This Week in Emergency Medicine — May 21, 2026

Welcome to This Week in Emergency Medicine. This week we're covering 10 notable papers spanning new findings in resuscitation, strategies for optimizing cardiac arrest care, and important updates on improving our clinical systems and workflows. Let's dive in.

Our first theme covers major intervention trials that challenge some of our current practices and offer new directions for care. We'll start with a highly anticipated pragmatic trial from The New England Journal of Medicine looking at prehospital resuscitation for trauma [5].

The Study This was a multicenter, phase 3, cluster-randomized trial across 44 air medical bases. They compared giving up to 2 units of type O whole blood versus standard blood components—that is, plasma, red cells, or both—for prehospital transfusion in trauma patients with hemorrhagic shock. The primary outcome was 30-day all-cause mortality.

Results Among over 1000 eligible patients, 30-day mortality was 25.9% in the whole-blood group and 20.5% in the component group. After adjustment, this difference was not statistically significant. The odds ratio was 1.24, with a confidence interval that crossed one, indicating no clear benefit for whole blood. There were also no substantial differences in adverse events. An observational substudy found no difference in mortality based on the storage age of the whole blood.

Conclusions In injured patients with hemorrhagic shock, prehospital transfusion with whole blood did not result in lower 30-day mortality compared to using blood components. This is a crucial negative trial that may temper the widespread push towards prehospital whole blood, suggesting that a well-run component therapy program can achieve similar outcomes.

Also in The New England Journal of Medicine, another study challenges a common practice, this time in pediatrics [6]. Investigators looked at using azithromycin for preschoolers with moderate-to-severe wheezing. Observational data has suggested a link between certain bacteria and recurrent wheezing, creating a rationale for antibiotic use. This multicenter randomized trial assigned over 800 children aged 18 to 59 months to receive either azithromycin or placebo for five days. The primary outcome was symptom severity over those five days. The trial was stopped early for futility. There was no significant difference in symptom scores between the azithromycin and placebo groups, even when they analyzed the subgroup of children who tested positive for pathogenic bacteria. While azithromycin did clear the bacteria more effectively, it did not translate into clinical improvement. Secondary outcomes like length of stay and return visits were also similar. This provides strong evidence that azithromycin does not reduce symptom severity in young children with acute wheezing and should not be used for this indication.

In contrast to those two negative trials, a study in The Lancet provides a new, positive treatment direction for ischemic stroke [8]. The TAPIS trial investigated adding early dual antiplatelet therapy, or DAPT, to intravenous thrombolysis. This was a randomized, double-blind, placebo-controlled trial in China involving patients with moderate ischemic stroke. Patients received either aspirin plus ticagrelor or aspirin plus placebo within 6 hours of stroke onset. The primary outcome was an excellent functional outcome, defined as a modified Rankin Scale score of 0 to 1, at 90 days. The results were positive. 68.7% of patients in the early DAPT group achieved an excellent outcome, compared to 62.0% in the placebo group. This represented a statistically significant 11% relative increase in the likelihood of an excellent outcome. Importantly, there was no significant difference in the primary safety outcome of symptomatic intracranial hemorrhage, which occurred in less than 1% of both groups, although the wide confidence intervals mean a small increased risk cannot be completely ruled out. For patients with moderate ischemic stroke receiving thrombolysis, this study supports initiating DAPT with ticagrelor within 6 hours of onset.

Next, we turn to a series of papers focused on optimizing cardiac arrest care, from the bystander to the ECPR center.

First, a proof-of-concept study in JAMA Internal Medicine explores the potential for artificial intelligence to improve bystander CPR [7]. Researchers evaluated how well widely available AI models like ChatGPT could provide CPR instructions in simulated emergency scenarios. While existing models performed reasonably well, the team then developed a purpose-built AI CPR instructor agent called ChatCPR. In simulated scenarios, this specialized agent achieved 100% adherence to both minimally viable and maximally effective CPR criteria. More impressively, when tested against transcripts of real 911 calls, the AI agent achieved nearly perfect adherence to guidelines, representing a significant improvement over the instructions provided by human dispatchers. This suggests AI-enabled CPR instruction is a promising and scalable tool for improving the quality of bystander CPR, though further real-world validation is needed.

Moving from the bystander to the in-hospital team, a quality improvement study in the journal Resuscitation looked at how to improve CPR quality in a pediatric emergency department [2]. The department's baseline compliance with AHA recommendations was low, particularly for compression depth at just 14%. Their intervention was a simulation-based curriculum for CPR coaches, supplemented with compressor training and metronome-guided feedback. The results were impressive. Median depth compliance for all ages increased from 14% to 61%. For infants, mean depth increased significantly. Rate compliance also improved, rising from 69% to 80%, all while maintaining a high chest compression fraction. This study demonstrates that a dedicated, multimodal training program centered on a CPR coach can tangibly improve pediatric CPR quality.

Finally, for the sickest cardiac arrest patients, another paper in Resuscitation addresses access to ECPR for rural populations [3]. The FLIGHT-to-ECPR study prospectively evaluated a helicopter-facilitated pathway for transporting patients with refractory VF/VT out-of-hospital cardiac arrest to a hospital-based ECMO center. They compared outcomes to a matched cohort who arrived via standard ground transport. The study found that the helicopter pathway was feasible. Among 27 patients who were cannulated for ECPR via the helicopter pathway, the rate of survival to discharge with a favorable neurologic status was 25.9%. This was identical to the 25.9% favorable survival rate in the matched ground-transport ECPR cohort. The study concluded that a helicopter-based strategy can safely expand rural access to advanced resuscitation, suggesting that with optimized systems, geography alone doesn't have to be a barrier to ECPR.

Our final theme is on improving clinical systems and workflows. Four papers this week offer insights into diagnostics, crowding, education, and prehospital care for opioid use disorder.

First, from the Emergency Medicine Journal, a study confirms what many of us believe about point-of-care ultrasound [1]. This prospective, blinded study compared emergency physician-performed compression ultrasound to radiologist-performed Doppler ultrasound for diagnosing lower-extremity DVT. With radiology as the gold standard, the emergency physician ultrasound was found to have a sensitivity of 97% and a specificity of 99.3%, with an excellent negative predictive value of 99.6%. The interobserver agreement was excellent. Critically, the median time-to-diagnosis was just 1 hour for the bedside ultrasound, compared to 4 hours for the radiology-performed study. These findings strongly support the use of point-of-care ultrasound by trained emergency physicians for the rapid and accurate diagnosis of DVT.

Next, a paper in The American Journal of Emergency Medicine provides a crucial insight into the problem of ED crowding [4]. This large, multicenter study from Brazil analyzed data from 160 hospitals to distinguish between two key components of crowding: occupancy, which is the total number of patients divided by beds, and boarding, which is the proportion of ED patients who have already been admitted. They found that while crowding was persistent, only one of these metrics was associated with poor outcomes. A higher boarding proportion was significantly associated with an increase in early in-hospital mortality, with each 10% increase in boarding linked to a 14% higher rate of death within 24 hours. In contrast, the overall occupancy ratio was not associated with mortality. This finding suggests that our efforts to combat crowding should be laser-focused on improving patient flow and reducing exit block.

From JAMA, we have a fascinating randomized clinical trial on a novel educational tool [10]. The study tested whether a 'serious game'—a purpose-driven video game—could improve emergency physician adherence to trauma triage guidelines for older adults, a group where undertriage is common. Eight hundred physicians at non-trauma centers were randomized to either receive the game-based training or usual education. The results showed that physicians in the game-based training group had a significantly lower rate of undertriage. 49% of their severely injured older adult patients were undertriaged, compared to 57% in the control group. This 7% absolute reduction is a meaningful improvement. The game did not affect overtriage or a composite clinical outcome of death or readmission. This trial demonstrates that a theory-based serious game can be an effective, low-cost method to improve physician adherence to critical, time-sensitive guidelines.

Finally, the journal Prehospital Emergency Care published a new position statement from the National Association of EMS Physicians on the role of EMS in addressing the opioid crisis [9]. The statement emphasizes that opioid use disorder is a treatable medical disease and that EMS is uniquely positioned to initiate care. Key recommendations include: ensuring EMS scope of practice allows for the administration of buprenorphine; establishing low-barrier protocols for prehospital buprenorphine initiation; developing pathways for patient navigation to alternate destinations or for treatment-in-place; and collaborating on harm reduction strategies like naloxone distribution. This document provides a clear framework for EMS systems to evolve beyond simple overdose reversal and become a critical entry point for long-term treatment.

If you only have time for one paper this week, make it the TAPIS trial from The Lancet [8]. It provides strong, positive evidence that for patients with moderate ischemic stroke receiving thrombolysis, adding early dual antiplatelet therapy with ticagrelor improves the odds of an excellent functional outcome at 90 days. This is a directly practice-changing finding for a common and critical presentation.

Here are the key takeaways from this week in Emergency Medicine.

First, in patients with moderate ischemic stroke treated with thrombolysis, adding early dual antiplatelet therapy with ticagrelor within 6 hours improves 90-day functional outcomes, as shown in the TAPIS trial [8].

Second, two major trials in The New England Journal of Medicine provide strong negative evidence. Prehospital whole blood did not improve 30-day mortality compared to components in trauma [5], and azithromycin does not improve symptoms in preschoolers with acute wheezing [6].

Third, when advocating for resources to fight crowding, focus on exit block. A large study found that the proportion of boarded inpatients, not overall ED occupancy, is the metric associated with increased early in-hospital mortality [4].

Fourth, for suspected DVT, a trained emergency physician-performed compression ultrasound is not only highly accurate, with a sensitivity of 97% and specificity over 99%, but it can also reduce the time to diagnosis from 4 hours to 1 hour [1].

Fifth, innovative approaches can improve care. A 'serious game' was shown in a randomized trial to significantly reduce the undertriage of severely injured older adults [10], and a purpose-built AI agent shows promise for delivering higher-quality CPR instructions than human dispatchers [7].

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.

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