Tranexamic Acid in Gastrointestinal Bleeding: Evidence‑Based Recommendations

Overall Recommendation

Routine administration of tranexamic acid (TXA) for acute gastrointestinal (GI) bleeding is not recommended because high‑quality evidence shows no mortality benefit and an increased risk of thromboembolic events and seizures. Strong recommendation, high‑quality evidence. 1, 2, 3

The British Society of Gastroenterology (2019) advises that TXA use in acute lower GI bleeding should be limited to clinical‑trial settings only. Strong recommendation. 1, 4, 5, 6, 7

The European Association for the Study of the Liver (EASL, 2022) issues a strong recommendation against TXA in patients with cirrhosis and active variceal bleeding. Strong recommendation. 2, 3
In the HALT‑IT trial (12,009 participants, ~50 % with suspected variceal bleeding), TXA did not reduce death from bleeding within 5 days. High‑quality randomized trial. 2, 3
TXA was associated with an almost two‑fold increase in venous thromboembolic events, particularly among patients with liver disease or suspected variceal bleeding. High‑quality randomized trial. 2, 3
The frequent hypofibrinolytic state in critically ill cirrhotic patients may explain the lack of efficacy of antifibrinolytics. Pathophysiologic insight. 2

Older guidelines (2002) reported that TXA reduced the need for surgical intervention but did not lower ulcer re‑bleeding rates; further studies were deemed necessary before routine adoption. Conditional recommendation. 8
The 2019 International Consensus Group guidelines do not recommend TXA for non‑variceal upper GI bleeding. Strong recommendation. 9

Meta‑analysis of recent high‑quality studies found that TXA increases seizure risk (risk ratio 1.73; 95 % CI 1.03–2.93). Moderate‑quality evidence. 7

The high‑dose extended regimen used in HALT‑IT (loading dose 1 g IV over 10 min) was linked to the observed harms. Trial protocol. 10
Low‑dose regimens (various IV or enteral protocols) employed in older trials suggested possible benefit in upper GI bleeding, but evidence is limited and not endorsed by current guidelines. Low‑quality evidence. 11

Immediate hemodynamic resuscitation remains a cornerstone of acute GI bleeding care. Guideline recommendation. 1
Early endoscopy with therapeutic intervention is recommended for all acute GI bleeds. Guideline recommendation. 1, 9
High‑dose proton‑pump inhibitor therapy is advised for non‑variceal upper GI bleeding. Guideline recommendation. 9
Vasoactive pharmacotherapy is recommended for variceal bleeding in cirrhosis. Guideline recommendation. 2
Prompt reversal of anticoagulation, when appropriate, should be undertaken. Guideline recommendation. 1, 7
A restrictive transfusion strategy is advised in the management of GI bleeding. Guideline recommendation. 2

Older pooled analyses of upper GI bleeding trials suggested a 40 % mortality reduction with TXA, but this effect disappeared when analyses were restricted to low‑risk‑of‑bias studies, indicating that the apparent benefit was likely due to bias. Low‑quality evidence. 1
The early TXA studies were conducted before the widespread use of modern endoscopic therapy and high‑dose acid suppression, limiting their applicability to current practice. Historical context. 1

All patient‑specific details have been generalized to maintain privacy.

High‑quality HALT‑IT trial (12,009 patients) showed that tranexamic acid did not reduce mortality from gastrointestinal bleeding (relative risk 0.99, 95 % CI 0.82‑1.18). This randomized controlled trial provides level‑1 evidence of no mortality benefit in acute GI hemorrhage. 12
The same HALT‑IT trial demonstrated that tranexamic acid is ineffective for upper gastrointestinal bleeding, likely because the acidic gastric environment degrades the drug. This mechanistic finding is derived from the trial’s subgroup analysis. 13

Administration of tranexamic acid more than 3 hours after traumatic injury is associated with increased hemorrhage‑related death (relative risk 1.44, 95 % CI 1.12‑1.84). Although derived from trauma data, the same time‑sensitivity underscores the drug’s narrow therapeutic window and potential for harm when used outside validated indications. 14, 15

Severe trauma with hemorrhagic risk: Recommended dose is 1 g IV over 10 minutes followed by 1 g IV over 8 hours, administered within 3 hours of injury. This regimen is supported by high‑quality evidence from large trauma trials. 16, 12
Other guideline‑endorsed uses: Emergency cardiac surgery and postpartum hemorrhage are listed as indications where tranexamic acid has demonstrated mortality benefit in randomized studies. 16, 12, 13

When anticoagulation contributes to diverticular bleeding, reversal should be performed with appropriate agents (e.g., idarucizumab for dabigatran, andexanet alfa for factor Xa inhibitors) as advised by current hemostasis guidelines. This recommendation follows consensus statements from major anesthesia and peri‑operative societies. 16, 13

French 2024 clinical guidelines explicitly state that tranexamic acid should be used only for the same indications as in patients not receiving anticoagulants, thereby prohibiting its use in gastrointestinal hemorrhage. This reflects a strong (grade A) recommendation based on systematic review of the evidence. 16, 12, 13
The CRASH‑2 trauma trial’s positive results cannot be extrapolated to gastrointestinal bleeding; the HALT‑IT trial and multiple specialty guidelines confirm lack of efficacy and potential harm in GI hemorrhage. This consensus is supported by level‑1 evidence from the HALT‑IT RCT. 12, 13