Abstract Background Venoarterial extracorporeal membrane oxygenation (VA-ECMO) is an effective therapeutic modality for patients with cardiogenic shock (CS) including STEMI. The VA-ECMO maintains end-organ perfusion, however, it significantly increases damaged left ventricular (LV) afterload. Combined treatment of VA-ECMO and a micro-axial Impella pump, ECPELLA, simultaneously provides systemic circulatory support and LV loading reduction (LV unloading). Studies in ischemic animal models displayed that LV unloading reduced myocardial size. However, it remains unknown whether LV unloading effect by ECPELLA during percutaneous coronary intervention (PCI) can reduce myocardial damage in human STEMI patients with severe cardiogenic shock. Purpose This study was to assess whether PCI on ECPELLA support can reduce myocardial damage in STEMI patients with severe cardiogenic shock. Methods Sixteen consecutive patients with STEMI and lethal CS (SCAI stage E) were enrolled. All patients suffered from cardiac arrest. Fifteen patients (94%) underwent extracorporeal cardiopulmonary resuscitation using VA-ECMO. From October 2018 to November 2019, PCI was carried out on VA-ECMO support and the mechanical circulatory support (MCS) modality during PCI was switched to ECPELLA since December 2019. There were 8 patients who underwent PCI on ECPELLA support (ECPELLA-PCI) and 8 patients who underwent PCI on single VA-ECMO support (ECMO-PCI). All patients in ECMO-PCI group were escalated to ECPELLA after PCI. Therefore, the difference between groups was timing of LV unloading, i.e., pre-PCI or post-PCI condition. We assessed total MCS flow during PCI, and peak and time course changes in CK-MB levels from days 0 to 3. Results There were no significant differences in age, rate of male sex, body surface area, shockable rhythm, serum lactate levels, and door to MCS time. Patients in ECPELLA-PCI had significantly longer door to recanalization time (73min in ECPELLA vs. 49min in VA-ECMO, p<0.05), and higher total MCS flow during PCI (3.1 L/min/sqM vs. 2.0 L/min/sqM, p<0.05). Peak and the area under the curves (AUC) from day 0 to day 3 of CK-MB in ECPELLA-PCI were significantly lower than ECMO-PCI group (peak CK-MB 120 vs. 524 IU/L and CK-MB AUC 143 vs. 464 IU/L*Day, respectively, p<0.05). While all patients received red blood cell (RBC) and plasma (PL) transfusions, ECPELLA-PCI required larger amount of RBC and PL transfusions (p<0.05). Conclusion The ECPELLA-PCI appeared to reduce myocardial damage shown by peak and AUC of CK-MB levels. Lager amount of blood transfusion is a major drawback of this treatment modality. Further studies are necessary whether ECPELLA-PCI is an effective revascularization approach for myocardial damage reduction with favorable mid- and long-term outcome in patients with STEMI with severe cardiogenic shock. Funding Acknowledgement Type of funding sources: None. Total MCS FLow and Ttoal MCS flow indexPeak CK-MB and CK-MB AUC