This week in Emergency Medicine

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Emergency Medicine. This week we're covering 10 notable papers spanning diagnostic rule-out pathways, resuscitation and airway performance, and the evidence behind everyday emergency therapeutics. Let's dive in.

We start with rule-out pathways, where several papers this week question whether faster or more sensitive tests change what happens to patients. The largest is in The Lancet, where Boeddinghaus and colleagues pooled individual patient data from all five randomised trials comparing the guideline-endorsed 0/1 hour troponin pathway with the older 0/3 hour approach, covering about 111,000 presentations [1]. Death or new type 1 myocardial infarction at 30 days was rare with both strategies, under one percent in each arm, and the pooled safety estimate was imprecise enough that no difference could be confirmed either way. More surprising for anyone who championed the faster pathway, earlier retesting did not shorten emergency department length of stay, which was just over five hours in both groups, and did not clearly increase direct discharge. The trials varied considerably, which the authors read as a sign that the benefit of a faster pathway depends heavily on the local system. Their conclusion is that hospitals weighing a switch should consider broader diagnostic and organisational factors, not assume throughput gains will follow. A similar tension appears in Annals of Emergency Medicine, where Pandey and colleagues validated the GFAP and UCH-L1 blood test for mild traumatic brain injury in 104 adults at a single Level I trauma centre [2]. The test caught ten of the eleven patients with intracranial injury on CT, but it was positive in roughly four out of five patients overall, so specificity was only about one in five. False positives clustered in older patients, with a median age in the seventies. The result came back about two hours faster than CT, and the authors frame the test as a potential initial rule-out tool, while acknowledging that a small single-centre cohort cannot show whether it actually reduces scanning or length of stay. Completing the theme, Pannu and colleagues, in Internal and Emergency Medicine, applied pulmonary embolism decision tools to 408 patients after clinicians had already decided to order CT pulmonary angiography [3]. Clinically significant embolism was found in about a third of those patients. Wells discriminated better than the simplified revised Geneva score, yet roughly one in five patients labelled unlikely by either two-level score actually had a segmental or larger clot. YEARS and PEGeD stayed highly sensitive, at about 96 percent, but would have spared imaging in only around one in seven patients. The takeaway is that in a high-prevalence, already-selected population these tools offer limited imaging savings, and the counterfactual design means this is not a test of how they perform when applied up front.

Our second theme is resuscitation and procedural performance, where three papers measure how well critical interventions are actually delivered. In Resuscitation, Ramage and colleagues report TETRiS, a prospective twelve-month evaluation of resuscitative thoracotomy across 20 helicopter services and 26 major trauma centres in the United Kingdom [4]. Of 116 thoracotomies, about seven in ten were performed prehospital, and patients were mostly young men with penetrating injury. Return of circulation occurred in about 43 percent of patients, but only eight patients survived to discharge. Tamponade was the commonest indication, yet exsanguinating haemorrhage was what operators usually found, and survival was far better with tamponade than with bleeding. Patients treated in the emergency department had shorter arrests and better physiology than prehospital patients, and there were no survivors beyond ten minutes of traumatic arrest. As an observational service evaluation it cannot define selection criteria, but it adds prospective national data, including prehospital deaths, to a literature dominated by retrospective series. Also in Resuscitation, Aufderheide and colleagues examined bag-valve-mask ventilation in out-of-hospital cardiac arrest in Wisconsin, using a before-and-after design [5]. With end-tidal carbon dioxide monitoring alone, only about seven percent of ventilations met guideline targets for both rate and tidal volume, and most breaths delivered too little volume. After introduction of a real-time rate and volume feedback device, the proportion of ventilations hitting both targets rose more than thirty points, to roughly four in ten. That is a process outcome rather than survival, and the before-and-after design leaves room for confounding, but it suggests that end-tidal monitoring does not tell providers whether they are ventilating correctly. Rounding out the theme, Annals of Emergency Medicine also published a NEAR4PEM registry analysis from Jackson and colleagues on 642 pediatric intubations performed by trainees in academic pediatric emergency departments [6]. Overall first-attempt success was about 73 percent. Emergency medicine residents and pediatric emergency medicine fellows performed similarly, at around three quarters, and accounted for most attempts. Pediatric residents succeeded on the first attempt only about 58 percent of the time, and after adjustment had less than half the odds of first-pass success and roughly double the odds of an adverse airway event compared with emergency medicine residents. The pediatric resident group was small and the data are observational, but the findings bear on how training programmes allocate airway opportunities.

Our final theme covers common emergency therapeutics, and several papers this week revisit drugs we reach for daily. In Internal and Emergency Medicine, Yılmaz and colleagues ran a double-blind, three-arm trial in 251 adults with renal colic, comparing intravenous ibuprofen, dexketoprofen and paracetamol [7]. The primary outcome, halving of pain at 30 minutes, did not differ significantly, occurring in roughly two thirds to three quarters of patients in each group. After adjusting for higher baseline pain in the dexketoprofen arm, dexketoprofen produced the lowest pain scores and lower adjusted odds of needing rescue fentanyl than paracetamol. Those are secondary and adjusted findings from a single unregistered trial, so they support rather than establish a ranking, with paracetamol trailing the two anti-inflammatories. Magnesium is the subject of a narrative review in the American Journal of Emergency Medicine by Peacock and Singer on adult acute asthma [8]. Pooled data suggest intravenous magnesium modestly reduces admissions, with roughly fourteen patients treated to prevent one admission, but the largest single trial found a similar reduction that was not statistically significant, along with more minor adverse events. Nebulised magnesium showed no convincing benefit in adults. The authors characterise the intravenous benefit as modest and uncertain, consistent with guideline positioning as a second-line adjunct in severe refractory exacerbations. Also in the American Journal of Emergency Medicine, Royce and colleagues present a scoping review of subcutaneous insulin for mild-to-moderate diabetic ketoacidosis, drawing on 13 studies [9]. The evidence suggests similar time to resolution compared with intravenous insulin in selected patients, alongside less intensive care use and shorter emergency department stays. The studies were small, mostly single-centre, protocol-heterogeneous, and excluded higher-risk patients, so this is an implementation framework built on early evidence rather than definitive proof. Finally, magnesium returns in Prehospital Emergency Care, where Maffetone and colleagues analysed nearly 40,000 emergency medical services responses to pregnant patients beyond 20 weeks [10]. About one in 32 patients had persistent severe hypertension, yet fewer than one percent of those patients received both magnesium and an antihypertensive, the combination endorsed in new 2025 prehospital guidance. This pre-guideline benchmark comes from documentation data and cannot capture every clinical nuance, but the gap is striking.

If you only have time for one paper this week, make it the Lancet individual patient data meta-analysis of 0/1 hour versus 0/3 hour troponin pathways [1]. It reopens a question many departments considered settled, showing that the faster pathway's promised throughput gains did not materialise on average across the randomised evidence.

Here is what this week's evidence adds up to in Emergency Medicine. First, faster rule-out tests do not automatically translate into faster departments: pooled trial data show no average reduction in length of stay with the 0/1 hour troponin pathway, and the brain injury biomarker and pulmonary embolism data likewise suggest modest real-world sparing, though the biomarker evidence remains single-centre and early. Second, prospective national thoracotomy data reinforce that survival is concentrated in tamponade and short arrest durations, an observational finding that informs but does not define selection. Third, ventilation quality during cardiac arrest appears poor without real-time volume feedback, though improvement so far is shown only in process measures. Fourth, for renal colic and asthma the evidence favours anti-inflammatories over paracetamol and keeps intravenous magnesium as a modest, uncertain adjunct, while subcutaneous insulin for mild ketoacidosis rests on small studies awaiting larger confirmation. And fifth, prehospital treatment of severe hypertension in pregnancy started from a very low baseline, setting a benchmark against which the new guidelines can be measured.

That's your roundup for This Week in Emergency Medicine. The full transcript and references are available on the episode page. 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 if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

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