Hyperoxia After Pediatric Cardiac Arrest: Association with Survival and Neurological Outcomes

Abstract

Objective: To evaluate the association between hyperoxia in the first 24 hours after in-hospital pediatriccardiac arrest and mortality and poor neurological outcome. Design: Retrospective cohort study. Setting: Tertiary care freestanding children’s hospital. Patients: Patients younger than 18 years of age with in-hospital cardiac arrest between December 2012 and December 2019, who achieved return of circulation (ROC) for longer than 20 minutes, survived at least 24 hours after cardiac arrest, and had documented PaO2 or SpO2 during the first 24 hours after ROC. Interventions: None. Main Results: There were 187 patients who met eligibility criteria, of whom 48% had hyperoxia (PaO2 > 200 mmHg) during the first 24 hours after cardiac arrest. In-hospital mortality in this cohort of patients was 41%, with similar mortality between oxygenation groups (hyperoxia 45% vs no hyperoxia 38%). We did not observe an association between hyperoxia and in-hospital mortality or poor neurological outcome after adjusting for confounders (odds ratio 1.2, 95% confidence interval 0.5-2.8). On sensitivity analysis using two additional cutoffs of PaO2 (>150 mmHg and > 300 mmHg), there was also no association with in-hospital mortality or poor neurological outcome after adjusting for confounders. Similarly, on multivariable logistic regression using SpO2 > 99% as the exposure, there was no difference in the frequency of death or poor neurological outcome at hospital discharge. Conclusions: In this cohort of patients, hyperoxia after pediatric cardiac arrest was common and was not associated with worse in-hospital outcomes.