The SWiFT trial, published in the New England Journal of Medicine in March 2026, found that up to two units of whole blood given to bleeding trauma patients before hospital did not reduce death or massive transfusion at 24 hours compared with red cells and plasma given separately. It is the largest randomised trial of pre-hospital whole blood so far, and it was run entirely in UK air ambulance services, so it speaks directly to UK practice.
Why this matters
Whole blood is blood from a single donor kept as one product, rather than split into red cells, plasma and platelets. Military teams have used it, and observational studies of pre-hospital red cell transfusion had reported a 12 to 14 per cent absolute reduction in 30-day mortality. These were not whole blood studies.
UK practice was mixed. A 2020 survey of UK air ambulance services found 67 per cent gave red cells plus plasma, 22 per cent red cells only and 11 per cent plasma only. More than 80 per cent said whole blood would be their preferred product. The only previous randomised trial of whole blood in trauma had 107 patients.
What the researchers did
SWiFT (Study of Whole Blood in Frontline Trauma) was a pragmatic, phase 3, open-label randomised controlled trial. Pragmatic means real working conditions; open-label means crews knew which product they were giving.
Ten air ambulance services in England took part. Recruitment opened on 15 December 2022 and ran for about two years. NHS Blood and Transplant led the trial, with Professors Jason Smith and Laura Green as co-chief investigators, funded with the Ministry of Defence and the participating air ambulance charities.
Patients of any age with a traumatic injury who needed a pre-hospital transfusion under their service’s usual criteria were eligible, unless there was no vascular access, a known objection to blood products, or blood had already been given before the air ambulance arrived.
Randomisation was by pre-packed blood boxes; each aircraft carried only the box it was allocated. The whole blood arm received up to two units of leukocyte-depleted whole blood, meaning white cells had been filtered out, stored at 2 to 6°C for up to 21 days. The control arm received up to two units of red cells and two units of plasma. Both groups then received normal hospital care.
The primary outcome was a composite of death from any cause or massive transfusion within 24 hours. Massive transfusion meant ten or more units of any blood component in adults, or 40 mL/kg in children. In total 942 patients were randomised. After excluding 225 in traumatic cardiac arrest, 33 with non-traumatic bleeding and some withdrawals, 616 patients had primary outcome data: 314 given whole blood and 302 given standard care. These were very sick patients, with a median Injury Severity Score (a scale of anatomical injury, higher being worse) of 33 and 34, and a median heart rate of 110 in both arms. Some 71 per cent had blunt trauma and 21 were children.
What they found
The primary outcome occurred in 48.7 per cent of the whole blood group (153 of 314) and 47.7 per cent of the standard care group (144 of 302). The adjusted relative risk was 1.02, with a 95 per cent confidence interval of 0.80 to 1.31 and a P value of 0.84. The confidence interval rules out any large benefit.
Neither part of the composite differed on its own. Massive transfusion occurred in 44.9 per cent versus 42.4 per cent. Death at 24 hours was 10.1 per cent versus 10.0 per cent, at 30 days 27.5 per cent versus 26.0 per cent, and at 90 days 30.4 per cent versus 27.2 per cent.
One finding stood out. A prothrombin time above the normal range on hospital arrival, a sign of impaired clotting, was more common after whole blood: 40.7 per cent versus 30.5 per cent, relative risk 1.31 (95 per cent CI 1.10 to 1.56). The authors reported it did not affect the primary outcome.
Safety was similar. Serious adverse events occurred in 24 patients (7.3 per cent) given whole blood and 31 patients (9.9 per cent) given standard care. Thrombotic events within 30 days were 14.2 per cent versus 15.9 per cent. Median scene time was 38 minutes and median transport time 28 minutes.
Strengths and limitations
This is the first large randomised trial of whole blood in trauma. The groups were well balanced, the patients were clearly shocked, and both parts of the composite pointed the same way. For UK helicopter emergency medical services (HEMS), the results apply directly because the trial was run in those systems.
The limitations are real. Two units is a small dose, and both groups then received identical components in hospital. The sample size assumed whole blood would cut the composite outcome from 68 per cent to 56 per cent. Commentators have called that optimistic, so a smaller true benefit cannot be ruled out. The trial was unblinded. Around a third of randomised patients were excluded from the main analysis, mostly for traumatic cardiac arrest, so the sickest were left out.
Few offshore, diving or remote settings carry blood at all, so the trial does not change what is available to you. It does suggest the product matters less than getting an appropriate transfusion started quickly and safely.
What it means for your practice
- All roles: do not change practice on the basis of this single trial. Local protocols, JRCALC and your service’s transfusion governance remain the standard. SWiFT shows neither product is better.
- HEMS and critical care paramedics: if your service carries whole blood, whether to continue is a governance decision, not an urgent safety issue. The trial supports a focus on scene time, haemorrhage control and rapid transfer to definitive care.
- Ambulance clinicians and first responders: you will rarely give blood, but you shape the timeline. Early recognition of major haemorrhage, early request for an enhanced care team, external haemorrhage control and vascular access all shorten the time to transfusion, whatever the product.
- Offshore and diver medics: blood is unlikely to be in your kit. Your contribution is bleeding control, preventing hypothermia, structured handover and a clear evacuation plan.
Reflect
- What does my service carry for major haemorrhage, and what is the evidence behind that choice?
- Where in my own practice could I shorten the time from injury to transfusion, even if I do not give blood myself?
- How would I explain to a colleague why a well-run trial with a null result still matters?
- What would I want to see in a future trial before accepting that whole blood should, or should not, replace components?
Read the original
- New England Journal of Medicine: Prehospital Whole Blood in Traumatic Hemorrhage, a Randomized Controlled Trial (SWiFT Trial Group), 2026
- Trials: The SWiFT trial study protocol for a multi-centre randomised controlled trial, 2023
- Critical Care Reviews: SWiFT trial summary and critical appraisal, 2026
- St Emlyn’s: Journal club on prehospital whole blood in traumatic haemorrhage, the SWiFT trial, 2026
- First10EM: No benefit from whole blood, the SWiFT trial, 2026
- London’s Air Ambulance Charity: Results published from the SWiFT trial, 2026
- Great Western Air Ambulance Charity: SWiFT trial published, 2026
- AABB News: SWiFT trial finds no advantage to prehospital whole blood in patients with life-threatening trauma, 2026
Published 25 October 2026. Last reviewed October 2026.
These summaries are for education and CPD only. They are not clinical guidelines. Always practise within your scope of practice and follow your employer’s protocols and national guidance such as JRCALC and Resuscitation Council UK.