Abstract 183: Prevalence and Development of Clinical States During Resuscitation from In-Hospital Cardiac Arrest

Abstract

Background During resuscitation of a patient in cardiac arrest the clinical state determines the management. The state will change either spontaneously (e.g. pulseless electrical activity [PEA] - asystole [ASY]), or due to intervention (e.g. ventricular fibrillation/-tachycardia [VF/VT] - return of spontaneous circulation [ROSC] following shock). The aim of this study was to describe and analyze the development of clinical states in patients receiving advanced life support (ALS). Methods Defibrillator files from 261 in-hospital cardiac arrests at the University of Chicago Hospital (Il, USA) and St.Olav University Hospital (Trondheim, Norway) were analyzed. The clinical states ASY, PEA, VF/VT, ROSC and death were annotated along the time axis. The average transition probabilities over all 2-minute periods (recommended CPR “loop” duration), were estimated using a Markov model. Results The figure shows the development of clinical states during the first 30 minutes of ALS. The inserted matrix shows the estimated probabilities of moving from one state (rows) to another (columns), during the next two minutes of ALS. E.g. a patient in PEA has a probability of 0.16 of gaining ROSC during (any) two minutes. Relapse to PEA from ROSC has a probability of 0.17. The prevalence of ROSC continues to increase until about 25 minutes of ALS, where about 45% of the patients have obtained ROSC and 35% have been declared dead. Discussion We provide a basic description of the dynamics of resuscitation from in-hospital cardiac arrest. Understanding these dynamics enables the provider to focus on important transitions (e.g. relapse from ROSC), and may provide a basis for improved tailored therapy.