In acute respiratory distress syndrome (ARDS), extracorporeal membrane oxygenation (ECMO) without invasive mechanical ventilation (IMV) is particularly challenging.

To study ARDS patients treated with ECMO to avoid IMV—‘primary awake ECMO’—or with extubation during ECMO support – ‘extubated ECMO’.

International retrospective cohort of adult ARDS patients treated with ECMO without IMV at 14 centers in 8 countries (2015-2024). The primary outcome was mortality 90 days after ECMO initiation.

Among 307 adult patients with ARDS, 113 received ‘primary awake ECMO’ and 194 were extubated on ECMO. Ninety-day mortality was 30.1% in the ‘primary awake ECMO’ group and 14.9% in the ‘extubated ECMO’. Strategy failure occurred in 46 patients (40.7%) and 47 patients (24.2%), respectively, most frequently within the first 10 days. In multivariate analysis, strategy failure was associated with 90-day mortality (hazard ratio 7.67 (3.44-17.11); P < .001 in ‘extubated ECMO’; hazard ratio 5.95 (2.63-13.46); P < .001 in ‘primary awake ECMO’), while higher age and longer time from ICU admission to ECMO cannulation were associated with 90-day mortality in ‘extubated ECMO’ and ‘primary awake ECMO’, respectively. The leading cause of strategy failure was worsening of respiratory failure, followed by agitation/delirium in ‘primary awake ECMO’ and inability to clear secretions in ‘extubated ECMO’

Patients selected for ‘primary awake ECMO’ and ‘extubated ECMO’ presented different baseline characteristics, strategy failure, and 90-day mortality rates. However, strategy failure was consistently associated with 90-day mortality in both groups.