Extracorporeal Perfusion for Resuscitation of Marginal Grafts

Principles of Organ Preservation and Machine Perfusion

Interrupting the blood flow to an organ is an unavoidable step in the transplantation process. Even when the procedures in the donor and the recipient are perfectly coordinated, a period of ischemia necessarily transpires. As the supply of metabolic substrates declines, adenosine triphosphate (ATP) is depleted, sodium accumulates intracellularly, and the transmembrane electrochemical gradient is lost. Cell swelling ensues, which in the liver results in narrowing of the sinusoidal lumen and impairment of the hepatic microvasculature. Hypoxia also triggers a switch from aerobic to anaerobic metabolism and the development of lactic acidosis. Severe acidosis activates phospholipases and proteases, ultimately leading to cellular damage and death ( Fig. 106-1 ).

Cold storage is used in organ preservation to curb cellular metabolism and catabolic enzymes during the ischemic phase. As the temperature of the organ is lowered to around 4° C, molecular motion slows, and deleterious chemical reactions are suppressed. However, because hypothermia never halts metabolic activity completely, the cascade of events triggered by ischemia continues during cold preservation. Though the reactions occur at a slower rate, they nonetheless damage the cells of the organ in progressive fashion. Furthermore, rapid cooling of the organ to near-freezing temperatures leads to dysfunctional cation regulation in the endothelial cells, independent of cellular ischemia. An increase in the cytoplasmic concentration of calcium activates intracellular proteases and phospholipases, leading to actin disassembly and endothelial cell rounding. Matrix metalloproteinases are also released, and cell adhesion molecules are expressed on the endothelial cell surface. In addition, Kupffer cells, the resident hepatic macrophages, are activated in the cold ( Fig. 106-2 ). Although special preservation solutions have been formulated to curtail the detrimental side effects of ischemia and hypothermia, they are all, unfortunately, limited in their potential to halt these processes.

Shortcomings aside, static cold storage is a simple, convenient, and inexpensive means of preservation. As long as it is maintained for relatively short periods of time (<12 hours), it is sufficient for the preservation of most standard quality livers, and cold storage is the only form of ex vivo preservation used to any significant extent in clinical liver transplantation today. Cold storage becomes inadequate, however, when it is used to maintain grafts of marginal quality. The significantly higher rates of both immediate and delayed complications seen with older, steatotic, and donation after circulatory determination of death (DCD) livers are evidence of this fact. Moreover, given that the need for livers for transplantation expands as the number of ideal organ donors simultaneously declines, simply avoiding the use of marginal livers for transplantation is no longer an option.

In contrast to static cold storage, which merely slows organ degradation during ischemia, machine perfusion is designed to continuously deliver metabolic substrates, eliminate toxic metabolites, and actually restore graft quality before transplantation. In addition, machine perfusion offers the opportunity to assess organ viability in real time and may even offer the chance to intervene and treat grafts with cytoprotective and immune-modulating substances before reperfusion in the recipient.

History of Machine Perfusion in Solid Organ Transplantation

The concept of isolated organ perfusion is older than transplantation itself. In the early to mid 1900s, normothermic perfusion was studied as a way of sustaining the function of organs in isolation, both in situ and outside the body. When clinical transplantation activity began in the 1960s, however, interest turned from normothermic to hypothermic methods for maintaining organs, in particular the kidney. Continuous perfusion with cooled, oxygenated blood or plasma was initially the most successful form of extracorporeal renal maintenance.

During subsequent years, however, machine perfusion in organ preservation gradually fell out of favor. Although the first human liver transplants were performed using grafts arising from DCD donors, the Ad Hoc Committee of Harvard Medical School produced the definition of irreversible coma in 1968, making it possible for the first time to recover organs from donation after neurological determination of death (DND) donors with intact circulation at the time of donation. Given the fact that organs recovered from DND donors had not suffered warm ischemic injury and generally functioned better than those recovered from DCD donors, organ maintenance via extracorporeal machine perfusion became less of a necessity. Solutions for static cold storage also underwent dramatic improvements in terms of composition and ability to preserve organs in an adequate state of viability for periods of 24 hours and longer, and studies performed in the 1970s and 1980s indicated that there was little to no added benefit for machine perfusion over static storage in cold solution.

Over the past two decades, however, donors have become progressively older and less healthy in general. Given an improved ability to take care of patients with traumatic brain injury, fewer donors may be declared dead based on neurological criteria. Instead, many more now have life support intentionally removed to provoke cardiac arrest, converting what was a relatively orderly process of organ donation into a rush to cool down the organs as quickly as possible using the so-called super-rapid recovery technique. And in spite of our best efforts to limit ischemic injury in these grafts, we nonetheless continue to fail to adequately maintain their viability using the means of preservation most readily available to us. Allograft dysfunction leads to increased transplantation costs and significantly higher rates of recipient morbidity and mortality and has been recognized as a major problem associated with the use of marginal livers. Taking the aforementioned facts into account, as well as the fact that preservation technology has advanced considerably in recent years and smaller, more portable machines have been developed, it is no surprise that machine perfusion in solid organ transplantation is once again at the forefront.

Recirculation in Situ

Before discussing extracorporeal machine perfusion as a means of resuscitating marginal livers, it is worthwhile to discuss what has been learned from experiences with regional perfusion (RP) of abdominal organs in situ. RP is particularly relevant to the DCD process, in which the heart is arrested and therefore incapable of perfusing and oxygenating the organs during evaluation and recovery. If RP is not performed in this setting, the period of warm ischemia may become prohibitively long.

RP is performed using cardiopulmonary bypass or extracorporeal membrane oxygenation technology. Hypothermic RP has been used to maintain Maastricht category 3 DCD donors (life support intentionally removed to provoke cardiac arrest), with the purpose of recovering primarily kidneys and pancreata but also, on occasion, livers for transplantation. Results using this method, however, have been variable, and rates of delayed graft function have remained relatively high in some series. In addition, the group from La Coruña, Spain, has described their series of Maastricht category 2 DCD liver transplants (donors declared dead after unexpected cardiac arrest and failed attempts at resuscitation), in which potential donors were maintained according to diverse methods, including both hypothermic and normothermic RP.

Our group in Barcelona has been working in the normothermic perfusion of livers since the mid-1990s. We performed some of the first studies on the use of normothermic RP to improve the cellular energy load and posttransplantation graft function in porcine livers previously injured by periods of warm ischemia, and we demonstrated the superiority of normothermic RP over total body cooling in this setting. The restorative effects of normothermic RP were found to lie in its capacity to convert the episode of cardiac arrest into a period of ischemic preconditioning. In 2002 we applied our preclinical experience to the clinical setting and initiated a protocol to transplant livers arising from human Maastricht category 2 DCD donors treated with normothermic RP. Since then we have performed close to 40 of these transplants, and normothermic RP has arguably become the gold standard for the maintenance of DCD livers in situ. Nevertheless, we have observed that, in spite of a progressively increasing number of potential donors from year to year, the applicability of Maastricht category 2 DCD liver transplantation remains relatively low, arguing for the need for more advanced forms of extracorporeal organ maintenance, such as machine perfusion.

Two other groups have also described their use of normothermic RP to maintain DCD donors. Since 2000 the University of Michigan has been using normothermic RP to preserve Maastricht category 3 DCD donors and has described the transplantation of kidneys, a few livers, and a pancreas in this setting, while Saint Louis Hospital in Paris, France, recently began using normothermic RP in Maastricht categories 1 (cardiac arrest outside the hospital and declaration of death) and 2 DCD donors. Primarily kidneys have been used thus far, but as experience with normothermic RP in DCD expands, more livers should be available for transplantation as well.

Hypothermic Extracorporeal Perfusion of the Liver

The application of hypothermic machine perfusion (HMP) in solid organ transplantation is based on the principle that energy production via the electron transport chain can be sustained with a modest supply of oxygen and other metabolic precursors and a reasonably low flow rate at low temperatures. The generation of ATP during HMP helps restore cellular homeostasis and prevent mitochondrial collapse. Improving the state of the mitochondria is regarded as the key benefit of HMP, because cells are better able to cope with the oxidative burst at graft reperfusion when their oxygen-handling organelles are healthy.

Overall, HMP is relatively easy to perform. The temperature exchanger can be as simple as melting ice, obviating the need for a separate module and the energy source to run it. Also, there is no requirement for a specific oxygen carrier in the perfusion solution, because the demand for oxygen under hypothermic conditions is significantly reduced. Finally, at 4° to 10° C, concerns about graft infection resulting from bacterial growth are low.