• 
• 

Three Part Question

In a [paediatric trauma patient presenting to the emergency department], does using [tranexamic acid] improve outcomes [in terms of mortality, transfusion requirements, re-bleeding rates and need for surgery]?

Clinical Scenario

A 12-year-old child presents to the emergency department following a major trauma. Primary survey reveals signs of shock so you commence initial resuscitation for this. Tranexamic acid is widely used for adult trauma patients and you wonder if using it in this child would reduce the bleeding and improve outcomes.

Search Strategy

A literature search was conducted using:
Medline 1964 to present via NICE Healthcare Databases
EMBASE 1974 to present via NICE Healthcare Databases

[exp *CHILD or exp *ADOLESCENT or exp *MINORS or (adolescen* OR child* OR p?ediatric*).ti,ab]
AND
[exp *”TRANEXAMIC ACID” or ("tranexamic acid" OR TXA).ti,ab]
AND
[exp *"WOUNDS AND INJURIES"/ or (trauma OR injur*).ti,ab]
AND
[exp *MORTALITY or (mortality OR death).ti,ab]

Trip Pro (date unknown) was also searched using the following search strategy:
[paediatric OR child OR pediatric OR adolescent]
AND
["tranexamic acid" OR TXA]
AND
[trauma OR injury]
AND
[reduced mortality OR death]

Search Outcome

Medline: 35 papers. Of these, 4 were relevant.
EMBASE: 66 papers. Of these, 9 were relevant once de-duplicated.
Trip: 7 papers. Of these 1 was relevant but duplicated with Medline & EMBASE.

Total: 67 papers once deduplicated. Of these, 9 were relevant. The majority of excluded papers were looking at the use of TXA in paediatric surgery or following PPH and were not relevant to the aims of this BET therefore excluded.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Hamele et al. 2020 USA	507 trauma patients <18 years from Iraq & Afghanistan requiring > 40ml/kg blood products in 24hrs (excluding burns & fatal head injuries)	Retrospective cohort study	In-hospital mortality	In patients requiring massive transfusions, those receiving TXA were less likely to die in hospital (8.5% vs 18.3%) although not statistically significant (p=0.055)	None of the results statistically significant although TXA reduced odds of death on logistic regression Retrospective study design can only show association not causation Limited data around dose + timing of TXA administration as well as lack of data on complications e.g. thromboembolic events + seizures Subject to survival bias i.e. those that received TXA self-selected as they remained alive to have it
			Hospital, ventilator & ICU-free days	Similar outcomes for hospital, ventilator & ICU-free days
			Total blood product volume	Those receiving TXA had a higher 24-hr blood product administration (p.0.015)
Eckert et al. 2014 USA	All paediatric trauma admissions to Camp Bastion, Afghanistan from 2008-2012 766 injured patients aged <18	Retrospective cohort study	Patient survival before discharge	TXA administration independently associated with decreased mortality among all patients OR 0.27, p=0.03 (correcting for demographics, injury type & severity, vitals & lab parameters) (p=0.03)	Relatively small sample size Retrospective study design can only show association not causation Lack of data regarding blood loss & coagulation studies to compare the effect of TXA on blood loss & coagulation 1g adult dosing administered regardless of child’s weight or age No data on TXA dose timing or subsequent dosing of TXA Limited longer-term outcome data as significant percentage transferred early after stabilisation Combat setting may not be representative of trauma in civilian setting
			Neurologic status at discharge	Patients receiving TXA were significantly more likely to have a near-normal GCS (14-15) at discharge & less severe brain injury (GCS <9) at discharge
			Mechanical ventilation at discharge	Significantly lower percentage of patients receiving TXA requiring mechanical ventilation at discharge (6% vs 22%, p<0.01)
				9% received TXA - usually in severe abdominal or extremity trauma
				No adverse safety/ medication related complications
Thomson et al. 2019 USA	48 patients aged <16 admitted to a single level 1 trauma centre having MTP ordered on admission to ED or ICU	Retrospective observational analysis	Survival to hospital discharge	No difference in survival to hospital discharge in those receiving TXA vs those who didn’t (19% vs 14%, p=0.55)	Observational study design can only show association not causation Small study size & not powered May be a difference in mechanism of injury or unmeasured confounder not accounted for in analysis Only 1/3rd of patients receiving TXA received a subsequent infusion Selection bias - needing MTP so may reflect worse injury severity or haemorrhage than general trauma population Information bias – reliant on documentation of complications e.g. thrombosis
			Surgical intervention	In group not receiving TXA went on to have surgical procedure more often (84% vs 69% but not statistically significant)
			Blood transfusion in 1st 24hrs; type & volume of blood product administered; length of hospital stay; safety based on occurrence of thrombosis	No identifiable difference between groups for blood transfusion, type of blood product administered, length of hospital stay or thrombosis
				60% received TXA
Maeda et al. 2018 Japan	61,779 patients aged <12 admitted to hospital with diagnosis of trauma between 2010-2014	Retrospective cohort study Propensity matched to get 1914 pairs of patients with/without TXA	of TXA – including seizure incidence, thromboembolic events, renal dysfunction	Proportion of seizures significantly higher in group receiving TXA (0.37% vs. 0%, p=0.008)	Retrospective study design can only show association not causation Propensity score matching may have missed unobserved confounding factors Lack of information regarding disease status & medical information e.g. use of anticoagulants & antiepileptics Lack of information on timing of TXA administration – there may be patients coded has having seizure who had this before TXA administered
			In-hospital mortality	No significant difference between other outcomes between groups
Lock et al. 2021 USA	20 patients identified by trauma registry at an academic hospital in Jan 2013. Matched by age, gender, ISS + mechanism of injury	Retrospective cohort study	In hospital mortality	No significant difference for in-hospital mortality (0% vs 6.7%, p=1.00)	Retrospective study design can only show association not causation Small cohort size No statistical significance of results Older population (median age 17) and low baseline ISS (median ISS 12) therefore results may not be applicable to general trauma population
			Definitive bleeding control	Better in TXA group (40% vs 26.7%, p=0.61)
			Length of stay (LOS)	Reduced LOS in TXA group (6 days vs 9.1 days, p=0.83)
			Blood products required in 24hrs	Reduced in TXA group (9.8ml/kg vs 15.2ml/kg, p=0.38)
				No thromboembolic events or seizures
Al-Jeabory et al. 2021 Poland	4 studies comparing TXA vs. non-TXA treatment for paediatric trauma patients excluding articles relating to head injuries. Total of 8751 paediatric patients.	Meta-analysis	In hospital mortality	No statistical difference for in-hospital mortality in TXA vs non-TXA (1.8% vs 3.6% respectively, OR =1.77, p=0.51)	Results not statistically significant Head injuries excluded and TXA may be beneficial for this subset of trauma patients
			Complications: seizures, thromboembolism, renal dysfunction	Increased risk of seizures with TXA administration (0.4% vs 0%, OR =15.06, p=0.06)
				Comparative risk of thromboembolism in TXA vs non-TXA (0.3% vs 0.2%, P=0.72, p=0.62)
				Comparative risk of renal dysfunction in TXA vs non-TXA (0.2% vs 0%, P=7.01, p=0.2)
Booth et al. 2015 UK	16 with traumatic injury between 2006-2013 coded as traumatic injury on joint theatre trauma registry	Retrospective cohort study	Mortality	TXA associated with higher injury severity score (ISS) + consequently overall mortality higher in this group (21.5% vs 16%, p=0.47)	Retrospective study design can only show association not causation Small cohort size High p values therefore not statistically significant
			Sub-group analysis for injury severity score (ISS) >15 and severe TBI (GCS <5)	Lower mortality in children receiving TXA with severe TBI (24% vs 43%, p=0.06) Lower mortality in children receiving TXA with high ISS >15 (27% vs 38%, p=0.18)
				65 cases received TXA (13.4%)
Borgman et al. 2014 UK	4327 paediatric trauma patients without burn or fatal head trauma from defence trauma registry between 2006-2013 receiving massive transfusion (>40ml/kg)	Retrospective cohort study	In-hospital mortality	Lower overall mortality in group receiving TXA (8.5% vs 18.5%, p=0.055)	Retrospective study design can only show association not causation Small cohort size Significantly different mechanism of injury between groups therefore may be confounding results High p values therefore not statistically significant
			24-hour mortality	Lower 24-hr mortality in group receiving TXA (3.4% vs 6.7%, p=0.33)
			Blood products transfused	Those receiving TXA tended to receive more blood products in 1st 24hrs
				59 cases received TXA (11.6%)
				A group had more cases of blast injury (83% vs 57%, p<0.001) whereas group not receiving TXA tended to have more blunt/ penetrating trauma
Thorpe et al. 2021 UK	27,385 patients <16 from TARN data between 2008-2020 + no clinical exclusions	Retrospective analysis of TARN data	Change in prevalence of TXA use over time	TXA use increased year-on-year from 1.2% in 2012 to 10.1% in 2020	Retrospective study design can only show association not causation No p-values therefore unclear if results statistically significant
			Impact of change e.g., thrombotic events, blood product use, surgical intervention, evaluation of injury patterns	No change in 30-day mortality over time
				No significant increase in thrombotic complications over time
				4.9% received TXA in total, 54.6% of these aged 12-15y/o
				RTCs (64%) + penetrating injuries (12.7%) most common mechanisms of injury in those receiving TXA
				Median ISS in those receiving TXA was 20 compared to 9 in untreated patients
				61% of patients treated with TXA were admitted to ICU compared to 18.6% without
				Blood products were transfused in 18.2% patients receiving TXA compared to 0.4% without

Comment(s)

The data is inconsistent as to whether TXA offers any benefit and reduction in mortality, as well as association with complications such as seizures and VTE. All of the studies were retrospective cohort analyses and therefore cannot determine causality. Furthermore, most of the studies had a small sample size therefore unlikely to be representative of the wider population. There is an ongoing RCT (TIC-TOC) comparing the use of TXA vs placebo. Results are expected in 2027. https://clinicaltrials.gov/ct2/show/study/NCT04387305

Clinical Bottom Line

Although there is a wealth of evidence to suggest that TXA administration in adult trauma patients reduces mortality, there is not the same evidence for the paediatric population. The available evidence is conflicting, with some studies demonstrating a reduction in mortality whilst others show no significant difference with an increased risk of complications such as seizures and thrombosis. More research is needed to draw conclusions about the use of TXA in the paediatric trauma population. The RCPCH released a statement in 2012 advising that “TXA administration, preferably within 3 hours of trauma, for children is likely to be beneficial”. They “recommend a pragmatic dosage schedule of 15mg/kg” loading dose “over 10 minutes followed by 2mg/kg/hr.”

References

1. Hamele et al. Tranexamic acid in pediatric combat trauma requiring massive transfusions and mortality Journal of Trauma and Acute Care Surgery April 2020; S242-S245
2. Eckert et al. Tranexamic acid administration to pediatric trauma patients in a combat setting The pediatric trauma and tranexamic acid study (PED-TRAX) Journal of Trauma and Acute Care Surgery December 2014; 852-858
3. Thomson et al. Experience in an Urban Level 1 Trauma Center With Tranexamic Acid in Pediatric Trauma: A Retrospective Chart Review Journal of Intensive Care Medicinee February 2020; 413-418
4. Maeda et al. Safety of Tranexamic Acid During Pediatric Trauma: A Nationwide Database Study Pediatric Critical Care Medicine December 2018; 637-642
5. Lock et al. Efficacy and Safety of Tranexamic Acid in Civilian Pediatric Trauma Patients Receiving Transfusion Critical Care Medicine Jan 2021; 662
6. Al-Jeabory et al. Efficacy and safety of tranexamic acid in pediatric trauma patients: Evidence from meta-analysis The American Journal of Emergency Medicine November 2021; 404-405
7. Booth et al. Use of Tranexamic Acid in the Paediatric Trauma Patient April 2015; 307-308
8. Borgman et al. Tranexamic Acid for Paediatric Trauma Critical Care Medicine December 2014; 1607-1608
9. Thorpe et al. Tranexamic acid in paediatric major trauma: a TARN data analysis 2008–2020 Archives of Disease in Childhood September 2021; 383