Trends and outcomes of veno-arterial extracorporeal membrane oxygenation for acute myocardial infarction-cardiogenic shock

Abstract

Abstract Background Mortality for acute myocardial infarction-cardiogenic shock (AMI-CS) remains high. Trends of veno-arterial extracorporeal membrane oxygenation (VA-ECMO) use and outcomes as a function of ECMO requirement are not well described. Purpose To identify the trends and outcomes of VA-ECMO use for AMI-CS in the United States. Methods We used the Nationwide Readmissions Database (NRD) and the Healthcare Cost and Utilization Project (HCUP) to obtain our cohort. We identified patients with AMI-CS requiring VA-ECMO using the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD 10 CM) and procedure (ICD 10-PCS) codes. We excluded patients with missing data on in-hospital mortality. Results During the study period (2016–2019), 2,005,607 records with myocardial infarction (ST-elevation myocardial infarction [STEMI] and non-ST-elevation myocardial infarction [NSTEMI]) were identified (national estimate of 3,674,083 records), 127,939 records (6.4%) had concomitant CS (national estimate of 234,630 patients), and 3,469 records (national estimate of 6,365 records) (2.7%) received VA-ECMO for CS support. Impella and Intra-aortic balloon pump were used within the same admission in 34.5% and 38.1% of VA-ECMO cases, respectively. VA-ECMO use for AMI-CS significantly increased over the study period (from 2.3% in the first quarter of 2016 to 3.3% in the last quarter of 2019, p=0.001) (Figure 1). In unadjusted analysis, patients undergoing VA-ECMO for AMI-CS were more likely to be male, have hypertension, congestive heart failure, and coagulopathy, present with STEMI, and receive multivessel PCI. Other comorbid conditions including diabetes, atrial fibrillation, end-stage renal disease, and prior MI, CABG, and stroke were less common in the VA-ECMO cohort. Medicare patients were less likely to receive VA-ECMO (39.4% vs. 63.8%, p<0.001), whereas patients at large or teaching hospitals were more likely. Patients requiring VA-ECMO had higher in-hospital mortality, length of stay, and rates of complications (Table 1). Conclusion Utilization of VA-ECMO for AMI-CS is increasing, and in one out three patients, VA-ECMO is used in combination with other MCS devices. Patients with underlying medical comorbidities and Medicare are less likely to receive VA-ECMO. Critically ill patients requiring VA-ECMO for AMI-CS are more likely to present with STEMI and have high morbidity and mortality. Funding Acknowledgement Type of funding sources: None.