ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2018

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1464

Mechanical Circulatory Support inLeft Ventricular Heart Failure

PETTER SCHILLER

ISSN 1651-6206ISBN 978-91-513-0330-7urn:nbn:se:uu:diva-347427

Dissertation presented at Uppsala University to be publicly examined in Enghoffsalen, Ingång50 bv, Akademiska sjukhuset, Uppsala, Wednesday, 6 June 2018 at 09:15 for the degree ofDoctor of Philosophy (Faculty of Medicine). The examination will be conducted in Swedish.Faculty examiner: Docent Göran Dellgren (Sahlgrenska Akademin).

AbstractSchiller, P. 2018. Mechanical Circulatory Support in Left Ventricular Heart Failure. DigitalComprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1464.68 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0330-7.

Short-term mechanical circulatory support (MCS) with ventricular assist devices or veno-arterial extracorporeal membrane oxygenation (VA ECMO) has become the standard treatmentin patients with cardiogenic shock unresponsive to pharmacological treatment. However, thehaemodynamic effects of these devices are not yet fully described, nor are their effects onventricular function and myocardial recovery.

The aims of this thesis are to increase knowledge of the haemodynamic changes during MCSin different settings and to provide new insights into how MCS therapy should be guided in thespecific patient.

In Studies I and II, we developed experimental animal models to investigate the effectof VA ECMO on left ventricular (LV) performance and size of myocardial infarction indifferent cannulation strategies. In Study I, we found that the LV performance was negativelyaffected by VA ECMO in both centrally and peripherally cannulated animals. In Study II, wespecifically studied the effect of VA ECMO with and without the addition of LV drainage onthe size of experimentally induced myocardial infarction. The results showed that active LVdecompression had no effect on infarct size in the acute setting.

Studies III and IV are retrospective studies on patients in cardiogenic shock treated with short-term mechanical support with either Impella® (Studies III and IV) or VA ECMO (Study IV). InStudy IV, we concluded that treatment with Impella® has excellent effects on haemodynamicparameters and an acceptable mortality and complication rate. The studied pre-implantationpatient parameters did not significantly affect outcome. In Study IV, we compared the outcomeof patients treated with Impella® with those treated with VA ECMO. After adjustment for pre-implantation patient status, as defined by SAVE score, no difference in short- or long-termmortality was seen between the two groups.

In conclusion, VA ECMO, whether central or peripheral, negatively affects the LV, and theaddition of a LV drain has no effect on infarct size in these experimental models. Both Impella®

and VA ECMO offer good haemodynamic results with acceptable mortality and complicationrates in patients with refractory cardiogenic shock. When adjusted for the SAVE score, theoutcomes of both treatment modalities are comparable.

Keywords: Mechanical Circulatory Support, Left Ventricular Heart Failure, Left VentricularAssist Device, Extracorporeal Membrane Oxygenation, ECMO

Petter Schiller, Department of Surgical Sciences, Thoracic Surgery, Akademiska sjukhuset,Uppsala University, SE-75185 Uppsala, Sweden.

© Petter Schiller 2018

ISSN 1651-6206ISBN 978-91-513-0330-7urn:nbn:se:uu:diva-347427 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-347427)

Elegance is refusal

Coco Chanel

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Schiller, P., Vikholm, P., Hellgren, L. Experimental Venoarteri-al Extracorporeal Membrane Oxygenation Induces Left Ven-tricular Dysfunction. ASAIO Journal 2016; 62:518-524.

II Schiller, P., Vikholm, P., Jönsson, A., Hellgren, L. Left Ven-tricular Decompression During Extracorporeal Membrane Oxy-genation (ECMO) Treatment does not Reduce Size of Myocar-dial Infarction in an Experimental Model. Manuscript

III Schiller, P., Vikholm, P., Hellgren, L. The Impella® Recover

Mechanical Assist Device in Acute Cardiogenic Shock: A Sin-gle-centre Experience of 66 Patients. Interact Cardiovasc Thorac Surg. 2016; 22:452-458.

IV Schiller, P., Vikholm, P., Hellgren, L. Survival After Refractory

Cardiogenic Shock is Comparable in Patients with Impella® and Veno-arterial Extracorporeal Membrane Oxygenation (ECMO) – when Adjusted for SAVE Score. Manuscript

Reprints were made with permission from the respective publishers.

Contents

Introduction ................................................................................................... 11 

Background ................................................................................................... 12 Heart failure .............................................................................................. 12 Cardiogenic shock .................................................................................... 13 Heart transplantation ................................................................................ 13 Mechanical circulatory support ................................................................ 14 

History and current use of MCS therapy ............................................. 15 Short-term ventricular assist devices ................................................... 16 The Impella® LVAD ............................................................................ 17 Extracorporeal membrane oxygenation (ECMO) ................................ 18 

Myocardial recovery ................................................................................ 20 Cardiac remodelling............................................................................. 20 Reverse remodelling ............................................................................ 21 

Grounds for ventricular decompression ................................................... 22 VA ECMO and LVAD in AMI ................................................................ 23 Predicting outcome after short-term LVAD and VA ECMO ................... 24 

Aims .............................................................................................................. 27 General aims ............................................................................................. 27 Specific aims ............................................................................................ 27 

Material and methods .................................................................................... 28 Experimental studies (Papers I and II) ..................................................... 28 

Study design ........................................................................................ 28 Animals ................................................................................................ 28 Anaesthesia .......................................................................................... 28 Surgical preparation ............................................................................. 29 The mechanical circulatory assist system ............................................ 29 Measurements ...................................................................................... 29 Experimental protocol ......................................................................... 30 Statistical methods ............................................................................... 31 

Clinical studies (Papers III and IV) .......................................................... 32 Study design ........................................................................................ 32 Statistical methods ............................................................................... 33 

Results ........................................................................................................... 34 Study I ...................................................................................................... 34 

Effects of VA ECMO on left-sided pressures ..................................... 34 Effects of VA ECMO on left ventricular function and volumes ......... 34 Blood gases and biochemical markers during VA ECMO .................. 36 

Study II ..................................................................................................... 36 Study III ................................................................................................... 39 

Patient characteristics and devices ....................................................... 39 Haemodynamic and biochemical changes during support in all patients (n=66) ..................................................................................... 40 Outcome of all patients (n=66) ............................................................ 44 Survivors versus non-survivors ........................................................... 44 

Study IV ................................................................................................... 44 Pre-implantation patient characteristics and MCS system .................. 44 Outcome ............................................................................................... 46 

Discussion ..................................................................................................... 49 The effects of VA ECMO on the left ventricle ........................................ 49 Clinical use of VA ECMO and Impella® in cardiogenic shock ............... 52 

Haemodynamic effects during Impella® support ................................. 52 Survival ................................................................................................ 53 

Limitations ............................................................................................... 54 

Conclusions ................................................................................................... 56 Study I ...................................................................................................... 56 Study II ..................................................................................................... 56 Study III ................................................................................................... 56 Study IV ................................................................................................... 56 

Acknowledgements ....................................................................................... 57 

References ..................................................................................................... 59 

Abbreviations

ACEi Angiotensin converting enzyme inhibitor ALT Alanine transpherase AMI Acute myocardial infarction BiVAD Biventricular assist device BTC Bridge-to-candidacy BTD Bridge-to-decision BTR Bridge-to-recovery BTT Bridge-to-transplant CO Cardiac output CS Cardiogenic shock CPR Cardiopulmonary resuscitation DAP Diastolic arterial pressure DT Destination therapy ECG Electrocardiography ECM Extracellular matrix ECMO Extracorporeal membrane oxygenation ECPR Extracorporeal cardiopulmonary resuscitation EDV End-diastolic volume EF Ejection fraction ESV End-systolic volume HF Heart failure HR Heart rate IABP Intra-aortic balloon pump LA Left atrium LV Left ventricle LVAD Left ventricular assist device LVEDV Left ventricular end-diastolic volume LVESV Left ventricular end-systolic volume LVPsys Left ventricular systolic pressure LVSF Left ventricular shortening fraction LVSV Left ventricular stroke volume LVSW Left ventricular stroke work MAP Mean arterial pressure MCS Mechanical circulatory support PA Pulmonary artery PAP Pulmonary artery pressure PCWP Pulmonary capillary wedge pressure

RA Right atrium RV Right ventricle RVAD Right ventricular assist device SAP Systolic arterial pressure SV Stroke volume SVC Superior vena cava SvO2 Mixed venous oxygen saturation TAH Total artificial heart VA Veno-arterial VAD Ventricular assist device VCF Velocity of circumferential fibre shortening VV Veno-venous

11

Introduction

Most patients with heart failure (HF) can be treated pharmacologically on an outpatient basis, but a proportion of the patients with severe HF are candi-dates for more advanced therapies such as heart transplantation. However, healthy hearts are hard to find, and the supply of donor hearts is not suffi-cient to meet the needs of patients on the transplant list. However, in the last few decades mechanical circulatory support (MCS) devices, commonly re-ferred to as heart pumps, have been developed to provide temporary relief for patients who would otherwise not survive waiting for an available donor organ. The encouraging results of these ventricular assist devices (VADs) have resulted in an increased use, not only to bridge patients to transplanta-tion, but also as destination therapy for patients who are not candidates for heart transplantation. Short-term MCS therapy has become the standard treatment in patients with acute HF unresponsive to pharmacological treat-ment, often with excellent outcomes, and a high proportion of patients re-cover cardiac function and do not need further mechanical support or heart transplantation.1-8 Yet, the haemodynamic effects of these devices are not yet fully described, nor are their effects on ventricular function and myocardial recovery.

12

Background

Heart failure HF is a global epidemic, with a prevalence of more than 26 million world-wide.9 It can be defined as an abnormality of cardiac structure or function leading to the failure of the heart to deliver oxygen at a rate commensurate with the requirements of the metabolising tissues, despite normal filling pressures.10 Clinically, HF is defined as a syndrome in which patients have typical symptoms (e.g., breathlessness, ankle swelling, and fatigue) and signs (e.g., elevated jugular venous pressure and pulmonary crackles) result-ing from an abnormality of cardiac structure or function.11 HF can be caused by myocardial ischaemia, volume or pressure overload (as in heart valve disease and hypertension), infections (myocarditis), or congenital and he-reditary defects in the myocardium (cardiomyopathies).

Although advances in the care of patients with myocardial ischaemia have resulted in a lower incidence of myocardial infarction, ischaemic heart dis-ease remains the leading cause of death on the planet, and the prevalence of ischaemic heart failure is increasing (see Figure 1).12-14 However, improve-ment in the medical care of HF patients has resulted in a steady decrease in HF-related mortality over the past few decades, and the estimated five-year survival rate is now around 50%.15-20 In people over the age of 65, HF is the leading cause of hospital admission in the Western world.21 For advanced HF, when pharmacological therapy fails, the only treatment options are heart transplantation or MCS therapy.

13

Figure 1. Reduction in myocardial infarction and increase in HF. Percentage chang-es in event rate, case fatality rate, and mortality rate for acute myocardial infarction in England from 2002 to 2010 (A). Hospital discharges for HF by sex in the US from 1980 to 2010 (B). Reprinted from Braunwald 22 with permission from Wolters Kluwer Health Inc.

Cardiogenic shock Cardiogenic shock (CS) is the most advanced form of HF, and an immedi-ately life-threatening situation. CS is defined as hypotension (systolic blood pressure < 90 mmHg) despite adequate filling status with signs of hy-poperfusion.23,24 The low cardiac output (CO) results in an inadequate circu-lation and end-organ perfusion, where tissue oxygen demand becomes high-er than the available oxygen supply.24-26 Untreated CS rapidly leads to multi-organ failure and subsequently cardiac arrest and death. CS can result from damage to the heart muscle, as in myocardial infarction, myocarditis or my-opathies, but also from arrhythmias, structural valve disease or congenital cardiac malformations. However, the most common cause is acute myocar-dial infarction (AMI).27 CS treatment is focused on the underlying cause, but inotropes and mechanical circulatory support play an important role in stabi-lising the patient, improving end-organ function, and gaining time for recov-ery or treatment procedures. Although advances in revascularisation proce-dures have been associated with improved outcome, early mortality from CS remains high (29-48%).25,28-30

Heart transplantation Although some experimental work was conducted as early as the beginning of the last century, including heterotopic heart transplantation, it was not until the early 1940s that the concept of rejection was established as the cause of organ malfunction after transplantation.31 After World War II, there

14

was an increasing interest in the field of organ transplantation and substantial research in kidney transplantation was initiated, leading to a better under-standing of the immunological mechanisms of rejection. Based on the new knowledge, the first successful kidney transplant was performed between identical twins in 1960.32 After the introduction of immunosuppressive ther-apy, the first successful kidney transplant between unrelated donor and re-cipient was performed in 1962. Subsequently, the results in kidney trans-plantation improved with the introduction of new immunosuppressive agents such as azathioprine.

Cardiac surgeons were inspired by the encouraging progress in the field of kidney transplantation, and, in 1960, the first successful orthotopic heart transplantation in dogs was performed at the University of Minnesota.33 Six years later, Christian Barnard performed the first human heart transplanta-tion and thereby initiated the concept of heart transplantation as a clinical possibility.34 However, the survival rates were discouraging, and it was not until the introduction of the immunosuppressant cyclosporine in the early 1980s that heart transplantation became a treatment option with an accepta-ble outcome. Since then, the research in the field of transplantation immu-nology has evolved and new, more effective, therapies have been developed. The survival rate is continuously increasing and AB0 incompatible trans-plantation no longer results in poor outcomes.35,36 By the early 1990s, the annual transplantation volume was well over 5,000 worldwide and has re-mained in that region ever since.35

Mechanical circulatory support MCS comprises a variety of different devices that augment or replace the function of one or both of the pumping chambers of the heart. Left ventricu-lar assist devices (LVADs) enhance or replace the left ventricular output by pumping blood from the left atrium (LA) or left ventricle (LV) into the as-cending or descending aorta, while right ventricular assist devices (RVADs) work in the same way on the right heart, pumping blood from the right atri-um (RA) or right ventricle (RV) into the pulmonary artery (PA). Biventricu-lar devices (BiVADs) are essentially a combination of LVAD and RVAD, while the total artificial heart (TAH) actually replaces the entire heart, both anatomically and functionally.

In veno-arterial extracorporeal membrane oxygenation (VA ECMO), on the other hand, venous blood from the RA is pumped through a membrane oxygenator, where the blood is oxygenated and carbon dioxide is removed and returned into the arterial circulation, thus bypassing both the heart and the lungs.

Implantable LVADs are well suited for long-term use or even destination therapy (DT). However, despite recent improvements in the construction of

15

RVADs, BiVADs and TAHs, they are still not optimal for long-term use and are thus only suitable for patients eligible for heart transplantation.

Short-term VADs or VA ECMO are primarily used to support critically ill patients for a limited period of time such as bridge-to-transplant (BTT), bridge-to-candidacy (BTC), bridge-to-decision (BTD), bridge-to-recovery (BTR), or bridge-to-LVAD.37,38

History and current use of MCS therapy Although the clinical use of LVADs to bridge patients to transplantation has expanded in the recent decades, the concept of mechanically pumping the blood, and thereby saving the mortally ill heart failure patient until a match-ing donor heart was found is actually fairly well established. The first im-plantable LVAD was developed and implanted in dogs in 1962.39 Some years later, the idea of intra-aortic balloon counterpulsation was formed, and the first patient series with intra-aortic balloon pumps (IABP) for cardiogen-ic shock was reported in 1968.40 The first TAH was implanted as a BTT device in 1969 and supported the patient until a donor heart was available some 30 hours later.41 In 1984, the first long-term TAH was implanted and the patient was supported for 112 days.42 The first devices were only used in patients in cardiogenic shock or when cardiopulmonary bypass could not be discontinued after heart surgery, but the development of more durable devic-es broadened the indications to include patients with severe HF awaiting transplantation.43 It was soon clear that the use of LVADs in bridging pa-tients to transplant improved haemodynamic, end-organ dysfunction, quality of life, and mortality.44-49 As donor organs were lacking and the survival rate of LVAD patients improved, the concept of DT was formed. The efficacy of LVADs as DT was soon thereafter evaluated in the REMATCH study which demonstrated a significant survival benefit with LVAD compared to medical therapy, and, in addition, an improved quality of life.50 In that era, large, pulsatile devices were used, but the further development of smaller continu-ous-flow devices with longer durability and easier implantation led to an enormous increase in LVAD implantations worldwide. The results from the HeartMate II trial demonstrated improved survival and lower complication rates with the continuous-flow devices compared to pulsatile LVADs, and the outcomes of this therapy are steadily improving.51-53 Most patients in the earlier decades were implanted as BTT, but in later years, the DT patients largely outnumber the patients listed for transplant.54 Still, many patients are implanted as BTT, but, due to the lack of donor organs, in reality many of them are DT patients. In 2013, for example, there were 590 patients implant-ed with a LVAD as BTT in the US, while 1,053 were DT (see Figure 2).22

16

Figure 2. Ventricular assist device implants for prolonged support per year in the US by device strategy. Reprinted from Braunwald 22 with per-mission from Wolters Kluwer Health Inc.

Short-term ventricular assist devices Historically, the most commonly used form of circulatory support is the IABP. Diastolic balloon inflation results in improved coronary and peripher-al perfusion, while systolic balloon deflation results in reduced LV afterload and reduced cardiac work. As the IABP does not actively pump the blood forward, its use is limited to situations where the LV can still eject sufficient volumes, and, for optimal performance, the IABP requires a stable cardiac rhythm, all of which limits the use of IABP in severe CS. Whereas earlier studies suggested a beneficial role for IABP in patients with myocardial infarction and CS, later randomised trials have failed to reproduce these re-sults, and, with the introduction of other percutaneous mechanical devices, the role for IABP in CS is unclear.25,55-62

Centrifugal pumps for short-term LV support with central cannulation of the left atrium or ventricle and the aorta, such as the Biopump or Centrimag pump, have been used for decades, but the invasiveness of the procedure has limited their use to situations where ECC cannot be discontinued after open-heart surgery.

In the last few decades, the role of minimally invasive, percutaneous VADs has increased, and the relatively easy implantation technique has re-sulted in widespread use even outside the cardiac surgery departments. The TandemHeart pump is a paracorporeal centrifugal pump used for left atrial to femoral artery bypass, which can be inserted percutaneously through the trans-septal approach. Pump flows of up to 4.5 L/min can be achieved with this device. The Impella device, which will be discussed in detail later, is an axial-flow pump that is inserted percutaneously through the femoral or axil-lary route, and the pump itself is positioned within the LV cavity with the outflow distal to the aortic valve.

17

VA ECMO provides both perfusion and oxygenation to the organs, and can be used as a short-term support for patients with very severe CS, which will also be discussed later.

Indications for short-term support Short-term MCS devices are designed for use from days to weeks, which limits the indications. The goal with short-term support is to bridge the criti-cally ill patient to recovery or to another device, and rarely direct to heart transplantation. Therefore, the patient must be a candidate for either of these situations. End-organ dysfunction, such as kidney or liver failure, can often be reversed with optimised perfusion over a limited period of time, but car-diac recovery depends on the underlying cause of CS and the probability of recovery must be considered before device implantation, especially when the patient is not eligible for further support with destination devices or heart transplantation.

For this reason, short-term MCS may be considered, and is often lifesav-ing in carefully selected patients with severe CS, where the etiological back-ground may vary. The most common diagnosis leading to CS requiring short-term MCS are: AMI, myocarditis, chronic HF with severe deteriora-tion, and inability to wean from ECC after cardiac surgery.8,37

Another field of use is in high-risk percutaneous coronary interventions (PCI), where the MCS can be used prophylactically during the procedure. In this situation, the support should be started beforehand, and, if the patient is stable, removed immediately after completion of the intervention.8,63,64

The Impella® LVAD Minimally invasive LVADs allow emergent insertion and effective ventricu-lar unloading for a limited time period.65-67 The Impella® LVAD (Abiomed Inc., Danvers, MA) is a micro-axial pump designed to provide short-term left ventricular support with a flow of 2.5-5 L/min (Impella® 2.5, CP or 5.0). The device consists of a catheter-integrated pump that drives blood from the left ventricular cavity to the ascending aorta (see Figure 3). The pump is positioned through the aortic valve into the LV through the femoral or axil-lary artery, and the surgical trauma is therefore minimal. The device is con-nected to a bedside control unit and the speed can be set in 9 stages (P1-P9). As less invasive short-term LVADs such as the Impella® have become more widely available, a wider range of patients in need of mechanical cardiac support may qualify for this treatment.

18

Figure 3. The Impella® axial flow pump positioned across the aortic valve into the LV. Reprinted from Thiele, et al. 67 with permission from Oxford University Press.

Extracorporeal membrane oxygenation (ECMO) The invention of the pump-oxygenator by John Gibbon in the early 1950s allowed cardiac surgeons to arrest the heart during surgery, while maintain-ing circulation and oxygenation. Further development of the heart-lung ma-chine enabled longer support times, and, in the early 1970s, the first patient was treated with prolonged extracorporeal circulatory support for respiratory failure.68 A few years later, the first neonatal patients were successfully treated with ECMO.69 Although most of the early use of ECMO was in the paediatric and neonatal field (to a great extent developed by Mr Bartlett at the University of Michigan), the use of ECMO for various diagnoses has expanded rapidly and today is an established treatment modality for patients of all ages with pulmonary or cardiac failure.

19

There are two main forms of ECMO systems: veno-arterial (VA) and veno-venous (VV). In VA, venous, desaturated blood is pumped from the venous system, oxygenated outside the body and returned into the arterial system, ultimately supporting both the heart and the lungs (see Figure 4). In veno-venous (VV), the desaturated blood is oxygenated outside the body and pumped back into the venous system, thus no cardiac support is provided and this modality is only used for respiratory failure. In the following sec-tions, only VA ECMO will be discussed.

Figure 4. VA ECMO circuit via femoral arterial and venous cannulation. Venous cannulation of the right internal jugular vein is also possible. The venous blood is pumped through the oxygenator and oxygenated blood is returned into the arterial circulation. Reprinted from Lawler, et al. 70 with permission from Wolters Kluwer Health Inc.

Indications for VA ECMO The main indications for VA ECMO are respiratory and/or circulatory fail-ure that are unresponsive to conventional therapy. As ECMO is not well suited for long-term use, the nature of the underlying disease often qualifies or disqualifies a patient for this treatment. Either the condition must be cura-ble so that the ECMO treatment can be weaned within a short time period, or the patient must be a candidate for heart or lung transplantation or DT with LVAD. The indications for VA ECMO expand continuously and a growing field is extracorporeal cardiopulmonary resuscitation (ECPR), where VA

20

ECMO is used when conventional CPR is unsuccessful. Recent studies show promising results of survival to discharge in around 50% of patients if ECMO was initiated within 30 minutes and around 30% when initiated with-in 30-60 minutes.71-75 Due to the easy and rapid implantation procedure and relatively low cost, VA ECMO has, to a large extent, replaced conventional short-term VADs in the treatment of acute cardiogenic shock.

Physiology of VA ECMO In VA ECMO, the blood is pumped from the venous system, oxygenated, and returned into the arterial system, either peripherally into the femoral artery, or centrally in the ascending aorta. In most cases, the peripheral route is used, as it is the least invasive.

Cannulation of the femoral artery results in a reversed blood flow in the aorta, which can lead to an increased LV afterload, which increases myocar-dial oxygen demand. If the heart ejects and the pulmonary function is re-duced, de-oxygenated blood can reach the coronary arteries and the upper body.

Central cannulation by midline sternotomy is a highly invasive procedure associated with major complications such as bleeding and infection, and is almost only performed after cardiac surgery when the sternum is already opened. By central cannulation of the aorta, the blood-flow can be directed in the normal direction, and upper-body desaturation can be avoided.

There have also been speculations that the antegrade flow in the aorta produced by central arterial cannulation would have a positive effect on left ventricular unloading, and some centres advocate for conversion to central cannulation in case of left ventricular dysfunction, although evidence for this strategy is lacking in the literature.

The effect of the increased afterload on the LV has been a concern, and there are reports of cardiac stun, LV dilatation, pulmonary oedema, and elec-tro-mechanical dissociation during VA ECMO.76-82 The success rate with regard to myocardial recovery in patients treated with VA ECMO alone due to cardiogenic shock is, unfortunately, low.83 Different strategies have been developed to unload the LV during VA ECMO. Successful LV decompres-sion has been achieved using: LA or LV vent insertion,79,84,85 atrial sep-tostomy,86-88 IABP, or by the addition of a LVAD.80,82,89-92

Myocardial recovery Cardiac remodelling The term remodelling is used to describe the changes in myocardial mass, shape, volume, and composition of the LV as a result of either physiological or pathological mechanisms. These are primarily adaptive mechanisms serv-

21

ing to enlarge the ventricle and thereby increase CO. However, ventricular dilatation is associated with increased wall stress, as explained by the law of Laplace, which facilitates further dilatation in a vicious cycle. The increase in wall stress leads to cellular and stromal degeneration. The myocardium consists of cardiac myocytes and the connective tissue between the myo-cytes. The underlying causes of ventricular remodelling are changes in myo-cyte number, morphology and function, or changes in the stroma cells or extracellular matrix (ECM) surrounding the cardiac myocytes. On the myo-cyte level, these changes include myocytolysis, myocyte elongation and hypertrophy, beta-adrenergic desensitisation, changes in contractility, and alterations in myocyte mitochondrial function. Changes in stroma include fibroblast proliferation, inflammation, degradation of the ECM, alterations in collagen type and content, and ultimately fibrosis.93

These changes at the cellular level lead to dilatation of the LV. The shape of the LV changes from elliptical to spherical, and the ventricular wall di-lates and becomes thinner, resulting in an increased LV end-diastolic volume (EDV), reduced stroke volume (SV) and thereby reduced CO. Further dilata-tion occurs as the ventricular wall stress increases as a result of increase in EDV.

Reverse remodelling In the last two decades, it has been noted that various treatments of heart failure (pharmacological or mechanical) can induce normalisation of LV size and geometry. This process has been referred to as reverse remodelling,94,95 and can be defined as the normalisation of LV geometry and function with associated improvement in HF symptoms.96 Clinically important reverse remodelling can be assessed by measuring LV volumes and ejection fraction (EF), where reduced volumes and increased EF are signs of reverse remodel-ling.97 Several studies demonstrated reverse remodelling in response to pharmacological therapy with angiotensin converting enzyme inhibitors (ACEi), beta-adrenergic receptor blockers, aldosterone antagonists, and vas-odilators.96 Cardiac resynchronisation therapy with biventricular pacing has also shown promising results with regard to reverse remodelling and mortali-ty.98,99 As LVAD implantations increased, it became clear that some patients recovered cardiac function and the LVAD could be explanted. The marked reduction in LV filling volumes and pressures together with changes in the neuro-hormonal balance are believed to mediate the reverse remodelling seen both clinically and at the cellular level after prolonged LVAD support.100,101 It has been demonstrated that using a strict pharmacological protocol, including beta-adrenergic blockers, ACEi, aldosterone antagonists, and digoxin, and the beta-adrenergic agonist clenbuterol and strict echocar-diographic follow-up, in a not insignificant portion of the patients, the LVAD can be explanted with acceptable survival and morbidity rates.102

22

Grounds for ventricular decompression Decompression of the cardiac ventricles can be accomplished by means of pharmacological therapy such as inotropes or by diuretics to decrease intra-vascular volume. Treatment of hypertension (actually reducing intravascular resistance) or use of IABP reduces afterload and thereby facilitates ventricu-lar ejection. The most effective means of ventricular decompression is, how-ever, mechanical unloading of the ventricle by actively pumping blood from the ventricle with a VAD or by an additional drain to the ECMO circuit.

There are two main motives for ventricular decompression in HF patients. The first and most obvious reason to decompress the ventricle is in the case of congestion leading to either pulmonary oedema (LV failure) or venous congestion with secondary liver and kidney dysfunction (RV failure). The second reason is to minimise cardiac work, cardiac oxygen demand, and ventricular wall stress in order to optimise the probability for ventricular reverse remodelling and recovery.

Addition of a LVAD to the failing ventricle reduces LV pressures and volumes, which results in decreased cardiac work, as seen in the pressure-volume (PV) loops in Figure 5 A.

The use of inotropic agents can be lifesaving by increasing CO in the acute situation, but on the other hand, this results in increased cardiac work and oxygen consumption, which can be deleterious for the failing heart mus-cle, especially in the ischaemic situation.

Figure 5. Cardiac effects of mechanical support. Illustrations of PV loops after acti-vation of device therapy (grey loops). (A) Percutaneous LV assist devices (pLVAD: Impella and TandemHeart) significantly reduce LV pressures, LV volumes, and LV stroke volume. The net effect is a significant reduction in cardiac work-load. (B) Veno-arterial extra-corporeal membrane oxygenation (VA ECMO) without a LV venting strategy increases LV systolic and diastolic pressure, while reducing LV stroke volume. The net effect is an increase in arterial elastance (Ea). Adopted from Rihal, et al. 8 with permission from Elsevier.

23

Several studies reported signs of myocardial recovery after LVAD implanta-tion at the clinical, echocardiographic, histological, and molecular levels.100-

104 Whether those changes are induced by the mechanical unloading of the ventricle or by the normalisation of perfusion, oxygenation and neuro-hormonal pathways induced by LVAD treatment remains to be clarified. In one study, it was shown that LVAD support induced a 28% decrease in LV myocyte volume, with no significant change in RV myocyte volume, indi-cating that the decompression of the LV per se induces myocardial reverse remodelling.105 Although randomised trials are lacking, clinical experience and a few retrospective comparative studies with LV unloading during ECMO suggest a beneficial role for myocardial recovery and improved sur-vival.80,82,84,85,89,90,92,106-108

VA ECMO and LVAD in AMI AMI is still a leading cause of mortality and morbidity. Early reperfusion is the most effective treatment to reduce infarct size, mortality, and morbidity in this group of patients, and the advances in PCI have largely contributed to the improved results during the last few decades. The reported 30-day mor-tality in AMI is, however, 5-15%, and the mortality in those with AMI com-plicated by CS is still around 50%.58,109 The concept of improving myocardi-al oxygen and nutrient delivery after AMI by early revascularisation is well established. Several methods to reduce infarct size by reducing myocardial metabolic activity and oxygen demand such as hypothermia, left ventricular decompression, and β-blockers, have also been developed and have proven to be effective in clinical and experimental studies.110-116

VA ECMO has become a standard treatment for patients in cardiogenic shock, replacing other ventricular assist devices (VAD) in the acute setting.83,117,118 As AMI is a leading cause of cardiogenic shock, an increas-ing number of patients with AMI and cardiogenic shock are being treated with VA ECMO. Although no randomised clinical trials have been per-formed, there is growing clinical experience for increased survival in pa-tients with AMI treated with VA ECMO. However, VA ECMO does not reduce LV work (Figure 5B), and the negative effects seen on LV volumes and pressures during VA ECMO would theoretically affect infarct develop-ment in a negative way. In experimental settings, mechanical decompression of the LV after AMI has been effective in reducing infarct size and ischae-mia/reperfusion injury.114,116,119-121 Recent reports even suggest a shift to-wards early LV decompression and delayed revascularisation in the AMI setting complicated by CS.63,122

24

Predicting outcome after short-term LVAD and VA ECMO Experience from long-term LVAD implantations has demonstrated that with modern devices, the short-term outcome is largely dependent on the timing of device implantation and pre-existing comorbidities, rather than the type of mechanical circulatory device chosen.123 It can also be assumed that the same is true for patients receiving short-term MCS, given that the device can deliver sufficient blood-flow.7,124

These days, VA ECMO and short-term MCS are relatively easy to im-plant and can save patients who, quite recently, would have been beyond rescue. The short-term nature of these therapies, however, makes end-organ and cardiac recovery vital for the next step in patient care. If the patient can be weaned from MCS or bridged to another device the result can be an ex-cellent quality of life. If, on the other hand, the patient does not recover or is ineligible for further device therapy, the decision to discontinue already ini-tiated life-supportive treatment can be extremely difficult with many ethical considerations, especially in the conscious patient. Moreover, the cost of the advanced intensive care of these patients is considerable.

Predicting outcomes in this patient group is therefore of critical im-portance in order to give those patients who have a good chance to recover the best possibilities, as well as to reduce the unnecessary suffering of pa-tients who do not benefit from the MCS therapy.

Recently, the survival after VA ECMO (SAVE) score was constructed to predict hospital survival in patients with acute cardiogenic shock considered for VA ECMO treatment.125 This score is calculated based on 13 pre-implantation variables such as diagnosis, age, haemodynamic and ventilatory parameters, secondary organ dysfunction, and pre-existing comorbidities, Table 1. Thus, the SAVE score provides an opportunity to grade disease severity in these patients in a more objective manner, and has proved to have a better predictive ability among patients on VA ECMO compared with tra-ditional predictive scores for patients in intensive care (i.e., SOFA, APACHE). This is important, as patients receiving short-term MCS are a heterogeneous cohort with different diagnoses and varying degrees of comorbidities, which makes it challenging to compare outcomes between different studies.

25

Table 1. The SAVE score.

Parameter Score

Acute cardiogenic shock diagnosis group (select one or more)

Myocarditis 3

Refractory VT/VF 2

Post-heart or lung transplantation 3

Congenital heart disease −3

Other diagnoses leading to cardiogenic shock requiring

VA ECMO

0

Age (years)

18–38 7

39–52 4

53–62 3

≥63 0

Weight (kg)

≤65 1

65–89 2

≥90 0

Acute pre-ECMO organ failures (select one or more if required)

Liver failurea −3

Central nervous system dysfunctionb −3

Renal failurec −3

Chronic renal failured −6

Duration of intubation prior to initiation of ECMO (h)

≤10 0

11–29 −2

≥30 −4

Peak inspiratory pressure ≤20 cmH2O 3

Pre-ECMO cardiac arrest −2

Diastolic blood pressure before ECMO ≥40 mmHge 3

Pulse pressure before ECMO ≤20 mmHge −2

HCO3 before ECMO ≤15 mmol/Le −3

Constant value to add to all calculations of the SAVE

score

−6

Total score −35 to 17

26

Table 1 cont.

Total SAVE score Risk Class Survival (%)

Hospital survival by risk class

>5 I 75

1–5 II 58

−4 to 0 III 42

−9 to −5 IV 30

≤−10 V 18

An online calculator is available at www.save-score.com.

VT (ventricular tachycardia); VF (ventricular fibrillation). aLiver failure was defined as bilirubin ≥33 µmol/L or elevation of serum aminotransferases (ALT or

AST)>70 UI/L. bCNS dysfunction combined neurotrauma, stroke, encephalopathy, cerebral embolism, as well as seizure

and epileptic syndromes. cRenal dysfunction is defined as acute renal insufficiency (e.g., creatinine >1.5 mg/dL) with or without

RRT. dChronic kidney disease is defined as either kidney damage or glomerular filtration rate <60 mL/min/1.73

m2 for ≥3 months. eWorse value within 6 h prior to ECMO cannulation.

Reprinted from Schmidt, et al. 125 with permission from Oxford University Press.

27

Aims

General aims The general aims of this thesis are to increase knowledge in the haemody-namic changes during MCS in different settings, and to provide new insights into how MCS therapy should be guided in the specific patient.

Specific aims

Study I To experimentally investigate the impact of VA ECMO inserted centrally or peripherally on left ventricular function.

Study II To experimentally study the impact of VA ECMO combined with left ven-tricular decompression on the size of experimentally induced myocardial infarction.

Study III To examine the immediate haemodynamic changes and long- and short-term mortality and complication rate in patients treated with the Impella® Recover device for acute cardiogenic shock.

Study IV To compare short- and long-term outcomes after the use of Impella® or VA ECMO in refractory cardiogenic shock.

28

Material and methods

Experimental studies (Papers I and II) Study design Both studies were experimental. In both studies the animals were random-ised. The study designs were approved by the Uppsala University Ethical Committee on Laboratory Animal Research. All animals received humane care in compliance with Swedish legislation on animal experimentation; (Animal Welfare Act SFS1998:56) and the European Convention on Animal Care (Convention ETS123 and Directive 86/609/EEC).

Animals In both studies, Swedish country-bred pigs with a weight between 33-47 kg were used. Study I comprised 10 animals, whereas Study II comprised 19 animals. All animals survived the preparation and were included in the stud-ies. At the end of the studies, the animals were euthanised with intravenous potassium chloride.

Anaesthesia Anaesthesia was induced by a subcutaneous injection of xylazine 2.2 mg/kg and tiletamine/zolazepam 6.0 mg/kg, and maintained by infusion of a buff-ered 25 mg/mL glucose solution carrier with ketamine 30 mg/kg/h, fentanyl 0.04 mg/kg/h, midazolam 0.1mg/kg/h and pancuronium bromide 0.3 mg/kg/h.

The animals were intubated and mechanically ventilated with 35-40% ox-ygen by a Siemens Servo-I ventilator. Volume-controlled ventilation was used aiming for an arterial pCO2 within the range of 5.0-5.5 kPa and a PEEP of 5 cm H2O was applied. The bladder was catheterised with a silicone catheter. Body temperature was controlled with a heating pad, aiming for a core temperature of 37 C.

Two catheters were inserted through the external jugular vein: one for central venous pressure, blood sampling and drug administration, and the other was advanced into the RV for pressure measurements. A PA catheter was introduced through the right jugular vein and advanced into the PA. For

29

continuous monitoring of the arterial blood pressure, a catheter was inserted through the left femoral artery and advanced into the descending aorta. In Study I, a 5 Fr 12 electrode PV conductance catheter (Ventri.Cath, Millar Instruments, Oxford UK) was introduced through the left carotid artery and advanced into the LV.

Surgical preparation Study I For peripheral cannulation, after administration of 7500 IU of heparin, the femoral artery was cannulated with a 16 Fr artery cannula. For distal perfu-sion, a small cannula was introduced distally into the femoral artery and connected to the main arterial cannula. A 19 Fr venous cannula was intro-duced into the femoral vein and advanced so that the tip of the cannula was located in the right atrium.

For central cannulation, a midline sternotomy was performed. After hepa-rinisation, a 16 Fr arterial cannula was inserted into the ascending aorta, directed towards the descending aorta. The RA was cannulated with a 28 Fr venous cannula.

Study II After midline sternotomy, additional catheters for pressure monitoring were placed directly into the RV, LA, and LV. A 16 Fr arterial cannula was in-serted into the left femoral artery and a 28 Fr venous cannula was inserted directly into the RA. The cannulae were filled with saline and connected to the ECMO circuit.

The mechanical circulatory assist system Arterial and venous cannulae were de-aired and connected to an ECMO system comprising an adult membrane oxygenator and a Bio-Pump. The oxygenator was connected to a heater-cooler for maintaining body tempera-ture.

Measurements Standard lead electrocardiography (ECG), heart rate (HR), systemic arterial blood pressure, PA pressures, and intracardiac pressures were monitored continuously and recorded. Samples for blood gas analysis and plasma bio-chemistry were taken regularly according to the specific protocols. Blood flow in the ECMO system was measured using the Bio-Pump flow probe. For left ventricular drain flow an ultrasonic flow probe was used (Study II).

30

PV curves For PV analysis in Study I, a five Fr 12-electrode pressure-volume conduct-ance catheter was introduced through the left carotid artery, advanced through the aortic valve, and placed in the LV. The catheter was fixed and connected to the Millar MPVS Ultra analysing system (Millar instruments, Michigan, USA).

Proper catheter positioning was determined by monitoring individual segmental pressure-volume loops. Extraventricular segments were excluded from analysis.

The pressure-volume catheter was calibrated for blood resistivity using a rho-cuvette, and the alpha value was calculated using thermodilution-derived CO. PV loops from the LV or RV were recorded for 25 consecutive heart-beats with the ventilation suspended. The following measures were extracted from the pressure volume loops: systolic ventricular pressure, EDV, end-systolic volume (ESV), SV, stroke work (SW), ejection fraction (EF), and maximum rate of ventricular pressure change (dP/dtmax).

Cardiac output In the baseline, CO was measured using the thermodilution technique to calibrate the pressure-volume catheter. Thereafter, in Study I, the pressure-volume catheter-derived CO was used to avoid errors related to the fact that the RV was partially bypassed during ECMO.

Experimental protocol Study I After baseline measurements, peripheral or central VA ECMO was initiated at a blood flow of 100 mL/kg/min. Sweep gas through the oxygenator was set to FiO2 1.0 and the gas-flow was adjusted to achieve normo-ventilation. Heparin was administered at the dose of 5000 IU every hour on ECMO. VA ECMO was continued for five consecutive hours. Measurements were rec-orded every 20 minutes during the experiment. Post-mortem examination was performed to verify correct positioning of the cannulae.

31

Study II After baseline measurements a suture was placed around the LAD to close the vessel. After 60 minutes the ligature was removed and, according to ran-domisation, a LV cannula was placed through the LV apex and connected to the venous line. The animals were randomised into three groups:

1. VA ECMO support and no LV drainage (n=7) 2. VA ECMO support and moderate LV drainage (n=7) 3. VA ECMO support and maximal LV drainage (n=5)

ECMO was initiated and continued for 120 minutes. The animals were eu-thanised and the hearts were perfused with normal saline to remove residual blood. The ligature on LAD was re-applied and the heart was perfused with Evan’s blue until a clear demarcation was seen on the surface of the heart. The heart was explanted and cut into 5 mm thick slices. The slices were then treated with 2,3,5-triphenyltetrazolium chloride (TTC), as previously de-scribed.126,127 The slices were scanned, and the infarct area, as well as the area at risk, was determined with computerised image analysis. The infarct ratio expressed as a percentage of the infarcted area relative to the area at risk was calculated.

Statistical methods Study I For haemodynamic measurements in Study I, the value for each pig was calculated as the mean of all measurements in a treatment period. For con-ductance catheter derived values, the mean value of the loops for each pig in each treatment period was used.

Significance testing was performed using related samples Wilcoxon signed-rank test, comparing the median values between the groups at base-line and ECMO start. Testing for significant changes during ECMO was performed using the repeated measures analysis of variance. P-values equal to or less than 0.05 were considered statistically significant.

Study II Differences in haemodynamics between the three groups during ECMO/reperfusion were analysed by a repeated measurements linear mixed-effect model. The residuals were found to be normally distributed, as ana-lysed by a visual Q-Q plot. In this model the groups were treated as fixed effects, while the respective individual was treated as a random effect. Dif-ferences in the median ratio of infarction size relative to the area at risk were analysed with the Wilcoxon rank-sum test. P-values equal to or less than 0.05 were considered statistically significant.

32

Clinical studies (Papers III and IV) Study design Papers III and IV were retrospective, single-centre observational cohort stud-ies based on patients treated with Impella® or VA ECMO at the Department of Cardiothoracic Surgery, Uppsala University Hospital. The studies comply with the Declaration of Helsinki and were approved by the Regional Ethical Review Board of Uppsala, Sweden.

Paper III The study was designed to investigate the haemodynamic changes, as well as long- and short-term mortality and complication rates in patients treated with the Impella® Recover device for acute CS. This study comprised all patients (n=66) receiving the Impella® Recover mechanical assist device for acute CS at our department between 2003 and 2014. The decision to institute MCS followed the department’s guidelines, including discussion with the multi-disciplinary mechanical assist team. Implantation of the Impella® device was considered in patients with refractory CS, unresponsive to high doses of inotropes. Patients with advanced biventricular heart failure, imminent car-diac arrest or ongoing CPR were supported with VA ECMO and excluded from the study.

Haemodynamic and biochemical data were extracted retrospectively from the patient charts. Survival data were obtained from the population registry so that all patients could be assigned a death date or identified as being alive.

Paper IV In this retrospective study, the outcome of patients with CS treated with ei-ther the Impella® LVAD or VA ECMO was compared. All patients receiving short-term MCS with Impella® or VA ECMO for refractory CS at the de-partment between 2003 and 2015 were included in the study. Patients who received VA ECMO during CPR (ECPR) (n=18) or prophylactic Impella® support during high-risk PCI (n=8) were excluded from the study. Those who received VA ECMO in addition to existing Impella® support were grouped into the ECMO category. One patient in the Impella® group was excluded due to lack of sufficient baseline characteristics to calculate the SAVE score. The remaining 94 patients constituted the study cohort.

Pre-implantation patient characteristics and post-implantation ICU data were extracted from the patient charts. Survival data were obtained from the population registry, as in Study III.

33

Statistical methods Differences in continuous variables were analysed with Student’s t-test, and categorical variables were compared using the χ2-test or, if not applicable, Fisher’s exact test. For comparison of differences in survival, Kaplan-Meier plots and log-rank test (Mantel-Cox) were performed. In Study IV, the prob-ability of ICU survival given the SAVE score was calculated using Cox re-gression analysis. The sensitivity and specificity of the SAVE score to pre-dict survival was also plotted as receiver operator characteristics (ROC) curves, and the area under the curve (AUC) was calculated. P-values of ≤ 0.05 were considered statistically significant.

34

Results

Study I

All pigs were successfully cannulated and included in the study protocol.

Effects of VA ECMO on left-sided pressures During the five hours of treatment, there was a constant decrease in systolic arterial pressure (SAP) in both groups from 104 mmHg (98-113) to 81 mmHg (70-85) (p<0.01); there was no difference in this pattern between groups (p=0.97). The diastolic arterial pressure (DAP) increased from 58 mmHg (53-62) at baseline to 86 mmHg (78-94) at ECMO start in both groups (p<0.01). Thereafter, during the five hours of ECMO, there was a decrease in DAP from 86 mmHg (78-94) to 67 mmHg (61-73) at the end of the experiment (p<0.01), with no difference between groups (p=0.31). The left ventricular systolic pressure (LVPsys) decreased comparably in both groups from 95 mmHg (81-99) at ECMO start to 74 mmHg (54-83) after five hours of ECMO (p=0.03), with no difference between groups (p=0.08). Pulmonary capillary wedge pressure (PCWP) reflecting LA pressure de-creased from 12 mmHg (9.0-14) at baseline to 8.0 mmHg (5.0-11) at ECMO start in both groups (p<0.01), but then remained constant during the five hours of ECMO.

Effects of VA ECMO on left ventricular function and volumes LVSV was unchanged when baseline was compared to ECMO start in the peripherally cannulated group. However, there was a drop in LVSV in the centrally cannulated group from 68 ml (60-71) at baseline to 25 ml (24-28) at ECMO start. Thereafter, there was no significant change in LVSV during the five-hour ECMO period in both groups, and there was no difference in this pattern between the groups (p=0.52). There was no significant change in LV CO during the five hours of ECMO in either group (p=0.99). The LVSW decreased from 2968 mmHg*ml (1480-3940) at baseline to 1661 mmHg*ml (945-1944) at ECMO start in both groups (p=0.03). Thereafter, the LVSW gradually decreased comparably in both groups during five hours of ECMO treatment from 1661 ml*mmHg (945-1944) to 608 ml*mmHg (287-789) at

35

the end of the experiment (p=0.05) (see Figure 6). There was no difference in this pattern between groups (p=0.47).

Figure 6. Left ventricular stroke work in the two groups during VA ECMO.

There was an increase in LV volumes during the five hours of ECMO treat-ment. LVESV increased from 31 ml (28-44) at ECMO start to 55 ml (44-91) (p<0.01) (see Figure 7), and LVEDV from 61 ml (48-71) to 84 ml (60-104) (p<0.01) at the end of the experiment, with no significant difference between groups (p=0.74, LVESV; p=0.46, LVEDV).

The combination of a lower LVSV and increased LVEDV resulted in a decrease in LVEF during ECMO treatment as a function of time on ECMO in both groups. LV dP/dtmax decreased gradually during the five hours of ECMO treatment from 859 mmHg/sec (676-1174) to 516 mmHg/sec (368-567) in both groups (p<0.01) with no difference between groups (p=0.07).

36

Figure 7. Left ventricular ESV in the two groups during VA ECMO.

Blood gases and biochemical markers during VA ECMO Blood gas analysis revealed increased pO2, SaO2, and SvO2 as ECMO was initiated. Plasma analysis revealed a continuous increase in troponin I levels during the experiment from 0.1 µg/L (0.12-0.20) at baseline to 1.65 µg/L (0.84-3.0) at the end of experiment (p<0.01), with no significant difference between the groups (p=0.37). There were minor changes in plasma creati-nine, alanine transpherase (ALT), and total plasma protein levels when base-line values were compared to ECMO start, but thereafter no significant changes were seen.

Study II All 19 pigs survived and remained haemodynamically stable throughout the experiment, and were included in the study protocol.

There were no significant differences in haemodynamic and physiological parameters between the groups during baseline and after LAD ligation. The parameters recorded during reperfusion and ECMO are presented in Table 2. During the reperfusion phase, there were no significant differences in CVP, RV pressures, or aortic pressure (AP) between the groups. The systolic and diastolic PAP, however, were higher in the groups with LV drainage (see Table 2).

37

Table 2. Haemodynamic parameters during reperfusion/VA ECMO in the three different groups: without left ventricular decompression (LVD) and with moderate or maximal LVD.

Reperfusion/VA ECMO

w/o LVD Mod LVD Max LVD P-value

RAP (mmHg) 7 (5-7) 7 (6-7) 5 (4-9) 0.82

RVPdia (mmHg) 5 (4-6) 4 (2-5) 3 (3-8) 0.78

RVPsys (mmHg) 16 (15-17) 16 (13-17) 18 (17-20) 0.48

PAPdia (mmHg) 9 (7-9) 7 (6-8) 11 (10-14) 0.02

PAPsys (mmHg) 15 (14-16) 14 (13-15) 18 (17-21) <0.01

LAP (mmHg) 10 (7-11) 10 (9-10) 7 (2-8) <0.05

LVPdia (mmHg) 8 (7-10) 8 (5-9) 9 (8-10) 0.54

LVPsys (mmHg) 70 (63-73) 60 (47-65) 42 (30-43) <0.01

APdia (mmHg) 51 (49-57) 60 (47-65) 58 (53-60) 0.52

APsys (mmHg) 67 (60-72) 70 (58-75) 65 (60-66) 0.44

CPP (mmHg) 41 (38-50) 52 (39-59) 49 (47-50) 0.39

CI (L/min/m2) N/A N/A N/A -

ECMO-flow

(L/min)

3.0 (2.8-3.1) 3.0 (2.9-3.0) 3.4 (3.3-3.5) -

LV drain flow

(L/min)

N/A 0.17 (0.14-0.21) 0.75 (0.60-0.80) -

pO2 (kPa) 42 (30-70) 64 (44-69) 53 (44-45) 0.80

SvO2 (%) 59 (54-61) 62 (55-63) 45 (45-46) 0.01

Diuresis (ml/h) 114 (51-135) 78 (66-117) 57 (51-57) 0.11

w/o LVD (group without left ventricular decompression) n=7; Mod LVD (group with moderate LVD) n=7; Max LVD (group with maximal LVD) n=5; RAP (right atrial pressure); RVPdia (diastol-ic right ventricular pressure); RVPsys (systolic right ventricular pressure); PAPdia (diastolic pulmo-nary artery pressure); PAPsys (systolic pulmonary artery pressure); LAP (left atrial pressure); LVPdia (diastolic left ventricular pressure); LVPsys (systolic left ventricular pressure); APdia (diastol-ic aortic pressure); APsys (systolic aortic pressure); CPP (coronary perfusion pressure); CI (cardiac index); LV drain flow (left ventricular drainage flow). Values are shown as medians with inter-quartile range within brackets.

The mean LAP was lower in the Max LVD group compared to the control group during the ECMO phase (see Figure 8). The LV systolic pressure also decreased with LV decompression (see Figure 9). There was no difference in coronary perfusion pressure between groups. There was no significant difference in the infarct size in relation to the area at risk between the w/o LVD group and the other groups (w/o LVD: 38.4% (31.1-45.3); Mod LVD: 38.0% (34.2-57.3), P=0.48; and Max LVD: 47.1% (43.6-49.9), P=0.25) (see Figure 10).

38

Figure 8. Left atrial pressure during the reperfusion/ECMO phase without (n=7), moderate (n=7), or maximal (n=5) LVD.

Figure 9. Systolic left ventricular pressure during the reperfusion/ECMO phase with none (n=7), moderate (n=7), or maximal (n=5) LVD.

39

Figure 10. Size of myocardial infarction in relation to area at risk with none (n=7), moderate (n=7), or maximal (n=5) LVD.

Study III Patient characteristics and devices Selected patient characteristics and haemodynamics are presented in Table 3. The mean age of patients was 55 (±1.7) and 65% of the patients were male. The distribution of diagnosis was: 39% AMI; 17% suffered from dilated cardiomyopathy; 17% had myocarditis, and 21% received the device due to post-ECC heart failure. The remaining 6% of the patients received the device for various reasons. The Impella® 5.0 was used in the majority of the patients (66%), Impella® 2.5 in 22%, and Impella® RD in the remaining 12% of the patients (due to RV failure after complex surgical procedures).

For 82% of the patients preoperative echocardiographic data were present. Of these, 68% of the patients had a LV-EF of less than 0.3 and 24% of them had a LV-EF of less than 15% at admittance. The patients received on aver-age 1.3 (±0.1) inotropic drugs before device implantation.

40

Table 3. Characteristics of patients, haemodynamics, and results in the study cohort (n=66) All patients,

n=66 Survivors, n=38

Non-Survivors, n=28

P-value

Age 55 ± 1.7 52 ± 2.4 59 ± 2.0 0.03 Male sex, n (%) 43 (65) 25 (66) 18 (64) 1.00 Hypertension, n (%) 14 (21) 6 (16) 8 (29) 0.34 Diabetes, n (%) 13 (20) 7 (18) 6 (21) 1.00 Duration of support (days) 7.4 ± 0.8 8.5 ± 1.2 5.9 ± 0.9 0.08 Outcome <0.01 Weaned, n (%) 34 (52) 29 (76) 5 (18) Bridged-to-VAD, n (%) 11 (17) 9 (24) 2 (7) Deceased on device, n (%) 21 (32) 0 (0) 21 (75) Complications Stroke, n (%) 3 (5) 0 (0) 3 (11) 0.07 Bleeding, n (%) 23 (35) 13 (34) 10 (36) 1.00 Leg ischemia, n (%) 8 (12) 4 (11) 4 (18) 0.71 Infection, n (%) 20 (30) 12 (32) 8 (29) 1.00 Dialysis, n (%) 29 (44) 13 (34) 16 (57) 0.11 ICU time (days) 24 ± 4.0 33 ± 6.2 11 ± 2.4 <0.01 Follow-up time (years) 2.9 ± 0.4 4.9 ± 0.6 N/A N/A Survival 30 days, % (CI 95%) 58 (47-71) 100 (100-100) 0 (0-0) N/A 90 days, % (CI 95%) 55 (44-68) 95 (88-100) N/A N/A 1 year, % (CI 95%) 47 (36-61) 82 (70-95) N/A N/A 3 years, % (CI 95%) 47 (36-61) 82 (70-95) N/A N/A 5 years, % (CI 95%) 36 (26-50) 63 (48-81) N/A N/A Median survival, months (CI 95%)

7.8 (0.8-65) 118 (45 – N/A) N/A N/A

MAP (mmHg) 73 ± 1.6 74 ± 2.2 73 ± 2.5 0.83 CVP (mmHg) 17 ± 0.8 16 ± 1.0 17 ± 1.2 0.59 Heart rate (min-1) 96 ± 3.1 100 ± 4.6 91 ± 3.9 0.16 CI (L/min/m2) 2.2 ± 0.2 2.1 ± 0.1 2.5 ± 0.3 0.31 CPO (W) 0.66 ± 0.04 0.61 ± 0.03 0.73 ± 0.09 0.40 Number of inotropes 1.3 (±0.1) 1.2 ± 0.2 1.4 ± 0.2 0.67 SvO2 (%) 56 ± 2.0 52 ± 1.2 62 ± 2.4 0.02 Hb (g/dL) 12.1 ± 0.3 12.0 ± 3.8 12.3 ± 3.7 0.66 Creatinine (mmol/L) 129 ± 8 128 ± 11 131 ± 12 0.84 ALT (U/L) 4.3 ± 1.3 4.8 ± 1.9 3.5 ± 1.1 0.59 NT-proBNP (ng/L) 11686 ± 2138 10367 ± 1600 14911 ± 3044 0.45 Lactate (mmol/L) 2.7 ± 0.4 2.8 ± 0.4 2.7 ± 0.5 0.87 If not stated otherwise, values are shown as mean ± SEM. VAD (ventricular assist device); ICU (intensive care unit); CI 95% - 95% (confidence interval); MAP (mean arterial pressure); CVP (central venous pres-sure); CI (cardiac index); CPO (cardiac power output); SvO2 (mixed venous oxygen saturation); Hb (hae-moglobin concentration); ALT (alanine aminotranspherase); NT-proBNP (N-terminal pro-brain natriuretic peptide). P-values for continuous variables were calculated using the Student’s t-test, while p-values for categorical variables were calculated using the chi-square test or, if not applicable, Fisher’s exact test.

Haemodynamic and biochemical changes during support in all patients (n=66) The changes in haemodynamics and biochemical parameters are presented in Table 4. From pre-implantation to day 7, CI increased from 2.1 L/min*m-2 (±0.20) to 3.8 L/min*m-2 (±0.2), respectively, and SvO2 increased from 56% (±2.0) to 68% (±1.3), respectively (see Figure 11 and Figure 12). There was also an immediate and continuous decrease in CVP from 17 mmHg (±0.8) at

41

device implantation to 9.7 mmHg (±0.8) at day 7, Figure 13. Plasma lactate decreased from 2.9 mmol/L (±0.45) at device implantation to 1.2 mmol/L (±0.1) at day 7. Urine output increased from 69 ml/h (±9.1) at device inser-tion to 105 ml/hour (±19) after 7 days on device.

Figure 11. CI before implantation of the Impella® Recover LVAD and during the first 7 days of support in all patients (n = 66). (CI: cardiac index; LVAD: left ven-tricular assist device).

42

Tab

le 4

. Pat

ient

var

iabl

es f

rom

pre

-im

plan

tatio

n to

day

7 o

n su

ppor

t.

P

re

Day

1

Day

2

Day

3

Day

4

Day

5

Day

6

Day

7

MA

P (

mm

Hg)

73

± 1

.6

73 ±

1.1

72

± 1

.3

72 ±

1.3

71

± 1

.3

73 ±

1.7

72

± 1

.9

73 ±

1.8

C

VP

(m

mH

g)

17 ±

0.8

14

± 0

.6

12 ±

0.6

12

± 0

.6

11 ±

0.7

9.

6 ±

0.6

9.6

± 0.

7 9.

7 ±

0.8

CI

(L/m

in/m

2 ) 2.

2 ±

0.2

2.8

± 0.

1 2.

9 ±

0.1

2.9

± 0.

1 3.

2 ±

0.2

3.2

± 0.

2 3.

4 ±0

.2

3.8

± 0.

2 S

vO2

(%)

56 ±

2.0

64

± 1

.2

62 ±

1.2

63

± 1

.2

64 ±

1.5

63

± 1

.8

65 ±

1.2

68

± 1

.3

Uri

ne

outp

ut

(mL

/h)

69 ±

9.1

10

1 ±

12

102

± 11

10

2 ±

12

115

± 14

10

9 ±

14

113

± 16

10

5 ±

19

Inot

rop

es, n

1.

3 ±

0.1

1.8

± 0.

1 1.

5 ±

0.1

1.3

± 0.

1 1.

0 ±

0.1

0.9

± 0.

1 0.

9 ±

0.1

1.0

± 0.

2 H

b (

g/d

L)

12.1

± 0

.3

10.8

± 0

.2

10.1

± 0

.1

10.0

± 0

.1

10.0

± 0

.1

10.1

± 0

.1

9.9

± 0.

1 9.

9 ±

0.2

Cre

atin

ine

(mm

ol/L

) 12

9 ±

8 14

2 ±

10

136

± 10

12

4 ±

9.4

126

± 11

11

3 ±

9.4

128

± 14

12

7 ±

15

AL

T (

µka

t/L

) 4.

3 ±

1.3

17.3

± 5

.7

12.7

± 3

.5

11.4

± 3

.9

8.6

± 2.

8 5.

8 ±

1.8

5.0

± 1.

3 4.

2 ±

1.2

Lac

tate

(m

mol

/L)

2.7

± 0.

4 2.

8 ±

0.4

2.3

± 0.

3 2.

0 ±

0.2

1.7

± 0.

2 1.

6 ±

0.2

1.6

± 0.

2 1.

9 ±

0.5

If n

ot s

tate

d ot

herw

ise,

val

ues

are

show

n as

mea

n ±S

EM

. MA

P (

mea

n ar

teri

al p

ress

ure)

; CV

P (

cent

ral v

enou

s pr

essu

re);

CI

(car

diac

inde

x); S

vO2

(mix

ed v

enou

s ox

ygen

sat

ura-

tion

); H

b (h

aem

oglo

bin

conc

entr

atio

n); A

LT

(al

anin

e am

inot

rans

pher

ase)

.

43

Figure 12. SvO2 before implantation of the Impella® Recover and during the first 7 days of support in all patients (n = 66). (SvO2: mixed venous oxygen saturation).

Figure 13. CVP before implantation of the Impella® Recover and during the first 7 days of support in all patients (n = 66). (CVP: central venous pressure).

44

Outcome of all patients (n=66) The mean duration of circulatory support with the Impella® device was 7.4 days (±0.8), and mean pump flow during the support time was 3.3 L/min (±0.17). In six patients, the device was replaced due to device malfunction. Mean ICU stay was 24 days (±4.0). Out of the 66 patients in total, 52% were successfully weaned from the device, 17% were bridged to long-term VADs and 32% died on support. Mean time to follow-up was 2.9 years (±0.4).

At 30 days, 58% of the patients were alive. Mean time to follow-up was 2.9 years (±0.4). Analysis of outcomes related to diagnosis showed 50% 30-day mortality among AMI patients and 29% 30-day mortality among DCMP and post cardiotomy patients.

Although patients experienced complications during Impella® treatment, including bleeding in 35%, infection in 30%, and leg ischemia in 12%, none of these complications was associated with increased 30-day mortality, as seen in Table 3. Analysis of long-term survival in those who survived 30 days showed 95% at 90 days, 82% at 1 year, 82% at 3 years, and 64% at 5 years. Median survival in those who survived the first 30 days was 9.8 years.

Survivors versus non-survivors When evaluating the pre-implantation haemodynamic parameters and bio-chemical markers, a higher SvO2 was found in non-survivors (62%, ±4.0) compared to survivors (52%, ±1.2), and non-survivors were also older (59 years, ±2.0) than survivors (52 years, ±2.4) (p=0.03). None of the other pre-implant variables differed between survivors and non-survivors.

Study IV Pre-implantation patient characteristics and MCS system The patient characteristics, pre-implantation status, and MCS details of the included patients are summarised in Table 5. The VA ECMO (n=46) and Impella® patients (n=48) were similar in age and sex distribution. The diag-nosis leading to initiation of MCS differed between the VA ECMO and Im-pella® patients (p<0.01). Among the VA ECMO patients, a majority (48%) received MCS due to failure to wean from cardiopulmonary bypass after cardiac surgery.

The prevalence of renal failure, liver dysfunction or CNS symptoms was similar in VA ECMO and Impella® patients. However, the patients who re-ceived VA ECMO were generally in a worse pre-implantation status as de-fined by prolonged ventilator dependency, higher airway pressures, lower

45

blood pressure, and lower SAVE score. Consequently, the VA ECMO pa-tients generally had a worse SAVE class (p<0.01).

Table 5. Pre-implantation characteristics and MCS details among VA ECMO and Impella® patients. VA ECMO

(n=46) Impella®

(n=48) p-value

Patient characteristics - Age, years 57 (15) 52 (14) 0.13 - Male, n (%) 35 (76) 32 (67) 0.43 - Weight, kg 86 (18) 79 (17) 0.05 ‐ BSA, m2 2.1 (0.2) 2.00 (0.2) 0.03 Diagnosis, n (%) <0.01 ‐ AMI 11 (24) 15 (31) ‐ DCMP 2 (4) 8 (17) ‐ Myocarditis 3 (7) 13 (27) ‐ Post-cardiotomy 23 (48) 9 (20) ‐ Arrhythmia 1 (2) 2 (4) ‐ Other 6 (12) 1 (2) ECMO characteristics, n (%) ‐ Central cannulation 13 (28) - - ‐ LV decompression 18 (39) - - Impella® type, n (%) - ‐ 2.5 3 (7) 8 (17) ‐ 5.0 5 (10) 36 (78) ‐ CP 0 (0) 2 (4) ‐ LD 0 (0) 2 (4) Secondary organ dysfunction, n (%) ‐ Acute renal failure 22 (48) 16 (33) 0.22 ‐ Chronic renal failure 5 (11) 6 (13) 1.00 ‐ Liver dysfunction 33 (72) 35 (73) 1.00 ‐ CNS dysfunction 4 (9) 3 (6) 1.00 Ventilation n (%) ‐ Intubated <10 h 24 (52) 37 (77) 0.04 ‐ Intubated 11-29 h 8 (17) 3 (7) ‐ Intubated >30 h 14 (30) 8 (17) ‐ PIP <20 cmH2O 11 (24) 21 (44) 0.07 Circulation, n (%) ‐ Pre-MCS cardiac arrest 11 (24) 11 (23) 1.00 ‐ Diastolic BP >40 mmHg 32 (70) 43 (90) 0.03 ‐ Pulse pressure <20 mmHg 20 (43) 10 (21) 0.03 ‐ HCO3

- <15 mmol/L 6 (13) 3 (6) 1.00 SAVE score, mean -0.4 (6.5) 4.1 (5.4) <0.01 SAVE class, n (%) <0.01 ‐ I 12 (26) 23 (48) ‐ II 7 (15) 10 (22) ‐ III 10 (22) 14 (29) ‐ IV 14 (29) 1 (2) ‐ IV 3 (7) 0 (0) BSA (body surface area); AMI (acute myocardial infarction); DCMP (dilated cardiomyopathy); LV (left ventricle); CNS (central nervous system); PIP (peak-inspiratory pressure); BP (blood pressure)

A considerable proportion (39%) of the VA ECMO patients had some type of concomitant support to prevent left ventricular distention (e.g., IABP, Impella® or left ventricular drainage). Among the Impella® patients, the ma-jority (78% were supported with the Impella® 5.0.

46

Outcome There were no differences in duration of MCS (7 days versus 8 days) or ICU time (24 days versus 23 days) between the VA ECMO and Impella® patients. The proportion of patients weaned from MCS, bridge-to-VAD or heart-transplanted was similar between the two groups (p=0.96).

There was no difference in ICU survival between the VA ECMO (65%) and Impella® patients (63%) (p=0.95). The long-term survival was also simi-lar between the two groups (p=0.39) (see Figure 14). However, there was a tendency to better survival among the Impella® patients one year after MCS treatment: 58% (46-74%) versus 48% (35-65%). This trend persisted in the following years with HR 0.78 (0.45-1.36, p=0.39) for mortality among the Impella® patients versus VA ECMO patients.

The value of the SAVE score to predict ICU mortality was fair, with an AUC in ROC curve of 62% (51-74%). The prognostic value of the SAVE score improved when the outcome was changed to one-year mortality, with an AUC in ROC of 72% (61-82%).

In Cox regression, the SAVE score was found to be a significant covariate affecting long-term survival (p<0.01). In the SAVE score adjusted for sur-vival, the previous tendency to better survival with Impella® treatment di-minishes (see Figure 15). Instead, the SAVE score adjusted for HR for mor-tality was 1.05 (0.58-1.91, p=0.87) for the Impella® patients compared with the VA ECMO patients.

Survival in the two groups was also stratified into SAVE classes I-II ver-sus SAVE classes III-V. This demonstrated that there was no difference in survival between the VA ECMO patients and Impella® patients in neither the best nor worst SAVE classes (p=0.83).

47

Figure 14. Kaplan-Meyer curve demonstrating five-year survival among the veno-arterial ECMO and Impella® patients.

48

Figure 15. Cox regression curve demonstrating five-year survival among the veno-arterial ECMO and Impella® patients adjusted for the mean SAVE score at zero (i.e., 50% expected in-hospital survival).

49

Discussion

Despite recent advances in the pharmacological and interventional treatment of patients in cardiogenic shock, mortality remains as high as 50-59%.58,128 Although widely used, the IABP has proven ineffective in reducing mortality in this patient population.58 LVADs have recently become more easily avail-able and experienced an increased use in acute heart failure patients.7 Im-plantable LVADs or paracorporeal mechanical assist systems such as the Centrimag pump are, however, associated with major surgery and are at a high risk for complications. In the past decade, VA ECMO has experienced an increased use in patients in cardiogenic shock.118 Although VA ECMO gives sufficient support to prevent or even reverse multi-organ failure, clini-cal experience suggests that the LV is often not sufficiently unloaded, and therefore left ventricular recovery might be impaired.76-79,82,92 Minimally invasive LVADs allow emergent insertion and effective ventricular unload-ing for a limited time period.65-67 The Impella® Recover is a micro-axial pump designed to provide short-term left ventricular support with a flow of 2.5 to 5 L/min. The pump is positioned in the LV through the femoral artery and the surgical trauma is thus minimal. As less invasive short-term MCS systems such as the Impella® Recover or VA ECMO become more widely available, a wider range of patients in need of mechanical cardiac support may qualify for this treatment. Studies of the haemodynamic and biochemi-cal changes during support, as well as follow-up data in relation to pre-implantation characteristics in patients supported with VA ECMO or the Impella® device are scarce.

The effects of VA ECMO on the left ventricle Several ECMO centres experienced good clinical outcomes such as reversal of pulmonary oedema when volume unloading of the left atrium or ventricle has been added to the conventional VA ECMO.129 LV volume unloading can be achieved in several ways, including left atrial or ventricular vent inser-tion,79,84,85,106 atrial septostomy,86,87 IABP, or by addition of a LVAD such as the Impella® or IVAC® pump.80,89,90 Although there are several reports on positive effects of LV unloading during VA ECMO, most of them are exclu-sively in the form of case reports. Recently, retrospective studies combining VA ECMO with LV unloading have shown promising results; however,

50

randomised studies are still lacking.88,107,108 In the field of LVADs for long-term use, there are data suggesting that unloading of the LV induces reverse remodelling and can contribute to ventricular recovery.104 It remains to be clarified if those findings are applicable to short-term ventricular support such as VA ECMO.

In Study I, we developed an experimental model of peripheral and central VA ECMO in pigs with normal cardiac function, which enabled us to further investigate the haemodynamic changes during VA ECMO support. We found that the LV was not unloaded during VA ECMO, regardless of cannu-lation site. In complete contrast, the LVESV increased during the ECMO course and the LVEDV showed a similar pattern, resulting in a continuous dilatation of the LV. As a consequence of LV distension, the LVEF was gradually decreasing during ECMO. Both LV dP/dTmax and LVSW de-creased constantly during ECMO support in both groups, indicating that VA ECMO negatively affected LV contractility.

Some centres advocate conversion to central cannulation in selected pa-tients on VA ECMO to facilitate LV unloading, but, according to our results, this strategy cannot be supported. The findings in our study suggest that VA ECMO, whether the cannulation is central or peripheral, induces LV disten-sion even in the normal heart. This could be deleterious as ventricular dilata-tion per se can induce a vicious cycle reducing ventricular performance even further. Most patients receiving VA ECMO have at least some degree of LV dysfunction at the time of ECMO initiation, which would likely result in more severe LV distension than that observed in this study. Therefore, un-loading the LV in the setting of VA ECMO, regardless of cannulation mode, seems to be an appealing strategy, at least if there are signs of LV distension. This is also the case in clinical practice, as seen in Study IV, where approx-imately 40% of the VA ECMO patients received some sort of device to facil-itate LV unloading.

LV decompression during VA ECMO for acute myocardial infarction In the special, but not infrequent, case where VA ECMO is initiated due to AMI with cardiogenic shock, optimal LV pressure and volume status is cru-cial in order to ameliorate the potential for ventricular recovery. High intra-ventricular pressure combined with low arterial blood pressure leads to de-creased coronary perfusion pressure and coronary blood flow, at the same time as myocardial oxygen demand increases. There is increasing evidence that in AMI, early reduction of intracardiac pressures and volumes, thereby reducing ventricular wall stress, have a role in reducing infarct size, reducing ischemia/reperfusion injury, and improving myocardial recovery.63,112,114-

116,119,120,130-135 In Study II, we hypothesised that active LV unloading during the reperfu-

sion phase after AMI and VA ECMO would reduce infarct development. The main finding was that additional LV decompression did not reduce size

51

of infarction during VA ECMO. However, the size of the infarction in rela-tion to area at risk was smaller in all groups than previously described when mechanical support was not used.119,136 Previous studies show a reduction in infarct size relative to the area at risk from 65-72% without support to 17.5-41% with mechanical ventricular support during reperfusion, which is in a similar range as our findings on VA ECMO.114,119,136 A possible explanation for our results would be that VA ECMO per se unloads the LV sufficiently during the first few hours to reduce the infarct size, and that additional LV decompression has no added value in reducing infarct size. The increase in CPP and coronary blood flow during ECMO might also contribute to re-duced infarct size.

In clinical practice, ECMO is now an established treatment for patients in cardiogenic shock, and is used more frequently in those showing refractive cardiac arrest.71,73,117,122,137,138 A large proportion of these patients suffer from myocardial infarction.72,74,139 In patients with cardiac failure, there is increas-ing evidence towards mechanical LV decompression during VA ECMO in order to improve myocardial recovery.92,129,140,141 However, the impact of VA ECMO and LV decompression on infarct development has not previously been studied. Our results indicate that additional mechanical decompression of the LV during VA ECMO in patients with AMI does not reduce the in-farct size. That said, the role for ventricular decompression in respect to re-covery of myocardial function has not been investigated in this study.

In our experience, a large proportion of the patients showing AMI and cardiogenic shock have extensive coronary disease and/or a history of heart failure. In these patients LV contractility is frequently extremely reduced to the extent where there is no aortic valve opening during VA ECMO. Active LV decompression is in our opinion indicated in this subgroup of patients, not only to reduce infarct development, but also to prevent pulmonary oe-dema and intracardiac thrombus formation.

Dilatation of the LV during VA ECMO is slowly progressive (as seen in Study I) and LV decompression may still be beneficial to prevent long-term negative effects of VA ECMO on infarct size. The clinical implications of our findings might be that immediate decompression of the LV may not be important in the setting of AMI and established VA ECMO, but may still be indicated if there are clinical or echocardiographic signs of LV distension, or when there is no aortic valve opening. Thus, LV decompression may be de-layed in most patients and in certain cases not needed at all.

52

Clinical use of VA ECMO and Impella® in cardiogenic shock The use of short-term MCS in patients with refractory cardiogenic shock has increased over the last few decades.124 Although the results have improved, mortality still remains high.124 Thus, efforts must be made to improve out-comes in these patients.

The most commonly used MCS systems in refractory cardiogenic shock are short-term LVADs and VA ECMO, since the IABP has proven ineffec-tive in randomised clinical trials (i.e., IABP-SHOCK II).7,25 Their indications are largely overlapping, with exception for severe RV failure, severe respira-tory dysfunction, and severe haemodynamic instability, where VA ECMO is the preferred choice.142,143 Over the past few years, VA ECMO has experi-enced an increased use in emergent situations,74,117,139 but the negative effect on the left ventricular loading conditions may, however, impede ventricular recovery in conditions where recovery otherwise is a possibility.76,82,144 Im-plantable LVADs for long-term use have shown very good haemodynamic effects and good results in elective patients. Yet, in emergent situations, like cardiogenic shock, the results are worse.54,145

Nonetheless, the evidence for these therapies is scarce, and therapeutic strategies have largely emerged through local traditions at cardiothoracic centres.7,142 As a result, patient selection and preferable assist devices, and indications might vary considerably amongst clinics.142 Experience from long-term LVADs has demonstrated that with modern devices, short-term outcome is largely dependent on the optimal timing of implantation of sup-port and pre-existing comorbidities, rather than the type of mechanical circu-latory device chosen.123 It is likely to assume that the same is true for out-comes in patients receiving short-term MCS, given that the device can deliv-er sufficient flow.7,124

Haemodynamic effects during Impella® support In Study III, we aimed to determine the immediate haemodynamic effects of the Impella® Recover device. We found that circulation could be restored, shown by increased CI, decreased CVP, increased SvO2, and decreased need for inotropic support. Signs of improved end-organ function such as in-creased diuresis and decreased serum lactate levels were shown during sup-port.

In small patient series, previous studies show similar improvements in haemodynamic parameters as in our study, during support with the Impella® Recover device. Four clinical studies demonstrate increased MAP, CI or CO in patients on Impella® support.6,61,63,146 Furthermore, the decrease in serum lactate seen in our study was previously demonstrated.61,146 The decreasing need for inotropic support that we found has previously been described in

53

one study.6 To our knowledge, no previous study has demonstrated the re-duction in right-sided pressures over time on support that we observed. This is an important finding, since increased central venous pressures (not only impairment of CO) is gaining in importance as a factor strongly associated with worsening of renal function among patients with advanced heart fail-ure.147

Survival In Study III, we report one of the largest single-centre experiences with the Impella® device, consisting of 66 patients. Although this is a retrospective analysis, we believe that the substantial number of patients adds additional information with regard to haemodynamic effects and outcome.

In most studies of patients in acute cardiogenic shock the 30-day mortali-ty rates are approximately 40-60% regardless of the chosen device.3,4,7,64,66,83,148 The early mortality rate in Study III was 42%, and, in Study IV, was 35% for the Impella® patients and 41% for VA ECMO pa-tients, which is in line with, or less than, earlier reports in these categories of high-risk patients.64,118,149 Interestingly, in Study III, none of the pre-implant parameters alone were found to have a significant impact on outcomes. In a recent report only blood lactate levels had an impact on outcomes.64 The survivors had higher urine output and lower blood lactate levels during the first few days of support, whereas none of the other parameters measured during the first days of support differed between survivors and non-survivors. It should be noted that once 30-day survival was achieved, the prognosis of 5-year survival was relatively good, which was confirmed in Study IV.

Patients receiving short-term MCS are a heterogeneous group, and have, up until recent years, been relatively few in number.124 Furthermore, indica-tions and type of short-term MCS to be chosen in refractory cardiogenic shock have been a matter of debate and still vary amongst clinics.7,142 Given these circumstances, solid evidence such as from large randomised trials is lacking.7,142 This is not only for selection of device, but also for the use of MCS in cardiogenic shock in general, which currently only has a Class IIB/C recommendation in European and American guidelines (i.e., expert opinions, largely based on retrospective studies, experimental evidence and clinical experience).7,8 Until randomised trials are in place, the aim should be to cor-rect for disease severity in studies comparing different MCS approaches.

In Study IV, we present the first report comparing outcomes between VA ECMO and Impella® LVAD adjusted for disease severity. The main finding was that both short- and long-term survival was similar among patients treat-ed with Impella® or VA ECMO due to cardiogenic shock, after adjustment for disease severity through the SAVE score. Moreover, MCS duration, ICU time, proportion of patients weaned, bridged-to-VAD, and heart-transplanted

54

was similar between the LVAD and VA ECMO patients. Thus, outcomes seem independent of type of device, and are most likely a complex result of the diagnosis and general status of patients at device implantation. Although VA ECMO patients needed ventilator support for a longer time and required CRRT to a greater extent, this is most certainly related to poorer baseline characteristics rather than type of treatment. In many cases with severe cir-culatory instability and respiratory dysfunction, Impella® treatment would not have been a viable option, which is illustrated by the fact that a few pa-tients who initially received Impella® treatment had to be converted to VA ECMO.

However, our study gives no support for choosing VA ECMO treatment over short-term LVAD only based on poor SAVE score if the patients are relatively circulatory stable and have a manageable respiratory function. Although a greater proportion of the VA ECMO patients were in SAVE classes III-V, the Impella® patients in SAVE classes III-V demonstrated just as good survival rate as the VA ECMO patients.

Moreover, to our knowledge, this is the first study to demonstrate that the SAVE score is not only valid in ECMO patients, but can also predict surviv-al in patients receiving Impella® treatment due to cardiogenic shock. Given these findings, it is reasonable to believe that the SAVE score can be used as a tool to score disease severity in severe cardiogenic shock in general. This offers future studies a more accurate comparison between different patient cohorts and interventions.

Limitations These studies have, of course, several important limitations. Firstly, Studies I and II are experimental animal studies, and even if the porcine anatomy and physiology resembles that of humans and the porcine heart is extensively studied in ischemia/reperfusion studies,150 the findings cannot be directly translated into clinical practice. Secondly, due to the nature of these animal studies, only the acute effects were studied. Patients receiving these kinds of treatment are treated for longer time periods, ranging from days to weeks, and our results cannot be extrapolated over time. Additionally, we have stud-ied the effects on these treatment modalities in otherwise healthy hearts, whereas in reality patients without HF are not candidates for these types of treatments for obvious reasons.

Studies III and IV are retrospective in their nature, and therefore have several important limitations. Although most haemodynamic and biochemi-cal data are routinely recorded in our department, due to the often emergent nature of Impella® or VA ECMO implantation, some patients have missing data, which affected the statistical analysis. Specifically, detailed infor-mation on haemodynamic status before device implantation (e.g., cardiac

55

index, mixed venous saturation) was also lacking since no preimplantation right heart catheterisations were performed in acute patients, and were not possible to perform in those patients with failure to wean from cardiopulmo-nary bypass. Moreover, the comparison of patient cohorts based on a score in Study IV cannot replace the need for randomised trials; it can simply serve as a complement.

56

Conclusions

Study I Regardless of cannulation site, LV function and volumes were progressively negatively affected during VA ECMO treatment. Therefore, central cannula-tion should not be used as a means to improve LV function or reduce LV dysfunction during VA ECMO. Other means may be needed in order to un-load the LV during VA ECMO.

Study II Although LV decompression during VA ECMO significantly reduced left-sided pressures, there was no effect on size of infarction in this experimental AMI model and cardiogenic shock. A possible explanation could be that VA ECMO unloads the LV sufficiently to reduce the infarct size, and that addi-tional LV decompression in the acute setting does not provide further infarct size reduction.

Study III The Impella® device offers a minimally invasive treatment option in critical-ly ill heart failure patients, with excellent effects on haemodynamic parame-ters and with an acceptable mortality and complication rate. None of the analysed pre-implantation parameters alone had a significant impact on out-comes for the studied patient group.

Study IV Short- and long-term survival is similar in patients treated with Impella® or VA ECMO due to refractory cardiogenic shock, after adjustment for disease severity through the SAVE score. Furthermore, the predictive ability of the SAVE score is not only valid in VA ECMO patients, but also in patients receiving Impella® treatment due to cardiogenic shock.

57

Acknowledgements

This thesis would not have been possible without the help and support that I have received from many people over the years. In particular, I would like to express my sincere gratitude to:

Laila Hellgren, primary supervisor, professional mentor and friend who both encourage and criticise me when I need it the most! Per Vikholm, my co-author, and great friend, for helping me with every-thing, in particular for your invaluable help in finding statistical means to test our hypothesis (and finding out we were almost always wrong…). Your never-ending enthusiasm makes life at work much easier. My co-supervisor Anders Jönsson, for invaluable input and for sharing all your stories with us at work, you make the daily work much more interest-ing. My boss, Vilyam Melki, for giving me the time to finish my thesis and for always believing in me. Colleagues and staff at the department of Cardiothoracic Surgery, especially I would like to thank Karl-Henrik Grinnemo, for valuable help both in the process of completing my doctoral studies and in clinical practice every day, and Rafael Astudillo, for sharing your clinical experience and being a great tutor.

All the personnel at the Hedenstierna Laboratory for technical assistance. My parents, Pelle and Lullan, who always believe in me and make this all possible. All my dear friends, in particular Petra, Mattias, Zion, Saga, Yasmine, Andrew, Jakob, Noël, Emma, and Ellen. Spending time with you is always lovely, and I hope we can meet more often now.

58

Finally, my family, who is always there for me. Our fantastic children Victor and Agnes, you brought so much joy into my life. Samuel, you re-main my dearest and best friend and the love of my life. I love you all!

59

References

1. Mohite, P.N., et al. Short-term ventricular assist device in post-cardiotomy cardiogenic shock: factors influencing survival. J Artif Organs 17, 228-235 (2014).

2. Takayama, H., Takeda, K., Doshi, D. & Jorde, U.P. Short-term continuous-flow ventricular assist devices. Curr. Opin. Cardiol. 29, 266-274 (2014).

3. Takayama, H., et al. Clinical outcome of mechanical circulatory support for refractory cardiogenic shock in the current era. J. Heart Lung Transplant. 32, 106-111 (2013).

4. Lauten, A., et al. Percutaneous left-ventricular support with the Impella-2.5-assist device in acute cardiogenic shock: results of the Impella-EUROSHOCK-registry. Circ. Heart Fail. 6, 23-30 (2013).

5. Lemaire, A., et al. The Impella device for acute mechanical circulatory sup-port in patients in cardiogenic shock. Ann. Thorac. Surg. 97, 133-138 (2014).

6. Griffith, B.P., et al. The RECOVER I: A multicenter prospective study of Impella 5.0/LD for postcardiotomy circulatory support. J. Thorac. Cardio-vasc. Surg. 145, 548-554 (2013).

7. Werdan, K., Gielen, S., Ebelt, H. & Hochman, J.S. Mechanical circulatory support in cardiogenic shock. Eur. Heart J. 35, 156-167 (2014).

8. Rihal, C.S., et al. 2015 SCAI/ACC/HFSA/STS Clinical Expert Consensus Statement on the Use of Percutaneous Mechanical Circulatory Support De-vices in Cardiovascular Care: Endorsed by the American Heart Assocation, the Cardiological Society of India, and Sociedad Latino Americana de Cardi-ologia Intervencion; Affirmation of Value by the Canadian Association of In-terventional Cardiology-Association Canadienne de Cardiologie d'interven-tion. J. Am. Coll. Cardiol. 65, e7-e26 (2015).

9. Ponikowski, P., et al. Heart failure: preventing disease and death worldwide. ESC Heart Fail 1, 4-25 (2014).

10. Dickstein, K., et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the diagnosis and treatment of acute and chronic heart failure 2008 of the European Society of Cardiolo-gy. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur. J. Heart Fail. 10, 933-989 (2008).

11. McMurray, J.J., et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur. Heart J. 33, 1787-1847 (2012).

12. Moran, A.E., et al. The global burden of ischemic heart disease in 1990 and 2010: the Global Burden of Disease 2010 study. Circulation 129, 1493-1501 (2014).

60

13. GBD 2013 Mortality and Causes of Death Collaborators. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 385, 117-171 (2015).

14. Roth, G.A., et al. Global, Regional, and National Burden of Cardiovascular Diseases for 10 Causes, 1990 to 2015. J. Am. Coll. Cardiol. 70, 1-25 (2017).

15. Roger, V.L. Epidemiology of heart failure. Circ. Res. 113, 646-659 (2013). 16. Roger, V.L., et al. Trends in heart failure incidence and survival in a commu-

nity-based population. JAMA 292, 344-350 (2004). 17. Levy, D., et al. Long-term trends in the incidence of and survival with heart

failure. N. Engl. J. Med. 347, 1397-1402 (2002). 18. MacIntyre, K., et al. Evidence of improving prognosis in heart failure: trends

in case fatality in 66 547 patients hospitalized between 1986 and 1995. Cir-culation 102, 1126-1131 (2000).

19. Mosterd, A., et al. The prognosis of heart failure in the general population: The Rotterdam Study. Eur. Heart J. 22, 1318-1327 (2001).

20. Cowie, M.R., et al. Survival of patients with a new diagnosis of heart failure: a population based study. Heart 83, 505-510 (2000).

21. Ambrosy, A.P., et al. The global health and economic burden of hospitaliza-tions for heart failure: lessons learned from hospitalized heart failure regis-tries. J. Am. Coll. Cardiol. 63, 1123-1133 (2014).

22. Braunwald, E. The war against heart failure: the Lancet lecture. Lancet 385, 812-824 (2015).

23. Ponikowski, P., et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the diagnosis and treat-ment of acute and chronic heart failure of the European Society of Cardiolo-gy (ESC)Developed with the special contribution of the Heart Failure Asso-ciation (HFA) of the ESC. Eur. Heart J. 37, 2129-2200 (2016).

24. van Diepen, S., et al. Contemporary Management of Cardiogenic Shock: A Scientific Statement From the American Heart Association. Circulation 136, e232-e268 (2017).

25. Thiele, H., et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock. N. Engl. J. Med. 367, 1287-1296 (2012).

26. Reynolds, H.R. & Hochman, J.S. Cardiogenic shock: current concepts and improving outcomes. Circulation 117, 686-697 (2008).

27. Hollenberg, S.M., Kavinsky, C.J. & Parrillo, J.E. Cardiogenic shock. Ann. Intern. Med. 131, 47-59 (1999).

28. Jeger, R.V., et al. Ten-year trends in the incidence and treatment of cardio-genic shock. Ann. Intern. Med. 149, 618-626 (2008).

29. Kolte, D., et al. Trends in incidence, management, and outcomes of cardio-genic shock complicating ST-elevation myocardial infarction in the United States. J Am Heart Assoc 3, e000590 (2014).

30. Goldberg, R.J., et al. Decade-Long Trends (2001-2011) in the Incidence and Hospital Death Rates Associated with the In-Hospital Development of Cardi-ogenic Shock after Acute Myocardial Infarction. Circ. Cardiovasc. Qual. Outcomes 9, 117-125 (2016).

31. Medawar, P.B. The behaviour and fate of skin autografts and skin homografts in rabbits: A report to the War Wounds Committee of the Medical Research Council. J. Anat. 78, 176-199 (1944).

32. Merrill, J.P., Murray, J.E., Harrison, J.H. & Guild, W.R. Successful homo-transplantation of the human kidney between identical twins. J Am Med As-soc 160, 277-282 (1956).

61

33. Lower, R.R. & Shumway, N.E. Studies on orthotopic homotransplantation of the canine heart. Surg. Forum 11, 18-19 (1960).

34. Barnard, C.N. The operation. A human cardiac transplant: an interim report of a successful operation performed at Groote Schuur Hospital, Cape Town. S. Afr. Med. J. 41, 1271-1274 (1967).

35. Lund, L.H., et al. The registry of the International Society for Heart and Lung Transplantation: thirty-first official adult heart transplant report--2014; focus theme: retransplantation. J. Heart Lung Transplant. 33, 996-1008 (2014).

36. Bergenfeldt, H., et al. Outcomes after ABO-incompatible heart transplanta-tion in adults: A registry study. J. Heart Lung Transplant. (2015).

37. Peura, J.L., et al. Recommendations for the use of mechanical circulatory support: device strategies and patient selection: a scientific statement from the American Heart Association. Circulation 126, 2648-2667 (2012).

38. Feldman, D., et al. The 2013 International Society for Heart and Lung Trans-plantation Guidelines for mechanical circulatory support: executive sum-mary. J. Heart Lung Transplant. 32, 157-187 (2013).

39. Liotta, D., et al. Prolonged partial left ventricular bypass by means of an intrathoracic pump implanted in the left chest. Trans. Am. Soc. Artif. Intern. Organs 8, 90-99 (1962).

40. Kantrowitz, A., et al. Initial clinical experience with intraaortic balloon pumping in cardiogenic shock. JAMA 203, 113-118 (1968).

41. Cooley, D.A., et al. Orthotopic cardiac prosthesis for two-staged cardiac replacement. Am. J. Cardiol. 24, 723-730 (1969).

42. DeVries, W.C., et al. Clinical use of the total artificial heart. N. Engl. J. Med. 310, 273-278 (1984).

43. Kormos, R.L., et al. Experience with univentricular support in mortally ill cardiac transplant candidates. Ann. Thorac. Surg. 49, 261-271; discussion 271-262 (1990).

44. Frazier, O.H., et al. Multicenter clinical evaluation of the HeartMate 1000 IP left ventricular assist device. Ann. Thorac. Surg. 53, 1080-1090 (1992).

45. Pennington, D.G., McBride, L.R., Peigh, P.S., Miller, L.W. & Swartz, M.T. Eight years' experience with bridging to cardiac transplantation. J. Thorac. Cardiovasc. Surg. 107, 472-480; discussion 480-471 (1994).

46. Korfer, R., et al. Single-center experience with the thoratec ventricular assist device. J. Thorac. Cardiovasc. Surg. 119, 596-600 (2000).

47. McBride, L.R., Naunheim, K.S., Fiore, A.C., Moroney, D.A. & Swartz, M.T. Clinical experience with 111 thoratec ventricular assist devices. Ann. Thorac. Surg. 67, 1233-1238; discussion 1238-1239 (1999).

48. Moskowitz, A.J., Weinberg, A.D., Oz, M.C. & Williams, D.L. Quality of life with an implanted left ventricular assist device. Ann. Thorac. Surg. 64, 1764-1769 (1997).

49. Mehta, S.M., Aufiero, T.X., Pae, W.E., Jr., Miller, C.A. & Pierce, W.S. Combined Registry for the Clinical Use of Mechanical Ventricular Assist Pumps and the Total Artificial Heart in conjunction with heart transplanta-tion: sixth official report--1994. J. Heart Lung Transplant. 14, 585-593 (1995).

50. Rose, E.A., et al. Long-term use of a left ventricular assist device for end-stage heart failure. N. Engl. J. Med. 345, 1435-1443 (2001).

51. Slaughter, M.S., et al. Advanced heart failure treated with continuous-flow left ventricular assist device. N. Engl. J. Med. 361, 2241-2251 (2009).

62

52. Park, S.J., et al. Outcomes in advanced heart failure patients with left ven-tricular assist devices for destination therapy. Circ. Heart Fail. 5, 241-248 (2012).

53. Jorde, U.P., et al. Results of the destination therapy post-food and drug ad-ministration approval study with a continuous flow left ventricular assist de-vice: a prospective study using the INTERMACS registry (Interagency Reg-istry for Mechanically Assisted Circulatory Support). J. Am. Coll. Cardiol. 63, 1751-1757 (2014).

54. Kirklin, J.K., et al. Sixth INTERMACS annual report: a 10,000-patient data-base. J. Heart Lung Transplant. 33, 555-564 (2014).

55. Sanborn, T.A., et al. Impact of thrombolysis, intra-aortic balloon pump coun-terpulsation, and their combination in cardiogenic shock complicating acute myocardial infarction: a report from the SHOCK Trial Registry. SHould we emergently revascularize Occluded Coronaries for cardiogenic shocK? J. Am. Coll. Cardiol. 36, 1123-1129 (2000).

56. Anderson, R.D., et al. Use of intraaortic balloon counterpulsation in patients presenting with cardiogenic shock: observations from the GUSTO-I Study. Global Utilization of Streptokinase and TPA for Occluded Coronary Arteries. J. Am. Coll. Cardiol. 30, 708-715 (1997).

57. Barron, H.V., et al. The use of intra-aortic balloon counterpulsation in pa-tients with cardiogenic shock complicating acute myocardial infarction: data from the National Registry of Myocardial Infarction 2. Am. Heart J. 141, 933-939 (2001).

58. Thiele, H., et al. Intra-aortic balloon counterpulsation in acute myocardial infarction complicated by cardiogenic shock (IABP-SHOCK II): final 12 month results of a randomised, open-label trial. Lancet 382, 1638-1645 (2013).

59. Manzo-Silberman, S., et al. Percutaneous left ventricular assistance in post cardiac arrest shock: comparison of intra aortic blood pump and IMPELLA Recover LP2.5. Resuscitation 84, 609-615 (2013).

60. O'Neill, W.W., et al. A prospective, randomized clinical trial of hemodynam-ic support with Impella 2.5 versus intra-aortic balloon pump in patients un-dergoing high-risk percutaneous coronary intervention: the PROTECT II study. Circulation 126, 1717-1727 (2012).

61. Seyfarth, M., et al. A randomized clinical trial to evaluate the safety and efficacy of a percutaneous left ventricular assist device versus intra-aortic balloon pumping for treatment of cardiogenic shock caused by myocardial in-farction. J. Am. Coll. Cardiol. 52, 1584-1588 (2008).

62. Thiele, H., et al. Randomized comparison of intra-aortic balloon support with a percutaneous left ventricular assist device in patients with revascularized acute myocardial infarction complicated by cardiogenic shock. Eur. Heart J. 26, 1276-1283 (2005).

63. O'Neill, W.W., et al. The current use of Impella 2.5 in acute myocardial in-farction complicated by cardiogenic shock: results from the USpella Registry. J. Interv. Cardiol. 27, 1-11 (2014).

64. Jensen, P.B., et al. Single-centre experience with the Impella CP, 5.0 and RP in 109 consecutive patients with profound cardiogenic shock. Eur Heart J Acute Cardiovasc Care 7, 53-61 (2018).

65. Ouweneel, D.M. & Henriques, J.P. Percutaneous cardiac support devices for cardiogenic shock: current indications and recommendations. Heart 98, 1246-1254 (2012).

63

66. Kar, B., Basra, S.S., Shah, N.R. & Loyalka, P. Percutaneous circulatory sup-port in cardiogenic shock: interventional bridge to recovery. Circulation 125, 1809-1817 (2012).

67. Thiele, H., Smalling, R.W. & Schuler, G.C. Percutaneous left ventricular assist devices in acute myocardial infarction complicated by cardiogenic shock. Eur. Heart J. 28, 2057-2063 (2007).

68. Hill, J.D., et al. Prolonged extracorporeal oxygenation for acute post-traumatic respiratory failure (shock-lung syndrome). Use of the Bramson membrane lung. N. Engl. J. Med. 286, 629-634 (1972).

69. Bartlett, R.H., et al. Extracorporeal membrane oxygenation (ECMO) cardio-pulmonary support in infancy. Trans. Am. Soc. Artif. Intern. Organs 22, 80-93 (1976).

70. Lawler, P.R., et al. Extracorporeal Membrane Oxygenation in Adults With Cardiogenic Shock. Circulation 131, 676 (2015).

71. Teschendorf, P. & Bernhard, M. A bridge to life: ECPR who, when, where and why? Resuscitation 85, 709-710 (2014).

72. Stub, D., et al. Refractory cardiac arrest treated with mechanical CPR, hypo-thermia, ECMO and early reperfusion (the CHEER trial). Resuscitation 86, 88-94 (2015).

73. Fagnoul, D., Combes, A. & De Backer, D. Extracorporeal cardiopulmonary resuscitation. Curr Opin Crit Care 20, 259-265 (2014).

74. Cardarelli, M.G., Young, A.J. & Griffith, B. Use of extracorporeal membrane oxygenation for adults in cardiac arrest (E-CPR): a meta-analysis of observa-tional studies. ASAIO J. 55, 581-586 (2009).

75. Conseil francais de reanimation, c., et al. Guidelines for indications for the use of extracorporeal life support in refractory cardiac arrest. French Ministry of Health. Ann. Fr. Anesth. Reanim. 28, 182-190 (2009).

76. Bavaria, J.E., et al. Changes in left ventricular systolic wall stress during biventricular circulatory assistance. Ann. Thorac. Surg. 45, 526-532 (1988).

77. Walther, F.J., van de Bor, M., Gangitano, E.S. & Snyder, J.R. Left and right ventricular output in newborn infants undergoing extracorporeal membrane oxygenation. Crit. Care Med. 18, 148-151 (1990).

78. Martin, G.R., Short, B.L., Abbott, C. & O'Brien, A.M. Cardiac stun in infants undergoing extracorporeal membrane oxygenation. J. Thorac. Cardiovasc. Surg. 101, 607-611 (1991).

79. Swartz, M.F., Smith, F., Byrum, C.J. & Alfieris, G.M. Transseptal catheter decompression of the left ventricle during extracorporeal membrane oxy-genation. Pediatr. Cardiol. 33, 185-187 (2012).

80. Koeckert, M.S., Jorde, U.P., Naka, Y., Moses, J.W. & Takayama, H. Impella LP 2.5 for left ventricular unloading during venoarterial extracorporeal mem-brane oxygenation support. J. Card. Surg. 26, 666-668 (2011).

81. Rosenberg, E.M. & Cook, L.N. Electromechanical dissociation in newborns treated with extracorporeal membrane oxygenation: an extreme form of car-diac stun syndrome. Crit. Care Med. 19, 780-784 (1991).

82. Truby, L.K., et al. Incidence and Implications of Left Ventricular Distention During Venoarterial Extracorporeal Membrane Oxygenation Support. ASAIO J. 63, 257-265 (2017).

83. Truby, L., et al. Contemporary Outcomes of Venoarterial Extracorporeal Membrane Oxygenation for Refractory Cardiogenic Shock at a Large Ter-tiary Care Center. ASAIO J. 61, 403-409 (2015).

64

84. Aiyagari, R.M., Rocchini, A.P., Remenapp, R.T. & Graziano, J.N. Decom-pression of the left atrium during extracorporeal membrane oxygenation us-ing a transseptal cannula incorporated into the circuit. Crit. Care Med. 34, 2603-2606 (2006).

85. Barbone, A., Malvindi, P.G., Ferrara, P. & Tarelli, G. Left ventricle unload-ing by percutaneous pigtail during extracorporeal membrane oxygenation. In-teract. Cardiovasc. Thorac. Surg. 13, 293-295 (2011).

86. Pantalos, G.M., et al. Mock circulation simulation of extracorporeal mem-brane oxygenation support for systemic ventricular failure in an infant: the ef-fect of atrial septostomy. ASAIO J. 58, 415-419 (2012).

87. Dahdouh, Z., Roule, V., Lognone, T., Sabatier, R. & Grollier, G. Percutane-ous blade and balloon atrioseptostomy as a supplement to extracorporeal membrane oxygenation as a bridge to heart transplantation. Cardiovasc. Re-vasc. Med. 13, 69-71 (2012).

88. Alhussein, M., et al. Percutaneous left atrial decompression in adults with refractory cardiogenic shock supported with veno-arterial extracorporeal membrane oxygenation. J. Card. Surg. 32, 396-401 (2017).

89. Chaparro, S.V., et al. Combined use of Impella left ventricular assist device and extracorporeal membrane oxygenation as a bridge to recovery in fulmi-nant myocarditis. ASAIO J. 58, 285-287 (2012).

90. Anastasiadis, K., Chalvatzoulis, O., Antonitsis, P., Tossios, P. & Papa-konstantinou, C. Left ventricular decompression during peripheral extracor-poreal membrane oxygenation support with the use of the novel iVAC pulsa-tile paracorporeal assist device. Ann. Thorac. Surg. 92, 2257-2259 (2011).

91. Lim, H.S. The Effect of Impella CP on Cardiopulmonary Physiology During Venoarterial Extracorporeal Membrane Oxygenation Support. Artif. Organs 41, 1109-1112 (2017).

92. Patel, S.M., et al. Simultaneous Venoarterial Extracorporeal Membrane Oxy-genation and Percutaneous Left Ventricular Decompression Therapy with Impella Is Associated with Improved Outcomes in Refractory Cardiogenic Shock. ASAIO J. (2018).

93. Cohn, J.N., Ferrari, R. & Sharpe, N. Cardiac remodeling--concepts and clini-cal implications: a consensus paper from an international forum on cardiac remodeling. Behalf of an International Forum on Cardiac Remodeling. J. Am. Coll. Cardiol. 35, 569-582 (2000).

94. Kass, D.A., et al. Reverse remodeling from cardiomyoplasty in human heart failure. External constraint versus active assist. Circulation 91, 2314-2318 (1995).

95. Levin, H.R., et al. Reversal of chronic ventricular dilation in patients with end-stage cardiomyopathy by prolonged mechanical unloading. Circulation 91, 2717-2720 (1995).

96. Saraon, T. & Katz, S.D. Reverse Remodeling in Systolic Heart Failure. Car-diol. Rev. (2015).

97. Udelson, J.E. & Konstam, M.A. Relation between left ventricular remodeling and clinical outcomes in heart failure patients with left ventricular systolic dysfunction. J. Card. Fail. 8, S465-471 (2002).

98. St John Sutton, M., et al. Cardiac resynchronization induces major structural and functional reverse remodeling in patients with New York Heart Associa-tion class I/II heart failure. Circulation 120, 1858-1865 (2009).

65

99. Linde, C., et al. Randomized trial of cardiac resynchronization in mildly symptomatic heart failure patients and in asymptomatic patients with left ventricular dysfunction and previous heart failure symptoms. J. Am. Coll. Cardiol. 52, 1834-1843 (2008).

100. Drakos, S.G., et al. Magnitude and time course of changes induced by con-tinuous-flow left ventricular assist device unloading in chronic heart failure: insights into cardiac recovery. J. Am. Coll. Cardiol. 61, 1985-1994 (2013).

101. Madigan, J.D., et al. Time course of reverse remodeling of the left ventricle during support with a left ventricular assist device. J. Thorac. Cardiovasc. Surg. 121, 902-908 (2001).

102. Birks, E.J., et al. Reversal of severe heart failure with a continuous-flow left ventricular assist device and pharmacological therapy: a prospective study. Circulation 123, 381-390 (2011).

103. Drakos, S.G., et al. Impact of mechanical unloading on microvasculature and associated central remodeling features of the failing human heart. J. Am. Coll. Cardiol. 56, 382-391 (2010).

104. Drakos, S.G., et al. Left ventricular assist device unloading effects on myo-cardial structure and function: current status of the field and call for action. Curr. Opin. Cardiol. 26, 245-255 (2011).

105. Barbone, A., et al. Comparison of Right and Left Ventricular Responses to Left Ventricular Assist Device Support in Patients With Severe Heart Failure. Circulation 104, 670-675 (2001).

106. Avalli, L., et al. Percutaneous left-heart decompression during extracorporeal membrane oxygenation: an alternative to surgical and transeptal venting in adult patients. ASAIO J. 57, 38-40 (2011).

107. Schmack, B., et al. Extracorporeal life support with left ventricular decom-pression-improved survival in severe cardiogenic shock: results from a retro-spective study. PeerJ 5, e3813 (2017).

108. Pappalardo, F., et al. Concomitant implantation of Impella(R) on top of veno-arterial extracorporeal membrane oxygenation may improve survival of pa-tients with cardiogenic shock. Eur. J. Heart Fail. (2016).

109. Popovic, B., Fay, R., Cravoisy-Popovic, A. & Levy, B. Cardiac power index, mean arterial pressure, and Simplified Acute Physiology Score II are strong predictors of survival and response to revascularization in cardiogenic shock. Shock 42, 22-26 (2014).

110. Hale, S.L. & Kloner, R.A. Mild hypothermia as a cardioprotective approach for acute myocardial infarction: laboratory to clinical application. J. Cardio-vasc. Pharmacol. Ther. 16, 131-139 (2011).

111. Gerczuk, P.Z. & Kloner, R.A. An update on cardioprotection: a review of the latest adjunctive therapies to limit myocardial infarction size in clinical trials. J. Am. Coll. Cardiol. 59, 969-978 (2012).

112. Kloner, R.A. Can myocardial infarct size be reduced by mechanically unload-ing the left ventricle? Circulation 128, 318-321 (2013).

113. Kloner, R.A. Current state of clinical translation of cardioprotective agents for acute myocardial infarction. Circ. Res. 113, 451-463 (2013).

114. Axelrod, H.I., et al. A comparison of methods for limiting myocardial infarct expansion during acute reperfusion--primary role of unloading. Circulation 76, V28-32 (1987).

115. Laschinger, J.C., et al. Adjunctive left ventricular unloading during myocar-dial reperfusion plays a major role in minimizing myocardial infarct size. J. Thorac. Cardiovasc. Surg. 90, 80-85 (1985).

66

116. Axelrod, H.I., et al. Percutaneous cardiopulmonary bypass limits myocardial injury from ischemic fibrillation and reperfusion. Circulation 78, III148-152 (1988).

117. Matsumiya, G., Saitoh, S. & Sawa, Y. Extracorporeal assist circulation for heart failure. Circ. J. 73 Suppl A, A42-47 (2009).

118. Thiagarajan, R.R., et al. Extracorporeal Life Support Organization Registry International Report 2016. ASAIO J. 63, 60-67 (2017).

119. Meyns, B., Stolinski, J., Leunens, V., Verbeken, E. & Flameng, W. Left ventricular support by catheter-mounted axial flow pump reduces infarct size. J. Am. Coll. Cardiol. 41, 1087-1095 (2003).

120. Kapur, N.K., et al. Mechanically unloading the left ventricle before coronary reperfusion reduces left ventricular wall stress and myocardial infarct size. Circulation 128, 328-336 (2013).

121. Wei, X., et al. Short-term mechanical unloading with left ventricular assist devices after acute myocardial infarction conserves calcium cycling and im-proves heart function. JACC Cardiovasc. Interv. 6, 406-415 (2013).

122. Esposito, M., et al. The role of acute circulatory support in ST-segment ele-vation myocardial infarction complicated by cardiogenic shock. Indian Heart J. 69, 668-674 (2017).

123. Kirklin, J.K., et al. Seventh INTERMACS annual report: 15,000 patients and counting. J. Heart Lung Transplant. 34, 1495-1504 (2015).

124. Stretch, R., Sauer, C.M., Yuh, D.D. & Bonde, P. National trends in the utili-zation of short-term mechanical circulatory support: incidence, outcomes, and cost analysis. J. Am. Coll. Cardiol. 64, 1407-1415 (2014).

125. Schmidt, M., et al. Predicting survival after ECMO for refractory cardiogenic shock: the survival after veno-arterial-ECMO (SAVE)-score. Eur. Heart J. 36, 2246-2256 (2015).

126. Csonka, C., et al. Measurement of myocardial infarct size in preclinical stud-ies. J. Pharmacol. Toxicol. Methods 61, 163-170 (2010).

127. Khalil, P.N., et al. Histochemical assessment of early myocardial infarction using 2,3,5-triphenyltetrazolium chloride in blood-perfused porcine hearts. J. Pharmacol. Toxicol. Methods 54, 307-312 (2006).

128. Goldberg, R.J., et al. Temporal trends in cardiogenic shock complicating acute myocardial infarction. N. Engl. J. Med. 340, 1162-1168 (1999).

129. Fuhrman, B.P., Hernan, L.J., Rotta, A.T., Heard, C.M. & Rosenkranz, E.R. Pathophysiology of cardiac extracorporeal membrane oxygenation. Artif. Or-gans 23, 966-969 (1999).

130. Achour, H., et al. Mechanical left ventricular unloading prior to reperfusion reduces infarct size in a canine infarction model. Catheter. Cardiovasc. In-terv. 64, 182-192 (2005).

131. Catinella, F.P., et al. Left atrium-to-femoral artery bypass: effectiveness in reduction of acute experimental myocardial infarction. J. Thorac. Cardio-vasc. Surg. 86, 887-896 (1983).

132. Fonger, J.D., Zhou, Y., Matsuura, H., Aldea, G.S. & Shemin, R.J. Enhanced preservation of acutely ischemic myocardium with transseptal left ventricular assist. Ann. Thorac. Surg. 57, 570-575 (1994).

133. Sjauw, K.D., et al. Left ventricular unloading in acute ST-segment elevation myocardial infarction patients is safe and feasible and provides acute and sus-tained left ventricular recovery. J. Am. Coll. Cardiol. 51, 1044-1046 (2008).

67

134. Smalling, R.W., et al. Improved regional myocardial blood flow, left ventric-ular unloading, and infarct salvage using an axial-flow, transvalvular left ven-tricular assist device. A comparison with intra-aortic balloon counterpulsa-tion and reperfusion alone in a canine infarction model. Circulation 85, 1152-1159 (1992).

135. Tamareille, S., et al. Left ventricular unloading before reperfusion reduces endothelin-1 release and calcium overload in porcine myocardial infarction. J. Thorac. Cardiovasc. Surg. 136, 343-351 (2008).

136. Yoshitake, I., et al. The effect of combined treatment with Impella((R)) and landiolol in a swine model of acute myocardial infarction. J Artif Organs 15, 231-239 (2012).

137. Huang, S.C., et al. Extracorporeal membrane oxygenation rescue for cardio-pulmonary resuscitation in pediatric patients. Crit. Care Med. 36, 1607-1613 (2008).

138. Massetti, M., Gaudino, M., De Paulis, S., Scapigliati, A. & Cavaliere, F. Extracorporeal membrane oxygenation for resuscitation and cardiac arrest management. Heart Fail Clin 10, S85-93 (2014).

139. Schopka, S., et al. Single-center experience with extracorporeal life support in 103 nonpostcardiotomy patients. Artif. Organs 37, 150-156 (2013).

140. Rupprecht, L., et al. Cardiac decompression on extracorporeal life support: a review and discussion of the literature. ASAIO J. 59, 547-553 (2013).

141. Subramaniam, K., Boisen, M., Shah, P.R., Ramesh, V. & Pete, A. Mechani-cal circulatory support for cardiogenic shock. Best Pract. Res. Clin. Anaes-thesiol. 26, 131-146 (2012).

142. Rihal, C.S., et al. 2015 SCAI/ACC/HFSA/STS Clinical Expert Consensus Statement on the Use of Percutaneous Mechanical Circulatory Support De-vices in Cardiovascular Care (Endorsed by the American Heart Association, the Cardiological Society of India, and Sociedad Latino Americana de Cardi-ologia Intervencion; Affirmation of Value by the Canadian Association of In-terventional Cardiology-Association Canadienne de Cardiologie d'interven-tion). J. Card. Fail. 21, 499-518 (2015).

143. Abrams, D., et al. Position paper for the organization of ECMO programs for cardiac failure in adults. Intensive Care Med. (2018).

144. Kawashima, D., et al. Left ventricular mechanical support with Impella pro-vides more ventricular unloading in heart failure than extracorporeal mem-brane oxygenation. ASAIO J. 57, 169-176 (2011).

145. Yoshioka, D., et al. Predictor of early mortality for severe heart failure pa-tients with left ventricular assist device implantation: significance of IN-TERMACS level and renal function. Circ. J. 76, 1631-1638 (2012).

146. Meyns, B., et al. Initial experiences with the Impella device in patients with cardiogenic shock - Impella support for cardiogenic shock. Thorac. Cardio-vasc. Surg. 51, 312-317 (2003).

147. Mullens, W., et al. Importance of venous congestion for worsening of renal function in advanced decompensated heart failure. J. Am. Coll. Cardiol. 53, 589-596 (2009).

148. Lemaire, A., et al. Outcome of the impella device for acute mechanical circu-latory support. Innovations (Phila) 8, 12-16 (2013).

149. Sieweke, J.T., et al. Mortality in patients with cardiogenic shock treated with the Impella CP microaxial pump for isolated left ventricular failure. Eur Heart J Acute Cardiovasc Care, 2048872618757393 (2018).

68

150. Heusch, G., Skyschally, A. & Schulz, R. The in-situ pig heart with regional ischemia/reperfusion - ready for translation. J. Mol. Cell. Cardiol. 50, 951-963 (2011).

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1464

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-347427

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2018