Background-Observational studies in adults have shown a worse outcome associated with hyperoxia after resuscitation from cardiac arrest. Extrapolating from adult data, current pediatric resuscitation guidelines recommend avoiding hyperoxia. We investigated the relationship between arterial partial oxygen pressure and survival in patients admitted to the pediatric intensive care unit (PICU) after cardiac arrest. Methods and Results-We conducted a retrospective cohort study using the Pediatric Intensive Care Audit Network (PICANet) database between 2003 and 2010 (nϭ122 521). Patients aged Ͻ16 years with documented cardiac arrest preceding PICU admission and arterial blood gas analysis taken within 1 hour of PICU admission were included. The primary outcome measure was death within the PICU. The relationship between postarrest oxygen status and outcome was modeled with logistic regression, with nonlinearities explored via multivariable fractional polynomials. Covariates included age, sex, ethnicity, congenital heart disease, out-of-hospital arrest, year, Pediatric Index of Mortality-2 (PIM2) mortality risk, and organ supportive therapies. Of 1875 patients, 735 (39%) died in PICU. Based on the first arterial gas, 207 patients (11%) had hyperoxia (PaO 2 Ն300 mm Hg) and 448 (24%) had hypoxia (PaO 2 Ͻ60 mm Hg). We found a significant nonlinear relationship between PaO 2 and PICU mortality. After covariate adjustment, risk of death increased sharply with increasing hypoxia (odds ratio, 1.92; 95% confidence interval, 1.80 -2.21 at PaO 2 of 23 mm Hg). There was also an association with increasing hyperoxia, although not as dramatic as that for hypoxia (odds ratio, 1.25; 95% confidence interval, 1.17-1.37 at 600 mm Hg). We observed an increasing mortality risk with advancing age, which was more pronounced in the presence of congenital heart disease. Conclusions-Both severe hypoxia and, to a lesser extent, hyperoxia are associated with an increased risk of death after PICU admission after cardiac arrest. (Circulation. 2012;126:335-342.)