Abstract 322: Post-cardiac Arrest Care Variations in Michigan Hospitals and Their Impact on Survival

Jaemin Song,David Berger,Nai-Wei Chen,James Paxton,Joseph Miller,James Pribble,Robert A Swor

DOI: https://doi.org/10.1161/circ.142.suppl_4.322

IF: 37.8

2020-11-17

Circulation

Abstract:Introduction: Targeted temperature management (TTM) is an important component of post cardiac arrest care. Current guidelines recommend deferring prognostic decisions for 72 hours post TTM. Most non survivors have care withdrawn and some hypothesize that early withdrawal of care adversely impacts survival. Our objective is to characterize hospital length of stay (LOS) post cardiac arrest, its relationship to survival and variation by hospitals using a statewide registry. Methods: We probabilistically linked out of hospital (OHCA) patients from 2014-2017 in a statewide Cardiac Arrest to Enhance Survival Registry (CARES) and the Michigan Inpatient Database (MIDB). We included resuscitated adult OHCA patients. To assess variation in LOS (arrival-death or discharge), those with > 30 subjects/hospital were included. We abstracted subject demographics, arrest characteristics, in hospital care, and hospital LOS. Our outcome is survival to discharge. We define early death as hospital LOS< 72 hours (LOS<72). We calculated Pearson’s correlation coefficient (r) to characterize the relationship of LOS<72 and patient survival. Results: We included 3,644 patients from 38 hospitals. They were predominately male (2160, 59.3%), mean age 62.4, 1716 (47.1%) received TTM, mean LOS was 6.1 days, and 1078 (29.6%) survived to discharge. Median LOS was longer in patients who were survivors (9.0 vs 2.0 days, p<0.001). Median LOS was greater in patients who received TTM (5.0 vs 3.0 days, p< 0.001) and almost half (1813, 49.8%) of all patients had a LOS <72. We observed variation in LOS<72 by hospitals (39.4% to 61.3%, Figure 1), and a negative relationship between LOC<72 and survival (r = -0.36). Conclusion: We observed that nearly half of subjects had a LOS<72, and a modest negative correlation of LOS<72 and outcome. These data suggest that further work is needed to clarify whether early hemodynamic collapse or early withdrawal of care explains differences in hospital LOS and survival.

cardiac & cardiovascular systems,peripheral vascular disease