Abstract Background Extracorporeal cardiopulmonary resuscitation (E-CPR) is considered for cardiac arrest (CA) patients when conventional CPR is ineffective. While veno-arterial extracorporeal membrane oxygenation (VA-ECMO) robustly maintains systemic perfusion, it also increases left ventricular (LV) afterload. Recently, the transcatheter LV assist device Impella has been introduced, and the potential benefits of its combined use with VA-ECMO (ECPELLA) are anticipated. However, the utility and effective management of ECPELLA remain unclear. Purpose This single-site cohort study aimed to assess the impact of ECPELLA compared to VA-ECMO with or without intra-aortic balloon pump (IABP) in patients with refractory CA who received E-CPR. Methods We retrospectively reviewed 207 consecutive patients who underwent E-CPR between January 2012 and September 2023 at our institute. The patients were divided into two groups, ECEPLLA (n = 65) and VA-ECMO with or without IABP (n = 141). A propensity score analysis was conducted, utilizing 1:1 score matching based on variables including age, the rate of male gender, acute coronary syndrome, out-of-hospital CA (OHCA), in-hospital CA, shockable rhythm, and the time from collapse to VA-ECMO initiation, known as low flow time (LFT). We evaluated 180-day survival rates, hemodynamic data, and safety profiles. Result After propensity score matching, 57 ECPELLA patients and 57 VA-ECMO patients were included for analysis. No significant differences in the adjusted factors and comorbidities were found between the two groups. In the VA-ECMO group, 42 patients (74%) received concomitant use of IABP. During the first 72 hours following the initiation of VA-ECMO, the ECPELLA group received significantly higher total mechanical circulatory support flow compared to the VA-ECMO group, correspondingly received less catecholamine administration, and had more patients started on vasodilators and cardioprotective medications. There were no statistical differences in safety profiles, except for hemolysis and embolic cerebral infarction, between the two groups. Kaplan-Meier analysis demonstrated that the 180-day survival rates were significantly higher in the ECPELLA group than in the VA-ECMO group. In a multivariate Cox proportional hazard analysis that included age, shockable rhythm, OHCA, initial lactate levels, LFT, and ECPELLA, it was found that age (hazard ratio [HR], 1.28 [for 10-year increase]; 95% confidence interval [CI], 1.09â1.51; p = 0.002), initial lactate level (HR, 1.06; 95% CI, 1.01-1.11; p = 0.04), and ECPELLA (HR, 0.55; 95% CI, 0.36â0.85; p = 0.007) were significantly associated with 180-day mortality. Conclusion ECPELLA was associated with improvement of 180-day mortality in patient who received E-CPR. The results suggest that not only the choice of ECPELLA but also sufficient circulatory support during the acute phase, along with modulating catecholamines and vasodilators, may have influenced the prognosis.