What is High-Risk PCI, and How Do You Safely Perform It?

You can liken performance of high-risk percutaneous coronary intervention (PCI) to attempting to repair a damaged car engine while it is turned on. Moreover, that car is trying to slowly drive away from a steep cliff with the ocean crashing at its base. If you can’t repair the engine, the car will stall and will slide backward and over the cliff into the ocean. Although slightly melodramatic, this analogy does summarize the consequences of failure after high-risk PCI.

As angioplasty has evolved over the last 30 years, the procedure has gotten markedly safer and more effective.¹ In particular, the advent of highly flexible, highly deliverable coronary stents has eliminated the most catastrophic failure mode of PCI — abrupt coronary occlusion. As PCI has gotten technically safer, anatomic lesion complexity (ACC/AHA lesion Type C) has become less of an issue and patient complexity has become the overriding element in defining high-risk PCI.

Today, high-risk PCI can be likened to the damaged car engine analogy. If transient ischemia or coronary occlusion occurs, rapid cardiovascular collapse ensues. The underlying vulnerability is caused by severe myocardial dysfunction with new ischemia of the residual, functioning myocardium. In 2010, this concept had not been codified. Fortunately, two large randomized trials^2,3 have been recently reported and can be used as a starting point for definition of high-risk PCI. One area where there is broad consensus is that PCI in cardiogenic shock is high risk. Mortalities of 30–50% are reported.⁴ This discussion will not elaborate on acute myocardial infarction (MI) cardiogenic shock PCI, but rather concentrate on urgent or elective non-MI interventions.

Perera et al have reported on a 301 patient, randomized trial of prophylactic balloon pump use.² In this trial, anatomic criteria were used to define high risk. Patients with ejection fraction (EF) ≤30% and jeopardy scores ≥8/12 were included. Importantly, since this trial randomized intra-aortic balloon pump (IABP) use, operators had to have equipoise about need for hemodynamic support. This group was successful in defining a high-risk group since in-hospital mortality (4/301) and 6-month mortality (18/301) is much higher than reported in standard elective PCI trials. Very importantly, while unlike elective PCI trials, a 4-fold increased risk of death occurred in the first 6 months after discharge.

The second randomized trial Protect II³ has been presented recently at the ACC and SCAI sessions. In this trial, high risk was defined as left main intervention with EF <35% or intervention with 3-vessel disease patents with EF <30% or intervention on the last patent coronary conduit. Since the trial tested efficacy of Impella 2.5^™ versus IABP, operators had to conclude that hemodynamic support was required. Thirty-day mortality was 6% and 90-day mortality was 10%.

Thus in 2010, these two studies demonstrated that high-risk patients can be prospectively identified. Two interrelated variables are required: first, underlying severe myocardial dysfunction must exist. Second, extensive zones of myocardial jeopardy during PCI must occur. The other important variable and the feature that differentiates BCIS from Protect II is that operators must feel that hemodynamic support is required. This fact appears to significantly accentuate risk since mortality was markedly higher in the balloon pump arm of the Protect II trial where hemodynamic support was required compared to the BCIS balloon pump arm where uncertainty about the need for support existed.

Now that high-risk PCI can be defined, what can be done to improve outcomes? In this issue of the Journal, a tiered approach to hemodynamic support for high-risk PCI is reported.⁵ Clinicians at Good Samaritan Hospital describe the use of IABP, Impella 2.5, and Tandem Heart. The authors use the probability of circulatory collapse during PCI as the driver of decision making with respect to device selection. In addition, operators chose Tandem Heart when aortic stenosis or severe mitral regurgitation was present and chose Impella when arterial vascular access was problematic for the Tandem Heart Device. When risk of hemodynamic collapse was low, IABP was used; when risk was intermediate, Impella was used; and when risk was high, Tandem Heart was used. Overall mortality for non-shock patients was 2.6%.

The 6-month mortality trends in BCIS and the 90-day MACE trends in Protect II suggest that hemodynamic support does improve intermediate-term clinical outcomes. Choice of device use is complex. IABP has the advantage of long-term experience of use and low profile for device insertion. True augmentation of forward cardiac output is minimal compared to the percutaneous left ventricular assist devices. Impella 2.5 has the advantage of superior forward cardiac flow and direct left ventricular unloading. It requires ability to place a 13 Fr sheath. Tandem Heart supplies greater forward cardiac flow and can be used when aortic stenosis or mitral regurgitation are present. This device requires skill and experience with transeptal puncture and requires a 15 Fr sheath for arterial access.

When hemodynamic support is required, thoughtful planning regarding vascular access is essential. Knowledge of aorta-femoral anatomy is mandatory. Occasionally, performance of peripheral angioplasty is required for large sheath access. This certainly should be done before the coronary procedure and not emergently when the patient is already in a state of hemodynamic collapse.

If only one femoral vessel is adequate, this site should be used for support and radial access for coronary intervention should be obtained. When uncertainty about the need for Tandem Heart support occurs, we have placed a transeptal sheath in the left atrium electively. Thus, if the patient collapsed, the left atrial cannula was already in place. We have found that ease of use and the ventricular unloading effect of the Impella is ideal for most high-risk cases and use this device predominantly unless vascular access is an issue. One final element of planning is communication with anesthesiology and circulatory support staff. The presence of skilled anesthesiology support makes sedation and airway management far safer. Since these devices are infrequently used in most cath labs, dedicated staff trained in use of all three devices is required. Nothing is more frustrating or dangerous than placing a support device and not having it work immediately! For all these reasons, we believe that elective, planned use of a support device is far safer than prophylactic use. It’s far more effective to repair that damaged car engine before the car is falling over the cliff!

References

1. O’Neill WW, O’Neill BP. The first generation of angioplasty. Circ Cardiovasc Intervent. 2009;2(1):6-13. Epub 2008 Dec 15.
2. Perera D, Stables R, Thomas M, et al. Elective intra-aortic balloon counterpulsation during high-risk percutaneous coronary intervention: a randomized controlled trial. JAMA. 2010;304(8):867-874.
3. Protect II: A randomized clinical trial of Impella vs. IABP during high-risk PCI. ACC 2011 Late Breaking Clinical Trials.
4. Jeger RV, Urban P, Harkness SM, et al. Early revascularization is beneficial across all ages and a wide spectrum of cardiogenic shock severity: a pooled analysis of trials. Acute Card Care. 2011;13(1):14-20.
5. Schwartz BG, Ludeman DJ, Mayeda GS, Kloner RA, Economides D, Burstein S. High-risk percutaneous coronary intervention with the Tandem Heart and Impella devices: a single-center experience. J Invasive Cardiol. 2011;23(10):417–424.

____________________________________

From the Office of the Executive Dean for Clinical Affairs, University of Miami Miller School of Medicine, Miami, Florida.
Disclosure: The author has completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr. O’Neill reports research support from Abiomed.
Address for correspondence: William W. O’Neill, MD,  Office of the Executive Dean for Clinical Affairs, University of Miami Miller School of Medicine, 1600 NW 10th Ave – RMSB #1122A (R696), Miami, FL 33136. Email: woneill@med.miami.edu