Cardiac arrest witnessed by emergency medical services personnel: Descriptive epidemiology, prodromal symptoms, and predictors of survival

Study objective: The Utstein guidelines recommend that emergency medical services (EMS)-witnessed cardiac arrests be considered separately from other out-of-hospital cardiac arrest cases. The objective of this study was to assess EMS-witnessed cardiac arrest and to determine predictors of survival in this group. Methods: This prospective cohort included all adults with an EMS-witnessed cardiac arrest in the 21 communities of the Ontario Prehospital Advanced Life Support (OPALS) study. Systems provided a basic life support with defibrillation (BLS-D) level of care. Case and survival definitions followed the Utstein style. Descriptive and univariate methods (χ² and t test) were used to characterize EMS-witnessed cardiac arrest. Multivariate logistic regression was undertaken to assess predictors of survival to hospital discharge. Results: From January 1, 1991, to December 31, 1996, there were 9,072 cardiac arrest cases in the study communities. Of these, 610 (6.7%) were EMS-witnessed. The majority had pre-existing cardiac or respiratory disease (81.5%) and experienced prodromal symptoms before EMS personnel arrived (91.4%). An initial rhythm of pulseless electrical activity was present in 50.1% of the patients, ventricular fibrillation/ventricular tachycardia in 34.2%, and asystole in 15.7%. Survival to discharge was 12.6%. Multivariate analysis identified the following as independent predictors of survival (odds ratio with 95% confidence intervals [Cls]): Nitroglycerin use before EMS arrival: 2.3 (95% Cl 1.2 to 4.5), prodromal symptoms of chest pain: 2.5 (95% Cl 1.4 to 4.5) or dyspnea: 0.5 (95% Cl 0.3 to 1.0), and unconsciousness on EMS arrival: 0.5 (95% Cl 0.2 to 0.9). Patients with chest pain were more likely than dyspneic patients to experience ventricular fibrillation/ventricular tachycardia (62% versus 17%, P < .0001), and were 5 times more likely to survive (30.6% versus 6.3%, P < .0001). Conclusion: EMS-witnessed cases are clearly an important subset of out-of-hospital cardiac arrest and are characterized by 2 distinct symptom groups: Chest pain and dyspnea. These symptoms are important predictors of survival and may also help determine underlying mechanisms before patient collapse. A later phase of the OPALS study will prospectively evaluate the impact of out-of-hospital advanced life support on the survival of victims of EMS-witnessed cardiac arrest.