The Lancet JULY 03, 2010

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

The CRASH-2 trial collaborators (led by Ian Roberts and Haleema Shakur)

Bottom Line

The CRASH-2 trial demonstrated that early administration of tranexamic acid significantly reduces all-cause mortality in bleeding trauma patients without increasing the risk of fatal or non-fatal vascular occlusive events.

Key Findings

1. Tranexamic acid (TXA) significantly reduced all-cause mortality compared to placebo (14.5% vs 16.0%; relative risk 0.91; 95% CI 0.85-0.97; p=0.0035).

2. The risk of death due to bleeding was significantly reduced with TXA (4.9% vs 5.7%; relative risk 0.85; 95% CI 0.76-0.96; p=0.0077).

3. There was no significant increase in the incidence of fatal or non-fatal vascular occlusive events (including myocardial infarction, stroke, pulmonary embolism, and deep vein thrombosis) in the TXA group compared to the placebo group.

4. Subsequent analysis indicated that the treatment effect is time-dependent; administration within 1 hour of injury showed the greatest benefit, while administration after 3 hours appeared to increase the risk of death due to bleeding.

Study Design

Design

RCT

Double-Blind

Sample

20,211

Patients

Duration

4 wk

Median

Setting

Multicenter, international

Population Adult trauma patients with, or at risk of, significant hemorrhage (defined as systolic BP <90 mmHg, HR >110 bpm, or both) presenting within 8 hours of injury.

Intervention Loading dose of 1 g of tranexamic acid infused over 10 minutes, followed by an intravenous infusion of 1 g over 8 hours.

Comparator Matching placebo (0.9% normal saline) administered as an identical loading dose and infusion.

Outcome All-cause mortality in hospital within 4 weeks of injury.

Study Limitations

• The use of the 'uncertainty principle' for enrollment allowed for potential selection bias, as clinicians randomized only when they were unsure of the benefit.

• The trial was conducted primarily in low-to-middle-income countries, raising questions about the generalizability of results to settings with advanced trauma systems.

• The study did not elucidate the precise biological mechanism by which TXA reduces mortality, as there was no significant difference in the total number of blood products transfused between the two groups.

Clinical Significance

The CRASH-2 trial transformed trauma care by establishing TXA as a cheap, safe, and effective intervention for life-threatening hemorrhage, leading to its inclusion on the WHO Model List of Essential Medicines.

Historical Context

Following the negative findings of the CRASH-1 trial regarding corticosteroids in head injury, the CRASH-2 investigators sought to evaluate an alternative pharmacological approach to improve outcomes in trauma, focusing on the prevention of fibrinolysis in the setting of traumatic hemorrhage.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

What is the pharmacological mechanism of action of tranexamic acid, and how does this relate to the physiological process of clot degradation in trauma patients?

Key Response

Tranexamic acid (TXA) is a synthetic analogue of the amino acid lysine. It competitively inhibits the activation of plasminogen to plasmin by binding to the lysine-binding sites of plasminogen molecules. In trauma, tissue injury triggers hyperfibrinolysis—the premature breaking down of fibrin clots. By blocking plasmin, TXA stabilizes formed clots and reduces hemorrhage, which is the primary mechanism for reducing mortality in the CRASH-2 trial.

Resident

A trauma patient arrives 45 minutes after a motor vehicle accident with a heart rate of 115 bpm and a blood pressure of 105/70 mmHg. Does this patient meet the inclusion criteria for TXA administration used in the CRASH-2 trial?

Key Response

Yes. The CRASH-2 inclusion criteria were broad and pragmatic: adult trauma patients with significant hemorrhage (systolic BP < 90 mmHg or heart rate > 110 bpm), or those considered by the clinician to be at risk of significant hemorrhage. This patient's tachycardia (HR > 110) qualifies them for treatment, even though they are not yet hypotensive.

Fellow

The CRASH-2 trial reported no significant increase in vascular occlusive events (MI, stroke, PE, DVT). How does this finding integrate with the subsequent WOMAN trial and the HALT-IT trial regarding the safety profile of antifibrinolytics?

Key Response

CRASH-2 established safety in trauma, and the WOMAN trial confirmed this in postpartum hemorrhage. However, the HALT-IT trial (gastrointestinal bleeding) found no mortality benefit and a possible increase in venous thromboembolic events. This suggests that the risk-benefit ratio of TXA is context-dependent, potentially favoring use in acute tissue trauma and obstetric hemorrhage over spontaneous mucosal bleeding where the pathophysiology of fibrinolysis may differ.

Attending

In the context of 'Damage Control Resuscitation,' how should the results of CRASH-2 and the follow-up 'time to treatment' analysis influence the coordination between pre-hospital EMS and the trauma bay?

Key Response

The follow-up analysis showed that every 15-minute delay in TXA administration decreases the survival benefit, and administration after 3 hours may be harmful. This necessitates the implementation of pre-hospital protocols where TXA is given by paramedics ('the golden hour'). It serves as a teaching point that TXA is not a 'rescue' medication for the OR, but an 'induction' medication for the resuscitation phase.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

The CRASH-2 trial used a 'large simple trial' design. What are the statistical and methodological advantages of this approach compared to a smaller, highly controlled trial utilizing Thromboelastography (TEG) or Rotational Thromboelastometry (ROTEM) for inclusion?

Key Response

A 'large simple trial' (20,211 patients) maximizes external validity and provides sufficient power to detect small but clinically significant differences in all-cause mortality, which is the gold-standard endpoint. While TEG/ROTEM designs provide mechanistic insights into hyperfibrinolysis, they are often underpowered for mortality and limit generalizability to settings (including many global trauma centers) where advanced viscoelastic testing is unavailable.

Journal Editor

As a reviewer for The Lancet, what are the implications of the study's decision to use 'all-cause mortality' as the primary endpoint rather than 'death due to bleeding'?

Key Response

All-cause mortality is the most rigorous and unbiased endpoint as it avoids 'adjudication bias,' where researchers might misattribute a death to bleeding or non-bleeding causes. However, a reviewer would note that if TXA only affects bleeding deaths, the effect might be diluted by deaths from head injury or multi-organ failure. The fact that CRASH-2 showed a significant difference in all-cause mortality despite this dilution underscores the potency of the intervention.

Guideline Committee

How do the CRASH-2 findings compare with current NICE and ACS-TCC guidelines regarding the '3-hour rule,' and should these guidelines be updated to include specific contraindications for late-arrival patients?

Key Response

Current guidelines (like the European Guideline on Management of Major Bleeding) strongly recommend TXA (Grade 1B) as soon as possible, but strictly within 3 hours. The CRASH-2 subgroup analysis (and later the 2017 meta-analysis) demonstrated that TXA given after 3 hours actually increased the risk of death due to bleeding (OR 1.44). Guideline committees use this to mandate that TXA should not be initiated if the injury occurred more than 300 minutes prior, unless hyperfibrinolysis is proven by viscoelastic testing.

Clinical Landscape

Noteworthy Related Trials

2012

MATTERS Trial

n = 896 · Arch Surg

Tested

Tranexamic acid

Population

Combat-related injury patients requiring massive transfusion

Comparator

No tranexamic acid

Endpoint

In-hospital mortality

Key result: The use of tranexamic acid was associated with reduced mortality in patients requiring massive blood transfusion in a combat setting.

2017

WOMAN Trial

n = 20,060 · Lancet

Tested

Tranexamic acid 1g IV

Population

Women with postpartum hemorrhage

Comparator

Placebo

Endpoint

Death from all causes or hysterectomy

Key result: Tranexamic acid significantly reduced death due to bleeding in women with postpartum hemorrhage when administered early.

2019

CRASH-3 Trial

n = 12,737 · BMJ

Tested

Tranexamic acid 1g IV loading dose followed by infusion

Population

Adults with traumatic brain injury

Comparator

Placebo

Endpoint

Head injury-related death in hospital within 28 days

Key result: Tranexamic acid did not significantly reduce mortality in the overall population, but showed potential benefit in patients with mild-to-moderate head injury.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis