Related Experiment Video
Updated: Jan 8, 2026
Utilizing Percutaneous Ventricular Assist Devices in Acute Myocardial Infarction Complicated by Cardiogenic Shock
Published on: June 12, 2021
Impella vs. IABP in Non-emergent High-Risk PCI: Outcomes and LVEF Recovery from a Large US EHR Study
Jason R Wollmuth1, Aditya S Bharadwaj2, Nudrat Noor3
1Providence Heart Institute, 9427 SW Barnes Rd Ste 495, Portland, OR, 97225, USA. jason.wollmuth@providence.org.
Introduction:
High-risk percutaneous coronary intervention (HR-PCI) involves patients with complex coronary disease, adverse hemodynamics, and/or severe comorbidities who are often ineligible for surgery. Mechanical circulatory support (MCS) may reduce procedural risk and facilitate complete revascularization. However, comparative data on outcomes and left ventricular ejection fraction (LVEF) recovery are limited. We hypothesized that the benefits of MCS-supported PCI observed in prior studies would extend to contemporary, real-world, non-protocolized all-comer datasets, where outcomes and LVEF recovery in patients undergoing non-emergent Impella- or intra-aortic balloon counterpulsation (IABP)-supported HR-PCI can be evaluated.
Methods:
We synthesized an MCS-specific cohort using de-identified electronic health record data (2017-2025). Adults undergoing non-emergent HR-PCI supported with cardiac unloading via a continuous, forward high-flow pump (Impella) or IABP were included. Admissions with emergent status, right heart failure, cardiogenic shock, or ST-elevation myocardial infarction, or those undergoing coronary artery bypass grafting, were excluded. Propensity score matching (1:1) adjusting for baseline differences was performed. Outcomes included all-cause mortality (7, 30, and 90 days) and 30-day, medical code-derived (cd) adverse events (acute kidney injury [cd-AKI] and cd-bleeding requiring transfusion). LVEF change within 1 year was assessed in a predefined subgroup.
Results:
Before matching, patients supported with Impella had more comorbidities, lower baseline LVEF, and more complex procedural characteristics than IABP-supported patients. After matching, baseline characteristics were balanced. Impella was associated with lower all-cause mortality at 30 days (12.7% vs. 16.6%) and 90 days (15.2% vs. 19.6%), and reduced cd-AKI (15.7% vs. 20.3%). In patients matched for baseline LVEF values and collection timing, both groups demonstrated LVEF improvement (+ 7% for Impella; + 3% for IABP).
Conclusion:
In contemporary, non-protocolized, non-emergent HR-PCI, the use of Impella was associated with improved outcomes and greater LVEF recovery compared to IABP.
