18794: Will Extending Time to Neuroprognostication Improve Cardiac Arrest Outcomes

Document Type


Publication Date



Introduction: Delayed awakening is not rare after cardiac arrest resuscitation. Literature also identifies that most post cardiac arrest deaths in hospital are due to withdrawal of care.

Objective: We sought to describe the frequency of early death after resuscitation as a means of estimating the potential benefit of delayed prognostication.

Methods: We performed a retrospective analysis of all cardiac arrest admissions to acute care hospitals in one state during a 3 year period (2010-2013). Cases were identified with ICD-9 CM codes for cardiac arrest (427.5) or VF (427.41). Patient demographics, Inpatient length of stay (LOS), outcome (survival to discharge), secondary diagnoses were recorded on all patients. Length of stay was dichotomized as <=3 days (early death) or 4 or more days, and descriptive statistics calculated. To assess the relationship between hospital survival rates and rates of early death, we evaluated these variables in state hospitals with the highest quartile of cardiac arrest volumes.

Results: During the study period 4,306 patient records met the inclusion criteria, of which 1650 (38.3) survived to hospital discharge. The patients were primarily male (57.6)%, with mean (range) age was 64.0 (18-112) years. Of all patients, the mean LOS was 6.1 days, with 2247 (52.2%) dying in <=3 days. Patients with early death, when compared to those with late deaths, were older 64.2 vs 61.7 years, p < 0.001, more likely female (56.3% vs 39.3%, p <0.001), or Non VT/VF (58.4 % vs 39.8%, p< 0.001). Hospital survival rates varied from 24.4%-55.7%, and rates of early death varied from 36.5%-60.7%. Hospital early death rates were negatively associated with hospital survival rates (pearson correlation p=0.001).

Conclusion: In the time period before the AHA 2015 Guidelines, a majority of inpatient post arrest deaths occurred in <=3 days. Early deaths were associated with factors typically predictive of poor cardiac arrest outcome. We observe that high hospital survival rates were associated with low rates of early death. Whether these data reflect an expected association, or clinically important variation of practice requires further scrutiny.


American Heart Association Resuscitation Science Symposium, Anaheim CA, November 11-13, 2017.