Abstract 4364333: Circulatory Support Devices in ST-Elevation MI Cardiogenic Shock Patients: Real-World Outcomes from Nationwide Inpatient Sample

Background: Recent guidelines for ST-elevation myocardial infarction (STEMI) complicated by cardiogenic shock (CS) support the use of impeller percutaneous endovascular devices (PED), in addition to percutaneous coronary intervention (PCI). Other mechanical circulatory support (MCS) devices, such as the intra-aortic balloon pump (IABP) or extracorporeal membrane oxygenation (ECMO), have also been used in some cases. We aim to evaluate the real-world use of these MCS devices in STEMI-CS patients. Methods: This study utilized Nationwide Inpatient Sample (NIS) data from 2017 to 2022, the largest publicly available inpatient database in the United States. We used ICD-10 codes to identify STEMI-CS patients and types of MCS devices used, stratifying them into No device (n=8531), IABP (n=4804), and PED (n=3123) groups. The STEMI-CS patients not receiving PCI were excluded. ECMO patients were also excluded (n=407). To alleviate the differences identified on initial analysis, a propensity matching was done between No Device, IABP, and PED groups (ECMOs excluded, being too small). In-hospital mortality and charges were compared as outcomes post-matching. Results: There were 16,458 STEMI-CS patients included in the final cohort. The patients in PED group were more likely male (74% v. 71% (IABP) v. 65% (No device), p<.01) and with resuscitation (25% v. 16% (IABP) v. 15% (No device), p<.01). Hypertension (67% v. 70% (IABP) v. 71% (No device), p<.01) was more common in No device group (Table 1A). The 1:1:2 propensity matching, respectively for PED (n=2900), IABP (n=2883), and No Device (n=5783) groups, reduced the differences in baseline characteristics between the groups (Table 1B). After propensity matching, the in-hospital mortality in impeller group was 56% which was significantly higher than IABP (38%) and No device (30%) groups (p<.01, Figure 1) The in-hospital charges were significantly higher ($285k v. $160k (IABP) v. $128k (No device), p<.01) for the PED (Impeller) group compared to others (Figure 1). Conclusion: The real-world outcomes of STEMI-CS patients show that in-hospital mortality remains high. Even after propensity matching, use of Impeller devices (PED) was associated with higher in-hospital mortality and charges, raising questions about its effectiveness and pertinent use. Although these patient populations are likely different, caution shall prevail while using Impeller devices in STEMI-CS patients.