This Week in Emergency Medicine — Sep 22, 2026

Welcome to This Week in Emergency Medicine. This week we're covering 10 notable papers spanning sepsis and resuscitation controversy, prehospital and critical care transport medicine, and everyday emergency department risk stratification. Let's dive in.

We'll start with sepsis, where a guideline critique and a prehospital cohort pull in interesting directions. In the American Journal of Emergency Medicine, Spiegel and colleagues have published what they frankly call a dissenting opinion on the 2026 Surviving Sepsis Campaign guidelines [2]. Their argument is less about any single recommendation and more about the architecture of modern sepsis care: that successive iterations have drifted from evidence-based guidance meant to support clinical judgement toward mandated bundles that quietly become the legal and administrative standard of care. It is an opinion piece, not data, and should be read as such, but it is a useful articulation of a discomfort many emergency physicians feel about metric-driven, protocolised resuscitation, and it argues for individualised, physiology-led decisions on fluids, timing and antibiotics rather than clock compliance. Set against that, Prehospital Emergency Care brings us an observational cohort from Suttapanit and colleagues in Thailand, asking whether pushing antibiotics even earlier, out in the field, saves lives [4]. Across 364 adults with suspected sepsis transported by a hospital-based emergency medical operation unit, a little over four in ten received prehospital empirical antibiotics. Crude 28-day mortality was similar between groups, around a quarter in each, but after adjustment for comorbidity, warning scores, organ dysfunction, lactate, fluids, vasopressors and transport variables, prehospital antibiotics were associated with roughly halved odds of death, an adjusted absolute risk reduction of about nine percentage points. Antibiotics arrived roughly two hours earlier, at the cost of about eight and a half extra minutes on scene. This is a single-centre observational study with substantial potential for confounding by indication, and the gap between the crude and adjusted results should give you pause, but it is a reasonable signal that the time-to-antibiotic question does not stop at the emergency department door.

Moving to airway management, two papers approach the same problem from different ends of the system. Lacy and colleagues, also in the American Journal of Emergency Medicine, offer an evidence update on the physiologically difficult airway [3], and the central message is that peri-intubation catastrophe in the emergency department is usually driven by physiology rather than anatomy. They frame four phenotypes plus one: hypotension, hypoxaemia, metabolic acidaemia, right ventricular dysfunction, and increasingly intracranial hypertension. Each demands a different pre-intubation plan. Hypotension is the strongest predictor of peri-intubation cardiac arrest, so resuscitate first with fluid, blood or vasopressors depending on the cause. For the hypoxaemic patient, non-invasive ventilation and high-flow nasal oxygen reduce desaturation. In severe metabolic acidaemia, minute ventilation must be preserved and apnoea time minimised. In right ventricular failure, preload, afterload and ventilatory strategy all need attention. Across every phenotype, first-attempt success and a rehearsed plan for deterioration are what matter. How well does that translate into the field for children? Cloyd and colleagues, in Prehospital Emergency Care, report a national multisite cohort of paediatric air medical intubations, 1,040 encounters from 39 sites [8]. Overall performance was respectable, with first-pass success just under 87 percent and overall success around 93 percent, and most crews used a sedative-plus-paralytic strategy, ketamine being the commonest induction agent. But performance fell sharply at the youngest end: in infants under one year, first-pass success was about 69 percent compared with just under 90 percent in adolescents, overall success was about 81 percent versus 94 percent, and more than half of infants had post-attempt physiologic instability, compared with under a third of adolescents. End-tidal carbon dioxide was documented after the attempt in only about six in ten infants. Video laryngoscopy was used in under nine percent of intubations overall, and medication practice varied enormously between sites, with ketamine use ranging from under one in ten cases at some sites to more than four in five at others. The practical implication is that infant prehospital airway management is where targeted training, capnography discipline and site benchmarking should be concentrated.

The transport theme continues with three more papers, all in Prehospital Emergency Care. George and colleagues studied medical management of aortic emergencies during critical care transport, over 1,530 encounters spanning two decades [6]. Guidelines favour beta-blockade first with calcium channel blockade added for pressure control, but in practice nicardipine alone was the most frequent choice, used in about half of encounters. Compared with esmolol alone, every other strategy was about equally likely to reach heart rate or systolic pressure targets. The differences were in adverse events: labetalol alone had the lowest rate of hypotension, at roughly one in fifty, while esmolol alone had the highest, around one in eleven, and the esmolol-plus-nicardipine combination had the lowest rate of tachycardia. The take-home is permissive rather than prescriptive: if the agent you have is esmolol, labetalol or nicardipine, you can reasonably get to target with any of them, so choose on the patient's haemodynamic vulnerability rather than on guideline hierarchy. Kenney and colleagues describe something newer, 100 consecutive non-intubated patients transferred between hospitals on dexmedetomidine infusions for suspected medetomidine withdrawal, the alpha-2 agonist now appearing as an adulterant in the opioid supply [10]. This toxidrome is dominated by agitation, vomiting and sympathetic overdrive: about three quarters of patients were hypertensive during transport and nearly four in ten were profoundly so. Reassuringly, these transfers were uneventful in respiratory terms. Only three percent of patients became hypoxic, none required advanced airway management, and only one patient needed a vasoactive infusion for hypotension. A minority needed extra antiemetic or sedative doses, and no patient required a different sedative infusion. So continuing dexmedetomidine through interfacility transfer appears safe, and the practical watchpoint is hypertension and nausea, not airway loss. Rounding out the transport section, Gaither and colleagues offer not a trial but a methodological foundation: an expert-informed causal model linking prehospital blood transfusion to mortality in traumatic haemorrhagic shock [7], mapping confounders like system response times, protocols and product availability, and mediators like reduced crystalloid exposure and shorter time to transfusion. It is a tool for designing and interpreting the observational studies that will inevitably dominate this field, and it is worth knowing about if you appraise prehospital blood literature.

Finally, three papers on bread-and-butter emergency department decisions. The largest, from O'Brien and colleagues in the American Journal of Emergency Medicine, addresses a question you face on most shifts: the anticoagulated patient with minor head trauma [1]. Using a global federated network spanning a decade and more than 1.8 million emergency department encounters with head computed tomography, and three propensity-matched comparisons, direct oral anticoagulant use came out consistently better than warfarin. Immediate intracranial haemorrhage occurred in just under four percent of patients on a direct oral anticoagulant versus five percent on warfarin, delayed haemorrhage after an initially negative scan in about two and a half percent versus just over three percent, and 30-day mortality was likewise about a fifth lower. Most strikingly, compared with patients on no anticoagulant at all, direct oral anticoagulant use showed no difference in haemorrhage or mortality. This is retrospective and observational, and residual confounding is always possible, but it supports what many of us already practise: routine mandatory observation and repeat scanning for direct oral anticoagulant patients after a negative initial scan is hard to justify, whereas warfarin patients remain a genuinely higher-risk group. Kamiloğlu and colleagues, in the Journal of Emergency Medicine, report 774 adults with posterior canal benign paroxysmal positional vertigo managed with an Epley-first strategy [5]. Roughly four in five patients had symptom resolution within an hour without further intervention, and no new focal neurological deficits emerged during evaluation. Older age and longer symptom duration independently predicted early failure, and the model discriminated well. The practical point is that the bedside manoeuvre should come before the scanner in appropriately selected patients. And in the American Journal of Emergency Medicine, Potter and colleagues analysed two decades of national injury surveillance data on frostbite, around 42,000 weighted emergency department visits [9]. Injuries peaked in January, with the winter months accounting for over half, and mechanisms split roughly between environmental and contact cold. About one in nine cases ended in admission, transfer or observation, and the risk of that serious disposition was nearly eight times higher in patients aged 65 and over, around triple in men and in those with social exposure such as homelessness, and lower for contact injuries. Exposure burden and clinical severity, in other words, follow different patterns, which matters for how we target prevention and disposition.

If you only have time for one paper this week, make it the head trauma anticoagulation cohort in the American Journal of Emergency Medicine [1]. It is the largest comparative dataset yet on a decision you make weekly, and it directly informs whether a patient on apixaban with a negative scan needs to stay.

Here are the key takeaways from this week in Emergency Medicine. First, after minor head trauma, direct oral anticoagulant patients look much more like unanticoagulated patients than like warfarin patients, so reserve your most intensive observation strategies for warfarin. Second, sepsis care is being pulled in two directions at once, with a published dissent against bundle-driven guidelines and an observational signal that field antibiotics may lower mortality; both argue for thinking rather than clock-watching. Third, peri-intubation collapse is usually a physiology problem, and in prehospital paediatrics that problem concentrates in infants under one year, where first-pass success and capnography documentation both fall short. Fourth, in aortic emergencies during transport, esmolol, labetalol and nicardipine all reach targets similarly, so pick your agent by adverse-event profile. And fifth, dexmedetomidine infusions for medetomidine withdrawal can be safely continued through interfacility transfer, with hypertension rather than hypoxia as the dominant issue.

That's your roundup for This Week in Emergency Medicine. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And think of one colleague — in any specialty — who never has time to keep up with the literature. Tell them about AudioScholar: a free ten-minute weekly for every specialty, to listen to in any podcast app, or to read at audioscholar dot C C.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.