Start with the bleeding patient
A 36-year-old man arrives intubated after a motorcycle crash: temperature 36 °C, pulse 120/min, BP 70/40 mmHg, distended abdomen and positive FAST. The immediate problem is probable intra-abdominal haemorrhage with profound shock. Activate the major-haemorrhage protocol and mobilise definitive source control. Do not wait for coagulation results or repeat imaging when intervention is already indicated. FAST neither identifies every injury nor excludes bleeding elsewhere. NICE trauma guidance.
This review concerns adults with major traumatic bleeding. Paediatric and obstetric haemorrhage need separate protocols.
Why a clot fails
Trauma-induced coagulopathy can begin before substantial fluids or the full lethal triad. Injury and hypoperfusion disturb endothelial signalling and clot regulation; blood loss, dilution, cold and acidosis amplify the problem. A temperature of 36 °C does not exclude it. Early trauma-coagulopathy research.
A normal platelet count does not guarantee normal platelet function. Fibrinolysis also varies: some patients break down clot rapidly, while others show reduced measurable breakdown. These observational findings explain heterogeneity; they do not justify withholding indicated early TXA or allowing reassuring tests to overrule ongoing bleeding. Platelet-function study; fibrinolysis phenotypes.
Resuscitate and control the source together
Control external bleeding, apply a pelvic binder when indicated, obtain rapid vascular access and alert the blood bank. In the intubated patient, confirm waveform capnography, reassess ventilation and keep looking for reversible causes of shock. An initially reassuring chest radiograph does not end reassessment.
Send a blood count, PT/INR, aPTT, fibrinogen, blood gas with lactate/base deficit, ionised calcium and compatibility samples. Initial haemoglobin may underestimate the significance of acute blood loss. Begin the local balanced blood-product strategy promptly; do not await a low haemoglobin. Adult critical-bleeding guidance.
In PROPPR, a plasma:platelets:red-cell strategy of 1:1:1 improved haemostasis and reduced early exsanguination deaths compared with 1:1:2, without a significant difference in the primary all-cause mortality outcomes. Follow blood-bank packs: an adult platelet dose is not one bag equivalent to one red-cell unit. PROPPR; product-dose guidance.
Low-titre group O whole blood is an option within an established programme, not proven universally superior. The 2026 prehospital SWiFT and TOWAR trials did not show superiority on their respective primary outcomes; they do not settle every in-hospital transfusion question.
Avoid repeated litres of crystalloid, which replace neither oxygen-carrying capacity nor haemostatic components. NICE advises against crystalloid for active in-hospital bleeding. If blood is unavailable, use the emergency bridging protocol while urgently obtaining blood and source control.
Before haemostasis, the European guideline supports temporary SBP 80–90 mmHg in selected bleeding adults without brain injury. Do not apply this automatically to suspected TBI or spinal cord injury. ACS TBI guidance uses initial SBP ≥110 mmHg, then individualised cerebral-perfusion goals.
Protect temperature, perfusion and calcium
Prevent exposure and heat loss, remove wet clothing, actively warm the patient and monitor core temperature. Warm fluids and appropriate blood products using approved equipment and product instructions. Cold impairs platelet and coagulation function; acidemia can amplify this impairment. Correcting the pH number alone cannot restore oxygen delivery or stop haemorrhage. Experimental clotting evidence.
Citrate in transfused blood binds calcium; shock can impair clearance. Hypocalcaemia can also precede transfusion and impairs coagulation and cardiac contraction. Monitor ionised calcium and maintain the normal range, approximately 1.1–1.3 mmol/L, using the local regimen and repeat measurements. European guideline.
Calcium chloride and calcium gluconate are not interchangeable gram for gram. Avoid unmonitored overcorrection. The best dosing strategy and independent survival benefit remain uncertain; recent observational findings vary by population and cannot establish causation. Transfused cohort; minimally transfused cohort.
Give timely TXA; replace demonstrated deficits
TXA: For significant traumatic bleeding, give as early as possible and within 3 hours of injury: adult regimen 1 g IV over 10 minutes, then 1 g over 8 hours. Check prehospital doses. Do not wait for viscoelastic evidence of hyperfibrinolysis. CRASH-2 timing analysis; European guideline.
Do not routinely start TXA beyond 3 hours: the timing analysis found a late bleeding-mortality harm signal. Late documented hyperfibrinolysis needs specialist/protocol guidance. TXA reduces fibrin breakdown; it does not replace fibrinogen or repair vessels, and is not indicated for every injured patient. Timing evidence.
Fibrinogen: With major bleeding, Clauss fibrinogen ≤1.5 g/L or a viscoelastic functional deficit supports cryoprecipitate or fibrinogen concentrate. The guideline suggests 3–4 g of fibrinogen initially, then reassessment. Use product-specific cryoprecipitate equivalents. Platelets: Suggested counts are >50 × 109/L during bleeding and >100 × 109/L with TBI; evidence is low certainty. European guideline.
Routine extra high-dose cryoprecipitate did not improve mortality in CRYOSTAT-2; this does not negate treating a demonstrated deficit. As results arrive, target products instead of continuing fixed ratios indefinitely. Check anticoagulant exposure and use the appropriate reversal pathway.
Reassess the patient, not one number
Repeat coagulation tests, fibrinogen, blood gas and ionised calcium at a pace matched to bleeding, transfusion and deterioration. During rapid transfusion, an hour may be too long. Critical-bleeding guidance. Viscoelastic testing can guide treatment through a validated device-specific algorithm, but ITACTIC did not establish overall superiority over conventional-test guidance.
Look for improving peripheral perfusion, pulse/pressure trends, lactate/base deficit, temperature and falling transfusion needs as bleeding is controlled. Sedation limits mental-status assessment. Narrowing pulse pressure is a warning, not recovery; supporting evidence is observational. Pulse-pressure study.
If operating-room access is delayed, escalate immediately through the trauma, surgical and anaesthetic teams and the hospital emergency pathway. Arrange a definitive haemostasis plan while resuscitation continues. Better laboratory results never replace stopping the bleeding.
