What is the role of TXA in head trauma, upper GI bleeding, epistaxis, and hemoptysis?
Bottom line up top: TXA clearly benefits major trauma patients presenting within 3 hours of injury. In other settings the evidence of benefit is weak (epistaxis, TBI, hemoptysis, post-tonsillectomy hemorrhage) or absent (upper GI bleeding).
Clinical scenario
You are working a busy overnight shift with four bleeding patients:
Patient 1: a 24-year-old man hit in the head with a baseball bat, with a subdural hemorrhage on non-contrast head CT. He is intubated and minimally responsive to pain.
Patient 2: a 53-year-old woman with anterior epistaxis not controlled by direct pressure.
Patient 3: a 48-year-old man with cirrhosis and two episodes of hematemesis.
Patient 4: a 64-year-old woman with COPD and hemoptysis.
As you enter orders, you wonder whether all four should get tranexamic acid (TXA).
What your gut says: Just give the TXA. If it works in major trauma, it probably works here too.
What the evidence says
Major trauma: the benefit is clear
CRASH-2 (2010) showed a 1.5% absolute reduction in mortality (NNT = 68) in major trauma patients given TXA within 3 hours of injury (CRASH-2).
Later research suggests the benefit is greatest within 90 minutes of injury, underscoring the importance of rapid administration (Ali 2025).
These results, plus TXA's low cost and wide availability, led to its use in many other bleeding presentations. We previously covered dysfunctional uterine bleeding, where TXA is standard gynecologic care. Here are the other indications.
Isolated traumatic intracranial hemorrhage
CRASH-3, the first well-done RCT of TXA in isolated head injury, found no benefit across all patients with TBI (CRASH-3).
A subgroup analysis suggested benefit in mild-to-moderate head injury, but not in severe head injury.
In spontaneous (non-traumatic) intracerebral hemorrhage, a 2021 meta-analysis found that TXA may cause less hematoma expansion, mainly when given within 4.5 hours of symptom onset, but no difference in 3-month mortality or functional outcome (Guo 2021).
No hospital-based study has shown harm in this group. Given that, patients with TBI and traumatic extracranial hemorrhage should still receive TXA.
Epistaxis
Between 2013 and 2021, several studies reported that topical TXA in anterior epistaxis reduced rebleeding, sped control, and reduced nasal packing.
In 2021, the higher-quality NoPAC RCT found no reduction in nasal packing with TXA in anterior epistaxis (Reuben 2021).
No study has suggested harm from topical TXA.
Gastrointestinal bleeding
Acute GI bleeding is life-threatening, and definitive care is outside the emergency physician's scope, which made it seem the perfect setting for TXA.
HALT-IT, a well-done RCT, found no improvement in 5-day mortality with TXA versus placebo (HALT-IT 2020).
Unlike earlier TXA trials, HALT-IT showed harm: more seizures and more venous thromboembolism in the TXA group.
One likely explanation is the difficulty of pinpointing when GI bleeding started, so TXA may often have been given outside the 3-hour window established by CRASH-2.
Hemoptysis and post-tonsillectomy hemorrhage
Small studies of inhaled and IV TXA for mild to moderate hemoptysis showed less hemoptysis and faster resolution with inhaled TXA (Wand 2018; Gopinath 2023). Massive hemoptysis was not studied.
Evidence for post-tonsillectomy hemorrhage is even weaker. There are no high-quality studies. The available data support use in mild to moderate bleeding but are at high risk of bias, and there are no safety data (Alghamdi 2025).
Clinical bottom line
TXA has been applied to a wide range of bleeding presentations, but evidence outside major trauma and dysfunctional uterine bleeding is sparse.
| Patient | Evidence of benefit | Evidence of harm |
|---|---|---|
| 1. Isolated intracranial hemorrhage | None (strong data) | None |
| 2. Epistaxis | Mixed; the best evidence shows no benefit | None |
| 3. GI bleeding | None (strong data) | Yes |
| 4. Hemoptysis | Yes (moderate-quality data) | None |
Clinical pearls
In major trauma, time is everything: give TXA within 3 hours of injury, and ideally within 90 minutes.
Do not give TXA for GI bleeding. HALT-IT showed no benefit and more seizures and VTE.
For TBI and epistaxis, TXA appears safe but has not been proven to help.
Read more
The Bottom Line: CRASH-3
REBEL EM: TXA for hemoptysis
References
CRASH-2 Trial Collaborators. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet. 2010;376(9734):23–32. PMID: 20554319
Ali A, et al. Tranexamic acid timing and mortality impact after trauma. Ann Emerg Med. 2025. PMID: 40751727
CRASH-3 Trial Collaborators. Effects of tranexamic acid on death, disability, vascular occlusive events and other morbidities in patients with acute traumatic brain injury (CRASH-3): a randomised, placebo-controlled trial. Lancet. 2019;394(10210):1713–1723. PMID: 31623894
Guo Y, et al. Tranexamic acid for acute spontaneous intracerebral hemorrhage: a meta-analysis of randomized controlled trials. Front Neurol. 2021. PMID: 34987465
Reuben A, et al. The use of tranexamic acid to reduce the need for nasal packing in epistaxis (NoPAC): randomized controlled trial. Ann Emerg Med. 2021;77(6):631–640. PMID: 33612282
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. PMID: 32563378
Gopinath B, et al. Nebulized vs IV tranexamic acid for hemoptysis: a pilot randomized controlled trial. Chest. 2023;163(5):1176–1184. PMID: 36410494
Wand O, et al. Inhaled tranexamic acid for hemoptysis treatment: a randomized controlled trial. Chest. 2018;154(6):1379–1384. PMID: 30321510
Alghamdi AS, et al. Nebulized tranexamic acid for treatment of post-tonsillectomy bleeding: a systematic review and meta-analysis. Eur Arch Otorhinolaryngol. 2025;282(3):1135–1146. PMID: 39356357
