What is the role of tranexamic acid (TXA) in the management of patients with upper Gastrointestinal Bleed (UGIB) or lower Gastrointestinal Bleed (GI bleed)?

Tranexamic Acid in Gastrointestinal Bleeding

Tranexamic acid should NOT be used for either upper or lower gastrointestinal bleeding in routine clinical practice, as the highest quality evidence from the HALT-IT trial demonstrates no mortality benefit and an increased risk of venous thromboembolism. 1, 2

Evidence from the Definitive HALT-IT Trial

The HALT-IT trial (2020-2021) is the largest and most rigorous study on this topic, enrolling 12,009 patients across 164 hospitals in 15 countries. This randomized, double-blind, placebo-controlled trial provides definitive evidence that directly contradicts earlier meta-analyses:

• No mortality benefit: Death due to bleeding within 5 days occurred in 3.7% of tranexamic acid patients versus 3.8% of placebo patients (risk ratio 0.99,95% CI 0.82-1.18) 1, 2
• Increased venous thromboembolism risk: Deep vein thrombosis or pulmonary embolism occurred in 0.8% of tranexamic acid patients versus 0.4% of placebo patients (risk ratio 1.85,95% CI 1.15-2.98) 1, 2
• Increased seizure risk: Seizures occurred in 0.6% of tranexamic acid patients versus 0.4% of placebo patients (risk ratio 1.73,95% CI 1.03-2.93) 1
• Not cost-effective: Economic analysis demonstrated tranexamic acid resulted in slightly poorer health outcomes and was not cost-effective 1

Current Guideline Recommendations

The British Society of Gastroenterology (2019) explicitly recommends against using tranexamic acid for acute lower GI bleeding outside of clinical trials, pending HALT-IT results. 3 This guideline acknowledged:

• Several older trials of tranexamic acid in UGIB showed a 40% risk reduction in mortality on pooled analysis 3
• However, this treatment benefit disappeared when analysis was limited to trials at low risk of bias 3
• The older studies were conducted before routine use of high-dose acid suppression and endoscopic therapy, making them not applicable to modern care 3
• Studies were too small to assess thromboembolic risk adequately 3

Why Earlier Meta-Analyses Were Misleading

The systematic reviews from 2008 and 2021 that suggested benefit were based on small, methodologically flawed trials:

• A 2008 meta-analysis of 1,754 patients suggested mortality reduction (RR 0.61,95% CI 0.42-0.89), but only one trial included modern endoscopic treatments or proton pump inhibitors, and 21% of patients were excluded from analysis 4
• A 2021 meta-analysis confirmed reduced bleeding rates and mortality in older trials, but explicitly noted these findings applied only to upper GI bleeding and that effects on lower GI bleeding required clarification 5
• The HALT-IT trial, with 12,009 patients and modern methodology, supersedes all previous evidence 1, 2

Clinical Algorithm: What to Do Instead

For Hemodynamically Unstable Patients (Shock Index >1):

• Calculate shock index (heart rate ÷ systolic blood pressure); if >1, patient is critically unstable 3
• Perform CT angiography immediately as first-line investigation (sensitivity 79-95%, specificity 95-100%) 3
• CTA can localize bleeding in upper GI tract, small bowel, or colon without bowel preparation 3
• If CTA negative, perform upper endoscopy immediately, as 10-15% of apparent lower GI bleeds originate from upper GI sources 3

For Hemodynamically Stable Patients:

• Apply Oakland score for lower GI bleeding risk stratification 3
• Score ≤8 with self-terminating bleed: consider outpatient investigation 3
• Score >8 or major bleed: admit for inpatient workup 3
• Upper GI bleeding: Perform esophagogastroduodenoscopy within 24 hours after hemodynamic stabilization 6
• Lower GI bleeding: Perform colonoscopy during hospital stay; no evidence that urgent colonoscopy (<24 hours) improves outcomes versus elective timing (36-60 hours) 3

Critical Pitfalls to Avoid

• Do not use tranexamic acid based on older meta-analyses or trauma protocols—the HALT-IT trial definitively shows no benefit and increased harm in GI bleeding 1, 2
• Do not delay imaging in unstable patients while attempting bowel preparation for colonoscopy 3
• Do not assume lower GI presentation excludes upper GI source—always consider upper endoscopy if CTA negative in unstable patients 3
• Do not place nasogastric tubes routinely in suspected UGIB—they do not reliably aid diagnosis, do not affect outcomes, and cause complications in one-third of patients 3

Bottom Line

Tranexamic acid has no role in the management of gastrointestinal bleeding based on the highest quality evidence available. The focus should remain on rapid hemodynamic resuscitation, appropriate imaging (CTA for unstable patients), and timely endoscopic or radiological intervention. 1, 2