Whole blood has become one of the most talked-about shifts in prehospital trauma care — a single bag offering balanced resuscitation with simpler logistics than juggling separate components. The observational data looked promising, but the randomised evidence was thin. That’s now changed. Within two months of each other, two large trials on opposite sides of the Atlantic — the UK’s SWiFT trial and the US TOWAR trial — put prehospital whole blood to the test against standard blood components, and both landed in the same place. Here’s what they found, and what it means for how we resuscitate bleeding patients before they reach hospital. This post summarises the findings from the two trials, and discusses what we have learned, and what we still don’t know, about pre-hospital whole blood.
SWiFT: Prehospital whole blood was no better than blood components
Source: Smith JE, Cardigan R, Sanderson E, et al., for the SWiFT Trial Group. Prehospital Whole Blood in Traumatic Hemorrhage — a Randomized Controlled Trial. N Engl J Med. Published online 17 March 2026. doi: 10.1056/NEJMoa2516043
Bottom line
In the UK SWiFT trial, up to 2 units of prehospital whole blood was not superior to red cells plus plasma for the composite of death or massive transfusion at 24 hours, and the safety profile was similar. At this dose, the case for whole blood is a logistic one, not a survival advantage.
What they did
A pragmatic, phase 3, unblinded, multicentre superiority RCT across 10 English air ambulance services. Patients of any age with major traumatic haemorrhage needing prehospital transfusion were randomised to up to 2 units of leukocyte-depleted whole blood or standard care — up to 2 units each of red cells and plasma. Sealed, identical blood boxes meant crews couldn’t pre-select. The primary outcome was a composite of death from any cause or massive transfusion within 24 hours. Of 942 randomised, 641 formed the modified intention-to-treat population after excluding traumatic cardiac arrest and non-traumatic haemorrhage.
What they found
A primary-outcome event occurred in 48.7% of the whole-blood group versus 47.7% of standard care (adjusted relative risk 1.02, 95% CI 0.80–1.31; P=0.84). Death at every timepoint out to 90 days, massive transfusion, and the other clinical secondary outcomes were similar, and the per-protocol and subgroup analyses agreed. Prothrombin time was more often above normal with whole blood (40.7% vs 30.5%, RR 1.31, 95% CI 1.10–1.56), but this didn’t translate into any clinical difference — most likely the older plasma in stored whole blood. Safety was comparable, with no whole-blood signal.
Caveats
This was a superiority trial, so “not superior” is the honest read — it doesn’t prove the two are equivalent, though the tight confidence interval around 1.0 and the consistency across outcomes argue against a large missed benefit. The pragmatic design left the decision to transfuse to clinical judgement, adding heterogeneity and probably including some patients without truly life-threatening bleeding. The authors note that up to 2 units may be too small a dose to reveal a difference, and a 2023 cyberattack on the national trauma database cost some data.
How this changes practice
For services already carrying whole blood, the reassurance is real — it performs as well as components and looks just as safe, while keeping the practical advantages of a single bag: simpler logistics, faster administration, fewer transfusion errors, and easier use when access is limited. For services weighing it up, SWiFT reframes the decision around logistics, supply, and cost rather than a survival edge. It doesn’t address higher-dose whole blood or specific subgroups, which is where the open questions now sit.
TOWAR: Prehospital whole blood didn’t lower mortality either
Source: Sperry JL, Guyette FX, Cotton BA, et al., for the TOWAR Study Group. Prehospital Resuscitation with Type O Whole Blood for Trauma and Hemorrhage. N Engl J Med. Published online 18 May 2026. doi: 10.1056/NEJMoa2602167
Screenshot
Bottom line:
In the US TOWAR trial, prehospital low-titre group O whole blood did not reduce 30-day mortality compared with blood components in injured patients with haemorrhagic shock. Stored whole blood up to 21 days old performed no differently from fresher units. Coming weeks after the UK SWiFT trial, two large RCTs now point the same way.
What they did
A pragmatic, phase 3, cluster-randomised trial across 44 US air medical bases, assigned 2:1 in monthly blocks to carry either up to 2 units of low-titre group O whole blood or as-indicated components (red cells, plasma, or both). Eligible patients had hypotension (SBP ≤90) with tachycardia (HR ≥108), or severe hypotension (SBP ≤70). The primary outcome was death from any cause within 30 days. A prespecified observational substudy compared whole blood stored 15–21 days against 1–14 days. Of 1020 enrolled, 993 entered the primary analysis (695 whole blood, 298 components).
What they found
30-day mortality was 25.9% with whole blood versus 20.5% with components (adjusted odds ratio 1.24, 95% CI 0.87–1.76; P=0.24) — no significant difference, with a confidence interval spanning both directions. Mortality at 3, 6 and 24 hours, in-hospital death, organ failure, ARDS, transfusion volumes and coagulation measures were all similar. The storage-age substudy found no difference (27.1% for 15–21 days vs 26.4% for 1–14 days; adjusted OR 0.99, 95% CI 0.74–1.32). Adverse events were comparable.
Caveats
This is the honest part. The point estimate sat numerically against whole blood, but it wasn’t significant and the trial cannot exclude a benefit or a harm — it isn’t a clean “no difference.” The authors flag several features that push toward a type II error: substantial crossover (about a third of the whole-blood group also received components, and some component patients got whole blood), low prehospital transfusion volumes relative to in-hospital, and an unmasked, cluster-randomised design. Pre-randomisation transfusion was allowed. So the two arms may have been more alike than the labels suggest.
How this changes practice
TOWAR doesn’t overturn the case for prehospital blood — early transfusion still beats delay, and overall mortality has fallen markedly since the 2018 PAMPer era. What it does, alongside SWiFT, is undercut the idea that whole blood specifically delivers a survival edge over components at these prehospital doses. For services running whole blood, the reassurance holds: it’s safe, it works as well, and stored units up to three weeks old are fine — so the rationale stays logistic. The unsettled question is whether higher volumes, earlier delivery, or specific subgroups would tell a different story.
So in summary:
Two trials, on different continents, with different designs, arriving at the same answer: at the doses tested, prehospital whole blood didn’t lower mortality compared with blood components. That convergence matters more than either result alone. It doesn’t argue against prehospital transfusion — getting blood to bleeding patients early still saves lives — but it does reframe whole blood as a question of logistics, supply and cost rather than a survival advantage. For services already carrying it, both trials are reassuring: it’s safe, it works as well as components, and stored units hold up. The open questions now sit elsewhere — higher transfusion volumes, earlier delivery, and whether specific subgroups might still benefit. For now, the message is that whole blood is a reasonable and practical option, but not the decisive advantage many had hoped for.
Also, as highlighted by Justin Morgenstern over at First10EM, it’s curious as to why these trials were done prehospital and not in hospital, in the setting of patients receiving massive transfusion, where one might expect to see a more notable effect/difference before being taken into the prehospital scene, and only giving a couple of units. To quote Justin: “We don’t know if whole blood has any effect, but even if it does, it is extremely unlikely that the effect will be seen from just the first 2 units”.