Abstract 4148085: Transfer to ‘Hub’ Hospitals and Outcomes in Cardiogenic Shock

Introduction: Patients with cardiogenic shock (CS) have high mortality. Many CS patients present to non-tertiary care centers. Limited data exist on outcomes for patients transferred to “hub” CS centers for both acute myocardial infarction (AMI) and non-AMI etiologies. This study sought to compare the characteristics and outcomes of transferred CS patients compared to patients presenting to a hub center. Methodology: Adults (≥18 years) with a primary or secondary diagnosis of CS were identified from the National Readmissions Database (2016-2020) and stratified by transfer status. Overlap propensity score (PS) weighting was performed to assess the association between in-hospital mortality, length of stay (LOS) and transfer status. Results: Of 314,098 CS patients (AMICS 27%, non-AMICS 73%), 30,630 (9.8%) were transferred. Compared to non-transferred patients, transferred patients were on average younger (65 [interquartile range {IQR} 55-73] vs. 68 [IQR 58-77] years), more likely to be female (34% vs. 37%), less frequently with Medicare as primary payor (55% vs. 62%), and with higher comorbidity (Elixhauser index 8 [IQR 6-9] vs. 6 [IQR 5-8]) (all p <0.001). Transferred patients had higher rates of respiratory failure (70% vs. 64%), renal failure (71% vs. 60%), hepatic failure (25% vs. 17%), bleeding complications (28% vs. 17%), and arrhythmias (59% vs. 52%) (all p <0.001). Transferred patients were also more likely to undergo percutaneous coronary intervention (22% vs. 19%), coronary artery bypass grafting (13% vs. 8%), pulmonary artery catheter insertion (36% vs.16%), placement of intra-aortic balloon pump (25% vs.13%), percutaneous left ventricular assist device (13% vs. 7%), and extracorporeal membrane oxygenation (7% vs. 2%), receive invasive mechanical ventilation (3% vs. 2%), and undergo hemodialysis (17% vs. 11%) (all p <0.001). In-hospital mortality was significantly lower in transferred patients – 39% vs. 47.1%; unadjusted odds ratio 0.71 (95% confidence interval 0.70-0.73); adjusted PS weighting odds ratio 0.73 (95% confidence interval 0.71-0.76), p <0.001. Transferred patients had a longer median LOS (13 [IQR 7-22] vs. 6 [IQR 3-11] days, p <0.001). Conclusion: Transferred CS patients had higher use of cardiovascular procedures and organ support therapies and lower in-hospital mortality suggesting patients were appropriately selected for transfer.