Timing and Cause of Death in Children Following Return of Circulation After Out-of-Hospital Cardiac Arrest: A Single-Center Retrospective Cohort Study

Abstract

Objectives: To determine timing and cause of death in children admitted to the PICU following return of circulation after out-of-hospital cardiac arrest.

Design: Retrospective observational study.

Setting: Single-center observational cohort study at the PICU of a tertiary-care hospital (Erasmus MC-Sophia, Rotterdam, The Netherlands) between 2012 and 2017.

Patients: Children younger than 18 years old with out-of-hospital cardiac arrest and return of circulation admitted to the PICU.

Measurements and results: Data included general, cardiopulmonary resuscitation and postreturn of circulation characteristics. The primary outcome was defined as survival to hospital discharge. Modes of death were classified as brain death, withdrawal of life-sustaining therapies due to poor neurologic prognosis, withdrawal of life-sustaining therapies due to refractory circulatory and/or respiratory failure, and recurrent cardiac arrest without return of circulation. One hundred thirteen children with out-of-hospital cardiac arrest were admitted to the PICU following return of circulation (median age 53 months, 64% male, most common cause of out-of-hospital cardiac arrest drowning [21%]). In these 113 children, there was 44% survival to hospital discharge and 56% nonsurvival to hospital discharge (brain death 29%, withdrawal of life-sustaining therapies due to poor neurologic prognosis 67%, withdrawal of life-sustaining therapies due to refractory circulatory and/or respiratory failure 2%, and recurrent cardiac arrest 2%). Compared with nonsurvivors, more survivors had witnessed arrest (p = 0.007), initial shockable rhythm (p < 0.001), shorter cardiopulmonary resuscitation duration (p < 0.001), and more favorable clinical neurologic examination within 24 hours after admission. Basic cardiopulmonary resuscitation event and postreturn of circulation (except for the number of extracorporeal membrane oxygenation) characteristics did not significantly differ between the withdrawal of life-sustaining therapies due to poor neurologic prognosis and brain death patients. Timing of decision-making to withdrawal of life-sustaining therapies due to poor neurologic prognosis ranged from 0 to 18 days (median: 0 d; interquartile range, 0-3) after cardiopulmonary resuscitation. The decision to withdrawal of life-sustaining therapies was based on neurologic examination (100%), electroencephalography (44%), and/or brain imaging (35%).

Conclusions: More than half of children who achieve return of circulation after out-of-hospital cardiac arrest died after PICU admission. Of these deaths, two thirds (67%) underwent withdrawal of life-sustaining therapies based on an expected poor neurologic prognosis and did so early after return of circulation. There is a need for international guidelines for accurate neuroprognostication in children after cardiac arrest.