B55-12 Association of Out-of-Bed Mobility With Hospital Survival in Veno-Venous Extracorporeal Membrane Oxygenation (VV ECMO) Patients: A Retrospective Single-Center Analysis

Rationale Mobilization on veno-venous extracorporeal membrane oxygenation (VV ECMO) has been associated with improved outcomes, but the association between hospitalization survival and ambulation remains unclear. This retrospective study aimed to identify the clinical variables associated with hospitalization survival and out-of-bed mobility in non-transplant VV ECMO patients. Methods A retrospective review of all patients supported with VV ECMO between 2019 to 2025 at Inova Fairfax Hospital was performed. Patients bridging to lung transplantation were excluded (n = 43). Out-of-bed mobility, defined as maximal nurse-rated achievement of steps on a standardized 7-point scale (chair, walking in room, or walking out of room), and survival to discharge were the primary outcomes. Diagnosis and cannula site were included as categorical variables. Age, ECMO duration, maximum ECMO flow, and time to maximum mobility were continuous variables. Correlation coefficients and p-values were calculated for each numerical variable. Cannula site and diagnosis were each compared to hospitalization survival, out-of-bed mobility, and time to maximum mobility. Multivariable Firth logistic regression models were used to evaluate predictors of survival and mobility. Results 179 patients were included in the analysis. 107 patients survived to discharge (59.8%). Out-of-bed mobility was achieved in 31 patients (17.3%). Univariate analyses revealed significant negative correlations between ECMO duration (r=-0.37, p < 0.001), maximum flow (r=-0.27, p < 0.001), age (r=-0.24, p < 0.001), and time to max mobility (r=-0.24, p < 0.001), with survival. ECMO duration (r = 0.37, p < 0.001) and time to max mobility (r = 0.42, p < 0.001) were positively correlated with achieving out-of-bed mobility. In the survival logistic regression model, out-of-bed mobility (OR = 5.31, p = 0.007), age (OR = 0.94, p < 0.001), ECMO duration (OR = 0.91, p < 0.01), and a diagnosis of COVID-19 compared to ARDS (OR = 2.54, p = 0.038) were significantly associated with survival. In the mobility logistic regression model, ECMO duration was significantly associated with out-of-bed mobility (OR = 1.05, p < 0.001). Cannula site (dual lumen jugular, n = 55; femoral/femoral, n = 101; femoral/jugular, n = 20; dual drainage cannulas, n = 3) did not impact mobility or survival. Conclusions Out-of-bed mobility independently predicts survival during VV ECMO. Although longer ECMO runs reduce survival, they facilitate ambulation. These findings support implementing early mobilization strategies and call for future prospective studies to evaluate causality and long-term functional outcomes. This abstract is funded by: None