Tranexamic Acid for Gastrointestinal Bleeding: Time to HALT-IT?

Although tranexamic acid reduces the risks of surgical bleeding and bleeding-related death in patients with trauma, it does not reduce risk of death from gastrointestinal bleeding (GI), according to results from the HALT-IT study published in Lancet. It also appeared to increase the risk of seizures and venous thromboembolic events, compared with placebo.

“Our results caution against a uniform approach to the management of patients with major hemorrhage and highlight the need for randomized trials targeted at specific pathophysiological processes,” the study authors, led by Ian Roberts, MD, of the London School of Hygiene & Tropical Medicine,” concluded. They added that these findings could have regulatory applications, as GI bleeding “is a licensed indication for tranexamic acid.”

The HALT-IT trial was designed to evaluate the effects of tranexamic acid on death and thromboembolic events in acute GI bleeding, after an earlier meta-analysis of randomized trials suggested that the drug substantially reduced mortality risk, the researchers explained.

HALT-IT included 12,009 patients with high-risk upper or lower GI bleeding. Participants were randomized to receive either tranexamic acid (n=5,994) or matching placebo (n=6,015). Tranexamic acid was administered intravenously via a loading dose of 1 g added to a 100-mL infusion bag of sodium chloride 0.9%, followed by a maintenance dose of tranexamic acid 3 g added to 1 L of any isotonic solution. In the placebo group, patients received a sodium chloride 0.9% solution. All patients, caregivers, and outcome assessors were masked to randomization.

By day 28 after randomization, 1,112 patients died, with a median time to death of 55 hours following randomization. The rate of bleeding-related death within 5 days of treatment allocation was similar between the tranexamic acid and placebo groups: 3.7% versus 3.8%, respectively (risk ratio [RR] = 0.99). The all-cause mortality rate within 28 days of randomization also was slightly higher with tranexamic acid (9.5% vs. 9.2%; RR=1.03), but rates of arterial thromboembolic events, including myocardial infarction or stroke, were slightly lower (0.7% vs. 0.8%; RR=0.92).

However, the rates of venous thromboembolic events, such as deep vein thrombosis (DVT) or pulmonary embolism, nearly doubled in patients who received tranexamic acid compared with those who received placebo (0.8% vs. 0.4%; RR=1.85). Treatment with tranexamic acid also was associated with a higher risk of death related to a thromboembolic event, the authors added (TABLE).

Looking at complications associated with treatment, the investigators also reported that more patients in the tranexamic acid group experienced seizures (0.6% [n=38] vs. 0.4% [n=22]; RR=1.73).

“We found no evidence that tranexamic acid decreases the risk of death in patients with gastrointestinal bleeding,” the authors concluded. However, “although this trial can rule out the large mortality reduction suggested by the meta-analysis of previous small trials, it cannot rule out more modest treatment effects.” Future studies, they suggested, should focus on individual patient data meta-analyses to assess whether the effectiveness and safety of tranexamic acid varies by the site and cause of bleeding.

The investigators noted the possibility of misclassification of cause of death in this study population as a potential limitation. There also were varying degrees of certainty regarding the sites and causes of bleeding among patients, which could have affected the findings.

The authors report no relevant conflicts of interest.


Roberts I, Shakur-Still H, Afolabi A, et al; HALT-IT Trial Collaborators. Effects of a high-dose 24-h infusion of tranexamic acid on death and thromboembolic events in patients with acute gastrointestinal bleeding (HALT-IT): an international randomised, double-blind, placebo-controlled trial. Lancet. 2020;395:1927-1936.