EM Evidence Rundown — Issue 30

EM EVIDENCE RUNDOWN — ISSUE 30 — 18 SEPTEMBER 2026

EM Evidence Rundown

Emergency medicine evidence for UK clinicians — weekly — emevidence.org

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

Lead (Trauma & Airway): VL vs DL in trauma intubation — DEVICE trial secondary analysis (n=338, J Trauma 2026): VL achieves first-pass success in 88% vs 68% with DL, ARD +20 percentage points. Consistent with DAS 2025. VL should be first-line in trauma RSI. Change This Week: Hyperpronation for radial head subluxation now confirmed superior to supination-flexion across 11 RCTs: first-attempt failure 9.4% vs 25%. Adopt hyperpronation as default. Also: calcium pre-treatment before diltiazem in AF-RVR reduces diltiazem-induced hypotension without compromising rate control. UK Safety Alert: HSSIB report on mental health crisis care in EDs: 3% of all ED attendances, patients twice as likely to wait >12 hours, documented waits up to 110 hours. A significant legal gap — no clear power to hold patients awaiting Mental Health Act assessment. Calls for urgent legislative change. Paediatric EM: AHA paediatric cardiogenic shock framework; AAP influenza 2026-27 guidance (antiviral regardless of vaccination status in hospitalised/high-risk children); hyperpronation for radial head subluxation. Core Revision: Diabetic Ketoacidosis — JBDS diagnostic criteria, fixed-rate insulin, potassium targets, when to escalate to critical care, BSPED vs JBDS. FRCEM-relevant.

BOTTOM LINE UP FRONT — 18 SEPTEMBER 2026

ACT ON THIS NOW

LEAD VL in trauma (DEVICE, n=338): First-pass success 88% vs 68% with DL (ARD +20%; 95% CI 11–29%). VL should be the default device for all trauma RSIs where available. DAS 2025 supports this. Novice operators benefit most.

CHANGE Hyperpronation for nursemaid's elbow: First-attempt failure 9.4% vs 25% with supination-flexion across 11 RCTs. Switch to hyperpronation as your default technique. Brief all SHOs and nurses who reduce these.

CHANGE Calcium before diltiazem in AF-RVR: Pre-treat with IV calcium chloride 90–180 mg before diltiazem to reduce diltiazem-induced hypotension. No compromise in rate control efficacy.

UK SAFETY HSSIB mental health in EDs: No clear legal power to hold patients awaiting MHA assessment. Know your trust's escalation policy. Involve senior staff and document competing risks explicitly.

KNOW FOR NEXT TIME

INFORMING POCUS-driven dyspnoea pathway (Danish RCT, n=663): Formal POCUS pathway did NOT improve 24h discharge or LOS vs standard care. POCUS is a tool for the clinician, not a standalone pathway substitute.

GUIDELINE RhIg before 12 weeks: ACOG/ASRM confirm routine Rh testing and RhIg not required before 12 weeks gestation for pregnancy loss or abortion. Update your early pregnancy protocol.

PAEDS EM AHA Paeds Cardiogenic Shock: New formal definition, severity staging aligned with INTERMACS, and escalation-to-MCS framework. Any child deteriorating over 24h with suspected cardiogenic shock: low threshold for PICU referral.

CORE REVISION DKA: Diagnostic criteria, FRIII 0.1 units/kg/hr, potassium targets, when to call critical care. FRCEM.

This week is anchored by a piece of evidence that DAS 2025 already anticipates but that UK trauma teams need to act on: VL doubles first-pass success in trauma intubation compared with DL, with an absolute risk difference of 20 percentage points in the DEVICE trial. The radial head subluxation meta-analysis settles a long-standing clinical debate in favour of hyperpronation — the evidence has been building for years, and 11 RCTs is now a robust signal. The HSSIB report on mental health in EDs is uncomfortable reading but clinically essential: the legal framework around holding patients in mental health crisis is genuinely unclear, and understanding your position is a patient safety and medico-legal matter. Core revision covers DKA — one of the most common and high-stakes resus bay diagnoses, with FRCEM exam significance and frequent trainees' errors around insulin dosing and potassium management.

WHAT'S INSIDE — ISSUE 30

• Lead: VL vs DL in trauma — DEVICE analysis
• Change: Hyperpronation for radial head subluxation
• Change: Calcium before diltiazem in AF-RVR
• SAVE-O2 AI: Autonomous O2 titration RCT
• POCUS-driven dyspnoea pathway — negative RCT
• Resuscitative hysterotomy OHCA — SR outcomes
• HSSIB: Mental health crisis care in EDs
• RCEM: 59 new mental health EDs in England
• Guideline: RhIg — not required before 12 weeks
• AAP Influenza 2026-27 guidance
• PEM: AHA Paediatric Cardiogenic Shock statement
• PEM: AAP Influenza 2026-27 (paeds antiviral)
• PEM: Hyperpronation for radial head subluxation
• FOAMed: EMCrit RACC September 2026

• Core Revision: DKA — ED Management

Contents: 1. Key Trials & Articles — 2. Guidelines & UK Updates — 3. Paediatric Emergency Medicine — 4. FOAMed & Critical Appraisal — 5. Action Points — 6. Trials to Watch — 7. Core Revision: DKA

1 — KEY TRIALS & ARTICLES

LEAD SECONDARY ANALYSIS RCT FRCEM

Videolaryngoscopy vs Direct Laryngoscopy in Trauma Intubation — DEVICE Trial Secondary Analysis

Trent SA, Schauer SG, Prekker ME, Driver BE, et al; Pragmatic Critical Care Research Group. J Trauma Acute Care Surg. 2026 May 7; online ahead of print. doi:10.1097/TA.0000000000005021. PMID: 42112946. JournalFeed EM Speed Read, 14 September 2026.

Design: Pre-planned secondary analysis of the DEVICE trial — a multicentre, pragmatic, parallel-group RCT comparing VL with DL for emergency intubation of critically ill adults. This secondary analysis includes the 338 patients (24% of the parent DEVICE population) intubated in the setting of traumatic injury: VL group n=171, DL group n=167. Primary outcome: successful intubation on the first attempt. Unadjusted intention-to-treat analysis. Level II evidence.

Key results: VL achieved first-attempt intubation success in 88% of patients (151/171) vs 68% with DL (114/167). Absolute risk difference: +20 percentage points (95% CI 11–29%). This is a large, statistically robust effect. Severe complications during intubation and in-hospital outcomes did not significantly differ between groups, though the secondary analysis was underpowered for these endpoints.

• 88% VL FIRST-PASS SUCCESS (151/171)
• 68% DL FIRST-PASS SUCCESS (114/167)
• +20pp ARD (95% CI 11–29%)
• NNT=5 APPROXIMATE: 1 EXTRA FIRST-PASS SUCCESS PER 5 PATIENTS INTUBATED WITH VL

Critical appraisal: This is a pre-planned secondary analysis rather than the primary DEVICE trial analysis, which limits its inferential power somewhat — randomisation was not stratified by trauma status, and there is no formal power calculation for this subgroup of 338 patients. However, the subgroup was pre-specified, the effect size is large (ARD 20%), and the 95% CI excludes zero comfortably. The parent DEVICE trial itself showed a significant VL benefit in the overall critically ill population, so this trauma subgroup finding is consistent and biologically plausible. The study was conducted in US academic EDs and ICUs, not UK EDs or trauma bays specifically, and the operators were experienced clinicians with access to multiple VL devices. The finding is nonetheless directly applicable to UK practice, where DAS 2025 already recommends VL as first-line wherever available — this secondary analysis provides specific trauma-population evidence to reinforce that recommendation. Importantly, the secondary outcomes (severe complications, in-hospital outcomes) were not significantly different, but the study was underpowered to detect differences in these rarer events. This should not be interpreted as evidence of equivalent safety rather than a neutral result on an underpowered endpoint.

UK ED practice: VL should be your default laryngoscope for all trauma RSIs where available — DAS 2025 already mandates this, and the DEVICE data provide direct trauma-specific evidence. An ARD of 20 percentage points and NNT of approximately 5 is a clinically significant benefit. The advantage is likely greatest for less experienced operators. Ensure VL is at the bedside for every trauma RSI, pre-positioned before induction. Do not default to DL because of habit or familiarity — have DL as backup only. Novice operators should receive VL simulation training specifically in the trauma scenario, where airway anatomy is often distorted by haemorrhage, swelling, or C-spine immobilisation.

CHANGE THIS WEEK SR/MA 11 RCTS PAEDS EM

Hyperpronation Outperforms Supination-Flexion for Radial Head Subluxation — SR and Meta-Analysis of 11 RCTs

Aksel G, Corbacıoglu SK, Akoglu H, Islam MM et al. Comparative effectiveness of supination-flexion and hyperpronation maneuvers in radial head subluxation: A systematic review and meta-analysis. Am J Emerg Med. 2025;92:68–78. doi:10.1016/j.ajem.2025.03.011. PMID: 40086092. JournalFeed EM Speed Read, 17 September 2026.

Design: Systematic review and meta-analysis of RCTs comparing supination-flexion with hyperpronation for reduction of radial head subluxation (nursemaid's elbow) in children aged 0–6 years. Databases searched: PubMed, Embase, Web of Science (1980–2024). Eleven RCTs included. Risk of bias was high in 10 of 11 studies.

Key results: Hyperpronation demonstrated a significantly lower first-attempt failure rate than supination-flexion: 9.42% vs 25% respectively. This corresponds to an absolute risk difference of approximately −15.6 percentage points in favour of hyperpronation (NNT approximately 6 to prevent one first-attempt failure). Pain during the manoeuvre was comparable between techniques, with slight advantages for hyperpronation in some subjective assessments. The authors conclude that hyperpronation should be the preferred technique because of higher first-attempt success and simplicity of the manoeuvre.

• 9.4% HYPERPRONATION FIRST-ATTEMPT FAILURE RATE
• 25% SUPINATION-FLEXION FIRST-ATTEMPT FAILURE RATE
• ~NNT 6 TO PREVENT ONE FIRST-ATTEMPT FAILURE VS SUPINATION-FLEXION

Critical appraisal: The meta-analysis pools 11 RCTs, which is a reasonably large evidence base for this clinical question. However, 10 of 11 studies are rated as high risk of bias — largely due to allocation concealment failures, open-label designs (blinding operators to technique is impossible), and limited outcome reporting. The heterogeneity in technique execution (forearm position, speed, angle of elbow flexion at pronation), patient age range, and operator experience introduces clinical heterogeneity beyond what standard meta-analytic techniques can fully address. The paediatric population is largely under 5 years old — a group with limited capacity to report pain reliably, making pain comparison imprecise. Despite these limitations, the direction and magnitude of the effect is consistent across included studies and across multiple prior meta-analyses; this SR updates and reinforces an established evidence signal. The simplicity of hyperpronation (no elbow flexion required, single smooth motion) also reduces the risk of iatrogenic injury during the attempt.

UK PEM practice: Adopt hyperpronation as your default technique for radial head subluxation reduction. The technique: hold the child's elbow at 90 degrees of flexion with one hand applying gentle traction; with the other hand, grasp the child's wrist and forearm and smoothly pronate the forearm (palm down) in one slow, continuous motion. Success is indicated by a palpable or audible "click" at the elbow and the child rapidly resuming normal arm use. If the first attempt fails, try again or switch to supination-flexion. Brief all SHOs and ED nurses who manage this common paediatric presentation.

CHANGE THIS WEEK DOUBLE-BLIND RCT

IV Calcium Pre-Treatment Prevents Diltiazem-Induced Hypotension in AF with Rapid Ventricular Response

Az A, Sogut O, Dogan Y, et al. Reducing diltiazem-related hypotension in atrial fibrillation: Role of pretreatment intravenous calcium. Am J Emerg Med. 2025;88:23–28. doi:10.1016/j.ajem.2024.11.033. PMID: 39577214. JournalFeed EM Speed Read, 14 September 2026.

Design: Randomised, double-blind, placebo-controlled trial. Adults with AF or atrial flutter with rapid ventricular response (HR >120 bpm), n=217. Three arms: IV NaCl 0.9% placebo then diltiazem (n=73); 90 mg IV calcium chloride then diltiazem (n=71); 180 mg IV calcium chloride then diltiazem (n=73). Outcomes: SBP and HR at 5, 10, 15 minutes; need for additional diltiazem doses; adverse events. NCT06494007.

Key results: Both calcium chloride doses significantly attenuated diltiazem-induced hypotension compared with placebo at 5 minutes (C180D) and at 15 minutes (both calcium groups vs placebo). Critically, calcium pre-treatment did not compromise heart rate control — rate-lowering efficacy of diltiazem was preserved in both calcium groups (HR significantly lower in both calcium pre-treatment groups at 10 and 15 minutes). No significant difference in the need for additional diltiazem doses or adverse events. The full numerical SBP values are not reported in the available abstract, but the directional findings are consistent and statistically significant.

Critical appraisal: This is a well-designed RCT with double blinding and appropriate allocation. The three-arm design allows dose comparison (90 mg vs 180 mg CaCl2), though the abstract does not report a statistically significant difference between the two calcium doses on either SBP or HR. The trial is relatively small at 217 patients and was conducted in a Turkish emergency department, which may differ from UK AF-RVR populations in baseline cardiac comorbidity, diltiazem dose used, and resuscitation practices. The primary outcome is a surrogate — blood pressure at 5, 10, 15 minutes — rather than a clinical outcome such as cardioversion rate, ICU admission, or mortality. The mechanism is pharmacologically sound: calcium channel blockade by diltiazem at the atrioventricular node is relatively selective, while pre-loading with calcium gluconate or chloride can blunt the peripheral vasodilatory effects without fully reversing the nodal rate control. The practical implication is a useful addition to managing a common clinical scenario, though the magnitude of the haemodynamic benefit cannot be precisely quantified from the available abstract data.

UK ED practice: For haemodynamically stable AF with RVR where you plan IV diltiazem for rate control: pre-treat with IV calcium chloride 90–180 mg (approximately 1–2 mL of 10% CaCl2 solution) over 2 minutes before administering diltiazem. This attenuates the diltiazem-induced vasodilation and hypotension without compromising rate control. Note: if the patient is already hypotensive or borderline, diltiazem remains contraindicated regardless of calcium pre-treatment — consider amiodarone or digoxin instead. Check your trust formulary for diltiazem IV availability, which varies across UK trusts; some use verapamil as an alternative.

INFORMING MULTICENTRE RCT

SAVE-O2 AI — Autonomous Oxygen Titration Substantially Improves Normoxemia in Acutely Ill Adults

Douin DJ, Rice JD, Xiao M, et al; SAVE-O2 AI Investigators. JAMA Intern Med. 2026 Aug 3;e264023. doi:10.1001/jamainternmed.2026.4023. PMID: 42546017. JournalFeed EM Speed Read, 15 September 2026.

Design: Multicenter, unblinded, parallel-group RCT. Four US hospitals. Adults hospitalised for acute respiratory illness, trauma, burn, or acute care surgery receiving supplemental oxygen. Randomised 300 patients: autonomous titration (n=152) vs usual manual care (n=148). Intervention period: first 72 hours. Target SpO2 range 90–96%. Primary outcome: proportion of time in the target normoxemia range.

Key results: Autonomous oxygen titration achieved substantially more time in the normoxemia target range: 85% vs 63% with usual care (adjusted risk difference +21 percentage points; 95% CI 18–25%; p<.001). Time in hypoxemia (SpO2 <88%) was reduced: 2.0% vs 3.6% (adjusted RD −1.3 pp; 95% CI −2.0 to −0.5; p=.002). Hyperoxemia (SpO2 >96%) was markedly reduced: 9.2% vs 29.1%. Results consistent across prespecified subgroups.

• 85% TIME IN NORMOXEMIA WITH AUTONOMOUS TITRATION
• 63% TIME IN NORMOXEMIA WITH USUAL CARE
• +21pp ADJUSTED RD FOR NORMOXEMIA (95% CI 18–25%)
• 29% HYPEROXEMIA WITH USUAL CARE (VS 9.2% AUTONOMOUS)

Critical appraisal: This RCT demonstrates impressive improvements in a key physiological process measure — time in target SpO2 range. The trial is well-designed with prospective randomisation across four hospitals and consistent results across subgroups. However, the critical limitation is the absence of patient-centred outcomes: mortality, ICU admission, duration of respiratory support, and hospital LOS are not reported. It is unclear whether the improved normoxemia translates into meaningful patient benefit, or whether it represents an improvement in monitoring precision without clinical impact. The trial was conducted over 72 hours only, in a US hospital setting with specific device infrastructure. The autonomous titration system is not described in detail, and it is not yet commercially available or approved in the UK. The substantial reduction in hyperoxemia (from 29.1% to 9.2%) is noteworthy — oxygen excess carries real harms including absorption atelectasis and worsened outcomes in specific populations (myocardial infarction, stroke, COPD). This is an emerging technology to watch rather than an immediate practice change.

UK context: Autonomous oxygen titration devices are not currently standard in UK EDs or wards. However, the trial highlights an important finding regardless of technology: manual oxygen management results in patients spending 37% of their time outside the target SpO2 range, including 29% in hyperoxia. Review your department's oxygen prescribing culture — oxygen should always be prescribed with a target range (typically 94–98% for most patients; 88–92% for COPD with known hypercapnic risk) and the flow rate adjusted to maintain that range, not simply left at the admission flow rate.

INFORMING MULTICENTRE RCT NEGATIVE

POCUS-Driven Diagnostic Pathway for Dyspnoea — No Improvement in 24h Discharge or LOS (Danish RCT, n=663)

Ovesen SH, Skaarup SH, Aagaard R, et al. A point-of-care ultrasound-driven diagnostic pathway for emergency department patients with dyspnoea: a randomised controlled trial. Eur Respir J. 2026;67(3):2500070. doi:10.1183/13993003.00070-2025. PMID: 41412717

Design: Multicentre RCT across 10 Danish EDs. Adults presenting with dyspnoea as their chief complaint. Randomised 674 patients (663 analysed): POCUS-driven pathway (focused lung and cardiac POCUS added to standard care) vs standard care (without planned POCUS). Primary outcome: proportion of patients discharged alive within 24 hours. Secondary outcomes: hospital LOS, chest imaging utilisation, 72-hour alive-and-revisit-free status.

Key results: Discharge alive within 24 hours: 42.6% with POCUS pathway vs 45.5% standard care. Risk difference: −2.9 percentage points (95% CI −10.4 to 4.7; p=0.45). Overall hospital LOS was not significantly altered (HR 0.93; 95% CI 0.79–1.08; p=0.35). The POCUS-driven pathway did not improve either primary or key secondary outcomes compared with standard care.

Critical appraisal: This is a well-conducted pragmatic RCT with a clinically relevant patient population and an important negative result. The primary outcome (24h discharge rate) was chosen as a pragmatic marker of diagnostic efficiency — the assumption being that faster, more accurate diagnosis would enable earlier disposition decisions. The fact that POCUS did not improve this suggests either that POCUS does not materially change diagnostic accuracy (perhaps because Danish ED physicians already perform POCUS routinely), or that diagnostic speed is not the rate-limiting step to discharge in dyspnoea patients, where treatment response and social factors dominate. The trial was conducted in Denmark, where POCUS integration in ED practice may be more embedded than in many UK EDs, potentially creating a ceiling effect — meaning the benefit of adding a formal POCUS pathway was minimal above an already high baseline. This does not mean POCUS is unhelpful for individual clinical decisions in dyspnoea — it means a formalised pathway protocol does not improve population-level flow outcomes. POCUS remains a valuable clinical tool when it changes a specific management decision (e.g., identifying pleural effusion, assessing LV function, ruling in pneumothorax).

UK ED practice: This trial argues against mandating a blanket POCUS-driven diagnostic protocol for all dyspnoea presentations. Use POCUS purposefully for specific clinical questions where the result changes your management — not as a screening tool on every dyspnoeic patient to tick a pathway box. Where POCUS is genuinely useful (suspected pneumothorax, AHF vs infective, pericardial effusion), the scan should be targeted and prompt. The throughput benefit of POCUS comes from the scan informing a decision that would otherwise require a longer investigation pathway, not from adding it to standard care universally.

INFORMING SYSTEMATIC REVIEW

Resuscitative Hysterotomy for OHCA — Maternal Survival 4.5%, Neonatal Survival 45%, Longer Time Window Than Previously Taught

Leech C, Nutbeam T, Chu J, Knight M, et al. Maternal and neonatal outcomes following resuscitative hysterotomy for out-of-hospital cardiac arrest: A systematic review. Resuscitation. 2024. doi:10.1016/j.resuscitation.2024.110479. Reviewed by St Emlyn's Feb 2025; JournalFeed EM Speed Read, 17 September 2026.

Design: Systematic review registered with PROSPERO. 42 publications included (one cohort study, three case series, 38 case reports): 66 women, 68 neonates. Narrative synthesis due to evidence heterogeneity. Evidence certainty: very low. PROSPERO CRD42023445064.

Key results: Maternal survival to hospital discharge: 4.5% (3/66). Neonatal survival to hospital discharge: 45.0% (27/60 viable neonates with follow-up data). Prehospital procedures: neonatal survival 62.5% vs in-hospital 42.9%. Median gestational age 35 weeks (range 24–41). Medical cause (PE, AFE, cardiac, haemorrhage): 68%. Trauma: 32%. Longest collapse-to-hysterotomy interval with maternal survival and normal neurological function: 29 minutes. Longest collapse-to-delivery interval with good neonatal outcome: 47 minutes. Good neurological neonatal outcomes documented at 26 weeks gestation. No maternal survivors in trauma-related cardiac arrests.

Critical appraisal: This systematic review synthesises the best available evidence on a rare and high-stakes procedure. The evidence quality is very low (case reports and small case series dominate), and publication bias is certain — successful outcomes are more likely to be published. The 4.5% maternal survival figure is likely an overestimate of the true benefit, since failed procedures and cases where the procedure was not attempted are systematically missing. However, the key clinical messages are pragmatically important: the previously taught 4-minute rule (perform within 4 minutes of arrest) is not supported by the data — maternal and neonatal survivors have been documented at intervals considerably longer than this. The neonatal survival benefit (45%) substantially exceeds the maternal benefit and should drive the decision to perform the procedure in pregnancies at or above 24 weeks where survival of either patient is possible. The finding that prehospital performance is associated with higher neonatal survival (62.5% vs 42.9% in hospital) may reflect case selection, timing, or both — it should not be used as a reason to delay in-hospital performance awaiting a prehospital team.

UK ED practice: Resuscitative hysterotomy should be considered in any pregnant patient ≥24 weeks gestation with OHCA where standard resuscitation is failing. Do not withhold or delay because of prolonged downtime alone — the evidence does not support a rigid time cutoff. Primary goal is neonatal survival (45% in this SR); maternal survival is possible but rare. UK RCUK guidance: perform within 4 minutes if ROSC not achieved. The 29-minute maternal survivor in this SR and the 47-minute neonatal survivor challenge strict time limits. Ensure your resus team knows the procedure, equipment location, and that surgery/neonatology should be alerted immediately.

2 — GUIDELINES & UK UPDATES

UK SAFETY HSSIB REPORT

HSSIB Report: Mental Health Crisis Care in Emergency Departments — Legal Gap, Prolonged Waits, Calls for Legislative Change

HSSIB. Mental health crisis care: care of patients in Emergency Departments. Investigation Report 1 of 2. Published 17 September 2026. hssib.org.uk. RCEM response: Dr Ian Higginson, 17 September 2026.

Key findings: Around 3% of all ED attendances are mental-health-related, equating to approximately 9,000 per week nationally. Patients in mental health crisis are twice as likely to wait more than 12 hours. Of all patients waiting more than 72 hours in EDs, 24% have mental health presentations. Cases of patients waiting 5–15 days in EDs for a mental health bed have been documented, with the longest reported case 110 hours. Timely MHA assessment (within 1 hour) was described as "nearly always impossible" due to AMHP and Section 12 doctor shortages.

The legal gap: The HSSIB report identifies a significant and unresolved legal problem: there is no clear general legal power to hold a patient in an ED who is awaiting a Mental Health Act assessment or an inpatient mental health bed, unless they meet criteria for deprivation of liberty under the MCA or can be held under Section 136 (time-limited, police-dependent). Staff described being forced to "choose the least harmful way to break the law" to keep patients safe. Section 5 MHA cannot be used in an ED (applies only to admitted hospital inpatients). Even when two medical recommendations for detention are made, formal detention does not occur until a bed is identified — and bed delays can take days. Unlawful deprivation of liberty carries substantial damages risk.

HSSIB recommendations: (R/2026/082) Department of Health and Social Care must urgently review the legal framework and clarify powers for holding patients in mental health crisis in EDs. (R/2026/083) CQC must produce a position statement on existing legal powers and expectations for care of patients in mental health crisis in EDs.

UK ED practice: This is a patient safety and medico-legal matter. Key points: (1) You have no clear general legal power to prevent a capacitous patient in mental health crisis from leaving the ED while awaiting MHA assessment. (2) Section 136 lasts up to 24 hours (or 36 with extension for clinical reasons, not bed delays alone); Section 5 MHA does not apply to ED patients. (3) If police leave, responsibility for safety may transfer to the ED — discuss explicitly with police whether and when they leave. (4) Document all competing risks under ECHR Articles 2, 3, 5, and 8. (5) Escalate every prolonged MHA assessment delay through your trust's escalation policy. (6) Know and use the least restrictive approach available. Section 4B of the MCA may authorise brief emergency deprivation of liberty for patients lacking capacity where life-sustaining treatment is needed, but this has limitations. Seek senior advice in complex cases.

UK RCEM

UK Government Announces 59 New Dedicated Mental Health Services in English EDs — RCEM Response

Royal College of Emergency Medicine. Dr Ian Higginson statement. August 2026. rcem.ac.uk

The UK Government announced 59 new dedicated mental health services within English emergency departments — part of an expansion of provision across the health service. RCEM's President, Dr Ian Higginson, welcomed the announcement as a step toward providing more appropriate care environments for people in mental health crisis, but warned that without adequate legal powers, staffing, and inpatient bed capacity, even purpose-built spaces cannot resolve the systemic problems highlighted by HSSIB. The announcement coincides with the HSSIB report publication — together they represent an acknowledgement at government and safety-body level that the current position is unsustainable.

GUIDELINE UPDATE ACOG/ASRM

RhIg Prophylaxis Not Required Before 12 Weeks — ACOG/ASRM Update 2026

ASRM Practice Committee. Position Statement on Rho(d) Immune Globulin Administration in the First Trimester. Fertil Steril. 2026;126:492–493. Based on ACOG Clinical Practice Update (published December 2024/online October 2024). asrm.org. JournalFeed EM Speed Read, 17 September 2026.

Key recommendation: Patients at less than 12+0 weeks gestation who experience vaginal bleeding, pregnancy loss (spontaneous or managed), or abortion do not require routine Rh testing or RhIg prophylaxis. Rh testing and RhIg may still be considered on an individual basis after shared decision-making, particularly if the patient has a strong preference, the institution has different policies, or clinical circumstances are more complex. At or beyond 12+0 weeks, routine Rh testing and RhIg prophylaxis remain recommended for Rh-negative unsensitised patients with vaginal bleeding or pregnancy loss.

Rationale: Fetal red blood cell concentrations remain below the estimated sensitisation threshold in 99.8% of pregnancies before 12 weeks. Historical estimates of Rh sensitisation risk after early pregnancy loss were 1.5–2% after spontaneous miscarriage and 4–5% after surgical evacuation — but more recent evidence suggests these estimates were substantially overstated. The risk of harm from withholding RhIg before 12 weeks is extremely low, and the benefit of giving it is effectively unproven at this gestational age.

UK ED practice: This guidance is ACOG/ASRM (American). UK practice currently follows RCOG guidance, which is undergoing review in light of the same evidence base. Before changing your practice, confirm whether your trust has adopted an updated protocol. Many UK EDs still routinely give anti-D (RhIg) in first-trimester bleeding and miscarriage regardless of gestation — a change to this practice should be policy-driven rather than individual-clinician-driven. Flag this evidence to your early pregnancy unit or gynaecology team for a local guideline review. Do not unilaterally change practice pending UK guidance update.

GUIDELINE PAEDS EM

AAP Influenza 2026-27: Antiviral Treatment for Hospitalised and High-Risk Children Regardless of Vaccination Status

Committee on Infectious Diseases. Recommendations for Prevention and Control of Influenza in Children, 2026–2027. Pediatrics. 2026;online ahead of print. doi:10.1542/peds.2026-078778. PMID: 42572140. JournalFeed Paeds Speed Read, 17 September 2026.

Key recommendations: The AAP updates annual influenza guidance for 2026-27. Core messages: (1) Annual influenza vaccination for all children from 6 months without contraindications, using any age-appropriate licensed vaccine as soon as available in the season. (2) Antiviral treatment (oseltamivir) is recommended for children who are hospitalised, have severe or progressive disease, or are at high risk for complications (chronic conditions, immunocompromised) — treatment should begin as soon as possible regardless of duration of illness and regardless of vaccination status. (3) In ambulatory settings, antiviral treatment is an option for all children with suspected or confirmed influenza.

The critical point for UK ED practice is the vaccination-status independence of the antiviral recommendation: do not withhold oseltamivir from a hospitalised or high-risk child with influenza because they have been vaccinated. While this is UKHSA guidance rather than RCOG/NICE, the underlying evidence base is shared and the principle is consistent with UK antiviral prescribing guidance.

UK PEM practice: In UK EDs, treat all children admitted with confirmed or strongly suspected influenza per UKHSA guidance: oseltamivir for hospitalised patients and all at-risk children. Do not use vaccination status as a reason to withhold antiviral treatment. As influenza season approaches (October onwards), ensure your ED has a clear influenza pathway, testing protocol, and oseltamivir prescribing guidance at the bedside. Children presenting with rapidly progressive respiratory symptoms and a positive flu swab in the immunocompromised or chronically ill population should be escalated early.

3 — PAEDIATRIC EMERGENCY MEDICINE

PAEDS EM AHA STATEMENT

AHA Scientific Statement: Paediatric Cardiogenic Shock — Definition, Severity Staging, and Escalation to Mechanical Circulatory Support

Puri K, Allen K, Jentzer J, Soderstrom R, et al; American Heart Association. Surviving Pediatric Cardiogenic Shock: Clinical Approach, Improving Outcomes, and Future Directions. Circulation. 2026 Sep 8;154(10):e374–e392. doi:10.1161/CIR.0000000000001461. PMID: 42558062. JournalFeed Paeds Speed Read, 16 September 2026.

What this is: An AHA Scientific Statement proposing the first formal definition of paediatric cardiogenic shock, severity staging aligned with existing validated systems (analogous to INTERMACS staging used in adult cardiogenic shock), a diagnostic framework for the bedside clinician, and guidance on when to escalate to mechanical circulatory support (MCS) including ECMO and ventricular assist devices. The statement identifies that a significant proportion of children with cardiogenic shock continue to deteriorate in the first 24 hours, emphasising the need for serial clinical assessment and readiness for escalation.

Key clinical framework: Paediatric cardiogenic shock is characterised by cardiac dysfunction resulting in inadequate systemic oxygen delivery, with or without hypotension. The proposed staging system grades severity from mild (NYHA-equivalent, compensated) through severe (refractory shock requiring escalating support), allowing structured decision-making about MCS timing. Early referral to a specialist paediatric cardiac centre is recommended for any child showing signs of progression. Haemodynamic monitoring strategies (non-invasive: POCUS, near-infrared spectroscopy; invasive: arterial line, central venous pressure) are discussed, with POCUS highlighted as the first-line bedside assessment tool in the ED.

Critical appraisal: This is a consensus scientific statement rather than a primary study — it synthesises expert opinion and existing evidence rather than generating new outcome data. The absence of paediatric RCT evidence in cardiogenic shock means much of the guidance is extrapolated from adult data, observational studies, and registry analyses. The statement's value is in standardising terminology and providing a structured framework for what is currently a highly variable, experience-dependent clinical decision. UK applicability requires consideration of the differential access to paediatric cardiac surgery, ECMO, and VAD services between major centres and district general hospitals, and the need for early referral and transfer.

UK PEM practice: Paediatric cardiogenic shock is a high-stakes, low-occurrence presentation. Any child with signs of cardiogenic shock — poor perfusion, tachycardia disproportionate to illness severity, hepatomegaly, cool peripheries, raised JVP or CVP, abnormal respiratory effort without primary respiratory cause — should trigger urgent POCUS (LV function, pericardial effusion, IVC assessment) and an early call to your paediatric intensive care or regional cardiac centre. Do not wait for deterioration to escalate. Standard ED sepsis protocols may be harmful in cardiogenic shock (aggressive fluids can precipitate pulmonary oedema). Inotropic support, cautious fluid management, and early PICU/cardiac centre referral are the priorities.

4 — FOAMED & CRITICAL APPRAISAL

EMCRIT

EMCrit RACC Lit Review — September 2026

Scott Weingart. EMCrit. Published 12 September 2026. emcrit.org (members only)

Weingart's monthly literature review for September 2026. The accessible commenter discussion highlights the first study using capnography to confirm NG tube placement — using a specific adapter to connect the NG tube to an end-tidal CO2 monitor. The waveform is interpretable if the tube is in the airway (positive CO2 waveform) but provides limited information about gastric vs oesophageal placement in the absence of CO2. Full article content is restricted to EMCrit members. The September 2026 lit review covers the major trials and educational content from the month. Members can access the full review at emcrit.org. The broader EMCrit archive for September includes EMCrit 433 (HALO procedures, covered in Issue 29) and the RACC lit review.

UK context: Capnography for NG tube confirmation: the current gold standard remains chest X-ray for initial confirmation before feeding commences. Capnography can rapidly exclude tracheal placement (CO2 waveform positive = trachea, not stomach) but cannot confirm gastric position. This is a useful adjunct for avoiding tracheal tube insertion, but does not replace radiological confirmation for post-pyloric or high-risk feeding tubes.

5 — ACTION POINTS

6 — TRIALS TO WATCH

ONGOING / EXPECTED

CORE REVISION — ISSUE 30 — 18 SEPTEMBER 2026

Diabetic Ketoacidosis (DKA)

JBDS diagnostic criteria, fluids, fixed-rate insulin, potassium, critical care escalation, BSPED vs JBDS — FRCEM

Exam goal: State the three JBDS diagnostic criteria; describe the initial fluid regimen; explain fixed-rate IV insulin infusion (FRIII) rate and targets; state when to add 10% dextrose and reduce insulin; give potassium replacement thresholds; list criteria for severe DKA requiring critical care; distinguish JBDS (adult) from BSPED (paediatric) key differences; recognise euglycaemic DKA and SGLT2 inhibitor association. JBDS March 2023 is the current reference.

1. DIAGNOSTIC CRITERIA (JBDS — ALL THREE REQUIRED)

All three of the following must be present to diagnose DKA:

2. EUGLYCAEMIC DKA — DO NOT MISS

Euglycaemic DKA: DKA in a patient with diabetes where blood glucose is below 11.0 mmol/L (often 6–10 mmol/L). Increasingly common because of SGLT2 inhibitor use (dapagliflozin, canagliflozin, empagliflozin, ertugliflozin, sotagliflozin). Patient may not look hyperglycaemic — clue is the ketosis and acidosis. Treatment: start 10% glucose at 125 mL/hour immediately, begin insulin at 0.1 units/kg/hour, reduce insulin to 0.05 units/kg/hour if glucose keeps falling. Stop the SGLT2 inhibitor and complete a Yellow Card.

3. INITIAL FLUID MANAGEMENT

Fluid choice: JBDS recommends 0.9% sodium chloride with pre-mixed potassium chloride as the default. Balanced crystalloid (Hartmann's) may be used in critical care settings per local policy. Do not use colloid.

Use more cautious/slower fluid replacement in: young adults 18–25 years (cerebral oedema risk), elderly, pregnant, heart failure, renal failure. Consider HDU/central line in these groups.

4. FIXED-RATE INTRAVENOUS INSULIN INFUSION (FRIII)

Always start fluids before insulin. FRIII: 0.1 units/kg/hour. Prepare 50 units human soluble insulin (Actrapid or Humulin S) in 50 mL 0.9% NaCl = 1 unit/mL. No priming bolus required. Continue patient's long-acting basal insulin at their usual dose and time (or start glargine 0.25 units/kg SC if newly diagnosed). Do NOT stop basal insulin during FRIII.

FRIII targets — if not meeting targets, check pump first, then increase by 1 unit/hour increments:

Stop FRIII when: Ketones <0.6 mmol/L AND pH >7.3 AND/OR bicarbonate >18 mmol/L, AND patient is eating and drinking. Give SC fast-acting insulin with a meal and overlap by 30–60 minutes before stopping the infusion.

5. POTASSIUM REPLACEMENT

Potassium falls rapidly during insulin therapy. Do not prescribe potassium if the patient is anuric. In ESKD: less potassium loss but hyperkalaemia is common due to acidosis — cardiac monitoring mandatory, consider urgent dialysis.

6. SEVERE DKA — CRITERIA FOR CRITICAL CARE INVOLVEMENT

Any ONE of the following indicates severe DKA requiring HDU or ITU consideration:

7. CEREBRAL OEDEMA

Rare in adults (<25 years old at highest risk); more common in children. Risk factors: younger age, marked hyperosmolarity (>330 mOsm/kg), failure of serum sodium to rise appropriately during treatment (dilutional hyponatraemia from excessive hypotonic fluid). Warning signs during treatment: decreasing GCS, headache, worsening nausea, bradycardia. Action: give mannitol or hypertonic saline immediately without waiting for imaging. Consider nasogastric tube with airway protection in any drowsy DKA patient.

8. JBDS VS BSPED: KEY DIFFERENCES

Use JBDS (adult guideline) for patients ≥18 years managed by adult teams. For 16–17-year-olds: use BSPED if managed by paediatric team, JBDS if managed by adult team — follow whichever guideline the ward staff know. The key difference is fluid management: paediatric DKA fluids are calculated by body weight and deficit, given more cautiously to reduce cerebral oedema risk. The paediatric protocol uses the BSPED DKA calculator and specifies deficit-based fluid resuscitation over 48 hours. Adult JBDS uses fixed-volume regimens per the table above. Never use the adult fluid regimen in a paediatric patient.

Memory aid — DKA severe criteria (KBPAGO): Ketones >6 — Bicarb <5 — PH <7.0 — Admission K <3.5 — GCS <12 — O2 sat <92%. Plus: SBP <90, HR >100 or <60, anion gap >16.

FRCEM exam focus: (1) All three JBDS diagnostic criteria including the euglycaemic variant. (2) Start fluids before insulin. (3) FRIII = 0.1 units/kg/hour; reduce to 0.05 units/kg/hour and add 10% dextrose when glucose <14 mmol/L. (4) Do not stop insulin until ketones <0.6, pH >7.3, bicarb >18 AND eating and drinking. (5) Potassium thresholds: >5.5 = no K, 3.5–5.5 = 40 mmol/L, <3.5 = stop insulin/senior review. (6) Severe DKA criteria (any one = HDU). (7) Euglycaemic DKA: SGLT2 inhibitor association, treat with 10% glucose + insulin, do NOT wait for hyperglycaemia. (8) BSPED vs JBDS: different fluid protocols; never use adult volumes in children.

EM Evidence Rundown — Issue 30 — 18 September 2026 Curated by Jake Turner, Senior Registrar in Emergency Medicine (ST6). Produced with the assistance of AI (Perplexity). All content editorially reviewed. Not a substitute for clinical judgement or local guidelines. Feedback: Submit feedback | Archive: emevidence.org Unsubscribe: Reply UNSUBSCRIBE to this email. To update your preferences, visit emevidence.org