Abstract
BACKGROUND: To investigate the impact of hyperoxia that developed immediately after extracorporeal membrane oxygenation (ECMO)-assisted cardiopulmonary resuscitation (ECPR) on patients' short-term neurological outcomes after out-of-hospital cardiac arrest (OHCA). METHODS: This study retrospectively analyzed data from the Japanese OHCA registry from June 2014 to December 2017. We analyzed adult patients (≥ 18 years) who had undergone ECPR. Eligible patients were divided into the following three groups based on their initial partial pressure of oxygen in arterial blood (PaO(2)) levels after ECMO pump-on: normoxia group, PaO(2) ≤ 200 mm Hg; moderate hyperoxia group, 200 mm Hg < PaO(2) ≤ 400 mm Hg; and extreme hyperoxia group, PaO(2) > 400 mm Hg. The primary and secondary outcomes were 30-day favorable neurological outcomes. Logistic regression statistical analysis model of 30-day favorable neurological outcomes was performed after adjusting for multiple propensity scores calculated using pre-ECPR covariates and for confounding factors post-ECPR. RESULTS: Of the 34,754 patients with OHCA enrolled in the registry, 847 were included. The median PaO(2) level was 300 mm Hg (interquartile range: 148-427 mm Hg). Among the eligible patients, 277, 313, and 257 were categorized as normoxic, moderately hyperoxic, and extremely hyperoxic, respectively. Moderate hyperoxia was not significantly associated with 30-day neurologically favorable outcomes compared with normoxia as a reference (adjusted odds ratio, 0.86; 95% confidence interval: 0.55-1.35; p = 0.51). However, extreme hyperoxia was associated with less 30-day neurologically favorable outcomes when compared with normoxia (adjusted odds ratio, 0.48; 95% confidence interval: 0.29-0.82; p = 0.007). CONCLUSIONS: For patients with OHCA who received ECPR, extreme hyperoxia (PaO(2) > 400 mm Hg) was associated with 30-day poor neurological outcomes. Avoidance of extreme hyperoxia may improve neurological outcomes in patients with OHCA treated with ECPR.