BACKGROUND AND PURPOSE This study sought to investigate demographic, clinical, and procedural determinants of outcomes in patients treated with percutaneous veno-arterial (VA) extracorporeal membrane oxygenation (ECMO) initiated in the cardiac catheterization laboratory with a portable system. METHODS We performed a retrospective review of patients treated with percutaneous VA-ECMO during the study period at our institution. A logistic regression model was applied to investigate the association between sequential organ failure assessment (SOFA) score and survivor status. Fisher's exact test was used to examine the association between survivor status and cannula size (15 Fr vs >15 Fr). RESULTS Percutaneous VA-ECMO was initiated in 25 patients. At 30 days, 10 patients were alive (40%). Fifteen patients had cardiac arrest (CA) prior to ECMO initiation, of which 5 were alive at 30 days (33%). Survivors had a lower baseline median SOFA score (9 vs 16; P=.02; odds ratio, 0.577). Use of a smaller cannula was associated with survival (P=.01). There was an association between the size of the arterial cannula and increased blood transfusions (P<.01). CONCLUSIONS Lower presenting SOFA score and smaller cannula size were associated with increased survival in patients with cardiogenic shock (CS) or CA who underwent percutaneous VA-ECMO placed in the cardiac catheterization laboratory using a portable system. Calculation of the SOFA score at presentation may help physicians determine which patients may derive benefit from ECMO. Smaller cannula size, while decreasing the amount of flow, may result in decreased bleeding and increased survival.