Cath Lab Essentials : LV Assist Devices for Hemodynamic Support (IABP, Impella, Tandem Heart) Michael A. Gibson, MD Assistant Professor of Medicine University of California, Irvine Division of Cardiology Talk briefly describes the functioning and assets of the most common devices used today. It gives an overview of the current evidence and indications for left ventricular assist device use in cardiogenic shock and high-risk percutaneous coronary intervention.

Goals To compare and contrast mechanical LV assistance and percutaneous support devices in terms of their designs and ideal applications Review current indications for commonly used devices Describe the factors that should be considered when choosing the most appropriate devices

Causes of Cardiogenic Shock Devices are predominantly used in the management of HF and CS. Most common cause of CS is LV failure. CS complicates 5-10% of acute MI cases. ~40,000-50,000 patients per year in the US. Mechanical complications of MI (VSD, free wall rupture, acute MR) are less frequent causes of CS. Cardiology Clinics, 2013; 31(4): 567-580,

Physiology of Cardiogenic Shock: A Downward Spiral Myocardial Infarction Damaged heart muscle Blood pressure [ BP] 1 Vasodilation [SVR ] Inflammatory activation (TNF-α, IL-6) 6 2 5 NO synthesis Cardiac output [ CO] 4 Myocardial perfusion 3 7 Hemodynamic support 11 12 Myocardial ischemia Venous return 8 BP SVR 10 Hypoperfusion causes release of catecholamines, which increase contractility and peripheral blood flow, but catecholamines also increase myocardial oxygen demand and have proarrhythmic and myocardiotoxic effects. The decrease in CO caused by MI and sustained by ongoing ischemia triggers release of catecholamines, which constrict peripheral arterioles to maintain perfusion of vital organs. Vasopressin and angiotensin II levels increase in the setting of MI and shock, which leads to improvement in coronary and peripheral perfusion at the cost of increased afterload, which may further impair myocardial function. Activation of the neurohormonal cascade promotes salt and water retention; this may improve perfusion but exacerbates pulmonary edema. The reflex mechanism of increased systemic vascular resistance (SVR) is not fully effective, as demonstrated by variable SVR, with median SVR during CS in the normal range. In some patients, SVR may be low, similar to septic shock. In fact, sepsis was suspected in 18% of the SHOCK trial cohort, 74% of whom developed positive bacterial cultures. SVR was lower in these patients, and low SVR preceded the clinical diagnosis of infection and culture positivity by days. Cardiogenic shock can cause the systemic inflammatory response syndrome (SIRS) and suggest that inappropriate vasodilation as part of SIRS results in impaired perfusion of the intestinal tract, which enables transmigration of bacteria and sepsis. SIRS is more common with increasing duration of shock even though levels of interleukin-6 and tumor necrosis factor have been found to be elevated on admission among MI patients who were initially in Killip class I and later developed CS. Cytokine levels rise more dramatically over the 24 to 72 hours after MI. Tumor necrosis factor and interleukin-6 have myocardial depressant action. TNF also induces coronary endothelial dysfunction, which may further diminish coronary flow. Excess NO may also contribute to SIRS. MI is associated with increased expression of inducible NO synthase, which leads to excess NO, which causes vasodilation, myocardial depression, and interference with catecholamine action. Although isoform-nonselective NO synthase inhibitors appeared to improve hemodynamics and outcome in small studies of CS patients, NG-monomethyl-L-arginine at the same dose and duration did not reduce mortality in a large multicenter trial. NG-monomethyl-L-arginine did, however, result in an early blood pressure rise in patients with persistent shock despite vasopressors and opening of the infarct artery, which suggests that excess NO contributes to hypotension. 13 9 Coronary artery perfusion Reperfusion : PCI or CABG CO 14 Reducing inflammatory response: ? Myocardial ischemia Death 15 CO & BP

Mortality CS remains the leading cause of death in MI with mortality rates still 40–60%. GUSTO-I trial 11-year survival in patients with cardiogenic shock complicating MI who survived 30 days. Survival was 67% in non-shock and 28% in shock patients.

Mortality rates decreased in overall cardiogenic shock (from 62 Mortality rates decreased in overall cardiogenic shock (from 62.8% to 47.7%), cardiogenic shock on admission (from 73.8% to 46.6%) and cardiogenic shock developing during hospitalization (from 60.9% to 48.9%)

Rates of PCI and intraaortic balloon counterpulsation increased, whereas rates of fibrinolytic therapy decreased. Rates of PCI increased to greater than 60% and balloon pump use increased to greater than 30%. In contrast, rates of coronary artery bypass graft surgery remained stable.

Emergency revascularisation - SHOCK Trial p=0.03 p=0.11 85% of survivors NYHA Class I/II at 12 months after early revascularization or initial medical stabilization Hochman JAMA 2000;285:190

Heart muscle can recover with support 9 High Potential of heart muscle recovery, Gain in Ejection Fraction Low Potential of heart muscle recovery, Loss in Ejection Fraction The medical community recognizes that, in the immediate hours after an acute MI or heart attack, the heart muscle, if supported properly, has a high potential to recover and regain a healthy ejection fraction. Ejection fraction is essentially a measurement of the heart‘s pumping efficiency and power. The heart muscle is the only muscle in the body that needs constant supply of blood with oxygen or it will switch into a hibernation mode to protect itself. (explain darkened area) In this stage, it will either recover with help or die within hours to days—the dead muscle is called an infarct. If this muscle support is applied too late in the process, more muscle will likely die. Then, the heart will dilate and the muscle wall will thin out (-point to figure-))—resulting in loss of heart muscle which will affect the patient‘s life. In its most basic sense, heart muscle recovery allows a patient to leave the hospital with their own heart. Some causes of heart attacks include clots in the heart from plaque, adverse events from stents or viruses that attack the muscle. The community recognizes this problem- but does not directly treat it. Today, the cardiologists focus on opening up the blockage with PCI—or angioplasty and stents. However, this only addresses the plumbing for the heart and not the pumping Remember, the heart only has one function—and that is to pump blood. The dilemma for the last 40 years has been – How do you support or pump blood with the heart muscle quickly, percutaneously and safely–without causing more harm to the critical patient? THIS is where IMPELLA makes the difference. New England Journal of Medicine: 2003; 348:2007-18

Intra-Aortic Balloon Pump Introduced in 1968 (Kantrowitz) First “true percutaneous” support device Cheapest, commonest (20% of all cardiogenic shock cases), CO 0.5L/min Stabilize pt, but not full support No outcome benefit Hemodynamic Effects: Diastolic pressure  CO/cardiac workload  MAP LV Wall Tension  PCWP Oxygen Demand  LV Volume Coronary Blood Flow 

Onset systole Adjust timing deflate In early inflation or late deflation- increased afterload and increased myocardial O2 consumption In late inflation or early deflation- Decreased physiologic benefit Onset systole deflate Adjust timing

Nair et al Journal of Invasive Cardiology 2011

Nair et al Journal of Invasive Cardiology 2011

Early Trials and Registry Data for IABP

IABP in Cardiogenic Shock Primary PCI Retrospective analysis of 23,180 patients from NRMI database 7268 treated by IABP (trend towards improved mortality)

Time After Randomization (Days) IABP-Shock II Trial: Results Primary Study Endpoint: 30-day Mortality (IABP in Cardiogenic Shock and Primary PCI) Mortality (%) P=0.92 by log-rank test Relative risk 0.96; 95% CI 0.79-1.17; P=0.69 by Chi2-Test The use of intraaortic balloon counterpulsation did not significantly reduce 30-day mortality in patients with cardiogenic shock complicating acute myocardial infarction for whom an early revascularization strategy was planned. METHODS In this randomized, prospective, open-label, multicenter trial, we randomly assigned 600 patients with cardiogenic shock complicating acute myocardial infarction to intraaortic balloon counterpulsation (IABP group, 301 patients) or no intraaortic balloon counterpulsation (control group, 299 patients). All patients were expected to undergo early revascularization (by means of percutaneous coronary intervention or bypass surgery) and to receive the best available medical therapy. The primary efficacy end point was 30-day all-cause mortality. Safety assessments included major bleeding, peripheral ischemic complications, sepsis, and stroke. Thiele H et al. NEJM 2012;367:1287. Time After Randomization (Days)

Indications for IABP High Risk PCI Cardiogenic Shock Refractory Ischemia Left Main 3 Vessel CAD VT/VFib MR or VSD after MI Severe CHF--? Bridge to Transplant Pre-operative stablization Weaning therapy after CABG

Contraindication to IABP Severe Peripheral vascular disease Aortic regurgitation Aortic Dissection PDA HOCM Heparin intolerance Bleeding Diathesis Sepsis

Complications of IABP Vascular Access bleeding/complications Limb Ischemia Infection Thrombocytopenia Migration and aortic arch trauma Other non-vascular (CVA, embolization of cholesterol, balloon rupture) Air embolism risk (reduced by using helium gas)

Impella Continuous axial flow pump Simple insertion Increases cardiac output & unloads LV LP 2.5 – CO 2.5 L/min CP 4.0 L/min 14 F percutaneous approach LP 5.0 21 F surgical cutdown; Maximum 5L flow

Impella insertion The Impella CP is built on the same foundation as the Impella 2.5, but provides more than a 50% increase in pumped blood volume (approx. 4L/min)

Principles of Impella Design Mimic Heart’s Natural Function Inflow (ventricle) Outflow (aortic root) aortic valve EDV, EDP AOP Flow O2 Demand O2 Supply Cardiac Power Output Myocardial Protection Systemic Hemodynamic Support Naidu S S Circulation 2011;123:533-543 24

Systemic Hemodynamic Support CO Increase … Valgimigli et al.,Cath Cardiov Interv (2005) 7.4 L/min Impella 2.5 High risk PCI patient Demonstrated net CO increase with simultaneous ventricular unloading CO Increase 6.0 L/min Impella (2.4) LV unloading Native CO 5.0 25

26 patients with CS secondary to MI randomized to IABP vs Impella 2. 5 26 patients with CS secondary to MI randomized to IABP vs Impella 2.5. The assigned device was implanted following revascularization and measurement of baseline hemodynamic parameters.

Significant rise in serum free hemoglobin due to hemolysis which improved with weaning of the device.

Tandem Heart Left atrial-to-femoral arterial LVAD 21F venous transseptal cannula 17F arterial cannula Maximum flow 4-5L/min Hemodynamic Effects Percutaneously inserted circulatory assist device that pumps blood from the left atrium through a extracorporeally through a centrifugal pump to the femoral arterial system via a transseptally placed left atrial cannula, thereby bypassing the LV. CO  MAP  PCWP  35

Transseptal puncture 21 F cannula in LA 21-Fr transseptal cannula

Tandem Heart and cannulae 15-19 Fr femoral arterial cannula Venous cannula arterial return cannula

Days after Randomization TandemHeart Shock study: Randomized Comparison of IABP with PTVA (VAD) Device in Patients with Cardiogenic Shock Kaplan–Meier Survival Estimates for 30 Day Survival 100 90 80 70 60 50 40 30 20 10 PTVA-VAD: Cardiac Power Index (CO x MBP) Hemodynamic parameters Metabolic parameters Vascular complications PTVA-VAD 57% % IABP 55% P = 0.86 0 5 10 15 20 25 30 Thiele et al, Eur Heart J 2005;26:1276 Days after Randomization

57 yo F w/ PMH of dilated cardiomyopathy diagnosed 4 years ago, DMII, HTN presenting in cardiogenic shock from ECRMC. At this time she had acute kidney injury requiring CRRT, acute liver injury, on Bipap, high dose milrinone. CXR from morning at 9:55

CXR from afternoon at 19:14

CXR from just over a week later.

Compared with the MI-alone group, the percent of total LV myocardium infarcted was reduced by 42% in the MI+unload group (49±14% of all myocardium versus 28±7% of all myocardium, MI versus MI+unload; P=0.03) To account for regional perfusion of the LAD, no differences in LAD length (80±3 versus 78±3 mm, MI versus MI+unload; P=0.4) or LV mass (32±6 versus 31±6 g, MI versus MI+unload; P=0.72) distal to the stent were observed between the MI and MI+unload groups. Circulation. 2013;128:328-336

In both the MI and MI+unload groups, LAD occlusion led to a similar increase in both creatine kinase-MB (4+2% versus 4+1%, MI versus MI+unload; P=0.53) and troponin-I (0.3+0.4 versus 0.6+0.4, MI versus MI+unload; P=0.6) values after 120 minutes of ischemia. In contrast, after 120 minutes of reperfusion, creatine kinase-MB (8.7±1% versus 5.1±0.5%, MI versus MI+unload; P=0.03) and troponin-I (9.7±2 versus 4.5±1.4, MI versus MI+unload; P=0.001) values were significantly lower in the mechanically supported group Mechanically unloading the LV despite delaying coronary reperfusion reduces LV wall stress and may promote myocardial salvage, leading to significantly reduced myocardial injury. This leads into the upcoming TRIS (TandemHeart to Reduce Infarct Size) Trial which will be similar to CRISP AMI. THis will be a prospective, randomized, controlled trial designed to evaluate the use of TandemHeart prior to revascularization as compared to PCI alone in acute STEMI patients. Cardiac MRI will be used to determine MI size and 1 year mortality will be looked at as a secondary endpoint.

Approach to cardiogenic shock Systolic BP > 70-85 mm Hg and good mentation: consider IABP to help prevent shock. Consider IABP in: Bridge to surgery Severe HF Cardiogenic shock (mild to moderate) BP < 70, or on inotropes and vasopressors: consider Impella (2.5, CP, 5L) or TandemHeart (4-5L) Biventricular Cardiogenic Shock: Extracorporeal membrane oxygenation

Optimal Support Device Optimal Timing (early, late, futility) Optimal Support Device Optimal Therapy Optimal management of device (avoiding complications) Euro Heart J(2015) 36, 1223–1230

Questions What are the complications of IABP? vascular complications CVA embolization of cholesterol balloon rupture All of the above

Questions An IABP achieve its action through a counter pulsation A. Deflates during systole B. Inflates during diastole What is used to inflate and deflate the balloon? helium used (reduced chance of air embolism if the balloon were to rupture)