Effects of Tranexamic Acid On Death, Vascular Occlusive Events, And Blood Transfusion In Trauma Patients With Significant Haemorrhage (CRASH-2).
F1000: Ranked "Exceptional" and "Changes Clinical Practice"

Jerrold Levy; Patrick Ray; Nicola Latronico; Julia Neely; Alain Vuylsteke


Faculty of 1000. 2010;376 

F1000: Ranked "Exceptional" and "Changes Clinical Practice"

CRASH-2 trial collaborators
Lancet 2010 Jun 14 376

Commentary from Faculty Member Jerrold Levy,[1] Emory University, USA, Anesthesiology & Pain Management

This article is important because this is the first study to show an improvement in outcome following trauma using a pharmacological therapy with antifibrinolytic agents. It is the first to demonstrate a reduction in mortality following any therapy in trauma patients, it has a large sample of patients studied (~20,000) -- an impressive accomplishment -- and a worldwide database which further enhances the unique aspects of the study.

This impressive study, called CRASH-2, evaluated 20,211 trauma patients randomized and treated within 8 hours of injury with either 2 grams of tranexamic acid (1 gram load, then 1 gram over 8 hours) or placebo. The primary outcome was in-hospital mortality within 4 weeks of injury; vascular occlusive events, transfusions or surgical interventions were secondary outcomes. The authors demonstrated a reduction in all-cause mortality, 14.5% in the tranexamic acid group (1463/10,060) compared to 16% in the placebo group (1613/10,067), with P=0.0035; bleeding-related mortality was also reduced by 4.9% vs. 5.7%, respectively, without an increase in fatal or nonfatal vascular occlusive events. This important finding further supports the role of antifibrinolytic therapy following massive tissue injury as a therapeutic modality. We and others have demonstrated the importance of using antifibrinolytic therapy to reduce bleeding following surgery, and this important finding further supports the application of tranexamic acid in a trauma setting. Trial registration: NCT00375258

Commentary from Faculty Member Patrick Ray,[2] Groupe Hospitalier Pitié-Salpêtrière, France, Critical Care & Emergency Medicine

Based on these results, medical teams caring for trauma patients should think about reviewing their protocols. In this randomized study, the authors nicely suggest that use of tranexamic acid could reduce in-hospital mortality from all causes within 4 weeks of injury (14.5% versus 16.0%; relative risk [RR] 0.91) in patients with active hemorrhage or significant risk for recurrent or ongoing bleeding.

Tranexamic acid is inexpensive, simple to administer and widely used for decreasing blood loss and transfusion requirements during surgery. Tranexamic acid (loading dose of 1g during 10 minutes followed by an infusion of 1g during 8 hours) or placebo was administered within 8 hours of injury in 20,211 adult trauma patients with (or at risk for) bleeding. The risk for hemorrhagic death was reduced (4.9% versus 5.7%; RR 0.85). Furthermore, the incidence of vascular occlusive events was not increased in the tranexamic acid group (about 2% in the two groups).

Commentary from Faculty Member Nicola Latronico,[3] University of Brescia, Italy, Critical Care & Emergency Medicine

Changes Clinical Practice: Trauma patients with a significant hemorrhage (resulting in a systolic blood pressure <90mmHg or heart rate >110 beats per min, or both), or those considered by the clinician to be at significant risk of hemorrhage, should be given a loading intravenous dose of 1g of tranexamic acid infused over 10min, followed by an intravenous infusion of 1g over 8h as soon as possible after trauma.

This is an exceptionally important study, the first to demonstrate that tranexamic acid, a cheap anti-fibrinolytic drug, significantly reduces mortality in adult trauma patients.

This was a multinational, multicenter randomized controlled trial (RCT) enrolling 20,211 adult trauma patients admitted to 274 hospitals in 40 countries. Patients were enrolled in cases of significant hemorrhage (systolic blood pressure <90mmHg or heart rate >110 beats per min, or both), or if at risk of significant hemorrhage within 8h of injury. The uncertainty principle was used to govern randomization, i.e. patients were randomized only if the doctor responsible for treatment was substantially uncertain about whether or not to treat with tranexamic acid. Patients received a loading dose of 1g of tranexamic acid infused over 10min, followed by an intravenous infusion of 1g over 8h, or a placebo. Primary outcome was death in hospital within 4 weeks of trauma, and was available for 20,127 patients. Secondary outcomes were vascular occlusive events, surgical intervention, and blood transfusion. Dependency was measured at hospital discharge, or on day 28 if still in hospital, with the 5-point Modified Oxford Handicap Scale. The relative risk (RR) of death with tranexamic acid was 0.91 (95% confidence interval [CI] 0.85-0.97, p=0.0035); the number needed to treat to avoid 1 hospital death was 68. Mortality was lower in all examined subgroups, including head trauma patients, who represented 50% of the patients enrolled. Vascular occlusive events, including acute myocardial infarction, ischemic stroke, pulmonary embolism and deep vein thrombosis, did not differ significantly (1.7% in the tranexamic acid group and 2.0% in the placebo group). Trial registration: ISRCTN86750102, NCT00375258 and DOH-27-0607-1919

Commentary from Faculty Member Julia Neely and Alain Vuylsteke,[4] Papworth Hospital, UK, Anesthesiology & Pain Management

Changes Clinical Practice: To reduce mortality, trauma patients with or at risk of haemorrhage should be given 1g of tranexamic acid within 8 hours of the trauma injury, followed by 1g as an infusion over 8 hours.

This is an amazing study showing that a simple intervention after trauma could save many lives. This multinational double-blind randomised study of more than 20,000 patients shows that early administration of tranexamic acid to trauma victims with or at risk of haemorrhage leads to reduced mortality without placing patients at increased risk of vascular occlusive events.

A large consortium of investigators, the CRASH-2 collaborators, used the premises that (1) tranexamic acid improves surgical outcomes through reduction in bleeding[1] and (2) tissue injury associated with trauma is not dissimilar to that induced by surgery. They explored this hypothesis in a double-blind randomised control study incorporating 20,211 patients from hospitals in 40 countries. Patients at risk of haemorrhage or who were clinically identified using physiological parameters as suffering from haemorrhage within 8 hours of a traumatic injury were randomised to well-balanced intervention and placebo groups. All were analysed on an intention-to-treat basis and very few patients were lost to follow-up. The authors demonstrated that administering 1g of tranexamic acid within 8 hours of traumatic injury, followed by 1g as an infusion over 8 hours led to a relative risk reduction in all-cause mortality compared to placebo of 0.91 (95% CI 0.85-0.97, p=0.0035). There was no significant difference between patients receiving tranexamic acid and those receiving placebo regarding vascular occlusive events (including myocardial infarction, cerebrovascular events and pulmonary embolism). Tranexamic acid showed a significant reduction in the probability of death due to bleeding -- relative risk of 0.85 (95% CI 0.76-0.96, p=0.0077) -- but was not associated with a reduction in transfusion rates or the number of units transfused. The authors suggest this may be associated to either early decisions to increase intravascular volume or reduced mortality especially on day 0 for the intervention group, leading to increased time during which transfusions could be given. The benefits of giving tranexamic acid probably go further than reducing haemorrhage; preventing plasmin-activated inflammatory mediators dampens the overall immune response to trauma via the activated protein C pathway,[2] but the study failed to show a significant difference in multiple organ failure between the two groups. The dose of tranexamic acid used was extrapolated from surgical studies; these recommend dose calculations based on weight, but in the interest of standardizing treatment in situations where patients' weight may be difficult to obtain, and ensuring the dose lay within therapeutic and safe ranges previously reported, the authors decided on 1g loading dose and 1g maintenance dose, which showed the significant benefits described. This well-designed, well-executed trial gives food for thought about the provision of tranexamic acid as standard practice for trauma patients at risk of haemorrhage in the acute setting worldwide.
Trial registration: ISRCTN86750102