Background
The Association between hyperoxia and mortality in patients receiving extracorporeal membrane oxygenation (ECMO) remains controversial. This study aimed to systematically evaluate the impact of arterial hyperoxia on mortality in adult ECMO patients.

Methods
We conducted a systematic review and meta-analysis following PRISMA guidelines (PROSPERO registration: CRD420251048959). PubMed, Web of Science, Embase, and the Cochrane Library were searched for relevant studies published up to April 1, 2025. Pooled odds ratios (ORs) with 95 % confidence intervals (CIs) were calculated for 28-day mortality. Heterogeneity was assessed using the I2 statistic.

Results
Twelve studies reporting a combined total of 17,529 study records were included. Hyperoxia was significantly associated with increased mortality (OR 1.41, 95 % CI 1.22–1.64; I2 = 89.96 %, p < 0.01). Subgroup analyses showed higher mortality with PaO₂ ≥200 mmHg and ≥ 300 mmHg, particularly in patients receiving VA-ECMO and ECPR. No significant association was found in VV-ECMO patients. Sensitivity analyses supported the stability of most results, though potential publication bias could not be ruled out.

Conclusions
In this systematic review and meta-analysis of twelve studies investigating arterial oxygenation and mortality in adult ECMO patients, early hyperoxia—particularly at very high PaO₂ levels—was associated with increased mortality in several observational cohorts, most notably among VA-ECMO and ECPR populations. However, substantial heterogeneity, residual confounding, and the scarcity of randomized evidence limit the strength of these conclusions. Future prospective, adequately powered randomized trials with standardized oxygen metrics and rigorous adjustment for key physiological confounders are needed to define optimal oxygenation targets in ECMO patients.