Is tranexamic acid helpful in upper GI bleeds?

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

The Lancet. 2020; 395(10241):1927-1936. DOI: 10.1016/S0140-6736(20)30848-5 NCT01713101


  • TXA (tranexamic acid) has been extensively studied in multiple patient populations.
  • Benefit of early administration of TXA has been demonstrated in the bleeding trauma patient (CRASH II), traumatic brain injury (CRASH III) and post-partum haemorrhage (WOMEN).


  • Determine the effect of early administration of tranexamic acid on mortality, morbidity (non-fatal thromboembolic event, re-bleeding), surgical/endoscopic intervention, blood transfusion and health status in patients with significant GI bleeding.


  • International, multi-centre, randomised, placebo-controlled, double blind trial.
  • 164 hospitals in 15 countries.
  • July 2013 – June 2019.
  • Outcomes are assessed at death, discharge or 28 days after randomisation whichever occurs first.
  • Sub group analysis planned in 4 groups; time to treatment (<3 hours, >3 hours), upper vs lower, clinical Rockall Score and suspected variceal/co-morbid liver disease.


  • Adults (Age ³16 or ³18 depending on country where patient enrolled).
  • Responsible clinician must be unsure whether treatment with TXA would benefit the patient.
  • Bleeding had to be deemed ‘significant’ by the responsible clinician. Definition broadly included as patients with hypotension, tachycardia, signs of shock, those likely to require endoscopic/surgical intervention or risk of bleeding to death.
  • No exclusion criteria.
  • 12009 patients randomised. 5956 to TXA group, 5981 to placebo.
  • Patients randomly assigned to TXA or placebo.
  • Intervention: Loading 1 gram dose TXA given over 10 minutes, with a maintenance infusion of 3 grams TXA given over 24 hours.
  • Control: Matched isotonic solution. Infusion schedule identical to intervention.

Primary Outcome:

  • Death due to bleeding within 5 days of randomisation.
    • Outcome was changed (November 2018 – 10000 patients randomised) from all cause mortality.
    • Decision made blinded to accumulated data
    • Initially expected most deaths to be due to bleeding. During trial, observed that half of all deaths were due to non-bleeding causes.
    • Often patients with GI bleeding re bleed. Patients received TXA only for their initial bleed. Given TXA half life (2 hours) there was no expected effect on later rebleeding episodes.

Secondary Outcomes:

  • Multiple secondary outcomes including: death due to bleeding within 24 hours (and within 28 days, rebleeding within 5 days (and 28 days), all-cause and cause specific mortality at 28 days, surgery, endoscopic or radiological intervention, blood product transfusion, thromboembolic events, seizures, other complications (renal, liver and respiratory failure, pneumonia, sepsis and other cardiac events).


  • Primary outcome of death due to bleeding within 5 days. No significant difference between the groups; TXA 222 (3.7%) vs placebo 226 (3.8%), RR 0.99 (95% CI 0.82-1.18).
  • No significant difference in all cause mortality at 28 days (original primary outcome); TXA 9.5% vs placebo 9.2%, RR 1.03 (95% CI 0.92-1.16).
  • Increased incidence of venous thromboembolic events as a whole; TXA 0.8% vs placebo 0.4%, RR 1.85 (95% CI 1.15-2.98).
  • Increased incidence of seizures; TXA 0.8% vs placebo 0.4%, RR 1.85 (95% CI 1.15-2.98).

Study Limitations:

  • Equipoise of responsible clinician of using TXA. Patients with larger GI bleeds may have been given TXA, rather than randomised, given previous positive outcomes in TXA trials and low risk of harm. This is represented in data. 87% of patients had a systolic BP >90 at randomisation.
  • Time from bleeding onset to randomisation. Mean 21.4 hours in TXA group and 22.5 in placebo group. Previous studies have shown that most beneficial effect of TXA is when it is given < 3 hours from onset of bleeding. Only 16% were randomised within 3 hours. Sub-group analysis of this group showed no significant difference in primary outcome. However population relatively small.
  • Change in primary outcome can appear as a methodological weakness.


  • Insidious onset of GI bleeding. Difficult to establish an onset time, this is in contrast to trauma/PPH where bleeding is caused by a definitive event.
  • Trial not powered to look for venous thromboembolism. No information if patients were screened/symptomatic. Difficult to put VTE at day 27 to TXA given on day 1.
  • Increased maintenance dose used compared to other trials (3 grams over 24 hours, rather than 1 gram over 8 hours). Idea behind this was to cover re-bleeding period. Increase incidence in seizures may be dose dependent.

Does this trial change practice?:

  • Good validity, minimal loss to follow up and pragmatic design.
  • Yes – TXA should not routinely be used in the treatment of GI bleeding

Summary by Dr Thomas Burt

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s