Factors affecting survival after prehospital asystolic cardiac arrest in a Basic Life Support-Defibrillation system

Objectives: Previous studies have shown a low but meaningful survival rate in cases of prehospital cardiac arrest with an initial rhythm of asystole. There may be, however, an identifiable subgroup in which resuscitation efforts are futile. This study identified potential field criteria for predicting 100% nonsurvival when the presenting rhythm is asystole in a Basic Life Support-Defibrillation (BLS-D) system. Methods: This prospective cohort study, a component of Phases I and II of the Ontario Prehospital Advanced Life Support (OPALS) Study, was conducted in 21 Ontario communities with BLS-D level of care, and included all adult arrests of presumed cardiac etiology according to the Utstein Style Guidelines. Analyses included descriptive and appropriate univariate tests, as well as multivariate stepwise logistic regression to determine predictors of survival. Results: From 1991 to 1997, 9899 consecutive cardiac arrest cases with the following characteristics: male (67.2%), bystander-witnessed (44.7%), bystander CPR (14.2%), call-response interval (CRI) ≤ 8 minutes (82%) and overall survival (4.3%) were enrolled. Of 9529 cases with available rhythm strip recordings, initial arrest rhythms were asystole in 40.8%, pulseless electrical activity in 21.2% and ventricular fibrillation or ventricular tachycardia in 38%. Of 3888 asystolic patients, 9 (0.2%) survived to discharge; 3 of these cases were unwitnessed arrests with no bystander CPR. There were no survivors if the CRI exceeded 8 minutes. Logistic regression analysis demonstrated that independent predictors of survival to admission were "CRI in minutes" (odds ratio [OR] = 0.87; 95% confidence interval [CI], 0.77-0.98) and "bystander-witnessed" (OR = 2.6; 95% CI, 1.5-4.4). Conclusions: In a BLS-D system, there is a very low but measurable survival rate for prehospital asystolic cardiac arrest. CRIs of over 8 minutes were associated with 100% nonsurvival, whereas unwitnessed arrests with no bystander CPR were not. These data add to the growing literature that will help guide ethical decision-making for protocol development in emergency medical services systems.