Haemorrhagic shock, trauma resuscitation and massive

• Haemorrhage is the leading preventable cause of death after trauma. The bleeding patient dies from blood loss within minutes to hours, usually before the underlying injuries have been fixed, so care is built around stopping bleeding and replacing what was lost in the right way.
• The Advanced Trauma Life Support (ATLS) classification divides blood loss into four classes. A single vital sign is enough to say "this patient is bleeding seriously"; the full picture is base deficit, lactate, shock index and response to fluids.
• The plan is damage-control resuscitation: control haemorrhage early, keep the patient warm and not too acidotic, allow permissive (controlled) hypotension until the bleeding source is controlled, give blood products in balanced ratios rather than crystalloid, and add tranexamic acid within three hours of injury. Every major trauma unit should have a massive transfusion protocol.
• In adults the usual starting blood-product ratio is about 1 unit of plasma to 1 unit of red cells, with platelets added early, then switch to a goal-directed strategy guided by standard laboratory tests and, where available, viscoelastic (TEG/ROTEM) assays.
• Point-of-care viscoelastic assays are more sensitive than conventional coagulation tests for the lethal coagulopathy of trauma, but they guide adjunct therapy rather than replace tranexamic acid or blood.

By the end of this page you should be able to:

1. State the four ATLS classes of haemorrhagic shock from the table, including approximate blood loss, vital signs, urine output and base deficit for each.
2. Explain controlled (permissive) hypotension: the target blood pressures, who must not receive it, and the rationale.
3. Outline damage-control resuscitation and the steps of damage-control surgery.
4. State the contents and ratios of an inpatient massive transfusion protocol and the laboratory targets that guide later therapy.
5. Give the tranexamic acid dose and timing in trauma and explain why it is given early.
6. Name two point-of-care coagulation tests, what they add, and one test that is not recommended for routine use.
7. Define the lethal triad and state why each part feeds the others.

Haemorrhagic shock is not a blood pressure number. Significant blood can be lost before the pressure falls, because heart rate rises and the peripheral circulation clamps down to protect the central organs. The way to miss it is to reassure yourself with a normal blood pressure in a pale, tachycardic, cold patient. See the MedGrasp topic "Shock" for the wider framework of hypovolaemic, cardiogenic, obstructive and distributive shock.

Blood is lost faster than it can be replaced. Venous return falls, stroke volume falls, cardiac output falls. Tissue oxygen delivery (cardiac output times arterial oxygen content) drops. Cells switch to anaerobic glycolysis, lactate rises, and the pH falls. Red-cell and platelet loss dilutes the clotting proteins; clotting slows; bleeding worsens. Acidosis itself impairs thrombin generation and platelet function, so the slower the bleeding is controlled, the harder it becomes to stop.

A severe bleed drives every system in the same direction at once. The body compensates with tachycardia, vasoconstriction and a shift of interstitial fluid into the vessels, which is why haemoglobin and haematocrit can look normal early: the lost volume is masked by haemoconcentration then haemodilution. Repeat the haemoglobin and search for the source, rather than trusting a single early value.

The Advanced Trauma Life Support (ATLS) classification of blood loss, reproduced as table 2 of the European guideline on management of major bleeding and coagulopathy following trauma (sixth edition, 2023), groups blood loss into four classes with their clinical signs. Classes III and IV represent haemorrhagic shock. ^{[1] [2]}

A narrow pulse pressure (under 40, or under 30 mmHg) is an ATLS class II signal and has been independently associated with massive transfusion (odds ratio 3.74, 95% confidence interval 1.8 to 7.7) and emergent surgery. ^[1]

Clinical judgement is the starting point. The European guideline recommends that the extent of traumatic haemorrhage be assessed using the patient's physiology, anatomical injury pattern, mechanism of injury and their response to initial resuscitation, and that the shock index and/or pulse pressure be used to assess the degree of hypovolaemic shock and transfusion requirements. ^[1]

• Shock index (SI) = heart rate divided by systolic blood pressure. In healthy adults it is typically 0.5 to 0.7. ^[1]

• SI 0.9 to 1.0 or higher has been retrospectively associated with massive transfusion in 25% of bleeding trauma patients, interventional radiology in 6.2% and operative intervention in 14.7%. Retrospective studies have used a range of thresholds between SI 0.8 and 1.0 or higher, with areas under the curve between 0.73 and 0.89. At a cut-off of 0.81, SI predicted massive transfusion with a sensitivity of 85%, a specificity of 64%, a positive predictive value of 16% and a negative predictive value of 98%. At a cut-off of 0.91, the sensitivity was 81% and the specificity 87%. ^[1]

• Base deficit. The ATLS table uses base deficit (0 to minus 2 in class I up to minus 10 or less in class IV) as a metabolic sign of shock. Lactate is preferred, but where lactate cannot be measured, base deficit is a suitable alternative. ^[1]

• Lactate. Blood lactate is produced by anaerobic glycolysis and reflects the severity of haemorrhagic shock and of tissue hypoperfusion; serial measurements are useful to track the response to treatment, especially in penetrating trauma. ^[1]

• Haemoglobin and haematocrit. An initial value in the normal range does not exclude serious bleeding, because it takes time for the interstitial fluid to shift. Repeated measurement is recommended. ^[1]

The decision rests on physiological criteria, the patient's haemodynamic status and their response to initial volume resuscitation. ^[2]

Severe trauma drives three problems that worsen each other. This is the lethal triad. ^[1]

• Coagulopathy. About one third of severely injured patients arrive at hospital already coagulopathic, and trauma-induced coagulopathy is associated with higher transfusion requirements, multi-organ failure and death. It is defined in the European guideline by a prothrombin time ratio above 1.2, with a ratio above 1.5 indicating severe coagulopathy. Low fibrinogen and raised D-dimers are commonly found and are associated with increased mortality. ^[1]
• Hypothermia. Core temperature below 35 degrees Celsius is consistently associated with higher mortality and more blood-product use. Every 1 degree Celsius fall in temperature is associated with about a 10% fall in clotting factor function. The coagulation defect may not show on laboratory tests performed at 37 degrees Celsius, which is standard practice. ^[1]
• Acidosis. Acidosis from anaerobic metabolism impairs thrombin generation and platelet function.

The loop is the point. Hypothermia and acidosis each worsen clotting, the resulting bleeding worsens hypothermia and acidosis (volume loss, exposure, cold fluids), and the patient enters a downward spiral. This is the main argument for damage control: stop the cycle early rather than pursuing a tidy repair in a patient with exhausted physiology.

Hypocalcaemia is so common during massive transfusion that many now speak of a lethal diamond, adding ionised calcium below the normal range (1.1 to 1.3 mmol/L) to the triad. See the component-therapy section. ^[1]

Major trauma in most networks uses the xABCDE primary survey, with the "C" placed first as critical bleeding: exsanguinating haemorrhage, airway, breathing, circulation, disability and exposure, with temperature control. The secondary survey (head-to-toe examination and AMPLE history) follows only once life threats are controlled; the MedGrasp topics "Shock", "Initial trauma assessment: primary survey", "Secondary survey and penetrating trauma" and the abdominal and thoracic trauma topics all sit alongside this page.

The priorities in the bleeding patient, supported by the European guideline and NICE NG39, are: ^{[1] [2]}

• Direct pressure with simple dressings on any external bleed; a commercial tourniquet for limb bleeding that does not stop with pressure. ^{[1] [2]}
• Pelvic binder early for suspected major pelvic fracture. ^{[1] [2]}
• Early imaging. Contrast-enhanced whole-body CT for detection of bleeding and injury in the stabilised patient; limit plain films and ultrasound to the minimum needed to direct intervention in patients not responding to fluids. A negative FAST does not exclude intraperitoneal or retroperitoneal bleeding. ^{[1] [2]}
• Damage-control surgery if the patient is haemodynamically unstable and not responding to volume; consider definitive surgery if they are responding, and use definitive surgery when the haemodynamic state is normal. ^[2]
• Airway and breathing as usual; avoid hypoxaemia and avoid routine hyperoxaemia except in imminent exsanguination, and aim for normoventilation unless there are signs of cerebral herniation. ^[1]

The idea is counter-intuitive: in a patient who is still bleeding, pushing the blood pressure up with fluid just pushes more blood out of the uncontrolled wound. Controlled resuscitation keeps the pressure low enough to slow bleeding but high enough to keep the brain and heart perfused.

The European guideline recommends, in the initial phase after trauma, a restricted volume replacement strategy with a target systolic blood pressure of 80 to 90 mmHg (and a mean arterial pressure of 50 to 60 mmHg) until major bleeding has been stopped, provided there is no clinical evidence of brain injury. ^[1]

In patients with severe traumatic brain injury (Glasgow Coma Scale 8 or less), a higher target applies: maintain a mean arterial pressure of 80 mmHg or more, because the injured brain cannot tolerate low perfusion pressure. ^[1]

NICE NG39 agrees: use a restrictive approach to volume resuscitation for patients with active bleeding until definitive early control of bleeding, titrating in pre-hospital care to a palpable central (carotid or femoral) pulse and in hospital to the central circulation while moving rapidly to haemorrhage control; if haemorrhagic shock is the dominant condition with traumatic brain injury, continue restrictive resuscitation, and if traumatic brain injury is dominant, use a less restrictive approach to maintain cerebral perfusion. ^[2]

The controlled-hypotension strategy was triggered in its modern form by a 1990s randomised trial in penetrating trauma showing increased survival, and a restricted strategy is now replacing conventional aggressive fluid resuscitation. ^[1]

Damage-control resuscitation is the strategy that matches the damage-control surgical approach: stop bleeding, do not wash it out with crystalloid, and keep physiology working long enough for definitive repair. ^[1]

Its elements: control haemorrhage aggressively; replace lost volume with blood products in balanced ratios rather than crystalloid; keep the patient warm; prevent acidosis with prompt bleeding control and adequate ventilation; and allow permissive hypotension until the source is controlled. ^{[1] [2]}

Damage-control surgery is recommended in the severely injured patient who presents with haemorrhagic shock, signs of ongoing bleeding, coagulopathy and/or combined abdominal vascular and pancreatic injury; and in anyone with hypothermia, acidosis, an inaccessible major anatomical injury or a need for time-consuming procedures. Otherwise use primary definitive surgical management. ^[1]

NICE NG39 says the same: use damage-control surgery in patients who are haemodynamically unstable and not responding to volume resuscitation; consider definitive surgery when they are responding; and use definitive surgery when the haemodynamic state is normal. ^[2]

A typical sequence of damage-control surgery: gain rapid control of bleeding (packing, temporary clamping or shunting), control contamination (stapling bowel, packing the pelvis), leave the abdomen open (temporary abdominal closure), resuscitate in intensive care to normalise temperature, acid-base and coagulation, and return to theatre 24 to 48 hours later for definitive reconstruction. ^[1]

Early haemostatic monitoring and measures to support coagulation should be started immediately on arrival: the type and degree of coagulopathy in the individual patient determines what is replaced. ^[1]

Crystalloid is for bridging, not for staying. In hospital, crystalloids are not to be used for patients with active bleeding. ^[2] The European guideline recommends 0.9% sodium chloride or a balanced crystalloid to start fluid therapy in the hypotensive bleeding trauma patient, and restricting the use of colloids because of their adverse effects on haemostasis. ^[1]

The European guideline suggests that if a restrictive strategy does not reach the target blood pressure, noradrenaline may be added to fluids to maintain the target arterial pressure, with dobutamine in the presence of myocardial dysfunction. ^[1]

Keep the patient warm throughout: minimise heat loss pre-hospital and in hospital, and rewarm the hypothermic patient. ^{[1] [2]}

A massive transfusion protocol activates early, on physiological criteria and on the patient's response to initial volume resuscitation. ^[2] Every hospital that receives major trauma should have protocols for adults and for children. ^[2]

For the initial phase of expected massive haemorrhage, the European guideline recommends choosing one of two strategies: ^[1]

• Fibrinogen concentrate or cryoprecipitate together with red cells, or
• Fresh frozen plasma or pathogen-inactivated plasma in a plasma-to-red-cell ratio of at least 1:2 as needed.

In addition, a high platelet-to-red-cell ratio is suggested. ^[1]

The randomised PROPPR trial in 680 trauma patients compared early plasma, platelet and red cell transfusion at a 1:1:1 ratio with a 1:1:2 ratio. Mortality was similar, but the 1:1:1 group showed improved haemostasis and fewer deaths from exsanguination. ^[1]

NICE NG39 recommends, for adults, a starting ratio of 1 unit of plasma to 1 unit of red blood cells to replace fluid volume. The protocol should then start with a fixed-ratio blood-component strategy and move at the earliest opportunity to a protocol guided by laboratory coagulation results. ^[2]

The replacement of large volumes dilutes clotting factors and platelets, so late therapy is driven by the deficit found. Contributors to the component problem are dilution (crystalloid and red cells without plasma), consumption (clotting at multiple injuries), hypothermia (see above), and the acidosis and calcium losses already described.

Prothrombin concentrate 20 to 50 U/kg is appropriate for vitamin-K antagonist reversal or selected direct-factor-Xa reversal rather than for routine trauma. ^[1] Recombinant activated factor VII is not recommended as first-line treatment; it should be considered only if major bleeding and coagulopathy persist despite all other measures, including optimal haemostasis, near-normal pH and temperature. ^[1]

Tranexamic acid is an antifibrinolytic (it blocks the activation of plasminogen to plasmin) that reduces fibrinolysis, which is common after severe trauma. The European guideline recommends giving tranexamic acid to the trauma patient who is bleeding or at risk of significant bleeding, as soon as possible, if feasible en route to hospital, and within three hours of injury, at a loading dose of 1 g infused over 10 minutes followed by an intravenous infusion of 1 g over 8 hours; the administration must not wait for viscoelastic assessment. ^[1]

NICE NG39 states the same clinical position for major trauma: use intravenous tranexamic acid as soon as possible in patients with major trauma and active or suspected active bleeding, and do not use it more than 3 hours after injury unless there is evidence of hyperfibrinolysis. ^[2]

This recommendation rests on the CRASH-2 trial, which showed a 1.5% reduction in overall mortality and a roughly one-third reduction in deaths from bleeding with tranexamic acid in bleeding trauma patients. ^[1] Later studies examined tranexamic acid in traumatic brain injury (the largest being CRASH-3, with 12,737 patients comparing a 1 g bolus), different dosing regimens and pre-hospital administration. ^[1] In brain-injured patients, note that traumatic brain injury dominates the volume strategy: when brain injury is the dominant condition, a less restrictive resuscitation targets adequate cerebral perfusion while other measures continue.

Early coagulation monitoring is recommended immediately upon arrival, using either traditional laboratory tests (prothrombin time with INR, Clauss fibrinogen, platelet count), or point-of-care PT/INR, or a viscoelastic method. ^[1] Viscoelastic assays such as thromboelastography (TEG) and rotational thromboelastometry (ROTEM) are more sensitive than conventional coagulation tests for detecting traumatic coagulopathy, can identify hyperfibrinolysis (the most lethal and resource-intensive fibrinolysis phenotype in trauma), and provide thresholds for goal-directed therapy with blood products, coagulation factors and haemostatic agents. ^[1]

Coagulation point-of-care testing is therefore additive: it selects what to replace after the first blood products, rather than replacing tranexamic acid or the fixed-ratio start. Use it to run a goal-directed strategy guided by standard laboratory values and/or viscoelastic results. ^[1]

Systematic review evidence (15 randomised trials, 1,238 patients, including one trauma trial) suggests that viscoelastic-guided algorithms reduce transfused red cells and fresh frozen plasma and bleeding volume compared with clinician discretion and/or conventional coagulation tests; platelet transfusion requirements and mortality were not significantly different. ^[3]

Routine point-of-care platelet-function testing is not recommended in trauma patients on antiplatelet therapy or with suspected platelet dysfunction: the devices measure different parameters of activation, are not interchangeable, and results lose value when the platelet count is low. ^[1]

Also note that routine prophylactic platelet transfusion is not recommended in patients with ongoing bleeding who were taking antiplatelet agents. ^[1] The co-concept is to reverse anticoagulation appropriately instead: rapid reversal with prothrombin complex concentrate and 5 to 10 mg intravenous vitamin K for vitamin-K antagonists; for direct factor-Xa inhibitors, plasma-level measurement and specific reversal (andexanet alfa where available, otherwise prothrombin complex concentrate) as in the guideline. ^[1]

Traumatic brain injury. Brain-injured patients need a higher blood-pressure target (mean arterial pressure at least 80 mmHg) and often a higher platelet target (above 100 × 10^9/L). Damage control and a restrictive fluid policy still apply when haemorrhagic shock is the dominant problem; when brain injury dominates, perfusion of the injured brain takes precedence. ^{[1] [2]}

Anticoagulated patients. Reverse anticoagulation rapidly only when there is active bleeding. For warfarin, use prothrombin complex concentrate plus vitamin K rather than plasma. NICE NG39 lists the trust-level protocol requirement and the need for a haematologist's advice for DOAC or childhood reversal. ^{[1] [2]}

Older adults and chronic hypertension. These patients fall off the table before the numbers do: their tolerance of hypotension is poor, and their bleeding response may be blunted by beta-blockers. Avoid permissive hypotension in these groups and pursue early haemorrhage control.

• Clinical assessment plus repeated simple measurements still stage the bleed: pulse, blood pressure, mental state, urine output and, where available, base deficit and haemoglobin. The ATLS classes, shock index and lactate thresholds remain usable where laboratory support exists.
• Where blood products are limited, tranexamic acid (a cheap, stable ampoule) is arguably the best value given within three hours, plus warming the patient, direct pressure, tourniquets and pelvic binding.
• In smaller hospitals without a defined massive transfusion protocol, agree one in advance with the nearest trauma centre and the blood service, including the criteria for activation, so that nobody improvises at 2 a.m.
• Viscoelastic testing is unlikely to be available. Goal-direction then relies on prothrombin time, fibrinogen (Clauss), platelet count and haemoglobin repeated regularly.
• Ratios of 1:1 plasma to red cells may be difficult to maintain if frozen plasma thawing is slow; use whatever blood-component logistics the hospital has, while moving as fast as possible toward component-based, ratio-driven resuscitation. ^[2]

• Hypotension or relative hypotension in a trauma patient after mechanisms that can bleed: treat as haemorrhagic shock until proven otherwise, and do not attribute it to head injury.
• Tachycardia out of proportion to pain or fever, with a narrow pulse pressure.
• Confusion, agitation or falling consciousness, even before hypotension appears.
• Oliguria (under a few tens of millilitres per hour in adults).
• Rising lactate or a deepening base deficit.
• Cold, clammy or mottled peripheries; high neck veins with hypotension suggest tamponade or tension pneumothorax, not simple bleeding.
• Continuing bleeding with a platelet count falling during resuscitation.

• Haemorrhage kills in minutes to hours; treatment is control of bleeding plus the right replacement, done together rather than in sequence. ^{[1] [2]}
• ATLS classes I to IV group blood loss by approximate percentage (under 15%, 15 to 30%, 31 to 40%, over 40%) with characteristic changes in heart rate, blood pressure, pulse pressure, respiration, urine, Glasgow score and base deficit; classes III and IV are haemorrhagic shock. ^[1]
• Shock index (heart rate divided by systolic blood pressure) in healthy adults is 0.5 to 0.7; 0.9 to 1.0 or higher predicts massive transfusion in a substantial minority of bleeding patients; 0.81 has a very high negative predictive value. ^[1]
• Controlled (permissive) hypotension targets a systolic pressure of 80 to 90 mmHg and a mean arterial pressure of 50 to 60 mmHg until bleeding is controlled, unless there is brain injury, where the mean arterial pressure is kept at 80 mmHg or above. ^{[1] [2]}
• Damage-control resuscitation: early haemorrhage control, blood products in balanced ratios rather than crystalloid, warmth, prevention of acidosis, and avoidance of large volumes before control. ^{[1] [2]}
• Damage-control surgery is reserved for the patient with ongoing bleeding and exhausted physiology: pack, control contamination, close temporarily, restore physiology, then reconstruct. ^{[1] [2]}
• A fixed-ratio blood-product start (about 1:1 plasma to red cells in adults, with early platelets) prevents both hypovolaemia and coagulopathy better than red cells alone, and should be switched to a goal-directed strategy as early as possible. ^{[1] [2]}
• Tranexamic acid 1 g over 10 minutes then 1 g over 8 hours, within 3 hours of injury, in bleeding or at-risk trauma patients; do not wait for viscoelastic results. CRASH-2 showed a 1.5% absolute mortality reduction and roughly one third fewer deaths from bleeding. ^{[1] [2]}
• The lethal triad is coagulopathy, hypothermia and acidosis, each worsening the others; hypocalcaemia is often added as a fourth point.
• Fibrinogen is supplemented when the Clauss level falls to 1.5 g per litre or less, with an initial dose of 3 to 4 g; platelets are maintained above 50 × 10^9 per litre in bleeding and above 100 × 10^9 per litre in brain injury; ionised calcium is kept at 1.1 to 1.3 mmol per litre; haemoglobin is targeted at 70 to 90 g per litre. ^[1]
• Viscoelastic assays (TEG, ROTEM) are more sensitive than conventional coagulation tests for traumatic coagulopathy and hyperfibrinolysis, and guide goal-directed replacement; trial evidence shows reduced red cell and plasma use but no proven mortality difference. ^{[1] [3]}
• Routine point-of-care platelet function testing is not recommended after trauma, and routine prophylactic platelet transfusion is not recommended in patients on antiplatelet agents. ^[1]
• Cryoprecipitate or fibrinogen concentrate, not plasma, is the correct treatment for hypofibrinogenaemia. ^[1]
• Keep the patient warm, use appropriate fluids (isotonic crystalloid as a bridge, never crystalloid for active hospital bleeding) and aim for total haemorrhage control as early as possible. ^{[1] [2]}
• In pregnancy: blood loss is tolerated differently because of physiological change; local obstetric and trauma teams must lead, but the same principles apply.

• Shock
• Initial trauma assessment: primary survey
• Secondary survey and penetrating trauma
• Surgical bleeding, coagulopathy and blood products
• Transfusion reactions
• Damage-control surgery and abdominal compartment syndrome
• Head and spinal injury in trauma
• Thoracic trauma
• Abdominal and pelvic trauma

(may be helpful): hypovolaemic shock, massive transfusion protocol, damage control resuscitation, permissive hypotension, tranexamic acid CRASH-2, lethal triad, TEG ROTEM trauma.