Prediction of Outcome in Out-of-Hospital Cardiac Arrest — PREDOHCA  

Cardiac Arrest Clinical Trial

Official title: Prediction of Outcome in Out-of-Hospital Cardiac Arrest

Status Not yet recruiting

Clinical Trial Summary

In the course of prehospital respiratory and circulatory arrest, approximately 1000 persons are resuscitated by cardiopulmonary resuscitation in Upper Austria every year. Despite constant further development of methods, equipment and continuous training of the rescue and emergency medical teams working on site, the majority of patients who have to be resuscitated prehospital still die. However, even patients whose circulatory function can be restored during prehospital resuscitation (Return of Spontaneous Circulation, ROSC) require intensive medical care for days to weeks and often find it very difficult to return to a normal, independent life. The success of resuscitation measures depends on the quality of the resuscitation performed as well as on patient-specific factors. Evaluation scales such as the Cerebral Performance Category score (CPC) allow a posteriori assessment of resuscitation success. Nowadays, it is very difficult to estimate the outcome of resuscitation a priori. In many cases, it is not at all clear at the beginning of the treatment pathway whether the individual patient is expected to have an unfavorable prognosis in the context of respiratory arrest or whether a restitutio ad integrum is possible. Thus, the decision to continue or discontinue resuscitation can only be made on the basis of an individual physician's assessment. In addition to the primary concern of stopping resuscitation too early, there is also the risk that medical resources are used beyond the normal level after resuscitation without expecting a successful outcome. Estimating and categorizing the subsequent outcome is difficult and emotionally stressful for the treating team in the acute situation. Some factors that influence outcome are now known: As cerebral hypoperfusion increases, the probability of survival decreases sharply with each passing minute. In this context, potentially reversible causes have been identified in different works, allowing causal therapy to improve neurological outcome. In addition to the most important therapy bridging hypoperfusion, chest compression, with the aim of ensuring minimal perfusion of the brain, immediate defibrillation should be mentioned in particular, which now allows medical laypersons to use defibrillators as part of the Public Access Defibrillation Network. Despite all efforts, however, it is not yet possible to make reliable statements about the probable outcome of persons with respiratory and circulatory arrest with a high degree of certainty in a large number of cases at an early stage. Artificial intelligence refers to the ability of machines to perform cognitive tasks, such as recognizing objects in images and classifying them. For a long time, many processes were too complex to explore through sufficient computing power, storage capacity, and understanding. More recently, however, technological advances have brought machine learning (ML) and the constructs behind it, including those based on so-called neural networks (known since about 1950), back to the fore. Not only the development of theoretical models, but after extensive testing also devices applicable in daily routine operation are available. Modern machine learning methods are enabling a variety of new approaches to assessing operations, including modeling complex systems and finding relationships between models.

Clinical Trial Description

n/a

Related Conditions & MeSH terms

• Cardiac Arrest
• Heart Arrest
• Out-of-Hospital Cardiac Arrest

• NCT number: NCT06030986
• Study type: Observational
• Source: Kepler University Hospital
• Contact: Thomas Tschoellitsch, MD
• Phone: +435768083
• Email: thomas.tschoellitsch@jku.at
• Status: Not yet recruiting
• Start date: May 31, 2024
• Completion date: December 31, 2024