Transfusion & Perioperative Coagulation — Thresholds, Components, Anticoagulants, Reversal, HIT, Reactions, TXA

The anticoagulated, anaemic surgical patient is the viva examiner's favourite trap — so learn transfusion as a set of defensible defaults: transfuse red cells at 7 g/dL in the stable patient, platelets at 50 × 10³/µL before major surgery, plasma only for active coagulopathic bleeding, and fibrinogen when it falls first; interrupt warfarin and DOACs without routine bridging; continue aspirin only when thrombosis outweighs bleeding; reverse by drug-specific ladder; score suspected HIT with 4Ts and switch heparin the same hour; separate TACO from TRALI at the bedside; and give prophylactic tranexamic acid in major general surgery where the elective evidence now sits.^{[2][7][15][17][33][40][48][55][69][56][62][73]}

A 68-year-old woman on warfarin for atrial fibrillation with a haemoglobin of 118 g/L and a symptomatic incisional hernia awaits repair; elsewhere on the list sit a DOAC-treated colectomy, a thrombocytopenic liver resection, and a Jehovah's-Witness-adjacent restrictive-threshold debate. The examiner will ask for the transfusion trigger and its exceptions, the platelet number that permits the knife, whether plasma corrects the INR, what fibrinogen level demands replacement, whether old blood harms, what iron buys before surgery, how to stop and restart each anticoagulant, how to reverse the bleeder, how to recognise HIT and transfusion reactions, what VTE prophylaxis to prescribe, and whether tranexamic acid helps. This page answers each question with every number taken from the paper named beside it.^{[1][13][14][24][27][43][52][63][66][74]}

Red-Cell Thresholds — the Restrictive Standard

The default is restrictive: transfuse the haemodynamically stable inpatient when haemoglobin falls below 7 g/dL — the strong AABB 2023 recommendation at moderate certainty, with 7.5 g/dL acceptable in cardiac surgery and 8 g/dL in orthopaedic surgery or pre-existing cardiovascular disease.^[2] The Cochrane 2025 update (61 trials, 27,639 participants) defines the two strategies precisely — restrictive most commonly 7.0-8.0 g/dL against liberal 9.0-10.0 g/dL — and finds restriction cuts transfusion exposure by 42% (RR 0.58, high certainty) without moving 30-day mortality (RR 1.01, high certainty) or myocardial infarction, stroke, thromboembolism, or infection.^[1] Context still rules the individual decision: good practice weighs overall clinical context and alternative therapies for every patient rather than transfusing to a number alone.^[2]

TRICC is the anchor trial behind the rule: 838 euvolemic critically ill patients under 9.0 g/dL randomised to a below-7.0 g/dL trigger (maintained 7.0-9.0) versus a below-10.0 g/dL trigger (maintained 10.0-12.0), with 30-day mortality 18.7% versus 23.3% (P = 0.11) and hospital mortality significantly lower under restriction (22.3% versus 28.1%, P = 0.05).^[3] Restriction won outright in the less ill (APACHE II ≤20: 8.7% versus 16.1%, P = 0.03) and the under-55s (5.7% versus 13.0%, P = 0.02), tied in clinical cardiac disease (20.5% versus 22.9%, P = 0.69) — and the authors' own caution survives verbatim: restrictive is at least as effective and possibly superior, with the possible exception of acute myocardial infarction and unstable angina.^[3]

Threshold Exceptions — Brain, Gut Bleed, Bone

Two Cochrane exceptions modify the default in opposite directions. In gastrointestinal bleeding, restriction lowers 30-day mortality (RR 0.63 across 4 trials and 1574 participants) — the bleeder gets less blood, with resuscitation mechanics fenced to the GI-bleed topics.^[1] In brain injury, the direction reverses: liberal thresholds improve 6-to-12-month neurological outcome (RR 1.14 for unfavourable outcome under restriction, moderate certainty) — so the head-injured laparotomy patient keeps a higher trigger.^[1] Liberal strategy also collects more transfusion reactions (Peto OR 0.47, 18 studies, 11,505 participants), which is a second quiet argument for restriction wherever exceptions do not apply.^[1]

The orthopaedic caution tempers the elderly surgical patient: restrictive (mostly 8.0 g/dL or symptomatic anaemia) versus liberal (mostly 10.0 g/dL) raised cardiovascular events (RR 1.51, 8 trials, 3618 participants, no heterogeneity) regardless of pre-existing cardiovascular disease, driven by hip-fracture surgery (RR 1.51, P = 0.02) while elective arthroplasty did not reach significance — with infections, 30-day mortality, thromboembolism, and stroke identical between arms.^[6] And the thrombosis meta-analysis (30 trials, 17,334 participants) slightly favours restriction for venous events (RR 0.65, very-low certainty) with no arterial signal either way (stroke RR 0.83, MI RR 1.05) — supportive but too uncertain to mandate anything alone.^[5]

TRISS — the Septic Surgical Trigger

TRISS extends the 7 g/dL floor into septic shock: 998 ICU patients at ≤9 g/dL randomised to a single leukoreduced unit at ≤7 versus ≤9 g/dL, with the lower-trigger arm receiving a median of 1 unit against 4 units in the higher arm.^[4] Ninety-day mortality tied at 43.0% versus 45.0% (RR 0.94, P = 0.44), and ischemic events, severe adverse reactions, and life-support needs were identical — fewer units, same survival, same safety.^[4] Quote TRISS only for the trigger: fluids, vasopressors, and bundles belong to shock-surgical and postoperative-sepsis, and this topic claims no volume strategy from a transfusion trial.^[4]

Platelets — the AABB Ladder and Restrictive Policy

The 2025 AABB/ICTMG international guideline (GRADE over 21 randomised trials plus 13 observational studies) sets the ladder: transfuse below 10 × 10³/µL in hypoproliferative thrombocytopenia without bleeding, below 20 × 10³/µL for lumbar puncture, below 20/50 × 10³/µL for low-/high-risk interventional radiology, below 50 × 10³/µL before major non-neuraxial surgery — and withhold in dengue without major bleeding, in non-thrombocytopenic bypass without major haemorrhage, and in non-operative intracranial haemorrhage above 100 × 10³/µL even on antiplatelets.^[7] The pattern statement is explicit: restrictive strategies probably do not raise mortality or bleeding, and restriction cuts reactions, eases shortages, and saves costs.^[7] The 2015 AABB version already held the same surgical number (below 50 × 10⁹/L for major elective non-neuraxial surgery, weak, very-low certainty), the same single-apheresis-unit dose (greater doses no more effective; half-doses equally effective), and the below-20 × 10⁹/L central-line suggestion.^[8]

The Blood 2022 bedside review supplies the humility: platelets are the component most implicated in reactions, count increments vary, bleeding occurs despite prophylaxis policies, and trial harm signals may run through inflammation beyond haemostasis — so policy moves restrictive and risk-adapted rather than count-only.^[9] The 153-manuscript surgical and critical-care review confirms how wide practice variation remains and how much guideline development is still owed.^[11]

Perioperative Platelet Decisions — Therapeutic, Not Numeric

All pre-intervention platelet thresholds rest on weak evidence or expert opinion, and prophylactic perioperative transfusion evidence is very low — so where bleeding can be controlled until platelets arrive, a therapeutic strategy — transfuse for observed bleeding rather than prophylaxis by count — is the defensible alternative to routine pre-procedure correction, with the liberal-platelet harm signal in the critically ill reinforcing restraint.^[10][12] Decide on four inputs, not one: count plus function, added bleeding risks, and antiplatelet pharmacokinetics.^[10] The German review operationalises the same discipline: respond to first bleeding signs rather than the morning count below 10,000/µL, and reserve therapeutic-only strategies for stable low-risk patients.^[12]

For the bleeding surgical patient specifically, the UGIB systematic review (18 of 803 articles; only haemato-oncology empirical data) admits no high-level threshold evidence in GI bleeding and proposes 50 × 10⁹/L on expert opinion, rising to 100 × 10⁹/L after bypass or CNS trauma — quoted here for threshold endpoints only, with GI-bleed resuscitation fenced to upper-gi-bleeding.^[13]

Plasma — Overused, Narrowly Indicated

Plasma is overused worldwide on scant data; its defensible core is coagulopathy in the actively bleeding patient, with pre-procedure optimisation to coagulation-profile targets as the contested margin — and dosing must be weight-based because factor concentration tracks INR poorly.^[14] The perioperative FFP rule is two-sided: support exists for active bleeding with multiple factor deficiencies and for dilutional coagulopathy in major haemorrhage (where current dosing may even be inadequate), but transfusing the non-bleeder for mild screen prolongation is unsupported, likely benefit-free, and buys febrile, allergic, TACO, and TRALI risk — TRALI worst of all, lessened by male-donor-predominant plasma.^[15] Warfarin reversal by FFP depends on context and alternatives (read: PCC where available), and routine FFP for liver-disease numbers needs broad reconsideration.^[15] The adult-surgical systematic review closes with equipoise: profound RCT lack even for massive bleeding, and concentrate-based goal-directed therapy may beat ratio-driven resuscitation without powered-trial proof — so every plasma unit needs a benefit-risk sentence.^[16]

Fibrinogen and Cryoprecipitate — First to Fall

Fibrinogen is the first factor to critical levels (below 1.0 g/L) in major haemorrhage against a normal 2.0-4.5 g/L, and guidelines hold the line above 1.5 g/L — which makes fibrinogen, not factors in general, the therapeutic target.^[17] Concentrate is the engineered answer to plasma and cryoprecipitate: standardised dose, small volume, rapid administration, very good safety, viral inactivation as standard — with use associated with reduced or avoided allogeneic transfusion across cardiovascular surgery, PPH, and trauma.^[17] Prescribe it goal-directed and early — by A5/A10-FIBTEM or documented hypofibrinogenaemia — never pre-emptively in non-bleeders, on moderate-to-high-grade trauma and cardiac evidence with lower-grade PPH and liver-disease support.^[18] Then face the emergency-FC meta-analysis honestly (9 RCTs, 701 patients, emergency haemorrhage): mortality RR 1.24, zero 24-hour RBC sparing, no thrombotic difference, all very-low certainty with unbalanced severity — concentrate enthusiasm stays algorithmic, never default.^[19]

Storage Lesion — Biology Real, Age Irrelevant

A unit may be stored up to 42 days, and the Cochrane storage review (22 trials, 42,835 participants) finds age does not move mortality: shorter-versus-standard-issue in-hospital mortality RR 1.05, ICU RR 1.06, 30-day RR 1.04 — all moderate certainty — with trial sequential analysis now able to reject even a 5% mortality shift either way.^[20] Oldest-available-issue practice continues safely, no further RCTs are required, and future work belongs to multi-unit subgroups and quality factors rather than calendar age.^[20] The 16-trial meta-analysis (31,359 patients) concurs on mortality (RR 1.04, high certainty, information size reached for a 10% shift) while adding the two paradoxes: fresher blood carried more transfusion reactions (RR 1.35) and possibly more infection (RR 1.08) — demanding fresh units is not the safe move.^[21] The 64-paper synthesis (462,581 patients, mostly observational) explains why the scare started and ended: RCT mortality OR 0.91 (non-significant) while observational subsets in ICU, cardiac, and trauma suggested storage harm — randomisation dissolved the signal.^[22] The ICU-only RCT meta-analysis (6 trials, 18,240 patients) is flat on mortality (OR 1.04) and on ICU and hospital stay alike.^[23]

The two landmark negatives name the trials: ABLE randomised critically ill adults to under-8-day versus standard-issue blood (mean 6.1 versus 22.0 days) and found 90-day mortality 37.0% versus 35.3% (HR 1.1, P = 0.38) with no secondary-outcome differences.^[24] RECESS randomised complex cardiac surgery to ≤10-day versus ≥21-day blood (median 7 versus 28 days) and found MODS change +8.5 versus +8.7 (P = 0.44), with 7- and 28-day mortality tied and only hyperbilirubinaemia favouring fresher units.^[25] Keep the biology without overclaiming: storage lesions — oxidised lipids, extracellular vesicles, free heme and iron — are biochemically real, TRIM immunology runs through redox, inflammasome, and immune-metabolism pathways, and donor genetics, diet, and environment shape unit quality — but metabolic age does not yet select units, so mechanism informs vigilance, not inventory.^[26]

Preoperative Anaemia and IV Iron — Treat Early, Expect Honestly

Anaemia complicates 5-76% of preoperative courses and predicts transfusion, stay, and mortality, with iron deficiency the leading correctable cause — so work the anaemia up before booking the list.^[27] The Cochrane iron review (6 small RCTs, 372 participants, delivery 48 hours to 3 weeks pre-op) is a humility verdict: no transfusion-proportion difference versus placebo or standard care (RR 1.21, moderate certainty), though IV beats oral for preoperative haemoglobin rise (MD 1.23 g/dL) and ferritin — underpowered throughout, with adequately powered trials still demanded.^[27] The larger preop IV-iron meta-analysis (10 RCTs, 1039 participants) is modestly positive: transfusion down 16% (RR 0.84), haemoglobin up 7.15 g/L pre-op and 6.46 g/L past 4 weeks post-op, with no adverse-effect excess — at moderate certainty with bias caveats.^[28]

PREVENTT is the negative landmark every viva expects: 1000 mg ferric carboxymaltose (or saline placebo) over 15 minutes, 10-42 days before major open abdominal surgery — transfusion-or-death 29% versus 28% (RR 1.03, P = 0.84), transfusion counts 105 versus 111 (rate ratio 0.98, P = 0.93), safety endpoints tied.^[29] Late single-dose iron does not rescue the anaemic laparotomy patient. Timing reconciles the literature in the 2025 network meta-analysis (22 RCTs, 3026 patients): postoperative IV iron reduces transfusion (RR 0.80) and lifts day-30 haemoglobin, while preoperative iron rebuilds haemoglobin best by day 30 (MD 6.67 over postoperative) without moving transfusion rates — so individualise, start weeks early, and never promise PREVENTT-defying rescue.^[30] Postoperative IV iron raises haemoglobin (MD 4.51 g/L) where oral iron does nothing (MD 0.61 g/L, P = 0.66), and even non-anaemic major-surgery patients show fewer transfusions with preop iron (OR 0.54) plus a small day-1 haemoglobin edge — with standardised protocols still unwritten.^[31][32]

Warfarin Interruption and Bridging — BRIDGE Ends Routine Bridging

BRIDGE randomised 1884 AF patients interrupting warfarin (stopped 5 days pre-procedure, resumed within 24 hours after) to dalteparin 100 IU/kg twice daily versus placebo from 3 days before until 24 hours before and 5-10 days after: arterial thromboembolism 0.4% versus 0.3% (noninferior, P = 0.01) against major bleeding 1.3% versus 3.2% (RR 0.41, superior, P = 0.005).^[33] The default follows: simple interruption without bridging in the average AF patient — a practice-changing proof of concept that questions bridging beyond AF populations too.^[35] The bridging meta-analysis (13,808 AF patients, mean CHADS2 ~2.4) agrees: no mortality or stroke difference, major bleeding significantly less without bridging (OR 0.41, P = 0.0006).^[37]

The BRIDGE bleed-predictor analysis (41 major bleeds, median day 7) builds the high-risk counsel: bridging itself (OR 2.4), renal disease (2.9), high-bleed-risk procedure (2.9) at baseline, plus perioperative aspirin (3.6) and post-procedure INR >3.0 (2.1) as time-dependent predictors — with bleeding front-loaded into the first 10 days (OR 3.5).^[34] The RE-LY substudy warns that bridging raised major bleeding after both dabigatran interruption (6.5% versus 1.8%) and warfarin interruption (6.8% versus 1.6%) with no stroke or embolism gain — bridging is not the safe default under either drug.^[36] Guidance synthesis reserves bridging for high thromboembolic risk: most chronic-warfarin patients interrupt and resume safely, accepting a consistent two-to-threefold bleeding excess with no stroke reduction whenever bridging is used.^[39] The valve exception gets its own trial: AMBER's 553 bileaflet mechanical-aortic-valve patients showed thromboembolism under 0.5% across all strategies while full bridging drove major bleeding (3.0% versus 0%, P = 0.018) and clinically relevant bleeding (9.5% versus 2.0%) — preoperative-only bridging did not predict bleeding, but prospective confirmation is still owed.^[38]

DOAC Pause and Resume — PAUSE Without Bridging or Levels

PAUSE (3007 AF patients on apixaban, dabigatran, or rivaroxaban across 23 centres) standardised the protocol by pharmacokinetics, bleed risk, and creatinine clearance: omit 1 day before low-bleed-risk and 2 days before high-bleed-risk procedures, resume at 1 day and 2-3 days respectively — with 30-day major bleeding 1.35%, 0.90%, and 1.85% and arterial thromboembolism 0.16%, 0.60%, and 0.37% across the three cohorts, all without bridging or routine coagulation testing.^[40] The bleeding-predictor analysis adds two risk factors worth chasing (hypertension OR 1.79, prior bleeding OR 1.71) and one test worth skipping: the preoperative DOAC level did not improve prediction (AUC 0.71 unchanged), so no protocol adjustment is needed.^[41] The periprocedural synthesis is blunt: diminishing bridging role overall, no bridging role for DOAC interruption, bleed-risk-plus-renal-function management preferred over level testing — with the interruption plan communicated explicitly to patient and team.^[43]

Adjust at the edges from the update review: pause 24-96 hours by drug, renal and hepatic function, and surgical bleed risk; hold longer and more conservatively around neuraxial anaesthesia for haematoma-paraplegia risk; omit LMWH switching (bleeding up, thromboembolism unchanged); resume within 24 hours after minor procedures but 24-72 hours after major or high-bleed-risk surgery.^[42] Minimal-bleed-risk procedures may continue anticoagulation outright — one of four practical scenarios (continue, interrupt DOAC, manage dual antiplatelets, emergency surgery) the case-based review structures.^[44] General-surgery data ground the ward rule: the DOAC-cessation survey (98 operations, median 2 days off and 3 days to restart) saw 9.2% complications with clots clustering before resumption (mean day 3) and bleeds after (mean day 4), concentrated in heart failure, combined antiplatelets, high HAS-BLED, and low baseline haemoglobin.^[45] Mayo resumption data set the clock: withhold 48-72 hours pre-op, resume 48-72 hours post-op when safe, and do not drift past 72 hours — later resumption raised DVT.^[46] For high-bleed-risk and neuraxial cases, the PAUSE-2 pilot found residual levels under 30 ng/mL in ~95% of patients under both ASRA- and PAUSE-based approaches — feasibility data for the definitive trial, not yet a protocol.^[47]

Perioperative Antiplatelets — POISE-2 Restraint, Surgical Rules

POISE-2 randomised 10,010 at-risk noncardiac-surgery patients (initiation and continuation strata) to perioperative aspirin or placebo: death-or-MI tied at 7.0% versus 7.1% (HR 0.99, P = 0.92) while major bleeding rose (4.6% versus 3.8%, HR 1.23, P = 0.04), identically in both strata — routine perioperative aspirin neither protects nor spares.^[48] The Arch Surg ward rules then set daily practice: continue aspirin unless bleeding risk clearly exceeds thrombotic risk; stop clopidogrel at least 5 days before most elective surgery (recent drug-eluting stent excepted); stop GPIIb/IIIa inhibitors more than 12 hours pre-op — because premature withdrawal carries ~10% vascular-event risk with stent thrombosis potentially fatal after DES.^[49] The stent patient is the exception that proves restraint: recent DES means delay elective surgery or continue dual therapy by joint cardiology decision, never unilateral cessation.^[49] CABG-specific aspirin and clopidogrel timing stays fenced to cardiac surgery; this topic claims only the noncardiac general-surgery decision.^[48]

Reversal — Drug-Specific Ladders Plus a Bundle

Reversal has grown complex with the drugs themselves, and urgent or emergent reversal is where complexity bites — so run one ladder per drug class.^[51] For vitamin K antagonists: vitamin K IV or oral reverses effect within 12-48 hours for any bleeding, INR >10, or INR 4.5-10 with added bleeding risk — plus PCC for major bleeding, since PCC without vitamin K is incomplete.^[50] The dosing review is explicit for severe VKA bleeding: stop the VKA, give 10 mg IV vitamin K plus 50 units/kg 4-factor PCC promptly (3-factor PCC or FFP only as inferior substitutes).^[55] For heparins: protamine fully reverses unfractionated heparin and partially reverses LMWH — place it before PCC in the heparinised bleeder.^[50] For dabigatran: 5 g IV idarucizumab after stopping the drug; for factor-Xa inhibitors: andexanet alfa for apixaban and rivaroxaban where approved and available, otherwise 50 units/kg 4-factor PCC as the haemostatic bridge.^[55] The comprehensive DOAC-reversal review sets expectations honestly: idarucizumab is the agent of choice for dabigatran in major bleeding or urgent surgery, no commercial specific reversal existed for direct Xa inhibitors at review, and PCC or activated PCC (FEIBA) may contribute to haemostasis on limited in-vitro, animal, volunteer, and case-series evidence with uncertain safety, efficacy, and dosing.^[53]

The emergency logic precedes the drug: assess site, onset, severity, renal function, and concurrent antiplatelets/NSAIDs; note last intake and residual concentration; localise the bleed, interrupt anticoagulation, apply local measures — and where residual DOAC is expected and surgery cannot wait, give PCC and/or specific antidote preoperatively.^[54] And run the bundle, because reversal without resuscitation fails: fast INR and DOAC levels, a hospital bleeding team spanning haemostasis, lab, trauma, emergency, endoscopy, radiology, and surgery, volume and homeostasis control, and planned anticoagulation resumption — against a literature the reviewers admit is mostly uncontrolled and panel-judged.^[52]

Heparin-Induced Thrombocytopenia — 4Ts, Assay, Switch

HIT is a PF4-heparin antibody-mediated platelet-activating reaction with markedly raised thromboembolism risk — thrombocytopenia plus clots, never low platelets alone — and the 33-recommendation ASH 2018 guideline (McMaster GRADE, public comment) is the spine.^[69] Diagnose by 4Ts pretest probability, never gestalt (strong recommendation): skip laboratory testing and empiric treatment at low probability, and proceed stepwise through risk tools into immunoassay then functional assay at intermediate-high probability.^[69][70] On suspicion, stop all heparin immediately and start a non-heparin anticoagulant the same hour: argatroban, bivalirudin, danaparoid, fondaparinux, or a DOAC (conditional choices) — recognising that only argatroban holds regulatory approval on both sides of the Atlantic with monitoring burden and cost, that fondaparinux is increasingly used off-label, and that acute-phase DOACs belong only in stable patients without severe thrombotic complications and never in severe renal insufficiency (fondaparinux likewise contraindicated there).^[69][71] Cardiovascular-surgery, PCI, dialysis, and VTE-prophylaxis HIT scenarios sit inside the ASH special-situations section — quoted here only for general-surgical recognition and switching.^[69]

Transfusion Reactions — the Bedside Drill

Acute reactions present alike — febrile, allergic, haemolytic, bacterial, TRALI, TACO — so the BCSH drill is symptom-guided, not classification-first: stop the transfusion, assess, resuscitate, investigate the cause, plan future transfusion, and report to haemovigilance.^[56] Two absolutes: adrenaline first-line for anaphylaxis (SHOT still receives 30-40 anaphylaxis reports yearly), and transfuse only where patients are directly observed by staff trained in complications, with allergist follow-up after anaphylaxis.^[56] Prevention is exposure discipline: minimise component use as the fundamental avoidance, transfuse one unit at a time slowly with TACO-risk assessment (rate reduction plus possible diuretics), warm the massive-transfusion patient against hypothermia, limit pre-transfusion pulmonary inflammation where TRALI risk looms, and cross-check patient against unit rigorously because acute haemolysis is commonly human error.^[57] Haemovigilance numbers set expectations: 0.44 reactions per 1000 labile products over five years and 435,651 units in the Rabat series, febrile (41.36%) and allergic (35.60%) predominant, grade 1 in 87% — with the three deaths from ABO incompatibility (two) and TRALI (one) carrying the mortality lesson.^[58]

TRALI and TACO both strike as acute respiratory distress within 6 hours of transfusion (TACO's window extends to 12), jointly lead transfusion fatalities, and have no specific therapies — yet their physiologies oppose: TACO is hydrostatic cardiogenic oedema, TRALI is permeability noncardiogenic injury, both plausibly two-hit (patient condition first, transfusion product second).^[60] The 2019 redefinition drops "possible TRALI" for TRALI Type I (no ARDS risk factor) versus Type II (ARDS risk factor or mild existing ARDS); cases whose pre-transfusion 12-hour deterioration implicates the ARDS risk factor classify as ARDS instead; and TRALI stays a clinical diagnosis requiring no antibody detection — with every post-transfusion pulmonary oedema reported to the transfusion service for classification.^[59] Mechanism (exam-depth, not bedside-actionable): anti-leukocyte antibodies or biological response modifiers trigger multiple interlinked cellular pathways that converge on reactive-oxygen endothelial damage and leak.^[61]

TACO is the commoner killer to suspect first: the leading worldwide transfusion morbidity-mortality cause at 1-12% of at-risk recipients, defined as circulatory-overload pulmonary oedema within 6-12 hours.^[62] In the United States it caused over 30% of FDA-reported transfusion fatalities (2016-2020) yet stays underdiagnosed and underreported through two definition updates since 2018 that never reached every bedside.^[63][64] Biomarkers help where physiology overlaps: post-to-pre NT-proBNP ratio above 1.5 supports TACO, while BNP below 300 or NT-proBNP below 2000 pg/mL within 24 hours argues against — with poor specificity admitted in the critically ill and cytokines unable to split TACO from TRALI.^[63] The GI-bleed cohort warns surgeons directly: 12.3% cumulative TACO across 811 components, risk concentrated at age 80-plus, cardiac disease, and renal disease, 3-year survival 52% without versus 20% with TACO (HR 2.19), and marked under-reporting in exactly our transfusion-heavy population.^[65]

VTE Prophylaxis in General Surgery — LMWH or UFH

In general surgery specifically, LMWH and UFH are equally effective: 33,068 participants across 12 studies, VTE 1.3% versus 3.1% crude, pooled OR 0.77 (P = 0.078) and sensitivity-analysis OR 0.86 with heterogeneity eliminated — choose by patient and procedural factors rather than efficacy league tables.^[66] After hepatic resection, pharmacologic prophylaxis works and is safe (VTE OR 0.39 with bleeding and mortality unchanged), with the ethnicity subgroup (stronger Asian signal) informing guideline development rather than bedside selection.^[67] The ICU network meta-analysis (13 RCTs, 9619 patients) clarifies the agent comparison for generalisable endpoints only: LMWH reduces DVT versus control (OR 0.59, high certainty), UFH may (OR 0.82, low certainty), and LMWH probably beats UFH (OR 0.72, moderate certainty) — quoted for drug choice, with ICU care itself fenced out.^[68]

Tranexamic Acid in Elective Surgery — Prophylactic, Not Trauma-Timed

POISE-3 randomised 9535 noncardiac-surgery patients to 1 g IV tranexamic acid at the start and end of surgery versus placebo: composite bleeding (life-threatening, major, or critical-organ) 9.1% versus 11.7% (HR 0.76, superiority P < 0.001) against composite cardiovascular events 14.2% versus 13.9% (HR 1.02) — where noninferiority required a one-sided 97.5% upper boundary below 1.125 and returned 1.14 (P = 0.04): bleeding down, cardiovascular noninferiority not established.^[72] The general-surgery subgroup (3260 patients) confirms the benefit where surgeons operate: bleeding 8.0% versus 10.5% (HR 0.74, P = 0.01) with safety tied (HR 0.95, P = 0.63), including hepatopancreaticobiliary (HR 0.55) and colorectal (HR 0.67) signals without subtype interaction.^[73] The 2026 general-surgery meta-analysis (26 RCTs, 6976 patients) quantifies the same profile: blood loss down 35.85 mL, transfusion need RR 0.75, major bleeding RR 0.72 — with VTE (RR 1.09), mortality (RR 1.08), and stay all neutral.^[74] The TRACTION hospital-policy cluster trial (8273 high-transfusion-risk patients, 60.5% oncologic surgery) makes it operational: a TXA policy cut transfusion (7.4% versus 9.8%, RR 0.73) with 90-day VTE noninferior at a 1.46 margin (2.1% versus 2.1%, RR 0.96).^[75] Dose with the seizure caveat: high TXA doses associate with seizures and standardisation is still owed, though severe events stay rare and reviews find TXA safe across procedures.^[76] Topical TXA is the adjunct with limits: reduced postoperative blood loss across 16 major-surgery publications, but few studies cutting transfusions, ICU stay, or hospitalisation — promising, with safety data still needed.^[77] Trauma-timed TXA (CRASH-2 logic) belongs to massive-transfusion and PPH dosing to obstetrics; this topic claims elective prophylaxis only.^[72]

Exam Synthesis & Fence Map — what this topic owns

This topic owns elective perioperative transfusion science and coagulation management: restrictive RBC thresholds with TRICC/TRISS/AABB anchors and brain, gut-bleed, and orthopaedic exceptions; platelet thresholds with the 2025 AABB ladder and therapeutic discipline; plasma restraint with weight-based dosing honesty; fibrinogen-first replacement with concentrate discipline; storage-age evidence with ABLE/RECESS negatives and TRIM biology; preop anaemia with Cochrane humility and PREVENTT timing lessons; warfarin interruption without routine bridging per BRIDGE/AMBER; DOAC pause-resume per PAUSE without bridging or levels; POISE-2 antiplatelet restraint with surgical stop rules; drug-specific reversal ladders with the bleeding-team bundle; 4Ts HIT with same-hour switching; TRALI/TACO bedside separation with biomarker support; general-surgery VTE prophylaxis; and elective TXA from POISE-3 through TRACTION.^{[1][7][15][17][20][29][33][40][48][55][69][59][62][66][72]} Massive transfusion owns MTP triggers, ratios, whole blood, trauma TXA timing, citrate calcium, and resuscitation VHA; DIC owns ISTH scoring and shutdown phenotypes; damage-control resuscitation owns permissive hypotension; shock, sepsis, and multiorgan topics own resuscitation bundles; fluids owns crystalloids; acid-base owns lactate and gaps; wound-healing owns closure and leak biology; antimicrobials own every antibiotic; GI-bleed topics own bleed resuscitation; and haemostasis-thrombosis (pending) will own cascade biology.^[16][68] The one-paragraph viva story: transfuse red cells at 7 and platelets at 50, correct no mild INR with plasma, replace fibrinogen first and goal-directed, accept old blood, optimise anaemia weeks early because PREVENTT punishes lateness, interrupt warfarin and DOACs without bridging and resume within days, stop clopidogrel at 5 days but continue aspirin when thrombosis dominates, reverse by ladder (vitamin K plus 4F-PCC, protamine, idarucizumab, andexanet-or-PCC), score 4Ts and switch heparin immediately, separate TACO from TRALI by volume plus peptides, prevent VTE with heparin of either weight, and give prophylactic TXA in major general surgery.^{[2][13][18][24][30][34][47][49][55][70][63][66][75]}

Exam Pearls — the one-liners that score

• Restrictive 7-8 g/dL versus liberal 9-10 g/dL: 42% less exposure, mortality RR 1.01 — transfuse the stable patient below 7, allow 7.5 cardiac and 8 orthopaedic/CVD.^[1][2]
• TRICC (838 patients, below 7 vs below 10): 18.7% vs 23.3% at 30 days, hospital deaths down; TRISS (septic shock, ≤7 vs ≤9): 43.0% vs 45.0% at 90 days on 1 vs 4 units.^[3][4]
• Exceptions: brain injury wants liberal (RR 1.14 harm under restriction); GI bleed wants restrictive (RR 0.63); hip-fracture restriction raises CV events (RR 1.51).^[1][6]
• Platelets: below 10 hypo-proliferative, below 20 lines and taps, below 50 major surgery; no platelets for dengue, bypass without bleeding, or ICH above 100 — restrictive and risk-adapted.^[7][9]
• Periop platelets are weak-evidence expert opinion: transfuse therapeutically for bleeding, weigh function and drug kinetics, propose 50 × 10⁹/L in GI bleeding without high-level proof.^[10][13]
• Plasma: overused on scant data; yes for active multi-factor bleeding, no for mild screens or liver numbers; dose by weight, fear TRALI most.^[14][15]
• Fibrinogen falls first (below 1.0 critical, hold above 1.5): concentrate early and goal-directed, never pre-emptive; emergency-FC mortality RR 1.24 at very-low certainty tempers zeal.^[17][18][19]
• Storage: units last 42 days; Cochrane finds no mortality effect and oldest-issue continues; ABLE (HR 1.1) and RECESS (MODS tie) confirm; fresher units paradoxically react more (RR 1.35).^{[20][24][25][21]}
• Iron: anaemia in 5-76%, Cochrane underpowered (RR 1.21); PREVENTT 1000 mg at 10-42 days changed nothing (RR 1.03); postop IV iron cuts transfusion (RR 0.80) while preop rebuilds Hb best — start early.^[27][29][30]
• BRIDGE (dalteparin 100 IU/kg vs placebo): clots 0.4% vs 0.3%, bleeds 1.3% vs 3.2% — interrupt without bridging; bleed predictors are bridging, renal disease, high-risk procedure, aspirin, INR >3, first 10 days.^[33][34]
• AMBER valves: clots below 0.5% all strategies, full bridging bleeds 3.0% vs 0% — even mechanical aortic valves get minimal-bridging individualisation.^[38]
• PAUSE: omit 1/2 days by bleed risk, resume 1/2-3 days, no bridging or levels (bleeds ≤1.85%, clots ≤0.60%); hypertension and prior bleed predict; resume within 72 hours.^[40][41][46]
• POISE-2 (10,010 patients): aspirin changed nothing on death/MI (HR 0.99) and raised major bleeding (HR 1.23) — continue aspirin only when thrombosis dominates; stop clopidogrel 5 days, GPIIb/IIIa 12 hours out.^[48][49]
• Reversal: VKA bleed gets 10 mg IV vitamin K plus 50 U/kg 4F-PCC; heparin gets protamine (full UFH, partial LMWH); dabigatran gets 5 g idarucizumab; Xa inhibitors get andexanet or 50 U/kg PCC — inside a bleeding-team bundle.^[55][50][52]
• HIT: PF4-antibody clots-with-low-platelets; 4Ts over gestalt, skip testing when low; stop all heparin and switch (argatroban the dual-approved one) the same hour; mind renal limits.^[69][71]
• Reactions: BCSH symptom-guided drill with adrenaline first-line in observed settings; one-unit-at-a-time slow transfusion; cross-checks stop ABO deaths; febrile/allergic dominate at 0.44 per 1000.^[56][57][58]
• TRALI Type I/II replaces possible TRALI (clinical, no antibodies needed); TACO is 6-12-hour hydrostatic overload at 1-12% of at-risk; NT-proBNP ratio above 1.5 supports, BNP below 300 or NT-proBNP below 2000 argues against.^[59][62][63]
• VTE prophylaxis in general surgery: LMWH equals UFH (OR 0.77, NS) — pick by kidneys, timing, and HIT history; hepatic resection prophylaxis works (OR 0.39).^[66][67]
• Elective TXA (1 g start and end): POISE-3 bleeding HR 0.76 with CV noninferiority unestablished; GS subgroup HR 0.74; 2026 meta transfusion RR 0.75 and major-bleed RR 0.72; TRACTION policy transfusion RR 0.73 with VTE noninferior.^{[72][73][74][75]}

References77ShowHide

1. [1]Carson JL, et al. Transfusion thresholds and other strategies for guiding red blood cell transfusion. Cochrane Database Syst Rev, 2025.PMID 41114449
2. [2]Carson JL, et al. Red Blood Cell Transfusion: 2023 AABB International Guidelines. JAMA, 2023.PMID 37824153
3. [3]Hébert PC, 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
4. [4]Holst LB, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. N Engl J Med, 2014.PMID 25270275
5. [5]Maimaitiming M, et al. Impact of restrictive red blood cell transfusion strategy on thrombosis-related events: A meta-analysis and systematic review. Vox Sang, 2022.PMID 35332942
6. [6]Gu WJ, et al. Restrictive Versus Liberal Strategy for Red Blood-Cell Transfusion: A Systematic Review and Meta-Analysis in Orthopaedic Patients. J Bone Joint Surg Am, 2018.PMID 29664857
7. [7]Metcalf RA, et al. Platelet Transfusion: 2025 AABB and ICTMG International Clinical Practice Guidelines. JAMA, 2025.PMID 40440268
8. [8]Kaufman RM, et al. Platelet transfusion: a clinical practice guideline from the AABB. Ann Intern Med, 2015.PMID 25383671
9. [9]Stanworth SJ, et al. How I use platelet transfusions. Blood, 2022.PMID 35926105
10. [10]Thiele T, et al. Platelet Transfusion in Perioperative Medicine. Semin Thromb Hemost, 2020.PMID 31830766
11. [11]Etchill EW, et al. Platelet Transfusion in Critical Care and Surgery: Evidence-Based Review of Contemporary Practice and Future Directions. Shock, 2017.PMID 27849676
12. [12]Wandt H, et al. Platelet transfusion in hematology, oncology and surgery. Dtsch Arztebl Int, 2014.PMID 25512006
13. [13]Razzaghi A, et al. Platelet transfusion threshold in patients with upper gastrointestinal bleeding: a systematic review. J Clin Gastroenterol, 2012.PMID 22688143
14. [14]Benson MA, et al. Plasma: indications, controversies, and opportunities. Postgrad Med, 2024.PMID 38362605
15. [15]Kor DJ, et al. Perioperative coagulation management--fresh frozen plasma. Best Pract Res Clin Anaesthesiol, 2010.PMID 20402170
16. [16]Adam EH, et al. Plasma Transfusion Practice in Adult Surgical Patients: Systematic Review of the Literature. Transfus Med Hemother, 2020.PMID 33173453
17. [17]Levy JH, et al. Fibrinogen as a therapeutic target for bleeding: a review of critical levels and replacement therapy. Transfusion, 2014.PMID 24117955
18. [18]Leal-Noval SR, et al. Current perspective on fibrinogen concentrate in critical bleeding. Expert Rev Clin Pharmacol, 2020.PMID 32479129
19. [19]Itagaki Y, et al. Emergency administration of fibrinogen concentrate for haemorrhage: systematic review and meta-analysis. World J Emerg Surg, 2023.PMID 36998084
20. [20]Shah A, et al. Transfusion of red blood cells stored for shorter versus longer duration for all conditions. Cochrane Database Syst Rev, 2018.PMID 30578732
21. [21]McQuilten ZK, et al. Effect of age of red cells for transfusion on patient outcomes: a systematic review and meta-analysis. Transfus Med Rev, 2018.PMID 29526337
22. [22]Ng MS, et al. Effects of packed red blood cell storage duration on post-transfusion clinical outcomes: a meta-analysis and systematic review. Intensive Care Med, 2015.PMID 26438223
23. [23]Zhou X, et al. Association between storage age of transfused red blood cells and clinical outcomes in critically ill adults: A meta-analysis of randomized controlled trials. Med Intensiva (Engl Ed), 2019.PMID 30241932
24. [24]Lacroix J, et al. Age of transfused blood in critically ill adults. N Engl J Med, 2015.PMID 25853745
25. [25]Steiner ME, et al. Effects of red-cell storage duration on patients undergoing cardiac surgery. N Engl J Med, 2015.PMID 25853746
26. [26]D'Alessandro A, et al. From metabolomics to transfusion-associated immunomodulation. Curr Opin Immunol, 2025.PMID 40848579
27. [27]Ng O, et al. Iron therapy for preoperative anaemia. Cochrane Database Syst Rev, 2019.PMID 31811820
28. [28]Elhenawy AM, et al. Role of preoperative intravenous iron therapy to correct anemia before major surgery: a systematic review and meta-analysis. Syst Rev, 2021.PMID 33485392
29. [29]Richards T, et al. Preoperative intravenous iron to treat anaemia before major abdominal surgery (PREVENTT): a randomised, double-blind, controlled trial. Lancet, 2020.PMID 32896294
30. [30]Liu C, et al. Timing of intravenous iron for treatment of anaemia in surgical patients: a systematic review and network meta-analysis. EClinicalMedicine, 2025.PMID 40727015
31. [31]Lim J, et al. The use of iron after surgery: a systematic review and meta-analysis. Anaesthesia, 2025.PMID 40127959
32. [32]Toledo FV, et al. Preoperative iron supplementation in non-anemic patients undergoing major surgery: a systematic review and meta-analysis. Braz J Anesthesiol, 2025.PMID 40189047
33. [33]Douketis JD, et al. Perioperative Bridging Anticoagulation in Patients with Atrial Fibrillation. N Engl J Med, 2015.PMID 26095867
34. [34]Clark NP, et al. Predictors of perioperative major bleeding in patients who interrupt warfarin for an elective surgery or procedure: Analysis of the BRIDGE trial. Am Heart J, 2018.PMID 29224638
35. [35]Nazha B, et al. The BRIDGE trial: What the hospitalist should know. J Hosp Med, 2016.PMID 27098835
36. [36]Douketis JD, et al. Perioperative bridging anticoagulation during dabigatran or warfarin interruption among patients who had an elective surgery or procedure. Substudy of the RE-LY trial. Thromb Haemost, 2015.PMID 25472710
37. [37]Ayoub K, et al. Perioperative Heparin Bridging in Atrial Fibrillation Patients Requiring Temporary Interruption of Anticoagulation: Evidence from Meta-analysis. J Stroke Cerebrovasc Dis, 2016.PMID 27289185
38. [38]Godoy A, et al. Perioperative Heparin Bridging in Patients With Mechanical Aortic Valves Undergoing Elective Surgery: The AMBER Study. JACC Adv, 2026.PMID 41579829
39. [39]Nikolakopoulos I, et al. Heparin Bridging Therapy for Patients on Chronic Oral Anticoagulants in Periprocedural Settings. Semin Thromb Hemost, 2020.PMID 31830765
40. [40]Douketis JD, et al. Perioperative Management of Patients With Atrial Fibrillation Receiving a Direct Oral Anticoagulant. JAMA Intern Med, 2019.PMID 31380891
41. [41]Tafur AJ, et al. Predictors of Bleeding in the Perioperative Anticoagulant Use for Surgery Evaluation Study. J Am Heart Assoc, 2020.PMID 32969288
42. [42]Koscielny J, et al. Update on Direct Oral AntiCoagulants (DOACs). Hamostaseologie, 2017.PMID 29582929
43. [43]Shaw JR, et al. Periprocedural Management of Oral Anticoagulation. Med Clin North Am, 2020.PMID 32505262
44. [44]Chen AT, et al. Perioperative management of antithrombotic therapy: a case-based narrative review. Intern Emerg Med, 2022.PMID 34652572
45. [45]Kawabata M, et al. A Survey of Direct Oral Anticoagulant Cessation in General Surgery and Outcomes in Patients with Nonvalvular Atrial Fibrillation. Int Heart J, 2020.PMID 32921660
46. [46]Irizarry Alvarado JM, et al. Perioperative Resumption of Direct Oral Anticoagulants: Review of the Mayo Clinic Experience. South Med J, 2021.PMID 34215899
47. [47]Douketis JD, et al. Perioperative management of direct oral anticoagulants in patients having a high-bleed-risk surgery or neuraxial procedure: the Perioperative Anticoagulant Use for Surgery Evaluation (PAUSE)-2 pilot randomized trial. J Thromb Haemost, 2025.PMID 40086754
48. [48]Devereaux PJ, et al. Aspirin in patients undergoing noncardiac surgery. N Engl J Med, 2014.PMID 24679062
49. [49]O'Riordan JM, et al. Antiplatelet agents in the perioperative period. Arch Surg, 2009.PMID 19153328
50. [50]Koscielny J, et al. How Do I Reverse Oral and Parenteral Anticoagulants? Hamostaseologie, 2020.PMID 32455460
51. [51]Holzmacher JL, et al. Indications and Methods of Anticoagulation Reversal. Surg Clin North Am, 2017.PMID 29132510
52. [52]Moia M, et al. Reversal agents for oral anticoagulant-associated major or life-threatening bleeding. Intern Emerg Med, 2019.PMID 31446606
53. [53]Shaw JR, et al. Pharmacological reversal of the direct oral anticoagulants-A comprehensive review of the literature. Res Pract Thromb Haemost, 2018.PMID 30046727
54. [54]Lindhoff-Last E, et al. Direct oral anticoagulants (DOAC) - Management of emergency situations. Hamostaseologie, 2017.PMID 29582928
55. [55]Raimondi P, et al. Reversal Agents for Oral Antiplatelet and Anticoagulant Treatment During Bleeding Events: Current Strategies. Curr Pharm Des, 2017.PMID 27917717
56. [56]Tinegate H, et al. Guideline on the investigation and management of acute transfusion reactions. Prepared by the BCSH Blood Transfusion Task Force. Br J Haematol, 2012.PMID 22928769
57. [57]Aubron C, et al. How clinicians can minimize transfusion-related adverse events? Transfus Clin Biol, 2018.PMID 30197000
58. [58]Lemssahli I, et al. Review of haemovigilance at the Rabat Regional Blood Transfusion Centre in Morocco (2017-2021). Pan Afr Med J, 2024.PMID 38646139
59. [59]Vlaar APJ, et al. A consensus redefinition of transfusion-related acute lung injury. Transfusion, 2019.PMID 30993745
60. [60]Semple JW, et al. Transfusion-associated circulatory overload and transfusion-related acute lung injury. Blood, 2019.PMID 30808638
61. [61]Tung JP, et al. Transfusion-related acute lung injury (TRALI): Potential pathways of development, strategies for prevention and treatment, and future research directions. Blood Rev, 2022.PMID 35065815
62. [62]Bosboom JJ, et al. Transfusion-Associated Circulatory Overload: A Clinical Perspective. Transfus Med Rev, 2019.PMID 30853167
63. [63]Klanderman RB, et al. Transfusion-associated circulatory overload-a systematic review of diagnostic biomarkers. Transfusion, 2019.PMID 30488959
64. [64]Smith CM, et al. CE: Recognizing Transfusion-Associated Circulatory Overload. Am J Nurs, 2023.PMID 37882401
65. [65]Piccin A, et al. Transfusion-associated circulatory overload in gastroenterology. Blood Transfus, 2021.PMID 33000753
66. [66]Nair P, et al. Low-molecular weight vs. unfractionated heparin for prevention of venous thromboembolism in general surgery: a meta-analysis. Updates Surg, 2021.PMID 32880870
67. [67]Zhang W, et al. Unfractionated heparin or low-molecular-weight heparin for venous thromboembolism prophylaxis after hepatic resection: A meta-analysis. Medicine (Baltimore), 2022.PMID 36401460
68. [68]Fernando SM, et al. VTE Prophylaxis in Critically Ill Adults: A Systematic Review and Network Meta-analysis. Chest, 2022.PMID 34419428
69. [69]Cuker A, et al. American Society of Hematology 2018 guidelines for management of venous thromboembolism: heparin-induced thrombocytopenia. Blood Adv, 2018.PMID 30482768
70. [70]May J, et al. Practical guide to the diagnosis and management of heparin-induced thrombocytopenia. Hematology Am Soc Hematol Educ Program, 2024.PMID 39644042
71. [71]Hvas AM, et al. Heparin-induced thrombocytopenia: pathophysiology, diagnosis and treatment. Expert Rev Hematol, 2021.PMID 33736552
72. [72]Devereaux PJ, et al. Tranexamic Acid in Patients Undergoing Noncardiac Surgery. N Engl J Med, 2022.PMID 35363452
73. [73]Park LJ, et al. Safety and Efficacy of Tranexamic Acid in General Surgery. JAMA Surg, 2025.PMID 39813061
74. [74]Delgado LM, et al. Perioperative Use of Tranexamic Acid in General Surgery: A Systematic Review and Meta-Analysis. JAMA Surg, 2026.PMID 41405985
75. [75]Houston BL, et al. Hospital Policy of Tranexamic Acid to Reduce Transfusion in Major Noncardiac Surgery. N Engl J Med, 2026.PMID 42267805
76. [76]Colomina MJ, et al. Clinical use of tranexamic acid: evidences and controversies. Braz J Anesthesiol, 2022.PMID 34626756
77. [77]Ipema HJ, et al. Use of topical tranexamic acid or aminocaproic acid to prevent bleeding after major surgical procedures. Ann Pharmacother, 2012.PMID 22202494