A Prospective Randomized Clinical Trial of Hemodynamic Support With Impella 2.5 vs. Intra-Aortic Balloon Pump in Patients Undergoing High-Risk PCI - PROTECT II


The PROTECT II trial sought to compare outcomes between Impella 2.5L percutaneous left ventricular assist device (LVAD) and intra-aortic balloon pump (IABP) in patients undergoing high-risk percutaneous coronary intervention (PCI).


Use of the Impella device would be superior to IABP in patients undergoing high-risk PCI who were thought to require hemodynamic support.

Study Design

  • Randomized
  • Parallel
  • Stratified

Patient Populations:

  • Age ≥18 years
  • Undergoing nonemergent PCI on an unprotected left main or last patent coronary vessel with a LVEF ≤35%
  • Patients with three-vessel disease and LVEF <30%

    Number of enrollees: 452
    Duration of follow-up: 90 days
    Mean patient age: 67.5 years
    Percentage female: 19%
    Ejection fraction: 24%
    NYHA class: III/IV (66%)


  • Recent MI with persistent elevation of cardiac enzymes
  • LV thrombus
  • Platelet count ≤75,000/mm3
  • Creatinine ≥4 mg/dl (patients already on dialysis were eligible)
  • Severe peripheral vascular disease that precluded passage of the Impella 2.5L catheter

Primary Endpoints:

  • Major adverse events at 30 days (death/MI/stroke/ repeat revascularization procedure/need for a cardiac or a vascular operation/acute renal insufficiency/severe intraprocedural hypotension requiring therapy/CPR or VT requiring cardioversion/aortic valve damage/angiographic failure of PCI)

Secondary Endpoints:

  • In-hospital efficacy and safety endpoints consisting of efficacy of hemodynamic support assessed by maximal decrease of cardiac power output from baseline
  • Creatinine clearance change from baseline 24 hours post-PCI
  • Device failure assessed as Impella flow <1 L/min for more than 5 minutes at the performance level 5 or higher (out of 9)
  • Rate of in-hospital major adverse events

Drug/Procedures Used:

Patients were randomized to receive either Impella or IABP. Impella implantation required a 12F sheath. Therapeutic anticoagulation with activated clotting time >250 seconds was required. Revascularization was performed and hemodynamics, aortic pressure, and cardiac output were measured every 15 minutes. Investigators were asked to discontinue hemodynamic support prior to discharge from the catheterization laboratory if the patient was deemed hemodynamically stable.

Concomitant Medications:

For IABP versus Impella: glycoprotein IIb/IIIa inhibitors (26% vs. 13.8%), heparin (83.3% vs. 93.3%)

Principal Findings:

A total of 448 patients were randomized, 223 to IABP and 225 to Impella. The trial had been powered to enroll 654 patients, but was terminated early by the Data Safety Monitoring Board after an interim analysis suggested futility. Baseline characteristics were fairly similar between the two arms. About 51% had diabetes mellitus, 26% had peripheral arterial disease, and 33% had undergone prior coronary artery bypass grafting (CABG). The average left ventricular ejection fraction (LVEF) was 24% and two thirds of the patients had New York Heart Association (NYHA) class III/IV symptoms. The mean SYNTAX score was 30, with a mean Society of Thoracic Surgeons (STS) score of 6%. Approximately two thirds of the patients had been deemed inoperable. A mean of 2.9 lesions were attempted per patient, and 12% underwent rotational atherectomy. The total support time was significantly longer in the IABP arm, as compared with the Impella arm (8.4 hours vs. 1.9 hours); 36.7% patients were discharged from the cath lab on an IABP versus 5.9% with an Impella.

In-hospital events were similar between the two arms. The composite primary endpoint of major adverse events at 30 days (death/myocardial infarction [MI]/stroke/ repeat revascularization procedure/need for a cardiac or a vascular operation/acute renal insufficiency/severe intraprocedural hypotension requiring therapy/cardiopulmonary resuscitation [CPR] or ventricular tachycardia [VT] requiring cardioversion/aortic valve damage/angiographic failure of PCI) was similar between the IABP and Impella arms (40.1% vs. 35.1%, p = 0.28). Individual endpoints at 30 days including death (5.9% vs. 7.6%, p = 0.47), MI (10.4% vs. 13.8%, p = 0.27), and need for cardiovascular surgery (1.4% vs. 0.9%, p = 0.64) were similar; stroke rates were higher in the IABP arm (1.8% vs. 0%, p = 0.04). Outcomes were also similar at 90 days, including the composite endpoint (49.3% vs. 40.6%, p = 0.07) and stroke (2.7% vs. 0.9%, p = 0.14).

On per-protocol analysis (excluding patients who did meet protocol eligibility), there appeared to be a benefit with Impella for the composite endpoint at 90 days (51.0% vs. 40.0%, p = 0.023), driven mostly by a reduction in the need for repeat revascularization (8.1% vs. 3.7%, p = 0.06). There appeared to be effect modification by rotational atherectomy use, such that patients not undergoing this procedure had a significant reduction in major adverse event rate at 30 (p = 0.01) and 90 days (p = 0.003) with Impella, as compared with IABP.


The results of the PROTECT II trial in high-risk patients undergoing PCI indicate that the use of the Impella 2.5L percutaneous LVAD is not superior to IABP in reducing adverse events at 30 and 90 days. Rates of adverse events including need for vascular surgery and aortic valve damage were low and similar, despite the need for a 12F sheath for the Impella device.

The recent BCIS-1 trial found no difference in outcomes with mandatory IABP versus provisional IABP use in similar high-risk patients undergoing PCI. Taken together, these results suggest that, in appropriately selected high-risk patients, PCI can be safely conducted without the need for hemodynamic support. If hemodynamic support is deemed necessary, both IABP and Impella 2.5L may be acceptable alternatives. It is unclear if the Impella 5L, with more robust hemodynamic support than the 2.5L, will have a more salutary effect in high-risk PCI patients.

While the mortality and reinfarction rates noted in this trial were high, they were still similar to those noted in similar high-risk patients undergoing surgical revascularization (such as in the STICH trial). Given that two thirds of the patients in this trial were deemed surgically inoperable, these results also suggest that PCI may be an acceptable option in high-risk patients.


O’Neill WW, Kleiman NS, Moses J, et al. A prospective randomized clinical trial of hemodynamic support with Impella 2.5™ versus intra-aortic balloon pump in patients undergoing high-risk percutaneous coronary intervention: the PROTECT II study. Circulation 2012;Aug 30:[Epub ahead of print].

Clinical Topics: Arrhythmias and Clinical EP, Cardiac Surgery, Heart Failure and Cardiomyopathies, Invasive Cardiovascular Angiography and Intervention, Vascular Medicine, Atherosclerotic Disease (CAD/PAD), EP Basic Science, SCD/Ventricular Arrhythmias, Aortic Surgery, Cardiac Surgery and Arrhythmias, Cardiac Surgery and Heart Failure, Mechanical Circulatory Support, Interventions and Vascular Medicine

Keywords: Myocardial Infarction, Follow-Up Studies, Atherectomy, Coronary, Heart-Assist Devices, Arterial Pressure, Cardiopulmonary Resuscitation, Acute Kidney Injury, Electric Countershock, Hypotension, Clinical Trials Data Monitoring Committees, Peripheral Arterial Disease, Percutaneous Coronary Intervention, Tachycardia, Ventricular, Research Personnel, Catheterization, Stroke Volume, Medical Futility, Intra-Aortic Balloon Pumping, Coronary Vessels, Coronary Artery Bypass, Diabetes Mellitus

< Back to Listings