top of page

Tranexamic Acid in Patients Undergoing Noncardiac Surgery

N Engl J Med 2022;386:1986-97.

DOI: 10.1056/NEJMoa2201171

What's the question?

  • In patients aged 45 years or older undergoing noncardiac surgery who are at increased risk for bleeding and cardiovascular events, does tranexamic acid (TXA) reduce serious perioperative bleeding compared with placebo, and is it noninferior to placebo with respect to major cardiovascular complications within 30 days?

Background Information

  • Perioperative bleeding is a common and serious complication of noncardiac surgery, associated with increased morbidity and mortality.

  • TXA is an antifibrinolytic drug that inhibits fibrinolysis; large trials had demonstrated efficacy in caesarean section, cardiac surgery, trauma, and postpartum haemorrhage, but prior noncardiac surgery trials were small and insufficient to establish cardiovascular safety.

  • An annual global shortage of 30 million blood-product units exists, with surgical bleeding accounting for up to 40% of all transfusions — highlighting the major public health importance of reducing perioperative blood loss.

  • No large randomised trial had previously established whether TXA increases the risk of thrombotic events in noncardiac surgery.

The study

  • International, randomised, double-blind, placebo-controlled trial (POISE-3) with a partial factorial design; conducted at 114 hospitals in 22 countries across six continents, June 2018 to July 2021.

  • 9,535 patients randomised 1:1 to TXA (1g intravenous bolus) or placebo at the start and end of surgery; 96.3% in both groups received both doses; 30-day follow-up was complete in 99.9% of patients.

  • Eligible patients: aged ³45 years, undergoing inpatient noncardiac surgery, at risk for bleeding and cardiovascular complications. Exclusions included cardiac surgery, intracranial neurosurgery, creatinine clearance <30 ml/min, and planned systemic TXA.

  • Primary efficacy outcome (composite bleeding — life-threatening bleeding, major bleeding, or bleeding into a critical organ at 30 days): TXA 9.1% vs placebo 11.7% (hazard ratio [HR] 0.76, 95% CI 0.67–0.87; absolute difference –2.6 percentage points; p<0.001 for superiority).

  • Primary safety outcome (composite cardiovascular — myocardial injury after noncardiac surgery, nonhemorrhagic stroke, peripheral arterial thrombosis, or symptomatic proximal venous thromboembolism at 30 days): TXA 14.2% vs placebo 13.9% (HR 1.02, 95% CI 0.92–1.14; upper boundary of one-sided 97.5% CI 1.14; one-sided p=0.04). Noninferiority was not established (required upper boundary <1.125).

  • TXA also reduced transfusion of at least 1 unit of packed red cells (9.4% vs 12.0%; odds ratio 0.77) and major bleeding by ISTH criteria (6.6% vs 8.7%; HR 0.75). Results were consistent across orthopedic and nonorthopedic surgery subgroups.

  • Seizures occurred in 0.2% (TXA) vs <0.1% (placebo) (HR 3.35; 95% CI 0.92–12.20) — not statistically significant but a numerically higher rate. Death from any cause was similar (1.1% vs 1.2%).

  • Recruitment was stopped early due to a financial deficit from the COVID-19 pandemic; more than 95% of the planned sample was enrolled and the decision was made without knowledge of results.

Strengths and Weaknesses

  • Strengths: Large, international, double-blind RCT across 114 hospitals in 22 countries; near-complete 30-day follow-up (99.9%); high drug adherence (96.3% received both doses); statistical analysis plan finalised before unblinding; results consistent across all prespecified subgroups (surgery type, haemoglobin level, renal function, NT-proBNP level).

  • Weaknesses: Early recruitment cessation due to COVID-19 financial constraints, reducing power marginally; noninferiority for the cardiovascular outcome was not established, leaving uncertainty about thrombotic risk; limited clinical ability to identify all perioperative thrombotic complications; nausea and vomiting data were not collected; the noninferiority margin of 1.125 (a relative increase of 12.5%) may be considered wide by some.

PHRACC's Conclusions

  • TXA significantly reduced the composite incidence of serious perioperative bleeding (HR 0.76; absolute reduction 2.6 percentage points; p<0.001), consistently across all types of noncardiac surgery.

  • Noninferiority to placebo for the composite cardiovascular outcome was not established (upper 97.5% CI boundary 1.14, exceeding the pre-specified threshold of 1.125), though the absolute difference was small (0.3 percentage points).

  • Clinicians must weigh a clear, meaningful reduction in bleeding risk against a small and uncertain potential increase in cardiovascular risk on an individual patient basis.

  • Given 300 million surgeries annually worldwide and a global blood product shortage, wider routine use of TXA in high-risk noncardiac surgery has the potential for substantial public health benefit.

This page is part of the PHRACC AI-assisted education series. AI has been used to help synthesise information from carefully hand-picked peer-reviewed recent publications which relate to the delivery of healthcare in both the United Kingdom and Australia. Content has been reviewed and edited by the author to minimise the risk of error and incorrect information. PHRACC.com and the author(s) can not be held responsible or liable in any way for any errors in this article and resulting sequelae. The responsibility rests solely with the reader to confirm the accuracy of all information contained here with independent reputable sources before using it for any purpose.

Last updated 26/6/26 by JH

bottom of page