IABP Use: Do We Need to Think More Like Heart Failure Specialists?
You have a unique perspective on mechanical circulatory support. Can you tell us about your background?
I am an interventional cardiologist with advanced training in heart failure and cardiac transplantation. My background in heart failure certainly influences how I approach patients in the cath lab. On the heart failure service, we think about hemodynamics all the time, because that is how we guide patient therapy. Similarly, as an interventionalist, the root of our practice is founded in a comprehensive appreciation for invasive hemodynamics. Since Tufts has a large cardiac transplant and ventricular assist device (VAD) program, our patient population tends to be people with early or late stage heart failure from various causes, anything from inherited myopathies all the way to acute myocardial infarction (MI). For this reason, we have developed an interdisciplinary interventional/heart failure (IHF) team approach that allows us to implement a gamut of mechanical support devices, both percutaneous and surgical, when necessary. For the past five years, when selecting a percutaneous ventricular support device (Figure 1), we have used a decision-making algorithm that blends interventional criteria, like coronary anatomic risk, clinical characteristics, acute MI and acute coronary syndromes (ACS), and heart failure characteristics such as hemodynamic assessment, New York Heart Association (NYHA) class, and Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) class1.These multi-disciplinary evaluations often involve our interventionalists, advanced heart failure specialists, and cardiac surgeons.
My particular expertise focuses on mechanical support devices, cardiac remodeling, and the development of the IHF program at Tufts. The IHF program is a great example of interdisciplinary work leading to better patient care, because decisions often have to be made acutely in the cath lab and having a system in place to rapidly discuss complex patient scenarios, if necessary, is helpful. Everyone, including our technologists and nurses, who are an important part of the IHF team, recognizes that whether we are implanting a mechanical support device, performing a high risk coronary intervention in a patient with advanced heart failure, percutaneously treating an aortic or mitral valve, performing a septal alcohol ablation procedure, or interrogating a patient’s hemodynamic status. We are constantly asking: what is the root cause of their heart failure and what is the best approach to stabilize and improve their quality and quantity of life?
Can you tell us more about the algorithm you use to assess patients for high risk PCI?
For high risk coronary interventions, our algorithm involves four main criteria:
- Assessing the patient’s coronary anatomy, thereby laying out realistic expectations of how difficult the procedure may be;
- Determining the amount of myocardium at risk in a coronary territory;
- Considering other aspects of the patient’s characteristics such as ejection fraction and clinical presentation — is it acute MI, is it ACS, or is it purely an elective percutaneous coronary intervention (PCI)?;
- Determining the hemodynamic status of the patient at the time of PCI (perhaps one of the most important criteria).
In most complex PCI cases, we first assess cardiac filling pressures and function with a pulmonary artery (PA) catheter or at minimum, a pigtail left ventricular end diastolic pressure (LVEDP) measurement. If we find that the patient has massively elevated cardiac filling pressures and if it is an elective or semi-elective procedure, we will often opt to diurese the patient, seeking to normalize their hemodynamics as much as possible, before making the myocardium ischemic during a potentially high-risk coronary intervention. For example, if you take a patient who comes in with a LVEDP of 40 who needs to have a PCI done, that makes the procedure considerably higher risk than if the LVEDP is 10, even if each patient has the same ‘high-risk coronary anatomy’. This is because the first effect of ischemia is to increase LVEDP, or cardiac filling pressures, which can lead to congestion, heart failure, cardiogenic shock, etc.
Percutaneously delivered mechanical support devices are designed to minimize the workload of the heart, by reducing LV pressure and volume, while sustaining systemic perfusion when cardiac function is impaired. For this reason, patients who have low filling pressures and preserved cardiac function may not benefit from a mechanical support device. As I noted, we evaluate the patient from a coronary/anatomic risk perspective, from a clinical presentation perspective, and then also from a heart failure perspective (or a hemodynamic assessment of the patient). If the patient is fortunate enough to only require elective or semi-elective PCI, we will try to optimize their condition with medications and potentially bring them back when they are in better shape. If there is a mandate or need to do the intervention at that time and their hemodynamic status is deranged, as in the case of an acute MI presenting with decompensated heart failure, then we will consider pre-procedural implant of an IABP in those patients. We commonly use an IABP for patients with decompensated heart failure, high-risk PCI, as well as for cardiogenic shock, because of its ease of implantation and global familiarity with the device and how it is used. I also like IABPs because they do not penetrate the heart, whereas other support options require cannulation of the left atrium or left ventricle (Figure 2). Often we find that we do not need to advance beyond an IABP for a more aggressive mechanical support device. One caveat is that if a patient has severely impaired cardiac function and meets high-risk criteria in several of the areas we are measuring: coronary anatomy, clinical presentation and hemodynamics, then there are occasions where we will skip an IABP and go directly to an Impella device (Abiomed), and more rarely, to a TandemHeart device (CardiacAssist), for high-risk PCI. In cardiogenic shock, whether it is an acute MI or not, our general practice is to first put in an IABP. However, if we do not see clinical or hemodynamic improvement with IABP implantation, meaning that the patient is now considered to be in ‘balloon pump refractory cardiogenic shock’, we will then quickly move to a more aggressive mechanical support device, like an Impella device or a Tandem device. The key for most patients with impaired LV function presenting with cardiogenic shock is to assess the hemodynamic status of the patient first and then to take a stepwise approach in escalating mechanical support options beginning with an IABP, then moving towards more aggressive support systems as needed.
What is your take on where we are headed with IABP use?
The field of mechanical circulatory support in the cath lab has a long history, but it seems to be renewing itself. The IABP represents one of the original devices that came out for this purpose, but since then, there have been several other devices that are now widely available, including the Impella and TandemHeart devices, and extracorporeal membrane oxygenation (ECMO).
The IABP, for a long time, dominated the field, especially in the cath lab, because it was easy to implant and there were limited alternatives available. But over the past two decades, with the emergence of the Impella device and the TandemHeart device in cath labs around the world, it has become clear that there is no algorithm to guide use of these devices and worse, no consensus as to how these devices impact left ventricular function. Who should get what device and why we would choose one device over another remains poorly understood. Furthermore, most trials exploring the clinical utility of these devices for high risk PCI or cardiogenic shock do not show mortality benefit for the use of percutaneous mechanical support. The issue is made more complicated since most interventional operators have experienced clinical scenarios where an IABP, Impella, TandemHeart, or ECMO system has saved a patient who otherwise might have died. For this reason, the science behind mechanical support requires further exploration.
Clinically, cardiology is also changing. With advances in coronary intervention, we have essentially traded one disease for another. Nearly 25% of patients who survive an acute MI go on to develop chronic ischemic heart failure. As a result, we are experiencing a tsunami of heart failure patients that are increasing in numbers in the United States and around the world. Historically, in the 1960s and 1950s, the patients would come in with a heart attack, they would be treated medically and as a result, in-hospital mortality rates were about 50%. However, with advances in balloon angioplasty, medical therapy, drugs, etc., there is now a larger population of patients who survived their heart attack and now have heart failure. As an estimate, out of 300 million individuals in the U.S., about 2.6% have heart failure (7.8 million people). Fifty percent of this heart failure population has impaired contractile function or systolic heart failure (3.9 million people), and about 550,000 of those patients have advanced heart failure, meaning they are in NYHA class IIIb or IV. Eventually, this population of heart failure patients will be likely to show up in a cath lab.
As a result, there has been a reinvigorated interest in interventional and invasive-based therapies for these patients. This interest has led the field in terms of performing complex PCI in patients with low ejection fractions and multi-vessel disease: the classic ischemic heart failure patient. It has also sparked interest in taking patients who otherwise would not have survived cardiogenic shock, out-of-hospital arrest, or decompensated heart failure with an acute MI, and in now trying to aggressively manage these patients. In the past, many of these patients would come to the cath lab and get revascularized or undergo an invasive hemodynamic study that would guide drug therapy, but there was really no other exit strategy for patients in advanced heart failure. Today, surgical device technology has grown significantly, and so with more aggressive technological advancement in surgical left ventricular assist devices (LVADs), right ventricular assist devices (RVADs) and bi-ventricular assist devices (bi-VADs), there may be an exit strategy. We can now potentially stabilize and transition these patients to a more definitive therapy, like an LVAD, recovery, or heart transplant. To add to the so-called “tri-borough bridge,” an intermediate option is to stabilize the patient so that a decision can be made as to whether one of these three options is reasonable for this particular patient. We treat many patients who come in the midst of cardiogenic shock. Often, we do not have much information about this patient, and evaluating their candidacy for transplantation or a surgical LVAD can take time. In these cases, placing a percutaneous mechanical support pump, like an IABP, an Impella or a TandemHeart, may be a better way to temporize the situation until we as a caregiving team can get to know this patient and figure out the best treatment strategy. This is my current opinion on where the field of mechanical circulatory support has been and why it is now becoming an increasingly important question that is very clinically relevant to the cath lab.
What has happened to IABPs from a developmental perspective?
IABPs have been around for a long time. From my perspective, innovation in IABPs has focused on optimizing the timing of IABP inflation and deflation using fiberoptics and novel software algorithms. The Mega series (Maquet) represents one of the more recent innovations on the balloon side of the pump technology. An IABP functions very differently from an active percutaneous LVAD. IABPs are designed to displace blood volume in the descending aorta and thereby augment native cardiac function by reducing afterload, thus increasing stroke volume. The other proposed effect of the IABP is to enhance coronary perfusion. The purpose of the Mega series is to essentially take advantage of the mechanism of IABP function, by displacing more volume. The intent is for the Mega balloon to potentially generate a greater reduction in ventricular afterload, potentially further increasing native stroke volume and as a result, cardiac output. That is the premise behind the Mega series, but there are very limited published data to validate the clinical expectation for the device. It has been recently introduced and we use it clinically at Tufts.
How do you decide about pre-procedure use of an IABP?
As described above, this decision depends heavily on the type of procedure. Pre-procedure selection of a support device is continually evolving in our lab. Defining high-risk PCI, with all the advances in the coronary stents and balloon angioplasty designs, is an important moving target and we do not have a great definition in the modern era of what really constitutes a high-risk PCI. Trials attempting to answer this question have not provided clear guidance for operators and for that reason, pre-procedural implantation for support devices like an IABP, Impella or TandemHeart is done on a case-by-case basis (see sidebar).
What is the typical length of use?
For high-risk PCI, it is usually less than 24 hours. In many cases, the devices are used during the procedure, then removed before leaving the cath lab. An important caveat here is to evaluate the post-PCI hemodynamics of your patient before removing a support device. Starting a left main PCI with an LVEDP of 10 and finishing the PCI with an LVEDP of 30 means that the patient may still require prolonged use of the support device. For patients with heart failure and shock, it depends. We have had patients who have been on an IABP for up to 4 weeks or even longer, who are awaiting a surgical LVAD or transplant. The majority of cases, though, are 3-5 days on a support device.
What is the likelihood of patients developing lower-extremity ischemia from long-term IABP use?
The definition of “long-term” remains unclear, but there are important risk factors for someone who could develop limb ischemia from an IABP. Patients with known peripheral vascular disease and pro-thrombotic states are the ones to mostly think about. If you consider the profile of an IABP implant versus other active LVADs like Impella or TandemHeart, the arterial profile is smaller with an IABP, even with the larger volume Mega series. We often implant the IABP without a sheath to reduce the profile within the femoral artery itself, thereby limiting the likelihood of developing limb ischemia. We do about 100-150 IABPs/year. The development of critical limb ischemia, at least in our clinical practice, is likely less than 1-2% of those patients.
What are your impressions of the recent IABP clinical trials?
The IABP has gone through several rigorous studies over its existence (including CRISP-AMI, BCIS and SHOCK II), along with 30 years of registry data. Both CRISP-AMI2 as well as IABP-SHOCK II3 were fascinating. Many people wonder why we are doing these trials. I would reference data generated by Richard Smalling, out of Texas. Dr. Smalling performed preclinical studies that explored a simple question: whether putting in an IABP at the time of acute MI will reduce infarct size.4 He showed that if cardiac workload can be reduced with an IABP, that we can potentially reduce infarct size and improve outcomes in acute MI. In CRISP-AMI, the idea was to implant the IABP immediately before reperfusion of an anterior MI. CRISP-AMI overall was a negative study, but provided some interesting insight in terms of the long-term clinical outcomes, showing a trend towards benefit with IABP plus PCI, versus PCI alone. The IABP-SHOCK II study was another important and large trial that effectively showed that not all patients presenting with an acute coronary syndrome (ACS) with marginal blood pressures and clinical evidence of hypoperfusion should receive an IABP. In my opinion, the importance of this trial is that it confirms what most catheterization laboratories already practice, namely, to avoid non-discretionary use of an IABP in ACS. As a result, the findings of SHOCK II were not all that shocking, in my opinion. Furthermore, entry criteria for cardiogenic shock were based on hypotension and evidence of poor perfusion, without any quantitative measure of hemodynamic status. Many acute MI patients come to us after they have been sitting at home for 2 or 3 days, having intermittent chest pain, not feeling well, and not eating or drinking as well as they might. If we put a PA catheter into these patients, we might find a significant amount to be dehydrated, with low cardiac filling pressures contributing to impaired cardiac output.
For example, a patient who has a wedge pressure of 5, who is hypotensive and has a positive troponin from a non-STEMI, whether it is the right coronary, circumflex or left anterior descending artery, could have been studied in SHOCK II. As part of the trial, this patient could get an IABP. Yet, clinically, as our algorithm tells us, we would expect that patients with low filling pressures may not benefit from a mechanical support device like an IABP. As a result, in my opinion, one of the major limitations for SHOCK II was the fact that quantitative hemodynamic assessment of the patient was not mandated in the trial. Another way to think about this is that the IABP is designed to unload the left ventricle. If the ventricle itself is not loaded, meaning not dealing with high pressure or volume, then the benefit of an unloading device, in a ventricle with low pressure and volume, may be small. This is also why we rarely, if ever, use mechanical support devices in patients with normal cardiac function. Those patients who had a trend toward benefit with the IABP in SHOCK II were patients younger than 50, those having an anterior ST-elevation MI, and those who did not have a history of prior infarction. In other words, a young person having a large anterior infarct, whose heart is now potentially dealing with a massive injury, may be more likely to have elevated cardiac filling pressures and as a result, could potentially benefit from an IABP. For these reasons, IABP-SHOCK II does not change our clinical practice. What it did confirm, in my mind at least, is that non-discretionary use of an IABP in all patients presenting with an ACS and a marginal blood pressure less than 90 may not be beneficial.
Dr. Navin Kapur can be contacted at firstname.lastname@example.org.
- Holman WL. Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS): what have we learned and what will we learn? Circulation. 2012 Sep 11; 126(11): 1401-1406.
- Patel MR, Smalling RW, Thiele H, Barnhart HX, Zhou Y, et al. Intra-aortic balloon counterpulsation and infarct size in patients with acute anterior myocardial infarction without shock: the CRISP AMI randomized trial. JAMA. 2011 Sep 28; 306(12): 1329-1337.
- Thiele H, Zeymer U, Neumann FJ, Ferenc M, Olbrich HG, Hausleiter J, et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock. N Engl J Med. 2012 Oct 4; 367(14): 1287-1296.
- Smalling RW, Cassidy DB, Barrett R, Lachterman B, Felli P, Amirian J. Improved regional myocardial blood flow, left ventricular unloading, and infarct salvage using an axial-flow, transvalvular left ventricular assist device. A comparison with intra-aortic balloon counterpulsation and reperfusion alone in a canine infarction model. Circulation. 1992 Mar; 85(3): 1152-1159.