EM Evidence Rundown — Issue 11

EMERGENCY MEDICINE · UK EDITION

EM Evidence Rundown

Week of 7 May 2026 | Issue #11

Jake Turner

Curated with the assistance of AI (Perplexity). All content editorially reviewed.

Lead: FASTEST phase 3 (Lancet, n=626) — rFVIIa reduces haematoma expansion but does NOT improve mRS at 180 days (aOR 1.09, p=0.61) and triples thromboembolic events (RR 3.41). Trial stopped for futility. No role in spontaneous ICH. CHANGE tonight: rFVIIa in ICH (do not use) • Midazolam buccal/IM for paediatric SE (best drug, best route) • Single-shot regional for rib fractures • Penicillin allergy direct challenge 97% safe • Paediatric concussion: document fog, nausea, light/noise sensitivity

BOTTOM LINE UP FRONT — ISSUE #11

ACT ON THIS NOW

TONIGHT rFVIIa in ICH: Do NOT use. FASTEST phase 3 (Lancet, n=626): no functional benefit (aOR 1.09, p=0.61), thromboembolic harm (RR 3.41). Stopped for futility. Not for spontaneous ICH.

TONIGHT Paediatric seizures: Midazolam buccal or IM is first-line (NMA 9 RCTs, n=1,135). Superior to diazepam on success (RR 1.13), failure (RR 0.74), recurrence (RR 0.51). Use buccal or IM when IV access unavailable.

TONIGHT Rib fracture pain: Single-shot regional block (erector spinae, serratus anterior, or paravertebral) reduces pain at 4–8h and opioid requirements vs standard care alone. NMA of 738 patients. Offer as adjunct.

TONIGHT Penicillin allergy: Direct oral challenge is safe — 97% tolerance rate, anaphylaxis <1% across 6,000+ patients (iNAAN study). Label "penicillin allergy" should trigger direct challenge pathway, not avoidance.

TONIGHT Paediatric concussion: Mental fog, noise/light sensitivity, nausea, and ocular abnormalities raise the likelihood. Absence of headache lowers it. Document these 4 features at assessment.

THIS MONTH ITP critical bleeding: IVIG 1 g/kg (repeat if needed) • steroids • platelet transfusion for haemostatic failure • anti-D in Rh+ non-splenectomised patients. McMaster guideline — strong recommendations, weak evidence base.

UK Corridor care data: Trust-level figures go public May 2026. Document every escalation-area episode now using the formal NHS England definition.

KNOW FOR NEXT TIME

GUIDELINE WATCH Novel hsTnT assay: A new hs-cTnT assay outperforms current hsTnT at 0h and 1h for MI rule-out (JAMA Cardiol). Watch for NICE NG185 chest pain guideline update incorporating new assay thresholds.

GUIDELINE WATCH EMPHASIS — minocycline for AIS: Modest benefit (NNT 19, ARD +5.2%, p=0.006) in 1,724 Chinese patients. Not practice-changing yet — needs UK/Western replication. Watch for NICE stroke guideline response.

INFORMING CT head score (UK validation): The Emergency CT Head Score validated in UK multicentre cohort for non-trauma imaging reduction. Potential to reduce unnecessary CT-heads once formally adopted in local pathways.

INFORMING ED fall prevention: SR/MA finds no significant fall reduction from ED-initiated multifactorial programmes (primary outcome not met at 3, 6 or 12 months). Functional status may improve. Tempers expectations for ED-initiated pathways alone.

INFORMING Bicarb for acute hyponatraemia: 50 mL 8.4% NaHCO3 outperformed 100 mL 3% saline for achieving ≥4 mmol/L Na rise (48% vs 10%). Small retrospective — useful if hypertonic saline is unavailable.

INFORMING AI ambient scribe: EMA commentary: shorter documentation time but low adoption rate, quality/completeness unassessed. Not yet evidence for system-level rollout. Hypothesis-generating only.

INFORMING Core revision: Lactate — dynamic lactate clearance >10% per hour is the target, not absolute value. Failure to clear by 2h is the key ED decision point for sepsis escalation.

The FASTEST trial settles a question that has circulated in emergency medicine and neurology for nearly 25 years: can recombinant factor VIIa prevent haematoma expansion and improve outcomes in spontaneous intracerebral haemorrhage? The answer, from a well-powered phase 3 multinational RCT stopped for futility, is no. rFVIIa reduces haematoma growth by 3.7 mL — a real pharmacological effect — but this does not translate into improved function at 180 days, and comes at the cost of a threefold increase in thromboembolic events. FASTEST-2 continues in the spot-sign+ population where ongoing bleeding is confirmed, but for now rFVIIa has no routine role in ICH. Meanwhile, in paediatrics: a network meta-analysis of 9 RCTs confirms what many departments already practise but others still do not — midazolam buccal or IM is superior to diazepam for childhood status epilepticus in every meaningful outcome, and should be the default regardless of the route available.

CONTENTS

Key Trials & Articles

1. FASTEST Phase 3 RCT — rFVIIa in Spontaneous ICH: No Benefit, Thromboembolic Harm [LEAD] 2. EMPHASIS RCT (Lancet) — Minocycline for AIS: NNT 19, Modest Benefit, Not Yet Practice-Changing 3. ITP Critical Bleeding Guideline (McMaster/Blood Advances) — ED Management Algorithm 4. Novel High-Sensitivity Troponin T Assay for Early MI Rule-Out (JAMA Cardiol) 5. Single-Shot Regional Anaesthesia for Rib Fracture Pain — NMA of 738 Patients (Acad EM) 6. Penicillin Allergy Direct Oral Challenge: 97% Tolerance — iNAAN International Study 7. Emergency CT Head Score UK External Validation (Emerg Med J) 8. Sodium Bicarbonate vs 3% Hypertonic Saline for Severe Hyponatraemia

Guidelines & UK Updates

9. NHS England Corridor Care Trust-Level Data — First Publication Due May 2026 10. AI Ambient Scribe in the ED — Early Signals and Important Caveats (EMA Daily) 11. ED-Initiated Fall Prevention — No Significant Fall Reduction (SR/MA, Acad EM)

Paediatric EM

12. Paediatric Status Epilepticus: Midazolam Buccal/IM — Best Drug and Route (NMA 9 RCTs) 13. Does This Child Have a Concussion? — Rational Clinical Examination SR (JAMA) 14. Paediatric Abscess I&D: POCUS Improves Volume Prediction vs Clinical Assessment Alone 15. ICAF RCT — Caffeine Does Not Reduce Intermittent Hypoxia in Very Preterm Infants

Core Revision, Action Points, Trials to Watch

1 — KEY JOURNAL ARTICLES & TRIALS

EMA DAILY · LANCET 2026;407(10529):773–783 · PUBLISHED 1 MAY 2026 · PMID 41653933

FASTEST Phase 3 RCT: rFVIIa Reduces Haematoma Expansion in Spontaneous ICH But Does Not Improve Functional Outcomes — and Triples Thromboembolic Risk. Trial Stopped for Futility. [LEAD]

RCT CHANGE TONIGHT FRCEM

• −3.7 mL HAEMATOMA GROWTH (RFVIIA VS PLACEBO)
• aOR 1.09 MRS AT 180D — P=0.61 (NO BENEFIT)
• RR 3.41 THROMBOEMBOLIC EVENTS (CI 1.14–10.15)
• n=626 STOPPED AT 2ND INTERIM (FUTILITY)

Broderick JP, Naidech AM, Elm JJ, Toyoda K et al. (FASTEST Investigators). Multinational, randomised, placebo-controlled phase 3 trial. 626 patients with spontaneous ICH (excluding those on anticoagulants) randomised to rFVIIa 80 mcg/kg IV (given within 2 hours of onset) or placebo. The primary clinical outcome was mRS distribution at 180 days. The trial was stopped at the second interim analysis for meeting pre-specified futility criteria.

Outcome

rFVIIa (n=328)

Placebo (n=298)

mRS 0–2 at 180 days (primary)

Not reported separately

aOR 1.09 (CI 0.79–1.51, p=0.61)

Thromboembolic events within 4 days

15 (<5%)

4 (1%) — RR 3.41 (CI 1.14–10.15), p=0.020

Critical appraisal: EMA Daily editor Sanjay Arora: "rFVIIa was found to decrease the volume expansion of spontaneous ICH, but the trial was stopped early because this finding did not translate to a clinical benefit and in fact showed a signal of harm in the form of an elevated rate of significant thromboembolic events." This is the key dissociation in the trial: haemostasis achieved, but at a cost that outweighs the benefit. The mechanistic hypothesis — that haematoma growth drives secondary injury — is not disproved, but rFVIIa is not the right tool to test it in unselected ICH. FASTEST-2 is ongoing, targeting patients with a spot sign on CTA (confirming active extravasation), where haematoma growth is most likely to be occurring and most likely to be treatable. Watch for FASTEST-2 results as the definitive test of this hypothesis in the enriched population.

Why it matters: Spontaneous ICH is one of the most devastating presentations in emergency medicine — approximately 15,000 cases per year in the UK with a 30-day mortality of 40%. For decades, haematoma expansion has been identified as a key modifiable target. FASTEST closes the door on rFVIIa for this indication in unselected patients. The practical message for UK EDs is unambiguous: there is no role for rFVIIa in routine spontaneous ICH management. If a clinician or relative asks about "clotting factor treatment," you can now cite a well-powered phase 3 trial. The only evidence-based acute interventions for ICH remain: blood pressure control (INTERACT2/ATACH-2 targets: SBP <140 mmHg), anticoagulant reversal if applicable, and avoidance of harm (glucose, temperature, aspiration).

Tell your department: If rFVIIa appears on any local ICH protocol or is being requested for spontaneous ICH, cite FASTEST. The UK-relevant context: rFVIIa (NovoSeven) is licensed in the UK for haemophilia with inhibitors and acquired haemophilia — not ICH. This trial provides the definitive evidence against its off-label use in this context.

Source: Broderick JP et al. FASTEST Trial. Lancet 2026;407:773–783 (PMID 41653933) · EMA Daily commentary (1 May 2026)

EMA DAILY · LANCET 2026;407(10526):679–688 · PUBLISHED FEBRUARY 2026 · PMID 41628627

EMPHASIS RCT (Lancet): Oral Minocycline for Acute Ischaemic Stroke Modestly Improves mRS 0–1 at 90 Days (NNT 19) — Not Yet Practice-Changing Pending UK Replication

RCT INFORMING PRACTICE FRCEM

• +5.2% ARD MRS 0–1 AT 90 DAYS
• NNT 19 MRS 0–1 AT 90 DAYS
• aRR 1.11 CI 1.03–1.20, P=0.006
• n=1,724 58 HOSPITALS, CHINA, NIHSS 4–25

Parameter

Minocycline (n=850)

Placebo (n=851)

mRS 0–1 at 90 days

52.6% (447/850)

47.4% (403/851)

Serious adverse events

4.6%

5.9% (p=0.24)

Symptomatic ICH at 24h

0.1%

0%

Lu Y, Guan L, Wu J et al. (EMPHASIS Investigators). Minocycline 200 mg loading dose then 100 mg every 12h for 5 days, started within 72 hours of AIS onset. 58 hospitals in China. Median age 65, NIHSS 5 (moderate strokes). Oral administration throughout.

Critical appraisal: EMA Daily editor Mike Menchine: "This well-powered RCT suggested that minocycline might modestly improve functional outcomes after stroke, possibly via anti-inflammatory effects. However, the absolute benefit is small, and concerns about generalizability and subgroup variability limit immediate clinical adoption. Overall, the findings are somewhat promising but not yet practice changing." The key limitations are the Chinese single-health-system context, oral administration in a largely mild-to-moderate stroke population (NIHSS 5 median), and uncertainty about the mechanism. Minocycline is a readily available, inexpensive oral tetracycline antibiotic — but it is not currently part of any UK AIS pathway. A Western replication study would be needed before NICE or the RCP would consider any guideline change. The absolute NNT of 19 is modest but not trivial for a stroke intervention with a good safety profile.

Tell your department: No immediate practice change. This is hypothesis-confirming data that will inform future trial design. Watch for the NICE stroke guideline surveillance (expected Q3 2026) for any response to EMPHASIS. If asked by colleagues or patients about minocycline for stroke, the accurate answer is: promising signal in one well-conducted trial, not yet replicated in Western populations, not in current UK guidelines.

Source: Lu Y et al. EMPHASIS Trial. Lancet 2026;407:679–688 (PMID 41628627)

EMA DAILY · BLOOD ADVANCES · PUBLISHED 4 FEBRUARY 2026 · PMID 41637560

McMaster Guideline: Emergency Management of Critical Bleeding in ITP — First Dedicated ED Guideline, Strong Recommendations Based on Weak Evidence

GUIDELINE CHANGE THIS MONTH FRCEM

Chowdhury SR, Sirotich E, Guyatt GH et al. McMaster group guideline on emergency management of critical bleeding in immune thrombocytopenia (ITP). This is the first dedicated guideline for ED management of this condition. EMA Daily editor Sanjay Arora: "A meaningful, well-intentioned first step toward standardising the approach to ED patients with severe ITP and critical bleeding. However, clinicians should recognise that strong recommendations are based on much weaker evidence than the label implies."

Key Recommendations for ED Critical ITP Bleeding

Critical appraisal: Arora's critical appraisal is important: this guideline acknowledges that most of the evidence is extrapolated from non-emergency settings. The "strong recommendation, very low evidence" classification is a grading construct that means expert consensus drives the recommendation, not RCT data. Platelet transfusion in ITP may be less effective than in other causes of thrombocytopenia due to the same immune mechanism destroying transfused platelets, but Arora notes that in haemostatic failure, any platelet increment (however brief) may be life-saving. Some agents (TPO-RAs) may not be rapidly accessible in a UK district general hospital — discuss availability with your local haematology team in advance.

Tell your department: Develop a local ITP critical bleeding protocol with haematology. Ensure IVIG is accessible via emergency pharmacy. Know the anti-D availability and your hospital's policy. This guideline is a useful reference document for an uncommon but high-stakes ED scenario.

Source: Chowdhury SR et al. Blood Adv 2026 — ITP Critical Bleeding Guideline (PMID 41637560) · EMA Daily commentary (7 May 2026)

MCMASTER EVIDENCE ALERTS · JAMA CARDIOL · MAY 2026 · SCORE 6/7

Novel High-Sensitivity Cardiac Troponin T Assay for Early MI Rule-Out: Outperforms Current hsTnT at 0h and 1h in Head-to-Head Comparison

OBSERVATIONAL CHANGE WHEN GUIDELINE UPDATES FRCEM

McMaster Evidence Alerts (score 6/7 — highest clinical impact tier this week). Head-to-head comparison of a novel hs-cTnT assay versus the current Roche hsTnT Elecsys assay. The novel assay demonstrates improved analytical performance and better clinical sensitivity/specificity at the 0h and 1h time points for ruling out NSTEMI. This is directly relevant to UK NICE NG185 chest pain pathway which is built around specific hsTnT assay thresholds. A new assay with better performance at 1h could reduce the number of patients requiring the 3h sample and accelerate the 0/1h rule-out pathway.

Critical appraisal: This is an observational diagnostic accuracy study — not an RCT of outcomes. The clinical question is whether better assay performance translates to safer accelerated pathways (fewer missed MIs or fewer unnecessary admissions). The current NICE NG185 0/1h pathway was validated using specific assay thresholds for specific platforms — any new assay requires its own platform-specific validation before its thresholds can be applied. UK ED departments should not apply novel assay thresholds to current equipment. The clinical significance: this study signals that the next generation of hs-cTnT assays may enable genuinely safer 0/1h rule-out, and NICE NG185 surveillance (triggered by the LEGEND trial, expected Q3 2026) will likely incorporate this.

Source: JAMA Cardiol 2026 — Novel hs-cTnT Assay Early MI Rule-Out (PMC open access) · via McMaster Evidence Alerts

JOURNALFEED EM WED · ACAD EMERG MED · APRIL 2026 · PMID 41914895

Single-Shot Regional Anaesthesia for Rib Fracture-Associated Pain: NMA of 738 Patients — Reduces Pain at 4–8 Hours and Opioid Requirements vs Standard Care

NMA CHANGE TONIGHT FRCEM

• Reduced PAIN AT 4–8H VS STANDARD CARE
• Reduced OPIOID REQUIREMENTS
• n=738 NETWORK META-ANALYSIS

Partyka C, Farenden S, Tian D, Delaney A, Curtis K. Network meta-analysis of adult patients with rib fractures. Single-shot regional anaesthesia techniques (erector spinae plane block, serratus anterior plane block, paravertebral block, intercostal nerve block) as adjuncts to standard care. All single-shot techniques outperformed standard care alone on pain at 4–8h and opioid consumption. No significant differences between the individual techniques in this analysis.

Critical appraisal: Network meta-analysis pools indirect comparisons — there were no direct head-to-head RCTs comparing all techniques. The study population is adult trauma patients; the benefit profile in elderly patients with multiple rib fractures (the highest risk group for respiratory complications) is the most clinically relevant. Procedural duration and clinician competence are not captured in NMA data — the real-world benefit depends on access to POCUS-trained clinicians and timely block delivery. This is a strong signal that any single-shot technique is better than no regional anaesthesia. The erector spinae plane block (ESPB) is accessible with basic POCUS training and should be the most deliverable option in a UK ED context.

Why it matters: Rib fractures are one of the most common ED injury patterns, and inadequate analgesia is a major driver of pneumonia, ICU admission, and prolonged hospital stay — particularly in older adults. If your department does not have a pathway for early regional anaesthesia in rib fractures, this NMA is the evidence base to develop one. The ESPB requires a single needle pass with ultrasound guidance, injects 20–30 mL 0.25% bupivacaine in the erector spinae fascial plane, and provides reliable analgesia for 12–18 hours. Discuss with your trauma team and anaesthetics colleagues to build a shared protocol.

Source: Partyka C et al. Acad Emerg Med 2026 — Rib Fracture Regional NMA (PMID 41914895)

JOURNALFEED EM WED · ALLERGY/ANN EMERG MED · APRIL 2026 · PMID 41921035

Penicillin Allergy: Direct Oral Challenge Is Safe in 97% — iNAAN International Multicentre Study Across 6,000+ Patients

OBSERVATIONAL CHANGE TONIGHT FRCEM

• 97% TOLERANCE RATE ON DIRECT CHALLENGE
• <1% ANAPHYLAXIS RATE
• n>6,000 INAAN INTERNATIONAL MULTICENTRE

Mitri EA, Fletcher LR, Vogrin S et al. (iNAAN — International Network of Antibiotic Allergy Nations). Large multicentre observational study of direct oral amoxicillin/penicillin challenge in patients carrying a "penicillin allergy" label. 97% tolerated the challenge without reaction. Anaphylaxis occurred in <1% (all manageable). Over 90% of "penicillin allergy" labels are ultimately found to be incorrect — this study reinforces that label in a large international dataset.

Critical appraisal: This is an observational study — patients self-selected for challenge and may represent a lower-risk group than those who decline. The population excluded patients with recent severe reactions (within 1 year), severe cutaneous adverse reactions (SCAR, SJS/TEN, DRESS), or known IgE-mediated anaphylaxis — these represent a small but important subgroup for whom direct challenge is not appropriate. The 97% figure applies to the correctly risk-stratified "low-risk" population. The critical point for UK EDs: the national antibiotic stewardship programme strongly supports penicillin allergy de-labelling because the "penicillin allergy" label drives use of broader-spectrum alternatives (cephalosporins, carbapenems, fluoroquinolones) with worse antimicrobial resistance profiles.

Why it matters: Approximately 10% of NHS patients carry a "penicillin allergy" label. Over 90% are incorrect. This label is directly contributing to unnecessary use of broader-spectrum antibiotics in UK hospitals, contributing to Clostridioides difficile infections, antimicrobial resistance, and worse outcomes in conditions where penicillins are the preferred treatment (e.g. streptococcal sepsis, surgical prophylaxis, infective endocarditis). An ED-based pathway to identify low-risk patients and offer direct challenge (or refer for allergy outpatient review) is now strongly evidence-supported. Discuss with your pharmacy and infection team.

Source: Mitri EA et al. iNAAN Study 2026 — Penicillin allergy direct challenge (PMID 41921035)

MCMASTER EVIDENCE ALERTS · EMERG MED J 2026;43(5) · SCORE 5/7 · UK MULTICENTRE

Emergency CT Head Score External Validation for Non-Trauma Imaging Reduction: UK Multicentre Retrospective Study — Clinically Useful Across Settings

OBSERVATIONAL CHANGE WHEN GUIDELINE UPDATES UK DATA FRCEM

UK multicentre retrospective validation of the Emergency CT Head Score — a clinical decision rule designed to reduce non-trauma CT head imaging. External validation confirms the score performs consistently across different UK settings. The ED non-trauma CT head (for presentations including headache, seizure, confusion, and syncope) is one of the highest-volume imaging decisions in emergency medicine, and the Emergency CT Head Score provides structured criteria to identify which patients can safely avoid immediate CT. The NICE CG176 head injury guidance does not cover non-trauma presentations; this score fills a gap.

Tell your department: The Emergency CT Head Score may be suitable for formal adoption in your department's non-trauma CT head pathway. It is not yet in national guidelines — but this UK validation provides the evidence base for a local protocol. Discuss with your radiology colleagues. The Ottawa SAH Rule remains the primary tool for thunderclap headache (covered in Issue 9).

Source: De Wit et al. Emerg Med J 2026 — Emergency CT Head Score External Validation (PMID 42031528; DOI: 10.1136/emermed-2025-215110) · via McMaster Evidence Alerts

JOURNALFEED ARTICLE OF DAY · MAY 6, 2026 · 2-CENTRE RETROSPECTIVE COHORT

Sodium Bicarbonate vs 3% Hypertonic Saline for Severe Hyponatraemia (≤120 mmol/L): Bicarb Achieves ≥4 mmol/L Rise in 48% vs 10% After Single Dose

OBSERVATIONAL INFORMING PRACTICE FRCEM

• 48% NAHCO3 ACHIEVED ≥4 MMOL/L RISE
• 10% 3% HTS ACHIEVED ≥4 MMOL/L RISE
• n=42 RETROSPECTIVE, 2 CENTRES

Two-centre retrospective cohort. 42 patients with severe hyponatraemia (Na ≤120 mmol/L). 50 mL of 8.4% sodium bicarbonate (HTB) vs 100 mL 3% hypertonic saline (HTS). JournalFeed reviewer Doug Wallace: "Nearly half of the HTB group met this target compared to just 10% in the HTS group. Median sodium increase was also higher with HTB (3 mmol/L vs 0 mmol/L)." No cases of osmotic demyelination syndrome in either group.

Critical appraisal: Small retrospective, 2 centres, no clinical endpoint (symptom improvement not assessed). The groups were not randomised — provider choice of agent introduces confounding. The mechanism for bicarb being more effective than HTS at similar sodium loads is not clear and may reflect patient selection. This is hypothesis-generating, not practice-defining. The main message is practical: if your department does not have 3% saline readily available (it requires pharmacy preparation in many UK hospitals), 8.4% sodium bicarbonate in 50 mL boluses may be a reasonable emergency alternative for acute symptomatic hyponatraemia. This should be confirmed with your local pharmacy and clinical pharmacologist. Standard management remains 100 mL 3% saline for acute symptomatic severe hyponatraemia (covered in Core Revision, Issue 10).

Source: JournalFeed — No Hypertonic Saline? Try Bicarb (6 May 2026)

2 — GUIDELINES & UK UPDATES

NHS ENGLAND · DUE MAY 2026

NHS England Corridor Care Trust-Level Data: First Public Publication This Month — Every Department's Figures Will Be Visible

UK DATA CHANGE TONIGHT

Action required now: NHS England trust-level corridor care figures are due this month. If your department is not documenting corridor care using the formal NHS England definition — every patient receiving any form of care in a space not designed for patient care — the window to establish a baseline before publication is closing. As noted in Issue 10 (BMJ data: 493,751 patients waited ≥24h in England in 2025), corridor care is now a nationally benchmarked metric. GIRFT is already visiting highest-burden trusts. Ensure consistent documentation by all clinical staff now.

Source: NHS England — Corridor Care Programme Update (March 2026)

EMA DAILY · ANN EMERG MED 2026 · PMID 41665590

AI Ambient Scribe Adoption and Documentation Time in the ED: Shorter Notes with Ambient AI, But Multiple Important Caveats on Quality and Adoption

OBSERVATIONAL INFORMING PRACTICE

Retrospective observational study from a single US tertiary ED. Ambient AI scribe use was associated with shorter documentation time. EMA Daily editor Sanjay Arora: "A largely descriptive, hypothesis-generating study that confirms what most emergency clinicians already know anecdotally: we need to take time on shift to eat and drink, and take care of ourselves. Confounding by indication, inability to assess note quality, diagnostic accuracy, or medicolegal completeness are important variables. The very low proportion of clinicians who chose to use the ambient AI also limits the generalisability." For UK context: NHS England is piloting ambient AI scribes in several trusts. The evidence base for clinical quality and safety remains very early.

Tell your department: AI ambient scribes are coming to UK EDs. The key governance question before adoption is documentation completeness and medicolegal safety, not just documentation speed. If your trust is piloting an ambient scribe product, ensure there is a formal quality assurance process reviewing clinical note accuracy, completeness, and patient safety flags — not just documentation time as the success metric.

Source: AI Ambient Scribe ED Study (PMID 41665590) · EMA Daily (5 May 2026)

MCMASTER EVIDENCE ALERTS · ACAD EMERG MED · SCORE 5/7

ED-Initiated Multifactorial Fall Prevention: SR/MA Finds No Significant Reduction in Falls — But Functional Status Improvements Seen

SR/MA

CHANGE WHEN GUIDELINE UPDATES

FRCEM

SR/MA of multifactorial and multicomponent fall prevention interventions initiated from the ED (Southerland et al., Acad Emerg Med 2026; PMC13109610). 12 RCTs analysed. Primary outcome — falls at 3, 6, and 12 months — was NOT significantly reduced: RD 0.05 (95% CI 0.00–0.09) at 3 months, 0.07 (−0.04–0.18) at 6 months, −0.02 (−0.11–0.07) at 12 months. Injurious falls, ED revisits, hospitalisation and mortality were also unchanged. However, functional status showed

improvement in 4 of 5 reporting studies (low certainty evidence). UK context: NICE NG161 endorses multifactorial falls assessment, and NHS SDEC units are a natural delivery point for such pathways — but this SR/MA should temper expectations that ED-initiated programmes alone will reduce fall rates in the short term.

Source: Southerland L et al. Acad Emerg Med 2026 — ED Fall Prevention SR/MA (PMC13109610) · via McMaster Evidence Alerts

3 — PAEDIATRIC EMERGENCY MEDICINE

JOURNALFEED ARTICLE OF DAY · 7 MAY 2026 · NMA OF 9 RCTS, N=1,135

Paediatric Status Epilepticus: Midazolam Is the Best Drug — Buccal and Intramuscular Routes Best When IV Unavailable — NMA of 9 RCTs (n=1,135)

NMA CHANGE TONIGHT PAEDS FRCEM

• RR 1.13 THERAPEUTIC SUCCESS (MIDAZ VS DIAZ)
• RR 0.74 TREATMENT FAILURE (FAVOURS MIDAZ)
• RR 0.51 SEIZURE RECURRENCE (FAVOURS MIDAZ)
• n=1,135 9 RCTS, NMA + TRIAL SEQUENTIAL ANALYSIS

Nicholson C (JournalFeed). Systematic review and NMA of 9 RCTs comparing midazolam and diazepam for paediatric status epilepticus, with trial sequential analysis (TSA) added to assess whether the evidence base is sufficient to guide practice. "The core finding is straightforward: midazolam outperformed diazepam on therapeutic success (RR=1.13), treatment failure (RR=0.74), and seizure recurrence (RR=0.51). Route of administration was a significant effect modifier. Buccal midazolam was superior to rectal diazepam."

Route and Dose Reference Table

Critical appraisal: NMA allows comparison of all routes simultaneously but introduces indirect comparisons. The TSA confirms that the evidence is sufficient to support the primary conclusion (midazolam > diazepam) — no further trials of this specific question are needed. The RCUK 2021 paediatric guidelines and APLS protocols already recommend buccal midazolam as first-line for out-of-hospital seizures, but rectal diazepam is still stocked and used in some UK hospitals. This NMA is the clearest evidence yet to standardise to midazolam buccal or IM as the universal first-line regardless of setting.

Tell your department: Audit your paediatric seizure trolley and emergency drug protocols. Ensure Buccolam (buccal midazolam) is stocked and accessible in resus, majors, and the paediatric bay. If rectal diazepam is still listed as first-line, update the protocol. Brief your paediatric nursing staff on the buccal administration technique — it requires no cannula and can be given immediately on arrival.

Source: JournalFeed — Best Drug, Best Route for Paediatric Seizures (7 May 2026)

JOURNALFEED EM SPEEDREAD · JAMA 2026 APR 6 · PMID 41941197

Does This Child Have a Concussion? — Rational Clinical Examination SR (JAMA): Mental Fog, Noise/Light Sensitivity, Nausea, and Ocular Findings Increase Likelihood

SR/MA CHANGE TONIGHT PAEDS FRCEM

Shah SN, Chizuk HM, Fong HF, Hannon M, Mannix RC. Rational Clinical Examination SR (JAMA), 23 studies. Clinical findings increasing likelihood of concussion in children and adolescents: mental fog/difficulty concentrating, sensitivity to noise or light, nausea, and ocular/vestibular abnormalities. Absence of headache lowers the probability.

Feature

LR+ (approximate)

Direction

Ocular / vestibular abnormalities (VOMS)

Strongly positive

Increases probability

Absence of headache

Negative

Decreases probability

Critical appraisal: The RCE SR methodology pools studies that have varying concussion definitions, reference standards, and populations — exact LR values vary considerably across studies. The useful clinical message is not a precise LR number but the identification of which features are most discriminating. Headache alone is non-specific (present in many conditions). The combination of cognitive symptoms + sensory sensitivity + nausea is more specific. VOMS (Vestibular/Ocular Motor Screening) is increasingly used in paediatric concussion clinics but requires training. In the ED, document the four key features above at assessment for every suspected concussion in a child.

Source: Shah SN et al. JAMA 2026 Apr 6 — Does This Child Have a Concussion? (PMID 41941197)

JOURNALFEED POCUS · PEDIATR EMERG CARE 2026 APR 1 · PMID 41422417

Paediatric Abscess I&D: POCUS + Clinical Assessment Predicts Volume Better Than Either Alone — Retrospective 653 Patients

OBSERVATIONAL INFORMING PRACTICE PAEDS

Neal JT, Waltman EM, Miller AF et al. Retrospective chart review of 653 paediatric patients undergoing I&D. Combining POCUS and clinical assessment improved prediction of abscess volume compared with either method alone, potentially reducing unnecessary procedures in smaller abscesses and ensuring adequate drainage in larger ones. The integrated approach affects the decision about whether to proceed with I&D or manage conservatively, and about how wide an incision is needed.

Tell your department: For paediatric skin and soft tissue abscesses in the ED, routine POCUS assessment before I&D is best practice — it confirms the presence of fluid loculation, estimates size, and guides procedural planning. This study adds the evidence that the combination of clinical + POCUS assessment is superior to either alone for volume estimation. Ensure your paediatric nurses and junior doctors are familiar with point-of-care soft tissue ultrasound technique.

Source: Neal JT et al. Pediatr Emerg Care 2026 — Paeds Abscess POCUS (PMID 41422417)

JOURNALFEED PAEDS · ARCH DIS CHILD FETAL NEONATAL ED 2026 APR 17 · PMID 41285561

ICAF RCT: Caffeine Does Not Reduce Intermittent Hypoxia Events in Very Preterm Infants — Neonatal/PEM (Arch Dis Child Fetal Neonatal Ed)

RCT INFORMING PRACTICE PAEDS

Eichenwald E, Corwin M, McEntire B, Knoblach S et al. Very preterm infants randomised to caffeine vs placebo for prevention of intermittent hypoxia events. Caffeine did not reduce intermittent hypoxia events. This challenges the assumption that caffeine — established for apnoea of prematurity — also reduces hypoxia episodes in this population. The ED relevance: preterm infants who re-present to the ED with apnoea or hypoxia events may be on caffeine — this trial clarifies that caffeine is for apnoea, not the broader hypoxia syndrome.

Source: Eichenwald E et al. ICAF RCT. Arch Dis Child Fetal Neonatal Ed 2026 (PMID 41285561)

4 — CORE REVISION: LACTATE — BEYOND THE NUMBER

FRCEM REVISION FOCUS — LACTATE IN THE ED: INTERPRETATION, CLEARANCE, AND PITFALLS

The absolute value is the starting point, not the endpoint

Lactate clearance — the key dynamic target Clearance = [(initial lactate − follow-up lactate) / initial lactate] × 100

Causes of elevated lactate beyond sepsis (Type B lactic acidosis)

Key rule: Lactate is a dynamic biomarker. Check it on arrival, check it again at 2 hours. Failure to clear ≥10% by 2h in the context of infection = inadequate resuscitation or undertreated infection source. In a patient on salbutamol nebulisers, a raised lactate does not mean sepsis — stop the salbutamol, recheck. In alcohol use disorder with a raised lactate not responding to fluids: give IV thiamine before or with IV glucose.

5 — ACTION POINTS THIS WEEK

• rFVIIa in ICH: Remove from any local ICH protocol if present. FASTEST phase 3 definitively shows no functional benefit and threefold thromboembolic harm in unselected spontaneous ICH. Document this at your next departmental governance meeting. The only evidence-based ICH interventions remain BP control (SBP <140 mmHg), anticoagulant reversal, and avoidance of secondary harm.
• Paediatric seizures: Audit your paediatric seizure drug protocol and trolley. Buccal midazolam (Buccolam) must be stocked and accessible at the bedside, not locked in a controlled drug cupboard. If rectal diazepam is still listed as first-line, update the protocol now. Dose: buccal 0.5 mg/kg (max 10 mg).
• Rib fracture regional anaesthesia: Identify which POCUS-trained clinicians in your department can perform erector spinae plane block or serratus anterior plane block. Develop a referral/activation pathway for any patient with ≥2 rib fractures, inadequate oral analgesia, or at risk of respiratory complications (elderly, COPD, pre-existing respiratory disease).
• Penicillin allergy: Raise penicillin allergy de-labelling as a departmental quality improvement project. Liaise with pharmacy and infection/microbiology. A simple first step: ensure all junior doctors know to ask "what happened when you took penicillin?" rather than accepting the allergy label at face value.
• ITP critical bleeding: Ensure IVIG is available via emergency pharmacy with a clear out-of-hours ordering pathway. Develop a joint protocol with haematology covering initial doses, platelet transfusion criteria, and anti-D eligibility. Most DGHs will see <5 cases/year — a pre-agreed protocol prevents decision paralysis at 3am.
• AI ambient scribe: If your trust is piloting an ambient scribe product, ensure the governance framework includes documentation quality review (not just documentation time), patient consent processes, and a clear policy on medicolegal completeness. Speed is not the only metric.
• Lactate clearance: Review your department's sepsis resuscitation protocol to confirm it specifies repeat lactate at 2 hours for any initial lactate ≥2.0 mmol/L. If the protocol says "repeat at 4 hours," update to 2 hours in line with SSC 2026 and ANDROMEDA-SHOCK data. The 2h decision point matters.

6 — TRIALS TO WATCH

IMMINENT — Q2 2026

FASTEST-2: Follow-on to FASTEST, targeting patients with CTA spot sign (confirmed active extravasation). This is the subgroup where rFVIIa has a biological rationale. Results will determine whether haematoma expansion in the right patient remains a viable treatment target. On-Scene ECPR Netherlands Trial: Inclusion complete. Results expected Q2–Q3 2026. If positive, will define the prehospital ECPR-eligible patient and logistics for UK HEMS/ambulance systems. NHS England Corridor Care Trust-Level Data: First publication this month. Every trust's figures will be publicly visible. Documentation must be in place now.

2026–2027

HSSIB Mental Health Crisis Care Part 2 (summer 2026): Binding recommendations on liaison psychiatry response times and ED mental health assessment protocols. Follow-up to Part 1 (Issues 5–6). NICE NG128 Stroke Guideline Surveillance (Q3 2026): Expected to respond to AHA/ASA 2026 stroke guideline (tenecteplase, post-EVT BP, EMPHASIS minocycline data). NICE NG185 Chest Pain Surveillance (Q3 2026): Triggered by LEGEND trial and new hs-cTnT assay data. Will update 0/1h rule-out pathway thresholds.

LONGER HORIZON

CoMiTED — UK Chest Drain RCT (2027): Conservative vs immediate drain in traumatic pneumothorax. Directly relevant to ED pleural procedure and disposition decisions.

Jake Turner

Curated with the assistance of AI (Perplexity). All content editorially reviewed. EM Evidence Rundown — Issue #11 — Week of 7 May 2026 — UK Edition Published by EM Evidence. Sent to subscribers in emergency medicine across the UK. For clinical use only — verify against local guidelines before implementing changes in practice. Feedback form · emevidence.org · emevidence999@gmail.com