Extracorporeal Membrane Oxygenation (ECMO) Use in Heart Transplantation

Abstract

Extracorporeal membrane oxygenation (ECMO) use in patients both pre and post heart transplantation has become a life-saving tool in the armamentarium for physicians. Early developments in perfusion were hindered by the challenges of coagulation. Technological advances leading to contemporary management of ECMO began with the oxygenation of blood circulating through an artificial cistern. VA ECMO provides mechanical circulatory support (MCS) to patients not responding to medical treatment with primary cardiac dysfunction or combined respiratory and cardiac dysfunction failure. Management of the patient on ECMO is complex and involves multidisciplinary daily input from cardiology, cardiac surgery, and intensivist with ECMO specialization. Indications for ECMO use in heart transplantation include bridge to decision, bridge to transplant, rescue from PGD, and management of vasoplegia. The use of ECMO pre transplant has increased in the United States since the heart allocation changes implemented in 2018. Outcomes utilizing ECMO in pre and post heart transplant patients appear to be improving worldwide.

1. Introduction

Extracorporeal Membrane Oxygenation (ECMO) has become a vital life-saving technology in the management of advanced heart failure and heart transplant patients. Failure of optimal medical therapy in heart failure shifts the treatment options from neurohormonal modulation to surgery and palliative options. Historically, ECMO was for brief support for periods of 5–10 days however prolonged runs have been reported [1]. Contemporary use in heart failure may be classified as bridge to transplant, bridge to recovery, destination, or bridge to decision.

ECMO is employed pretransplant to provide support to patients with end-stage cardiac failure when conventional optimal medical therapy is insufficient. ECMO can be used in pure cardiac dysfunction with left, right, and biventricular failure, and unlike other mechanical options, in patients with concomitant pulmonary insufficiency due to pulmonary edema ECMO provides respiratory support. Veno-arterial extra corporeal membrane oxygenation (VA ECMO) may be utilized in neonatal, pediatric, and adult populations, making it a viable support modality for various patient populations. In the postoperative heart transplant patient, primary graft dysfunction is the most common indication for mechanical circulatory support using VA ECMO.

2. History of ECMO

Pioneering animal research on extracorporeal oxygenation began in the seventeenth century. Early experiments that included perfusion were initially limited by coagulation/thrombosis and only became possible with the development of anticoagulants. Direct contact oxygenation became available in the nineteenth century with agitated oxygenators and motorized oxygenators. Bubble-type and surface-type oxygenators were described by Hooker in the early twentieth century [2]. Following the discovery of anticoagulation, stationary screen oxygenators were developed, and subsequently modified pump oxygenators became commercially available. Rotating disc film oxygenators, which were initially developed for animal use and then adapted to human use was made possible in the mid-twentieth century. With the development of the DeWall bubble oxygenator, concerns about optimum bubble size arose. Small bubble sizes achieved a higher surface area for direct contact with blood and allowed better oxygenation; however, they had a longer transit time through the oxygenating column predisposing patients to air embolism. The best compromise was achieved with bubble sizes between 2 and 7 mm in diameter, a mixture of large- and small-sized bubbles and optimizing the rate of fresh gas flow. The use of these bubble oxygenators resulted in blood shearing with resultant hemolysis, activation of the inflammatory response, platelet activation, and increased thrombogenicity.

Technological advances that ushered in the era of contemporary ECMO began with the oxygenation of blood as it flowed through an artificial cistern. Kolf and Berk observed that the gas contents of blood approximated those of aerated dialysate across a cellophane tube [3]. This led to research on biomaterials with the hope of identifying materials with optimum gas exchange and mechanical properties. Initial materials that were tested include ethylcellulose and polyethylene, but these materials were limited by mechanical strength and presence of pinholes. The development of the membrane exchange system, the membrane lung, was made possible using polysiloxane or silicon polymers.

The development of hydrophobic, microporous membranes with pore sizes less than 1 micron, allowed direct contact of blood and air while plasmatic leak across the micropores limited by surface tension forces. These membranes allowed high-performance membrane oxygenators capable of being used for long ECMO runs.

Gibbon reported the first successful use of a mechanical extracorporeal lung and heart bypass during cardiac surgery in 1954 [4], and Hill et al. reported the first use of prolonged extracorporeal bypass in the context of post-traumatic respiratory failure in 1972 [5]. Extracorporeal support lasted for 75 hours in this case. In 1975, Bartlett went on to describe the first successful ECMO in a neonatal patient [6]. Following a premature termination of a multicenter trial of ECMO for ARDS, the development of ECMO was stalled for 20 years [7]. Thereafter, significant progress has been made in the circuit, management, indications and outcomes [8].

3. General concepts of ECMO

VA ECMO is a temporary modality of mechanical circulatory support (MCS) that can be used to support patients with pure cardiac dysfunction (RV, LV and Biventricular) and/or combined respiratory and cardiac dysfunction failure who are refractory to optimal medical treatment. In these scenarios, ECMO is performed via a veno-arterial route. The venous drainage is from the right atrium which is usually accessed percutaneously through the femoral vein, however, the subclavian, jugular veins or a surgical central venous system may be used. Arterial cannulation can be either central/aortic or peripheral using the femoral, axillary, or subclavian artery. A recent systematic review and meta-analysis comparing central versus peripheral canulation for ECMO reports that central canulation is associated with greater hospital mortality, reoperation, and transfusion [9]. The peripheral arterial cannula range in size from 15 to 19 Fr. Central arterial cannulas are larger: up to 22 Fr. Adult-size venous cannulas used for drainage are often reinforced to avoid kinking, about 50 to 70 cm in length with a diameter of 19 to 25 FR. The presence of multiple side cannula holes along the end of the cannula facilitates drainage [10]. In pediatric patients, the sizes of the cannula are matched with the body surface area. This matching is critical to optimal delivery of support because the resistance to blood flow in the extracorporeal circuit is inversely proportional to the diameter of the return cannula [11]. Hence, optimal flows are facilitated by larger bore cannulas [12]. The ECMO circuit consists of a centrifugal pump; the oxygenator; the heat exchanger; the tubing of the circuit; the control console; the gas supply and blender; sensors and monitors; safety systems; and the power supply.

4. ECMO cannulation and management

The procedure for the institution and management of VA-ECMO support for bridge to transplant is a sterile surgical procedure and requires intravascular access with drainage and inflow cannula. This intravascular access may be achieved via the Seldinger technique or via a cut down to the target vessels. Before the commencement of ECMO, ACT should be confirmed to be greater than 200 seconds. The oxygen line is connected to the oxygenator and the gas flow set at about 5–6 L/min. The sterile loop is handed to the cannulating physician; this is cut between clamps, and the connections of the tubing to the canulae are made to the exclusion of air. Flows are adjusted to maintain mean arterial pressure and adequate arterial oxygenation. Baseline ventilator requirements and anticoagulation sampling times are determined. The lines are secured as required using anchoring sutures and Opsite® (Smith and Nephew). Typical ventilatory setting may include FiO2 less than 70%, positive inspiratory pressure less than 35cmH2O, PEEP of <10, and respiratory rate < 10. Typical flow rates for a VA-ECMO would be in the range of 2.1 to 2.4 L/min/M2. Prolonged flow rates less than 2 liters should be avoided as this predisposes to thrombosis. Pre-membrane pressures should be kept below 300 mmHg to avoid catastrophic line accidents. Target transmembrane gradient should be less than 50 mmHg. The circuit should be changed if transmembrane gradient is above 150 mmHg as this may be due to clot formation and/or fibrin accumulation in the oxygenator. The ECMO pumps are almost exclusively centrifugal pumps and consist of a rotor that drives blood on rotation. Contemporary oxygenators are commonly Quadrox® (Getinge) or Nautilus® (Medtronic) oxygenators. The Quadrox® (Getinge) oxygenator uses hollow fiber membranes for gas exchange and have a compact design that integrates the heat exchanger with the oxygenator. The Nautilus® (Medtronic) oxygenator uses hollow fiber or flat sheet membranes for gas exchange and has a modular design with separate heat exchanger and oxygenator. Understanding these designs will assist healthcare providers customize the set up to individual patient needs.

Signs of limb mal perfusion as well as differential perfusion as seen in Harlequin syndrome should be monitored for and documented. Harlequin syndrome results from blood flow through the heart that is not oxygenated, because of lung pathology, being ejected to upper body. The oxygenated blood which is delivered back to body through a groin cannula meets the unoxygenated blood near the distal arch. Coronary and brain perfusion may be compromised. It is imperative to monitor PaO2/ SaO2 in right upper extremity. For ECMO-Impella combination therapy, the cannulation strategy involves a standard femoral or central cannulation for ECMO while the Impella device may be deployed through the femoral or subclavian artery. Typically, the Impella is deployed on the contralateral side of the ECMO cannulation site.

Other parameters monitored include temperature with a goal of maintenance of normothermia using the heater - cooler. ACT/ aPTT guided titration of heparin dosing should be done 6-hourly for the first 24 hours. Following the exclusion of reactionary hemorrhage, ACT may be lowered from commencement levels of above 200 to about 150 if platelet counts are stable. Measurement every six hours aPTT with a target of 40 to 80 seconds should be used to monitor patient subsequently. Direct thrombin inhibitors such as Bivalirudin may be used in cases of heparin resistance requiring frequent antithrombin III replacement or heparin-induced thrombocytopenia. Serum lactate, renal function hepatic function and daily wake up assessment should be monitored according to institutional protocols.

While VA ECMO improves systemic flow and unloads the right ventricle it does increase LV afterload. The LV is subjected to significant increases in afterload from blood being pumped backwards toward the aortic valve. This may lead to a loss of pulsatility, aortic valve closure, and if aortic insufficiency is present LV distension and transmural ischemia and increasing LVEDP and eventually pulmonary edema. Decreased aortic valve opening, LV distention and stasis can cause LV cavity thrombosis and subsequent strokes and peripheral artery embolic events. Careful monitoring for left-sided distension is required using chest x-rays, attention to ventilator plateau pressures and oxygenation needs, as well as assessment of pulmonary artery pressures using a pulmonary catheter, and frequent echocardiographic assessment. Early treatment with inotropes may be sufficient, however with continued LV distension, further interventions may be necessary. Intra-aortic balloon counter pulsation reduces after load and may be easily inserted in ICU. The placement of a LV axial flow pump, (Impella ® (Abiomed)) is an excellent way to achieve decompression. This may be done in the Cath lab or the OR under fluoroscopic guidance. If a patient has prosthetic aortic valve, percutaneous pulmonary artery vents can be placed or surgical vents which are less desirable because of the incision with bleeding consequences. Prevention, rather than treatment, of these complications should be the rule and routine use of LV unloading early during an ECMO run with an LV axial flow pump is strongly encouraged.

5. ECMO use in heart transplantation

Extracorporeal Membrane Oxygenation (ECMO) can play a crucial role in heart transplantation by providing temporary circulatory and, if necessary, respiratory support to patients pre- and/ or post transplantation.

5.1 Bridge to decision

In patients presenting in cardiogenic shock, emergency ECMO may be used to resuscitate and support the cardiovascular system while the patient is evaluated for multiple decision pathways including transplant, durable MCS, shifted palliative care or achieve recovery from shock and are weaned off VA ECMO. The availability and speed of deployment of VA ECMO makes it the circulatory support device of choice in emergent potentially terminal cardiac events such as refractory ventricular fibrillation arrests, refractory cardiogenic shock, and pulseless electrical activity. Use of ECMO prior to determination of subsequent treatment options, which may include transplantation, is referred to as bridge to decision. It is notable that, in a significant proportion of patients, sufficient recovery occurs to allow weaning from the ECMO support [13].

5.2 Bridge to transplant (BTT)

ECMO use as a bridge to transplantation is most commonly used for patients already on the wait list, as opposed to those placed on the wait list while already on ECMO [14, 15, 16, 17]. By providing mechanical support of the heart, ECMO allows patients to stabilize and recover from cardiogenic shock refractory to maximal medical management, while waiting for a suitable donor organ to become available. This not only increases their chances of successful transplantation but also reduces the risk of further complications and end organ deterioration. Notably, ECMO improves renal and hepatic perfusion, with the goal of normalization of function prior to transplant [18]. This support not only helps stabilize the patient’s condition but elevates their status on the transplant waiting list.

5.3 Rescue from primary graft dysfunction (PGD)

Primary graft dysfunction is identified when the transplanted allograft fails to function adequately in the early postoperative period [19]. PGD may be at least partially attributable to ischemia-reperfusion injury of the graft, immunological response and from a combination of other host and donor factors. It is recognized as a low cardiac output state, with potentially pulmonary edema and/ or arrhythmias. The right ventricle, left ventricle, or biventricular involvement can occur. Initial management involves expansion of intravascular volume and use of inotropes; however these may not suffice and, unhindered, may lead to multiple organ dysfunction. Rescue treatment by use of VA ECMO support maintains perfusion and oxygenation while the graft recovery occurs. Peripheral cannulation support in PGD was associated with better survival then central [20].

5.4 Management of Persistent Vasoplegia

Vasoplegia occurs when systemic vasodilation with consequent hypotension complicates the hemodynamics in the post CPB/post-transplant period. It is at least partially attributable to unregulated release of nitric oxide from reperfusion injury. This syndrome leads to severe hypotension refractory to high doses of vasopressors, despite normal appearing cardiac function. The hypotension and high dose vasopressors have deleterious effect on end organ perfusion. Medical management includes fluid therapy, Vitamin B12, vasopressors including Angiotensin II, and intravenous methylene blue. VA ECMO support should be instituted in persistent vasoplegia as a rescue therapy when medical management fails to restore adequate perfusion to prevent end organ failure.

6. ECMO as a bridge to transplant

Worldwide, changes to heart allocation guidelines prioritize patients on mechanical circulatory support, including ECMO. This new allocation system became effective in the United States in 2018, in the United Kingdom in 2016, and in France in 2004. Since then, there has been an increasing use of VA-ECMO as a bridge to heart transplant worldwide. Across the United States, the use of VA-ECMO as a direct bridge to heart transplant increased from 1.1% in 2012 to 5.2 in 2022 [21, 22]. Advanced heart failure has remained the most common indication in patients who receive VA-ECMO as a bridge to transplant. Unfortunately, the term “advanced heart failure” is not clearly defined. Patients with advanced heart failure can be identified by signs of circulatory failure with attendant renal and hepatic dysfunction, frequent malignant arrhythmias, and the need for continuous intravenous inotropes. Options for mechanical circulatory support include the Intra-Aortic Balloon Pump (IABP), various ventricular assist devices (VAD), VA ECMO and Impella ® (Abiomed). The choice of device for support is based on patient factors as well as available technology and expertise, residual left ventricular function, right ventricular function, pulmonary artery pressures, status of the aortic valve and need to unload the left ventricle. Bridging to transplant with ECMO has been considered controversial and not preferred because, compared to VAD, it requires immobilization, anticoagulation, and ICU care. Also, there is a high rate of waitlist mortality in patients bridged to transplant with ECMO. This practice is associated with a survival to hospital discharge rate of less than 50% [23].

The versatility and efficacy in delivering biventricular as well as pulmonary support has continued to earn its place as an option for MCS delivery. In addition, the speed of deployment makes it suitable for emergency situations, such as critical sudden circulatory collapse. In an analysis of the post 2018 allocation change in the United States, 3.6% of transplants were on ECMO (bridge to decision) at the time of listing and 5.2% of transplants in the same period were on ECMO at transplantation (bridge to transplant) [21]. The success of ECMO bridging to heart transplant or VAD is estimated to be between 44 and 79% with the most prevalent complications estimated to include bleeding (46%), arrhythmia (26%), device thrombus (24%), and device malfunction (23%) [24, 25, 26].

Several centers have reported bridging to transplant with ECMO as a risk factor for worse post-transplant survival. An analysis of the UNOS database by Moonsamy et al. showed that bridging with ECMO was a significant predictor of post-transplant mortality with lower 1 year survival rate of transplants bridged on ECMO (68 vs. 90%) [27]. In a similar analysis by Fukuhara et al. ECMO as direct bridge to transplant demonstrated lower posttransplant survival compared with continuous flow LVAD with survival rates of 73.1 versus 93.1% at 90 days, and 67.4 versus 82.4% at 3 years respectively [16]. These results reflect the acuity of patients on ECMO and the attendant issues with coagulation, end-organ perfusion, deconditioning, debility and inflammatory response associated with VA ECMO support.

Aimed at reducing the afterload, ECMO support may be combined with Impella support in a combined mechanical circulatory support known as ECMO-Impella combination therapy or ECPELLA, this has been used in cases of cardiogenic shock and was reported to be associated with activation of cardioprotective signaling [28]. The addition of Impella to ECMO support is based on the finding that, in contrast to ECMO, Impella reduces infarct size [29]. Hence the opinion that left ventricular unloading may mitigate myocardial injury in patients on ECMO. Delivery of hemodynamic support is done by titrating flow rates of the devices while avoiding complications. Weaning protocols for ECPELLA are individualized according to patients’ clinical context. Generally, in patients who are bridged to heart transplant and in whom ECMO component of combined therapy was used to provide respiratory support, weaning the ECMO first may be favored as the lung recovers. While in patients who bridged for heart and lung transplant due to pulmonary failure and failure of both ventricles, weaning the Impella first would be the goal.

In a retrospective review of ECPELLA use in cardiogenic shock, Patel et al. reported that 30-day mortality was significantly lower in the ECPELLA cohort compared with the VA-ECMO cohort, this difference remained significant after adjusting for STEMI and PCI [30]. Also, one-year all-cause mortality was significantly lower in the ECPELLA group. The use of ECPELLA was associated with higher rates of hemolysis and need for hemofiltration [31].

Reviews of the practice of VA ECMO bridge to heart transplant in the United States since the 2018 change in UNOS allocation policy shows that there has been an increase in the use of temporary MCS, including ECMO, following the policy change [21, 32]. This increase in the use of temporary MCS was noted to have occurred in transplant centers but not in other critical care units, despite similar patient characteristics during two time periods. The allocation system may have driven a change in practitioners’ behavior, not the patients [33]. Considering outcomes, Cogswell et al., Trivedi et al., as well as Kilic et al. reported a decrease in unadjusted post-transplant survival following the policy change [34, 35, 36]. Goff et al. reported no difference in unadjusted post-transplant survival following the policy change [37]. In contrast, patients supported on ECMO following the most recent allocation policy change were more likely to receive heart transplants, and less likely to die on the waitlist or be removed from the waitlist; they also had a lower average waitlist time [38].

These varying results may be attributable to informative censoring. More transplant candidates are waitlisted on ECMO since the most recent allocation policy change, however the demographics of patients waitlisted on ECMO did not change with the new policy. Comparatively, a greater percentage of ECMO bridges are undergoing transplant in the new era compared to the old allocation policy (78.6 vs. 37.6%). Following transplantation, 90-day survival in ECMO bridged patients were comparable in the two eras with a nonsignificant decrease in 1 year survival in the new era [14].

Kim et al. noted that there was improved waitlist and post-transplant outcomes in patients bridged with ECMO following the most recent change in heart allocation policy [39]. In their report, bridging with ECMO was strongly associated with worse 1-year mortality prior to the policy change, while ECMO patients transplanted after the policy change have similar post-transplant outcomes as non-ECMO patients. They concluded that their findings suggest that ECMO may be most effective as a bridging modality when patients are transplanted promptly following initial listing. There is also a possibility that nonrestrictive patient selection for ECMO bridging may be a possible alternative explanation. Despite these issues, when deaths within first 30 days of transplant were excluded, the use of ECMO was not associated with increased mortality [40, 41].

Predictors of failure of the ECMO bridge to transplant have been reported to include left ventricular EF, older age, ischemic heart disease, pre-ECMO cardiopulmonary resuscitation, and high Sequential Organ Failure Assessment (SOFA) scores [42, 43]. Other factors associated with failure include the presence of female sex, dilated cardiomyopathy, decompensated heart failure, the occurrence of complications such as stroke, and the need for dialysis [24]. There is little hope that a complete consensus will soon be reached on the age limit for successful bridging. Smedira et al. recommend 60 years with decompensated heart failure, while Chung et al. [43, 44] recommend 50 years with a SOFA score of 10. Analyses of the organ procurement and transplant network data on ECMO bridges in the pediatric population in the US showed that 45% of the cohort bridged on ECMO survived to transplantation, 10% of the cohort recovered enough to be removed from the waitlist, 11% were removed from waitlist due to clinical deterioration and 28% died while on the waiting list [24].

In an analysis of use of ECMO bridge to transplant by Dong et al., median bridging time in the BTT group was 13 days (interquartile range [IQR], 7–19 days) [45]. Generally, ECMO support longer than 14 days is a risk factor for posttransplant mortality [24, 46]. It is important to note that while ECMO is reliable for temporary bridge to transplantation, it is not an option for long-term mechanical circulatory support to bridge transplant candidates. When full flow on ECMO for more than a few days is required to achieve adequate organ perfusion, escalation of ECMO to LVAD provides a safe and durable support that provides cardiac unloading, high flow, and physiological circulatory patterns [47].

Outside the United States, experience with the use of ECMO as bridge for heart transplantation showed that, comparatively, temporary left ventricular assist devices were associated with a lower risk of death in the first year of the bridge when compared with VA-ECMO [25]. The higher mortality noted in patients who were bridged with VA-ECMO could be attributable to the higher prevalence of critically ill patients and more adverse events in the ECMO group [35].

In the United Kingdom, VA-ECMO bridge-to-transplant is uncommon, while temporary LVAD such as Impella ® (Abiomed), make up the vast majority of bridged patients. In a review following the change in allocation system, only a handful of patients were bridged using VA-ECMO [48].

Analysis of the French national registry shows ECMO bridged patients to have higher priority and shows that patients bridged with ECMO have higher incidence of being dependent on intravenous inotropes, mechanical ventilation, and dialysis. Despite these features suggesting higher acuity among heart transplant patients bridged on ECMO, French patients bridged on VA-ECMO had a comparably good outcome with equivalent mortality in single center analyses [49] and even lower risk of mortality in national database analysis [50]. Notably, the duration of extracorporeal life support was short with a mean duration of 6.3 ± 4.6 days [15].

7. Indications for ECMO bridge to transplant

Indications for ECMO bridge to transplant are not different from published indications for mechanical circulatory support, namely, advanced heart failure with progression of failure despite optimal medical therapy. The American Heart Association admits that there is no consensus on the definition of advanced heart failure [51]. This leaves practitioners to identify patients with advanced heart failure by the presence of subjective features such as the requirement for critical care, continuous ionotropic requirements, and consideration for heart transplantation, MCS, or hospice. Objective assessments, where used, include measures of functional limitation such as peak Vo2 of ≤14 (or < 50% of expected) mL/kg/min and a 6-minute walk of less than 300 meters. In practice, it is often not feasible to determine these objective assessments at the time the decision to institute ECMO support is made. The ideal scenario would be for the shock team to make the decision even in emergent cases, based on the clinical state of the patient.

After failure of medical therapy, options for effective therapy are limited to mechanical circulatory support, heart transplant and palliative care. Due to the scarcity of donor organs, temporary or durable mechanical circulatory support becomes the default option in the absence of donor hearts. When patients require both cardiac and pulmonary support, VA ECMO is the only option.

8. Post-operative ECMO use in heart transplantation

Indications for the use of ECMO mechanical circulatory support following a heart transplant include the occurrence of hyperacute graft rejection, post-operative bleeding, severe vasoplegia, and primary graft dysfunction.

Hyperacute rejection occurs within minutes to hours following a heart transplant, results from an antibody-mediated rejection (AMR) in response to HLA donor specific antibodies, and results in rapid damage to the graft with deposition of C4d complement. The diagnosis of antibody mediated rejection was previously, made in cases of graft dysfunction in the absence of cellular rejection. Recently, the requirement of histological evidence of antibody mediated injury was included [52, 53]. This inclusion helps to differentiate hyperacute rejection from non-immunological causes (NIC) of graft dysfunction. Notably, transplant guidelines recognize that AMR may occur when neither donor specific antibodies nor C4d deposition is present. Both AMR and NIC may present with severe pulmonary edema, profound coagulopathy, and biventricular failure. With current practice of HLA matching, hyper-acute rejection due to AMR has become rare [54, 55]. There are still instances of hyperacute graft dysfunction in orthotopic heart transplantation in the presence of non-HLA antibodies. These have been attributed to anti-angiotensin type I receptor antibodies [56].

The entities of vasoplegia and PGD after orthotopic heart transplantation have significant overlap with both conditions having a common factor of severely reduced afterload. It is possible that either entity may be complicated by the other for example, the increased capacity of the vascular system resulting from loss of vascular tone and attendant vasodilation as seen in vasoplegia may, in the presence of a denervated heart and loss of cardioaccelerator response, lead to ineffective cardiac output and resulting fall in afterload. The resulting cardiovascular collapse leads to failure of coronary perfusion and other end-organs. PGD which may result from myocardial stunning, leading to inadequate cardiac output which is beyond the capacity of the initial vasoconstrictive response to maintain cardiac afterload. The report of increase in PGD with longer ischemic time supports the second theory [57, 58]. A third possibility is autonomic dyssynergia between the denervated cardioaccelerator beta 1 supply to the heart and the vasoconstrictive alpha supply to the vascular system.

Regardless of the pathogenesis, the use of a VA-ECMO in patients with severe primary graft dysfunction and/or vasoplegia following heart transplantation is recommended in situations of escalating inotropes and poor ventricular function [59, 60, 61]. VA-ECMO empties the right heart, allows systemic circulatory support and increases afterload makes this modality an appropriate technology for the management of PGD and vasoplegia. Pressure dynamics following institution of VA ECMO may result in failure of the passive aortic valve to open due to the continuous pressurization of the aorta above the left ventricular pressure. This could result in LV stasis and thrombosis [62]. Failure of left ventricular emptying may result in distension and increased wall tension, decrease of subendocardial perfusion and perpetuate the graft failure [63]. The target should be to prevent rather than treat ECMO related complications in the post-transplant period.

9. Weaning

Following satisfactory cardiac recovery evidenced by minimal inotropic support, maintenance of pulse pressure greater than 15 mmHg, and MAP above 65 mmHg at ECMO flow rate of 2 L/min; the patient is considered for weaning from the ECMO support. Prior to this it is imperative to ensure that bleeding has resolved, any indication for delayed sternal closure should have been resolved, and central venous pressure should be within acceptable limits.

Several successful protocols have been employed for successful VA ECMO weaning. Once such protocol involves a turn down test performed under TEE guidance. The process of weaning consists of reducing the flow rate by 50% for 10 minutes; then, if the right and left ventricular EF do not deteriorate and no mitral regurgitation or distension of the left ventricle occurs, an additional 25% reduction of flow support is done for 5 minutes. If at this time, the TEE demonstrates good function, then weaning may commence. Recognizing that even minimal ECMO flows decompress, and hence may not completely test, the right heart function; the interim ELSO guideline on VA ECMO recommends Pump Controlled Retrograde Trial where the arterial flow probe is reversed, and 1 liter of flow is returned to the right heart through the venous cannula. Successful weaning is expected following a turn down test evidenced by mean arterial pressure above sixty mm Hg, left ventricular outflow track velocity time integral (VTI) above 0.12 m/s, tissue Doppler lateral mitral annulus peak systolic velocity of 6 cm/second and above, central veinous pressure of ≤10 mm Hg, and left ventricular EF ≥ 25–30% on minimal doses of one or two inotropes or pressors [64].

Another weaning protocol consists of progressively reducing the support daily and at least a daily TEE check. In this protocol, the support is initially reduced to 75% support for 24 hours, if the hemodynamics remain stable, a TEE is performed to confirm that ejection fraction remains greater than 40% with no left ventricular distension and MR. Weaning is advanced by further reduction to 50% support for 24 hours, and then 40% while ensuring stable vitals, inotropes and labs. It is acceptable to decannulate if parameters remain stable with 25% support for one hour [59].

10. Summary

In summary, ECMO is a valuable tool for mechanical circulatory support in heart transplant patients. The development of ECMO has a rich history that continues to advance the care of critically ill patients before heart transplant, and in cases with post-op catastrophic complications that were hitherto certain mortality. Advancements in anticoagulation and extracorporeal oxygenation have been fundamental to the development of this technology.