Heterogeneity and confounding in the TRAUMOX2 randomised clinical trial—a reply

We appreciate the opportunity to respond to the editorial commentary regarding our trial (1).

The trial population was intentionally designed to be heterogenous, as the Advanced Trauma Life Support (ATLS) guidelines recommend that all severely injured trauma patients must receive supplemental oxygen in the initial phase after trauma (2). Thus, this universal oxygen approach was based on a severity assessment by the attending clinician, rather than sustained injuries, as the ATLS guidelines do not provide oxygen recommendations for specific anatomical lesions. Furthermore, because the evidence supporting this recommendation was extremely limited and our trial was the first major trial to address this research question, we found it crucial to adhere to the guideline’s rationale by applying the oxygen strategies to all severely injured trauma patients, irrespective of the specific lesions.

Importantly, we conducted a randomised controlled trial (3). Within the hierarchy of generating new clinical evidence, this study design is considered the most robust for evaluating interventions compared to other designs, as it minimises bias and accounts for both known and potential unknown confounding factors between the groups (4). Several remarks in the editorial related to issues commonly seen in observational studies, where groups receiving different interventions could be imbalanced and were subject to various forms of confounding, such as confounding by indication. These concerns are far less relevant in randomised controlled trials, which is why we used this methodology.

The editorial expressed concern about the lack of haemoglobin data. Nonetheless, haemoglobin levels were reported in the supplementary material as we agree this could impact the outcomes. The haemoglobin levels measured at two time points during the intervention, as shown in Suppl. 2 (Tab. S3), were similar between the groups. Furthermore, this has also been highlighted in a previous reply letter, as a post-hoc analysis of haemoglobin levels showed no effect on the primary outcome (5). Finally, it is important to note that baseline characteristics were generally well balanced between the groups, including the proportion of patients in shock (systolic blood pressure <90 mmHg). This supports the validity of the group comparisons, as no substantial systematic differences in patient characteristics were observed.

The editorial additionally noted that 30-day mortality was primarily used in drug-safety studies. Nevertheless, the 30-day endpoint is widely considered as a robust and clinically relevant outcome. It is superior to discharge-based observation time periods because of its standardised follow-up period, which avoids the variability in hospital length of stay that could introduce bias by underestimating event rates, as death and major respiratory complications may still occur after discharge and up to day 30. Furthermore, it is a key outcome across multiple medical disciplines, well beyond drug safety, including trials of plasma and tranexamic acid for trauma patients (6,7), and it is an Utstein core outcome for cardiac arrest (8). While our primary outcome of death and/or major respiratory complications (pneumonia and/or acute respiratory distress syndrome) within 30 days may indeed be affected by several factors, it is unlikely that these influences would differ systematically between the two randomised intervention groups.

We recognise that injury type and location may influence the intervention effect. All Abbreviated Injury Scale (AIS) scores used to calculate the Injury Severity Scores (ISS) were available to the investigators and formed the basis for the subgroup analyses of patients with moderate to severe injury (ISS >15) and for patients with traumatic brain injury (TBI) (3). No significant difference or interaction between groups was observed. We agree about the importance of this question, which is why dedicated manuscripts addressing TBI and thoracic injuries separately are currently being prepared. The protocols for both studies are available at the trial website www.traumox2.org.

In the editorial, it is stated that approximately 20% of patients in both groups required surgery at any point. This reflects a clear misunderstanding of our data, as this proportion represents the need for acute surgery immediately following trauma centre resuscitation. We did not present data on surgeries performed subsequently during hospital stay.

The Mega-ROX trial, including 40,000 intensive care unit (ICU) patients, including a subset of trauma patients, to differentiated oxygen strategies, will potentially be able to detect between-group differences of 1% to 2%, and provide insightful new knowledge to this research area (9). However, several differences in the Mega-ROX trial design limit direct comparability with our trial. We attempted to mimic the ATLS guidelines by administering the oxygen strategies as early as possible to all severely injured trauma patients, indicated by including 41% of the population in the prehospital setting, for a duration of 8 hours to reflect the acute trauma phase. In contrast, the Mega-ROX trial will intervene subsequently to trauma resuscitation over a longer period as mechanically ventilated ICU patients will be included, with the intervention continuing until day 90 or ICU discharge, whichever occurs first.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Annals of Translational Medicine. The article did not undergo external peer review.

Funding: This work was supported by Novo Nordisk Foundation (grant No. NNF20OC0063985).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://atm.amegroups.com/article/view/10.21037/atm-2025-1-169/coif). J.B. received grants from Novo Nordisk Foundation and Lundbeck Foundation outside the submitted work. J.S. received grants from the Norwegian Air Ambulance Foundation and Novo Nordisk Foundation outside the submitted work. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Mackenzie CF. Early restrictive versus liberal oxygen for trauma patients: does it make a difference? Ann Transl Med 2025;13:12. [Crossref] [PubMed]
2. American College of Surgeons: The Committee on Trauma. ATLS: Advanced Trauma Life Support (Student Course Manual), 10th Edition. Published online 2018.
3. Arleth T, Baekgaard J, Siersma V, et al. Early Restrictive vs Liberal Oxygen for Trauma Patients: The TRAUMOX2 Randomized Clinical Trial. JAMA 2025;333:479-89. [Crossref] [PubMed]
4. Murad MH, Asi N, Alsawas M, et al. New evidence pyramid. Evid Based Med 2016;21:125-7. [Crossref] [PubMed]
5. Baekgaard J, Siersma V, Steinmetz J. Early Restrictive vs Liberal Oxygen for Trauma Patients-Reply. JAMA 2025;334:181-2. [Crossref] [PubMed]
6. Sperry JL, Guyette FX, Brown JB, et al. Prehospital Plasma during Air Medical Transport in Trauma Patients at Risk for Hemorrhagic Shock. N Engl J Med 2018;379:315-26. [Crossref] [PubMed]
7. Guyette FX, Brown JB, Zenati MS, 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;156:11-20. Erratum in: JAMA Surg 2021;156:105. [Crossref] [PubMed]
8. Ringdal KG, Coats TJ, Lefering R, et al. The Utstein Trauma Template for Uniform Reporting of Data following Major Trauma: Data Dictionary. Version 1.1.1. European Trauma Registry Network, 2008. Available online: https://rcsyd.se/swetrau/wp-content/uploads/sites/10/2021/01/Revised-Utstein-Template-%E2%80%93-Data-Dictionary-v1.1.1.pdf
9. Young PJ, Arabi YM, Bagshaw SM, et al. Protocol and statistical analysis plan for the mega randomised registry trial research program comparing conservative versus liberal oxygenation targets in adults receiving unplanned invasive mechanical ventilation in the ICU (Mega-ROX). Crit Care Resusc 2022;24:137-49. [Crossref] [PubMed]