Dennis T Ko1, Feng Qiu2, Maria Koh2, Paul Dorian3, Sheldon Cheskes4, Peter C Austin2, Damon C Scales5, Harindra C Wijeysundera6, P Richard Verbeek7, Ian Drennan8, Tiffany Ng2, Jack V Tu6, Laurie J Morrison8. 1. Institute of Clinical Evaluative Sciences, Toronto, ON, Canada; Schulich Heart Centre, Toronto, ON, Canada; Department of Medicine, University of Toronto, Toronto, ON, Canada. Electronic address: dennis.ko@ices.on.ca. 2. Institute of Clinical Evaluative Sciences, Toronto, ON, Canada. 3. Department of Medicine, University of Toronto, Toronto, ON, Canada; Sunnybrook Health Sciences Centre; Rescu Keenan Research Centre, Li Ka Shing Knowledge Institute at St Michael's Hospital, Toronto, ON, Canada. 4. Department of Medicine, University of Toronto, Toronto, ON, Canada; Sunnybrook Health Sciences Centre; Rescu Keenan Research Centre, Li Ka Shing Knowledge Institute at St Michael's Hospital, Toronto, ON, Canada; Sunnybrook Centre for Prehospital Medicine, Toronto, ON, Canada. 5. Institute of Clinical Evaluative Sciences, Toronto, ON, Canada; Department of Medicine, University of Toronto, Toronto, ON, Canada; Department of Critical Care Medicine, Toronto, ON, Canada. 6. Institute of Clinical Evaluative Sciences, Toronto, ON, Canada; Schulich Heart Centre, Toronto, ON, Canada; Department of Medicine, University of Toronto, Toronto, ON, Canada. 7. Sunnybrook Centre for Prehospital Medicine, Toronto, ON, Canada; Sunnybrook Health Sciences Centre; Rescu Keenan Research Centre, Li Ka Shing Knowledge Institute at St Michael's Hospital, Toronto, ON, Canada. 8. Sunnybrook Health Sciences Centre; Rescu Keenan Research Centre, Li Ka Shing Knowledge Institute at St Michael's Hospital, Toronto, ON, Canada.
Abstract
BACKGROUND: Many patients with out-of-hospital cardiac arrest present with pulseless electric activity (PEA) rather than shockable rhythm. Despite improvements in resuscitation care, survival of PEA patients remains dismal. Our main objective was to characterize out-of-hospital cardiac arrest patients by initial presenting rhythm and to evaluate independent determinants of PEA. METHODS: A population-based study was conducted using the Toronto Rescu Epistry database with linkage to administrative data in Ontario, Canada. We included patients older than 20 years who had nontraumatic cardiac arrests from 2005 to 2010. Multivariable logistic regression models were constructed to determine factors predicting the occurrence of PEA vs shockable rhythm vs asystole. RESULTS: Of the 9,882 included patients who received treatment, 24.5% had PEA, 26.3% had shockable rhythm, and 49.2% had asystole. Patients with PEA had a mean age of 72 years, 41.2% were female and had multiple comorbidities, and 53.4% were hospitalized in the past year. As compared with shockable rhythm, PEA patients were older, were more likely to be women, and had more comorbidities. As compared with asystole, PEA patients had similar baseline and clinical characteristics, but were substantially more likely to have an arrest witnessed by emergency medical services (odds ratio 13) or by bystander (odds ratio 3.24). Mortality at 30 days was 95.5%, 77.9%, and 98.9% for patients with PEA, shockable rhythm, asystole, respectively. CONCLUSIONS: Patient characteristics differed substantially in those presenting with PEA and shockable rhythm. In contrast, the main distinguishing factor between PEA and asystole cardiac arrest related mainly to factors at the time of the cardiac arrest.
BACKGROUND: Many patients with out-of-hospital cardiac arrest present with pulseless electric activity (PEA) rather than shockable rhythm. Despite improvements in resuscitation care, survival of PEA patients remains dismal. Our main objective was to characterize out-of-hospital cardiac arrestpatients by initial presenting rhythm and to evaluate independent determinants of PEA. METHODS: A population-based study was conducted using the Toronto Rescu Epistry database with linkage to administrative data in Ontario, Canada. We included patients older than 20 years who had nontraumatic cardiac arrests from 2005 to 2010. Multivariable logistic regression models were constructed to determine factors predicting the occurrence of PEA vs shockable rhythm vs asystole. RESULTS: Of the 9,882 included patients who received treatment, 24.5% had PEA, 26.3% had shockable rhythm, and 49.2% had asystole. Patients with PEA had a mean age of 72 years, 41.2% were female and had multiple comorbidities, and 53.4% were hospitalized in the past year. As compared with shockable rhythm, PEA patients were older, were more likely to be women, and had more comorbidities. As compared with asystole, PEA patients had similar baseline and clinical characteristics, but were substantially more likely to have an arrest witnessed by emergency medical services (odds ratio 13) or by bystander (odds ratio 3.24). Mortality at 30 days was 95.5%, 77.9%, and 98.9% for patients with PEA, shockable rhythm, asystole, respectively. CONCLUSIONS:Patient characteristics differed substantially in those presenting with PEA and shockable rhythm. In contrast, the main distinguishing factor between PEA and asystole cardiac arrest related mainly to factors at the time of the cardiac arrest.