reduction would mean that about 1 in 40-45 patients avoids any red blood cell transfusion because of tranexamic acid use, which for a large institution performing thousands of such operations annually is not trivial. The benefit for an individual patient is probably modest since most patients never receive transfusions either way.”

Overall, the study “strengthens the case that tranexamic acid can be implemented at an institution as a routine perioperative blood-sparing intervention, not just as a selectively applied drug,” Abrams said. “However, uncommon thrombotic or neurologic harms cannot be completely excluded.”

On a side note, Houston added that the study design itself provided substantial cost savings by utilizing logic and simplicity.

“We integrated this trial into routine care, and because of this we were able to conduct the trial at a fraction of typical costs,” she explained. “Our trial cost approximately CAD $1.8 million, which equates to approximately $1.3 million.”

Comparatively, “a typical trial of this size would usually cost somewhere in the $10 million to $20 million range,” she said. The use of novel methods such as the study’s registry-based data ascertainment and limited manual data capture “allowed for the trial to be conducted efficiently at a fraction of the cost.”

Houston’s disclosures included consulting for AbbVie. Abrams had no disclosures to report.