Michael Sayre

EMS Medical Director, Emergency Physician, Board Member @ResuscAcademy, Opinions not policy of my employer, #UWashEMS

[JAMA @JAMA_current]

Among comatose survivors of out-of-hospital #CardiacArrest treated with #TherapeuticHypothermia at 33 °C, increasing cooling duration did not improve neurological outcomes or mortality.

https://t.co/SIYk6v7STH

[Jakob Nikolas Kather @jnkath]

Our paper "Towards autonomous medical artificial intelligence agents" is out in @Nature today! It was led by @Dykex6 who did the real work here.

We present MIRA, an autonomous AI agent that runs end to end through a clinical patient case. We build it on top of LLMs and provide it with scaffolding, tools, and most importantly, a thorough evaluation against human doctors. We show that our agent can solve a clinical case in a multistep process. It takes the history, orders labs and scans, selects medications, decides on procedures, and triages for admission.

For me the most important part is that we can now benchmark humans against AI for the whole decision pathway, not just the end result. Our evaluation shows very good results, almost no severe errors, these AI agents perform just really well on difficult medical cases.

As we move now to implement this in real clinical routine, and run clinical trials about it, we must first fix the infrastructure in our hospitals (and also ensure we still have access to the underlying LLMs...).

Kudos also to our colleagues from Google and Deepmind who published a very similar study back-to-back with us! (https://t.co/mnCsdMCEyE) We are happy that with <<0.01% of Google's R&D budget we can contribute to this space together.

Thanks to all co-authors, our institutions @UniHeidelberg @tudresden_de and also the great @EricTopol for covering us in his newsletter. Congratulations to Dyke and the whole team!

Full text link: https://t.co/eyzKZYYFSs

[Connor Boyack 📚 @cboyack]

https://t.co/4SMw7birVU

[Julio La Torre, MD @jclatorre]

I recently took @tszzl and @RichardHanania (true story lolz-and both advisors to my new venture in full disclosure) to one of the countries largest academic medical centers.

@tszzl has since wrote a viral X thread thats worth reading again https://t.co/2LCP2VOVdg

It speaks for itself. I'll just add if he's surprised by the missing data .....he'd be further surprised by the data we do capture: fragmented, delayed, and built for billing, not clinical insight. It tells us very little about human disease and how to truly treat it.

Epic Systems is at the center of this monumental failure.

Simply put: AI cannot reach its full potential in medicine while Epic Systems controls the underlying data architecture. The ideas are fundamentally oppositional at their core.

I explain why Epic's culture, regulatory inertia, and the staggering $1B sunk costs of its clients do not guarantee its dominance but if anything bring a slow march into irrelevance.

[Circ: PHO @CircOutcomes]

Mixed-methods study protocol to evaluate whether Resuscitation Academy training for EMS agencies improves cardiac arrest survival @saketgirotra @sayremr @saintlukeskc https://t.co/zTQk6V3l5S

[Cardiac Arrest Registry to Enhance Survival (CARES @CARESRegistry]

From 8.2% to 25.0% Utstein survival — Detroit's transformation in OHCA outcomes is turning heads nationwide. CARES data helped drive the change.
https://t.co/7aSTucvmSF

[NEJM @NEJM]

Presented at #ESOC2025:

Among 550 patients with stroke due to large-vessel occlusion who had presented within 4.5 hours after onset, 90-day functional outcomes were better with intravenous tenecteplase before thrombectomy than with thrombectomy alone.

Read the full BRIDGE-TNK trial results: https://t.co/jxq5kfKc1h

@ESOstroke

[Ken Milne MD @TheSGEM]

ICYMI:
Ketamine vs midazolam for pre-hospital agitation?

https://t.co/5TFl4HRkRU

@hp_ems @EMS1 @EMSWorldOFCL @MLPS911 @ACEPNation

[Anna Edgerton @annaedge4]

What if an Amazon delivery van could also help save lives? Scoop here, with @BenoitBerthelot @mattmday, on Amazon's experiment to get more AEDs into the residential areas where most heart attacks happen:

https://t.co/eYgtXGA0AS via @business @technology

[Nick Johnson @NickJohnsonMD]

Congrats to @jesunshine and team on developing this impressive technology that will certainly improve detection of unwitnessed cardiac arrest.

Nature paper here: https://t.co/19In4mYPeT

@uwmnewsroom @UWAnesthPainMed

[Hans van Schuppen @HansvanSchuppen]

Newsflash: “refractory” #VF is not really that refractory. In patients needing three shocks at rhythm checks, successful VF termination occurred in one or more shocks in 95%. In other words, it is recurrent VF. Check out our new @ARREST_study in the #EHJ: https://t.co/yDdX8J6RVp

[Michael Sayre @SayreMR]

@HansvanSchuppen @ARREST_study @HansvanSchuppen I love the Sankey chart. It took me a minute to realize that all patients on the chart were in VF at the end. On an annual population basis, how many patients were in constant VF through 3 shocks? My sense is that we see one or two a year in Seattle, pop 750k.

[Robert Herman, MD, PhD @RobertHermanMD]

Powerful Medical wins Best Overall and Best Science Pitch at the #AHA2024 Health Tech Competition!

Thank you for this wonderful recognition @American_Heart.

[AnnalsofEM @AnnalsofEM]

Hot off the Press: A Comparison of Ketamine to Midazolam for the Management of Acute Behavioral Disturbance in the Out-of-Hospital Setting https://t.co/5qhrTTA7Re

[Catherine R. Counts @CatherineCounts]

No difference in airway management (EMS or ED) for cases of acute behavioral disturbance treated with either midazolam or ketamine.

A high functioning EMS system can safety deliver ketamine to agitated patients. #UWashEMS https://t.co/oJYnoI7CsF

[Tony Walker ASM @agwalker01]

📢 Call for Engagement: Neonatal Global Resuscitation Alliance (Neonatal-GRA)

The Neonatal Global Resuscitation Alliance Leadership Team would like to invite interested organisations and individuals to solicit their feedback on a 10-step approach to improving neonatal resuscitation.

The broad objective of the Neonatal-GRA is to ensure knowledge and experience gained in each neonatal resuscitation setting can be shared and used by others in a truly global fashion to support neonatal resuscitation implementation and performance and improve outcomes worldwide.

The Neonatal-GRA will host a virtual meeting from 12:00-14:00 EST on September 24, 2024 to discuss the drafted 10-step approach.

You can register for the virtual meeting using this link: https://t.co/4lUO1dzWEX

Access the draft of the 10-step approach and provide your feedback using this link: https://t.co/8Ido9EnPUB

#ItTakesASystemToSaveALife

[Matt Sztajnkrycer @NoobieMatt]

Evidence is not mounting. There is already ample safety data.

It is time to allow this drug in the US!

[UW Medicine Newsroom @uwmnewsroom]

"Dr. Copass was a beacon of excellence, setting high standards for himself and those around him,” said Dr. Tim Dellit, CEO of @UWMedicine. "He was admired for always being ready for an emergency, whether it was carrying a radio to go to the scene of an accident himself, providing medical advice to paramedics, personally attending to patients or mentoring colleagues.” @harborviewmc @AirliftNW @UWashEM @MedicOneFndn @SayreMR
https://t.co/0NDu2i4aaN

[Michael Sayre @SayreMR]

Myth busting results found in this investigation. @larswik1 and Rose Yin did great work.

[Resuscitation @ResusJournal]

Recovery of arterial blood pressure after chest compressions pauses in patients with out-of-hospital cardiac arrest

https://t.co/VGJ1rrLCvB

#Resuscitation

[Ambulance Victoria Research @VACAR_AV]

😍 Proud of our MICA paramedics for another successful launch of a world-first clinical trial: Today the #PANDA trial enrolled its 1st patient! The #PANDA trial will end the long-standing debate around the optimal vasoactive agent for cardiogenic shock in the prehospital setting