Commentary: Intraaortic Balloon Pump Versus Percutaneous Ventricular Assist for the High-Risk Patient: Keep it Simple Except When You Can't

Mary Shields, MD and Sheldon Goldberg, MD 

Mary Shields, MD and Sheldon Goldberg, MD 

In this issue of the Journal of Invasive Cardiology, Shah et al analyzed whether two of the percutaneous left ventricular assist devices (pVADs), the Impella 2.5 (Abiomed, Inc) and the TandemHeart (TH; Cardiac Assist, Inc), provide clinical and cost-effective benefit compared with the intraaortic balloon pump (IABP). Based on the authors’ review of four small randomized trials and data from 11,000 patients retrospectively obtained from Medicare files, the investigators conclude that the pVADs offer no significant clinical advantage, while increasing health-care expenditures.1 In the current era, we must take into account the results of cost-effectiveness research, which seeks to balance technological advances with evidence-based results and financial considerations. However, for the clinical interventional cardiologist, the primary objective is always improving outcomes during high-risk percutaneous coronary intervention (PCI). 

In order to put this study into clinical perspective, we review the definitions of “high-risk PCI,” the mechanism of action of support devices for PCI patients, and we comment on the results of trials comparing these devices.

The concept of myocardial protection during PCI is as old as angioplasty itself. Andreas Gruentzig alluded to the use of a roller pump for his initial balloon angioplasty procedures to provide blood flow into the transiently occluded coronary artery; these procedures were performed in patients with predominantly single-vessel disease, which was considered a “high-risk intervention” at the time.2 This method was quickly laid aside when it was realized that transient coronary occlusion was generally well tolerated.

Over the ensuing decades, there have been major advances, most notably the development of flexible and deliverable stents, highly effective adjunctive techniques such as rotational atherectomy, and potent antiplatelet therapy. Because of these improvements, PCI is frequently offered to patients deemed to be at high risk. This population has been variously defined but in general includes a broad spectrum of patients with complex coronary anatomy (ie, unprotected left main, multivessel disease, calcified diffuse disease, bifurcation lesions), poor left ventricular (LV) function, acute myocardial infarction (MI), and cardiogenic shock.3 Although brief interruption of coronary flow is generally well tolerated in patients with sufficient myocardial reserve, the use of multiple dye injections, repeated passes with atherectomy catheters, and placement of multiple stents may tax the reserve of those patients with severe impairment of LV function.

Developed in 1968, the IABP is placed by a simple technique and provides enhanced coronary perfusion, a mild degree of LV unloading, and a modest increase in cardiac output. It has the advantages of familiarity of use and quick insertion through an 8 Fr sheath with the possibility of rapid control of ongoing myocardial ischemia, especially if there is residual flow to the affected myocardium.4 The IABP is contraindicated in the presence of significant aortic regurgitation. Complication rates have dropped through the years to approximately 2.8% and are related to the duration of therapy with the device.3

The Impella 2.5 consists of an axial flow pump inside a pigtail catheter, which is inserted via a 13-14 Fr sheath, and is placed into the LV through standard technique. Compared with the IABP there is greater LV unloading and improvement in cardiac output and cardiac power output. Coronary perfusion is also increased. However, this more complex device has drawbacks compared with the IABP: less rapid institution of support, the need for larger-bore sheaths, and the potential for hemolysis.2

The TH is a percutaneous left atrial to femoral artery bypass powered by an external centrifugal pump.5 Although the TH is capable of increasing the CO up to 3.5-4.5 L/min, it represents the most complex approach for LV support. Complications include cardiac tamponade, limb ischemia, major bleeding, sepsis, arrhythmia, and residual atrial septal defect. The TH is contraindicated in cases of aortic insufficiency, ventricular septal defect, and significant peripheral vascular disease.5

It would be expected that the greater hemodynamic effects of pVADs would translate into improved clinical outcomes. However, a superficial review of randomized comparisons suggest that there is no clear-cut benefit over the IABP. Once again, the idea that an advanced, novel, and more expensive approach would be superior to the old standby is not clearly supported by the available evidence. 

Why should this be the case concerning pVADs? The answers may lie partially in the definitions of what constitutes a high-risk patient, in the limitations of the devices themselves, and in the trial designs.

The definition of what constitutes “high risk” has varied greatly — we’ll give two examples from our recent experience of patients who would have met entry criteria for the PROTECT II trial, the largest randomized comparison of IABP and Impella 2.5.6 Our first patient had a severe ostial left main lesion and an ejection fraction of 30%. Given current stent technology and ease of delivery, we chose an IABP for support. We easily completed the procedure, and removed the IABP after the case. It seems highly unlikely that there would have been an advantage of the Impella 2.5 over the IABP. In fact, we could probably have used a provisional strategy of placing a small-bore sheath in the contralateral femoral artery. The second patient had overt congestive heart failure with an ejection fraction of 25%, hypotension, and ventricular irritability. Coronary angiography showed calcified, severely stenosed left main, left anterior descending, occluded circumflex, and right coronary arteries. We needed to reconstruct the coronary, which supplied the only remaining flow to the myocardium. This procedure required multiple passes with a Rotablator (Boston Scientific Corporation) and placement of five stents. The axial flow pump was chosen to support the circulation, while we concentrated on achieving the optimal angiographic result. We used the preclose technique and removed the Impella shortly after the procedure, thus minimizing the risk of vascular complications or hemolysis. 

As demonstrated by these two vignettes, the interventional cardiologist needs to choose the method of support based on a plethora of clinical considerations, rather than on bottom-line conclusions, which blur the details and complexity of the individual patient. If we examine the results of the PROTECT II trial more carefully, we find a strong trend in favor of the axial flow pump in terms of major adverse events at 90 days.6 Because of the great variability of the patients studied in Protect II, the potential benefits to the sickest patients may have been underestimated. Furthermore, subsequent analysis on the same cohort of patients demonstrated decreased readmission length of stay and consequent charges.7

Cardiogenic shock and acute MI represent a different patient population where LV unloading and circulatory support are more critical. A small randomized trial in patients with acute MI and cardiogenic shock showed greater hemodynamic improvement with the axial flow pump compared with the IABP.8 Although larger studies are needed to examine clinical endpoints, we should not exclude the possibility that pVADs would be beneficial in these critically ill patients. 

For the high-risk patient, the physician should meticulously plan the procedure and be prepared for potential obstacles, including the presence of significant peripheral arterial obstruction, significant aortic valve disease, and extremely complex coronary anatomy.

The physician performing these procedures should be well versed in peripheral angiography and intervention, as well as access and closure techniques when large-bore catheters are placed. If left atrial to femoral bypass is contemplated, transseptal experience is mandatory. The interventionalist needs to also consider whether a provisional strategy of placing a low-profile sheath in the contralateral femoral artery should be employed rather than prophylactic device placement. If LV support is indicated, the IABP will usually suffice. However, in patients with severe LV dysfunction and clinical instability, there is an important role for the pVADs.

In general, simpler is usually better, but for those patients on the far end of the spectrum, the pVADs may be the most efficacious and cost-effective option. The insights provided by retrospective studies from large databases may be of interest regarding general populations, but may not be entirely relevant in solving the problems of our most challenging patients. 


  1. Shah AP, Retzer EM, Nathan S, et al. Clinical and economic effectiveness of percutaneous ventricular assist devices for high-risk patients undergoing percutaneous coronary intervention. J Invasive Cardiol. 2015;27(3):148-154.
  2. Willis Hurst J. The first coronary angioplasty as described by Andreas Gruentzig. Am J Cardiol. 1986;57(1):185-186. 
  3. Jones HA, Kalisetti DR, Gaba M, McCormick DJ, Goldberg S. Left ventricular assist for high-risk percutaneous coronary intervention. J Invasive Cardiol. 2012;24(10):544-550. 
  4. Pepine CJ, Hill JA, Lambert CR. Diagnostic and Therapeutic Cardiac Catheterization. Williams & Wilkins; 1989.
  5. Naidu SS. Novel percutaneous cardiac assist devices: the science of and indications for hemodynamic support. Circulation. 2011;123(5):533-543. 
  6. O’Neill WW, Kleiman NS, Moses J, Henriques, et al. A prospective, randomized clinical trial of hemodynamic support with Impella 2.5 versus intraaortic balloon pump in patients undergoing high-risk percutaneous coronary intervention: the PROTECT II study. Circulation. 2012;126(14):1717-1727. 
  7. Gregory D, Scotti DJ, de Lissovoy G, et al. A value-based analysis of hemodynamic support strategies for high-risk heart failure patients undergoing a percutaneous coronary intervention. Am Health Drug Benefits. 2013;6(2):88.
  8. Seyfarth M, Sibbing D, Bauer I, et al. A randomized clinical trial to evaluate the safety and efficacy of a percutaneous left ventricular assist device versus intraaortic balloon pumping for treatment of cardiogenic shock caused by myocardial infarction. J Am Coll Cardiol. 2008;52(19):1584-1588. 


From the Division of Cardiology Pennsylvania Hospital, University of Pennsylvania Health System, Philadelphia, Pennsylvania.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Address for correspondence: Sheldon Goldberg, MD, FACC, FSCAI, Division of Cardiology Pennsylvania Hospital, University of Pennsylvania Health System, Philadelphia, PA 19107. Email: