Abstract P1057: Outcomes Of Impella Versus ECMO In Multivessel Percutaneous Coronary Intervention

Abstract

Background: Impella is a mechanical circulatory support device indicated in severe cardiogenic shock and high-risk percutaneous coronary intervention. ECMO is also used during the percutaneous coronary intervention (PCI) in an acute coronary syndrome that is complicated by refractory cardiogenic shock. There have been few studies examining short-term mechanical circulatory support devices for patients that undergo PCI, so the choice between Impella and ECMO is still a matter of debate. We investigated the in-hospital outcomes of Impella and ECMO in patients who underwent multi-vessel PCI (MVPCI). Methods: In a retrospective cohort study of National Inpatient Sample 2017 data, we used ICD-10 codes to identify patients who underwent MVPCI and received Impella or ECMO support. Multivariate logistic regression was used to adjust for patient demographics, hospital demographics, and relevant comorbidities. Primary outcomes (inpatient mortality, length of stay (LOS), and total hospital charges. Secondary outcomes, such as respiratory failure (RF), gastrointestinal hemorrhage, cerebrovascular accident (CVA)/transient ischemic attack (TIA), intracerebral hemorrhage (ICH), acute kidney injury (AKI), ventricular tachycardia (VT), and atrial fibrillation, were also analyzed. Results: Among 64,489 patients who had MVPCI, 102 (%) received ECMO support, and 927 (%) received Impella support. Both Impella and ECMO use was associated with high in-hospital mortality (Impella 21.8% vs ECMO (49.5%), high hospital LOS (ECMO 19.9 days vs Impella 9.28 days), and total hospital charges (Impella $3,200,000 vs ECMO $7,500,00). Mortality (OR 4.27, p<0.0001), hospital LOS (OR 5.9, p<0.0001), and total hospital charges (OR 7.52, p<0.0001) for patients undergoing MVPCI. There was no statistically significant difference in complications such as RF, gastrointestinal hemorrhage, CVA/TIA, ICH, AKI, VT, and atrial fibrillation. Hispanic patients had higher mortality with ECMO (OR 2.28, p <0.024) than all other ethnicities and races. Conclusions: Both patients with Impella support and ECMO support had significant inpatient mortality. Our analysis suggests that Impella patients had better outcomes than patients on ECMO for MVPCI.