Since 1963, when the first human liver transplantation (LT) was performed by Thomas Starzl, the world has witnessed 50 years of development in surgical techniques, immunosuppression, organ allocation, donor selection, and the indications and contraindications for LT. This has led to the mainstream, well-established procedure that has saved innumerable lives worldwide. Today, there are hundreds of liver transplant centres in over 80 countries. This review aims to describe the main aspects of LT regarding the progressive changes that have occurred over the years. We herein review historical aspects since the first experimental studies and the first attempts at human transplantation. We also provide an overview of immunosuppressive agents and their potential side effects, the evolution of the indications and contraindications of LT, the evolution of survival according to different time periods, and the evolution of methods of organ allocation.

Liver transplantation (LT) celebrated its 50^th anniversary in 2013, with the first human procedure being performed by Thomas Starzl in 1963. Since this time, many strategies have evolved worldwide regarding technical aspects, immunosuppressive agents, organ allocation, donor selection, indications and contraindications, prophylaxis of infection, and the prevention of recurrent diseases[1]. This review aims at describing some aspects of LT regarding the progressive changes that have occurred over the years.

In 1952, in Milan, Italy, Vittorio Staudacher was the first to perform a LT in a large animal model, a canine species[2]. Two years later, Jack Cannon was credited with the first animal orthotopic LT[2]. Still in experimental models, in 1960, in Colorado, United States, Thomas Starlz reported his experience with almost 80 canine liver transplants, where the maximum survival was 20 d[3]. In the same year, Francis Moore also reported his experience with over 30 canine homotransplants in Boston, MA, United States[4]. These experimental experiences expanded the available knowledge on many issues such as venovenous bypass, organ preservation, tissue matching and immunosuppression.

In humans, solid organ transplantation began in 1954 with a successful kidney transplantation between identical twin brothers[5]. In 1963, Starzl et al[6] published the first three attempts at human LT, but it was not until 1967 that the procedure resulted in an extended survival. It was the case of a 19-mo-old girl with hepatocellular carcinoma, who died 13 mo after surgery for metastatic disease[7]. Roy Calne, in Cambridge, United Kingdom, joined Roger Williams in London, United Kingdom, in 1968, and reported 5 cases of liver transplant, detailing the technical difficulties encountered[8]. Thomas Starzl and Roy Calne were later honoured with the Lasker-DeBakey Clinical Medical Research Award in 2012 for these pioneer procedures.

The acceptance of the concept of brain death in the United States in 1968 was an additional landmark for LT development, which allowed donor organ preservation in ideal, physiologic conditions and resulted in better graft quality and survival[9,10].

The introduction of cyclosporine in the late 1970s as part of the immunosuppressive regimen in organ transplantation permitted less toxicity and the prevention of rejection and severe opportunistic infections when compared to azathioprine[11,12]. Later on, with the introduction of tacrolimus, the world observed further improvements in survival[13,14].

A series of 540 cases united by 4 different liver transplant units was presented at the 1983 NIH Conference, showing greater survival in these patients compared to those who did not undergo transplantation, permitting the establishment of LT as a beneficial procedure for patients with end-stage liver disease, as opposed to an experimental procedure[15].

In 1984, Bismuth et al[16] (France) reported the first left-lobe LT in a child, and in 1988, Pichlmayr performed the first split-LT in Hannover, Germany[17]. The following year, in Sao Paulo, Brazil, Silvano Raia described the first attempt at a living donor graft in a child[18], with a successful procedure performed by Strong et al[19] in Brisbane, Australia, in 1990. Since this time, the greatest experience with living donor grafts has been found in countries where cadaveric donation is not culturally accepted such as Japan and South Korea.

Since the nineties, the field of LT has witnessed a huge expansion of the number of institutions performing the procedure, and today, there are hundreds of liver transplant centres in over 80 countries. In addition, an increasing number of conditions associated with end-stage liver disease are now referred to LT. This has led to the scarcity of donor organs, thus obliging transplant coordinators worldwide to adopt evolving organ allocation strategies[20]. Immunosuppressive, technical, infection risk and intensive care management advances have made LT a long-lasting, efficient therapy for end-stage liver disease, hepatocellular carcinoma and other hepatic cancers. Advances in organ procurement, preservation, and allocation have accompanied these improvements. The indications have progressively expanded, and the contraindications have slowly changed as technical issues have evolved.

During these 50 years, many surgical techniques and strategies were reported, involving donor and recipient operations. The surgical techniques employed in the first experiments and in clinical trials are described and the following variations or evolutions are discussed: piggy-back vs conventional technique, split liver, living donor liver transplantation (LDLT), and domino liver transplantation (DLT).

Graft rejection, either acute, late acute, or chronic, is an immune-mediated disease. The risk of graft rejection is counterbalanced by the permanent use of immunosuppression, which has been a crucial concern in LT since the early 1960s. The use of efficient immunosuppression starting in the late 1970s represented a pivotal change in acute rejection and LT survival[75]. Today, mortality due to graft rejection is an uncommon event[76].

However, over time, the late complications of immunosuppression have been considered a major threat to LT morbidity. In addition to the risk of opportunistic viral, bacterial, and fungal infections, the continued use of immunosuppression may lead to degenerative and metabolic diseases as well as de novo malignancies[76]. A list of the most frequently prescribed immunosuppressive drugs and potential associated adverse events can be found in Table 1.

The choice of the best immunosuppression regimen is an individual-based decision and should consider a combination of variables[76]: (1) Time since LT: after the first 90 d following LT, the necessity for immunosuppression is reduced, as the graft becomes somewhat tolerant to the recipient´s immune injury[76]. However, most recipients require lifelong immunosuppressive therapy. In a frequency that varies between 20% and 60% of recipients[77-80], a phenomenon called operational tolerance may occur, and graft rejection does not occur despite immunosuppression withdrawal. This is specific to LT compared to other organ transplants because livers have a unique microenvironment that promotes tolerance rather than immunity[81]. Several authors have recently discussed that the reduction or even discontinuation of immunosuppression may be planned and safely achieved in a percentage of recipients that varies between 41% and 62%[80,82-87]. This leads to the diminution of morbidity and mortality by preventing long-term adverse events and the occurrence of opportunistic diseases as well as the improvement of quality of life. Successful tolerance has been associated with a longer time period since LT[84,87-89], male gender[87] and lower lymphocyte reactivity, measured by the phytohaemagglutinin stimulation index[84]. Regarding recipient age, successful immunosuppression withdrawal has been achieved in higher percentages among paediatric patients[80], but older age at LT has also been associated with tolerance[87]. However, biomarkers have been studied to identify the select patients who could benefit from immunosuppression withdrawal. Some studies have shown that tolerant patients can be characterised by an enrichment in peripheral blood natural killer cell gene signatures along with an increased frequency of γδ T cells and regulatory T cells[89,90]. More recently, other genes have been found to possibly define tolerance, including an enrichment of iron-associated genes together with an anti-inflammatory gene set[88]; additionally, a 3-gene signature in peripheral blood leukocytes was reported in a paediatric population[91]. Apparently, liver tissue-derived biomarkers are more accurate than blood-related markers at predicting the success of drug withdrawal strategies. Currently, they constitute the most robust biomarkers of operational tolerance[88]. However, biomarker profiles need to be prospectively validated to identify these recipients before drug withdrawal[92,93]. Another issue is the induction of tolerance early following transplantation, and the link between this and gene signatures is not clear[45,47]. Additionally, the long-term histological consequences of immunosuppression withdrawal are also unclear; (2) Cause of end-stage liver disease. Disease recurrence may be affected by the type of immunosuppression. For example, in autoimmune hepatitis and primary biliary cirrhosis, the early reduction of anti-rejection drugs has been associated with increased recurrence rates[77,94-96]. However, excessive immunosuppression has been associated with increased HCV recurrence[97], and high exposure to calcineurin inhibitors during the first month post-LT, defined as a mean tacrolimus trough concentration > 10 ng/mL or a cyclosporine trough concentration > 300 ng/mL, is associated with an increased risk of HCC recurrence[98]; (3) History of rejection: recurrent or severe rejection history; (4) History or risk of cancer or infectious complications; and (5) Comorbidities and adverse drug events, including prior experiences with immunosuppressive drugs.

The indications for LT have changed over the years. Overall, LT is indicated for acute liver failure, chronic liver failure leading to cirrhosis, and inherited metabolic liver diseases. It is also indicated for hepatocellular carcinoma (HCC) and other hepatic cancers, including hepatoblastoma, epithelioid haemangioendothelioma, and hilar cholangiocarcinoma (CCA), in selected cases as well as some miscellaneous conditions. The main current indications for LT are listed in Table 2[99].

The Clichy and King’s College criteria are the two main scoring systems used to select patients in cases of acute liver failure[100,101]. Both models achieve high specificity but remain associated with limited negative predictive value.

Regarding HCC, initially, only patients with unresectable, large, or multinodular tumours or with other associated underlying liver dysfunction were selected for LT, resulting in low survival and high rates of recurrence[102]. The restrictive selection criteria widely known as the Milan criteria were first established by Mazzaferro et al[103] and significantly improved patient survival. Later, Yao et al[104] at the University of California, San Francisco, demonstrated that a set of less conservative criteria, known as the UCSF criteria, had similar outcomes.

According to the European LT Registry, in the past, cancers constituted almost half of all the indications, and currently, this indication accounts for approximately 15%[105]. Transplantation for primary biliary cirrhosis has also decreased over time, as opposed to the increasing number of indications for alcoholic and hepatitis C cirrhosis, in Europe as well as in the United States[106].

Over time, the contraindications for LT have also evolved. One example of the change in strategy is hepatitis B. Before the availability of antivirals and hepatitis B immunoglobulin, there was poor survival of patients infected with hepatitis B due to the uncontrolled viral replication. This led most centres to abandon transplantation for this indication for several years. Nonetheless, currently, transplant survival for these patients presents excellent survival.

It was also believed that HIV infection constituted an absolute contraindication in the past; however, with the advent of highly active antiviral therapy and the evolving knowledge of drug interactions, these patients are regarded as potential candidates[107].

Advanced age is another relative contraindication in which we have observed changing patterns over the years. In the nineties, the population of patients over 60 years old accounted for approximately 10% of all transplanted patients, whereas currently, they constitute almost 20% of the procedures[105]. As there is no universally accepted age limit for considering transplantation, centres have dealt with this issue on a case-by-case basis, according to the physiological and functional status of the individual.

Portal vein thrombosis was also initially considered an absolute contraindication. However, long after the first report of LT in this context[108], operative strategies such as simple thrombectomy, extra-anatomic venous graft, arterialisation of the portal vein, and cavoportal hemitransposition may now be employed.

The absolute contraindications have hardly changed over time and constitute circumstances in which short- and/or long-term survival is compromised. The current absolute and relative contraindications are listed in Table 3[99,109].

Reasonable and impartial allocation criteria were not an issue during the first year of LT, as organs had a very short viability period and transplants were restricted to candidate recipients who were lucky enough to be managed at the same institution as the deceased donor[10].

However, the acceptance of brain death criteria in several countries improved donor organ preservation and quality, making it possible to allocate donated livers at distant sites. At the same time, an increasing number of conditions associated with end-stage liver disease have inflated transplant waiting lines, making it essential to develop a fair and structured organ distribution system[10,110].

Similar to renal transplantation, the first listing systems were based solely on waiting time. In the 1990s in the United States, the listing criteria were substituted by government-regulated policies, which later established priority for candidate recipients based on disease severity, according to a medical status system. In 1998, the US Department of Health and Human Services (DHHS) defined the principles of allocation policies and procedures to guide the Organ Procurement and Transplant Network (OPTN). From 1996 to 1999, the more objective Child-Turcotte-Pugh score was also incorporated, but disparities in waiting lines across regions were mounting. For many years, the modified Child-Turcotte-Pugh score was the main tool for estimating survival without transplant and was originally developed to predict the survival rates of patients undergoing portosystemic shunt surgery[111].

Lastly, the Model for End-stage Liver Disease (MELD) was adopted for recipient ranking in February 2002 in the United States[10,110]. For paediatric patients, a similar scoring system (Pediatric End-Stage Liver Disease - PELD) was created.

Several studies have analysed the impact of the use of the MELD/PELD score[110,112,113] and have demonstrated both a reduced waiting list mortality and unchanged patient and graft survival, despite the fact that transplants are performed in patients with worse clinical conditions[10]. Figure 1 illustrates the history of organ allocation policies in liver transplant in the United States.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 History of organ allocation policies in liver transplant in the United States. ICU: Intensive care unit; MELD: Model for End-stage Liver Disease; PELD: Pediatric End-Stage Liver Disease.

In Europe, there are no uniform rules for organ allocation. Organ procurement organisations for different countries include Eurotransplant (participating countries include Germany, The Netherlands, Belgium, Luxembourg, Austria, Slovenia, and Croatia), United Kingdom Transplant, Organización Nacional de Trasplantes in Spain, Scandiatransplant (Sweden, Finland, Norway, Denmark, and Iceland), North Italian transplant, and Agence de la Biomédecine (previously Établissement français des Greffes) in France. Although the majority of organs are allocated and transplanted within each organisation, there is some degree of collaboration among them. Within Eurotransplant and in France, allocation is patient based, but in Spain, Scandiatransplant and United Kingdom Transplant, it is centre based.

In Eurotransplant, the allocation can be categorised based on two time periods. From 2000 until 2006, recipient selection was based on a scoring system that considered the degree of medical urgency, donor weight, ABO blood group, waiting list time, and donor region[114]. Due to an increasing waiting list and positive experience with the use of MELD/PELD scores in the United States, this allocation system was implemented in December 2006. In France, grafts were allocated to a transplant centre by rotation, except for emergencies, until 2007. Within each centre, patients generally underwent transplantation according to their waiting time. In March 2007, due to high rates of mortality in some geographic areas, the French Liver Allocation Score (FLAS) was adopted. This score considers the MELD score as well as other conditions such as HCC and retransplantation indication, which are not necessarily associated with high MELD scores[115].

In the United Kingdom, an analysis of factors predicting transplant list mortality in more than 1000 patients in the waiting list identified 4 independent predictive factors, resulting in the development and validation of the UKELD score, which includes sodium in addition to the MELD score factors (bilirubin, INR of prothrombin time and serum creatinine)[116].

In July 2005, Argentina was the first country after the United States to adopt the MELD score system[117]. Initially, liver allocation was based on the location of care and time on the waiting list, with 2 categories (emergency and non-emergency patients).

In Brazil, from 1997 until 2007, liver allocation was based on the chronological order of registration on the waiting list and was substituted by the MELD/PELD scores in 2007[118].

The two main objectives of LT are to prolong survival and improve quality of life.

In the 1970s, the overall 1-year survival was approximately 30%, and the majority of patients died from rejection and/or infection[119]. Currently, 10-year survival rates may exceed 70% in many indications. The indications with better survival are generally primary biliary cirrhosis and autoimmune cirrhosis, whereas malignant liver tumours and hepatitis C may have worse outcomes due to high rates of recurrence[120]. Despite improved survival rates, LT recipients have an estimated loss of 7 years compared with an age- and sex-matched general population, with increasing differences for recipients at younger age[121].

Factors that may influence survival rates depend on donor, recipient, perioperative, and post-operative characteristics[122]. Donor parameters that may result in poorer outcomes include advanced age, high BMI, length of hospitalisation, use of vasopressors, and the presence of infection. Recipient parameters include urgent indication, renal dysfunction, age, mechanical ventilation requirement, hepatitis C, poor nutritional status, and the presence of infection. Perioperative factors include cold and warm ischaemia time, blood product requirements, and surgical difficulties. Finally, postoperative factors include primary non-function, renal dysfunction, centre experience, need for mechanical ventilation, and prolonged stay in an intensive care unit.

The majority of deaths and retransplantations occur soon after LT. The causes may vary, according to time after LT. Infection and primary non-function predominate in the early period, with perioperative factors accounting for nearly 60% of deaths in the first post-LT year[76,120]. After this initial period, de novo malignancies and other comorbidities such as cardiovascular disorders prevail alongside the recurrence of pretransplant diseases such as hepatitis C and autoimmune disease.

Long-term survival also results in higher morbidity due to prolonged immunosuppression as a result of the reactivation of prior infections, newly acquired infections, metabolic disorders (hypertension, diabetes, dyslipidemia, obesity), and de novo hepatic or extrahepatic malignancies, including posttransplant lymphoproliferative diseases (PTLDs)[76]. Cardiovascular failure and renal failure are the leading nonhepatic causes of morbidity and mortality in the long term[76].

In post-transplant care, LT institutions must include close follow-up to reduce cardiovascular risk, to prevent infections, monitor for cancer, and prevent and provide early treatment for recurrent diseases[76]. These measures impact both the survival and quality of life of these recipients.

LT has seen enormous progress over the last 50 years and can currently be considered a true life-saving procedure for patients presenting with liver failure. The main aspects that have evolved since the first series are surgical techniques and methods, the balance of immunosuppressive therapy, the prevention of infectious and non-infectious complications, and organ allocation. Much remains to be refined, such as the prevention of complications that are associated with long-term immunosuppression, organ allocation with the aim of improved survival, and the worldwide problem of organ shortage.

Edson Abdala has received fees as a travel grant and for serving as a speaker for MSD, as a travel grant from Pfizer, and for serving as a speaker for Novartis.