1151: CARDIAC ARREST OUTCOMES AND TIME OF DAY: A SINGLE-CENTER STUDY

Abstract

Introduction: Previous studies of in-hospital cardiac arrest have shown lower survival rates in arrests which occur during the night time hours. These statistics have held true when looking at aggregate data throughout the hospital as well as specifically in the ICU. The majority of prior analyses have been on registry data from a wide range of facilities, including both academic teaching and community hospitals. This study looks at sentinel arrests occurring in the ICU of a community hospital primarily staffed by advanced practice providers overnight with emergency physician code team backup. Methods: This retrospective, observational analysis of a Code Blue database looked at the survival to hospital discharge of adult patients who received either CPR or defibrillation while in the ICU between January 1st, 2015 – December 31st 2019 based on timing of arrest (0700 – 1859, 1900 – 0659). Binary logistic regression was utilized to assess the relationship between time of day and survival to discharge, as well as chance of ROSC. Multivariable logistic regression was performed utilizing age, gender, initial rhythm, comorbidities, use of vasoactive medications as well as ventilatory status at time of arrest. Results: Over the 5-year period there were 219 Code Blue events meeting the prespecified criteria, of these 118 (54%) occurred during the night hours while 101 (46%) occurred during the day. Survival was 23/118 (19.5%) on nights and 29/101 (28.7%) on days. After adjusting for confounders there was no significant association between an arrest occurring between the hours of 0700 – 1859 p=0.173 OR 1.69[CI 95% 0.79 – 3.6] and survival to hospital discharge. The unadjusted analysis demonstrated an OR 0.24[CI 95% 0.15 - 0.38] of survival if the arrest occurred during this timeframe. An initial shockable rhythm exhibited a positive effect on survival p=0.011 OR 3.7[CI 95% 1.36 – 10.07], while acute kidney injury p=0.012 OR 0.36 [CI 95% 0.17 – 0.8] and the use of vasoactive medications p= < 0.001 OR 0.19 [CI 95% 0.09 – 0.44] at the time of arrest were associated with a negative outcome. Conclusions: When adjusted for confounding factors there was no difference in cardiac arrest outcomes based on time of day in the ICU of a large community hospital staffed by advanced practice providers at night.