Intervention Review

You have free access to this content

Antifibrinolytic drugs for acute traumatic injury

  1. Ian Roberts1,*,
  2. Haleema Shakur2,
  3. Katharine Ker1,
  4. Tim Coats3,
  5. on behalf of the CRASH-2 Trial collaborators2

Editorial Group: Cochrane Injuries Group

Published Online: 12 DEC 2012

Assessed as up-to-date: 14 JUL 2010

DOI: 10.1002/14651858.CD004896.pub3


How to Cite

Roberts I, Shakur H, Ker K, Coats T, on behalf of the CRASH-2 Trial collaborators. Antifibrinolytic drugs for acute traumatic injury. Cochrane Database of Systematic Reviews 2012, Issue 12. Art. No.: CD004896. DOI: 10.1002/14651858.CD004896.pub3.

Author Information

  1. 1

    London School of Hygiene & Tropical Medicine, Cochrane Injuries Group, London, UK

  2. 2

    London School of Hygiene & Tropical Medicine, Clinical Trials Unit, London, UK

  3. 3

    Leicester Royal Infirmary, Department of Emergency Medicine, Leicester, UK

*Ian Roberts, Cochrane Injuries Group, London School of Hygiene & Tropical Medicine, North Courtyard, Keppel Street, London, WC1E 7HT, UK. Ian.Roberts@Lshtm.ac.uk.

Publication History

  1. Publication Status: New search for studies and content updated (no change to conclusions)
  2. Published Online: 12 DEC 2012

SEARCH

 
Characteristics of included studies [ordered by study ID]
Auer 1979

MethodsProbable RCT: "Twenty patients were included in a double-blind study; nine patients were treated with Trasylol. Eleven received a placebo drug." Five additional (non-randomly allocated) patients were added to the study and received aprotinin treatment. These patients were not separated out in the analysis.


ParticipantsPatients with severe head injury who had remained comatose for seven days. Most of them had clinical brain stem signs.


InterventionsAprotinin group: 500,000 IE initially thereafter 200,000 IV every four hours.


OutcomesDeath.
Range of biochemical end points.


NotesBecause it was not possible to separate the 5 non-randomised patients from the 20 probably randomised patients, this study provides no useable outcome data.


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Unclear riskUnclear.

CRASH-2 2010

MethodsRCT: Randomisation was balanced by centre, with an allocation sequence based on a block size of eight, generated with a computer random number generator. Participants and study staff were blind to treatment allocation.


Participants20,211 adult (>16 years) trauma patients with, or at risk of, significant bleeding.


InterventionsTranexamic acid group: loading dose 1g over 10 minutes then infusion of 1g over 8 hours.

Matching placebo.


OutcomesDeath.
Vascular occlusive events.
Blood transfusion requirements.
Disability.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Low riskTXA and placebo were packaged in identical ampoules. Hospitals with a reliable telephone access used a telephone randomisation service, hospitals without used a local pack system.

McMichan 1982

MethodsRCT: Aprotinin and placebo were supplied in identical coded ampoules. Ampoules were in boxes of 50 with a code number assigned to each box. These numbers were randomised in groups of 20 and each batch was assigned in numerical order. The codes were not broken until the end of the study.

Patients excluded after randomisation were those who died within the first 24 hours or refused continuing investigation.


ParticipantsPatients with a combination of hypovolaemic shock and major fractures of the lower limb and or pelvis. Patients seen 12 or more hours after injury and those with major head or chest injuries were excluded.


InterventionsAprotinin group: 500,000 Kallikrein Inhibitor Units (KIU) IV statim followed by 300,000 KIU at 6-hour intervals for 96 hours.


OutcomesDeath.
Mean blood transfusion.
Respiratory function.


Notes77 patients were randomised but there were 7 post-randomisation exclusions. Among the 7 excluded patients, there were 3 deaths within the first 24 hours of injury. One patient was transferred to another hospital because of quadriplegia and died later, and three patients refused investigation.
It was noted in the results that the data on transfusion requirement was found to have a non-normal distrubution. Nevertheless, the mean and standard deviation were presented.


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Low riskAprotinin and placebo were prepared in "similar ampoules". All ampoules were in boxes of 50, with a code number assigned to each box. The nature of the content of the ampoules was not known to any of the investigators nor to the attending physicians. The codes were not broken until the end of the study.

Yutthakasemsunt 2010

MethodsRCT


Participants240 adults patients (>16 years) with moderate to severe traumatic brain injury (Glasgow Coma Scale 4 to 12) within 8 hours of injury.


InterventionsTranexamic acid group: 2g.

Matching placebo.


OutcomesDeath.
Progressive intracranial haemorrhage.
Disability (GOS).
Thromboembolic events.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Allocation concealment (selection bias)Unclear riskUnclear

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Gierhake 1971Types of patients: general surgery patients not trauma.

Husted 2003Types of patients: orthopaedic patients not trauma.

Klobow 1977aTypes of interventions: trasylol compared with heparin.

Klobow 1977bTypes of interventions: trasylol compared with heparin.

Kuiian 1999Types of studies: After Dr Vasiliy Vlassov, Director of the Russian Branch of the Nordic Cochrane Centre kindly translated the methods section it was clear that this study was not randomised.

Loew 1970Types of studies: alternation used not random allocation.

Nissen 1989Types of studies: review article not randomised controlled trial.

Schneider 1976Types of studies: randomisation in this trial was by allocating patients to the treatment group according to the day of admission. However, this procedure was subverted for large numbers (813) of patients in which case the study cannot be considered to be a randomised controlled trial.

 
Characteristics of ongoing studies [ordered by study ID]
CRASH-3

Trial name or titleClinical Randomisation of an Antifibrinolytic in Significant Head Injury (CRASH-3)

MethodsLarge, international, randomised, placebo controlled trial.

ParticipantsAdults with traumatic brain injury, who are within eight hours of injury, with any intracranial bleeding on CT scan or who have a GCS of 12 or less, and have no significant extra-cranial haemorrhage, are eligible for inclusion, except those for whom antifibrinolytic agents are thought to be clearly indicated or clearly contra-indicated

InterventionsLoading dose of tranexamic acid (1 gram by intravenous injection) or placebo (sodium chloride 0.9%) given as soon as possible after
randomisation. Maintenance dose of tranexamic acid (1 gram by intravenous injection) or placebo (sodium chloride 0.9%) given after the loading dose is finished.

OutcomesPrimary outcome is death in hospital within 28 days of injury. Secondary outcomes are vascular occlusive events (myocardial infarction, pulmonary embolism, clinical evidence of deep vein thrombosis), stroke, disability, seizures, neurosurgical intervention, days in intensive care, other adverse events.

Starting dateJuly 2012

Contact informationcrash@Lshtm.ac.uk

NotesCurrent Controlled Trials ISRCTN15088122; Clinicaltrials.gov NCT01402882.

The JP Moulton Charitable Trust, UK, is funding the run-in costs for the trial and up to 500 patients’ recruitment. Full funding is being sought from public funding organisations for the main trial.

 
Comparison 1. Tranexamic acid versus placebo

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 All-cause mortality220367Risk Ratio (M-H, Fixed, 95% CI)0.90 [0.85, 0.97]

 2 Proportion undergoing surgical intervention1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

 3 Proportion receiving blood transfusion1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

 4 Volume of blood transfused1Mean Difference (IV, Fixed, 95% CI)Subtotals only

 
Comparison 2. Aprotinin versus placebo

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Death1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

 2 Proportion undergoing surgical intervention1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

 3 Volume of blood transfused1Mean Difference (IV, Fixed, 95% CI)Subtotals only

 
Table 1. Deaths that could be avoided by the administration of TXA to bleeding trauma patients (ten countries with the highest numbers of avoided deaths shown)

CountryTrauma deathsHaemorrhage deathsDeaths averted with TXA

India714,73085,76812,865

China667,27780,07312011

Indonesia279,49933,5345030

Russia246,83629,6204443

Brazil122,95314,7542206

USA122,52914,7032206

Iraq99,96811,9961799

Nigeria87,81110,5371581

Bangladesh76,93892331385

DRC73,57988291324

World4,100,645492,07773,812

 Trauma and haemorrhage death estimates taken from the WHO Global Burden of Disease Study: http://www.who.int/healthinfo/global_burden_disease/en/