PHEM Evidence Rundown — Issue 10 (November 2026)

PHEM EVIDENCE RUNDOWN — ISSUE 10 — NOVEMBER 2026

PHEM Evidence Rundown

Pre-Hospital Emergency Medicine — UK Edition — monthly — emevidence.org

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 — bystander CPR: a French stepped-wedge cluster RCT (15 agencies randomised; intention-to-treat analysis of 12 agencies, n=1,684) of dispatcher training, a volunteer-responder app and audit raised bystander CPR from 61.4% to 70.9% (adjusted OR 1.54, 95% CI 1.09 to 2.19; not significant after adjusting for time trends), with no significant difference in 30-day survival (aOR 1.30, 0.71 to 2.38). Change this month: none. No paper or guideline this month was strong enough, or directly applicable enough, to justify a Change badge; the action points are checks against existing UK guidance. UK: TETRiS national thoracotomy evaluation • GFAP after prehospital RSI (UK HEMS) • ambulance ramping and 28-day mortality (UNCORKED) • AACE position statement on public bleed kits. Paediatric PHEM (4 items): births before arrival, paediatric cardiac arrest in Australia and New Zealand, video vs direct laryngoscopy in children, and TRiM peer support after distressing jobs. Core Revision: Paediatric resuscitation in the 2025 guidelines — what changed for prehospital teams.

BOTTOM LINE UP FRONT — NOVEMBER 2026

ACT ON THIS NOW

Nothing this month changes practice on its own. The action points at the end are checks against existing UK guidance (RCUK 2025, NICE NG39, JESIP, Safer Prehospital Anaesthesia 2026).

WORTH KNOWING ON SHIFT

UK Thoracotomy (TETRiS, abstract): 116 UK resuscitative thoracotomies, 70.7% on scene: ROSC 43.1% and survival to discharge 6.9%; none of the 8 survivors had been in arrest for more than 10 minutes. Too few survivors to change the RCUK limit of under 15 minutes.

OBS TXA: fibrinolytic phenotype is a post-randomisation subgroup; keep giving TXA to bleeding major trauma within 3 hours per NICE NG39.

DX GFAP after RSI (UK HEMS, n=65): sensitivity 97.0% but specificity 53.1% for intracranial haemorrhage; a positive result barely shifts probability (LR+ 2.07, calculated).

PAEDS Births before arrival (France, n=3,473): 62.7% of newborns with a recorded temperature were hypothermic on scene; start thermal care at once and measure the temperature.

REVIEW Chemical agents: decontaminate first; atropine with early pralidoxime for nerve agents; rocuronium rather than suxamethonium; empiric hydroxocobalamin for suspected cyanide.

KNOW FOR NEXT TIME

OBS Dispatcher CPR barriers (Sweden, n=3,263): 43.7% of recognised arrests then met a barrier, most often a caller unable to move the patient; 30-day survival 6.5% vs 14.8% (adjusted RR 0.57, 95% CrI 0.46 to 0.71).

PAEDS Rescue breaths (Japan, non-cardiac OHCA, n=154,137): linked to lower 30-day mortality in children (adjusted RR 0.88, 0.83 to 0.94) and after drowning, with no clear difference in adults.

OBS ECPR (Göttingen model): 20.5% met local criteria at EMS arrival; modelled collapse-to-flow was 37 minutes on scene vs 75 minutes in hospital.

UK Ramping (131 UK EDs): each 5-point rise in escalation care held in ambulances was associated with a 2.1% higher 28-day mortality hazard (HR 1.021, 1.002 to 1.041); fragile and post hoc.

OBS HEMS airway (Poland, n=8,980): a supraglottic airway as the final airway had higher unadjusted odds of death at handover, but confounding by indication makes this uninterpretable as device harm.

RCT Chest pain conveyance (PreHEART, n=5,170, abstract): non-inferior for 3-day serious events (win ratio 1.01, 0.89 to 1.16) with more patients left at home (34.6% vs 24.5%); not for UK use yet.

OBS Older adults in traumatic arrest (France, n=5,595): 30-day survival 0.5% vs 1.2% under 65; age alone is not a reason to withhold resuscitation.

PAEDS Paediatric OHCA (ANZ, n=3,139): 10.0% survived; each extra minute of EMS response was linked to 4% lower odds of survival.

PAEDS Video laryngoscopy in children (18 studies): higher first-pass success (RR 1.11, 1.06 to 1.16) at very low certainty; the two prehospital studies were inconclusive.

PHEM Evidence Rundown, Issue 10, November 2026. Page 1 of 31

November’s lead is a French trial that did what many UK services are trying to do: train dispatchers in telephone CPR, send volunteers by app and audit every call. Bystander CPR rose, but survival did not clearly follow. Two registry studies sit beside it: what stops callers starting CPR once the call handler has recognised the arrest, and whether rescue breaths matter in arrests that are not cardiac. Around it sit four UK papers: the national TETRiS evaluation of resuscitative thoracotomy, a HEMS study of point-of-care GFAP after RSI, the UNCORKED analysis of ambulance ramping and mortality, and AACE’s new position on public bleed kits. Elsewhere, ECPR logistics, a pooled look at TXA by fibrinolytic phenotype, prehospital fluids in major burns and olanzapine for stimulant agitation. Paediatrics has four items, from births before arrival to peer support for crews after distressing jobs, and the Core Revision walks through what the 2025 paediatric resuscitation guidelines changed for prehospital teams.

WHAT’S INSIDE — ISSUE 10

• Lead: dispatcher CPR, volunteer app and audit bundle (French stepped-wedge cluster RCT)
• Bystander and dispatcher CPR: barriers to telephone CPR (Sweden); rescue breaths vs compression-only in non-cardiac arrest (Japan)
• Cardiac arrest: ECPR eligibility and timings; HEMS airway devices in Poland; ambulance ramping in UK EDs
• Medical: GFAP after RSI (UK HEMS); PreHEART chest pain conveyance; scene blood gas; olanzapine for methamphetamine agitation
• Trauma: TETRiS thoracotomy (UK); TXA by fibrinolytic phenotype; signs of life in traumatic arrest; older adults in traumatic arrest; prehospital fluid in major burns
• UK: AACE position statement on public bleed kits
• Paediatric PHEM (4): births before arrival; paediatric OHCA in Australia and New Zealand; video vs direct laryngoscopy; TRiM after distressing jobs
• FOAMed and appraisal: St Emlyn’s on catastrophic midface bleeding; chemical warfare agents (BJA Education)
• Quick Hits (31): including CABARET head-up CPR, the ERC 2025 corrigenda, opioid-associated arrest, ventilation feedback, pelvic binders and mass casualty incidents
• Core Revision: paediatric resuscitation in the 2025 guidelines

CHANGE TONIGHT Act immediately CHANGE THIS MONTH Adjust practice SAFETY ALERT

GUIDELINE

INFORMING

PRACTICE PAEDS PHEM UK FOAMED DIPIMC/FIMC

Contents: 1. Key PHEM Articles & Trials — 2. Guidelines & UK Updates — 3. Paediatric PHEM — 4. FOAMed & Critical Appraisal — 5. Quick Hits — Core Revision — 6. Action Points — Trials to Watch

PHEM Evidence Rundown, Issue 10, November 2026. Page 2 of 31

1 — KEY PHEM ARTICLES & TRIALS

LEAD

INFORMING PRACTICE

RCT

DIPIMC/FIMC

Dispatcher Training, Volunteer App and Audit to Raise Bystander CPR (French Stepped-Wedge Cluster RCT, n=1,684): Bystander CPR Rose From 61.4% to 70.9%, but 30-Day Survival Did Not Differ Significantly

Debaty G, Ribes S, Ray P, et al. A multifaceted intervention to increase bystander CPR rates after out-of-hospital cardiac arrest in France: a stepped-wedge, cluster-randomised, controlled trial. Lancet Public Health. 2026;11:e878–e888. doi:10.1016/S2468-2667(26)00200-8. PMID: 42901186. ClinicalTrials.gov NCT03633370. Source route: PDF via BMA Library (open access, CC BY 4.0). Appraised from full text

Fifteen French physician-staffed EMS agencies were randomised to the order in which they switched from usual care to a three-part intervention: dispatcher training in arrest recognition and scripted compression-only telephone CPR, activation of trained volunteers through the SAUV Life geolocation app, and monthly audit and feedback. The primary outcome was bystander-initiated CPR before EMS arrival in consecutive adults with non-traumatic OHCA; the comparator was each agency’s own control period. Two agencies withdrew and were replaced and one enrolled no one, so 14 agencies enrolled 2,160 patients; the intention-to-treat analysis was limited to the 12 originally randomised agencies (686 control and 998 intervention patients), bystander CPR occurred in 61.4% vs 70.9% (adjusted OR 1.54, 95% CI 1.09 to 2.19), mostly through more telephone-directed CPR (51.3% to 61.9%); volunteer CPR rose from 0.4% to 10.9%. The effect lost significance after adjusting for agency-specific time trends (OR 1.38, 95% CI 0.98 to 1.94), and no clinical outcome differed significantly.

• 61.4% vs 70.9% BYSTANDER CPR, CONTROL VS INTERVENTION PERIODS (CRUDE; CONFOUNDED BY CALENDAR TIME)
• OR 1.54 BYSTANDER CPR, ADJUSTED (95% CI 1.09 TO 2.19)
• 6.9% vs 9.3% 30-DAY SURVIVAL, AOR 1.30 (0.71 TO 2.38)

Source: Debaty et al, Table 3. Denominators for AED use are the group totals; the table gives n (%) only for intervention 30-day survival and AED use. CRITICAL APPRAISAL Registered primary; analysis plan

Partly NCT03633370, primary unchanged; but the analysis plan was amended after outcome data were available to make the scheduled-period ITT primary

Design: RCT | Certainty: low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing new: UK services already train call handlers in telephone CPR; the trial supports auditing recognition and telephone-CPR rates rather than adding an app alone | Confidence in that: moderate Main concerns: analysis plan changed after outcome data and estimate sensitive to time trends, so the 1.54 likely overstates the effect; more bystander-witnessed arrests in intervention periods overstate benefit (adjusted OR 1.39, CI crosses 1); the app’s co-founder is an author, a possible bias towards the intervention Strengths: large multicentre cluster trial in routine 24/7 practice with a complete primary outcome and transparent sensitivity analyses

UK practice: The RCUK Adult Basic Life Support Guidelines (2025) already ask services to train ambulance call handlers in arrest recognition and telephone CPR, to give compression-only instructions by default, and to integrate dispatch with volunteer first-responder systems to bring a nearby AED. This trial supports those priorities but shows no survival gain; in this trial, AED use before EMS arrival did not rise.

PHEM Evidence Rundown, Issue 10, November 2026. Page 3 of 31

INFORMING PRACTICE

OBSERVATIONAL

DIPIMC/FIMC

Barriers After the Call Handler Recognises Cardiac Arrest (Swedish Registry-Linked 112 Calls, n=3,263): Present in 43.7% of Calls, With Later Bystander CPR (Primary Outcome) and Lower 30-Day Survival (Adjusted RR 0.57, 95% CrI 0.46 to 0.71)

Boberg E, Jonsson M, Riva G, et al. Barriers to dispatcher-assisted CPR are associated with delayed bystander CPR and worse outcomes in out-of-hospital cardiac arrest: a retrospective cohort study. Resuscitation. 2026;228:111305. doi:10.1016/j.resuscitation.2026.111305. PMID: 42727702. Source route: PDF via BMA Library (open access, CC BY 4.0). Appraised from full text

Retrospective cohort of transcribed 112 emergency calls pooled from five Swedish OHCA studies (2017–2025), linked to the Swedish Registry for Cardiopulmonary Resuscitation, including EMS-treated arrests that the dispatcher had recognised and where CPR was not already under way. Reviewers blinded to outcome coded barriers arising after recognition (caller not with the patient, hung up or left the phone, language barrier, overly distraught caller, unable to move the patient, refusal), compared with barrier-free calls; the primary objective was time to bystander CPR and the secondary outcomes were ROSC and 30-day survival. Barriers occurred in 1,428 of 3,263 calls (43.7%), most often inability to move the patient (55.8% of barrier calls), and bystander CPR started in 68.8% vs 88.3%, with a median 270 vs 156 seconds to first compressions; each extra barrier added delay (103, 174 and 225 seconds for one, two and three or more barriers). Crude 30-day survival was 6.5% vs 14.8% (adjusted RR 0.57, 95% credible interval 0.46 to 0.71) and ROSC 32.4% vs 39.9% (adjusted RR 0.89, 0.81 to 0.98).

• 43.7% OF RECOGNISED ARRESTS HAD AT LEAST ONE BARRIER AFTER RECOGNITION (1,428/3,263)
• 68.8% vs 88.3% BYSTANDER CPR, BARRIER VS BARRIER-FREE CALLS
• 6.5% vs 14.8% 30-DAY SURVIVAL, CRUDE; ADJUSTED RR 0.57 (95% CRI 0.46 TO 0.71)
• +103 to +225 s ESTIMATED EXTRA DELAY TO BYSTANDER CPR WITH ONE TO THREE OR MORE BARRIERS

Source: Boberg et al, Tables 1 and 2 and Fig. 4. Bayesian logistic regression adjusted for age, sex, location, witnessed status, presumed medical cause and EMS response time; CrI, credible interval.

CRITICAL APPRAISAL Exposure measured reliably

Partly Structured checklist from call audio, reviewers blinded to outcome; inter-rater check on 20 calls, result not reported

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing new for crews; for ambulance call-handling leads, audit post-recognition barriers and reinforce starting CPR where the patient lies, including on a bed | Confidence in that: moderate Main concerns: barriers such as “caller not with the patient” also mark unwitnessed, later-found arrests, so residual confounding likely overstates the harm attributable to the barrier itself; missing CPR timestamps treated as no CPR, which the authors say may overstate non-initiation in barrier calls; a selected research cohort with 20% of calls unclassifiable, direction unknown Strengths: large audio-reviewed dataset linked to national registry outcomes, with a dose-response between barrier count and delay

UK practice: The RCUK Adult Basic Life Support Guidelines (2025) ask call handlers to use standardised protocols, to give CPR instructions to all callers (compression-only by default) and to integrate dispatch with volunteer first-responder systems. They also say rescuers should not move a person from a soft surface such as a bed to the floor, but start CPR on the bed and compress deeper, which addresses the commonest barrier here. This complements this issue’s Debaty trial, where a dispatcher-training bundle worked mainly through more telephone-directed CPR.

PHEM Evidence Rundown, Issue 10, November 2026. Page 4 of 31

INFORMING PRACTICE

OBSERVATIONAL

PAEDS

DIPIMC/FIMC

Rescue Breaths vs Compression-Only Bystander CPR in Non-Cardiac OHCA (All-Japan Utstein Registry, n=154,137): Adjusted RR for 30-Day Mortality 0.98 Overall, 0.88 in Children and 0.94 After Drowning

Iida Y, Obara T, Nojima T, et al. Rescue-breathing versus chest compression-only bystander cardiopulmonary resuscitation in non-cardiac out-of-hospital cardiac arrest: a nationwide retrospective cohort study. Resuscitation. 2026;228:111332. doi:10.1016/j.resuscitation.2026.111332. PMID: 42790858. Source route: PDF via BMA Library (open access, CC BY 4.0). Appraised from full text

Nationwide registry cohort (2012–2023) of non-cardiac, non-traumatic OHCA given bystander chest compressions with rescue-breathing status recorded; 20,124 (13.1%) had compressions plus rescue breaths and 134,013 compression-only CPR. The primary outcome was 30-day mortality, estimated with modified Poisson regression adjusted for age, sex, witnessed arrest, dispatcher instruction, bystander AED use, presumed cause and call-to-EMS-contact time (complete cases 61.2%). Crude mortality was 89.8% vs 94.1% (adjusted RR 0.98, 95% CI 0.98 to 0.99), so much of the crude difference of 4.3 per 100 disappeared after adjustment. The association was clearest in children under 18 (60.4% vs 78.3%; adjusted RR 0.88, 0.83 to 0.94), after drowning (adjusted RR 0.94, 0.92 to 0.96) and in children after drowning (adjusted RR 0.67, 0.55 to 0.82), with no clear difference in adults or older people.

• 13.1% HAD RESCUE BREATHS WITH COMPRESSIONS (20,124/154,137)
• 10.2% vs 5.9% 30-DAY SURVIVAL, RESCUE BREATHS VS COMPRESSION-ONLY (CRUDE, CALCULATED)
• 39.6% vs 21.7% CHILDREN: 30-DAY SURVIVAL (CRUDE, CALCULATED); ADJUSTED RR FOR MORTALITY 0.88 (0.83 TO 0.94)
• RR 0.98 30-DAY MORTALITY OVERALL, ADJUSTED (95% CI 0.98 TO 0.99)

Source: Iida et al, Tables 2 and 3. Crude rates are not adjusted; the large crude gaps shrink after adjustment. CRITICAL APPRAISAL Exposure measured reliably

Partly Registry field; 49,179 excluded because rescue-breathing information was missing; content of dispatcher instructions unknown

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing new: it supports existing UK advice to give rescue breaths to children and after drowning when trained and willing; it does not support changing compression-only telephone CPR for adults | Confidence in that: moderate Main concerns: rescue breaths came more often from non-family bystanders (likely trained) and in patients with a pulse at EMS contact, neither adjusted for, so the benefit is likely overstated; rescue breathing fell from 20.7% (2014) to 5.7% (2023) and year was not adjusted for, direction unknown; complete-case analysis of 61.2% could bias either way Strengths: nationwide 12-year population with prespecified age and cause subgroups and crude counts reported for every comparison

UK practice: The RCUK Paediatric Life Support (basic and advanced) guidelines (2025) ask call handlers to encourage rescue breaths with compressions in children of all ages, with five initial breaths and a 30:2 ratio for untrained bystanders, and compression-only CPR only if bystanders are unwilling or unable. The RCUK Adult Basic Life Support Guidelines (2025) keep compression-only telephone instructions as the default for adults, moving to 30:2 if the caller knows how to give rescue breaths. For drowning, the RCUK Special Circumstances Guidelines (2025) advise starting with five ventilations using 100% oxygen when available.

PHEM Evidence Rundown, Issue 10, November 2026. Page 5 of 31

INFORMING PRACTICE

DIAGNOSTIC

UK

DIPIMC/FIMC

Point-of-Care GFAP in Non-Traumatic Coma After Prehospital RSI (East Anglian Air Ambulance, n=65): Sensitivity 97.0% but Specificity 53.1% for Intracranial Haemorrhage on CT

Price J, Lachowycz K, Palti G, et al. Prehospital Point-of-Care GFAP for Detection of Intracranial Hemorrhage in Acute Nontraumatic Coma. JAMA Netw Open. 2026;9(10):e2639337. doi:10.1001/jamanetworkopen.2026.39337. PMID: 42879151. Source route: PMC full text (PMC13673314; open access, CC BY-NC-ND); the supplementary threshold table was not available. Appraised from full text

Diagnostic study of consecutive adults (16 or older) with non-traumatic coma (GCS 12 or less) who had prehospital rapid sequence induction (RSI) with suspected intracranial haemorrhage (ICH), attended by a UK physician-paramedic HEMS between January 2025 and April 2026. Whole-blood glial fibrillary acidic protein (GFAP) and UCH-L1 were measured on scene with the i-STAT Alinity traumatic brain injury cartridge (results within 15 minutes), against CT as the reference standard read blind to the biomarker. Of 65 patients (median GCS 4), 33 (50.8%) had ICH; GFAP above the manufacturer’s 65 pg/mL threshold had sensitivity 97.0% (95% CI 84.2% to 99.9%), specificity 53.1% (34.7% to 70.9%) and AUC 0.94 (0.87 to 0.99), and the one false negative was an isolated basal subarachnoid haemorrhage. UCH-L1 discriminated worse than chance (AUC 0.30), and calling either test positive gave sensitivity 100.0% (89.4% to 100.0%) at specificity 15.6% (5.3% to 32.8%).

• 97.0% GFAP SENSITIVITY (95% CI 84.2 TO 99.9)
• 53.1% GFAP SPECIFICITY (95% CI 34.7 TO 70.9)
• LR+ 2.07 CALCULATED (95% CI 1.42 TO 3.01)
• LR− 0.06 CALCULATED (95% CI 0.01 TO 0.40)
• n=65 33 WITH ICH ON CT

Calculated from 32 true positives, 1 false negative, 17 true negatives and 15 false positives (LR+ 2.07, LR− 0.057). At the upper CI of LR− (0.40), a negative result at a 30% prior still leaves about 15%. A negative result keeps residual risk at 2% or less only when the pre-test probability is 26% or lower (calculated). The 10% and 30% priors are our assumptions, not from the paper.

CRITICAL APPRAISAL Representative spectrum, consecutive enrolment

Partly Consecutive, but only RSI patients with suspected ICH (65 of 338 RSI patients with GCS 12 or less, after 61 traumatic cases were excluded); 13 of 32 without ICH had a prehospital cardiac arrest

Design: Diagnostic | Certainty: low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet: a positive GFAP (LR+ 2) barely shifts the destination decision, and a negative result should not be used to avoid a neurosciences centre outside a prospective evaluation | Confidence in that: moderate Main concerns: high prevalence (50.8%) in a selected RSI cohort from one service overstates the predictive values for wider use; only 6 subarachnoid haemorrhages were included and one was missed, so sensitivity for subarachnoid haemorrhage may be overstated; false positives were not broken down by diagnosis; with 13 post-arrest patients among the ICH-negative group, specificity in coma without arrest is uncertain (direction unknown) Strengths: consecutive UK HEMS cohort with a blinded CT reference standard and complete 30-day follow-up, reported to STARD

UK practice: NICE NG128 (stroke and TIA in over 16s, 2019, last updated April 2022) recommends a validated tool such as FAST outside hospital (1.1.1) and immediate non-enhanced CT for suspected stroke with a GCS below 13 (1.3.2); it makes no recommendation on blood biomarkers. Prehospital GFAP is therefore a service-evaluation tool in the UK, and destination decisions for comatose patients after RSI stay with local HEMS and network policy.

PHEM Evidence Rundown, Issue 10, November 2026. Page 6 of 31

INFORMING PRACTICE

OBSERVATIONAL

UK

Ambulance Ramping vs Corridor Care, UNCORKED Secondary Analysis (131 UK EDs, 18,744 Admissions): Each 5-Point Rise in the Share of Escalation Care Held in Ambulances Was Linked to 2.1% Higher 28-Day Mortality Hazard

McHenry RD, Birse F, Clarke B, Roberts T, on behalf of the Trainee Emergency Research Network. Ambulance or corridor? The association between site-level use of ambulance ‘ramping’ as Emergency Department escalation areas and 28-day mortality in admitted patients: a secondary analysis of the UNCORKED study. Scand J Trauma Resusc Emerg Med. 2026 (article in press). doi:10.1186/s13049-026-01710-5. PMID not yet assigned (PubMed checked by DOI, 1 Nov 2026). ISRCTN16396025. Source route: open-access PDF. Appraised from full text

Post hoc analysis of a prospective snapshot study of 131 type 1 EDs in England, Wales and Northern Ireland (five snapshots, March 2025; Scotland had no patient-level data), including 18,744 admitted patients aged 16 or over. The exposure was each site’s share of all escalation-area time spent in ambulances waiting more than 15 minutes to hand over; 82 sites (62.6%) used ambulances this way. The primary outcome was 28-day all-cause mortality (6.6% overall), in a Cox model adjusted for age, sex, trauma designation, admission rate and the site’s total escalation burden. Any ambulance use was not associated with mortality (HR 0.957, 95% CI 0.808 to 1.134), but among users each 5-point rise in the ambulance share was (HR 1.021, 95% CI 1.002 to 1.041); crude mortality was 8.9% at sites above 15% vs 6.4% at sites with no ambulance holding.

• 62.6% OF SITES HELD PATIENTS IN AMBULANCES (82/131)
• 8.9% vs 6.4% 28-DAY MORTALITY, INDEX >15% VS NO AMBULANCE USE (UNADJUSTED)
• HR 1.021 PER 5-POINT RISE IN AMBULANCE SHARE (95% CI 1.002 TO 1.041)
• HR 0.957 ANY AMBULANCE USE (95% CI 0.808 TO 1.134)

CRITICAL APPRAISAL Exposure measured reliably

Partly Site-level index from five snapshots; areas self-categorised by sites; ambulance time counted only beyond 15 min

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing for crews yet; it adds observational support for timely handover and release of held crews, which services already pursue | Confidence in that: moderate Main concerns: site-level exposure applied to all admissions may reflect hospital organisation rather than ambulance waits, likely overstating harm from ramping; no acuity measure, so residual confounding by severity is possible (direction unknown); post hoc analysis with a new index and a borderline result (lower CI 1.002; any-use HR null), so the finding is fragile Strengths: national prospective sample with an objective outcome and hospital-clustered analysis

UK practice: According to the AACE update on W45 rapid release (April 2025), the NHS England 2025/26 priorities and operational planning guidance asked systems to ensure that no handover takes longer than 45 minutes and to work towards handovers within 15 minutes; we could not open the NHS England document itself. The RCEM position statement on temporary escalation spaces (December 2024) states that safe, good-quality care is not possible in corridors, so this paper does not support moving waits from ambulances into corridors.

PHEM Evidence Rundown, Issue 10, November 2026. Page 7 of 31

INFORMING PRACTICE

OBSERVATIONAL

DIPIMC/FIMC

ECPR Eligibility in Göttingen, 10-Year OHCA Model (n=1,696): 20.5% Met Local Criteria at EMS Arrival, but Modelled Collapse-to-Flow Was 37 Minutes On Scene vs 75 Minutes After Intra-Arrest Transport

Grimm C, Illian H, Kunze-Szikszay N, et al. Modelling eCPR eligibility after out-of-hospital cardiac arrest: a 10-year retrospective temporospatial analysis from a mixed urban-rural region in Germany. Scand J Trauma Resusc Emerg Med. 2026;34:159. doi:10.1186/s13049-026-01681-7. PMID: 42823760. Source route: PMC full text (PMC13628803). Appraised from full text

Retrospective registry cohort of 1,696 resuscitation attempts (2015–2025) in the city and county of Göttingen (about 328,000 people), with each scene geocoded and road time to the ECPR centre modelled. At EMS arrival, 348 (20.5%) met the local protocol (age 16–75, witnessed, non-traumatic, no-flow 10 minutes or less) and 149 (8.8%) a stricter modified ELSO definition, about 46 and 20 patients a year. If transport had started after 10, 15 or 20 minutes of on-scene CPR, 54.0%, 48.0% and 43.1% of the locally eligible would still have arrived without ROSC within 60 minutes. Adding published procedural times, median collapse-to-ECPR was 37 minutes with an ECPR team dispatched at the same time as the ambulance, 54 minutes if dispatched after 10 minutes, and 75 minutes for in-hospital ECPR after transport, which met the 60-minute target in under 1%.

• 20.5% MET LOCAL ECPR CRITERIA AT EMS ARRIVAL (348/1,696)
• 47.4% STILL NEEDED AND COULD GET ON-SCENE ECPR WITHIN 60 MIN, EARLY TEAM DISPATCH (165/348)
• 31.6% SAME, TEAM DISPATCHED AFTER 10 MIN (110/348)
• 1 patient IN-HOSPITAL ECPR WITHIN 60 MIN AFTER TRANSPORT, WITHOUT PRIOR ROSC

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Registry no-flow time missing in 40.5% (classed ineligible; best case 24.9% eligible); no pH, lactate or end-tidal CO2

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet for crews; for HEMS and network planners, judge any intra-arrest transport ECPR pathway by modelled collapse-to-flow time, not by arrival at the door | Confidence in that: moderate Main concerns: fixed procedural time assumptions penalise in-hospital ECPR, overstating the on-scene advantage; eligibility without physiological, comorbidity or team-availability exclusions overstates programme volume; one German physician-staffed region with short transport times (median 9.3 min), so applicability to UK geography is uncertain Strengths: ten years of geocoded population data that separate hospital arrival from the start of extracorporeal flow

UK practice: The RCUK Adult Advanced Life Support Guidelines (2025) say ECPR may be considered as rescue therapy for selected adults when conventional CPR is failing, in settings where it can be implemented. In the London Sub30 feasibility study (Resuscitation 2025; PMID: 39645021), the prehospital team was despatched to 18 patients and 5 were cannulated, reaching full flow at a mean 47 minutes from collapse; none met the 30-minute target and 2 survived to discharge.

PHEM Evidence Rundown, Issue 10, November 2026. Page 8 of 31

INFORMING PRACTICE

OBSERVATIONAL

DIPIMC/FIMC

Airway Strategy in Polish National HEMS (n=8,980 Adults, 2020–2024): Half Were Intubated by HEMS, and an SGA Left as the Final Airway Was Linked to Worse Unadjusted Status at Handover

Grażewicz M, Ingielewicz A, Rzońca P, et al. Advanced airway management practice patterns and short-term survival in Poland’s national physician-staffed helicopter emergency medical service, 2020–2024: a retrospective observational study. Scand J Trauma Resusc Emerg Med. 2026 (article in press). doi:10.1186/s13049-026-01713-2. PMID not yet assigned (PubMed checked by DOI, 1 Nov 2026). Source route: PDF via BMA Library (open access). Appraised from full text

Retrospective registry study of 8,980 adults given a tracheal tube (ETI) or supraglottic airway (SGA) by ground EMS or the national physician-staffed HEMS, comparing airway strategies within four clinical groups, with HEMS ETI as the reference. The primary outcome was being alive at hospital handover, which in cardiac arrest required ROSC; 2,950 (32.9%) were not. HEMS ETI was the commonest strategy (49.8%), followed by EMS ETI (22.6%) and SGA then ETI (19.0%). In non-traumatic arrest, sequential SGA-to-ETI did not differ from HEMS ETI, whereas an EMS-placed SGA left as the final airway had higher unadjusted odds of not being alive (OR 2.17, 95% CI 1.66 to 2.84); in trauma without arrest the unadjusted OR was 22.59 (95% CI 14.69 to 34.73).

• 32.9% NOT ALIVE AT HANDOVER OVERALL (2,950/8,980)
• 49.8% HEMS INTUBATION, COMMONEST STRATEGY
• 71.5% vs 53.6% NON-TRAUMATIC ARREST, NOT ALIVE: EMS SGA FINAL VS HEMS ETI (CALCULATED)
• 66.4% vs 8.0% TRAUMA WITHOUT ARREST, NOT ALIVE: EMS SGA FINAL VS HEMS ETI (CALCULATED)

Source: Grażewicz et al, Tables 1 and 3. Odds of not being alive at hospital handover; no adjustment for any confounder. CRITICAL APPRAISAL Exposure measured reliably

Partly Registry strategy only; operator, attempts, drugs, video or direct laryngoscopy and timing relative to ROSC not recorded

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing: do not read this as evidence against SGAs; keep using a second-generation SGA as the stepwise or rescue airway and focus on first-pass success | Confidence in that: high Main concerns: confounding by indication: an SGA may be left in place when further escalation looks futile or after failed intubation, overstating harm from SGAs; the outcome requires ROSC, which also prompts intubation (reverse causation), overstating benefit from intubation; no adjustment for anything, so every OR is crude Strengths: national five-year dataset that describes the SGA-then-intubation hand-offs between ground crews and HEMS

UK practice: The Association of Anaesthetists’ Safer Prehospital Anaesthesia guidelines (Lockey et al, Anaesthesia 2026, which replace the 2017 version; endorsed by FPHC RCSEd, DAS and RCEM, among others) say videolaryngoscopy should always be available wherever prehospital emergency anaesthesia is done, list a second-generation SGA as rescue equipment, and require operators trained to the standard of independent emergency anaesthesia in the ED. The RCUK Adult Advanced Life Support Guidelines (2025) advise a stepwise airway approach in cardiac arrest, with intubation only by rescuers with over 95% success within two attempts and waveform capnography.

PHEM Evidence Rundown, Issue 10, November 2026. Page 9 of 31

INFORMING PRACTICE

RCT

DIPIMC/FIMC

PreHEART Score-Guided Ambulance Conveyance for Chest Pain (Dutch RCT, n=5,170): Non-Inferior for 3-Day Serious Events (Win Ratio 1.01) as Non-Conveyance Rose From 24.5% to 34.6%

Sagel DC, Vlaar PJ, Wang D, et al. PreHEART score for emergency medical services triage of chest pain: a randomised trial. Heart. 2026 (published online 29 September 2026). doi:10.1136/heartjnl-2026-328924. PMID: 42810856. NL-OMON27238. Source route: abstract only (BMA Library cannot supply current Heart); PubMed abstract, Medscape Medical News summary (7 October 2026) and the trial registry records NL-OMON27238 and NL-OMON48009. Abstract only

Pragmatic, open-label, blinded-endpoint non-inferiority RCT run by Dutch ambulance services: 5,170 adults with undifferentiated chest pain were randomised 1:1 to conveyance guided by the PreHEART score or usual care (crew judgement, usually the nearest ED). The score combines history, ECG, age, sex-specific cardiovascular risk and point-of-care troponin I (Medscape); 0–3 points means stay at home with GP review within 24 hours if the patient agrees, 4–7 the nearest ED and 8–10 a PCI centre (registry). The primary outcome, a hierarchical composite of unexpected serious adverse events at 3 days analysed by win ratio (non-inferior if the lower 95% CI bound exceeded 0.80), was met: win ratio 1.01 (95% CI 0.89 to 1.16), with no deaths within 3 days among the 1,526 non-conveyed patients. Non-conveyance rose (34.6% vs 24.5%), 30-day ED presentations fell (RR 0.88, CI not reported) and costs fell by €345, but Medscape reports that the per-protocol analysis did not show non-inferiority.

• WR 1.01 3-DAY SERIOUS EVENTS (95% CI 0.89 TO 1.16); MARGIN 0.80
• 34.6% vs 24.5% NON-CONVEYANCE, PREHEART VS USUAL CARE
• ARD 10.1 points NON-CONVEYANCE, CALCULATED; CI NOT REPORTED; ABOUT 1 EXTRA PATIENT LEFT AT HOME PER 10 ASSESSED
• RR 1.48 ACS DIAGNOSIS IN WOMEN WITH PREHEART (CI NOT REPORTED; INTERACTION NOT SIGNIFICANT AFTER HOLM CORRECTION)

CRITICAL APPRAISAL Focused question; registered primary outcome

Partly NL-OMON27238 registers unexpected serious adverse events at 3 and 30 days; the win-ratio hierarchy and the 0.80 margin are not on the registry pages we read

Design: RCT | Certainty: low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet: do not leave chest pain patients at home on a PreHEART score in UK practice; it is a signal for a UK evaluation with point-of-care troponin | Confidence in that: moderate Main concerns: Medscape reports that the per-protocol analysis failed and about half lacked returned consent, so the intention-to-treat result likely overstates safety; a 3-day primary horizon (event counts not published) could miss later harms, which would overstate safety; Medscape reports non-inferiority persisted to 12 months; more ACS diagnoses (RR 1.48) and angiography (RR 1.66) in women with PreHEART may mean better detection or overdiagnosis (direction unknown) Strengths: a large pragmatic randomised test of the conveyance decision itself, with blinded endpoint assessment Disagreement: PreHEART advises non-conveyance for scores of 0–3 (registry), whereas NICE CG95 advises emergency referral whenever an ACS is suspected and pain is current, or occurred in the last 12 hours with an abnormal or unavailable ECG.

UK practice: NICE CG95 (recent-onset chest pain of suspected cardiac origin, 2010, last updated November 2016) says refer people to hospital as an emergency if an ACS is suspected and they currently have chest pain, or are pain free but had pain in the last 12 hours and the resting 12-lead ECG is abnormal or not available (1.2.1.7). It also says do not assess symptoms of an ACS differently in men and women (1.2.1.5), which makes the sex-specific signal here worth watching.

PHEM Evidence Rundown, Issue 10, November 2026. Page 10 of 31

INFORMING PRACTICE OBSERVATIONAL

Prehospital Blood Gas Analysis by German Emergency Physicians (98 Patients vs 390 Matched Controls): Non-Conveyance 27.6% vs 8.7%, and None of 27 Left at Home Re-Contacted Services in 30 Days

Lux H, Roth J, Hemmer S, et al. Moving diagnostics upstream: prehospital blood gas analysis is associated with safe community care and improved patient selection for hospital admission. Emerg Med J. 2026 (published online 13 October 2026). doi:10.1136/emermed-2026-216096. PMID: 42869640. Not registered (service evaluation). Source route: PDF via BMA Library. Appraised from full text

Retrospective service evaluation in the physician-staffed ground EMS of Jena, Germany (July 2023 to May 2024): 98 adults in whom the emergency physician chose to run a venous blood gas on scene (epoc analyser; no protocol linked results to disposition) were compared with 390 propensity-matched EMS patients without one, matched on age, sex, NACA severity score and ICD-10 category. The primary outcomes were non-conveyance and, among transported patients, admission. Non-conveyance was 27.6% (27/98) vs 8.7% (34/390), OR 3.98 (95% CI 2.26 to 7.01), and 41 of 71 transported blood gas patients (58%) were admitted; no admission rate was reported for controls. None of the 27 non-conveyed blood gas patients had an ED reattendance or repeat EMS contact within 30 days, against 3 of 34 (8.8%) repeat EMS contacts among non-conveyed controls (p=0.248), and median scene time was 37.0 vs 28.5 minutes.

• 27.6% vs 8.7% NON-CONVEYANCE (OR 3.98, 95% CI 2.26 TO 7.01)
• 58% (41/71) ADMITTED AMONG TRANSPORTED BLOOD GAS PATIENTS; CONTROL RATE NOT REPORTED
• 0/27 vs 3/34 30-DAY REPEAT EMS CONTACT AFTER NON-CONVEYANCE (P=0.248); UPPER 95% CI FOR 0/27 IS 12.8%, CALCULATED
• 37.0 vs 28.5 min MEDIAN SCENE TIME (P<0.001)

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Device use recorded; re-contact traced only in local EMS and hospital systems, so events elsewhere would be missed and safety overstated

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet: a venous gas can inform a HEMS or critical care paramedic treat-and-refer decision, but any service doing so should audit 30-day re-contact and admission | Confidence in that: moderate Main concerns: physician-level and indication confounding likely overstate the effect of the blood gas itself; local-only follow-up of 27 patients likely overstates safety; no control admission rate, so “improved selection for admission” cannot be judged Strengths: a pragmatic, transparent evaluation with sensitivity analyses (adjusted OR 6.14 in the full unmatched cohort)

UK practice: The NHS England Urgent and emergency care plan 2025/26 (June 2025) notes that half of ambulance incidents convey patients to an ED and asks services to reduce avoidable conveyance and the variation in see and treat rates (currently 25.6% to 36.7%). This German model is physician-delivered, so the UK analogue is HEMS and critical care paramedic teams that already carry point-of-care analysers.

PHEM Evidence Rundown, Issue 10, November 2026. Page 11 of 31

INFORMING PRACTICE

OBSERVATIONAL

DIPIMC/FIMC

Paramedic Oral Olanzapine for Methamphetamine Intoxication (Winnipeg, n=1,041): More Transported to the ED (99% vs 88%; aOR 11.83), but Prehospital Restraint Was Also More Common

McDonald N, Enns JE, Friesen EL, et al. Impact of prehospital olanzapine for methamphetamine intoxication on patient and system-level outcomes: a population-based retrospective cohort study. BMJ Open. 2026;16(10):e122263. doi:10.1136/bmjopen-2026-122263. PMID: 42849953. Source route: PMC full text (PMC13656143; open access, CC BY-NC). Appraised from full text

Population-based retrospective pre-post cohort linking Winnipeg Fire Paramedic Service narratives to Manitoba health and social data: 1,041 people at their first methamphetamine-related contact (2016–2020) judged eligible for the paramedic protocol (use within 48 hours with a mental health complaint or mild to moderate agitation; severe agitation and acute medical problems excluded). From December 2018 paramedics could give olanzapine 10 mg as an orodispersible tablet; 222 treated patients were compared with 241 untreated patients after the policy and 578 before it, using inverse probability of treatment weighting (IPTW). Against untreated patients after the policy, treated patients were more often transported to the ED (99% vs 88%; adjusted OR 11.83, 95% CI 3.19 to 43.80) and less often left without being seen (11% vs 19%; adjusted OR 0.56, 0.32 to 0.97). Prehospital physical or chemical restraint was more common after olanzapine than before the policy (5% vs 1%; adjusted OR 3.89, 1.49 to 10.15), and admission or transfer did not differ (7% in each group).

• 99% vs 88% TRANSPORTED TO ED, TREATED VS UNTREATED AFTER POLICY (AOR 11.83, 3.19 TO 43.80)
• 11% vs 19% LEFT ED WITHOUT BEING SEEN (AOR 0.56, 0.32 TO 0.97)
• 5% vs 1% PREHOSPITAL RESTRAINT, TREATED VS PRE-POLICY (AOR 3.89, 1.49 TO 10.15)
• 7% vs 7% ADMITTED OR TRANSFERRED, TREATED VS UNTREATED AFTER POLICY (CRUDE)

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Eligibility judged retrospectively from free-text narratives by 2 paramedics blinded to era and treatment; no agitation scores, sedation depth or adverse effects recorded

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing: this does not support a UK paramedic olanzapine protocol; it shows more transport, not calmer or safer patients | Confidence in that: moderate Main concerns: confounding by acceptance (people who refuse the tablet also refuse transport) overstates the benefit; no safety data (sedation, airway, hypotension, QT) means harms are invisible, overstating net benefit; the higher restraint rate after olanzapine may reflect sicker treated patients or real treatment failure (direction unknown) Strengths: population-wide linkage of paramedic records to ED and hospital outcomes, with blinded eligibility adjudication

UK practice: NICE NG10 (violence and aggression: short-term management, 2015) defines rapid tranquillisation as parenteral medication when oral medication is not possible or appropriate, and recommends intramuscular lorazepam alone or intramuscular haloperidol with intramuscular promethazine in adults, taking possible intoxication into account (1.4.37). In the UK, orodispersible olanzapine is licensed for schizophrenia and moderate to severe manic episodes (Zyprexa Velotab 10 mg SmPC (eMC, revised February 2024)), so use for stimulant intoxication would be off-label.

PHEM Evidence Rundown, Issue 10, November 2026. Page 12 of 31

INFORMING PRACTICE

OBSERVATIONAL

UK

DIPIMC/FIMC

Resuscitative Thoracotomy Across UK HEMS and Major Trauma Centres (TETRiS, n=116, 70.7% Prehospital): ROSC 43.1%, Survival to Discharge 6.9%; None of the 8 Survivors Had Been in Arrest for More Than 10 Minutes

Ramage L, Tucker H, Marsden ME, et al; TETRiS Collaborative. Traumatic Emergency Thoracotomy for Resuscitation in Shock (TETRiS): a national service evaluation of resuscitative thoracotomy practice outside the operating theatre. Resuscitation. 2026 (online 28 October 2026):111371. doi:10.1016/j.resuscitation. 2026.111371. PMID: 42950975. Source route: PubMed abstract only (closed access; not in PMC). We found no published commentary on this paper (web search, 1 November 2026). Abstract only

Prospective 12-month service evaluation of resuscitative thoracotomy (RT) for traumatic cardiac arrest or peri-arrest performed outside the operating theatre by 20 UK HEMS and 26 major trauma centre emergency departments, run through the PHOTON, TERN and NaTRIC research networks; there was no comparator group and the abstract names no primary outcome. Of 116 patients (median age 30, 83.6% male, 69.8% penetrating injury), 82 (70.7%) had RT on scene. Suspected tamponade was the commonest indication (72 of 116, 62.1%), but exsanguinating haemorrhage was the operative finding in 82 (70.7%); 50 (43.1%) achieved ROSC and 8 (6.9%) survived to hospital discharge. Survival was associated with tamponade rather than exsanguination, shorter arrest and better physiology at the time of RT; ED thoracotomies had shorter arrests and higher survival than prehospital ones, and no patient survived after more than 10 minutes of traumatic cardiac arrest.

• 70.7% PERFORMED PREHOSPITAL (82/116)
• 62.1% vs 70.7% TAMPONADE SUSPECTED VS EXSANGUINATION FOUND AT SURGERY (82/116; THE SAME COUNT AS PREHOSPITAL RT, AS THE ABSTRACT REPORTS)
• 43.1% ROSC (50/116); CI NOT REPORTED
• 6.9% SURVIVAL TO DISCHARGE (8/116); CI NOT REPORTED

CRITICAL APPRAISAL Exposure and outcome measured reliably

Not reported Abstract gives no definitions of arrest duration, signs of life or how operative findings were recorded

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet beyond the Issue 9 action to audit your thoracotomy SOP; record time from loss of output and the operative finding for every case so local data can be set against TETRiS | Confidence in that: moderate Main concerns: 8 survivors make every subgroup comparison, including “no survivors beyond 10 minutes”, very imprecise (direction unknown; the abstract gives no denominator for arrests over 10 minutes); prehospital and ED patients differ in arrest duration and physiology, so the lower prehospital survival likely reflects case mix rather than setting; abstract only: no full-text check of definitions or missing data Strengths: The first prospective national UK dataset that captures prehospital thoracotomy, including patients who die on scene. Disagreement: TETRiS reports no survivors after more than 10 minutes of traumatic arrest, while the RCUK 2025 algorithm and the 2026 EMJ update use an elapsed-time threshold of under 15 minutes. With 8 survivors in total, the TETRiS finding cannot yet support a shorter cut-off.

UK practice: The RCUK adult traumatic cardiac arrest/peri-arrest algorithm (2025) in the RCUK special circumstances guidelines (2025) lists resuscitative thoracotomy after expertise, equipment, environment and elapsed time under 15 minutes; NICE NG39 (major trauma: assessment and initial management, 2016) makes no recommendation on thoracotomy (we searched the recommendations page). Issue 9 covered the 2026 EMJ update (Chesters et al, doi:10.1136/emermed-2026-216110), which gives the same 15-minute threshold and a governance checklist; TETRiS gives services a national benchmark for that audit.

PHEM Evidence Rundown, Issue 10, November 2026. Page 13 of 31

INFORMING PRACTICE

OBSERVATIONAL

DIPIMC/FIMC

Prehospital TXA by Fibrinolytic Phenotype (Pooled Data From Two Prehospital TXA Trials, n=1,249): Lower Hazard of Death With TXA in “Shutdown” (HR 0.32), a Post-Randomisation Subgroup; Other Phenotypes 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 (the BMA Library cannot supply this journal); no published commentary found (JournalFeed, SGEM, St Emlyn’s, REBEL EM, EMCrit and journal commentary searched 1 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 Study of Tranexamic Acid During Air and Ground Medical Prehospital Transport Trial), with 1,249 injured patients grouped by their initial thromboelastography (TEG) LY30, a measure of clot breakdown: fibrinolytic shutdown (below 0.9%), physiological (0.9–3%) and hyperfibrinolysis (above 3%). The primary outcome was 30-day mortality, which was 14.5% with shutdown, 6.7% with hyperfibrinolysis and 6.5% with physiological fibrinolysis. In a model adjusted for TXA, phenotype, age, mechanism, Injury Severity Score and shock index on hospital arrival, hyperfibrinolysis was linked to a higher hazard of death (HR 2.87, 95% CI 1.10 to 7.53; reference group not stated in the abstract), and patients in shutdown randomised to TXA had a lower hazard of death (HR 0.32, 95% CI 0.11 to 0.91). Group sizes, TXA-versus-placebo event rates within each phenotype, the timing of the TEG 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 “initial” TEG LY30 with fixed cut-offs; when the TEG was drawn relative to prehospital TXA is not reported; 30-day death is objective

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet; keep giving TXA as soon as possible to bleeding major trauma patients, because no crew can know the fibrinolytic phenotype at scene | Confidence in that: moderate Main concerns: stratifying on a post-randomisation TEG result breaks the randomised comparison, so the direction of bias is unpredictable; an unreported subgroup size with a wide CI, picked out of three phenotypes, 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 were prospective and randomised prehospital TXA, so dosing was protocolised and outcomes were collected prospectively

UK practice: NICE NG39 (major trauma: assessment and initial management, 2016) says to use intravenous tranexamic acid as soon as possible in patients with major trauma and active or suspected active bleeding (1.5.4), and not more than 3 hours after injury unless there is evidence of hyperfibrinolysis (1.5.5). NG39 does not mention fibrinolytic shutdown or ask for a viscoelastic test before giving TXA. This abstract gives no reason to withhold TXA at scene from a patient who meets the NG39 criteria.

PHEM Evidence Rundown, Issue 10, November 2026. Page 14 of 31

INFORMING PRACTICE

OBSERVATIONAL

DIPIMC/FIMC

Signs of Life in Traumatic Cardiac Arrest (Kanto Registry, n=462): 4.8% Survived Beyond the ED, Mostly Those With Electrical Activity on the First Monitor, and Only 1 Patient Had a Good Neurological Outcome

Yamashita I, Yamamoto R, Suzuki M, et al. Signs of life and clinical outcomes in patients with traumatic out-of-hospital cardiac arrest: a nationwide multicenter observational study. J Emerg Med. 2026;90:161–169. doi:10.1016/j.jemermed.2026.09.006. PMID: 42874196. Source route: PDF via BMA Library (open access, CC BY 4.0). Appraised from full text

Retrospective descriptive analysis of the SOS-KANTO 2017 prospective OHCA registry (42 hospitals in Tokyo and its suburbs, September 2019 to March 2021): 462 adults in traumatic cardiac arrest on ambulance arrival, grouped by three signs of life at the scene from ambulance records: pupillary light reflex, spontaneous respiration (agonal gasping) and cardiac electrical activity (any waveform other than asystole on the first monitor). The primary outcome was survival beyond the ED, which was 22 of 462 (4.8%) overall. It was 4 of 16 (25.0%) with electrical activity plus gasping, 10 of 117 (8.5%) with electrical activity alone and 8 of 318 (2.5%) with no signs of life; none of the 7 patients with only a pupil reflex or only gasping survived beyond the ED. One patient (0.2%), in the electrical-activity-only group, had a good neurological outcome (Cerebral Performance Category 1–2) at 30 days; there was no comparative or adjusted analysis.

• 10.2% vs 2.5% SURVIVAL BEYOND ED, ANY ELECTRICAL ACTIVITY VS NONE (14/137 VS 8/325, CALCULATED)
• 21.9% vs 6.5% ROSC, ANY ELECTRICAL ACTIVITY VS NONE (30/137 VS 21/325, CALCULATED)
• 2.5% SURVIVAL BEYOND ED WITH NO SIGNS OF LIFE (8/318; 95% CI 1.3 TO 4.9)
• 1 of 462 GOOD NEUROLOGICAL OUTCOME AT 30 DAYS (0.2%)

Source: Table 2 of the paper (CIs as printed). Rows 3 and 4 combine two small subgroups each. CRITICAL APPRAISAL Exposure measured reliably

Partly Signs taken from ambulance records at first contact; limb movement not recorded; missing values multiply imputed, with two sensitivity analyses giving similar results

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet; record the first rhythm, gasping and pupils at scene, and treat non-asystolic traumatic arrest as the group most likely to respond to reversal of causes | Confidence in that: moderate Main concerns: no injury data or adjustment, so electrical activity may simply mark less severe injury, overstating its value as an independent sign; the short-term primary outcome overstates the meaning of “survival” when only 1 of 462 patients survived neurologically well; ambulance crews recording signs during a resuscitation may miss gasping or pupil reactions, which would bias those signs towards the null Strengths: prospective Utstein-style registry data from 42 hospitals with predefined sign definitions

UK practice: The RCUK adult traumatic cardiac arrest/peri-arrest algorithm (2025) asks first whether traumatic cardiac arrest is likely, then prioritises reversible causes (haemorrhage control, airway and oxygenation, bilateral thoracostomies, relief of tamponade, pelvic splint, blood products) over chest compressions, and lists resuscitative thoracotomy after checking expertise, equipment, environment and an elapsed time under 15 minutes. If there is no ROSC it says to consider termination of resuscitation; it does not use individual signs of life. NICE NG39 (major trauma: assessment and initial management, 2016) makes no recommendation on traumatic cardiac arrest (we checked the recommendations page).

PHEM Evidence Rundown, Issue 10, November 2026. Page 15 of 31

INFORMING PRACTICE

OBSERVATIONAL

DIPIMC/FIMC

Traumatic Cardiac Arrest in Adults Aged 65 and Over (French Registry, n=5,595): More ROSC (22.5% vs 18.6%) but Fewer 30-Day Survivors (0.5% vs 1.2%) Than Younger Adults

Benhamed A, Morcel B, David JS, et al. Prehospital management and outcomes of traumatic cardiac arrest in older adults: a French nationwide registry study. Resusc Plus. 2026;32:101486. doi:10.1016/j.resplu.2026.101486. PMID: 42825123. Source route: PMC full text (PMC13629317). Appraised from full text

Retrospective cohort from the French national OHCA registry (RéAC): 5,595 adults with traumatic cardiac arrest resuscitated by a physician-led mobile medical team, comparing 1,346 aged 65 or over with 4,249 aged 18–64. Older patients more often fell (52.2% vs 28.5%) and had fewer thoracic injuries (43.8% vs 59.6%), and they received fewer thoracostomies (15.2% vs 23.9%), less fluid (28.3% vs 37.4%) and fewer intubations (87.5% vs 91.9%). For the primary outcome, ROSC was more frequent in older patients (303 of 1,346, 22.5%, vs 791 of 4,248, 18.6%; adjusted OR 1.22, 95% CI 1.03 to 1.45), but 30-day survival was lower (7 of 1,346, 0.5%, vs 51 of 4,249, 1.2%; adjusted OR 0.34, 95% CI 0.15 to 0.81). A good neurological outcome at 30 days was rare in both groups (0.4% vs 0.5%; OR 0.58, 95% CI 0.19 to 1.76).

• 22.5% vs 18.6% ROSC, 65 AND OVER VS 18–64 (ADJUSTED OR 1.22, 1.03 TO 1.45)
• 0.5% vs 1.2% 30-DAY SURVIVAL (7/1,346 VS 51/4,249; ADJUSTED OR 0.34, 0.15 TO 0.81)
• 0.4% vs 0.5% CPC 1–2 AT 30 DAYS (5/1,344 VS 23/4,236; OR 0.58, 0.19 TO 1.76)
• 15.2% vs 23.9% PREHOSPITAL THORACOSTOMY

CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Age and outcomes objective; injury sites from the scene doctor’s assessment without imaging or post-mortem confirmation; study years not given in the text

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet; age alone is not a reason to withhold resuscitation in traumatic arrest, but think of a medical cause in an older faller with minor injuries | Confidence in that: moderate Main concerns: medical arrests misclassified as traumatic likely overstate ROSC in older adults; no injury severity or frailty data, so the lower 30-day survival may reflect injury, reserve or treatment limitation; direction unknown; very few 30-day survivors make the adjusted odds ratio of 0.34 unstable Strengths: large national registry from a physician-staffed system with prespecified age bands and imputation sensitivity analyses

UK practice: The RCUK adult traumatic cardiac arrest/peri-arrest algorithm (2025) starts by asking whether traumatic cardiac arrest is likely and, if not, sends the team to the adult ALS algorithm, which fits the authors’ advice to consider a medical cause in older fallers. Where arrest is traumatic, it prioritises reversible causes over compressions and, after prehospital ROSC, immediate transport to an appropriate hospital. The RCUK Special circumstances guidelines (2025) add that ultrasound aids in identifying the cause of arrest. NICE NG39 (major trauma: assessment and initial management, 2016) makes no recommendation on traumatic cardiac arrest.

PHEM Evidence Rundown, Issue 10, November 2026. Page 16 of 31

INFORMING PRACTICE

OBSERVATIONAL

DIPIMC/FIMC

Prehospital Fluids in Severe Burns (Japan Trauma Data Bank, n=884): No Clear Link With 48-Hour Death After Weighting (OR 0.88, 95% CI 0.32 to 2.39), but Volume Was Not Recorded

Kanou S, Nakahara S, Chiba H, et al. Prehospital fluid resuscitation in patients with severe burns: a nationwide cohort study. Burns. 2026;52(9):108254 (journal pre-proof). doi:10.1016/j.burns.2026.108254. PMID: 42874523. Source route: PDF via BMA Library. Appraised from full text

Retrospective cohort from the Japan Trauma Data Bank (2019–2023): 884 patients with burns of at least 20% total body surface area (TBSA), or at least 10% under age 5, taken directly from scene, excluding those with a systolic pressure or heart rate of 0 at first ambulance contact and those transferred on. Of these, 94 (10.6%) had prehospital intravenous Ringer’s solution, recorded only as given or not, and 790 had none. The primary outcome, death within 48 hours of arrival, occurred in 22 of 94 (23.4%) vs 106 of 790 (13.4%) unadjusted, and in-hospital death in 42 (44.7%) vs 224 (28.4%). After propensity-score weighting and further adjustment, the odds ratios were 0.88 (95% CI 0.32 to 2.39) for 48-hour death and 1.30 (95% CI 0.75 to 2.25) for in-hospital death.

• 23.4% vs 13.4% 48-HOUR DEATH, FLUID VS NONE, UNADJUSTED (22/94 VS 106/790)
• OR 0.88 48-HOUR DEATH, WEIGHTED AND ADJUSTED (95% CI 0.32 TO 2.39)
• 44.7% vs 28.4% IN-HOSPITAL DEATH, UNADJUSTED (42/94 VS 224/790)
• OR 1.30 IN-HOSPITAL DEATH, WEIGHTED (95% CI 0.75 TO 2.25)

CRITICAL APPRAISAL Exposure measured reliably

No Fluid recorded as yes/no only, with no volume, rate or start time; TBSA estimated after hospital arrival and coded in bands

Design: Observational | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet; keep following the FPHC/BBA advice for burns over 20% TBSA, and hand over the volume given | Confidence in that: moderate Main concerns: confounding by indication (fluid given to older patients with larger flame burns) likely overstates harm or hides benefit; adjusting for arrival physiology, lactate, transfusion and surgery likely biases the estimate towards no effect; unstable weights in the fluid group (median 0.277, SD 1.302), although the text calls them “near 1.0”, make the weighted estimate fragile; direction unknown Strengths: nationwide registry with a stated primary outcome, multiple imputation, E-values and a stated minimum detectable effect Disagreement: The FPHC/BBA consensus advises starting prehospital fluid for burns over 20% TBSA, on level III evidence that delayed resuscitation is harmful. This registry found no association with 48-hour or in-hospital death, but it could not measure volume and could not exclude either a benefit or a harm.

UK practice: The FPHC and British Burn Association consensus: Management of Burns in Pre-Hospital Trauma Care (version 2, April 2020) recommends starting fluid resuscitation before hospital for adult and paediatric burns over 20% TBSA (grade D), guided by a threshold method that needs only TBSA, age and weight, and handing over the volume given. Its evidence for starting fluid early is level III. The FPHC consensus statements page lists the burns consensus as “currently under review” (checked 1 Nov 2026). NICE NG39 (major trauma: assessment and initial management, 2016) has no burns fluid recommendation.

PHEM Evidence Rundown, Issue 10, November 2026. Page 17 of 31

2 — GUIDELINES & UK UPDATES

INFORMING PRACTICE

GUIDELINE

UK

AACE Position Statement on Public-Access Trauma and Bleed Kits (October 2026): Direct Pressure First, Kits Where a Local Risk Assessment Supports Them, and Ambulance Services Do Not Govern Them

Association of Ambulance Chief Executives. AACE position statement: public-access trauma / bleed kits. AACE news, 15 October 2026. AACE position statement: public-access trauma / bleed kits (15 October 2026). No DOI or PMID. Source route: AACE website (whole statement read); the appended NHS England Major Trauma Clinical Reference Group consensus position (16 January 2026) also read. Appraised from full text

A one-page national position statement from the body representing UK NHS ambulance services, addressed to organisations considering public-access trauma kits (PAcT kits) or “Stop the Bleed”-style bleed kits. It states that direct pressure is the most important first aid for external bleeding, that a dressing or any clean cloth will do, and that locating or retrieving a bleed kit should not delay immediate life-saving action; 999 and 111 call handlers give bleeding-control instructions. Organisations may buy kits where a specific risk assessment of a location, workplace, venue or event supports them, with training for likely users, and AACE “strongly recommends” that equipment is kept unlocked and clearly identifiable. Responsibility for governance, training, maintenance and deployment of community bleed kits does not sit with NHS ambulance services, and AACE endorses no product. The appended NHS England Major Trauma Clinical Reference Group position (January 2026) recommends that NHS major trauma centres do not endorse, promote or encourage buying commercial “Stop the Bleed” kits, and advises education and training over kits.

CRITICAL APPRAISAL Developer and funding; conflicts managed

Partly AACE, the national ambulance chief executives’ body; funding and conflicts not stated; no product endorsed

Design: Guideline | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing for HEMS clinical practice; for ambulance services and community responder schemes, advise direct pressure first and treat kits as an adjunct whose governance sits with the owning organisation | Confidence in that: moderate Main concerns: consensus without an evidence review, so the weight placed on direct pressure over tourniquets for lay rescuers cannot be traced to data (direction unknown); the appended clinical reference group position is itself consensus and states that the evidence for public bleed kits is far less robust than for public access defibrillators, so neither document can say whether kits save lives (direction unknown) Strengths: A clear, single national ambulance-sector position that matches existing London and NICE advice.

UK practice: The Bystander bleed control: a London joint position statement (London Ambulance Service and partners, August 2024) says tourniquets can be used for limb bleeding not controlled by direct pressure, that dedicated bleed kits are not essential and should contain only standard gauze and an easily applied tourniquet, and that haemostatic dressings are not recommended for bystanders. NICE NG39 (major trauma: assessment and initial management, 2016) recommendation 1.5.2 (for clinicians) advises a tourniquet in major limb trauma if direct pressure has failed. The earlier AACE bleed kits and items of emergency equipment position statement (8 February 2024) also said bleed kits could be registered on the GoodSAM app.

PHEM Evidence Rundown, Issue 10, November 2026. Page 18 of 31

3 — PAEDIATRIC PHEM

INFORMING PRACTICE

OBSERVATIONAL

PAEDS

DIPIMC/FIMC

Unplanned Births Outside Hospital (French Registry, n=3,473): 62.7% of Newborns Whose Temperature Was Measured Were Hypothermic on Scene; Neonatal Death by Day 7 Was 1.1%

Duchanois C, Laurent C, Duarte K, et al; AIE Group Investigators. Unplanned out-of-hospital deliveries: Insights from 3473 births of a 13-year prospective cohort study in France. Acta Obstet Gynecol Scand. 2026. doi:10.1111/aogs.70357. PMID: 42886579. Source route: PMC full text (PMC13674680; open access, CC BY-NC-ND 4.0). Appraised from full text

Prospective registry from 79 French physician-staffed mobile intensive care units (Mobile ICUs), 2011–2024: 3,473 unplanned births outside a maternity unit, 30% with no medical assistance, 36% before the Mobile ICU arrived and 34% with the team on scene. Most mothers were multiparous (88%); there was no hospital-birth comparison group. On scene, 8.4% of mothers had a critical complication (postpartum haemorrhage 4.5%), oxytocin was given for placental delivery in 51.4% (519/1010), and of the 2,002 newborns with a recorded temperature, 62.7% were below 36.5°C. In multivariable models, complications were associated with birth before 32 weeks and with hypothermia on hospital admission, but not with home-to-hospital travel time, and birth before Mobile ICU arrival was associated with newborn hypothermia on scene (adjusted OR 2.02, 95% CI 1.57 to 2.61).

• 62.7% NEWBORNS HYPOTHERMIC ON SCENE (1256/2002 MEASURED)
• 56.9% STILL HYPOTHERMIC AT HOSPITAL ADMISSION (1141/2005)
• 1.1% NEONATAL DEATH BY DAY 7 (38/3456)
• 8.4% MATERNAL CRITICAL COMPLICATION ON SCENE (290/3472)

Source: Duchanois et al, Tables 4 and 5. CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Temperature recorded in 60.2% of newborns, route varied; haemorrhage judged clinically, with no blood-loss threshold

Design: Observational | Certainty: low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing new to protocol; at every birth before arrival, start thermal care at once, record the baby’s temperature and recheck it before handover | Confidence in that: moderate Main concerns: counting hypothermia as both a complication and a predictor overstates its link with complications; no temperature was recorded for about 40% of newborns (calculated), which biases the 62.7% in an unknown direction; neonatal severity captured only by Apgar, so hypothermia as a cause is likely overstated Strengths: Large prospective multicentre registry over 13 years with day-7 maternal and neonatal outcomes.

UK practice: RCUK Newborn Resuscitation and Support of Transition of Infants at Birth Guidelines 2025 (published 27 October 2025) says that, ideally, at least two double-crewed ambulances should be sent to every imminent birth and baby born before arrival, and that thermal care should start early and proactively whatever the room temperature, aiming for 36.5–37.5°C with regular temperature checks. NICE NG235 (intrapartum care, 2023, last updated June 2026) advises telling women that active management of the third stage, including a routine uterotonic, is associated with a lower risk of postpartum haemorrhage or transfusion (1.10.6).

PHEM Evidence Rundown, Issue 10, November 2026. Page 19 of 31

INFORMING PRACTICE

OBSERVATIONAL

PAEDS

DIPIMC/FIMC

Paediatric Out-of-Hospital Cardiac Arrest Across Australia and New Zealand (n=4,779): 10.0% of Children Resuscitated by EMS Survived to 30 Days or Discharge, Lowest in Infants and 1–5 Year Olds (Both 7.9%)

Saha A, Howell S, Nehme Z, et al. The epidemiology of paediatric out-of-hospital cardiac arrest in Australia and New Zealand: a report from the Australasian Resuscitation Outcomes Consortium Epistry (Aus-ROC). Resuscitation. 2026;229:111347. doi:10.1016/j.resuscitation.2026.111347. PMID: 42855035. Source route: PDF via BMA Library (journal pre-proof). Appraised from full text

Population-based registry study of every EMS-attended cardiac arrest in children under 18 across all ten Australian and New Zealand ambulance services, 2017–2023: 4,779 arrests (incidence 10.1 per 100,000 person-years), of whom EMS attempted resuscitation in 3,139 (65.7%). Most arrests were unwitnessed (66.9%), at home (74.3%) and in a non-shockable rhythm (91.1%); in arrests not witnessed by EMS, bystander CPR was given in 62.7% but a bystander shock in only 1.6%. Causes varied with age: drowning led in early childhood, while hanging and trauma dominated in adolescents. In EMS-treated children, survival to 30 days or hospital discharge was 10.0%, and after adjustment, bystander-witnessed arrest (adjusted OR 4.63, 95% CI 3.15 to 6.81), shockable rhythm (11.91, 7.67 to 18.50) and a shorter EMS response (0.96 per extra minute, 0.93 to 0.99) were linked with survival, while age group and bystander CPR were not.

• 10.1 EMS-ATTENDED ARRESTS PER 100,000 CHILD-YEARS
• 10.0% SURVIVAL TO 30 DAYS OR DISCHARGE, EMS-TREATED
• 7.9% vs 15.4% SAME OUTCOME, INFANTS VS 6–11 YEARS
• aOR 0.96 SURVIVAL PER EXTRA MINUTE OF EMS RESPONSE (95% CI 0.93 TO 0.99)

Source: Saha et al, Table 2 (aetiology and rhythm) and Figure 3 (survival). CRITICAL APPRAISAL Exposure and outcome measured reliably

Partly Utstein definitions with registry quality checks; aetiology is presumed by EMS; no neurological or functional outcome

Design: Observational | Certainty: low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet; expect hypoxic and traumatic causes in older children and keep the early-shock pathway (AED as soon as possible) in children of all ages | Confidence in that: moderate Main concerns: selective resuscitation of more salvageable adolescents likely overstates survival in older age groups; response time may stand in for remoteness and system factors, so the 4% per minute figure may overstate the effect of speed alone; the 30-day model used 2,544 of 3,139 treated children (complete cases, excluding EMS-witnessed arrests and two regions), with bias of unknown direction Strengths: Every ambulance service in two countries over seven years, reported by the 2024 Utstein age bands.

UK practice: RCUK Paediatric Life Support Guidelines 2025 (published 27 October 2025) tells ambulance call handlers to coach bystanders to give rescue breaths and chest compressions in children of all ages (30:2 with five initial breaths), and advises that a second rescuer should attach an AED as soon as possible. For traumatic arrest it asks professional rescuers to treat hypotension, hypoxia, tension pneumothorax and tamponade (HOTT), including bilateral thoracostomies and early blood products in haemorrhagic shock.

PHEM Evidence Rundown, Issue 10, November 2026. Page 20 of 31

INFORMING PRACTICE

SR/MA

PAEDS

DIPIMC/FIMC

Video vs Direct Laryngoscopy for Emergency Intubation in Children (ILCOR SR/MA, 18 Studies, 17 Observational): First-Pass Success RR 1.11 and Fewer Severe Events, Very Low Certainty; the Two Prehospital Studies Are Inconclusive

Couto TB, Acworth J, Rech L, et al. Video laryngoscopy versus direct laryngoscopy for pediatric emergency tracheal intubation: a systematic review and meta-analysis. Resusc Plus. 2026;32:101524. doi:10.1016/j.resplu.2026.101524. PMID: 42960018. PROSPERO CRD420251081694. Commissioned by the ILCOR Pediatric Life Support Task Force. Source route: PMC full text (PMC13717893); supplementary figures (setting subgroups) not available. Appraised from full text

Systematic review and random-effects meta-analysis of 18 studies (17 observational, 1 RCT) comparing video laryngoscopy (VL, most often C-MAC) with direct laryngoscopy (DL) for emergency intubation of infants and children (VL used as a rescue device excluded) in emergency departments, paediatric intensive care units and prehospital care; neonates at birth, NICU and theatre intubations were excluded. For first-pass success, a critical outcome, VL was associated with higher success (7,737 of 10,720, 72.2%, vs 9,651 of 13,994, 69.0%; RR 1.11, 95% CI 1.06 to 1.16) and with fewer severe tracheal intubation-associated events (7.7% vs 11.8%; RR 0.76, 95% CI 0.62 to 0.93), with overall success unchanged (RR 1.00, 95% CI 0.99 to 1.01); all outcomes were graded very low certainty. No study reported survival or children in cardiac arrest. The prehospital subgroup was 2 US studies of nurses and paramedics (135 and 55 children) and gave an imprecise estimate (RR 1.13, 95% CI 0.66 to 1.95; the text does not name the outcome).

Source: Couto et al, Table 4 and subgroup analysis text. Crude rates pool all arms; the pooled RR weights each study, which is why the authors’ table gives 76 more first-pass successes per 1,000 while the crude difference is 3.2 percentage points (calculated). No NNT: 17 of 18 studies are observational.

• RR 1.11 FIRST-PASS SUCCESS (95% CI 1.06 TO 1.16)
• RR 0.76 SEVERE EVENTS (95% CI 0.62 TO 0.93)
• RR 1.13 PREHOSPITAL SUBGROUP, 2 STUDIES (95% CI 0.66 TO 1.95)
• Very low GRADE CERTAINTY, ALL OUTCOMES

CRITICAL APPRAISAL Clear question; protocol registered

Yes PROSPERO CRD420251081694; ILCOR PICOST with cardiac arrest and peri-arrest children

Design: SR/MA | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet: use the laryngoscope your service trains with for paediatric RSI; the prehospital evidence neither supports nor argues against a switch to VL | Confidence in that: moderate Main concerns: confounding by indication and by era (VL used by more experienced operators in later years) likely overstates the VL benefit; registry overlap and self-reported adverse events may overstate precision and understate events; the one RCT pointed the other way for severe events (RR 1.24, 95% CI 0.51 to 3.01), so the safety signal is not confirmed by randomised data Strengths: ILCOR-commissioned, registered and GRADE-rated, with transparent sensitivity analyses for registry overlap.

UK practice: The RCUK paediatric life support guidelines (2025) advise video or direct laryngoscopy “depending on local protocols and rescuer experience” and not to use unfamiliar equipment, and to limit intubation attempts to 4. This review does not change that: it supports VL as a reasonable first-line device where crews are trained in it, but says nothing reliable about prehospital practice.

PHEM Evidence Rundown, Issue 10, November 2026. Page 21 of 31

INFORMING PRACTICE

FOAMED

PAEDS

UK

Trauma Risk Management (TRiM) After Distressing Paediatric Jobs (DFTB, October 2026): A Persuasive Case Built on Thin Evidence; the Only TRiM RCT (12 Warships, Few Incidents) Found No Change in Psychological Health

Hewison C. The Case for TRiM in Paediatric Pre-Hospital Care. Don’t Forget the Bubbles, 13 October 2026 (updated 25 October 2026). doi:10.31440/DFTB. 69065. dontforgetthebubbles.com/trim. No PMID (blog). Source route: full post read on the website; primary TRiM studies checked by us (Greenberg 2010, PMID: 20690168; Hunt, Occup Med 2013;63:549–55, PMID: 24174630; Billings 2023 full text, PMID: 38032045). Appraised from full text

A UK paramedic uses the case of a 7-year-old whose treatment was stopped after a prolonged seizure and respiratory arrest to argue that ambulance crews should be offered TRiM after distressing paediatric jobs. TRiM is a voluntary peer-support model from the UK armed forces: a trained peer, who need not be a clinician, assesses risk within 72 hours and again about four weeks later, and refers on if needed. The post quotes probable PTSD in 14% of first responders, notes that critical incident stress debriefing (CISD) has been criticised as possibly harmful, and calls TRiM “simple yet highly effective” and in line with NICE NG116. Its TRiM sources are two training providers’ websites and one evaluation in a paramedic journal; it does not cite the controlled studies. Of 7 TRiM studies in a 2023 systematic review, 2 found fewer PTSD symptoms (a highly rated non-randomised controlled trial in Royal Marines and Coldstream Guards, Frappell-Cooke 2010, and a cross-sectional survey), and 6 came from the TRiM originator’s group.

• 12 ships TRIM CLUSTER RCT (GREENBERG 2010): NO CHANGE IN PSYCHOLOGICAL HEALTH OR STIGMA AT 12–18 MONTHS; FEW CRITICAL INCIDENTS; ORGANISATIONAL FUNCTIONING MODESTLY BETTER
• 717 staff CUMBRIA POLICE AFTER A 2010 MULTIPLE-FATALITY INCIDENT (SERVICE EVALUATION): LESS SICKNESS ABSENCE, MAINLY IN JUNIOR RANKS; BILLINGS 2023 READ TRIM ALONE AS NOT SIGNIFICANT
• 7 TRiM studies WORKPLACE SYSTEMATIC REVIEW (BILLINGS 2023; 80 STUDIES, 7 ON TRIM): NO HARM, NO CONCLUSIVE BENEFIT

CRITICAL APPRAISAL Who wrote it and for whom

Yes A UK ambulance paramedic with a paediatric interest, writing for clinicians on a paediatric FOAMed site

Design: FOAMed | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing clinical; know how to reach your service’s peer-support or TRiM route after a distressing paediatric job, and keep operational hot debriefs separate from psychological debriefing | Confidence in that: moderate Main concerns: citing training-provider websites rather than controlled studies overstates TRiM’s effectiveness; the NWAS outcome quoted (more referrals) measures process, not mental health Strengths: A clear, practical account of how TRiM runs and why paediatric jobs weigh on crews. Disagreement: The post says TRiM works “in line with NICE guidance”; NG116 does not mention TRiM, and its peer-support recommendation is about groups for people with PTSD led by people with mental health training, whereas TRiM practitioners need not be clinicians. TRiM’s two-step check fits NG116’s active monitoring in shape, but that fit is the author’s inference, not NICE’s.

UK practice: NICE NG116 (post-traumatic stress disorder, 2018) recommends considering active monitoring for people with subthreshold PTSD symptoms within 1 month of a traumatic event, with follow-up within 1 month (1.6.4). It says do not offer psychologically focused debriefing for the prevention or treatment of PTSD (1.6.5). Its peer-support recommendation is for peer support groups that are facilitated by people with mental health training and supervision (1.4.1).

PHEM Evidence Rundown, Issue 10, November 2026. Page 22 of 31

4 — FOAMED & CRITICAL APPRAISAL

INFORMING PRACTICE

FOAMED

UK

DIPIMC/FIMC

Catastrophic Midface Bleeding (St Emlyn’s TTL Tips 21): Secure the Airway, Support the Facial Skeleton, Then Inflate Bilateral Nasal Balloons

Carley S. TTL tips 21: Catastrophic Bleeding from the midface. St Emlyn’s blog, 10 October 2026. stemlynsblog.org. No DOI or PMID. Source route: St Emlyn’s website (whole post read). Appraised from full text

A short trauma team leader tip from a UK emergency medicine and HEMS consultant on the “stacking” technique for catastrophic bleeding from midface fractures, described as a rescue technique for an intubated patient that needs experienced clinicians and the right kit. The sequence is: secure the airway early with an experienced anaesthetist and good suction; place bilateral nasal balloon catheters along the floor of the nose without inflating them; reduce the midface and place dental bite blocks between the molars either side of the tube; support the mandible with a collar or, failing that, a bandage under the chin and over the head; then inflate the posterior and then the anterior balloons on both sides. The post stresses that stacking is temporary, that maxillofacial and anaesthetic colleagues should be brought in early, and that teams should know where bite blocks are kept before they are needed.

CRITICAL APPRAISAL Who wrote it and for whom

Yes Simon Carley, EM consultant and HEMS consultant (North West Air Ambulance), for trauma team leaders

Design: FOAMed | Certainty: very low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing yet at scene beyond early definitive airway control; pre-alert the receiving trauma centre so anaesthetic and maxillofacial help, nasal balloon catheters and bite blocks are ready | Confidence in that: moderate Main concerns: no evidence is cited, so the benefit of the full sequence over simpler packing is assumed rather than shown (direction unknown); skull-base injury is flagged but the post gives no rule for when nasal devices should be avoided, which could understate the risk of misplacement Strengths: A clear, ordered sequence with a practical kit-readiness check.

UK practice: NICE NG39 (major trauma: assessment and initial management, 2016) contains no recommendation on maxillofacial or midface haemorrhage (we searched the recommendations page), so management follows local trauma network and maxillofacial protocols. We did not find a JRCALC or RCUK statement on this technique that we could read.

PHEM Evidence Rundown, Issue 10, November 2026. Page 23 of 31

INFORMING PRACTICE

REVIEW

DIPIMC/FIMC

Chemical Warfare Agents (BJA Education Review): Decontaminate First, Give Atropine and Pralidoxime Early for Nerve Agents, and Hydroxocobalamin for Suspected Cyanide

Malins WLE, Tong JL. Update on chemical warfare agents. BJA Educ. 2026;26(8):326–333. doi:10.1016/j.bjae.2026.05.001. PMID: 42453801. Source route: PDF via BJA Education. Appraised from full text

A continuing-education review for anaesthetists and intensivists of the four categories of chemical warfare agent (nerve, blistering, choking and blood agents), drawing on UK experience of the 2018 Salisbury novichok incident and on the 1995 Tokyo sarin attack, after which 38.7% of ICU staff at the main receiving hospital had symptoms. For nerve agents it stresses decontamination with personal protective equipment (the review’s minimum for healthcare staff is a gown, visor and two pairs of nitrile gloves; at scene, JESIP and ambulance HART arrangements apply), atropine titrated to dry secretions and to a heart rate target of 80 beats per minute, pralidoxime given with atropine as early as possible before the enzyme bond “ages”, and rocuronium rather than suxamethonium for RSI. Vesicants (mustard, Lewisite) need clothing cut off and early decontamination, burns-style care with cautious fluids, and airway monitoring; choking agents (chlorine, phosgene) have no antidote and are managed as ARDS. For cyanide it recommends empiric hydroxocobalamin, with sodium thiosulphate as an alternative or adjunct.

Agent class Examples Key specific treatment (per the review)

Source: Malins and Tong, BJA Education 2026. Doses are not reproduced here except where checked against a UK SmPC (below). CRITICAL APPRAISAL Who wrote it and for whom

Yes UK anaesthesia and ICM registrar and a US associate professor and director of trauma anaesthesia; CPD review for anaesthetists and intensivists

Design: Review | Certainty: low (GRADE-style, our judgement) | What it changes for UK prehospital care: Nothing new; use it to refresh CBRN knowledge alongside your service’s IOR training and antidote holdings | Confidence in that: moderate Main concerns: narrative review without a systematic search, so selection of sources may favour older military guidance (direction unknown); little on scene triage and decontamination, the decisions prehospital teams actually make, which limits direct applicability Strengths: Concise, UK-relevant synthesis that ties each agent class to its antidote and to anaesthetic implications.

UK practice: At scene, the JESIP Initial Operational Response to incidents suspected to involve hazardous substances or CBRN materials (version 1.1, June 2024) sets “Remove, Remove, Remove”: move people away from the hazard, remove outer clothing (avoid pulling it over the head) and remove the substance, with dry decontamination as the default and wet decontamination if a caustic substance, biological or radiological agent is suspected. The Cyanokit 5 g SmPC (eMC, last updated March 2026) gives an adult initial dose of 5 g IV over 15 minutes (70 mg/kg, maximum 5 g, in children), with a second dose possible up to a total of 10 g.

PHEM Evidence Rundown, Issue 10, November 2026. Page 24 of 31

5 — QUICK HITS

CARDIAC ARREST AND RESUSCITATION CABARET head-up CPR bundle (UK feasibility RCT, SCAS and Hampshire and Isle of Wight Air Ambulance, n=32 randomised, 30 analysed): head-up positioning, LUCAS-3 and an impedance threshold device were delivered on scene in 14 of 17 bundle patients, and all 32 planned patients were recruited, but only 9.5% of air ambulance attendances led to recruitment. Survival was 3 of 17 vs 2 of 13 with standard ALS, but per protocol only 1 of 14 bundle recipients survived; not powered for outcomes, and a definitive trial is needed before any UK service adopts the bundle. Plumb JOM, et al. Resusc Plus 2026;29:101328. PMID: 42125736 Time to ROSC and long-term recovery (Danish registry cohort, n=2,583 bystander-witnessed OHCA surviving 30 days): anoxic brain damage rose from 0.3% (ROSC within 10 minutes of the call) to 3.2% (over 18 minutes; adjusted HR 9.98, 95% CI 2.93 to 33.96), but 5-year survival (79%, 95% CI 77 to 81%) and return to work (about 60%) did not differ by time to ROSC; only 30-day survivors were studied, so this says nothing about when to stop. Oxwall A, et al. Resuscitation 2026. PMID: 42838497 Emergency physician training and OHCA (German single-region cohort, n=1,426 non-traumatic arrests): survival to discharge was 12.2% with residents, 14.4% with specialists without additional critical care training and 18.0% with specialists holding additional critical care training (adjusted OR 1.59, 95% CI 1.05 to 2.40, vs residents), with faster vascular access; observational, with no adjustment for unmeasured differences in which physician attended. Kreutz J, et al. Resuscitation 2026. PMID: 42600766 Family-member bystander CPR (US CARES registry, n=824,960 OHCA): survival to discharge was 7.3% without bystander CPR, 9.6% with CPR by a family member (adjusted OR 1.31, 95% CI 1.29 to 1.34) and 17.3% with CPR by a non-family member (adjusted OR 1.58, 95% CI 1.55 to 1.62); 95.7% of family CPR was at home, and CPR quality and dispatcher assistance were not captured. Tatavarthy M, et al. Resusc Plus 2026. PMID: 42927085 Community first responders in the North East (abstract only; NEAS retrospective cohort, n=7,916 OHCA, 177 with CFR attendance): public-access defibrillation was more frequent when CFRs attended (20.3% vs 12.2%), but survival to discharge was not (9.6% vs 10.2%; adjusted OR 1.15, 95% CI 0.51 to 2.61); CFRs attended only 2.2% of arrests; CFR-attended arrests were more often rural, with longer ambulance responses, so the study is underpowered. Finney O. Br Paramed J 2026. PMID: 42683050 Lights and sirens despite a DNAR order (Danish one-year observational study, Region of Southern Denmark, 1,695 dispatches to suspected OHCA, 1,348 analysed): 266 patients (15.7% of all dispatches) had a physician-issued DNAR order in the hospital record that dispatchers could not see, 47.4% of these at the patient’s request; one-day mortality was 90.6% vs 69.5% without a DNAR order, and the number of units sent did not differ. Single region, and the paper does not report how many were actually resuscitated; it argues for DNAR orders that ambulance services can see. Schmidt PH, et al. Resuscitation 2026. PMID: 42508598 Opioid-associated OHCA (British Columbia registry cohort, 2015–2020, n=10,786 adult EMS-treated arrests): 13% were opioid-associated; these patients were younger (median 40 vs 70 years), more often unwitnessed (83% vs 55%) and rarely shockable (2.8% vs 23%), yet favourable neurological outcome at discharge was similar (7.2% vs 8.3%; adjusted OR 1.02, 95% CI 0.77 to 1.35). Sex, initial rhythm and witnessed status related to outcome differently in opioid cases, so the usual prognostic cues may mislead here; classification relied partly on post-mortem toxicology. Mok V, et al. Resuscitation 2026. PMID: 42855034 CPR quality across the paramedic-to-ED handover (abstract only; retrospective defibrillator-data study, one Toronto academic ED, n=35 adult OHCA): chest compression fraction (89–92%) and rate (109.9–117.5/min) stayed within guideline targets through handover, but compression depth rose from 5.8–5.9 cm with paramedics to 6.4–6.8 cm with the ED team, beyond the upper limit. A small single-centre sample; the RCUK 2025 adult BLS guideline also sets depth at 5–6 cm, so watch for over-deep compressions after handover. Myers V, et al. Prehosp Emerg Care 2026. PMID: 42725800 Pulse-check duration and capnography (abstract only; quasi-randomised multicentre study, Türkiye, n=252): pulse checks of 5 seconds or less were associated with more ROSC than checks of more than 5 to 10 seconds (36.3% vs 19.5%; OR 2.39, 95% CI 1.35 to 4.24); the setting (in or out of hospital) and how groups were allocated are not clear from the abstract. Yılmaz G, et al. Resuscitation 2026. PMID: 42901369 Real-time BVM ventilation feedback (abstract only; before-and-after, Paris Fire Brigade BLS teams, n=163 OHCA, 4,300 ventilations): insufflated volumes of 400–600 mL rose from 42% to 68% of breaths with feedback, but measured tidal volume stayed low (median 243 vs 235 mL and in range in 15% vs 14%, with vs without feedback) and leak fell by 12% (mean ratio 0.88, 95% CI 0.77 to 0.99); no patient outcomes. Jost D, et al. Resuscitation 2026. PMID: 42901366 Real-time ventilation feedback in OHCA (abstract only; before-after study, Milwaukee and Kenosha, n=278): with the Real BVM Help feedback device active, the mean share of ventilations on target for both rate and tidal volume rose from 7.5% to 39.0% (p<0.0001); with ETCO2 monitoring alone, 58.2% of 22,037 ventilations were below target volume. Process outcome only, no survival data, and a non-randomised before-after design funded in part by ZOLL. Aufderheide TP, et al. Resuscitation 2026. PMID: 42956924

PHEM Evidence Rundown, Issue 10, November 2026. Page 25 of 31

Ventilation feedback during CPR (EOlife X training device; porcine method-comparison study, 11 pigs, 821 ventilations): the device agreed with reference flow and pressure sensors within prespecified limits (±50 mL, ±3/min) during normal ventilation, but with ongoing chest compressions 37.1% of inspiratory volumes, 86.2% of expiratory volumes and 60.4% of respiratory rates fell outside them, apparently because compression-induced airflow distorted the measurement. Intubated, mechanically ventilated pigs only; airflow-derived volumes shown during CPR should not be taken at face value until validated under compressions. Orlob S, et al. Resuscitation 2026. PMID: 42492618

ERC Guidelines 2025 corrigenda (four published errata, Resuscitation 2026): First Aid now says naloxone is “probably not effective” once cardiac arrest has occurred and should be tested in a large trial (previously “not effective”); Post-resuscitation Care corrects the PaCO2 target in Figure 3 to 4.7–6 kPa, as the RCUK 2025 guideline already states; Adult ALS corrects the ibutilide dose to 0.01 mg/kg IV over 10 minutes under 60 kg or 1 mg IV over 10 minutes at 60 kg or more (previously printed as 0.6 mg/kg), adds an omitted reference on CPR-induced consciousness and completes an incomplete reference. Special Circumstances replaces Figure 5 (accidental hypothermia algorithm) and Figure 14 (buried avalanche victim algorithm) without saying what was wrong. Djärv T, et al. PMID: 42766975; Lott C, et al. PMID: 42766976; Nolan JP, et al. PMID: 42766977; Soar J, et al. PMID: 42863166. Resuscitation 2026 (corrigenda to ERC Guidelines 2025: PMID: 41117568, PMID: 41117569, PMID: 41117575, PMID: 41117572)

TRAUMA, HAEMORRHAGE AND BURNS Shock index, RTS and ISS for trauma death (diagnostic meta-analysis, 12 studies, 234,129 patients, mainly hospital cohorts): shock index had a pooled sensitivity of 0.679 (95% CI 0.544 to 0.789) and specificity of 0.763 (95% CI 0.446 to 0.928), LR+ 2.86 and LR− 0.42, and the Revised Trauma Score did not differ from ISS in six paired studies; for all three scores, at a 20% prior a positive result raised the probability of death to only 42–50%. One registry of older patients supplied 217,190 of the patients and thresholds varied (shock index 0.86 to 1.08), so none of these scores should decide triage on its own. Raza AA, et al. Injury 2026. PMID: 42854533 Prehospital pulse pressure and active haemorrhage (abstract only; retrospective cohort, 3 Alberta level 1 centres, n=2,177): 68 patients (3%) had active haemorrhage, whose median prehospital pulse pressure was 40 vs 53 mmHg; pulse pressure discriminated about as well as systolic pressure (AUC 0.73, 95% CI 0.67 to 0.79, vs 0.74, 95% CI 0.67 to 0.81), and each 10 mmHg narrowing gave an odds ratio of 1.51 (95% CI 1.28 to 1.79). The authors propose a cut-off of 45 mmHg or less, and describe its negative predictive value as high, but the abstract gives no sensitivity, specificity or predictive values for it. Czarnecka Z, et al. Injury 2026. PMID: 42918714 Severely injured patients in routine mass casualty incidents (TraumaRegister DGU, 430 MCI patients among 65,152, with a 1:4 matched cohort for outcomes): in the full cohort, median prehospital time was 85 vs 67 minutes, intubation (33.6% vs 20.9%) and TXA (32.1% vs 15.0%) were more frequent, and HEMS carried 34.5% vs 20.7% of patients, while matched in-hospital mortality did not differ (13.5% vs 11.7%; adjusted OR 1.19, 95% CI 0.77 to 1.85, in a multivariable sensitivity analysis). Scene deaths are not in the registry and patients from the same incident could not be linked. Breuer F, et al. BMC Emerg Med 2026. PMID: 42907761 Helicopter vs ground physician-staffed EMS in Germany (abstract only; TraumaRegister DGU, n=129,783, 25.0% by HEMS): in indication-based subgroups HEMS patients more often had advanced airway management with impaired consciousness (96.5% vs 82.9%), prehospital chest drains (14.0% vs 5.5%), pelvic stabilisation (56.7% vs 44.0%) and TXA for suspected major bleeding (48.7% vs 22.8%). Descriptive analysis only, in a system where both arms are doctor-led; the authors note that higher rates do not prove better care. Faul P, et al. Eur J Trauma Emerg Surg 2026. PMID: 42905143 Pelvic binders and mortality (abstract only; systematic review and meta-analysis, 16 studies, 6,591 patients): in the 5 studies that adjusted for confounding, binder use was not linked to lower mortality (OR 0.90, 95% CI 0.73 to 1.10), nor in unadjusted studies (OR 1.31, 95% CI 0.86 to 1.99). The abstract does not describe the included study designs, and confounding by indication is likely in non-randomised data; the authors argue mortality may be an insufficient endpoint for a temporising device, and NICE NG39 still advises a binder for suspected pelvic bleeding after blunt high-energy trauma. Forssten MP, et al. Am Surg 2026. PMID: 42803114 Pelvic binder mechanics (abstract only; retrospective imaging review, single tertiary centre, n=298 CT-confirmed pelvic fractures): four orthopaedic consultants judged a binder mechanically favourable for only 17% (95% CI 13% to 22%) of fracture patterns, neutral for 28% and potentially unfavourable for 56%, and no prehospital variable discriminated well (best was ISS, AUC 0.65, 95% CI 0.58 to 0.73). Read with the Forssten meta-analysis above (no mortality difference with binders): this imaging study has no outcome data, and NICE NG39 still advises a binder for suspected active pelvic bleeding after blunt high-energy trauma. Forssten MP, et al. Am Surg 2026. PMID: 42856048 Converting prehospital tourniquets in the ED (abstract only; CONVERT, single level I trauma centre, n=647): 83.9% (543 of 647) were converted to packing, haemostatic dressings or pressure bandages without a procedure or operation, at a median tourniquet time of 40 minutes, and no complication was attributed to the tourniquet. Each extra 30 minutes cut the odds of successful conversion (OR 0.41, 95% CI 0.32 to 0.53), a reason to record the application time at scene and hand it over. Fox AL, et al. J Trauma Acute Care Surg 2026. PMID: 42371668

PHEM Evidence Rundown, Issue 10, November 2026. Page 26 of 31

Calcium on arrival after scene care (abstract only; pre-planned secondary analysis of a prospective cohort, three trauma centres, n=1,047 highest-level activations arriving from scene): 235 were hypocalcaemic and 39 hypercalcaemic on arrival; of 136 given prehospital blood, 77 (56.6%) were hypocalcaemic, but 116 of 718 (16.1%) who received no fluid or blood were too, and 5 of the 39 hypercalcaemic patients had been given prehospital calcium. The authors conclude prehospital assessment cannot predict calcium derangement in either direction; thresholds are not given in the abstract. Schauer SG, et al. Am J Emerg Med 2026. PMID: 42945810

Prehospital analgesia in adult trauma (abstract only; systematic review, 17 studies, 17,115 patients, 2014–2026): 6 RCTs and 11 non-randomised studies of opioids, ketamine, methoxyflurane and paracetamol; apart from one opioid study, no serious drug-attributed adverse events were reported, while minor effects such as nausea and dizziness were common. The abstract reports no pooled estimate, and protocols varied widely. Alkoby-Meshulam L, et al. Emerg Med Australas 2026. PMID: 42947661

Burns dispatches for a Dutch physician-staffed HEMS (Rotterdam, retrospective cohort, 976 dispatches, 2015–2023): 455 of 976 dispatches (46.6%) were cancelled, mostly because the ambulance crew found minor injuries, and HEMS-specific interventions (mainly drugs and advanced airway) were given in 175 of 521 completed missions (33.6%), less often in children than adults (14.2% vs 43.5%) and in scalds than flame burns (14.6% vs 45.6%). Descriptive single-service data; useful for burns dispatch criteria, especially for scalded children. Vianen NJ, et al. Burns 2026. PMID: 42874521

AIRWAY, MEDICAL AND TRANSFER GCS 8 as an intubation trigger (abstract only; two-era Japan Trauma Data Bank cohort, n=71,988 and 66,417): in trauma patients arriving without a prehospital advanced airway, recorded intubation was 9.1 and 10.6 percentage points more frequent at arrival GCS 8 than at GCS 9 (pooled adjusted OR 1.61, 95% CI 1.36 to 1.91), but model comparison favoured a graded GCS–outcome relationship over a unique step at 8/9, and differences in death and cranial surgery were imprecise. It uses hospital-arrival GCS and recorded intubation, so it cannot show whether keeping or dropping the rule changes outcomes. Senda A. Scand J Trauma Resusc Emerg Med 2026. doi:10.1186/s13049-026-01715-0. PMID not yet assigned Checklists for time-critical interhospital transfers (abstract only; service evaluation in one UK HEMS organisation): after pre-arrival and pre-departure checklists with simulation training, the referring-hospital turnaround target of under 45 minutes was met in 7 of 11 transfers in 2024 vs 12 of 23 in 2022 (64% vs 52%; not significant on Fisher’s exact test). Tiny numbers and a before-after design; the authors note outlier delays affected performance. Sheldrake I, et al. Air Med J 2026. PMID: 42946505 PAEDIATRICS AND BIRTHS First PaCO2 after ROSC in children (abstract only; Japanese multicentre OHCA registry cohort, 2014–2022, n=215 under 18 with a blood gas within 60 minutes): most were hypercapnic (149 above 6.0 kPa [45 mmHg], 39 at 4.7–6.0 kPa [35–45 mmHg], 27 below 4.7 kPa [35 mmHg]), but neither hypocapnia (adjusted OR 0.83, 95% CI 0.13 to 5.14) nor hypercapnia (adjusted OR 1.46, 95% CI 0.25 to 8.68) was clearly associated with favourable neurological outcome at 30 days. The intervals are so wide that the study cannot rule a meaningful effect in or out, so it neither supports nor undermines aiming for normocapnia after ROSC. Kikuchi N, et al. Resuscitation 2026. PMID: 42727701 Paediatric air medical intubation (abstract only; US multisite cohort, 1,040 encounters, 39 sites, 2021–2025): first-pass success was 86.5% overall but 69.1% (47/68) in infants under 1 year vs 88.7% (535/603) in adolescents, and post-attempt hypoxia, hypotension or bradycardia occurred in 53.6% vs 31.3%; video laryngoscopy was documented in only 8.6%. CIs not reported; registry documentation, so infant failures may be undercounted. Cloyd J, et al. Prehosp Emerg Care 2026. PMID: 42752295 Ventilation feedback for paediatric bag-mask ventilation (abstract only; randomised crossover simulation, 123 US EMS clinicians, toddler manikin): with a real-time feedback device, guideline tidal volumes were reached in 87% of breaths vs 16.2% without it, and guideline rate in 88.6% vs 65%. CIs not reported; a 2-minute manikin scenario, not patients. Finney JD, et al. Pediatrics 2026. PMID: 42896864 PREP score for 999 calls about breathing problems in children aged 0–4 (abstract only; Welsh Ambulance Service service evaluation with development and internal validation, 5,935 incidents): area under the curve 0.73; at the chosen threshold, sensitivity 53%, specificity 81% and negative predictive value 93%. The outcome was a pre-alert, a proxy for urgency rather than a patient outcome, and there is no external validation yet. Nicholls A, et al. Br Paramed J 2026. PMID: 42683303 Prehospital transfusion in injured children (abstract only; two US database studies): in the National Trauma Data Bank (79,764 children aged 1–15), five EMS-observable features identified a high-risk tier with sensitivity 36.8% and specificity 97.8% for transfusion within 4 hours; in national EMS data, 287 of 363,844 children with a positive shock index (0.1%) were transfused before hospital. Shock index does not confirm haemorrhage, so the second figure overstates the unmet need. Su J, et al. Prehosp Emerg Care 2026. PMID: 42886574; Hayes JM, et al. Prehosp Emerg Care 2026. PMID: 42836701 Births attended by ambulance crews in Southern Denmark (cross-sectional record review, 1,137,222 ambulance runs, 2016–2024; full text): 320 births or imminent births needed crew involvement; postpartum haemorrhage was recorded in 19.7% (63/320) of mothers and 4 newborns (1.3%) needed resuscitation. Crews see these cases rarely, so the authors call for routine training. Pedersen LL, et al. Acta Anaesthesiol Scand 2026. PMID: 42831496

PHEM Evidence Rundown, Issue 10, November 2026. Page 27 of 31

CORE REVISION — PHEM ISSUE 10 — NOVEMBER 2026

Core Revision — Paediatric Resuscitation in 2025 Guidelines: What Changed for Prehospital Teams

Recognition, the PLS sequence, airway, IO, drugs and shocks, reversible causes, the newborn born outside hospital, post-ROSC targets — DipIMC/FIMC

Exam goal: Run a paediatric cardiac arrest and an out-of-hospital newborn resuscitation to the Resuscitation Council UK (RCUK) 2025 guidelines, with weight-based doses, and know where RCUK and the European Resuscitation Council (ERC) 2025 guidelines differ. Where they differ, follow RCUK and your service’s procedures.

1. WHY CHILDREN ARREST AND HOW TO SPOT IT

2. THE SEQUENCE FOR PREHOSPITAL TEAMS

3. AIRWAY, ACCESS, DRUGS AND SHOCKS

PHEM Evidence Rundown, Issue 10, November 2026. Page 28 of 31

Memory aid — “4 Hs and 4 Ts” (RCUK and Stretch): Hypoxia (commonest: 100% oxygen, check chest rise); Hypovolaemia (10 mL kg−1 crystalloid boluses, blood for major haemorrhage); Hyper/hypokalaemia and metabolic, including hypoglycaemia (glucose 0.2 g kg−1, 2 mL kg−1 of 10%); Hypo/hyperthermia (below 30 °C: one dose of adrenaline, a maximum of three shocks, no amiodarone); Thrombosis; Tension pneumothorax; Tamponade; Toxins. Do not give calcium in hyperkalaemic arrest (Stretch: calcium and bicarbonate are no longer recommended for it). For sudden deterioration in a ventilated child, RCUK uses DOPES: Displacement, Obstruction, Pneumothorax, Equipment, Stomach/stacking/sedation.

4. THE NEWBORN BORN OUTSIDE HOSPITAL: RCUK NLS 2025

5. AFTER ROSC: TARGETS

DipIMC/FIMC focus: give 5 rescue breaths, then 15:2; know the ventilation rates with an advanced airway; adrenaline 10 mcg kg−1 (max 1 mg) every 3–5 min; 4 J kg−1 shocks, amiodarone 5 mg kg−1 after shocks 3 and 5; go early to IO; recite the 4 Hs and 4 Ts; for the newborn, dry and wrap (or bag if under 32 weeks), 5 inflation breaths at 30 cmH2O in air (32 weeks or more), 3:1 if the heart rate stays below 60 min−1, adrenaline 20 mcg kg−1. Sources: RCUK Paediatric Life Support and Newborn (NLS) guidelines 2025 with PALS, PBLS and NLS (March 2026) algorithms, resus.org.uk; Stretch B et al, Update on paediatric resuscitation, BJA Education 2026 (pre-proof), doi:10.1016/j.bjae.2026.08.002, PMID not yet assigned; Markel and Habicht, Notfall Rettungsmed 2025 (ERC 2025 PLS commentary); adrenaline and amiodarone SmPCs (eMC).

PHEM Evidence Rundown, Issue 10, November 2026. Page 29 of 31

6 — ACTION POINTS

PHEM Evidence Rundown, Issue 10, November 2026. Page 30 of 31

TRIALS TO WATCH

RECRUITING OR RUNNING

RESULTS AWAITED

ON-SCENE Dutch HEMS teams starting ECPR at the scene in adults aged 18 to 50 with refractory OHCA (NCT04620070): ClinicalTrials.gov lists it as completed (221 enrolled, primary completion July 2025; primary outcome survival to hospital discharge), but no results were in PubMed on 1 November 2026.

RECENTLY REPORTED

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

PHEM Evidence Rundown, Issue 10, November 2026. Page 31 of 31