Novel Approaches to Risk Stratification of In-Hospital Cardiac Arrest

AbstractPurpose of ReviewThis review aims to discuss strategies for patient risk stratification of in-hospital cardiac arrest and their challenges and limitations.Recent FindingsIn-hospital cardiac arrest is a significant cause of inpatient mortality, but survival to discharge rates remain low and have not significantly improved in the last three decades. As most patients are in a monitored setting and commonly show clinical deterioration prior to cardiac arrest, early intervention is thought to be the best way to both prevent and improve survival from cardiac arrest. However, detection of an impending cardiac arrest has proven to be particularly challenging.SummaryContemporary methods often rely on score-based early warning systems based on intermittently collected vitals and laboratory values that have seen some success in preventing in-hospital cardiac arrest but are neither sensitive enough to detect impending arrest without significant false alarms nor have the ability to uncover the underlying cause of an imminent arrest. Machine learning –derived early warning systems that utilize temporal trends in vital signs as well as continuous telemetry data are currently being developed. These new approaches show promise in addressing the issue of sensitivity and positive predictive value but require further clinical research and technologi cal advancements.

Source: Current Cardiovascular Risk Reports - January 31, 2021 Category: Cardiology Source Type: research