Abstract
Objectives
The aim of this study was to assess 24-month functional outcomes of patients randomly assigned to prehospital tranexamic acid (TXA) compared with those assigned to placebo.
Methods
The PATCH-Trauma trial assessed administration of TXA by prehospital clinicians and demonstrated greater survival at 28 days from injury among patients allocated to receive TXA, but no difference in favorable functional outcomes at 6 months.
This was a subgroup analysis of patients enrolled in the PATCH-Trauma trial. The primary outcome measure was the 8-point Glasgow Outcome Scale-Extended (GOS-E) at 24 months after injury, dichotomized to favorable functional outcomes (GOS-E 5-8) and dead or unfavorable functional outcomes (GOS-E 1-4). We also assessed mortality at 24 hours, 28 days, 6 months, and 24 months after injury.
Results
There were 584 patients eligible for inclusion, and 516 patients had data on 24-month outcomes available. Baseline characteristics were comparable. At 24 months, a favorable outcome was reported in 167 (64.0%; 95% CI: 57.8-69.8) patients in the TXA group and 149 (58.4%; 95% CI: 52.1-64.5) in the placebo group (P = .20). There was no difference in the change in GOS-E profile from 6 to 24 months (coeff −0.06; 95% CI: −0.30 to 0.19; P = .64). There was no difference in mortality at any timepoint.
Conclusion
Prehospital treatment and ongoing hospital infusion of TXA did not result in improved survival or statistically significant functional outcomes at 24 months compared to placebo. Consideration of longer-term outcomes after trauma is recommended for evidence to practice decisions.
Keywords: wounds and injuries, blood transfusion, tranexamic acid, death, shock, hemorrhagic, emergency medicine
The Bottom Line.
We followed 584 patients who were randomly allocated to receive prehospital tranexamic acid (TXA) or placebo as part of the PATCH-Trauma Trial. More patients in the TXA group reported a favorable outcome at 24 months after injury (64% vs 58%), but the difference did not reach statistical significance. Prehospital treatment and ongoing hospital infusion of TXA did not result in improved survival or statistically significant functional outcomes at 24 months compared to placebo.
1. Introduction
1.1. Background
Despite improvements in trauma systems, injury remains the leading cause of death among young adults, even in settings with advanced trauma systems.1 Although there have been some improvements in-hospital mortality, the overall burden of injuries on premature deaths remains relatively unchanged.2 The leading cause of preventable death after trauma is hemorrhage.3,4
Acute traumatic coagulopathy or trauma-induced coagulopathy due to pathological hyperfibrinolysis is commonly observed in severely injured patients with significant hemorrhage5 and has been associated with greater blood loss and adverse clinical outcomes.6 Antifibrinolytic medications are therefore a logical consideration during the management of patients with critical bleeding. Tranexamic acid (TXA) is the most commonly available antifibrinolytic agent, and early administration of TXA is one of the few resuscitative interventions that have been proven in a randomized controlled trial to reduce mortality after trauma.7
The PATCH-Trauma randomized controlled trial assessed administration of TXA by prehospital clinicians and demonstrated greater survival at 28 days from injury among patients allocated to receive TXA.8 However, there was no difference in favorable functional outcomes at 6 months as measured by the dichotomized Glasgow Outcome Scale-Extended (GOS-E 1-4 compared with GOS-E 5-8), and the results suggested that most additional survivors in the TXA group were still severely disabled (GOS-E 3) at 6 months after injury.
1.2. Importance
One unanswered question from the PATCH-Trauma trial was whether 6 months was long enough to assess recovery after major trauma. There is some evidence that more than 6 months of physical and psychological rehabilitation may be required for patients to achieve their best possible recovery.9
1.3. Goals of This Investigation
The aim of this study was to assess 24-month functional outcomes among a subgroup of patients enrolled in the PATCH-Trauma trial.
2. Methods
This was a secondary analysis of a subgroup of patients enrolled in the PATCH-Trauma trial, whose methods have previously been reported.10 In brief, adult patients (≥18 years of age) were enrolled at the scene of their incident by 15 emergency medical services in Australia, New Zealand, and Germany. Patients were eligible for inclusion in the PATCH-Trauma trial if they were assessed as being at high risk for trauma-induced coagulopathy and if the first dose of TXA or placebo could be administered within 3 hours after injury and before hospital admission. Prehospital assessment of coagulopathy risk was performed with the use of the Coagulopathy of Severe Trauma (COAST) score, with a score of ≥3 considered to be high risk and eligible for enrollment. COAST scores range from 0 to 7, with 1 point assigned for each of the following variables: entrapment in a vehicle, systolic blood pressure of <100 mm Hg, body temperature of less than 35 °C, suspected pneumothorax, and suspected intraabdominal or pelvic injury. Additional points were assigned if the systolic blood pressure was <90 mmHg or if the body temperature was <32 °C.11
In the PATCH-Trauma trial, randomization was stratified by state jurisdictions. The subgroup included in this study comprised those patients enrolled in the state of Victoria, Australia. This subgroup was analyzed by a joint research collaborative between The Alfred and Royal Melbourne Hospitals, which are the only 2 adult major trauma services in Victoria, Australia. Between them, they received more patients enrolled in PATCH-Trauma than any other state in Australia. Patients (or the person responsible) who were alive at 6 months were contacted by telephone, and the GOS-E was recorded using a structured interview.
2.1. Intervention
Prehospital clinicians administered one dose of TXA (1 g) or placebo intravenously as a bolus as soon as practicable at the scene or en route to the hospital. After hospital arrival, the second ampule in the trial pack, containing TXA (1 g) or placebo, as originally randomized, was infused over a period of 8 hours. In addition to TXA or placebo, patients received usual prehospital, in-hospital, and posthospital care.
2.2. Outcomes
The primary outcome was a favorable functional outcome at 6 months, as assessed with the use of the GOS-E, administered as a standardized questionnaire by trained telephone interviewers.12 For this study, we extended the timeframe to assess GOS-E at 24 months using methodology consistent with the primary analysis. Favorable GOS-E was defined as scores of 5 to 8. We also assessed mortality at 24 hours, 28 days, 6 months, and 24 months.
2.3. Statistical Analysis
Baseline characteristics summarized using categoric variables were assessed using the chi-square test or Fisher’s exact test if the value in a cell was <5. Continuous variables with near-normal distribution were compared using Student’s t test, whereas skewed variables were compared using the Wilcoxon rank sum test. Outcomes of mortality and favorable GOS-E were presented using risk ratios and 95% CIs.
Furthermore, as longitudinal data of participants were likely to be correlated, we conducted an analysis of response profiles to test if there was a difference in the pattern of change in GOS-E from 6 months to 24 months between the TXA and placebo groups.13 This corresponded to an interaction. A mixed effects model was developed that included an effect of TXA (placebo as the reference group), an effect of time (with 6 months as the reference group), and an interaction between TXA and time. We did not include any random effects and assumed that the residuals within a patient were related and that the covariance matrix for the residuals within a patient was unstructured. We used restricted maximum likelihood for estimation and tested for parallelism between the change in GOS-E from 6 months to 24 months among the 2 groups. P values of <.05 were defined to be statistically significant. All analyses were conducted using Stata v 18.0, College Station, TX, USA. This secondary analysis was approved by The Alfred Hospital Human Research and Ethics Committee, with site-specific authorization from the Royal Melbourne Hospital Ethics Committee.
3. Results
The PATCH-Trauma trial enrolled 1310 patients with treatment allocation and consent for data valid for 1300 patients. Among these patients, 584 patients were enrolled in the state of Victoria, randomized within the strata of the state, and included in this study. A comparison of patients included in this subgroup, compared to the overall population, is presented in Supplementary Table 1. Baseline characteristics of patients were comparable between groups (Table 1). During the study period, there were no patients who received prehospital transfusion of plasma in Victoria.
Table.
Baseline characteristics.
| Variable | TXA (n = 297) | Placebo (n = 287) |
|---|---|---|
| Age—y; mean (SD) | 43.6 (SD 19.8) | 45.8 (SD 19.4) |
| Male sex | 202 (68.0%) | 190 (66.2%) |
| Mechanism of injury—n (%) | ||
| Blunt | 276 (92.9%) | 270 (94.1%) |
| Penetrating | 19 (6.4%) | 17 (5.9%) |
| Burn | 2 (0.7%) | 0 |
| Median initial heart rate (IQR)—beats/min | 106 (87-123) | 109 (90-124) |
| Initial systolic blood pressure—mm Hg | 76.0 (29.7) | 75.3 (32.2) |
| Initial body temperature—°C | 35.3 (1.3) | 35.4 (1.3) |
| Initial Glasgow Coma Scale—n (%) | ||
| 3-8 | 110 (37.0%) | 98 (34.2%) |
| 9-12 | 17 (5.7%) | 28 (9.8%) |
| 13-15 | 170 (57.2%) | 161 (56.1%) |
| Coagulopathy of Severe Trauma score—n (%) | ||
| <3 | 0 | 1 (0.4%) |
| 3 | 98 (33.0%) | 108 (37.6%) |
| 4 | 111 (37.4%) | 97 (33.8%) |
| 5 | 65 (21.9%) | 59 (20.6%) |
| 6 | 23 (7.7%) | 22 (7.7%) |
| Median injury severity score (IQR) | 30 (19-43) | 34 (21-43) |
| Score >2 on the abbreviated injury scale for head or neck—n (%) | 134 (46.1%) | 157 (54.0%) |
| Previous anticoagulant use—n (%) | 5 (1.9%) | 8 (3.3%) |
| Previous antiplatelet use—n (%) | 20 (7.6%) | 15 (6.2%) |
| Prerandomization red-cell transfusion—n (%) | 90 (30.3%) | 99 (34.5%) |
| Time from injury to first dose of tranexamic acid or placebo—median (IQR)—min | 88 (57-126) | 91 (65-120) |
There were 516 patients who had follow-up data available for 24 months after injury. Favorable GOS-E at 6 months and 24 months after injury are illustrated in Figure 1. At 24 months after injury, favorable GOS-E was reported in 167 (64.0%; 95% CI: 57.8-69.8) patients in the TXA group and 149 (58.4%; 95% CI: 52.1-64.5) in the placebo group (P = .20). For the analysis of response profiles, there was no TXA and time interaction (coeff −0.06; 95% CI: −0.30 to 0.19). The p value for the test of parallelism in the change from 6 to 24 months was 0.64. The distributions of GOS-E at 24 months in the TXA and placebo groups are illustrated in Figure 2.
Figure 1.
Favorable Glasgow Outcome Scale-Extended (GOS-E) at 6 months and 24 months after injury.
Figure 2.
GOS-E at 6 and 24 months after injury. GOS-E, Glasgow Outcome Scale-Extended, TXA, tranexamic acid.
At 24 hours after injury, there were 31 (10.4%; 95% CI: 7.2-14.5) deaths in the TXA group and 42 (14.6%; 95% CI: 10.8-19.3) deaths in the placebo group (P = .13). At 28 days after injury, there were 51 (17.2%; 95% CI: 13.1-22.0) deaths in the TXA group and 63 (22.0%; 95% CI: 17.3-27.2) deaths in the placebo group (P = .15). At 6 months after injury, 575 patients had outcome data available and there were 58 (19.9%; 95% CI: 15.4-24.9) deaths in the TXA group and 66 (23.3%; 95% CI: 18.5-28.7) deaths in the placebo group (P = .31). At 24 months, 516 patients’ data were available. Among patients randomized to the TXA group, there were 61/261 (23.4%; 95% CI: 18.4-28.9) deaths, and among patients randomized to placebo, there were 71/255 (27.8%; 95% CI: 22.4-33.8) deaths (P = .24). Mortality outcomes at the different timepoints are illustrated in Figure 3.
Figure 3.
Mortality at different timepoints. TXA, tranexamic acid.
4. Limitations
This study was limited to being a post hoc subgroup analysis of patients from the PATCH-Trauma trial. Although patients were randomized to TXA within this subgroup, the smaller sample size has reduced statistical power and as such is at risk of type 2 error. However, ongoing improvement in participants with severe disability at 6 months after injury was a possibility that could affect the interpretation of the PATCH-Trauma results. The population enrolled from Victoria was all transported by helicopter emergency medical services with long prehospital times, and, therefore, the results cannot be readily translated to patients experiencing shorter prehospital times. This subgroup also had greater injury severity, and more patients had traumatic brain injury.
5. Discussion
At 24 months after injury, there were no differences in the proportion of patients with favorable functional outcomes among groups randomly assigned to TXA or placebo. Improvements in functional outcomes from 6 to 24 months were observed for the overall cohort but were not significantly different between the 2 groups. Furthermore, in this subgroup of patients from the PATCH-Trauma trial, a difference in mortality between the TXA and placebo groups was not demonstrated at any timepoint up to 24 months after injury.
Following the CRASH-2 trial that demonstrated a significant mortality benefit following TXA administration for trauma, no other trial investigating TXA for trauma has demonstrated a difference for the primary outcome in a multitrauma population. In the STAAMP trial, patients who received TXA compared with placebo in the prehospital setting did not have a significantly lower rate of 30-day mortality (8.1% vs 9.9%).14 For the primary outcome of the PATCH-Trauma trial, prehospital administration of TXA followed by an infusion over 8 hours did not result in a greater number of patients surviving with a favorable functional outcome at 6 months than placebo (risk ratio, 1.00; 95% CI: 0.90-1.12).8
In this follow-up study, we demonstrate that while there was ongoing recovery of the study cohort, the recovery rate among patients assigned to the TXA group was not significantly different. Thus, in the setting of critical bleeding, the core principles of recognition and early hemorrhage control remain essential for resuscitation teams. Following this, the maintenance of perfusion through transfusion of blood or components is physiologically supported for the maintenance of tissue perfusion. We continue to recommend that, as an adjunct therapy, early TXA should be considered, given the low cost, low rates of adverse events, and signals of benefit for survival. This is based on the potential of benefit demonstrated in the CRASH-2 trial and secondary outcome measures of the PATCH-Trauma trial and the STAAMP trial.
Following major trauma, persisting disability among survivors over the long term has been highlighted.15, 16, 17 However, longitudinal studies in posttrauma patients assessing long-term outcomes are uncommon. Providing estimates of the nonfatal injury burden is therefore often overlooked when assessing outcomes after trauma, and few registries or trials have assessed this measure. Where assessed, adverse long-term outcomes appear to be amplified in low- and middle-income countries, reinforcing the need to consider long-term function before treatment decisions.18 No other randomized trial of TXA has reported long-term outcomes.
Given the trajectory of recovery, assuming a similar pattern was maintained over time, it is possible that early TXA could result in better functional recovery at a longer time point. It is possible that the observed numerical difference in long-term favorable outcome, if consistent with the full study protocol, may have reached statistical significance if long-term outcomes were available for the complete data set.
The best patient-centered outcome for trauma resuscitation continues to be debated.19,20 The single intervention of prehospital TXA is postulated to improve trauma-induced coagulopathy and, following on, early mortality. However, there are potential effects on longer-term outcomes that are mediated by effects, such as immunomodulation, unrelated to coagulopathy. The importance of survival alone vs the quality of survival may well be an individual choice that cannot be determined during trauma resuscitation. Faced with this uncertainty, it appears that clinicians prefer to prioritize survival alone, given the persistence and expansion of TXA into guidelines for critical bleeding. With varied individual preferences in any population, patient and public involvement through consumer engagement would be valuable in weighing up these possible outcomes.
Prehospital TXA with ongoing infusion over 8 hours did not result in improved functional outcomes at 24 months compared to placebo. It is possible that differences in functional outcomes might be observed at a longer time point.
Author Contributions
BM and RG conceived this substudy. RG, BM, SB, MR, PM, and PC designed the trial and obtained research funding. RG supervised the conduct of the trial and data collection. SM, EB, RA, and AG undertook data collection and managed the data, including quality control. BM and PM provided statistical advice on study design and analyzed the data. BM drafted the manuscript, and all authors contributed substantially to its revision. BM takes responsibility for the paper as a whole.
Funding and Support
The PATCH-Trauma trial was supported by grants from the Australian National Health and Medical Research Council (NHMRC ID1044894 and ID165275), the Health Research Council of New Zealand (GA216F), the New Zealand Lottery Grants Board (34033), and the German Research Funding Agency (Deutsche Forschungsgemeinschaft) (MA 2569/6-1). P. Myles is funded by an Australian NHMRC Investigator Grant (ID2008079).
Conflict of Interest
All authors have affirmed they have no conflicts of interest to declare.
Acknowledgments
We would like to thank staff at The Royal Melbourne Hospital Trauma registry and The Alfred Intensive Care Unit for identification and long-term follow-up of patients.
Footnotes
Supervising Editor: Nicholas Caputo, MD, MSc
Supplementary material associated with this article can be found in the online version at https://doi.org/10.1016/j.acepjo.2026.100417.
Supplementary Materials
References
- 1.Choi J., Carlos G., Nassar A.K., Knowlton L.M., Spain D.A. The impact of trauma systems on patient outcomes. Curr Probl Surg. 2021;58(1) doi: 10.1016/j.cpsurg.2020.100840. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lockey D.J. Improved trauma outcomes after the introduction of a trauma system in England. eClinicalMedicine. 2018;2:3–4. doi: 10.1016/j.eclinm.2018.08.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Callcut R.A., Kornblith L.Z., Conroy A.S., et al. The why and how our trauma patients die: a prospective Multicenter Western Trauma Association study. J Trauma Acute Care Surg. 2019;86(5):864–870. doi: 10.1097/TA.0000000000002205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kalkwarf K.J., Drake S.A., Yang Y., et al. Bleeding to death in a big city: an analysis of all trauma deaths from hemorrhage in a metropolitan area during 1 year. J Trauma Acute Care Surg. 2020;89(4):716–722. doi: 10.1097/TA.0000000000002833. [DOI] [PubMed] [Google Scholar]
- 5.Carroll R.C., Craft R.M., Langdon R.J., et al. Early evaluation of acute traumatic coagulopathy by thrombelastography. Transl Res. 2009;154(1):34–39. doi: 10.1016/j.trsl.2009.04.001. [DOI] [PubMed] [Google Scholar]
- 6.Cardenas J.C., Wade C.E., Cotton B.A., et al. TEG lysis shutdown represents coagulopathy in bleeding trauma patients: analysis of the PROPPR cohort. Shock. 2019;51(3):273–283. doi: 10.1097/SHK.0000000000001160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.CRASH-2 trial collaborators. Shakur H., Roberts I., Bautista R., et al. 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. doi: 10.1016/S0140-6736(10)60835-5. [DOI] [PubMed] [Google Scholar]
- 8.PATCH-Trauma Investigators and the ANZICS Clinical Trials Group. Gruen R.L., Mitra B., Bernard S.A., et al. Prehospital tranexamic acid for severe trauma. N Engl J Med. 2023;389(2):127–136. doi: 10.1056/NEJMoa2215457. [DOI] [PubMed] [Google Scholar]
- 9.Gabbe B.J., Simpson P.M., Cameron P.A., et al. Long-term health status and trajectories of seriously injured patients: a population-based longitudinal study. PLoS Med. 2017;14(7) doi: 10.1371/journal.pmed.1002322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Mitra B., Bernard S., Gantner D., et al. Protocol for a multicentre prehospital randomised controlled trial investigating tranexamic acid in severe trauma: the PATCH-Trauma trial. BMJ Open. 2021;11(3) doi: 10.1136/bmjopen-2020-046522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mitra B., Cameron P.A., Mori A., et al. Early prediction of acute traumatic coagulopathy. Resuscitation. 2011;82(9):1208–1213. doi: 10.1016/j.resuscitation.2011.04.007. [DOI] [PubMed] [Google Scholar]
- 12.Gabbe B.J., Cameron P.A., Hannaford A.P., Sutherland A.M., McNeil J.J. Routine follow up of major trauma patients from trauma registries: What are the outcomes? J Trauma. 2006;61(6):1393–1399. doi: 10.1097/01.ta.0000225926.34180.47. [DOI] [PubMed] [Google Scholar]
- 13.Modeling the mean: analyzing response profiles. Applied Longitudinal Analysis. 2011:105–141. [Google Scholar]
- 14.Guyette F.X., Brown J.B., Zenati M.S., et al. Tranexamic acid during prehospital transport in patients at risk for hemorrhage after injury: a double-blind, placebo-controlled, randomized clinical trial. JAMA Surg. 2020;156(1):11–20. doi: 10.1001/jamasurg.2020.4350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gabbe B.J., Harrison J.E., Lyons R.A., Jolley D. Modelling long term disability following injury: comparison of three approaches for handling multiple injuries. PLoS One. 2011;6(9) doi: 10.1371/journal.pone.0025862. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tan A.L., Chiong Y., Nadkarni N., Cheng J.Y.X., Chiu M.T., Wong T.H. Predictors of change in functional outcome at six months and twelve months after severe injury: a retrospective cohort study. World J Emerg Surg. 2018;13:57. doi: 10.1186/s13017-018-0217-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Meakes S., Enninghorst N., Weaver N., Hardy B.M., Balogh Z.J. Long-term functional outcomes in polytrauma: a fundamentally new approach is needed in prediction. Eur J Trauma Emerg Surg. 2024;50(4):1439–1452. doi: 10.1007/s00068-023-02430-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ahmed A.N., Lysaght R., Addissie A., Zewdie A., Finlayson M. One-year outcomes of traumatic injuries among survivors in Ethiopia: a cross-sectional study on the employment outcomes and functioning state. Trauma Surg Acute Care Open. 2024;9(1) doi: 10.1136/tsaco-2023-001209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kaufman E.J., Richmond T.S., Wiebe D.J., Jacoby S.F., Holena D.N. Patient experiences of trauma resuscitation. JAMA Surg. 2017;152(9):843–850. doi: 10.1001/jamasurg.2017.1088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Laura N.G., Aaron R.J., Deborah M.S., Coalition for National Trauma Research Scientific Advisory Council Patient-centered outcomes research and the injured patient: a summary of application. Trauma Surg Acute Care Open. 2020;5(1) doi: 10.1136/tsaco-2019-000422. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.