Major Trauma Evidence Rundown — Issue 8 (November 2026)

MAJOR TRAUMA · SURGERY & CRITICAL CARE

Major Trauma Evidence Rundown

Issue 8 · November 2026 · UK Edition

Jake Turner — Senior Registrar in Emergency Medicine, ST6 — Curated with the assistance of AI (Claude). All content editorially reviewed.

Full archive and PDF downloads at emevidence.org

Lead — decompressive craniectomy: a systematic review and meta-analysis of 8 RCTs (n=1,326) found that decompressive craniectomy did not reduce unfavourable outcome in adults with moderate to severe TBI (61.5% vs 60.2%; RR 1.00, 95% CI 0.83 to 1.20). Any mortality gain from last-tier craniectomy may come as survival with severe disability, so the decision belongs in a frank conversation with the family. Change this month: nothing this month met the bar for a practice-change badge. Every main item is Informing Practice. UK: NHSBT O and B D-negative red cells still at Pre-Amber (O D-negative 2.9 days vs a 6-day aim, read 5 November) • SHOT 2025: 8 ABO-incompatible transfusions, 5 of them red cells • MHRA field safety notices, including a Traumafix dressing • no new TARN/NMTR, NHS England EPRR or JESIP output. Core Revision: spinal cord injury — ISNCSCI, spinal vs neurogenic shock, steroids, timing of decompression and MAP targets (FRCS / Final FRCA).

BOTTOM LINE UP FRONT — NOVEMBER 2026

WORTH KNOWING ON SHIFT

OBS TXA: in two pooled prehospital trials (n=1,249), 30-day death was highest with TEG shutdown (14.5%), where TXA was linked to a lower hazard of death (HR 0.32, 95% CI 0.11 to 0.91) in a subgroup likely defined after the prehospital dose; do not withhold TXA on an arrival TEG or ROTEM.

OBS Tourniquet conversion: 83.9% of 647 civilian prehospital tourniquets were converted in the ED without a procedure or operation, and each extra 30 minutes was associated with lower odds of success (OR 0.41, 95% CI 0.32 to 0.53); reassess every tourniquet on arrival.

OBS Burn size at referral: overestimated in 76% of 82 patients sent to a Swedish burn centre (median 10% vs 6% TBSA); recalculate with Lund and Browder before fixing fluids or the level of burn service.

OBS Subdural haematoma with GCS 13–15: 10% of those aged 65 or over died or were discharged to hospice vs 3% under 65 (n=1,157).

DX Shock index: a weak predictor of trauma death on its own (LR+ 2.86, LR− 0.42); a positive result lifts a 20% prior only to 42%, so use it as a flag, not a decision.

UK Blood supply: O and B D-negative red cells remain at Pre-Amber; O D-negative stock was 2.9 days on 5 November against a 6-day aim.

KNOW FOR NEXT TIME

SR/MA Decompressive craniectomy (Lead): no reduction in unfavourable outcome across 8 RCTs (RR 1.00, 95% CI 0.83 to 1.20).

RCT Ankle fracture weight bearing: 2 weeks after fixation was non-inferior to 6 weeks for reoperation (2.3% vs 3.4%; n=418) with no significant difference in return to work (HR 1.1, 95% CI 0.8 to 1.6).

OBS TFNA nails: no significant difference in any reoperation vs PFNA (17.2% vs 12.9%, p=0.194), but more major revisions (7.8% vs 3.2%) in one cohort and more mechanical failures (1.6% vs 0.3%) in another.

OBS Platelets and respiratory failure: each platelet unit in 24 hours was linked to 7% higher odds of respiratory failure by day 7 (OR 1.071, 95% CI 1.021 to 1.122) in a post hoc analysis likely confounded by bleeding severity; keep NG39 and local major haemorrhage protocols.

OBS Mass casualty incidents: severely injured patients took longer to reach hospital (median 85 vs 69 minutes) without a clear difference in in-hospital death (adjusted OR 1.19, 95% CI 0.77 to 1.85).

PAEDS Telemedicine for injured children: adding video to the phone referral did not change parent experience, distress or transfer in a 10-ED stepped-wedge RCT, but only 44.7% actually had a video call.

REVIEW Pregnancy (Quick Hits): UK practice considers resuscitative hysterotomy from 20 weeks (fundus at the umbilicus) and gives at least 500 IU anti-D within 72 hours after abdominal trauma from 12 weeks, plus a Kleihauer test from 20 weeks.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 1 of 24

November’s lead is a meta-analysis of the decompressive craniectomy trials in adult TBI, which found no reduction in unfavourable outcome and a trade between death and survival with severe disability. Haemorrhage gets four items: TXA by fibrinolytic phenotype, the transfusion burden at which mortality climbs, ED tourniquet conversion, and platelets and respiratory failure after severe trauma. Orthopaedics has a weight-bearing trial after ankle fracture fixation and a two-cohort comparison of TFNA and PFNA trochanteric nails. Burns are covered by a Swedish study of TBSA estimation at referral. There are three mass casualty items, two observational and one NATO doctrine update, plus a telemedicine trial for injured children. Two US reviews, on burns with combined injuries and on trauma in pregnancy, are in Quick Hits with the UK thresholds that differ. The UK update covers blood stocks, the SHOT 2025 ABO-incompatible figures and MHRA notices. The Core Revision covers spinal cord injury.

WHAT’S INSIDE — ISSUE 8

• Lead: decompressive craniectomy in adult TBI (SR/MA of 8 RCTs); shock index, RTS and ISS for predicting trauma death
• Guidelines and UK: UK trauma system update (NHSBT, SHOT, NMTR, EPRR, JESIP); Western Trauma Association haemothorax algorithm
• Haemorrhage and resuscitation: TXA by fibrinolytic phenotype; transfusion burden and death; ED tourniquet conversion; platelets and respiratory failure
• Orthopaedic: weight bearing at 2 vs 6 weeks after ankle fixation; TFNA vs PFNA trochanteric nails
• Neurotrauma and burns: subdural haematoma with GCS 13–15; burn size at referral
• Mass casualty: routine civilian MCIs; transport order at multi-patient crashes; NATO doctrine
• Paediatric trauma: telemedicine for injured children
• Quick Hits (12): including MHRA notices, GFAP and UCH-L1 cut-offs, paediatric craniectomy and pneumothorax observation, burns with combined injuries and trauma in pregnancy

• Core Revision: spinal cord injury

CHANGE TONIGHT CHANGE THIS MONTH SAFETY ALERT GUIDELINE INFORMING PRACTICE UK FRCS

FINAL FRCA

Contents: 1. Key Trials & Research — 2. Guidelines & UK Updates — 3. Haemorrhage & Resuscitation — 4. Orthopaedic & Musculoskeletal Trauma — 5. Neurotrauma & Burns — 6. Mass Casualty & Major Incident — 7. Paediatric Trauma — 8. Quick Hits — Core Revision — 9. Action Points — Trials to Watch

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 2 of 24

1 — KEY TRIALS & RESEARCH

LEAD

INFORMING PRACTICE

SR/MA

UK

FRCS

FINAL FRCA

Decompressive Craniectomy in Adults With Moderate to Severe TBI (SR/MA, 8 RCTs, n=1,326): No Reduction in Unfavourable Outcome (61.5% vs 60.2%; RR 1.00, 95% CI 0.83 to 1.20) and No Clear Mortality Benefit (RR 0.84, 0.59 to 1.19)

Bouras M, Lamothe-Boucher F, Abou Jamous J, et al. Decompressive craniectomy in adults with moderate to severe traumatic brain injury: a systematic review and meta-analysis of randomised controlled trials. Crit Care Resusc. 2026;28(4):100220. doi:10.1016/j.ccrj.2026.100220. PMID: 42893097. PROSPERO CRD42024565322. Source route: PMC full text (PMC13676264; open access, CC BY-NC-ND); supplementary figures not read. Appraised from full text

Systematic review (search to 1 August 2026) of 8 RCTs (1,326 patients; at least 80% of each trial aged 18 or over) comparing decompressive craniectomy (DC) with medical management (4 trials, including DECRA and RESCUEicp) or with a surgical comparator (craniotomy in 3, including RESCUE-ASDH and PREDICT-AEDH; cisternostomy in 1); 5 trials tested primary DC, performed without regard to intracranial pressure, and 3 secondary DC for refractory intracranial hypertension. The primary outcome, unfavourable outcome (GOS 1–3 or GOS-E 1–4) at the latest follow-up (3 to 12 months), occurred in 397 of 646 (61.5%) with DC vs 391 of 650 (60.2%) controls (7 trials; RR 1.00, 95% CI 0.83 to 1.20; I² 68%; low certainty). Mortality was 26.5% vs 33.7% (8 trials; RR 0.84, 0.59 to 1.19; I² 65%; low certainty); it was lower with DC against medical management (RR 0.59, 0.43 to 0.81; 4 trials) and with secondary DC (RR 0.61, 0.42 to 0.87; 3 trials), but no subgroup showed a functional benefit. Quality of life was reported by 2 trials and did not differ.

602 vs 602

UNFAVOURABLE OUTCOME PER 1,000 (RR 1.00, 95% CI 0.83 TO 1.20)

ARD 0

PER 1,000, CALCULATED FROM THE GRADE TABLE (95% CI −103 TO +120); NO NNT

283 vs 337

DEATHS PER 1,000, GRADE ESTIMATE FROM THE POOLED RR APPLIED TO THE CONTROL RISK (RR 0.84, 0.59 TO 1.19; RAW POOLED 26.5% VS 33.7%); ARD −54, CALCULATED (−138 TO +64)

RR 0.59

MORTALITY VS MEDICAL MANAGEMENT, 4 TRIALS (0.43 TO 0.81); SUBGROUP ONLY

CRITICAL APPRAISAL Clear question; protocol registered

Yes PROSPERO CRD42024565322; adult RCTs only; prespecified subgroups by comparator, DC type, haemorrhage and risk of bias

Design: SR/MA | Certainty: low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing new for theatre or neuro-ICU: keep secondary DC as a selective last-tier option for refractory ICP, and tell families before surgery that lower mortality may come with survival in a vegetative state or with dependence | Confidence in that: moderate Main concerns: the GOS-E 1–4 cut-off counts upper severe disability (independent at home) as unfavourable, so it likely understates any benefit of last-tier DC; using upper severe disability or better as favourable, RESCUEicp reported 45.4% vs 32.4% at 12 months; pooling early DC, last-tier DC and DC at haematoma evacuation (I² 68%) can hide benefit in one setting and harm in another; direction unknown; all 3 secondary-DC trials also sit in the 4-trial medical-comparator subgroup, so the two mortality signals are largely one finding, which likely overstates their strength Strengths: registered protocol, RCTs only, RoB 2 and GRADE, and the first synthesis to include RESCUE-ASDH and PREDICT-AEDH Disagreement: The authors argue that the Brain Trauma Foundation’s 2020 recommendation of secondary DC for late refractory intracranial hypertension is not supported by the pooled data. RESCUEicp, on which that recommendation largely rests, reported lower mortality, and more favourable outcomes at 12 months when upper severe disability was counted as favourable.

UK practice: RESCUEicp (2016; MRC-funded) and RESCUE-ASDH (2023; NIHR-funded) were led from Cambridge and recruited mostly UK patients. RESCUE-ASDH found similar 12-month GOS-E with craniectomy and craniotomy for acute subdural haematoma (common OR 0.85, 0.60 to 1.18), with fewer additional cranial operations within 2 weeks (6.9% vs 14.6%) but more wound complications (12.2% vs 3.9%). NICE NG232 (head injury: assessment and early management, May 2023) makes no recommendation on DC; it asks for a neurosurgical discussion about any new, surgically significant abnormality on imaging (1.4.15) and about persisting coma (GCS 8 or less) after resuscitation (1.4.16). The International Consensus Meeting statement on decompressive craniectomy in TBI (Cambridge meeting, published 2019) advises using secondary DC selectively and having frank discussions with the family about risks and prognosis beforehand; we found no SBNS guidance on DC (searched 3 Nov 2026).

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 3 of 24

INFORMING PRACTICE

DIAGNOSTIC

FRCS

FINAL FRCA

Shock Index, RTS and ISS for Predicting Trauma Death (Diagnostic Meta-Analysis, 12 Studies, n=234,129): Shock Index LR+ 2.86 and LR− 0.42, So a Positive Result Raises a 20% Prior Only to 42%

Raza AA, Ahmad M, Choudhary N, et al. Diagnostic accuracy of shock index, revised trauma score, and injury severity score for mortality prediction in trauma: a hierarchical meta-analysis. Injury. 2026;57(11):113760. doi:10.1016/j.injury.2026.113760. PMID: 42854533. No protocol registration reported. Source route: PDF via BMA Library; supplementary files not read. Appraised from full text

Hierarchical (bivariate random-effects) meta-analysis of 12 mainly retrospective studies published 2014–2025 (search to May 2026; 234,129 adults from emergency department, trauma centre, registry and ICU cohorts in Türkiye, Slovenia, India, the USA, Iran, Qatar, South Africa and Pakistan), testing shock index (SI, 5 studies), Revised Trauma Score (RTS, 6) and Injury Severity Score (ISS, 9) against death; in-hospital mortality was the prespecified endpoint, and some studies used 7-, 28- or 30-day mortality. Pooled SI sensitivity was 0.679 (95% CI 0.544 to 0.789) and specificity 0.763 (0.446 to 0.928; AUC 0.747) at cut-offs from 0.86 to 1.08; RTS gave 0.788 (0.454 to 0.943) and 0.801 (0.683 to 0.882; AUC 0.852), and ISS 0.788 (0.712 to 0.848) and 0.796 (0.709 to 0.862; AUC 0.855). In the 6 studies reporting both, RTS and ISS did not differ (likelihood-ratio test p=0.914; within-study sensitivity OR 1.59, 0.82 to 3.07; specificity OR 0.89, 0.55 to 1.45), and SI could not be compared directly with RTS. One US National Trauma Data Bank study of older patients supplied 217,190 of the 234,129 patients (92.8%, calculated), all in the SI analysis.

• 0.679 SI SENSITIVITY (95% CI 0.544 TO 0.789)
• 0.763 SI SPECIFICITY (95% CI 0.446 TO 0.928)
• 2.86 SI LR+ (FROM POOLED SN/SP; CI NOT REPORTED)
• 0.42 SI LR− (CI NOT REPORTED)
• n=234,129 12 STUDIES: SI 5, RTS 6, ISS 9

Post-test probability of death, calculated by us from the pooled sensitivity and specificity (the paper’s Table 4 gives the same values except for three negative results, RTS 1.3% at 5% and 6.1% at 20% and ISS 6.3% at 20%, because it used rounded LR−). Priors: 5%, lower-acuity ED population (paper); 10%, total mortality (death or hospice discharge) in adults aged 65 or over with acute subdural haematoma and GCS 13–15 (Orlando, this issue), an indirect anchor because these LRs are for death in general adult trauma; 20%, major trauma centre population (paper). LR CIs not reported; the wide SI specificity CI makes the SI LR+ imprecise. ISS is not available at triage.

CRITICAL APPRAISAL Representative spectrum

No Mostly retrospective cohorts, several restricted to older, pelvic, abdominal, thoracoabdominal or polytrauma patients; one geriatric registry supplied 92.8% of patients; I² up to 99.9%

Design: Diagnostic | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing new: no single score should decide trauma-team activation or MTC triage; keep a raised shock index as one physiological flag within the triage tool and serial assessment | Confidence in that: moderate Main concerns: cut-offs chosen from each study’s own ROC curve likely overstate accuracy (optimism bias); one US geriatric registry dominates the patient count and leave-one-out SI specificity ranged from 0.64 to 0.85, so the SI estimate is unstable; direction unknown; death as the target condition is indirect for triage, which aims to find injuries needing intervention; direction unknown Strengths: a bivariate hierarchical model with a paired within-study RTS–ISS comparison, rather than ranking separately pooled AUCs

UK practice: NICE NG40 (major trauma: service delivery, 2016) asks ambulance services to use a prehospital major trauma triage tool that includes physiology and anatomical injury and accounts for older patients, children and people taking anticoagulants (1.1.1, 1.1.2), and asks a senior nurse or trauma team leader to set the level of trauma team response from the pre-alert, using written local guidelines (1.3.3). NICE NG39 (major trauma: assessment and initial management, 2016) says to activate the major haemorrhage protocol on haemodynamic status and response to volume resuscitation (1.5.13), and not to rely on a haemorrhagic risk tool applied at a single time point (1.5.14). These likelihood ratios support that approach: a raised shock index should prompt a senior look, not decide the pathway on its own.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 4 of 24

2 — GUIDELINES & UK UPDATES

INFORMING PRACTICE

POLICY/NEWS

UK

FINAL FRCA

UK Trauma System Update, 3 October to 5 November 2026: O D-Negative Still at Pre-Amber (2.9 Days vs 6-Day Aim); 8 ABO-Incompatible Transfusions in 2025, 5 of Them Red Cells (SHOT); No New TARN, EPRR or JESIP Output

NHS Blood and Transplant. Blood stocks (undated; read 3 and 5 November 2026) and The Update. hospital.blood.co.uk. Serious Hazards of Transfusion. SHOT Newsletter October 2026 (published 6 October 2026). shotuk.org (PDF). NHS England NMTR and EPRR pages; JESIP news. Source route: each body’s website checked 3 November 2026, plus NICE, BOA, BAPRAS, RCSEng and the British Burn Association. Appraised from full text

On 5 November 2026 NHSBT showed Pre-Amber status for O and B D-negative red cells, with 2.9 days of O D-negative red cell stock against its aim of 6 days. The page is undated; our checks found 2.5 days on 22 October and 3.5 days on 3 November, and Issue 7 reported 3.2 days. SHOT’s October newsletter spotlights the 2025 Annual SHOT Report chapter on incorrect blood components transfused (IBCT): there were 8 ABO-incompatible transfusions in 2025 (5 red cell, 3 plasma). Red cell cases rose from 1 in 2024 to 5, all clinical errors (4 at collection, 1 at administration); plasma cases stayed at 3, all laboratory errors. IBCT errors caused 8 cases of major morbidity (2 in 2024). We found no new TARN/NMTR, NHS England EPRR, JESIP or NARU publication in the window.

Each body’s own website, read 3 November 2026 (NHSBT also 5 November). NICE, BOA (latest BOAST June 2026), BAPRAS, RCSEng and the British Burn Association had no trauma guidance in the window.

CRITICAL APPRAISAL Who wrote it and for whom

Yes NHSBT, SHOT, NHS England and JESIP, for hospital and responder teams

Design: Policy/News | Certainty: moderate (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing new to change; keep Pre-Amber stewardship of O D-negative red cells in the major haemorrhage protocol, and review bedside pre-administration checks against SHOT’s checklist | Confidence in that: moderate Main concerns: a single undated stock reading may misstate the trend (direction unknown); voluntary SHOT reporting likely understates ABO-incompatible transfusions; unreadable sites (NARU, MHRA notices) mean “nothing new” may understate activity Strengths: every row rests on the issuing body’s own page, checked 3 November 2026, with blood stocks re-read 5 November

UK practice: NHSBT Blood stocks (read 3 and 5 November 2026) defines Pre-Amber as “reduced availability of blood for a short or a prolonged period”. Annual SHOT Report 2025, chapter 10 (IBCT) names collection and administration as the weak points and says the new SHOT pre-administration checklist may help. NHS England 2026/27 EPRR annual assurance process (June 2026) sets 23 December 2026 for national ratings, and the concept of operations for managing mass casualties (July 2026) is the current national plan.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 5 of 24

INFORMING PRACTICE

GUIDELINE

FRCS

Western Trauma Association Traumatic Haemothorax Algorithm (12 Authors, Level V Consensus): Four Decision Inputs, but No Thresholds in the Abstract

Hynes AM, Brown CVR, Hartwell JL, et al. Western Trauma Association critical decisions in trauma: traumatic hemothorax. J Trauma Acute Care Surg. 2026 (online ahead of print). doi:10.1097/TA.0000000000005207. PMID: 42862903. Source route: PubMed abstract only (the BMA Library cannot supply this journal); no published commentary found (JournalFeed, SGEM, St Emlyn’s, REBEL EM, EMCrit, EAST/AAST journal club and journal commentary searched 3 Nov 2026). Abstract only

A Western Trauma Association consensus algorithm, by 12 authors from US centres, for traumatic haemothorax, whose management ranges from observation to tube thoracostomy and operation. The recommendations came from a review of the literature, including randomised and observational studies, supplemented by expert consensus where high-quality evidence is limited (Level V). The pathway uses four inputs: the patient’s physiology, the size of the haemothorax, whether pleural drainage is adequate, and evidence of ongoing haemorrhage. The abstract gives no size or drain-output thresholds, no timing for evacuation of retained haemothorax and no grading of individual recommendations, and the authors present it as support for, not a replacement of, clinical judgement and a template for local protocols.

CRITICAL APPRAISAL Developer and funding; conflicts managed

Partly WTA, 12 authors from US centres; funding and conflicts not in the abstract

Design: Guideline | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing yet; a possible starting point for an MTC haemothorax protocol once the full thresholds can be read | Confidence in that: moderate Main concerns: abstract only, so the decision points and thresholds cannot be checked; direction unknown; one association’s consensus may reflect US access to theatre and thoracoscopic surgery, which may not match every UK trauma unit; direction unknown Strengths: states the types of evidence used and the four clinical inputs plainly

UK practice: NICE NG39 (major trauma: assessment and initial management, 2016) has no recommendation on haemothorax; for the chest it covers open thoracostomy followed by a chest drain for tension pneumothorax (1.4.2) and urgent imaging, considering immediate CT for adults with suspected chest trauma without severe respiratory compromise (1.4.5). The BTS Guideline for pleural disease (Thorax 2023) covers spontaneous pneumothorax, pleural effusion, infection and malignancy, not trauma. We found no SCTS guidance on traumatic haemothorax, so UK MTCs have no national haemothorax algorithm.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 6 of 24

3 — HAEMORRHAGE & RESUSCITATION

INFORMING PRACTICE

OBSERVATIONAL

FRCS

FINAL FRCA

TXA by Fibrinolytic Phenotype (Two Pooled Prehospital TXA Trials, n=1,249): 30-Day Death Highest With TEG “Shutdown” (14.5%), Where TXA Was Linked to a Lower Hazard of Death (HR 0.32) in a Subgroup Likely Defined After the Prehospital Dose (TEG Timing Not Reported in the Abstract)

Kenny JE, Brito AM, Benchiba-Charron M, et al. Survival benefit of prehospital TXA in fibrinolytic shutdown: insights from two prospective trauma trials. J Trauma Acute Care Surg. 2026 (online ahead of print). doi:10.1097/TA.0000000000005092. PMID: 42853261. Source route: PubMed abstract only (journal not supplied by the BMA Library); no published commentary found (PubMed comment links, Europe PMC, JournalFeed, EMA Daily, SGEM, St Emlyn’s, REBEL EM, EMCrit, EAST literature reviews and journal commentary searched 3 Nov 2026). Abstract only

Secondary analysis of a harmonised dataset from two prospective prehospital TXA trials (the Prehospital Tranexamic Acid Use for Traumatic Brain Injury trial and the STAAMP air and ground transport trial): 1,249 injured patients grouped by their initial thromboelastography (TEG) LY30, the clot-lysis measure on the TEG trace, into fibrinolytic shutdown (below 0.9%), physiological (0.9–3%) and hyperfibrinolysis (above 3%). The primary outcome, 30-day mortality, was 14.5% with shutdown, 6.7% with hyperfibrinolysis and 6.5% with physiological fibrinolysis. Adjusted for TXA, phenotype, age, mechanism, Injury Severity Score and shock index on hospital arrival, hyperfibrinolysis carried a higher hazard of death (HR 2.87, 95% CI 1.10 to 7.53; reference group not stated), and patients in shutdown randomised to TXA had a lower hazard of death (HR 0.32, 95% CI 0.11 to 0.91). Phenotype group sizes, TXA-versus-placebo event rates within each phenotype, when the TEG was drawn relative to the TXA dose, and adverse events are not reported in the abstract.

• 14.5% 30-DAY MORTALITY, FIBRINOLYTIC SHUTDOWN (N NOT REPORTED)
• 6.5% / 6.7% 30-DAY MORTALITY, PHYSIOLOGICAL / HYPERFIBRINOLYSIS
• HR 2.87 DEATH WITH HYPERFIBRINOLYSIS, ADJUSTED (95% CI 1.10 TO 7.53)
• HR 0.32 DEATH WITH TXA IN SHUTDOWN, ADJUSTED (95% CI 0.11 TO 0.91)

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Phenotype from an “initial” TEG LY30 with fixed cut-offs; timing against the prehospital TXA dose not reported; 30-day death is objective

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing yet; do not delay or withhold TXA in the resus room because an arrival TEG or ROTEM shows low lysis, and keep the NG39 3-hour window | Confidence in that: moderate Main concerns: stratifying on a TEG result likely drawn after the prehospital dose (timing not reported in the abstract) would break the randomised comparison, with unpredictable direction of bias; an unreported subgroup size with a wide CI, one of three phenotypes examined, likely overstates the TXA effect in shutdown; pooling a brain-injury trial with a mixed-injury transport trial may mix different TXA effects; direction unknown Strengths: both source trials randomised prehospital TXA prospectively, so dosing was protocolised and outcomes were collected prospectively

UK practice: NICE NG39 (major trauma: assessment and initial management, 2016) says to give intravenous TXA as soon as possible in major trauma with active or suspected active bleeding (1.5.4), not more than 3 hours after injury unless there is evidence of hyperfibrinolysis (1.5.5), and to move from a fixed-ratio protocol to one guided by laboratory coagulation results at the earliest opportunity (1.5.27); it does not mention fibrinolytic shutdown. NICE HTG348, formerly DG13 (viscoelastometric point-of-care testing, 2014) finds insufficient evidence to recommend routine viscoelastic testing for the emergency control of bleeding after trauma (1.4). The British Society for Haematology guideline on viscoelastic haemostatic assays in major bleeding (2018) says TEG and ROTEM are insensitive to mild or moderate fibrinolysis and that TXA should not be withheld on TEG or ROTEM parameters (Grade 1B), which this abstract is consistent with.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 7 of 24

INFORMING PRACTICE

OBSERVATIONAL

FRCS

FINAL FRCA

Early Transfusion Burden and In-Hospital Death in US Trauma (TQIP, n=126,271): Mortality Rose Most Steeply Near 21 Red Cell-Equivalent Units in 4 Hours, Reached 50% Only Near 42 Units and Never Approached 100%

Mehta R, Zangbar B, Prabhakaran K, et al. Mortality risk across early RBC-equivalent transfusion burden in the whole blood era. J Am Coll Surg. 2026 (online ahead of print). doi:10.1097/XCS.0000000000002247. PMID: 42847426. Source route: PubMed abstract only (journal not supplied by the BMA Library); no published commentary found (PubMed comment links, Europe PMC, JournalFeed, EMA Daily, SGEM, St Emlyn’s, REBEL EM, EMCrit and EAST/AAST journal club searched 3 Nov 2026). Abstract only

Retrospective cohort from the American College of Surgeons Trauma Quality Improvement Program (TQIP, 2020–2023): 126,271 adult trauma patients who received at least 1 unit of packed red cells within 4 hours of admission (68.0% blunt, 32.0% penetrating). The exposure was “RBC-equivalent” burden, packed red cells plus whole blood units in the first 4 hours, and the primary outcome was in-hospital mortality, modelled with multivariable spline-based logistic regression. Mortality rose non-linearly, most steeply at 21.1 units (the 95th centile of burden); predicted mortality reached 50% at 41.9 units (the 99th centile) and plateaued from 55.3 units, without approaching 100%. Overall mortality, the share of patients or units that were whole blood, the model covariates and confidence intervals for these points are not reported in the abstract.

• n=126,271 TQIP 2020 TO 2023, AT LEAST 1 UNIT OF RED CELLS IN 4 H
• 21.1 units STEEPEST RISE IN MORTALITY, 95TH CENTILE (CI NOT REPORTED)
• 41.9 units PREDICTED MORTALITY 50%, 99TH CENTILE (CI NOT REPORTED)
• 55.3 units MORTALITY PLATEAU ONSET (CI NOT REPORTED)

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Units counted in a registry over 4 hours; plasma and platelets are not part of the burden measure; in-hospital death is objective

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing yet; no unit count should be used as a futility threshold, and MTC major haemorrhage protocols stay triggered by physiology and response | Confidence in that: moderate Main concerns: survivor selection at high unit counts likely overstates the chance of surviving extreme transfusion; a US registry with an unreported whole blood share may not transfer to UK component-based MTC practice; direction unknown; reference points set at the cohort’s own 95th and 99th centiles, without CIs or validation, may not hold in other cohorts; direction unknown Strengths: a very large contemporary national registry with continuous modelling instead of an arbitrary massive transfusion cut-off

UK practice: NICE NG39 (major trauma: assessment and initial management, 2016) says to activate the major haemorrhage protocol on physiological criteria, including haemodynamic status and response to immediate volume resuscitation (1.5.13), not on a risk tool applied at a single time point (1.5.14), to use 1 unit of plasma to 1 unit of red cells in adults (1.5.24), and to move to laboratory-guided replacement at the earliest opportunity (1.5.27); it sets no transfusion ceiling. In the NHSBT-sponsored SWiFT (Smith JE et al, N Engl J Med 2026, doi:10.1056/NEJMoa2516043, PMID: 41841706), prehospital whole blood across 10 English air ambulance services was not superior to red cells plus plasma for death or massive transfusion at 24 hours (48.7% vs 47.7%; RR 1.02, 95% CI 0.80 to 1.31). We found no published NHSBT decision on routine civilian whole blood supply since SWiFT (searched 3 Nov 2026).

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 8 of 24

INFORMING PRACTICE

OBSERVATIONAL

FINAL FRCA

FRCS

Blood Components and Early Respiratory Failure After Severe Trauma (RESTRIC Post Hoc, n=411): Each Platelet Unit in 24 Hours Linked to 7% Higher Odds (OR 1.071); Red Cells and Plasma Not Linked After Adjustment

Tanaka C, Tagami T, Fukuda R, Sakamoto T, Nakayama F, Kuno M. Effect of blood transfusion on respiratory failure in severe trauma patients: a post hoc analysis of a randomized multicenter trial in Japan. Injury. 2026 (corrected proof, online 19 October 2026):113781. doi:10.1016/j.injury.2026.113781. PMID: 42926186. RESTRIC: UMIN000034405. Source route: full text (PDF; licence listed as CC BY-NC-ND in Europe PMC); supplementary appendix not read. Appraised from full text

Post hoc analysis of RESTRIC, a cluster-randomised crossover non-inferiority trial of restrictive versus liberal red cell transfusion in Japan (May 2019 to October 2021): 411 adults arriving with severe or suspected severe bleeding and shock (422 registered; 11 excluded, 6 for death on the admission day). The exposures were units of red cells, fresh frozen plasma (FFP) and platelets in the first 24 hours, and the primary outcome was respiratory failure within 7 days of ICU admission (PEEP at least 5 cmH2O and PaO2:FiO2 below 200 mmHg [26.7 kPa, calculated]), which occurred in 40 patients (9.7%, calculated; the paper states 9.8%). Adjusted for sex, ISS, chest AIS and the three component volumes, each platelet unit carried an OR of 1.071 (95% CI 1.021 to 1.122), while red cells (OR 1.030, 0.968 to 1.096) and FFP categories were not associated. Patients with respiratory failure had lower 28-day survival (82.5% vs 94.2%), and the randomised red cell strategy did not differ between groups (restrictive in 55.0% vs 56.9%).

• 40 of 411 RESPIRATORY FAILURE WITHIN 7 DAYS (9.7%, CALCULATED)
• 60.0% vs 21.0% GIVEN PLATELETS IN 24 H, RESPIRATORY FAILURE VS NONE (24/40 VS 78/371)
• OR 1.071 PER PLATELET UNIT, ADJUSTED (95% CI 1.021 TO 1.122)
• OR 1.030 PER RED CELL UNIT, ADJUSTED (95% CI 0.968 TO 1.096)

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Volumes in product units as reported, with no UK dose equivalent given; respiratory failure defined by PEEP and PaO2:FiO2 alone, without imaging or TRALI/TACO adjudication (no TRALI diagnosed)

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing; keep platelets in fixed-ratio major haemorrhage packs and move to laboratory- or viscoelastic-guided replacement as early as possible | Confidence in that: moderate Main concerns: platelets are given mostly to the patients bleeding most, so confounding by indication likely overstates any platelet harm; a data-driven model with 40 events likely overstates the platelet association through overfitting; respiratory failure from chest injury or brain injury is not separated from transfusion lung injury; direction unknown Strengths: prospectively collected multicentre trial data with protocolised red cell thresholds and a stated respiratory failure definition Disagreement: The authors cite the PROPPR randomised trial, in which higher plasma and platelet ratios did not change safety outcomes including respiratory failure. This post hoc cohort links platelet volume with respiratory failure but cannot separate platelets from the bleeding severity that prompted them.

UK practice: NICE NG39 (major trauma: assessment and initial management, 2016) recommends a 1:1 plasma to red cell ratio in adults (1.5.24) and a fixed-ratio protocol that changes to one guided by laboratory coagulation results at the earliest opportunity (1.5.27); it does not specify platelet dosing. NICE NG24 (blood transfusion, 2015, last updated February 2026) sets general platelet thresholds for thrombocytopenia (1.7.1, 1.7.2) but explicitly excludes trauma and massive haemorrhage, so MTC platelet use rests on NG39 and local major haemorrhage protocols; this study gives no reason to change them.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 9 of 24

INFORMING PRACTICE

OBSERVATIONAL

FRCS

Tourniquet Conversion: 83.9% of 647 Civilian Prehospital Tourniquets Converted in the ED Without a Procedure or Operation, With Each Extra 30 Minutes Associated With Lower Odds (OR 0.41); 13 of 14 Converted in a Wartime Mass Casualty Series

Fox AL, Stuart M, Clemens M, Callaway DW. CONVERT: civilian outcomes of emergency department tourniquet conversion. J Trauma Acute Care Surg. 2026 (online 29 June 2026). doi:10.1097/TA.0000000000005106. PMID: 42371668. Paired with: Hofmann LJ, Hiles JM, Hetzler M, et al. “High and tight to pack and press”: prioritization of tourniquet conversion in unconventional warfare improves outcomes. J Trauma Acute Care Surg. 2026;101(2):291–296. doi:10.1097/TA. 0000000000004949. PMID: 41944821. Source route: PubMed abstracts only (journal not supplied by the BMA Library); secondary sources for CONVERT: JournalFeed (Patel K, 28 Aug 2026) and EMA Daily editor’s commentary (15 Oct 2026); none found for Hofmann (PubMed, Europe PMC, web and JournalFeed/EMA email searched 3 Nov 2026). Abstract only

CONVERT is a retrospective cohort of 647 patients arriving at one US level I trauma centre (March 2021 to December 2024) with prehospital extremity tourniquets; the outcome was successful conversion (removal with packing, haemostatic dressings or pressure bandages, without a procedure or operation), procedural intervention or operative intervention. Conversion succeeded in 543 (83.9%) at a median tourniquet time of 40 minutes (IQR 30–59); 60 patients (9.3%) had injury-related complications, none attributed to the tourniquet, more often after operative intervention than after successful conversion (44.4% vs 4.6%). On multivariable analysis each extra 30 minutes was associated with lower odds of successful conversion (OR 0.41, 95% CI 0.32 to 0.53); upper-limb tourniquets were also less often converted (OR 0.61, 95% CI 0.38 to 0.97; adjustment not stated in the abstract). In Hofmann’s descriptive series, a mobile surgical team at a mass casualty incident in unconventional warfare treated 46 casualties with 14 tourniquets on 12 patients: 13 (93%) were converted, 4 of them after surgical haemostasis, with no deaths from limb haemorrhage and no reported tourniquet complications.

• 83.9% CONVERTED IN ED WITHOUT PROCEDURE OR OPERATION (543/647)
• 40 min MEDIAN TOURNIQUET TIME (IQR 30 TO 59)
• OR 0.41 SUCCESSFUL CONVERSION PER EXTRA 30 MIN (95% CI 0.32 TO 0.53)
• 13 of 14 TOURNIQUETS CONVERTED IN THE WARTIME SERIES (93%)

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly How tourniquet time was recorded, and who judged that no complication was due to the tourniquet, are not reported

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing new to adopt; it supports reassessing every prehospital tourniquet on arrival in resus and attempting conversion early unless there is amputation or uncontrolled bleeding | Confidence in that: moderate Main concerns: including tourniquets that were never needed likely overstates the conversion success rate; confounding by injury severity likely overstates the effect of tourniquet duration itself; unreported follow-up may miss late ischaemic or reperfusion harm, understating tourniquet complications Strengths: a large civilian ED conversion cohort (647 patients) with predefined outcome categories, plus a mass casualty series that tests conversion where evacuation is delayed

UK practice: NICE NG39 (major trauma: assessment and initial management, 2016) says to use a tourniquet in major limb trauma if direct pressure has failed to control life-threatening haemorrhage (1.5.2), but says nothing on reassessment or conversion (we checked the recommendations page). The Defence Medical Services Clinical Guidelines for Operations: Tourniquet Use (last reviewed January 2026) says to review every tourniquet as soon as it is safe, convert ideally within 2 hours, convert up to 6 hours with reperfusion monitoring beyond 2 hours, reassess at every handover including arrival at a treatment facility, and record the application time; for high and tight or hasty tourniquets it advises: “Convert if possible; replace if not.” We found no civilian UK guideline on ED tourniquet conversion (NICE and FPHC searched 3 Nov 2026).

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 10 of 24

4 — ORTHOPAEDIC & MUSCULOSKELETAL TRAUMA

INFORMING PRACTICE

RCT

FRCS

Weight Bearing at 2 vs 6 Weeks After Ankle Fracture Fixation (US Multicentre RCT, n=418): Non-Inferior for Reoperation (2.3% vs 3.4%) but No Significant Difference in Return to Work (HR 1.1)

Obremskey WT, Archer KR, Gaski GE, et al. A prospective randomized trial to assess early versus delayed weight bearing following operatively treated ankle fractures in the United States. J Orthop Trauma. 2026 (online ahead of print). doi:10.1097/BOT.0000000000003299. PMID: 42873990. Trial registry id not given in the abstract. Source route: PubMed abstract only (open access on ovid.com, but the full text could not be retrieved; no PMC or Europe PMC copy), plus the Orthopaedic Trauma Association 2025 Annual Meeting abstract (Podium 25, ota.org); no journal editorial or blog commentary found (JournalFeed, SGEM, St Emlyn’s, REBEL EM, EMCrit and journal commentary searched 3 Nov 2026). Abstract only

Randomised non-inferiority trial in 19 US trauma centres (January 2020 to October 2023): adults with an OTA/AO 44 A, B or C ankle fracture fixed surgically were assigned to weight bearing from 2 weeks or from 6 weeks after surgery, excluding Gustilo-Anderson IIIB and IIIC open fractures, syndesmotic injuries with fixation and impaction injuries surgeons judged unsafe to load. Of 435 randomised (meeting abstract), 418 formed the modified intention-to-treat sample (214 early, 204 delayed). The co-primary outcomes, within 365 days, were reoperation for hardware failure, non-union, loss of reduction, infection or fusion (non-inferiority margin −5.0 percentage points) and, if that was met, time to return to work. Reoperation occurred in 5 early (2.3%) vs 7 delayed (3.4%) patients (difference in complication-free survival 1.5 points, 90% CI −1.7 to 4.8, so non-inferior), and median return to work was 54 vs 69 days (HR 1.1, 95% CI 0.8 to 1.6; p=0.475).

2.3% vs 3.4%

• REOPERATION, EARLY VS DELAYED (5/214 VS 7/204); ARD −1.1 POINTS, CALCULATED FROM CRUDE COUNTS, 95% CI −4.3 TO 2.1; THE PAPER’S 1.5-POINT FIGURE IS THE DIFFERENCE IN COMPLICATION-FREE SURVIVAL
• +1.5 points COMPLICATION-FREE SURVIVAL, EARLY MINUS DELAYED (90% CI −1.7 TO 4.8; MARGIN −5.0)
• 54 vs 69 days MEDIAN RETURN TO WORK (95% CI 42 TO 77 VS 47 TO 89)
• HR 1.1 RETURN TO WORK (95% CI 0.8 TO 1.6); NO NNT AS NO DIFFERENCE SHOWN

CRITICAL APPRAISAL Focused question; registered primary outcome

Partly Co-primary outcomes and a −5.0-point margin stated, with gated testing; no registry id in either abstract

Design: RCT | Certainty: low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing new; it supports weight bearing from 2 weeks after ankle fixation without syndesmotic fixation, in line with WAX and with BOAST’s advice that most patients bear weight as tolerated, and shows no return-to-work gain | Confidence in that: moderate Main concerns: a wide −5-point margin against a 3.4% control rate favours a finding of non-inferiority; no per-protocol analysis is reported (ITT-type analyses favour non-inferiority), and 17 randomised patients were excluded from the modified ITT for unstated reasons (direction unknown); return to work depends on US employment and insurance, so the null result may not transfer to UK sick leave (direction unknown) Strengths: a multicentre randomised design across 19 trauma centres with a prespecified non-inferiority margin and gated superiority testing

UK practice: The BOA BOAST: The Management of Ankle Fractures (August 2016) says most patients should be allowed to bear weight as tolerated in a splint or cast unless there are specific concerns about the stability of fixation or contraindications such as peripheral neuropathy or soft-tissue concerns (standard 13), with fracture clinic review within 6 weeks (standard 14). The UK WAX trial (Bretherton CP, et al. Lancet 2024;403:2787–2797; doi:10.1016/S0140-6736(24)00710-4) randomised 561 patients in 23 NHS hospitals and found weight bearing at 2 weeks non-inferior to 6 weeks for the Olerud and Molander Ankle Score at 4 months (adjusted mean difference 4.47, 95% CI 0.58 to 8.37, favouring early), with similar complications (16% vs 14%; adjusted OR 1.18, 0.80 to 1.75) and similar mean NHS costs (£725 vs £785; difference −£60, 95% CI −342 to 232). This US trial adds reassurance on reoperation but excluded syndesmotic fixation, so it says nothing about loading after a syndesmotic screw or device.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 11 of 24

INFORMING PRACTICE

OBSERVATIONAL

FRCS

TFNA vs Its Predecessor PFNA for Trochanteric Fractures (Two Before-After Cohorts, n=475 and n=1,970): No Significant Difference in Any Reoperation (17.2% vs 12.9%) but More Major Revisions (7.8% vs 3.2%) and More Mechanical Failures (1.6% vs 0.3%)

Kuttner H, Schläppi M, Meier C, et al. The new trochanteric nail TFN-A has a higher complication rate in the treatment of proximal femoral fractures than the predecessor model PFN-A. Eur J Trauma Emerg Surg. 2026;52(1). doi:10.1007/s00068-026-03304-3. PMID: 42878747. Source route: PMC full text (PMC13672456). Paired with: Ben Zoor I, Fogel I, Balziano S, et al. Mechanical failure and revision risk following transition from PFNA to TFNA: a single-center cohort study of 1,970 patients. J Orthop Trauma. 2026 (online ahead of print). doi:10.1097/BOT.0000000000003303. PMID: 42854683. Source route for the paired paper: PubMed abstract only (the BMA Library cannot supply this journal); no published commentary found on either paper (journal commentary and web searched 3 Nov 2026). Appraised from full text

Kuttner and colleagues reviewed all cephalomedullary nails for proximal femoral fractures at one Swiss hospital (2017 to 2021; median age 85): 492 nails in 475 patients, 248 PFN-A and 244 TFN-A (DePuy Synthes), which replaced it in May 2019, with 24-month follow-up and 4 patients lost. Any reoperation did not differ significantly (42 of 244, 17.2%, vs 32 of 248, 12.9%; p=0.194), but major revision (nail exchange, conversion to arthroplasty or peri-implant fracture fixation) was more frequent with the TFN-A (19, 7.8%, vs 8, 3.2%; p=0.004; HR 3.57, 95% CI 1.42 to 9.94, given in the abstract only); reoperations for cut-out (13 vs 5; p=0.086) and nail breakage (7 vs 3; p=0.143) were not significantly different. In the paired Israeli cohort (abstract only; 1,970 patients over 65 with OTA/AO 31A fractures, short helical-blade nails, logistical switch to TFNA in March 2023), mechanical failure was 1.6% with TFNA vs 0.3% with PFNA (RR 5.36, 95% CI 1.66 to 17.3), driven by cut-out (1.2% vs 0.3%; RR 4.18, 95% CI 1.23 to 14.2).

• 7.8% vs 3.2% MAJOR REVISION, TFN-A VS PFN-A AT 24 MONTHS (19/244 VS 8/248)
• 17.2% vs 12.9% ANY REOPERATION, TFN-A VS PFN-A (P=0.194)
• 1.6% vs 0.3% MECHANICAL FAILURE, TFNA VS PFNA, BEN ZOOR (RR 5.36, 95% CI 1.66 TO 17.3)
• 80 vs 244 days MEDIAN TIME TO FAILURE, TFNA VS PFNA, BEN ZOOR

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Kuttner: reoperations from records, phone, relatives and GPs, deaths cross-checked with the federal registry; radiographic measures by the first author alone, blinding not stated

Design: Observational | Certainty: low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing yet as protocol; units using TFNA should audit cut-out, breakage and major revision against their previous nail | Confidence in that: moderate Main concerns: before-after designs mix the nail change with secular changes in surgeons and case mix; direction unknown; small event counts and an unadjusted HR given only in the abstract likely overstate the TFNA excess; about 38% of Ben Zoor’s patients lacked orthopaedic follow-up, so failures may be missed in either arm; direction unknown Strengths: two independent centres found the same direction of effect, one with near-complete 2-year follow-up Disagreement: A US registry of 19,215 nailed hip fractures (Okike K, et al. J Bone Joint Surg Am 2025; PMID: 39812725) found no higher adjusted aseptic revision with TFN/TFNA than with the Gamma nail, but pooled TFN with TFNA and did not compare with the PFNA.

UK practice: NICE CG124 (hip fracture: management, 2011, updated January 2023) recommends a sliding hip screw in preference to an intramedullary nail for trochanteric fractures above and including the lesser trochanter, except reverse oblique fractures (1.6.9), and a nail for subtrochanteric fractures (1.6.10), yet about three quarters of Kuttner’s fractures were AO 31-A1 or A2. The MHRA field safety notice list for 11 to 15 March 2024 includes a DePuy Synthes notice on 7 lots of TFNA femoral nails (11 March 2024), a sterility recall unrelated to nail design (Swissmedic copy of FSN 2348044); we found no MHRA alert on TFNA breakage or cut-out and no National Hip Fracture Database report comparing nail models.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 12 of 24

5 — NEUROTRAUMA & BURNS

INFORMING PRACTICE

OBSERVATIONAL

FRCS

Acute Subdural Haematoma With GCS 13–15 (10-Centre Cohort, n=1,157): Death or Hospice Discharge in 10% of Patients Aged 65 and Over vs 3% Under 65

Orlando A, Mazumder H, Harbour LF, et al. In older adults, “mild” by GCS does not mean low risk: multicenter study of 1,157 patients with mild TBI and acute subdural hematoma. J Trauma Acute Care Surg. 2026 (online ahead of print). doi:10.1097/TA.0000000000005206. PMID: 42886856. Source route: PubMed abstract only (the BMA Library cannot supply this journal); no published commentary found (PubMed comment links, JournalFeed, St Emlyn’s, REBEL EM and journal commentary searched 3 Nov 2026). Abstract only

Retrospective cohort from 10 centres (2022–2023) of 1,157 adults with an emergency department GCS of 13 to 15 and isolated acute traumatic subdural haematoma (SDH) on the first CT, 72% aged 65 or over and 58% transferred in; median non-head ISS was 1 in both age groups. The primary outcome, total mortality (death plus discharge to hospice), was 10% at 65 or over vs 3% at 18–64, and 11% vs 8% had a neurosurgical intervention. In older patients, total mortality was independently associated with an advance directive (aOR 11.8, 95% CI 2.9 to 49.0), male sex (3.0, 1.3 to 7.0), midline shift (1.8, 1.3 to 2.6), shock index (1.5, 1.1 to 2.1) and worsening SDH on follow-up CT (2.9, 1.3 to 6.6); the units for midline shift and shock index are not reported in the abstract. Of older patients, 51% went home and 7% to hospice, vs 78% and 2% of younger patients.

• 10% vs 3% DEATH OR HOSPICE DISCHARGE, 65+ VS 18–64 (NUMERATORS NOT REPORTED)
• 11% vs 8% NEUROSURGICAL INTERVENTION, 65+ VS 18–64
• 51% vs 78% DISCHARGED HOME, 65+ VS 18–64
• aOR 2.9 WORSENING SDH ON REPEAT CT, 65+ (95% CI 1.3 TO 6.6)

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly CT features from PACS and registry data by automated and manual methods; the outcome counts hospice discharge, a care decision, as death; its time frame is not reported

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing new: older patients with SDH and GCS 13–15 still need a network neurosurgical discussion, repeat CT if they worsen and an early goals-of-care conversation; the study does not show that MTC care changes outcome | Confidence in that: moderate Main concerns: counting hospice discharge as death merges treatment-limitation decisions with injury severity, likely overstating the lethality of the injury itself (and explaining the advance-directive aOR of 11.8); backward selection from 21 covariates with about 80 events in older patients (calculated from 10% of 72% of 1,157) likely overfits and inflates the odds ratios; with 58% transferred in, the cohort is enriched for patients already judged to need specialist care, likely overstating risk compared with all older patients with a small SDH Strengths: contemporary multicentre data with CT features read from PACS and an age-stratified analysis Disagreement: The authors say older adults with SDH should continue to receive the highest-level care. The London Major Trauma System guidance (2021) keeps older patients with isolated TBI who need no immediate neurosurgical input at the trauma unit, with rapid discussion and critical transfer to the MTC if they deteriorate.

UK practice: NICE NG232 (head injury: assessment and early management, May 2023) asks for a CT head scan within 8 hours for adults aged 65 or over with any loss of consciousness or amnesia (1.5.9), suggests considering one within 8 hours for people on anticoagulants or antiplatelets other than aspirin alone with no other indication (1.5.13), and asks for a neurosurgical discussion about any new, surgically significant abnormality (1.4.15); its committee said age or frailty should not be the sole indicator for admission. The London Major Trauma System guidance on management of older major trauma patients (3rd edition, April 2021) advises a low threshold for a trauma call and head CT after a low-level fall or in patients on anticoagulants, and lists GCS below 15 (even if baseline) as a trigger in its screening tool. It says any acute intracranial blood on CT should prompt a discussion with the MTC, and it lets older patients with isolated TBI who need no immediate neurosurgical input stay at the trauma unit with a repeat scan in 48 hours.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 13 of 24

INFORMING PRACTICE

OBSERVATIONAL

FRCS

Burn Size at Referral vs Burn Centre Assessment (Swedish Cross-Sectional Study, n=82): Referring Doctors Overestimated TBSA in 76%, Median 10% vs 6%, and Were 5 or More Points Out in 38%

Kjölhede A, Dogan S, Karlsson M. Accuracy of burn size estimation in referred patients: a comparison between referring units and a Swedish National Burn Centre. Burns. 2026;52(9):108250 (journal pre-proof). doi:10.1016/j.burns.2026.108250. PMID: 42874527. Source route: PDF via BMA Library. Appraised from full text

Prospective cross-sectional study at the Linköping Burn Centre, one of two national burn centres in Sweden (April 2024 to April 2025): the total body surface area (TBSA) estimate given by the referring doctor on the referral call (method not recorded) was compared with the burn surgeons’ Lund & Browder estimate after arrival, a median of 1 day (IQR 1–2) later, in 82 adults and children (median age 33.5 years, 82% male, 63% with more than two separate burn areas). Referring doctors overestimated burn size in 62 of 82 (76%) and underestimated it in 15 (18%); the median TBSA was 10% (IQR 3–17) vs 6% (1.5–13), and the gap widened as burns got larger (Bland–Altman bias = 1.09 + 0.176 × mean TBSA). Estimates differed by at least 5 TBSA points in 31 patients (37.8%) and by at least 10 points in 8 (9.8%), and 3 patients (4%) did not meet referral criteria on specialist review. A further 95 transferred patients had no study form and were excluded; their burns were much smaller (median 0.4% vs 6%), and only 27% of transferred women were included, compared with 55% of men.

• 76% REFERRER ESTIMATE ABOVE BURN CENTRE ESTIMATE (62/82)
• 10% vs 6% MEDIAN TBSA, REFERRER VS LUND & BROWDER (IQR 3–17 VS 1.5–13)
• 37.8% DIFFERED BY 5 TBSA POINTS OR MORE (31/82)
• 9.8% DIFFERED BY 10 TBSA POINTS OR MORE (8/82)

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Referrers’ method not recorded; reference was a burn surgeon’s Lund & Browder chart, itself not a gold standard; whether the surgeon knew the referral estimate is not reported

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing new; expect the TBSA on a referral call to be an overestimate, especially for scattered burns, and recalculate it with a Lund & Browder chart before fixing fluid rates or the level of burn service | Confidence in that: moderate Main concerns: excluding mostly small burns and including more men (who were overestimated more) likely overstates the average error across all referrals; a reference estimate made a day later, after burns have evolved and been cleaned, makes the error partly a timing effect (direction unknown); a single centre with no outcome data cannot show that the overestimates changed fluids, transfers or harm Strengths: estimates captured prospectively on the live referral call, with paired data and regression-based limits of agreement

UK practice: The National Burn Care Referral Guidance (National Network for Burn Care, version 1, February 2012; hosted by the British Burn Association) sets referral levels by TBSA: for adults, a burn unit from 10% to under 40% and a burn centre from 40% (25% with inhalation injury); for children, a unit from 5% to under 30% and a centre from 30%. The NHS England Burns Clinical Network Specification (PRN231102, 2023) still cites this 2012 guidance, requires each network to approve and ensure use of an electronic telemedicine and tele-referral system for referrals and specialist advice, and asks for IT supporting image transfer. The FPHC and British Burn Association consensus: Management of Burns in Pre-Hospital Trauma Care (version 2, April 2020) recommends a Lund & Browder chart, counting only partial or full thickness burns, and notes that TBSA is frequently overestimated while patients are under-resuscitated; the FPHC consensus statements page lists it as currently under review (checked 3 Nov 2026).

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 14 of 24

6 — MASS CASUALTY & MAJOR INCIDENT

INFORMING PRACTICE

OBSERVATIONAL

FRCS

FINAL FRCA

Severely Injured Patients in Routine Civilian Mass Casualty Incidents (TraumaRegister DGU, 430 of 65,152; 1:4 Matched): Prehospital Time 85 vs 69 Minutes, With No Clear Difference in In-Hospital Death (13.5% vs 11.7%; Adjusted OR 1.19, 95% CI 0.77 to 1.85)

Breuer F, Beckers SK, Brune B, et al. Prehospital management and hospital outcomes of severely injured patients during routine civilian mass casualty incidents: a 1:4 matched cohort analysis of the TraumaRegister DGU. BMC Emerg Med. 2026;26:282. doi:10.1186/s12873-026-01818-y. PMID: 42907761. TR-DGU Project ID 2021-012. Source route: PMC full text (PMC13689070, open access). Appraised from full text

Retrospective cohort from the German trauma registry (614 hospitals, mid-2020 to 2023): 65,152 patients with a Maximum AIS of 3 or more, of whom 430 (0.7%) were flagged as treated in a declared mass casualty incident (MCI), 84.1% of them after traffic collisions. Each MCI patient was matched to four controls on age decade, traffic involvement and AIS 3 or more in the head, thorax, abdomen and extremities, giving 423 MCI patients and 1,692 controls. The primary outcome, in-hospital death, occurred in 57 of 423 (13.5%) vs 198 of 1,692 (11.7%; p=0.317); after excluding early interhospital transfers it was 14.4% vs 12.7%, against RISC III-predicted rates of 13.9% vs 13.3%, and the adjusted odds ratio was 1.19 (95% CI 0.77 to 1.85). Against matched rather than all controls, the prehospital differences shrink: median prehospital time 85 vs 69 minutes, intubation 33.4% vs 28.3%, tranexamic acid 31.9% vs 24.6% and HEMS transport 34.6% vs 27.5%, whereas the abstract’s 33.6% vs 20.9% intubation and 32.1% vs 15.0% tranexamic acid compare with all controls, whose mean age was 56.1 vs 39.2 years.

• 13.5% vs 11.7% IN-HOSPITAL DEATH, MATCHED MCI VS CONTROL (57/423 VS 198/1,692)
• OR 1.19 IN-HOSPITAL DEATH, ADJUSTED (95% CI 0.77 TO 1.85)
• 85 vs 69 min MEDIAN PREHOSPITAL TIME, MATCHED (IQR 66–100 VS 54–83)
• 3 vs 2 days MEDIAN ICU STAY, MATCHED (IQR 1–9 VS 1–7)

Source: Tables 3, 4 and 5 of the paper; p-values are for the matched comparison. Denominators vary with missing data. CRITICAL APPRAISAL Exposure measured reliably

Partly MCI flag recorded per patient after local declaration (DIN 13050, no numerical threshold); no incident identifier, incident size or triage category; about 95% of German MCIs are MANV 5 or MANV 10 (up to 5 or 10 casualties)

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing yet for mortality; for surge planning, expect MCI casualties to reach hospital about a quarter of an hour later than usual, more often by HEMS, and to stay longer in ICU | Confidence in that: moderate Main concerns: scene deaths are not in the registry, so the comparison likely understates any harm from MCI conditions; matching on three variables left more penetrating trauma and higher ISS among MCI patients, which may bias mortality against the MCI group; RISC III adjustment only partly offsets this; unlinked patients from the same incident are treated as independent, so the CIs are likely too narrow Strengths: prospectively collected national registry from 614 hospitals with a stated primary outcome, matching plus a RISC III-adjusted sensitivity analysis

UK practice: The NHS England Concept of operations for managing mass casualties (published 21 July 2026) asks major trauma centres, trauma units and local EDs to agree in advance with their NHS England region and ambulance trust how many casualties in each priority they can take. It also asks receiving hospitals to free 20% of their beds within 12 hours, and hospitals with level 3 intensive care to prepare to double their level 3 ventilated capacity for at least 96 hours. These German incidents were mostly small (MANV 5 or 10), far from the hundreds or thousands of casualties the concept of operations plans for, but the longer ICU and hospital stays (median 12 vs 10 days, matched) support planning for sustained critical-care and ward demand after the first wave.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 15 of 24

INFORMING PRACTICE

OBSERVATIONAL

FRCS

Who Leaves the Scene First in Multi-Patient Road Traffic Incidents (Japan, n=143,130): First-Away Patients Had More Severe Outcomes (4.35% vs 1.87%; Adjusted Risk Difference 2.2 Points), Most Marked at the Shortest Scene Times, Consistent With Triage Selection Rather Than Harm

Kubo A, Tamura S, Hiraide A, et al. Prehospital transport prioritization and time-varying selection bias in traffic-related multi-patient incidents: a nationwide retrospective cohort study. Crit Care. 2026;30:524. doi:10.1186/s13054-026-06196-2. PMID: 42850620. No trial registration (registry study). Source route: PMC full text (PMC13647583, open access). Appraised from full text

Nationwide retrospective cohort from Japanese ambulance transport records (January 2021 to December 2023): 143,130 patients transported with signs of life from traffic incidents with two or more patients from one emergency call, of whom 62,769 (43.85%) were “priority transport”, the first patient to leave the scene. The primary outcome, death or critical or severe injury as rated by the receiving physician on arrival, occurred in 2,730 of 62,769 (4.35%) priority patients vs 1,499 of 80,361 (1.87%) others; the doubly robust adjusted risk difference was 2.2 percentage points (95% CI 2.01 to 2.36). The difference was 6.4 points (95% CI 5.6 to 7.2) at 5 minutes on scene, fell to 1.4 points (95% CI 1.3 to 1.6) at about 32 minutes and rose to 2.4 points (95% CI 2.0 to 2.8) at 60 minutes. The lesson in plain terms: the pattern is consistent with crews sending the worst-injured first, most sharply when they decide quickest, so in registry data without physiology “went first” (or “got the intervention”) is a marker of severity, and even sophisticated adjustment can make a sensible triage decision look harmful.

• 4.35% vs 1.87% SEVERE OUTCOME, PRIORITY VS OTHER TRANSPORT (2,730/62,769 VS 1,499/80,361)
• +2.2 points ADJUSTED RISK DIFFERENCE, AIPW (95% CI 2.01 TO 2.36)
• 6.4 vs 1.4 points RISK DIFFERENCE AT 5 VS ABOUT 32 MIN ON SCENE
• E-value 3.76 UNMEASURED CONFOUNDING NEEDED TO EXPLAIN IT AWAY (LOWER BOUND 3.57)

Source: Table 3 and Figure 2 of the paper. Interaction p<0.001; on-scene time is a post-exposure variable, so the curve is descriptive. CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Exposure is first departure, a proxy for triage intent (37,296 patients came from incidents with simultaneous departures); outcome is a physician severity rating on arrival, 87.7% of events being “severe injury” (3,709 of 4,228, calculated), with deaths only 455

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing in clinical practice; when auditing MTC outcomes after a multi-casualty incident, treat arrival order as a severity marker, not a measure of care quality | Confidence in that: moderate Main concerns: unmeasured injury severity makes priority transport look harmful, the bias the authors set out to show, so the 2.2-point difference should not be read as harm; excluding patients who arrested before arrival removes the sickest first-away patients, likely understating the severity gap; an arrival-only severity grade cannot show whether going first changed survival; direction unknown Strengths: nationwide registry with a causal diagram, doubly robust estimation, incident-level cluster bootstrap and a TMLE sensitivity analysis giving 2.4 points

UK practice: The NHS England Concept of operations for managing mass casualties (published 21 July 2026) asks ambulance emergency operations centres to confirm the immediate receiving capacity of every possible receiving centre within an hour of declaration, and asks MTCs, trauma units and local EDs to set in advance how many casualties in each priority they can take, aiming for “right patient, right place, right time, first time”. This paper suggests the first ambulances will carry a concentrated share of the most severely injured, so MTC Priority 1 capacity is likely to be used early. Japan has no regionalised trauma network, and most incidents here were small multi-patient crashes rather than declared mass casualty incidents, so the size of the effect will not transfer directly to a UK network.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 16 of 24

INFORMING PRACTICE

GUIDELINE

FRCS

UK

NATO Military Major Incident and Mass Casualty Doctrine Update (Consensus of 16 Experts From 7 NATO Countries): The Main Challenge Is Systemic Rather Than Purely Clinical, Framed as “Augment-Adapt-Transform”

Walker J, Freeman J, Wurmb T, et al. Medical management of major incidents and mass casualty events during military operations: a consensus statement. BMJ Mil Health. 2026 (Online First, 8 October 2026). doi:10.1136/military-2026-003458. PMID: 42855200. Source route: publisher-page abstract only (PubMed has no abstract; full text not accessible); no published commentary found (journal, PubMed linked comments, JournalFeed, St Emlyn’s, EMCrit and general web searched 3 Nov 2026). Abstract only

Consensus statement from the Multinational Military Working Group on major incidents and mass casualty incidents (MMI/MASCAL), 16 experts from 7 NATO countries tasked with updating NATO doctrine, with UK Defence Medical Services authors. It defines a major incident as a spectrum of demand surges needing extraordinary activity across medical and operational systems, and a mass casualty incident as the subset with an irreconcilable gap between casualty load and medical capacity, made harder on operations by dispersed forces, contested logistics and hybrid threats. Its main finding is that the principal challenge is systemic rather than purely clinical, presented through a new Augment-Adapt-Transform framework. It lists elements of an effective response: synchronised command and control, casualty modelling, triage, evacuation, treatment capacity, logistics and information-sharing under uncertainty, time pressure and degraded infrastructure; the consensus method, evidence base and specific recommendations are not reported in the abstract.

• 16 EXPERTS IN THE WORKING GROUP
• 7 NATO COUNTRIES
• Not reported CONSENSUS METHOD, EVIDENCE SEARCH AND RECOMMENDATION GRADING (ABSTRACT)

CRITICAL APPRAISAL Developer and funding; conflicts managed

Partly NATO multinational working group; funding and competing interests “none declared”; how conflicts were managed is not reported

Design: Guideline | Certainty: very low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing yet for civilian MTC practice; useful background for MTCs that may receive military casualties or Defence Medical Services support in a UK mass casualty incident | Confidence in that: moderate Main concerns: abstract only, so the content of any recommendation is unverified; direction unknown; expert consensus without a reported evidence search may overstate how well-founded the framework is; military operational context (contested logistics, hybrid threats) is indirect for a civilian UK trauma network Strengths: a multinational group formally tasked with updating NATO doctrine, including UK Defence Medical Services clinicians

UK practice: For civilian incidents in England, the NHS England Concept of operations for managing mass casualties (published 21 July 2026) says DHSC can ask the MoD to make available uncommitted Defence Medical Services personnel with recent operational experience of blast and high-velocity injuries, with requests going via the regional EPRR team under Military Aid to the Civil Authorities (MACA). The NHS England EPRR Framework (version 3, July 2022) makes NHS England regional teams co-ordinate and endorse NHS requests for military assistance. The JESIP Joint Doctrine: The Interoperability Framework (edition 3.1, April 2024) treats the military as a non-categorised responder: civil authorities always lead, with Defence in support, and Defence personnel are expected to follow the JESIP principles where possible.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 17 of 24

7 — PAEDIATRIC TRAUMA

INFORMING PRACTICE

RCT

PAEDS

FRCS

Video Consultation Added to the Phone Call for Injured Children in Community EDs (Stepped-Wedge Cluster RCT, 10 EDs, n=595): No Difference in Parent Experience, Distress or Transfer (91.6% vs 93.8%)

Marcin JP, Rosenthal JL, Tancredi DJ, et al. Telemedicine consultation for pediatric trauma in community ED visits and parent experience of care: a stepped-wedge cluster-randomized clinical trial. JAMA Netw Open. 2026;9(10):e2636546. doi:10.1001/jamanetworkopen.2026.36546. PMID: 42905708. NCT04469036. Source route: full text (PDF; open access); supplements not read. Appraised from full text

Stepped-wedge cluster RCT in northern California (November 2020 to November 2022): 10 community EDs (2 adult level II and 3 level III trauma centres, 5 undesignated hospitals) switched in random order, at 8-week steps, from telephone-only consultation with the regional level I paediatric trauma centre to telephone plus a live bedside video link with the child and family, run mainly by trauma nurse practitioners after COVID-19 changed staffing; 595 injured children under 18 were included (369 in telemedicine periods, 226 in telephone periods). The primary outcomes, parent experience of care (selected ED CAHPS items, scored 0 to 1) and parent distress (State-Trait Anxiety Inventory) at 3 days or more, did not differ: adjusted mean differences 0.041 (95% CI −0.008 to 0.090) and −0.035 (−0.304 to 0.234; Table 2, where the abstract and text give −0.031, −0.297 to 0.235). Transfer to the paediatric trauma centre, a secondary outcome, occurred in 338 of 369 (91.6%) vs 212 of 226 (93.8%) (adjusted OR 0.59, 0.16 to 2.16). Only 165 of 369 (44.7%) in telemedicine periods actually had a video consultation, and the one positive result, the “leaving the ED” subscale (0.104, 0.023 to 0.186), was not adjusted for multiple comparisons.

• 0.768 vs 0.775 OVERALL PARENT EXPERIENCE, TELEMEDICINE VS TELEPHONE, UNADJUSTED MEANS; ADJUSTED MEAN DIFFERENCE 0.041 (95% CI −0.008 TO 0.090)
• 91.6% vs 93.8% TRANSFERRED TO PAEDIATRIC TRAUMA CENTRE (ADJUSTED OR 0.59, 0.16 TO 2.16)
• ARD −2.2 points TRANSFER, CALCULATED, UNADJUSTED AND IGNORING CLUSTERING (95% CI −6.4 TO 2.0); NO NNT
• 44.7% OF THE TELEMEDICINE ARM ACTUALLY HAD A VIDEO CONSULTATION (165/369)

CRITICAL APPRAISAL Focused question; registered primary outcome

Yes NCT04469036 with a published protocol (Trials 2022); two primary outcomes, parent experience and distress

Design: RCT | Certainty: low (GRADE-style, our judgement) | What it changes for UK trauma care: Nothing yet; adding a parent-facing video call to an MTC advice call did not reduce transfers or parent distress, so any network that adds video should measure both | Confidence in that: moderate Main concerns: video delivered to under half the telemedicine arm biases every comparison towards the null; lower survey response in the telemedicine periods (53.7% vs 64.6%) may bias the parent-reported outcomes in either direction; the single positive result among six subscales, unadjusted for multiplicity, likely overstates any benefit at discharge or transfer Strengths: randomised stepped-wedge design with a published protocol, time-adjusted mixed models and an independent funder (PCORI)

UK practice: NICE NG40 (major trauma: service delivery, 2016) asks trauma units and MTCs to have a paediatric trauma team available immediately for children with major trauma (1.4.3, 1.4.5), compatible records across the network such as an image exchange portal (1.7.2), a clear point of contact for families (1.9.2), and verbal and written information on the reason for any transfer (1.9.5). It makes no recommendation on video consultation between hospitals, and a search found no RCPCH or RCEM guidance on it (3 Nov 2026). This trial tested video added to the telephone referral, not a replacement for image transfer, and most children were transferred either way.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 18 of 24

8 — QUICK HITS

SAFETY NOTICES AND BLOOD MHRA Field Safety Notice: Reliance Medical Traumafix trauma dressing (FSN dated 30 September 2026; MHRA reference 40975235) (safety notice): listed in MHRA’s 28 September to 2 October roundup, published 9 October 2026. Traumafix is a pressure dressing sold for controlling severe bleeding; we could not open the notice itself, so the affected lots and the fault are not reported here, and ED, theatre and major incident stock should be checked against the manufacturer’s FSN. MHRA. Field Safety Notices: 28 September 2026 to 2 October 2026. gov.uk Other MHRA Field Safety Notices relevant to trauma theatres and IR, 5 to 23 October 2026 (safety notices): Philips Azurion interventional fluoroscopic X-ray system (6 October), Maquet Flow-i, Flow-c and Flow-e anaesthesia systems (5 October), B Braun Space infusion pumps (14 October), a Shimadzu X-ray high-voltage generator (14 October), Intersurgical spirometry sets and anaesthetic breathing systems (13 October) and Philips IntelliVue patient monitors (19 to 23 October roundup). None was listed for orthopaedic implants, tourniquets, rapid infusers, blood warmers, cell salvage, external fixators, pelvic binders, chest drains or CT; the notice texts were not read, so check them with your medical devices team. MHRA. Field Safety Notices: 5 to 9 October, 12 to 16 October and 19 to 23 October 2026. gov.uk SHOT: new position for Jehovah’s Witness patients (news): the Jehovah’s Witnesses governing body has decided that individuals can decide whether to receive blood components; SHOT says the position on whole blood is unchanged. Confirm each patient’s own wishes directly. SHOT accepts reports for any patient transfused despite withholding consent, under the avoidable transfusion category. Serious Hazards of Transfusion. SHOT Newsletter October 2026 (6 October 2026). shotuk.org (PDF) NHSBT/UKHSA Epidemiology Unit 2025 annual review (news): there were no confirmed transfusion-transmitted infections among 110 investigated cases. The risk of testing missing HBV, HIV or HCV because of a window-period donation is around 1 per million donations tested. This may help when discussing transfusion risk. NHSBT. The Update, 29 October 2026. hospital.blood.co.uk

NEUROTRAUMA GFAP and UCH-L1 cut-offs in TBI with GCS 13–15 (abstract only; TRACK-TBI secondary analysis, 18 US level I trauma centres, n=1,718): day-1 GFAP separated CT-positive from CT-negative patients with an AUC of 0.82 (95% CI 0.80 to 0.84), and MRI-positive from MRI-negative patients among those with a negative CT with an AUC of 0.76 (0.71 to 0.81), whereas UCH-L1 reached only 0.60. The median GFAP cut-off giving an adjusted positive predictive value of 0.9 for CT pathology was 2,879 pg/mL (95% CI 2,195 to 4,704) at an assumed prevalence of 36% (from the cohort; observed 37%), and 5,191 pg/mL (3,779 to 5,426) at 10%. These are rule-in cut-offs from a high-prevalence US cohort, and NICE NG232 (2023) made only a research recommendation on biomarkers for acute complications. Eagle SR, et al. J Neurosurg 2026. PMID: 42913506 Decompressive craniectomy in children with severe TBI (abstract only; retrospective analysis of the international ADAPT cohort via FITBIR, 2014–2017, n=686 with GCS 8 or less and an ICP monitor): 213 children (31.0%) had a craniectomy, which was not associated with good functional recovery (GOS-E Peds below 5) at 6 months or more after discharge (OR 0.64, 95% CI 0.39 to 1.05); in a secondary analysis of all admitted patients, it was not associated with in-hospital mortality (OR 1.02, 0.60 to 1.74). The comparison is non-randomised, so confounding by indication is likely; read alongside the adult RCT review in Section 1. Wong AM, et al. J Trauma Acute Care Surg 2026. PMID: 42853259 ICU management of traumatic brain injury (review): Tiered treatment now targets an ICP of 22 mmHg or less, and the lower CPP target has risen from 50 to 60 mmHg. On surgery, DECRA (n=155) found that early bifrontal decompressive craniectomy improved ICP control but worsened functional outcome at 6 and 12 months; RESCUEicp (n=408) found that late rescue craniectomy reduced mortality at the cost of more severe disability and vegetative state. The authors reserve it for carefully selected patients after medical therapy has been optimised. Plummer MP, et al. BJA Educ 2026;26:418–426. PMID: 42764890

CHEST AND CRITICAL CARE Observation vs chest drain for blunt traumatic pneumothorax (CONSTRAIN-Retro, retrospective cohort, 9 Dutch hospitals, n=275 clinically stable patients aged 16 or over): 126 (45.8%) were observed and 149 (54.2%) had a chest drain; practice varied most for pneumothoraces of 15% to 50% on chest radiograph (36.6% observed), of whom 14 of 48 (29.2%) observed patients later needed a drain. In that medium-sized subgroup (n=131), after weighting, complications did not differ (32.8% vs 30.2%; OR 1.14, 95% CI 0.40 to 3.28) and ICU admission, a secondary outcome, was lower with observation (2.4% vs 17.5%; OR 0.13, 95% CI 0.03 to 0.64). Before weighting, drained patients had larger pneumothoraces and higher ISS; weighting balanced pneumothorax size, but ISS, ventilation and mechanism stayed imbalanced. Hartman A, et al. J Trauma Acute Care Surg 2026;100:863–870. doi:10.1097/TA.0000000000005113. PMID: 42468511 Damage control thoracic surgery (review, Baltimore): There are no rigid indications; the authors list physiological triggers (pH below 7.2, lactate above 5 mmol/L, base deficit above 15, temperature below 34 °C, coagulopathy, massive transfusion) and injury patterns such as resuscitative thoracotomy with aortic occlusion. Most operative lung injuries can be managed with lung-sparing repair or non-anatomic resection, and stapled tractotomy is their first-line manoeuvre for deep non-hilar bleeding; after trauma pneumonectomy, reported mortality is 50% to 100%. Much of it describes the authors’ own institutional practice; published in 2025. Ghneim MH, et al. J Trauma Acute Care Surg 2025;98:11–19 (online October 2024). PMID: 39375907 What’s new in trauma critical care (review): The ESICM trauma section argues for managing major trauma as a multisystem condition, not a list of anatomical injuries. Its most useful point for the trauma team: haemorrhage is the most frequent reversible cause of trauma death, but obstructive causes such as pneumothorax and tamponade are frequently underdiagnosed, distributive shock after spinal cord injury may overlap with bleeding, and a catecholamine surge can cause acute heart failure, especially in severe TBI. Opinion piece, no new data. Gauss T, Dünser MW. Intensive Care Med 2026. PMID: 42911398

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 19 of 24

BURNS AND PREGNANCY Burns with combined injuries (US narrative review, full text): combined burn and trauma injuries are under 5% of civilian burn centre injuries but carry two to six times the mortality of either alone in US databank studies. The authors advise damage control resuscitation until haemorrhage control, with plasma-based burn resuscitation in the background. No search method was stated; the text says “4 years or older” where its tables say 14, and its high-dose vitamin C rescue option predates the VICTORY trial. In the UK, the 2012 national burn referral guidance advises discussion with the burn service by TBSA thresholds, with treatment agreed between the trauma and burn services. Stewart BT, Hunter MA, Johnson L, et al. J Trauma Acute Care Surg 2025;99(1):10–23. PMID: 40074715 Pregnancy and trauma (US narrative review, full text): a sound revision summary: blood volume rises by up to 50% by 32 weeks, so up to 2 L may be lost before signs of shock, and CT should not be withheld when indicated. Apply UK thresholds rather than the review’s: FPHC (November 2025) considers resuscitative hysterotomy when the fundus is at or above the umbilicus (over 20 weeks, not 23) and says pregnancy is not a contraindication to TXA; the SHOT anti-D aide-memoire gives at least 500 IU within 72 hours after abdominal trauma from 12 weeks, plus a Kleihauer test from 20 weeks. The review’s anti-D dose of 1,250 µg at 12 weeks or under looks like an error. Burruss S, Jebbia M, Nahmias J. J Trauma Acute Care Surg 2025;98(2):190–196. PMID: 39496074

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 20 of 24

CORE REVISION — MAJOR TRAUMA ISSUE 8 — NOVEMBER 2026

Core Revision — Spinal Trauma: ASIA Classification, Steroids and Surgical Timing

ISNCSCI and AIS grades, spinal vs neurogenic shock, NICE NG41 immobilisation and imaging, methylprednisolone, decompression timing, MAP targets — FRCS (T&O), FRCS (Gen), Final FRCA

Exam goal: Classify a spinal cord injury with the ISNCSCI steps and AIS grades, tell spinal shock from neurogenic shock, apply NICE NG41 to immobilisation and imaging in adults and children, give a balanced answer on methylprednisolone (NASCIS II and III against current guidance), and quote the evidence for decompression within 24 hours and for MAP targets.

1. ISNCSCI: LEVELS AND THE AIS GRADE

The ISNCSCI worksheet (ASIA/ISCOS) sets the order. (1) Sensory level on each side: the most caudal intact dermatome for both pin prick and light touch. (2) Motor level on each side: the most caudal key muscle function graded at least 3 (on supine testing), provided the key muscle functions above it are intact (graded 5); where there is no myotome to test, the motor level is presumed to be the same as the sensory level if testable motor function above it is normal. (3) Neurological level of injury (NLI): the most cephalad of the sensory and motor levels. (4) Complete or incomplete: the injury is complete only if voluntary anal contraction (VAC) is absent, all S4–5 sensory scores are 0 and deep anal pressure (DAP) is absent. (5) AIS grade (table). (6) Zone of partial preservation, used only when sacral motor or sensory function is absent. In an apparent AIS B, test non-key muscles more than 3 segments below the motor level on each side to separate B from C or D.

2. SPINAL SHOCK VS NEUROGENIC SHOCK

Spinal shock is the acute, transient loss of motor, sensory and reflex function below the level of injury. Neurogenic shock is tissue hypoperfusion from disruption of sympathetic control of vascular tone, especially with injuries above T6: hypotension with bradycardia, and the skin may be warm and pink. It has been documented to last up to 4 to 5 weeks (StatPearls 2023). In the multiply injured patient, distributive shock after spinal cord injury may overlap with haemorrhagic shock (ESICM 2026), so treat haemorrhage as the cause until proven otherwise.

3. NICE NG41: IMMOBILISATION, DESTINATION AND IMAGING

Canadian C-spine rule: high risk is age 65 or over, a dangerous mechanism (for example a fall from more than 1 metre or 5 steps, axial load to the head, high-speed or rollover collision, ejection) or paraesthesia in the limbs. A person with a low-risk factor who can actively rotate the neck 45 degrees left and right does not need imaging. Immobilisation in adults: manual in-line stabilisation, then a correctly sized semi-rigid collar unless the airway is compromised or there is a known deformity such as ankylosing spondylitis (keep the spine in its current position); reassess the airway; then scoop stretcher, head blocks and tape, ideally in a vacuum mattress. In uncooperative, agitated or distressed patients, including children, consider letting them find a comfortable position with manual in-line stabilisation. A longboard is for extrication only, never for transport. Destination: suspected spinal cord injury goes to a major trauma centre irrespective of transfer time unless an immediate life-saving intervention is needed, and never directly from scene to a spinal cord injury centre; children with suspected column injury go to a major trauma centre. Adults (16 or over): CT if the Canadian C-spine rule indicates imaging, or if there is strong suspicion of thoracic or lumbosacral injury with abnormal neurology. If a neurological abnormality could be due to cord injury, perform MRI after CT even when CT shows nothing (clarified May 2021). Suspected T1–L3 column injury without neurological signs: X-ray first, then CT if the X-ray is abnormal or there are clinical signs or symptoms of column injury. A new column fracture means imaging the rest of the spine. Children (under 16): MRI if there is strong suspicion of cervical cord or column injury; consider plain X-rays if MRI criteria are not met but suspicion remains after repeated clinical assessment, and discuss them with a consultant radiologist; do not routinely use whole-body CT. The spinal surgeon at the MTC or trauma unit should contact the linked spinal cord injury centre consultant within 4 hours of diagnosis.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 21 of 24

4. THE STEROID CONTROVERSY

NASCIS II (NEJM 1990, double-blind RCT): methylprednisolone 30 mg kg−1 bolus then 5.4 mg kg−1 h−1 for 23 h (n=162) vs naloxone (n=154) vs placebo (n=171). At 6 months, patients given methylprednisolone within 8 h had better motor change scores than placebo (16.0 vs 11.2, P=0.03); those treated after 8 h, and those given naloxone, did not differ from placebo. NASCIS III (JAMA 1997, n=499, all within 8 h and all given the bolus) compared 24 h with 48 h of methylprednisolone and with tirilazad. It advised 24 h if started within 3 h and 48 h if started at 3–8 h, but the 48-h regimen started at 3–8 h brought more severe sepsis and severe pneumonia. Guidance diverges. AANS/CNS moved from listing methylprednisolone as an option, with a harm caveat (2002), to a level I recommendation against it (2013). The AO Spine 2017 guideline wording is: “We suggest not offering a 24-hour infusion of high-dose MPSS to adult patients who present after 8 hours with acute SCI”; “We suggest a 24-hour infusion of high-dose MPSS be offered to adult patients within 8 hours of acute SCI as a treatment option”; and “We suggest not offering a 48-hour infusion of high-dose MPSS to adult patients with acute SCI.” All three are weak recommendations, and the motor benefit behind the second was small (3–4 motor points). So AO Spine leaves a weak option within 8 h, whereas the UK says no.

UK position (NICE NG41, 2016): do not use methylprednisolone, nimodipine or naloxone, aimed at neuroprotection, in the acute stage after traumatic spinal cord injury (recommendation 1.7.3). The management of each spinal cord injury should be agreed between spinal surgery and spinal cord injury specialists (1.7.2). nice.org.uk/guidance/ng41

5. TIMING OF DECOMPRESSION AND MAP TARGETS

STASCIS (PLoS One 2012) was a prospective cohort, not an RCT: 313 adults aged 16–80 with cervical cord injury at 6 North American centres, early surgery (under 24 h, n=182, mean 14.2 h) vs late (n=131, mean 48.3 h). Of the 222 followed up at 6 months, 19.8% vs 8.8% improved by 2 or more AIS grades (OR 2.57, 95% CI 1.11 to 5.97; adjusted OR 2.83, 1.10 to 7.28). The AO Spine 2017 guideline made a weak suggestion (low-quality evidence) to offer early surgery regardless of level, and in central cord syndrome. The AO Spine/Praxis 2024 update made it a strong recommendation (moderate-quality evidence): offer surgery within 24 hours, regardless of level, when medically feasible. Its pooled estimates were a risk ratio of 2.76 (95% CI 1.60 to 4.98) for a 2-grade AIS gain at 6 months and 4.50 extra ASIA motor points (95% CI 1.70 to 7.29). It made no recommendation on ultra-early surgery (under 12 h). NG41 does not address operative timing. MAP: the 2013 AANS/CNS guideline recommended maintaining MAP between 85 and 90 mmHg for the first 7 days. The AO Spine/Praxis 2024 haemodynamic guideline, which updated it, suggests augmenting MAP to a lower limit of at least 75–80 mmHg, not actively beyond an upper limit of 90–95 mmHg, for 3–7 days. Both are weak recommendations based on very low-quality evidence. Systolic pressure below 90 mmHg should be avoided or corrected; no recommendation was made on the choice of vasopressor. Older adults and associated injuries: age 65 or over is itself a high-risk factor for cervical injury in NG41, as is age 65 or over with thoracic or lumbosacral pain, and being known to have, or at risk of, osteoporosis is a factor for thoracolumbar injury.

FRCS / FINAL FRCA FOCUS NLI is the most cephalad of the sensory and motor levels; complete means no VAC, no DAP and zero

S4–5 sensation; C vs D turns on half the key muscles below the NLI reaching grade 3. Spinal shock is neurological; neurogenic shock is hypotension with bradycardia, typically above T6. NG41: CT for adults, MRI for children with strong suspicion of cervical injury, MRI after CT for any cord-type deficit, no longboard transport, no methylprednisolone. Decompress within 24 h (AO Spine 2024, strong); MAP at least 75–80 and no higher than 90–95 mmHg for 3–7 days (weak). Sources: [S1, S2] NICE NG41 2016 (updated May 2021); [S3] ASIA/ISCOS ISNCSCI worksheet REV 04/26; [S4] Bracken MB et al, NEJM 1990, PMID 2278545; [S5] Bracken MB et al, JAMA 1997, PMID 9168289; [S6] Fehlings MG et al, AO Spine MPSS guideline, Global Spine J 2017, PMID 29164025; [S7] Fehlings MG et al, STASCIS, PLoS One 2012, PMID 22384132; [S8] Fehlings MG et al, Global Spine J 2017, PMID 29164024; [S9] Fehlings MG et al, Global Spine J 2024, PMID 38526922; [S10] Kwon BK et al, Global Spine J 2024, PMID 38526923; [S11] Dave S et al, Neurogenic shock, StatPearls 2023; [S12] Gauss T, Dünser MW, Intensive Care Med 2026, PMID 42911398.

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 22 of 24

9 — ACTION POINTS

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 23 of 24

TRIALS TO WATCH

RESULTS EXPECTED SOON

RECRUITING

RECENTLY REPORTED

Jake Turner — Curated with the assistance of AI (Claude). All content editorially reviewed. Feedback form · emevidence.org · hello@emevidence.org

Major Trauma Evidence Rundown, Issue 8, November 2026. Page 24 of 24