LAUSR

Introduction/Hypothesis: Appropriate arterial partial pressure of oxygen (P a O2) remains unclear in cases of extracorporeal cardiopulmonary resuscitation (ECPR) for out–of–hospital cardiac arrest (OHCA). The aim of the present study is to investigate the relationship between hyperoxia and 1– month mortality in the patients who underwent ECPR. Methods: This single–center retrospective cohort study was conducted from January 2010 to December 2018. The OHCA patients who underwent ECPR were included. Exclusion criteria were 1) age < 18, 2) dead within 24 hours after admission, 3) return of spontaneous circulation at hospital arrival. Based on P a O2 at 24 hours after admission, the hyperoxia group was defined as PaO2 ≥ 201 mmHg, the moderate hyperoxia group was defined as P a O2 of 101 – 200 mmHg, and the normoxia group was defined as P a O2 of 61 – 100 mmHg. The primary outcome was 1–month survival after cardiac arrest. The secondary outcome was 1–month neurologically favorable outcome. Multivariate logistic regression analysis for 1–month mortality and P a O2 were performed after adjusting for multiple propensity scores. To estimate the multiple propensity score, we fitted a multinomial logistic regression model, which fell into one of the three groups as patient demographics. Results: Of the patients who underwent ECPR, 109 were eligible; the hyperoxia group included 35 cases, the moderate hyperoxia group included 44 cases and the normoxia group included 30 cases. There is no significant association between oxygenation and 1–month mortality in the patients who underwent ECPR (adjusted odds ratio [OR]: 1.27, 95% confidence interval [CI]: 0.32–5.01 for moderate hyperoxia, and adjusted OR: 1.19, 95% CI: 0.28–5.10 for hyperoxia using normoxia as reference, respectively). Furthermore, no association was seen between oxygenation and 1–month neurological outcome. Conclusions: No significant association between hyperoxia at 24 hours after admission and 1– month mortality in OHCA patients who underwent ECPR could be identified.