Transfusion — Products, Thresholds, TRALI & TACO

Overview & definition

The blood transfusion in the ICU requires the product knowledge (the RBC, the FFP, the platelets, the cryo), the restrictive thresholds (the TRICC trial — the Hb under 70), and the recognition of the transfusion reactions (the TRALI, the TACO, the acute haemolytic). The restrictive strategy (the lower the threshold) the reduces the complications without the worse the outcome. The TRALI and the TACO are the two the pulmonary reactions — the TRALI is the ARDS-like (the NO overload); the TACO is the overload (the diuretics).[1]

The products and thresholds

RBC (packed red blood cells)

• The 1 unit raises the Hb by about 10 g per L. The leukodepleted (the standard — reduces the febrile and the CMV).[1]
• The threshold — Hb under 70 for the most (the TRICC trial — the restrictive vs the liberal, the no the worse the outcome; the TRISS — the 70 vs 90 in the septic shock, the no the difference). The under 80 for the cardiovascular disease, the acute coronary syndrome, the elderly. The under 80 to 100 for the massive the bleeding or the ongoing.[1]

FFP (fresh frozen plasma)

• The clotting factors. The 10 to 15 mL per kg raises the factor levels the about 20 per cent.[1]
• The threshold — the PT or APTT over 1.5 times the normal AND bleeding (the 15 mL per kg). The takes the 30 min to thaw.[1]

Platelets

• The 1 adult dose (the 1 pool = the 4 to 6 units = the about 300 × 10⁹) raises the count by about 30 to 40 × 10⁹.[1]
• The threshold — the under 10 to 20 (the prophylactic, the non-bleeding); the under 50 (the active bleeding or the surgery); the under 100 (the CNS bleeding, the neurosurgery).[1]

Cryoprecipitate

• The fibrinogen, the factor VIII, the vWF, the factor XIII. The 2 pools (10 units) raises the fibrinogen by about 1.0 g per L.[1]
• The threshold — the fibrinogen under 1.5 g per L AND bleeding.[1]

PCC (prothrombin complex concentrate)

• The factors II, VII, IX, X. For the warfarin reversal (the faster than the FFP — the minutes vs the 6 to 12 h).[1]

Blood products — composition, dose, indication, risks

[1]

Restrictive vs liberal transfusion — the trial evidence

The single most-tested question in critical-care transfusion is the haemoglobin threshold at which to transfuse red cells. Six landmark RCTs and a large meta-analysis underpin the modern restrictive strategy (Hb under 70 g/L) for the majority of ICU patients. The exceptions — acute coronary syndrome, cardiac surgery, severe brain injury — are the high-yield exam edge cases.[1]

The age and brain-injury exceptions

The TRICC trial excluded patients with acute coronary syndrome and severe brain injury; these are the populations where higher thresholds have historically been recommended. The modern synthesis is more nuanced:[1]

• Acute coronary syndrome / acute MI: REALITY 2021 showed non-inferiority was borderline — most guidelines suggest Hb under 80 for AMI (some under 80-90 if ongoing ischaemia). The anaemic failing heart relies on the oxygen-carrying capacity; under 70 risks demand ischaemia. Yet excessive liberal transfusion in TRICC-era meta-analyses showed no benefit and possible harm.[5]
• Severe traumatic brain injury (TBI): cerebral perfusion pressure (CPP = MAP − ICP) depends on oxygen delivery; a low Hb worsens secondary brain injury. Common target Hb 90-100 g/L in acute severe TBI (Robertson 1999 EPO trial). However, no modern RCT has definitively confirmed benefit of Hb over 100, and over-transfusion raises thrombosis and ARDS risk.
• Acute cerebral ischaemia / ICH / subarachnoid haemorrhage: target Hb typically 90-100 g/L in the acute phase; some SAH centres transfuse up to 100-110 for DCI (delayed cerebral ischaemia).
• Sepsis / septic shock: TRISS 2014 settled the question — Hb 70 is safe.[2]
• Cardiac surgery: TRICS-3 2018 (n=5243) settled it — restrictive (Hb 75) non-inferior.[7]
• Elderly with CV comorbidity after hip surgery: FOCUS 2011 legitimised Hb under 80.[3]
• Upper GI bleeding: Villanueva 2013 — restrictive (Hb under 70) superior, less rebleeding.[8]

Massive transfusion and the MTP

Massive transfusion is defined as the replacement of one blood volume within 24 h (≈ 10 units PRBC in an adult), or over 4 units in an hour with ongoing haemorrhage, or the replacement of 50 per cent of blood volume in 3 h. It triggers the massive transfusion protocol (MTP) — a predefined ratio-based delivery of PRBC, FFP and platelets to prevent the lethal triad of acidosis, hypothermia and coagulopathy.[6]

The PROPPR trial — 1:1:1 vs 1:1:2

The Pragmatic Randomized Optimal Platelet and Plasma Ratios trial (Holcomb 2015, JAMA) randomised 680 patients with severe trauma and active bleeding to 1:1:1 (PRBC:FFP:platelets) vs 1:1:2. The primary outcome (24-h and 30-day mortality) was not significantly different, but the 1:1:1 arm achieved more haemostasis (86 vs 78 per cent) and fewer deaths from exsanguination (9 vs 15 per cent). The harm trade-off: no increase in complications. The pragmatic conclusion: activate a balanced ratio early in massive trauma haemorrhage.[6]

[1] [1]

Viscoelastic testing — TEG and ROTEM

Conventional coagulation tests (PT, APTT, INR, fibrinogen) are slow (30-60 min turnaround), performed at 37 °C in platelet-poor plasma, and tell you the time to fibrin formation but nothing about clot strength or breakdown. Viscoelastic tests (VHA — viscoelastic haemostatic assay) measure whole-blood clot formation and lysis in real time, at the bedside, and are the modern standard for goal-directed therapy in massive haemorrhage, cardiac surgery, liver transplantation and postpartum haemorrhage.[1]

[1] [1]

Special populations

Jehovah's Witnesses

Jehovah's Witnesses refuse transfusion of whole blood, PRBC, platelets, plasma and cryoprecipitate on religious grounds (the Watchtower doctrine). The refusal is usually absolute but highly individualised — never assume; document the specific products the patient will and will not accept. Key management principles:[1]

• Optimise endogenous red cell mass — intravenous iron (ferric carboxymaltose 1 g), erythropoietin (epoetin alfa 40 000 U weekly), vitamin B12, folate.
• Minimise iatrogenic blood loss — small-volume paediatric tubes, avoid arterial lines unless essential, point-of-care testing.
• Cell salvage (cell saver) — accepted by many (but not all) Witnesses provided the circuit forms a continuous loop with the patient's body (not stored separately). Confirm with the patient in writing.
• Acute normovolaemic haemodilution — also accepted by some if the circuit is continuous.
• Agents that may be accepted: factor concentrates (PCC, factor VIII, fibrinogen concentrate — derived fractions, usually accepted), recombinant factors, albumin, immunoglobulins ( Witnesses generally accept plasma fractions but again, document).
• Decision-making in incapacity: if the patient lacks capacity and has no documented refusal, the doctrine of necessity applies — transfuse to save life. If a valid Advance Decision (UK) or Advance Care Directive (ANZ) refuses transfusion, the courts usually uphold it; emergency life-saving transfusion in a child or pregnant patient may require judicial review.
• Mortality rises sharply below Hb 50 g/L in the Witness population — the case series by Viele (1994, n=195) and Shander (2017) define the steep inflection. [1]

Anticoagulated patients

The anticoagulated patient who needs transfusion is the rule, not the exception, in ICU. The transfusion question is rarely the Hb threshold alone — it is (1) does the anticoagulant need reversing now, and (2) which product?[1]

• Warfarin with life-threatening bleed: vitamin K 10 mg IV slow PLUS 4F-PCC 25-50 IU/kg (or Prothrombinex 25-50 IU/kg + FFP in ANZ). Reversal in minutes; INR check at 30 min. Do not give FFP alone — too slow, too much volume (INCH trial).
• Dabigatran: idarucizumab 5 g IV (Praxbind). Haemodialysis also works (low protein binding).
• Rivaroxaban / apixaban: andexanet alfa if available, otherwise 4F-PCC 50 IU/kg.
• Heparin (UFH): protamine 1 mg per 100 units of heparin (max 50 mg/dose, max 5 mg/min).
• LMWH: protamine partial reversal only (~60-75 per cent).
• Antiplatelet with ICH: platelet transfusion is contraindicated in the PATCH trial (unscheduled surgery for spontaneous ICH on antiplatelet — platelet transfusion worsened outcome). Exception: planned neurosurgery within 24 h (some centres still transfuse). [1]

Neonates and paediatrics

Neonates are uniquely vulnerable to three transfusion-specific risks: TA-GVHD (immature immune system, especially if donor is a first- or second-degree relative — the donor's T-cells engraft), CMV transmission (lifelong consequence in a low-birth-weight neonate), and cytopaenia from a single adult unit (volume and citrate load).[1]

• CMV-negative AND irradiated products for: intrauterine transfusion, exchange transfusion, neonates under 1200 g, neonates receiving components from first- or second-degree relatives, known immunodeficiency (SCID, DiGeorge).
• Irradiation prevents T-cell proliferation (TA-GVHD) — required for all directed donations from relatives, all HLA-matched units, all components for cellular immunodeficiency, all granulocyte concentrates, and within 14 days of collection for neonatal top-up transfusions.
• Volume: 10-20 mL/kg PRBC raises Hb by ~30-50 g/L; use a dedicated unit to limit donor exposure.
• Thresholds: neonatal top-up PRBC typically Hb under 110-120 with symptoms, or under 70-80 stable; platelet prophylactic threshold higher (under 30 in stable, under 50-100 if bleeding/procedure); FFP 10-20 mL/kg. [1]

The transfusion reactions

TRALI (transfusion-related acute lung injury)

• The donor anti-leukocyte antibodies → the neutrophil activation → the pulmonary capillary leak → the ARDS-like (the hypoxia, the bilateral infiltrates) within the 6 hours of the transfusion.[1]
• The NO the fluid overload (distinguishes from the TACO — the TRALI the euvolaemic or the hypovolaemic).[1]
• The treatment — the supportive (the oxygen, the ventilation). The NO the diuretics (the TRALI not the overload).[1]
• The mortality the about 10 per cent. The report (the donor the traced; the male-donor plasma preferred — the fewer the antibodies).[1]

TACO (transfusion-associated circulatory overload)

• The circulatory overload from the transfusion volume → the pulmonary oedema.[1]
• The diagnosis — the fluid balance (the positive), the BNP or NT-proBNP (the elevated), the hypertension (the distinguishing from the TRALI).[1]
• The treatment — the diuretics (the furosemide), the slow the transfusion, the reduce the volume.[1]
• The more common than the TRALI. The elderly and the cardiac the higher risk.[1]

The acute haemolytic (the ABO incompatibility)

• The wrong blood to the wrong patient → the intravascular haemolysis (the fever, the flank pain, the hypotension, the DIC, the haemoglobinuria).[1]
• The STOP the transfusion immediately. The supportive + the treat the DIC + the renal protection (the fluids, the alkalinisation).[1]
• The prevented by the patient identification and the cross-match.[1]

The other reactions

• The febrile non-haemolytic (the cytokines from the white cells — the leukodepletion the reduces). The antipyretic.[1]
• The allergic (the plasma proteins — the mild to the anaphylaxis). The antihistamine (mild); the adrenaline (severe).[1]
• The delayed haemolytic (the antibodies from the prior the sensitisation → the extravascular haemolysis, the days to the weeks).[1]

Transfusion reactions — the comprehensive differential

The seven reactions below are tested across CICM, FFICM and EDIC — and TRALI vs TACO is the single most-asked comparison. The mnemonic ABCDEF (Acute haemolytic, Bacterial, TACO circulatory overload, Delayed haemolytic, E [TRALI lung injury], Febrile non-haemolytic) covers the major ones; allergic/anaphylaxis and TA-GVHD round out the list. [1]

[1]

Bacterial contamination — the septic transfusion reaction

The single most important infectious complication is bacterial growth in stored components — most often platelets (stored at 20-24 °C, the only room-temperature product) and rarely PRBC (cold storage at 1-6 °C favours psychrophilic gram-negatives like Yersinia enterocolitica, Pseudomonas fluorescens, Serratia).[1]

• Clinical: high fever (over 39 °C or over 2 °C rise from baseline), rigors, severe hypotension, vomiting, DIC — out of proportion to the transfusion. Occurs during or within 1-4 h of infusion.
• Action: STOP the transfusion immediately. Send the unit AND the patient's blood cultures. Begin broad-spectrum antibiotics (vancomycin + piperacillin-tazobactam or local protocol). Resuscitate aggressively; vasopressors for shock.
• Prevention: platelet bacterial screening (culture or rapid test pre-release), short shelf-life (5-7 days), pathogen-reduction technologies (amotosalen/UVA — INTERCEPT, riboflavin/UV — Mirasol).
• Mortality: 25-60 per cent — the highest of the acute reactions. [1]

TA-GVHD — the almost universally fatal reaction

Transfusion-associated graft-versus-host disease is rare but almost universally fatal (over 90 per cent mortality). Donor T-lymphocytes in transfused cellular components engraft in a recipient who cannot reject them, then attack the recipient's skin, gastrointestinal tract, liver and bone marrow. Onset is 4-30 days post-transfusion.[1]

• Clinical: fever, maculopapular rash (palms, soles, spreads centrally), diarrhoea, hepatitis, profound pancytopenia from marrow aplasia.
• Diagnosis: skin biopsy (interface dermatitis with lymphocyte tagging along the basement membrane, dyskeratotic cells); molecular typing showing donor lymphocytes in recipient blood.
• Risk groups: cellular components (PRBC, platelets, granulocytes) given to:
  • Neonates (especially intrauterine or exchange transfusion)
  • Congenital immunodeficiency (SCID, DiGeorge)
  • Haematological malignancy on chemotherapy (Hodgkin's lymphoma classically)
  • Solid organ or stem-cell transplant recipients
  • Fludarabine / clofarabine / other purine analogues (impair T-cell function for months)
  • Directed donations from first- or second-degree relatives (donor HLA-homozygous, shares one haplotype with recipient → not recognised as foreign)
  • HLA-matched platelets
  • Granulocyte concentrates
• Prevention: gamma or X-ray irradiation of cellular components at 25 Gy (centre of the canister; min 15 Gy). Leukodepletion is NOT enough — leukocytes are reduced but viable T-cells persist.
• Treatment: no effective therapy. Corticosteroids, ciclosporin, anti-thymocyte globulin and eltrombopag have been tried — mortality remains over 90 per cent. [1]

Anaphylactic reaction — the IgA-deficient recipient

The classic exam patient is the IgA-deficient recipient with anti-IgA antibodies (1 in 500-1000 of the population, often asymptomatic until first transfusion). The recipient's IgE (or IgG) against donor IgA in plasma triggers massive mast-cell degranulation within minutes of starting the transfusion. Anhaptoglobulinaemia (more common in East Asian populations) is an emerging cause.[1]

• Clinical: hypotension, bronchospasm, angioedema, urticaria, gastrointestinal symptoms. No fever (distinguishes from haemolytic / bacterial). Under anaesthesia the only sign may be refractory hypotension.
• Action: STOP the transfusion. IM adrenaline 0.5 mg (repeat every 5-15 min; titrate to BP). IV fluids. Antihistamine. Hydrocortisone.
• Future transfusions: wash RBC and platelets (removes plasma); if plasma needed, use IgA-deficient donor plasma. Always document on the patient's record and MedicAlert. [1]

Prevention — the institutional safeguards

Modern transfusion safety rests on a layered set of safeguards that have driven the rate of serious reactions to historic lows:[1]

• Pre-storage leukodepletion: universal in most high-income countries — reduces FNHTR, CMV transmission (equivalent to CMV-negative for most indications), and may reduce TRALI and immunomodulation (TRIM).
• Male-predominant / never-pregnant-female plasma: dramatically reduces TRALI by minimising HLA-antibody exposure (the leading TRALI cause was multiparous female donors).
• Two-person bedside identity check: the single most effective intervention to prevent acute haemolytic (ABO incompatibility from clerical error). The wrong-blood-in-the-tube event is the precursor — patient misidentification at sample collection is the root cause.
• Group O plasma only from male or never-pregnant female donors: same TRALI rationale for plasma and apheresis platelets.
• Irradiation for the populations above (TA-GVHD prevention).
• Bacterial screening of platelets (culture or rapid nucleic-acid test).
• Pathogen reduction (INTERCEPT amotosalen/UVA; Mirasol riboflavin/UV) for plasma and platelets in some countries.
• Antenatal anti-D prophylaxis (prevents haemolytic disease of the newborn, the original transfusion-prevention programme). [1]

[1]

SaqBlocks — fellowship exam practice

[1] [1]

Red flags

Key trials and evidence

[1] [1] [1] [1] [1] [1] [1] [1]

Additional pearls and edge cases

Two-minute revision

[1]

References

1. [1]Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group N Engl J Med, 1999.PMID 9971864
2. [2]Holst LB, Haase N, Wetterslev J, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock N Engl J Med, 2014.PMID 25270275
3. [3]Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery N Engl J Med, 2011.PMID 22168590
4. [4]Murphy GJ, Pike K, Rogers CA, et al. Liberal or restrictive transfusion after cardiac surgery N Engl J Med, 2015.PMID 25760354
5. [5]Ducrocq G, Gonzalez-Juanatey JR, Puymirat E, et al. Effect of a Restrictive vs Liberal Blood Transfusion Strategy on Major Cardiovascular Events Among Patients With Acute Myocardial Infarction and Anemia: The REALITY Randomized Clinical Trial JAMA, 2021.PMID 33560322
6. [6]Holcomb JB, Tilley BC, Baraniuk S, et al. Transfusion of plasma, platelets, and red blood cells in a 1:1:1 vs a 1:1:2 ratio and mortality in patients with severe trauma: the PROPPR randomized clinical trial JAMA, 2015.PMID 25647203
7. [7]Mazer CD, Whitlock RP, Shehata N, et al. Restrictive versus Liberal Transfusion for Cardiac Surgery N Engl J Med, 2018.PMID 30586526
8. [8]Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding N Engl J Med, 2013.PMID 23281973