Veno-Arterial Extracorporeal Membrane Oxygenation for Refractory Cardiogenic Shock and Cardiac Arrest

Cardiogenic shock (CS) following acute myocardial infarction (AMI) occurs in 7% to 9% of patients affected by AMI with a high mortality rates. Despite all recent advanced treatments such as use of inotropes, vasoconstrictors and intra-aortic balloon pump (IAPB) therapy, revascularization techniques and application of different systems of mechanical circulatory support, CS is still the most common cause of hospital mortality ranging between 60%-70% compared to patients with AMI without advanced CS whose hospital mortality is about 10% [1]. Cardiac arrest is a major cause of unexpected death and complicates about 22% of patients with acute myocardial infarction [2]. Cardiac arrest has a poor prognosis, and despite con‐ ventional cardiopulmonary resuscitation (CPR) maneuvers, only a few patients can fully return to a normal lifestyle. The main reasons for very poor outcome and prognosis in CA are a lack of return of spontaneous circulation (ROSC), a long time of CPR [3],[4], re-arrest from hemodynamic instability after ROSC, hypoxic encephalopathy [5], out-of-hospital CA [6-8]. In both refractory CS and CA secondary AMI, which are very critical circumstances, VenoArterial Extracorporeal Membrane Oxygenation (V-A ECMO) has been proposed and utilized during the last decades to obtain rapid resuscitation, stabilization, and subsequent triage to a more permanent treatment strategy.


Introduction
Cardiogenic shock (CS) following acute myocardial infarction (AMI) occurs in 7% to 9% of patients affected by AMI with a high mortality rates. Despite all recent advanced treatments such as use of inotropes, vasoconstrictors and intra-aortic balloon pump (IAPB) therapy, revascularization techniques and application of different systems of mechanical circulatory support, CS is still the most common cause of hospital mortality ranging between 60%-70% compared to patients with AMI without advanced CS whose hospital mortality is about 10% [1]. Cardiac arrest is a major cause of unexpected death and complicates about 22% of patients with acute myocardial infarction [2]. Cardiac arrest has a poor prognosis, and despite conventional cardiopulmonary resuscitation (CPR) maneuvers, only a few patients can fully return to a normal lifestyle. The main reasons for very poor outcome and prognosis in CA are a lack of return of spontaneous circulation (ROSC), a long time of CPR [3], [4], re-arrest from hemodynamic instability after ROSC, hypoxic encephalopathy [5], out-of-hospital CA [6][7][8]. In both refractory CS and CA secondary AMI, which are very critical circumstances, Veno-Arterial Extracorporeal Membrane Oxygenation (V-A ECMO) has been proposed and utilized during the last decades to obtain rapid resuscitation, stabilization, and subsequent triage to a more permanent treatment strategy.
The aim of this chapter is to describe the more recent indications, techniques and results in the usage of the V-A ECMO in patients with refractory cardiogenic shock and cardiac arrest secondary to acute myocardial infarction.

Definition of V-A ECMO
Extracorporeal membrane oxygenation is essentially a modification of the cardiopulmonary bypass circuit, which is used routinely in cardiac surgery. Blood is removed from the venous system, oxygenated by an oxygenator and then returned back to the body by a pump. ECMO provides both full cardiac and respiratory support. In brief, every ECMO system is basically a closed tubing loop with the interpolation of a blood pump (centrifugal or roller) and an oxygenator. Two vessel cannulas complete the system. Technically the ECMO system is more complex and several configurations have been developed according to the primary etiology.
The veno-arterial ECMO configuration is a tubing loop with a venous arm connected to a venous cannula to allow the venous blood drainage and an arterial arm to return back the oxygenated blood inside the patient's circulatory system. This mode provides both cardiac and respiratory support and can be achieved by either peripheral or central cannulation (Figure1). The veno-venous ECMO mode refers to blood, which is drained from the venous system and returned back to the venous system. This mode only provides respiratory support and is obtained by peripheral cannulation, usually of both femoral veins and jugular vein. This ECMO configuration will be not discussed in this chapter.
c. a cardiac index less than 1.8 L/min/m2 without inotropic support and 2.0-2.2 L/min/m2 with inotropic and intra-aortic balloon pump (IABP) support; d. high left ventricular (LV) filling pressures (pulmonary capillary wedge pressure more than 18 mmHg).
End-organ hypoperfusion may be manifested clinically by: a. pale, cool, and clammy peripheries; b. alteration in mental status such delirium, confusion, clouded sensorium, psychomotor agitation; c. decreased urine output (less than 1 ml/kg/min); d. pulmonary congestion or edema; e. tachycardia; f. hyperlactacidemia (more than 3.mmol/L) as expression of impaired peripheral microcirculation; g. mixed venous saturation of less than 65%.

Definition of in-hospital and out-hospital cardiac arrest
Cardiac arrest is a major cause of unexpected death in developed countries with a low probability of patient survival. Survival is influenced by several variables common to both inhospital and out-of-hospital arrest, such as time to recognition of the cardiac arrest, time to initiation of CPR, rhythm presentation, first defibrillation [2,11,12]. In current resuscitation guidelines for in-hospital cardiac arrest (IHCA) patients [13], CPR using ECMO (E-CPR) has been assigned a low-grade recommendation. It is reported that ECMO for out-of-hospital cardiac arrest (OHCA) has worse outcomes compared with ECMO for IHCA patients [7,14].
In the United States, more than 166,000 patients experience an OHCA annually [15] and approximately 60% are treated by emergency medical services. OHCA survival to hospital discharge range from 0.3% in Detroit [16] to 20.4% in Slovenia [17].Five clinical criteria to predict survival from OHCA [18] have recently been reported. They are: cardiac arrest witnessed by a bystander, arrest witnessed by emergency medical personnel, provision of bystander CPR, shockable cardiac rhythm, and return of spontaneous circulation (ROSC) in the field. These criteria are applicable on IHCA too.

Indication and contraindications for v-a ECMO
Patient selection is a crucial point when the physician needs to take the decision to institute ECMO and several considerations must be focused up. Most importantly, it must be consider the likelihood of heart and end organs recovery. If the organs failure is thought to be reversible with ECMO, in such situation the device application is to be encouraged. If the likelihood of recovery of the heart or other end organs is thought to be very low or even impossible, then other factors must be taken into account. In such clinical scenario, the decision to institute ECMO should be based on an experienced ECMO team approach, which has to evaluate the patient's eligibility for heart transplantation or a definitive mechanical assist device (LVAD) implant as destination therapy.

Indications for v-a ECMO
The following factors need to be evaluated for the indications [19]: • Age of patient and body surface area; • sufficient medical expertise in the field of ECMO; • possibilities for myocardial revascularization therapy, such as a coronary artery bypass grafting or coronary angioplasty; • possibilities for heart transplantation or LVAD implant as destination therapy; • status of central organs such as kidney, liver, and brain.

Contraindications for v-a ECMO
Contraindications to the institution of v-a ECMO include [20]: • disseminated malignancy; • advanced age; • graft vs. host disease; • known severe brain injury; • unwitnessed cardiac arrest or cardiac arrest of prolonged duration; • aortic dissection aortic incompetence.

ECMO circuit
The ECMO circuit is made of PVC tubing and the diameter of lines varies from ¼ inch for a neonate to ½ inch for pediatric and adult patients. The length of the circuit is kept not more than 2 meters to avoid increasing of resistance within a tube and twisting, but the length should be suitable to allow the movements of the patient by ECMO staff.
Areas of turbulent flow can predispose to clot formation; therefore loop and connectors should be avoided or kept at minimum.

ECMO cannulas and cannulation techniques
Cannulation is one of the most challenging aspects of ECMO. Peripheral percutaneous approach [6, 21 is the most used in cardiogenic shock and cardiac arrest [7,22] because is quicker with less bleeding complications and easier decannulation. (Figure 2). The open surgical approach is considered for patients with severe peripheral vascular disease or for patients with postcardiotomy syndrome or failure of weaning from cardiopulmonary bypass [23,24] ( Figure 3). The open or central cannulation has more complications such as bleeding, infections, and mediastinitis.   The size of the cannulas depends on the size of the patient; usually the arterial cannula ranges between 17 Fr to 21 Fr and the venous cannula ranges between 21 Fr and 25 Fr. Cannulas of sufficient size are required to support high blood flow with low resistance. Local complications, particularly at the site of peripheral insertion of VA-ECMO can occur, of which the most concerning is leg ischemia. For this reason all attempts the limb perfusion is restored, after noted the absence of anterior and posterior tibial artery flow, by inserting a 9-Fr catheter distally to the arterial cannula by means of vascular ultrasound scan as soon as possible after ECMO implantation ( Figure 6). Some Authors [25] suggested to insert a catheter for a distal perfusion if the mean pressure of the superficial femoral artery is lower than 50 mmHg.

Figure 7.
Cannulation of right axillary artery (tube on the right). Cannula is tunneled below the skin to protect by accidental trauma.
Whatever the type of approach cannulation is considered, it requires always a highly skilled medical staff, usually a cardiac or vascular surgeon, who are able to undertake this procedure under often very difficult conditions as the patients are so unstable or even in cardiac arrest.

Pumps
The pump pushes the blood through the oxygenator and then back to the patient. The most used pump in adult patients is the centrifugal pump. These pumps consist of a polycarbonate housing with a one-point sapphire bearing linked to a magnetic field, which create a vortex flow at an adjustable rate ( Figure 8). Vortex creates a negative pressure in the pump head and this negative pressure pulls blood into the pump and then the blood is pushed towards the oxygenator.

Oxygenators
The silicone membrane has been the principal artificial lung used for ECMO for many years and introduced in the clinical practice by Kolobow [27] and Bartlett [28]. The silicone oxygenators were used until the diffusion of microporous hollow fibers oxygenators in the 90s. The silicone surface is homogeneous and does not contain micropores, which can cause plasma leakage. However, the silicone oxygenator has a very large membrane surface to ensure adequate gas exchange and needs both high prime volume and high pressure drop; moreover, the procedure to optimize the efficacy of the oxygenator is cumbersome and lengthy, requiring a CO2 gas flush. The hollow-fiber polypropylene membrane oxygenators had advantages over the silicone oxygenators, such as high gas exchange efficiency with a smaller change surface, lower prime volume, and lower pressure gradient. However, this generation of oxygenators has micropores causing plasma leakage for periods more than 6 hours, thus reducing the gas exchange. Recently, a new generation of poly-methylpentene (PMP) membrane oxygenators have been introduced with the aim of allowing longer support without the complications linked to the hollow-fiber oxygenators, such as plasma leakage [29] (Figure 9).
The adjunct equipment that completes the ECMO system includes a heat exchanger for temperature regulation, monitors that measure blood flow, venous and arterial saturation, hematocrit, and other variables. ECMO systems also can measure circuit pressures and changes in circuit resistance. Additional safety features include continuous monitoring of venous drainage and air detection. Recently a new miniaturized system for V-A ECMO was introduced in the clinical practice.
The system has the console directly connected with the oxygenator and the blood pump, which are integrated to each other. The console has a touch screen where is possible to monitor continuously several parameters such has hematocrit, hemoglobin, SVO2, resistance at the inlet and the outlet of oxygenator and the pressure drop ( Figure 10). This system is very useful because of his reduced dimension and weight (the weight of console is about 10 Kgs); for this reason this system can be used for transportation of V-A ECMO patients within the different sites of the hospital or from hospital to other hospital.

Management of v-a ECMO
Systemic heparinization is obtained with an intravenous bolus of 5,000 UI of heparin, 5 minutes prior to vessels cannulation. ECMO blood flow is calculated to maintain a Cardiac Output (CO) index of 2.5 L/min/m 2 , an SvO 2 of about 70% and a mean blood arterial pressure of 60-70 mmHg during the first 24-48 hours. Continuous intravenous heparin is administered in order to achieve an activate clotting time (ACT) of 160-180 seconds and a prothrombin time value of 50-60 seconds. Small doses of inotrope (dobutamine, 5 to 7 μg/Kg/min) are given to maintain the ventricular ejection, to allow the opening of aortic valve and to prevent the formation of clots inside the left ventricle (LV). Oxygenator is always connected with a heat exchange to maintain a body temperature of 36 °C. Those patients who had a cardiac arrest before ECMO implantation are gradually cooled to 32-34°C during the first 24-36 hours. The assessment of the neurologic status is initiated by electroencephalography after body rewarming; serial neurologic evaluations and cerebral computed tomography scan are always considered to assess cerebral hemorrhage, stroke or hypoxic encephalopathy. In those patients with criteria of irreversible brain damage, ECMO withdrawn is usually considered.
Multiple heart examinations by transesophageal echo are performed to monitor the LV pulsatility. Left ventricular venting is considered in case of irreversible pulmonary edema, LV distension or pulseless heart with blood stasis. If the heart needs to be decompressed, several techniques can be considered. An 18-20 Fr catheter could be inserted into the apex of LV after surgery when patient cannot be weaned from cardiopulmonary bypass ( Figure 11). Alternatively, the LV can be indirectly decompressed with a 16-Fr percutaneous venous cannula inserted in the right internal jugular vein and advanced into the main pulmonary artery; the cannula was connected to the venous arm of the ECMO circuit [30] (Figure 12). Other techniques of LV decompression have been described such as the use of Pulse-Cath [31], Insertion of a pigtail inside the LV through the aortic valve [32], use of a transeptal atrial cannula [33].
Red blood cells (RBCs) transfusions are given to achieve a hematocrit of 30-32% and platelets are infused when the platelet count is less than 50,000-60,000/μL. Mechanical ventilation is continued throughout the ECMO support with the same management for each patient. Ventilator setting is commonly set at a tidal volume of 8 ml/Kg, 4 breaths/min, positive end expiratory pressure of 10 cm H 2 0 and a FiO 2 of 0.40-0.60.
Intraortic balloon pump (IABP) [24,34] is employed with the aim to reduce the afterload, to increase the coronary and cerebral perfusion [35] and to maintain a pulsatile flow.
No attempts to wean off ECMO are usually considered during the first 48 hours.
Step by step is the main strategy for weaning off ECMO using transesophageal echocardiography monitoring. This consists to reduce the pump flow at 1.0 L/min/m 2 for about 40-60 minutes after having obtained an ACT of 180 seconds. In patients who are supported also with IABP, this is set to 1:1 ratio. If systemic pressure, LV contractility, central venous pressure, wedge pressure and SvO 2 had not significant changes without the addition of new inotropes, then heparin is stopped and ECMO is removed at patient's bedside or in operating room within the next few hours.
Transthoracic and transesophageal echocardiography play a very important role in the assessment of LV and RV function and during the delicate phase of weaning from ECMO. Echocardiographic knowledge and facilities are becoming mandatory to start and continue an ECMO program with the aim to improve the outcome. Such echocardiographic parameters, such as transmitral E velocity, E/e' ratio, LV ejection fraction, aortic valve velocity-time integral, tissue Doppler lateral Ea, Sa, and parameters derived from Velocity Vector Imaging, including lateral systolic velocity, strain, and strain rate, are now considered important data [36] to drive a safe weaning from ECMO and they should be frequently collected in each ECMO patient.

Results and discussion
Extracorporeal membrane oxygenation is now considered a validate tool to support very ill patients affected by refractory cardiogenic shock to conventional therapy or cardiac arrest [4,7,[37][38][39] and it is a well-established technology to provide a rapid and full circulatory support and to reverse the severe hypoperfusion organ injury. The ECMO system has several advantages: a) it can easily be implanted at patient's bedside, b) it can be initiated through a peripheral percutaneous cannulation, c) it is possible to stabilize the patient in the out-center hospital [6,22] d) it provides a full cardiopulmonary support, e) it allows to take time for diagnosis and further decision f) it is a relative low-cost support and g) it is a validate system for a "bridge strategy" [40]. However, the use of ECMO for cardiogenic shock has several limitations. All patients need to be anticoagulated during the ECMO support and complications such as neurological damage [41], infections [42], limb ischemia [43], bleeding and transfusion requiring [44] are frequently reported. However, the use of ECMO in patients with acute coronary syndrome complicated by advanced cardiogenic shock or by cardiac arrest is becoming an increasingly accepted procedure [3,38,39,45].
Although the results about the use of IABP before and during ECMO are not univocal, the use of IABP seems to affect positively the early outcome. Some Authors [23,24,34,46] found that the nonuse of IABP was one of the significant predictors of in-hospital death. On the other side, other Authors [39,40,47] could not find significant difference about the use of IABP during ECMO support. According to these different results, we cannot confirm whether the use of IABP has a determinant role in the cardiac function improvement. It can be argued that the use of IABP during ECMO, through the increase of coronary blood flow as reported by Madershahian [35], could favorite the cardiac recovery in ischemic patients.
Higher lactate levels are an index of severe acidosis and tissue hypoxia. The trends of blood lactate levels during the first three days of ECMO are considered as independent predictors of early mortality. Hyperlactatemia (level of blood lactate above 3 mmol/l) during cardiopulmonary bypass is associated with an increased mortality and morbidity, and appears to be related primarily to a state of inadequate perfusion [48]. We have already observed [44] that, when blood lactate level is > 3 mmol/l at 48 hours after ECMO initiation, the predicted probability of mortality is 52%. The earlier ECMO initiation should improve the organ perfusion and reduce dramatically the incidence of multi-organ failure. The persistence of hyperlactacidemia during the first days of ECMO support in nonsurvivors patients, even though the flow of the pump during the same period is similar to that of surviving patients, is likely to be referred to the persistent systemic and splanchnic hypoperfusion due to the extent of atherosclerotic disease or other unknown causes.
Bleeding and transfusion requiring can negatively affect the ECMO course and the early outcome and they are considered important complications during ECMO [23,24,47]. It is worthy to point out that lower number of RBCs transfused the number of RBCs units transfused was an independent predictor of in-hospital and late mortality. The need for RBCs transfusion depends not only by the fact that some patients on ECMO have undergone surgery; other factors such as systemic heparinization during ECMO and the use of platelet inhibitors after PTCA can cause bleeding and need for transfusions with increased risk of early mortality. Alternative therapy to conventional heparin anticoagulation therapy, such as bivalirudin or fondaparinux, to reduce the risk of bleeding and for the treatment of heparine induced thrombocitopenya have been recently published [49,50].
High incidences of central nervous system (CNS) injury meeting the criteria of brain death are reported. These patients usually are withdrawn from ECMO sooner than the rest of the other patients. Brain death is frequent in patients who presented with cardiac arrest and received V-A ECMO implantation during cardiopulmonary resuscitation maneuvers. The incidence of brain death is ranging between 10% and 40% [6,7,39,51]. Thiagarajan et al. [5] analyzing data of 297 patients supported by ECPR and extracted from the Extracorporeal Life Support Organization (ELSO) Registry reported an incidence of 33% of CNS damage and 21% had irreversible hypoxic encephalopathy. Other Authors [3,4] described a very low survival when cardiopulmonary resuscitation (CPR) time is 60 minutes and a survival approaching to 0% when the CPR was more than 90 minutes.
Left ventricular decompression during ECMO support is an important priority in cases in which the contractile activity of the heart is inadequate to allow the opening of the aortic valve. In such scenario, the risk of clotting formation inside the left cavities is very high and the clots may embolize.
Several techniques to unload the left ventricle such as atrial septostomy [4,33], direct LV apex cannulation [21], insertion of PulseCath iVAC [31], use of Impella [52,53], percutaneous insertion of a pigtail [32] or percutaneous pulmonary truck drainage [30] have been described. One of the most followed strategies is to use as soon as the IABP associated with low dose of inotrope (dobutamine 5 mcg/min/Kg) with the aim to reduce the systemic resistance, improve the coronary and cerebral flow and increase the cardiac contractility. Whether the use of IABP is a useful tool to dramatically reduce the afterload mainly in such patients with a peripheral retrograde arterial return [21], and whether the IABP simply increases the coronary blood flow [35], is still debated.
Peripheral percutaneous cannulation represents a big challenging for all ECMO teams. Several peripheral complications such as retroperitoneal hemorrhage, cannula dislocation, cannulation failure, leg ischemia and leg amputation are described [43,54]. According to early or late vascular complication following peripheral cannulation, Huang et al. [25] suggested measuring the mean pressure of the superficial femoral artery and they indicate to insert a catheter for a distal perfusion if the mean pressure is lower than 50 mmHg. It is extremely important to verify the pulsatility of the anterior and posterior tibial artery by an ultrasound vascular Doppler and to restore the limb perfusion signs of hypoperfusion are observed. In such case an 8-9 F catheter is placed distally to the arterial cannula by means of vascular ultrasound scan.
In female patients or in patients with a BSA less than 1.7 m 2 or in patients with a severe peripheral vascular disease, the distal catheter is inserted as soon as possible.
Patients with acute coronary syndrome complicated by advanced cardiogenic shock had a higher survival than patients presented with cardiac arrest. Kim et al. [45] reported an early survival of 59.2% in a group of 27 patients and described a long-term survival of 42.9% at 3 years. Bermudez et al. [39]

Future implications
Recently, some Authors have reported early results about the use of IABP in the setting of cardiogenic shock following acute myocardial infarction and in these articles the IABP seems to have not robust data to be still considered as the tool of first choice in the treatment of cardiogenic shock. Seyfart et al [57], in a randomized study of 25 patients with CS, randomly assigned to IABP (n=13) and percutaneous Impella 2.5 (n=12), reported a superior hemodinamic parameter and a significative increasing of cardiac index in patients treated with Impella; the 30 days mortality (46%) was not different in both groups. In a meta-analisys published by Sjauw et al. [58] about the use of IABP in the setting of ST-elevation myocardial infarction complicated by cardiogenic shock, the Authors could not find robust data in favours of the use of IABP. Different complications such as stroke and bleeding and increasing of 30days mortality in patients managed with IABP were observed. A very recent article by Thiele and al [59], 600 patients affected by CS following acute myocardial infarction, were randomly assigned to IABP therapy (n = 300) or conventional therapy (n = 298). The Authors could not find significant differences in 30-days mortality and in secondary end points or in process-ofcare measures, including the time to hemodynamic stabilization, the length of stay in the intensive care unit. No other significant differences with respect to the rates of major bleeding, peripheral ischemic complications and stroke were reported between the two groups. Howewer all these results have received different criticisms due to small number of patients [57] or a high number of patients with a relatively low mortality risk if treated with conventional therapy [59] and therefore these report could be influenced from some confounding factors.
According to these recent results, in the next close future, it can be argued that the use of ECMO could be more encouraged and anticipated in such patient who are in the setting of "pre-shock", in order to reduce the complications linked to the low cardiac output and to reduce the rate of very late application of ECMO. The current systems are safe and simple to apply, due to the advance in miniaturized centrifugal pumps and circuits, to the increased biocompatibility (heparin-coated system), but they are still associated with major complications in a relatively high percentage. Big efforts are still needed to improve the current techniques and devices.

Conclusion
The use of V-A ECMO in patients with acute myocardial infarction complicated by refractory cardiogenic shock and or cardiac arrest is widely increasing due to the improving in the early e mid-term results. The relatively low early survival rate in these very illness patients supported by ECMO should be considered an encouraging data, because in these patients the mortality without the ECMO support is dramatically higher. Bleeding, infections and CNS irreversible damage remain still serious complications and efforts to reduce or prevent them are necessary and strongly recommended to improve the outcome.