Risk Factors and Mortality Impact of Acute Kidney Injury in Different ECMO Modalities: A Multicenter Retrospective Study

Objective To identify independent risk factors for acute kidney injury (AKI) in critically ill patients undergoing extracorporeal membrane oxygenation (ECMO) and to evaluate the association between AKI severity and 30-day mortality. Methods A retrospective multicenter cohort study across three provincial ECMO medical centers was conducted between September 2019 and June 2024. Serum creatinine levels within the first 7 days following ECMO initiation were collected, and AKI was staged according to KDIGO criteria, with dynamic progression tracked over time. Multivariable logistic regression was used to identify predictors of moderate-to-severe AKI (AKI stages 2–3). Survival analysis was performed using Kaplan-Meier curves, and Cox proportional-hazards models were applied to assess the impact of AKI stage on 30-day mortality. Results Among 210 enrolled patients, 100 (47.6%) had AKI stages 0–1, while 110 (52.4%) developed AKI stages 2–3 within 7 days after ECMO initiation. Serial creatinine monitoring indicated a gradual increase in AKI stage 2, whereas the incidence of AKI stage 3 plateaued. Cox regression analysis demonstrated that moderate-to-severe AKI was independently associated with 30-day mortality, both before and after adjustment for APACHE II score (adjusted HR = 1.29, 95% CI 1.13–1.86, P  < 0.05). In the overall cohort, VA-ECMO modality, norepinephrine use, and low fibrinogen level were significant risk factors for AKI stages 2–3, while pre-existing cardiac disease was protective (AUC 0.78). Subgroup analysis showed that for VV-ECMO patients, elevated LAC, HCO₃⁻, CRP, PCT, and BUN, as well as decreased fibrinogen, TBIL, and WBC count, were associated with AKI risk (AUC = 0.89). Conclusions AKI is an independent risk factor for 30-day mortality in ECMO patients. VA-ECMO modality, vasopressor use, and low fibrinogen levels increase the risk of moderate-to-severe AKI, with risk profiles varying by ECMO modality. Early identification of high-risk phenotypes and mode-specific management strategies are essential to improving outcomes.