SWiFT and TOWAR, published together in NEJM this June, randomized nearly 2,000 bleeding trauma patients between whole blood and blood components in the air. Neither found whole blood superior, and that changes how we should argue for our blood programs.
Educational reading for clinicians. Follow your current protocols and local guidance.
The Trials We Were Waiting For
I've made the case for prehospital whole blood in rural EMS before, and I'd make most of it again. But the argument always leaned on physiology and military experience more than on big civilian randomized trials. In June 2026, the New England Journal of Medicine published two of them in the same issue. Both asked the same question: in a bleeding trauma patient, does whole blood before the hospital beat blood components?
Both said no.
SWiFT (England)
SWiFT was a pragmatic, phase 3, multicenter trial across 10 air ambulance services in England. Patients with major traumatic hemorrhage got either up to 2 units of whole blood or standard care with up to 2 units each of red cells and plasma before arrival.
- 942 patients were randomized. After excluding nontraumatic hemorrhage and traumatic cardiac arrest, 616 were analyzed.
- The primary outcome, death from any cause or massive transfusion (≥10 units) within 24 hours, occurred in 48.7% with whole blood and 47.7% with standard care (relative risk 1.02; 95% CI 0.80 to 1.31; P = 0.84).
- Prothrombin time was above normal in 40.7% of the whole-blood group versus 30.5% with components.
- Thrombotic events looked similar. Serious adverse events: 31 with whole blood, 37 with components.
TOWAR (United States)
TOWAR cluster-randomized 44 U.S. air medical bases, 2:1, to whole blood or as-indicated components (plasma, red cells or both) in one-month blocks, again capped at 2 units.
- 1,020 eligible patients; 695 whole blood and 298 components in the primary analysis.
- 30-day mortality was 25.9% with whole blood and 20.5% with components (adjusted OR 1.24; 95% CI 0.87 to 1.76; P = 0.24).
- An observational substudy found no difference by storage age: 27.1% mortality with whole blood stored 15 to 21 days versus 26.4% with 1 to 14 days.
| SWiFT | TOWAR | |
|---|---|---|
| Setting | 10 English air ambulance services | 44 U.S. air medical bases |
| Design | Individually randomized | Cluster-randomized, 2:1 |
| Analyzed | 616 | 993 |
| Primary outcome | Death or massive transfusion at 24 h | Death at 30 days |
| Whole blood vs components | 48.7% vs 47.7% | 25.9% vs 20.5% |
| Result | Not superior (P = 0.84) | Not superior (P = 0.24) |
What This Does and Doesn't Say
It doesn't say blood in the field is useless. TOWAR's own background states that transfusion before hospital arrival reduces mortality from traumatic hemorrhage. Both trials compared one blood strategy against another, not blood against crystalloid.
It doesn't say whole blood is dangerous. TOWAR's point estimate leaned toward components, but the confidence interval runs from benefit to harm. Reading that as "whole blood kills" is the same mistake as reading the old observational data as "whole blood saves."
It does take away our strongest outcome claim. If you've been pitching your county or medical director on whole blood because it saves more lives than components, two randomized trials now say we can't show that.
It's air medical, and it's two units. Neither trial studied ground crews, long rural transports or larger volumes. That's the population where I think the logistics argument is strongest, and it's still unstudied.
How I'd Argue for a Program Now
The case for whole blood moves from outcomes to operations. One bag instead of two, platelets included, fewer products to rotate through a two-truck service. TOWAR's storage data helps there too: blood stored into its third week did no worse, which matters for rotation and waste.
Components are now an evidence-equivalent option. If your blood bank partner can supply red cells and plasma more easily than low-titer O whole blood, that's a defensible program. And if a shelf-stable product fits better still, freeze-dried plasma just got its first U.S. license.
What doesn't change is the clock. Stop the bleeding you can see, give TXA early, get product in and get them to a surgeon.
Sources
Jonathan B, career paramedic
Published October 23, 2026