Publication

• Title: Viscoelastic haemostatic assay augmented protocols for major trauma haemorrhage (ITACTIC): a randomized, controlled trial
• Acronym: ITACTIC
• Year: 2021
• Journal published in: Intensive Care Medicine
• Citation: Baksaas-Aasen K, Gall LS, Stensballe J, et al. Viscoelastic haemostatic assay augmented protocols for major trauma haemorrhage (ITACTIC): a randomized, controlled trial. Intensive Care Med. 2021;47:49-59.

Context & Rationale

• Background

  • Major trauma haemorrhage remains a key cause of potentially preventable early death, and is strongly linked to trauma-induced coagulopathy.
  • Contemporary damage-control resuscitation (rapid haemorrhage control, minimising crystalloids, tranexamic acid, and balanced component therapy) improves outcomes, but does not guarantee timely correction of the specific coagulopathy phenotype in an individual patient.
  • Conventional coagulation tests (PT/INR, fibrinogen, platelet count) are commonly used to guide haemostatic therapy, but are plasma-based and can have clinically important turnaround delays during active bleeding.
  • Viscoelastic haemostatic assays (VHA; ROTEM/TEG) offer point-of-care whole-blood assessment of clot formation, strength and lysis, potentially enabling earlier and more targeted supplementation (e.g., fibrinogen, platelets, plasma) than CCT-based approaches.
  • The iTACTIC programme developed and published pragmatic, standardised goal-directed algorithms and the RCT protocol to test whether VHA-guided care improves clinically relevant outcomes versus CCT-guided care.¹²
  • High-quality evidence supporting balanced component resuscitation (e.g., fixed ratio approaches) existed, but the incremental value of adding “precision” haemostatic supplementation based on rapid assays remained uncertain in multicentre settings.³
  • Prior randomised evidence for VHA-guided trauma resuscitation was limited and predominantly single-centre; a pragmatic trial suggested benefit versus CCT-guided care, but generalisability to diverse trauma systems and modern massive haemorrhage protocols required validation.⁴
• Research Question/Hypothesis

  • In adult trauma patients with ongoing bleeding requiring activation of a massive haemorrhage protocol, does a VHA-augmented goal-directed algorithm improve the proportion of patients alive and free of massive transfusion at 24 hours compared with a CCT-augmented goal-directed algorithm?
  • Hypothesis: VHA-guided management would enable earlier targeted haemostatic therapy, reduce massive transfusion, and improve survival.
• Why This Matters

  • VHA implementation is resource-intensive (equipment, training, quality control) and can increase use of costly haemostatic products; an outcome benefit would justify adoption and standardisation.
  • A neutral (or harmful) result would support focusing on optimising conventional massive haemorrhage pathways and rapid laboratory processes rather than routine VHA expansion.
  • Defining whether (and in whom) “precision” haemostatic algorithms improve patient-centred outcomes informs guideline recommendations and trial design for trauma haemorrhage care.

Design & Methods

• Research Question: Among adult trauma patients with ongoing haemorrhage managed with a massive haemorrhage protocol, does VHA-guided goal-directed haemostatic therapy improve “alive and free of massive transfusion at 24 hours” versus CCT-guided goal-directed therapy?
• Study Type: Investigator-initiated, pragmatic, multicentre, international, parallel-group randomised controlled trial (1:1), stratified by centre (block randomisation); conducted in emergency department trauma resuscitation at 7 European major trauma centres; open-label to treating clinicians with blinded research personnel for safety/outcome data collection.
• Population:
  • Setting: Major trauma centres; enrolment in the emergency department during active haemorrhage with massive haemorrhage protocol (MHP) activation.
  • Inclusion criteria: Adult trauma; clinical signs of bleeding triggering local MHP activation; red blood cell (RBC) transfusion initiated; randomisation within 3 hours of injury and within 1 hour of emergency department admission.
  • Exclusion criteria: No additional exclusion criteria beyond not meeting inclusion thresholds; patients who did not provide/retain consent were excluded post-randomisation (withdrawal of consent).
• Intervention:
  • VHA-guided goal-directed algorithm (plus standard MHP care): baseline and repeat testing after every 4 RBC units until haemostasis; VHA performed with ROTEM Sigma or TEG 6s (RapidTEG/functional fibrinogen).
  • ROTEM thresholds: FIBTEM CA5 <10 mm → additional 4 g fibrinogen equivalent; (EXTEM CA5 − FIBTEM CA5) <30 mm → 1 pool of platelets; EXTEM CA5 ≥40 mm and EXTEM CT >80 s → 4 units plasma; EXTEM LI30 <85% → additional 1 g tranexamic acid (bolus).
  • RapidTEG thresholds: FF-TEG MA <20 mm → additional 4 g fibrinogen equivalent; (RapidTEG MA − FF-TEG MA) <45 mm → 1 pool of platelets; RapidTEG MA >65 mm and RapidTEG ACT >120 s → 4 units plasma; LY30 >10% → additional 1 g tranexamic acid (bolus).
  • Product approach: Fibrinogen delivered as cryoprecipitate and/or fibrinogen concentrate depending on centre (reported as “fibrinogen equivalent dose”); platelet pool defined as 4 single units.
• Comparison:
  • CCT-guided goal-directed algorithm (plus standard MHP care): baseline and repeat testing after every 4 RBC units until haemostasis; laboratory PT ratio (PTr)/INR, fibrinogen, platelet count used to trigger interventions.
  • CCT thresholds: Fibrinogen <2 g/L → additional 4 g fibrinogen equivalent; platelet count <100×10⁹/L → 1 pool of platelets; INR >1.2 with fibrinogen ≥2 g/L → 4 units plasma.
• Blinding: Treating clinicians were unblinded (intervention inherently test-dependent); research personnel collecting safety/outcome data were blinded to allocation (mitigating detection bias for adjudicated outcomes).
• Statistics: Power calculation: 392 patients required to detect an absolute improvement in the primary composite outcome from 72% to 85% (i.e., event rate reduction from 28% to 15%) with 80% power at a two-sided 5% significance level; sample inflated to 196 per group to allow 15% post-randomisation attrition. Primary analysis was intention-to-treat; effect estimates for binary outcomes reported as odds ratios from logistic regression; a per-protocol analysis was prespecified.
• Follow-Up Period: Outcomes assessed through 24 hours (primary endpoint), 28 days, and 90 days (mortality and health-status follow-up); ventilator-free and ICU-free days reported to day 28.

Key Results

This trial was not stopped early. A planned interim analysis occurred after 100 patients, and the data monitoring committee reviewed outcomes after every additional 50 patients.

• Overall, VHA-guided goal-directed therapy did not improve the primary composite endpoint (67% vs 64%) nor overall mortality through 90 days versus CCT-guided therapy.
• Despite similar RBC and plasma exposure, VHA guidance increased early delivery of protocolised haemostatic interventions and fibrinogen supplementation (see Internal Validity for quantified separation), without translating into improved haemostasis time or organ-support-free days.
• A prespecified severe TBI subgroup demonstrated a lower 28-day mortality with VHA guidance (44% vs 74%; interaction P=0.016), but this requires confirmatory evidence given small sample size and multiplicity.

Internal Validity

• Randomisation and allocation: Central 1:1 allocation using site-stratified block randomisation; allocation concealment via sealed opaque envelopes; baseline characteristics broadly similar between groups (e.g., median ISS 26 vs 26; pre-baseline RBC 2 vs 2 units).
• Dropout / post-randomisation exclusions: 411 randomised; 15 excluded after withdrawal of consent (8 CCT; 7 VHA), leaving 396 in the intention-to-treat population (195 CCT; 201 VHA); withdrawal was balanced but always carries potential for attrition bias.
• Performance / detection bias: Treating clinicians were unblinded, creating risk of performance bias (transfusion thresholds, escalation, and haemostasis timing); key endpoints include objective components (mortality; ≥10 RBC threshold) but transfusion exposure is still clinician- and context-dependent.
• Protocol adherence and analysed populations: Per-protocol population was 313/396 (79%) (163 CCT; 150 VHA); exclusions (n=83) included early haemostasis within 1 hour (54), death within 1 hour (16), wrong-arm treatment (8), and not meeting entry criteria (5).
• Baseline illness severity and “room to benefit”: Coagulopathy burden at baseline was lower than often assumed for MHP activations (PTr >1.2: 58/181 [32%] CCT vs 44/175 [25%] VHA); near-universal pre-baseline TXA bolus (98% vs 94%) and early RBC exposure (median 2 units in both) suggest contemporary high-quality resuscitation in both groups.
• Heterogeneity: Pragmatic multicentre design increases realism but introduces variability (site blood-product availability including cryoprecipitate vs fibrinogen concentrate; use of ROTEM vs TEG platforms); centre-stratified randomisation mitigates but does not remove treatment-effect heterogeneity.
• Timing: Randomisation occurred within 3 hours of injury and within 1 hour of ED admission, appropriate for haemorrhage physiology; however, a meaningful proportion of deaths occurred before haemostasis (19/50 [38%] VHA vs 24/54 [44%] CCT), limiting the window for assay-guided interventions in the sickest patients.
• Dose: Algorithm dosing used fixed increments (e.g., 4 g fibrinogen-equivalent; 4 units plasma; 1 pool platelets; 1 g TXA) which supports standardisation but may under-dose some phenotypes of severe coagulopathy.
• Separation of the variable of interest:
  • Received any protocol “study intervention” before haemostasis: 120/178 (67%) VHA vs 62/170 (36%) CCT.
  • Time to first study intervention before haemostasis (median [IQR]): 61 (48–85) minutes VHA vs 80 (60–106) minutes CCT.
  • Fibrinogen equivalent dose before haemostasis (median [IQR]): 4 (0–4) g VHA vs 0 (0–4) g CCT.
  • RBC and plasma exposure before haemostasis (median [IQR]): RBC 3 (1–6) vs 3 (1–6); plasma 4 (1–6) vs 4 (2–6).
  • By 24 hours (median [IQR]): RBC 6 (3–10) vs 6 (4–10); plasma 6 (3–10) vs 7 (4–11); fibrinogen 4 (0–4) vs 3 (0–4); platelets 2 (1–3) vs 1 (0–2).
• Adjunctive therapy and crossover: Wrong-arm treatment occurred in 8 patients total (6 CCT; 2 VHA) and contributed to per-protocol exclusions; otherwise, both groups were managed under similar local MHPs (balanced component therapy, TXA, limited crystalloids).
• Outcome assessment: Mortality outcomes are objective; thromboembolic events were reported as symptomatic events; functional outcomes (EQ-5D) had substantial missingness and may be subject to survivorship and response bias.
• Statistical rigor: The primary endpoint was analysed by intention-to-treat with logistic regression; however, the assumed effect size (28% to 15% event rate) was optimistic relative to observed event rates, increasing risk of type II error.

Conclusion on Internal Validity: Overall, internal validity is moderate: randomisation and baseline balance were strong, and outcome ascertainment for mortality was robust, but open-label delivery, limited early-phase separation beyond fibrinogen use, substantial per-protocol exclusions, and a lower-than-expected coagulopathy burden reduce confidence that the trial could detect modest but clinically important effects.

External Validity

• Population representativeness: Broad inclusion with no additional exclusion criteria increases generalisability to adult trauma patients triggering an MHP and receiving RBCs; a sizeable subgroup had severe TBI (18–20%) and baseline shock physiology (median SBP ~90–95 mmHg).
• System context: Conducted in well-resourced European major trauma centres with established MHPs, near-universal tranexamic acid, and ready access to blood components and laboratory services; applicability may differ where MHP activation thresholds, transfusion logistics, or factor availability vary.
• Intervention feasibility: VHA algorithms required ROTEM/TEG platforms and trained operators; centres without these resources may not replicate the intervention or turnaround times.
• Applicability to different transfusion strategies: Findings most directly apply to systems already practising balanced component therapy and early TXA; effects could differ where baseline care is less standardised or where fibrinogen concentrates/cryoprecipitate are used differently.

Conclusion on External Validity: External validity is moderate-to-high for high-income trauma systems with mature massive haemorrhage pathways; generalisability is more limited in resource-constrained settings or where prehospital/hospital transfusion logistics and haemostatic product access differ substantially.

Strengths & Limitations

• Strengths:
  • Pragmatic, multicentre randomised design across 7 major trauma centres, enhancing relevance to real-world trauma resuscitation.
  • Standardised, protocolised algorithms in both arms with repeated testing (baseline and after every 4 RBC units), enabling a fair comparison of VHA- versus CCT-augmented goal-directed strategies.
  • Clinically relevant primary endpoint anchored in early survival and transfusion intensity.
  • High uptake of contemporary co-interventions (e.g., TXA and balanced transfusion) reflecting modern care.
• Limitations:
  • Open-label design with clinician-driven transfusion decisions; the composite primary endpoint includes a component (massive transfusion) susceptible to practice variation.
  • Observed baseline coagulopathy prevalence and event rates were lower than assumed in the power calculation, increasing risk of a false-negative result for modest effects.
  • Limited early-phase separation beyond increased/earlier fibrinogen and study interventions; by 24 hours, RBC and plasma exposure were similar.
  • Substantial missing data for some longer-term outcomes (e.g., 90-day mortality status and EQ-5D), limiting inference for patient-centred recovery endpoints.
  • Heterogeneity in VHA platforms and haemostatic product availability (cryoprecipitate vs fibrinogen concentrate) may dilute device- or product-specific effects.

Interpretation & Why It Matters

• Clinical practice

  • In high-performing trauma systems already delivering early TXA and balanced component resuscitation, replacing CCT-guided goal-directed algorithms with VHA-guided algorithms did not improve early composite outcomes or mortality.
  • VHA guidance substantially increased early use of fibrinogen supplementation, without measurable improvements in time to haemostasis, organ-support-free days, or overall survival.
• Mechanistic inference

  • The trial demonstrates that achieving faster and more frequent goal-directed product delivery (particularly fibrinogen) is not, by itself, sufficient to improve outcomes when baseline care is strong and coagulopathy burden is modest.
  • Implementation remains a key “last mile” challenge: the pathway from test result to timely targeted therapy may be more outcome-relevant than assay choice alone.
• Research implications

  • The severe TBI subgroup signal (lower 28-day mortality with VHA) is hypothesis-generating and supports confirmatory trials focusing on clearly defined coagulopathy phenotypes and clinically plausible mechanisms in neurotrauma.
  • Future trials should explicitly measure and optimise the full implementation chain (sampling → turnaround → decision → delivery), not only diagnostic performance.

Controversies & Subsequent Evidence

• Composite primary endpoint and interpretability:
  • The primary outcome combined survival and freedom from massive transfusion; while clinically meaningful, the transfusion component is partially behaviour-dependent in an open-label trial and may be influenced by local MHP practice variation.
  • The accompanying editorial framed this as a key challenge for validating “precision medicine protocols” in urgent care, emphasising the gap between diagnostic granularity and measurable patient benefit.⁵
• Patient selection and dilution of treatment effect:
  • Baseline coagulopathy (PTr >1.2) occurred in ~25–32% at enrolment, lower than expected for an MHP population, raising the possibility that many participants were unlikely to benefit from assay-driven haemostatic supplementation.
  • A published commentary highlighted the neutral overall outcome and argued that better targeting (earlier enrolment, higher coagulopathy burden, or selected phenotypes) may be necessary to realise benefit from VHA-guided protocols.⁶
• Implementation gap (“last mile”) demonstrated in subsequent analysis:
  • A secondary analysis evaluating correction of trauma-induced coagulopathy found that, among 133 patients with serial sampling, 71% were coagulopathic on admission and 16% developed coagulopathy after admission.
  • VHA allocation increased the probability of receiving goal-directed haemostatic treatment (76% vs 47%; OR 3.73; 95% CI 1.62 to 8.59; P=0.002) and reduced time to first goal-directed treatment (68 vs 110 minutes; difference −37 minutes; 95% CI −62 to −14).⁷
  • Despite this, only 54% of patients received any goal-directed treatment and only 20% achieved coagulopathy correction at 24 hours, supporting the editorial’s emphasis that protocol execution may be the limiting step rather than assay availability alone.⁷
• Severe TBI signal: biologically plausible, statistically fragile:
  • The severe TBI subgroup showed a lower 28-day mortality with VHA guidance (44% vs 74%; OR 0.28; 95% CI 0.10 to 0.74; interaction P=0.016) and an association persisting after multivariable adjustment in supplementary analyses.
  • However, the subgroup was small, multiple subgroup comparisons were performed, and the trial was not powered for subgroup mortality effects; the finding should be treated as hypothesis-generating and prioritised for prospective confirmation.
• Guidelines after ITACTIC:
  • The 2023 European trauma bleeding guideline recommends early and repeated monitoring of coagulation with conventional laboratory tests and/or point-of-care/viscoelastic methods (Grade 1C) and recommends early, repeated goal-directed haemostatic therapy guided by coagulation monitoring (Grade 1B/1C depending on domain), noting ITACTIC’s neutral overall findings while still supporting VEM use within structured pathways where available.⁸
• Systematic review evidence base:
  • A 2022 systematic review (10 studies; 2 randomised, 8 observational) concluded that evidence supporting VHA-guided haemostatic resuscitation in trauma was not robust, with heterogeneity and limited study numbers constraining certainty; reported signals included reduced blood product use and mortality in individual included studies, but overall quality was downgraded.⁹

Summary

• In 396 adult trauma patients managed under contemporary massive haemorrhage protocols, VHA-guided algorithms did not improve the primary composite outcome (alive and free of massive transfusion at 24 h) compared with CCT-guided algorithms (67% vs 64%; OR 1.15; 95% CI 0.76 to 1.73; P=0.509).
• Overall mortality was similar at 6 hours, 24 hours, 28 days, and 90 days; time to haemostasis and organ-support-free days were also similar between groups.
• VHA guidance increased and accelerated protocolised haemostatic interventions (notably fibrinogen supplementation), demonstrating operational separation without overall outcome benefit.
• A prespecified severe TBI subgroup showed lower 28-day mortality with VHA guidance (44% vs 74%; OR 0.28; 95% CI 0.10 to 0.74; interaction P=0.016), but this finding remains hypothesis-generating.
• Subsequent analysis suggests the main limiting factor may be implementation of goal-directed therapy (only 54% received any goal-directed treatment; only 20% corrected coagulopathy at 24 h), reinforcing the importance of the “last mile” from test to treatment.

Overall Takeaway

ITACTIC provides high-quality multicentre evidence that, in modern trauma systems already delivering early tranexamic acid and balanced component resuscitation, substituting VHA-guided for CCT-guided goal-directed haemostatic algorithms increases early fibrinogen-directed interventions but does not improve overall early composite outcomes or survival. Its main legacy is in defining the limits of assay-driven “precision” transfusion in broadly enrolled trauma populations, while highlighting an important hypothesis-generating signal in severe traumatic brain injury and the critical implementation gap between diagnosis and timely targeted therapy.

Overall Summary

• VHA-guided goal-directed haemostatic algorithms improved process measures (earlier/more fibrinogen and study interventions) but did not improve overall outcomes versus CCT-guided algorithms; a severe TBI subgroup mortality signal warrants confirmatory trials.

Bibliography

• Baksaas-Aasen K, Gall L, Eaglestone S, et al. iTACTIC—implementing treatment algorithms for the correction of trauma-induced coagulopathy: study protocol for a multicentre, randomised controlled trial. Trials. 2017;18:486.
• Baksaas-Aasen K, van Dieren S, Balvers K, et al. Data-driven development of ROTEM and TEG algorithms for the management of trauma hemorrhage. Ann Surg. 2019;270:1178-1185.
• 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;313(5):471-482.
• Gonzalez E, Moore EE, Moore HB, et al. Goal-directed hemostatic resuscitation of trauma-induced coagulopathy: a pragmatic randomized clinical trial comparing a viscoelastic assay to conventional coagulation assays. Ann Surg. 2016;263:1051-1059.
• Casaer MP, De Jong A, Perner A, et al. Clinical validation of precision medicine protocols: the last mile is the longest. Intensive Care Med. 2021;47:80-82.
• Abraham P, Pasquier P, Rimmele T, David JS. Trauma patients do not benefit from a viscoelastic haemostatic assay-guided protocol, but why? Intensive Care Med. 2021;47:726-727.
• Shammassian BH, Saar M, Balvers K, et al. Correction of trauma-induced coagulopathy by goal-directed therapy: a secondary analysis of the ITACTIC trial. Anesthesiology. 2024;141(5):904-912.
• Rossaint R, Bouillon B, Cerny V, et al. The European guideline on management of major bleeding and coagulopathy following trauma: sixth edition. Crit Care. 2023;27:80.
• Zhu Z, Yu Y, Hong K, Luo M, Ke Y. Utility of viscoelastic hemostatic assay to guide hemostatic resuscitation in trauma patients: a systematic review. World J Emerg Surg. 2022;17:48.