LO018: The utility of ECG characteristics as prognostic markers in pulseless electrical activity arrests: a retrospective observational cohort study

Abstract

Introduction: Compared to pseudo-pulseless electrical activity (PEA with myocardial contractions present), true PEA is hypothesized to carry a poorer prognosis and to show bradycardia and a wide QRS complex on ECG. Our objective was to study the predictive potential of ECG characteristics on survival to hospital discharge (SHD) for out-of-hospital cardiac arrest (OHCA) patients with PEA initial rhythm. Methods: We studied a cohort of OHCA patients prospectively enrolled between Sept. 2007 and Oct. 2009 at the Ottawa/OPALS site (13 cities, 7 EMS, and 6 Fire services) of the ROC PRIMED study. We included adult (≥ 18) non-traumatic OHCA with PEA initial rhythm where resuscitation was attempted, and for which ECG characteristics were available. We measured mean heart rate (HR), mean QRS interval, and presence of P waves (each with kappa agreement) using the first six QRS complex available. We report patient and system characteristics using descriptive statistics and determined the impact of ECG characteristics (HR, QRS width, P waves) on return of spontaneous circulation (ROSC) and SHD using multivariate regression analysis. Results: Demographics of 332 included cases were: mean age 71.7; male 58.4%; home residence 76.5%; bystander witnessed 56.3%; bystander CPR 28.5%; interval from dispatch to paramedic arrival 6min:24sec; ROSC at ED arrival 26.5%; SHD 5.4%. Survivors had higher mean HR (66.1 vs. 52.0 bpm, p=0.83; kappa=0.69) and shorter mean QRS intervals (108.3 vs. 129.6 ms, p=0.01; kappa=0.74) compared to non-survivors. Presence of p waves could not reliably be ascertained (kappa=0.35). Predictors of ROSC were: ALS paramedic on scene (AdjOR=8.90, 95%CI 1.11-71.41; p=0.04), successful intubation (AdjOR=3.35, 1.75-6.39; p=0.0002), and use of atropine (AdjOR=0.27, 0.14 - 0.50; p<0.0001). Predictors of survival were: location of arrest (AdjOR=1.49, 1.11 - 1.99; p=0.007), and use of atropine (AdjOR=0.06, 0.02-0.22; p<0.0001). Despite various cutoff explorations, ECG characteristics were not predictive of ROSC or survival in multivariate analyses. Survivors had HR as low as 6 bpm and QRS as wide as 357 ms. Conclusion: Early ECG characteristics could not predict ROSC or SHD in a population of OHCA PEA victims, and should not be used to terminate resuscitation efforts. Atropine administration was consistently associated with decreased likelihood of ROSC and survival.