Abstract 194: ST Changes on Continuous Telemetry Monitoring Before In-Hospital Cardiac Arrests
In-hospital cardiac arrests (CA) remain a significant problem in the United States, affecting up to 750,000 patients annually. No more than 30% survive to discharge - this has improved little in the past 25 years, and is unlikely to improve further without major changes in treatment or better identification of at risk patients. Hence, there is significant interest in predicting impending CA to allow for intervention and possibly improve survival. In this retrospective study, we examine the characteristics of ST changes prior to CA to evaluate their utility in predicting CA.
We evaluated all “Code Blue” events at UCLA Ronald Reagan Medical Center, a 520 bed tertiary care hospital, from April 2010 to April 2012. Inclusion criteria were: age ≥ 18, actual cardiac arrest, continuous telemetry data available for at least 3 hours immediately prior to CA. Up to 24 hours of telemetry data prior to CA was obtained and analyzed for each case. The arrhythmic mode and likely underlying cause, and presence/absence/duration of ST changes prior to CA were assessed for each case.
Of 65 cases meeting inclusion criteria, 19(29%) had progressive ST changes leading up to CA. The most likely cause was respiratory failure in 5(26%), multiorgan failure in 7(37%), acute cardiac failure (pulmonary embolism or myocardial infarction) in 5(26%), metabolic acidosis in 1(5%), and hemorrhagic shock in 1(5%). The arrhythmic mode of CA was pulseless electrical activity in 14(73%), bradycardic in 4(21%) and torsade de pointes in 1(5%). ST changes were detectable on telemetry monitoring in 6(32%) more than 60 minutes prior to CA, in an additional 6(32%) between 10 and 60 minutes prior to CA, and only in the last 10 minutes in a further 6(32%).
ST changes leading up to CA occurred in one third of cases in this small series, mainly in PEA and bradycardic arrests. These were associated with multiple underlying medical conditions that may lead to cardiac hypoperfusion such as multiorgan failure on pressors, severe hypoxia, pulmonary embolism and myocardial infarction. In about two thirds of the cases in which ST changes were detectable, they were evident more than 10 minutes prior to onset of arrest, which may allow sufficient time for rapid intervention to prevent CA.
- © 2012 by American Heart Association, Inc.