To address the problem of out-of-hospital cardiac arrest, researchers at Osaka Metropolitan University have developed a novel scoring method that uses only data available from pre-hospital resuscitations to accurately predict neurological outcomes and enable doctors to make more informed decisions once a patient arrives at the hospital.
WHY IS THIS IMPORTANT
Following out-of-hospital transport (OHCA), adverse neurological outcomes occur, ranging from disability to death.
Developed by researchers at Osaka Metropolitan University, the novel R-EDByUS model relies on five key variables to underpin it – age, time to return of unplanned circulation or time to hospital, lack of bystander CPR, whether the person witnessed the cardiac arrest, and initial heart rhythm.
This model accurately predicted the neurological prognosis of cardiac arrest on arrival at hospital, consistent with tests published in the current issue.
“Our hypothesis is that a scoring model consisting solely of prehospital factors in the algorithm would be easy to use and could predict outcome at the earliest stage of care,” the researchers said.
They took advantage of the disadvantages of the American College of Cardiology algorithm:
- Arrest without witnesses.
- Initial non-shockable rhythm.
- No cardiopulmonary resuscitation was provided to witnesses to the incident.
- Time to ROSC > 30 min.
- Cardiopulmonary resuscitation in progress.
- Blood pH < 7.2, 7.
- Lactate level > 7 mmol/l.
- Age > 85 years.
- End-stage renal failure.
- Oncardiological causes in patients with cardiac arrest.
They used OHCA data collected between 2005 and 2019 from the All-Japan Utstein Registry for 942,891 adults with presumed cardiac origin. They divided the patients into two groups—those who achieved ROSC before hospital arrival or those who continued to receive CPR after arrival. They then used detailed regression-based models and simplified models to calculate R-EDByUS scores for each group.
Patients under 18 years of age, patients whose cardiopulmonary resuscitation was performed by bystanders, and several other factors were excluded.
In the prehospital ROSC group, 70.0% had an unfavorable neurologic outcome and 55.7% died. For those in whom CPR was continued by EMS after arrival at the hospital, 99.4% had an unfavorable neurologic outcome and 98.2% died.
“Our predictive model helps identify patients who are likely to benefit from intensive care while reducing unnecessary burden on those with poor prognosis,” Takenobu Shimada, a professor of medicine at the Graduate School of Medicine at Osaka Metropolitan University and the study’s lead author, said last week.
This article says the scoring model will become a valuable tool for healthcare providers, helping them quickly assess and treat patients undergoing resuscitation.
Scientists have developed online calculator found it to be uncomplicated to employ in a clinical setting and has potential for future validation.
BIGGER TREND
AI-assisted patient triage has the potential to create appropriate care channels, improve patient outcomes and experiences, and optimize resource utilization, but demanding regulations are needed to assess them, according to Piotr Orzechowski, founder and CEO of Infermedica, a medical company that uses AI for initial symptom analysis and digital patient triage.
“AI tools are not authorized to diagnose patients,” Orzechowski said in December during a talk about healthcare’s connections to artificial intelligence.
“Despite significant progress in generative AI, we must remain cautious about its practical application in healthcare,” he said.
IN THE DOCUMENT
“With this free calculation tool, one can easily estimate the probability of adverse neurological outcomes and mortality by checking each variable online, instead of performing calculations using nomograms,” the researchers said.
The HIMSS AI in Healthcare Forum will be held September 5-6 in Boston. Learn more and sign up.