Abstract
BACKGROUND: Frailty predicts adverse outcomes in critical illness, but its prognostic value in patients receiving extracorporeal membrane oxygenation (ECMO) remains unclear. METHODS: In this retrospective cohort study across 116 intensive care units (ICUs), we analysed data from the Australia and New Zealand Intensive Care Society Adult Patient Database on adults (≥ 16 years) who received ECMO between 2018 and 2024. Frailty was assessed using the Clinical Frailty Scale (CFS), and we defined patients as being frail if they had a CFS of ≥ 5. The primary outcome was one-year mortality, which we assessed using g-computation. We further analysed the association between frailty and death among several subgroups, including age, sex, and primary ICU admission diagnostic group (cardiac medical, cardiac surgical, cardiac arrest, respiratory, transplant, and others). Secondary outcomes included in-hospital and ICU mortality, lengths of hospital and ICU stay, delirium, and pressure injuries. RESULTS: We included 2,084 patients (51.3 years [IQR: 38.9–61.3], 64.2% male). Frailty was present in 254 (12.2%) patients. The adjusted one-year mortality was 41.9% (760/1830) in patients without frailty and 42.1% (113/254) in patients with frailty (adjusted risk difference: +0.27%, 95%-CI: -6.65% to +7.19%). These results were consistent across sensitivity and subgroup analyses, particularly amongst principal diagnostic groups (cardiac medical: -7.87%, cardiac surgical: + 3.70%, cardiac arrest: -1.66%, respiratory: + 10.02%, transplant: + 2.19%, others: -0.58%). CONCLUSIONS: Frailty was not associated with short or long-term mortality among patients who received ECMO. This may reflect patterns in patient eligibility for ECMO. More studies are required to investigate this relationship and validate these findings. GRAPHICAL ABSTRACT: [Image: see text] SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s13054-025-05801-0.