Your browser is unsupported

We recommend using the latest version of IE11, Edge, Chrome, Firefox or Safari.

What is the role of tranexamic acid in patients with acute gastrointestinal bleeding?

Acute gastrointestinal (GI) bleeding is one of the most common GI diagnoses associated with emergency department visits, hospital admissions, and readmissions.1 Current formal definitions designate esophageal, gastric, or duodenal bleeding sources as upper GI bleeding (UGIB), while lower GI bleeding (LGIB) refers to bleeding originating in the colon or rectum; bleeding originating in the portion of the bowel not falling into the upper or lower categories is referred to as small intestinal or “middle” GI bleeding, and is relatively uncommon.2-4 Upper GI bleeding constitutes the majority (59%) of emergency department visits involving GI bleeding, whereas LGIB is less common (27%); when GI bleeding is the principal diagnosis associated with an emergency department visit, the majority of patients with UGIB (approximately 72%) and a large proportion of patients with LGIB (approximately 42%) are subsequently admitted to the hospital.1 These events are significant sources of morbidity, mortality, and cost, with case-fatality rates ranging from 8% to 26% for UGIB and 3% to 9% for LGIB, and median inpatient costs incurred and charges billed per admission of approximately $7900 and $31,400, respectively.1,5

The management of acute GI bleeding hinges upon rapid assessment and resuscitation, addressing reversible causes (such as anticoagulants), and endoscopic (or, in some cases, radiological or surgical) identification of and intervention at the site, when possible, to stop the bleeding.2-4,6 Even when hemostasis is achieved, rebleeding remains an important potential complication, with variable rates according to site and etiology of the original hemorrhage and other risk factors. Tranexamic acid (TXA), an inhibitor of fibrinolysis commonly utilized as a hemostatic in postpartum and traumatic hemorrhage, has been proposed both as an adjunct treatment to improve outcomes in acute GI bleeding and as an option for management in settings where endoscopic intervention and other advanced care techniques may not be readily accessible.7 Guidelines published by the American College of Gastroenterology for the management of UGIB (2021) and LGIB (2016) are principally focused on endoscopic and other invasive interventions and do not address the use of TXA or other systemic hemostatic agents.3,4 This review summarizes key literature evaluating TXA use in patients with acute GI bleeding.

Literature Review
Important studies regarding use of TXA in GI bleeding are summarized in the Table. A 2014 Cochrane meta-analysis of randomized controlled trials of TXA for UGIB was performed with the aim of clarifying the mixed data available at the time regarding TXA efficacy and safety in this setting.8 The primary analysis indicated a 40% reduction in relative risk of mortality with TXA versus comparators, without an increase in risk of thromboembolic events; however, no mortality benefit was noted in a sensitivity analysis restricted to studies involving the use of endoscopic intervention and/or proton pump inhibitors, and availability of data for the analysis of thromboembolic events was limited. No benefit was identified in risk of rebleeding, subsequent surgery, or transfusion with TXA versus comparators. This study was limited by heterogeneity in TXA regimens utilized, unclear risk of bias in several trials, changes in primary management of UGIB over time, and the relatively small sample size of the individual trials.

The limitations of contemporary data, including those highlighted by the meta-analysis described above and the overall lack of data for TXA use in LGIB, formed the basis for the recently-reported Hemorrhage Alleviation with Tranexamic Acid-Intestinal System (HALT-IT) trial.9 In this large trial, adults with a clinical diagnosis of significant GI bleeding (11% of whom had suspected LGIB) for whom the treating clinician had substantial uncertainty whether TXA was indicated were randomized to receive relatively high doses of TXA or placebo over 24 hours. In the primary analysis of death due to bleeding within 5 days of randomization, there was no significant difference between TXA and placebo, and no secondary efficacy outcomes were significantly different between arms; however, relative risk of venous thromboembolism (VTE) was over 2-fold higher (absolute risk 0.85% vs 0.4%) and relative risk of seizure was 73% higher (absolute risk 0.6% vs 0.4%) with TXA compared to placebo. The difference in VTE appeared to be driven by the relatively high proportion of patients with suspected variceal bleeding or liver disease (risk ratio 7.26, 95% confidence interval 1.65 to 31.90, versus no significant difference in relative risk of VTE in patients with other sites of bleeding).

While the HALT-IT trial strongly refutes the mortality benefit identified in the 2014 Cochrane analysis, it raises additional questions regarding more attenuated benefits of TXA, if any, in GI bleeding. The dose of TXA used in HALT-IT was 2- to 4-fold higher than intravenous (IV) doses typically used for other hemorrhagic indications, while the treatment duration with TXA was relatively short compared to historical studies, suggesting the possibility that lower doses (and/or use of enteral preparations) for longer durations may underly previous findings of TXA benefit without increased toxicity.10 Additionally, the high proportion of patients in HALT-IT with variceal bleeding or liver disease, who are predisposed to thromboembolic events, suggests limited external validity of the study’s VTE risk findings.11

Shortly after publication of HALT-IT in 2020, Lee and colleagues published a meta-analysis of 13 randomized placebo-controlled trials (including 1 small randomized trial excluded from the 2014 Cochrane analysis because it evaluated TXA in patients with stable ulcerative colitis-related hematochezia, and 4 randomized trials published since the Cochrane analysis) evaluating TXA in patients with GI bleeding.12 This analysis indicated a similar mortality benefit to the 2014 Cochrane analysis, as well as reduced rates of continued bleeding (but not rebleeding) and urgent endoscopic intervention (but not overall endoscopic intervention or surgical intervention) with TXA versus placebo. However, data from HALT-IT were not available at the time of the literature search cutoff – thus, a limited follow-up meta-analysis was undertaken in 2022 using the same data as the earlier meta-analysis with the addition of HALT-IT results.13 The authors found a 24% reduction in relative risk of death even after incorporating HALT-IT data, and suggested that the relatively large proportion of patients enrolled in HALT-IT with risk factors for poor outcomes (eg, 43% of patients presenting with signs of shock and 38% with high-severity bleeds according to prognostic scoring, indicative of substantial risk of mortality or rebleeding) may have precluded a finding of benefit with TXA in patients with less emergent presentations in the HALT-IT trial.9,13

Another recent meta-analysis by Dionne and colleagues examined TXA use in GI bleeding through the lens of potential dose-related effects on efficacy or toxicity. In this study published in 2022, data from 12 randomized controlled trials (including all studies analyzed in the 2014 Cochrane analysis and 3 randomized trials published since; 1 randomized trial included in the 2020 meta-analysis by Lee et al was not included due to the timeframe of the original literature search) were evaluated by TXA dose strata, where high-dose TXA was defined as IV doses of more than 2 g administered over 24 hours and low-dose TXA was defined as IV doses of 2 g or less administered over 24 hours or regimens utilizing enteric TXA only.14 Although there was no mortality benefit over placebo for either high-dose or low-dose TXA regimens, the authors did identify significantly lower risks of rebleeding and subsequent surgical intervention in the low-dose TXA subgroup. The analysis also indicated increased VTE rates in patients receiving high-dose TXA regimens; however, there were no data available for the risk of VTE or arterial thrombotic events in trials utilizing low-dose TXA regimens, precluding any conclusion regarding dose-dependent risk of thrombosis.

Table. Key studies of TXA for GI bleeding.8,9,12-14
Study design and duration
Bennett et al (2014)8

SR/MA of 8 RCTs published between 1973 and 2011
N=1701 adults with suspected or confirmed UGIB (median n=204 per trial)
TXA 4 g to 8 g per day IV or PO for 2 to 7 days vs PPI and/or placebo
*Mortality RR=0.60 (95% CI 0.42 to 0.87)
Rebleeding RR=0.72 (95% CI 0.50 to 1.03)
Subsequent surgery RR=0.61 (95% CI 0.35 to 1.04)
Any transfusion RR=1.02 (95% CI 0.94 to 1.11)
Thromboembolic event RR=1.37 (95% CI 0.36 to 5.28)
No benefit of TXA for mortality when restricted to studies utilizing endoscopic intervention and/or PPI
HALT-IT (2020)9

International, multicenter, double-blind phase 3 RCT
N=12,009 adults with significant UGIB (n=10,681) or LGIB (n=1328)
TXA 1 g IV bolus, followed by continuous infusion of TXA 3 g over 24 hours (n=5,994) or matching placebo (n=6,015)
Primary: Death due to bleeding within 5 days RR=0.99 (95% CI 0.82-1.18)
All-cause mortality within 28 days RR=1.03 (95% CI 0.92 to 1.16)
Rebleeding within 28 days RR=0.97 (95% CI 0.82 to 1.15)
Any subsequent surgical, endoscopic, or radiological intervention RR=1.00 (95% CI 0.99 to 1.01)
Any transfusion RR=0.99 (95% CI 0.97 to 1.02)
Thromboembolic event RR=1.54 (95% CI 0.83 to 2.83)
*VTE event RR=2.11 (95% CI 1.24 to 3.59)
*Seizure RR=1.73 (95% CI 1.03 to 2.93)
Lee et al (2021)12
SR/MA of 13 RCTs published between 1973 and 2020
N=2271 patients with UGIB or unspecified site (12 trials, median n=152 per trial) or LGIB (1 trial, total n=108)
TXA 4 g to 8 g per day IV or PO for 2 to 7 days and/or 1 g to 2 g administered once via NGT vs placebo
*Mortality RR=0.60 (95% CI 0.45 to 0.80)
*Continued bleeding RR=0.60 (95% CI 0.43 to 0.84)
Rebleeding RR=0.84 (95% CI 0.61-1.15)
Subsequent surgery RR=0.70 (95% CI 0.44 to 1.10)
Subsequent endoscopic intervention RR=0.91 (95% CI 0.54 to 1.51)
*Urgent endoscopic intervention RR=0.35 (95% CI 0.24 to 0.50)
Any transfusion RR=0.94 (95% CI 0.76 to 1.16)
Tran et al (2022)13

SR/MA of 14 RCTs published between 1973 and 2020
N=14,280 patients with UGIB or LGIB (comprising populations studied by Lee et al12 and HALT-IT collaborators9, described above)
TXA 4 g to 8 g per day IV or PO for 1 to 7 days and/or 1 g to 2 g administered once via NGT vs placebo
*Mortality RR=0.76 (95% CI 0.59 to 0.95)

Rebleeding RR=0.89 (95% CI 0.68 to 1.17)
Dionne et al (2022)14

SR/MA of 12 RCTs published between 1973 and 2020
N=13,999 patients with UGIB (12 trials, median n=206 per trial) or LGIB (2 trials, total n=1424)
TXA vs placebo
Analysis stratified by dose intensity of TXA regimen: high-dose TXA (> 2 g IV in 24 hours, total n=13,109) vs low-dose TXA (≤ 2 g IV in 24 hours or enteral-only, total n=890)
High-dose TXA RR=0.98 (95% CI 0.88 to 1.09)
Low-dose TXA RR=0.62 (95% CI 0.36 to 1.09)
High-dose TXA RR=0.92 (95% CI 0.82 to 1.04)
*Low-dose TXA RR=0.50 (95% CI 0.33 to 0.75)
Subsequent surgery
High-dose TXA RR=0.92 (95% CI 0.76 to 1.09)
*Low-dose TXA RR=0.58 (95% CI 0.38 to 0.88)
Any transfusion
High-dose TXA RR=1.00 (95% CI 0.99 to 1.01)
Low-dose TXA RR=1.03 (95% CI 0.93 to 1.13)
Thromboembolic events
Only high-dose TXA trials reported thrombotic event rates
Arterial thrombosis: No difference between high-dose TXA and placebo
*DVT: High-dose TXA RR=2.01 (95% CI 1.08 to 3.72)
*PE: High-dose TXA RR=1.78 (95% CI 1.06 to 3.00)
*Statistically significant finding
Abbreviations: CI=confidence interval; DVT=deep venous thrombosis; GI=gastrointestinal; IV=intravenously; LGIB=lower gastrointestinal bleeding; NGT=nasogastric tube; PE=pulmonary embolism; PO=by mouth; PPI=proton pump inhibitor; RCT=randomized controlled trial; RR=risk ratio; SR/MA=systematic review with meta-analysis; TXA=tranexamic acid; UGIB=upper gastrointestinal bleeding; VTE=venous thromboembolism

Taken together, these recent studies present a complicated picture. While the largest and most rigorous study of TXA in GI bleeding clearly does not support clinical benefit of a high-dose regimen in terms of mortality, rebleeding, or other important outcomes, the meta-analysis by Dionne and colleagues also produces a strong signal that dose and/or route of administration may be of substantial importance.9,14 Additionally, these studies and the others comprising the evidentiary basis for TXA use in this setting carry important limitations related to populations and risk factors represented, heterogeneity in use of TXA and concurrent interventions, and incomplete reporting of safety data. While it is clear that uniform application of TXA in patients with acute GI bleeding, especially at high doses, is not warranted, it is not clear that there is no role for TXA in this setting, and concerted clinical studies addressing the knowledge gaps identified by recent literature are needed.

The role of TXA in patients with acute GI bleeding has been evaluated in a variety of settings over the last several decades. While a recent large, well-designed trial indicates that TXA is not associated with broad clinical benefit in these patients, meta-analyses suggest that there may be situations in which TXA has utility without introducing undue risk, and clinical trials addressing these new questions are awaited.


  1. Peery AF, Crockett SD, Murphy CC, et al. Burden and Cost of Gastrointestinal, Liver, and Pancreatic Diseases in the United States: Update 2021. Gastroenterology. 2022;162(2):621-644. doi:10.1053/j.gastro.2021.10.017
  2. Gerson LB, Fidler JL, Cave DR, Leighton JA. ACG Clinical Guideline: Diagnosis and Management of Small Bowel Bleeding. Am J Gastroenterol. 2015;110:1265-1287; quiz 1288. doi:10.1038/ajg.2015.246
  3. Laine L, Barkun AN, Saltzman JR, Martel M, Leontiadis GI. ACG Clinical Guideline: Upper Gastrointestinal and Ulcer Bleeding. Am J Gastroenterol. 2021;116:899-917. doi:10.14309/ajg.0000000000001245
  4. Strate LL, Gralnek IM. ACG Clinical Guideline: Management of Patients With Acute Lower Gastrointestinal Bleeding. Am J Gastroenterol. 2016;111:459-474. doi:10.1038/ajg.2016.41
  5. Zheng NS, Tsay C, Laine L, Shung DL. Trends in characteristics, management, and outcomes of patients presenting with gastrointestinal bleeding to emergency departments in the United States from 2006 to 2019. Aliment Pharmacol Ther. 2022;56(11-12):1543-1555. doi:10.1111/apt.17238
  6. Tokar JL, Higa JT. Acute Gastrointestinal Bleeding. Ann Intern Med. 2022;175(2):ITC17-ITC32. doi:10.7326/AITC202202150
  7. Wang K, Santiago R. Tranexamic acid – A narrative review for the emergency medicine clinician. Am J Emerg Med. 2022;56:33-44. doi:10.1016/j.ajem.2022.03.027
  8. Bennett C, Klingenberg SL, Langholz E, Gluud LL. Tranexamic acid for upper gastrointestinal bleeding. Cochrane Database Syst Rev. 2014;2014:CD006640. doi:10.1002/14651858.CD006640.pub3
  9. 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(10241):1927-1936. doi:10.1016/S0140-6736(20)30848-5
  10. Shah A, Kerner V, Stanworth SJ, Agarwal S. Major haemorrhage: past, present and future. Anaesthesia. 2022. doi:10.1111/anae.15866
  11. Sogaard KK, Horvath-Puho E, Gronbaek H, Jepsen P, Vilstrup H, Sorensen HT. Risk of venous thromboembolism in patients with liver disease: a nationwide population-based case-control study. Am J Gastroenterol. 2009;104:96-101. doi:10.1038/ajg.2008.34
  12. Lee PL, Yang KS, Tsai HW, Hou SK, Kang YN, Chang CC. Tranexamic acid for gastrointestinal bleeding: A systematic review with meta-analysis of randomized clinical trials. Am J Emerg Med. 2021;45:269-279. doi:10.1016/j.ajem.2020.08.062
  13. Tran QK, Tang K, Pourmand A. Tranexamic acid and Gastrointestinal bleed: Effect of the HALT-IT trial on current meta-analysis. Am J Emerg Med. 2022;59:165-167. doi:10.1016/j.ajem.2022.06.050
  14. Dionne JC, Oczkowski SJW, Hunt BJ, et al. Tranexamic Acid in Gastrointestinal Bleeding: A Systematic Review and Meta-Analysis. Crit Care Med. 2022;50:e313-e319. doi:10.1097/CCM.0000000000005362

Prepared by:
Michael Buege, PharmD, BCOP
Clinical Assistant Professor, Drug Information Specialist
University of Illinois at Chicago College of Pharmacy

December 2022

The information presented is current as November 23, 2022. This information is intended as an educational piece and should not be used as the sole source for clinical decision-making.