Liver transplantation has evolved since Dr. Thomas Starzl performed the first orthotopic liver transplant (OLT) over 4 decades ago. Advances in immunosuppressive therapy, medical management, surgical technique, and identification of appropriate indications for OLT have resulted in significant improvements in patients’ survival and universal recognition of the procedure as preferred therapy for those suffering from hepatic failure. The number of patients awaiting primary or repeat OLT in the United States has tripled to 18,000 in the last 2 decades. Over the same period, organ availability increased from 1700 to 6200 grafts annually ; however, the concurrent increase in organ availability has not significantly impacted the rate of wait-list mortality; deaths on the waiting list have increased 5-fold over the same period. The discrepancy between supply and demand and the increasing organ scarcity has motivated select transplant centers to relax customary restrictions to donation, creating the term “extended-criteria” donors (ECD) or “marginal” donors. The precise definitions of these terms remain elusive. There is no consensus as to what makes a graft “marginal” in one center and acceptable in another. The use of these ECD grafts often depends on the judgement of the transplant surgeon and the needs of the recipient.
Definitions
An ECD implies higher risk in comparison with a reference donor. Conceptually, this added risk may manifest as an increased incidence of early failure, namely, delayed allograft function or primary nonfunction (PNF), transmission of a donor-derived disease, or, in the case of adult-to-adult living-donor liver transplantation (LDLT), living-donor morbidity. To appreciate the components that define an ECD, it is important to recognize the criteria that define a reference (or ideal) donor: These include age below 40 years, death caused by trauma, donation after brain death (DBD), hemodynamic stability at the time of organ procurement, no steatosis or any other underlying chronic liver disease, and no transmissible disease (infectious or neoplastic).
Durand and colleagues, in a Report of the Paris Consensus Meeting on Expanded Criteria Donors in Liver Transplantation, draw a distinction between an “ideal allograft” and an “ideal donor.” They mention that the ideal allograft category may be influenced by variables that are introduced after procurement, such as prolonged cold ischemia time (CIT) or technical variants, such as those occurring with allograft reduction (split-liver allograft). These variables should ideally not be included in the definition of ECD, because the aim is to assess risk at procurement.
Organ allocation and distribution
Liver allograft allocation and distribution are dependent upon a calculated disease severity score, the Model for End-Stage Liver Disease (MELD) Score, and geographic location of the recipient, with the United States being divided into 11 United Network for Organ Sharing (UNOS) regions. This has resulted in significant discrepancies in waiting list times for OLT and MELD score at the time of transplant. Thus, locations with short candidate wait times, or low MELD at the time of OLT, have the luxury of practicing selective organ donor use. ECD allografts may then be exported out of the vicinity of donor origin, after being declined by all transplant centers within geographic proximity to the donor, either to be prioritized for a higher MELD patient or as an “open offer” for use in an area of donor scarcity. ECD distribution is distinctly different from distribution of optimal organs that remain locally and are allocated by MELD score. ECD recipients are often selected by the transplant center rather than being allocated according to regional wait-list priority.
Organ allocation and distribution
Liver allograft allocation and distribution are dependent upon a calculated disease severity score, the Model for End-Stage Liver Disease (MELD) Score, and geographic location of the recipient, with the United States being divided into 11 United Network for Organ Sharing (UNOS) regions. This has resulted in significant discrepancies in waiting list times for OLT and MELD score at the time of transplant. Thus, locations with short candidate wait times, or low MELD at the time of OLT, have the luxury of practicing selective organ donor use. ECD allografts may then be exported out of the vicinity of donor origin, after being declined by all transplant centers within geographic proximity to the donor, either to be prioritized for a higher MELD patient or as an “open offer” for use in an area of donor scarcity. ECD distribution is distinctly different from distribution of optimal organs that remain locally and are allocated by MELD score. ECD recipients are often selected by the transplant center rather than being allocated according to regional wait-list priority.
Indicators of impaired allograft function
Donor Age
Advanced age is a nontechnical and nonmodifiable donor variable that has a significant impact on early allograft function ( Table 1 ). Liver weight, liver volume, and blood flow are reduced with aging. Older liver allografts have a lower tolerance for preservation. Endothelial injury from cold ischemia occurs earlier in older allografts, increasing the risk for inflammation, thrombosis, and T-cell–mediated rejection. Donor age has steadily increased over recent decades. In 1994, only 20% of deceased donors were 50 years or older. This percentage has increased to 34% by the year 2010. Although early studies suggested that older donors (>50 years) conferred no additional risk to poor outcome compared with younger donors, this notion has been refuted by more recent publications using large United States [Scientific Registry for Transplant Recipients (SRTR)] and European (European Liver Transplant Registry) transplant databases. Feng and colleagues, in a study of donor age correlation with graft failure from UNOS Transplant Registry data of more than 20,000 OLTs, found an increased risk of graft failure, which was significantly higher among donors older than 60 (relative risk of 1.53) and 70 years (relative risk of 1.65) compared with donors younger than 40 years. Donor risk related to age thus represents a continuum. There is, however, no absolute limit of donor age for liver transplantation. Some reports have shown excellent graft survival with octogenarian donors, provided that there are no additional risk factors, such as steatosis or prolonged CIT. An important caveat to the utilization of elderly donors is in the setting of hepatitis C virus (HCV). There are convincing data demonstrating early HCV recurrence and decreased survival of patients among HCV recipients of donor allografts older than 60 years, and so older donor livers should be used for HCV recipients selectively and with caution.
| Risk of Impaired Graft Function | Risk of Disease Transmission |
|---|---|
| Donor age (>60 years) | Positive hepatitis B and C serologies |
| Donor obesity | Unexplained cause of death |
| Steatotic livers (>40% macro) | Known donor malignancy |
| Donation after cardiac death | “High-risk” lifestyle |
| Hypernatremia (serum Na > 155 mEq/L) | Active bacterial/viral infections |
| Hypotension and inotropic support | Elderly donors |
| Prolonged intensive care stay | |
| Long ischemia times (CIT > 12 hours) | |
| Partial liver grafts (split/live donor) |
Donor Gender, Weight, Height, and Race
Some studies have identified donor gender (female gender) as a risk factor for worse post-OLT outcome, whereas others have failed to confirm this. African-American donor race consistently seems to have worse recipient outcome. With regard to donor size, less height has been shown to be independently associated with graft failure.
Donor Hypernatremia
Hypernatremia is a frequent clinical finding within the donor population that has a negative impact on function of hepatic allografts. The cause of hypernatremia could be related to derangement of fluid balance and diabetes insipidus in potential donors. The impaired allograft function is postulated to occur owing to a process whereby hepatocytes increase their intracellular osmolality to minimize cellular damage associated with the extracellular hypertonic state. Avolio and colleagues were the first to report a direct correlation between donor serum sodium concentration and peak serum aminotransferase after OLT. Several additional studies have validated this finding, and in general a donor serum sodium exceeding 155 mEq/dL at procurement is thought to be the threshold for decreased actuarial graft survival. Hypernatremia should preferably be corrected before organ recovery.
Donor Hepatic Steatosis
Steatosis is among the most important factors affecting liver allograft function. With the obesity epidemic in developed countries, it is not uncommon to encounter significant degrees of hepatic steatosis when procuring donor livers. Early functional recovery and regenerative capacity are significantly impaired with steatotic allografts, mostly because of severe ischemia–reperfusion injury. Steatosis has traditionally been classified as microvesicular (which has not correlated with poor allograft function) and macrovesicular. Although the liver is carefully inspected by the surgeon at the time of procurement, a biopsy is the gold standard to obtain an objective assessment on the degree of macrovesicular steatosis, which is subcategorized as mild (<30%), moderate (30%–60%), or severe (>60%). Mild steatosis generally has minimal impact on post-OLT liver function, provided that the CIT is short. When macrovesicular steatosis exceeds 60%, there is consensus that such allografts be discarded because of unacceptably high rates of PNF. Utilizing grafts with moderate macrovesicular steatosis (30%–60%) is a challenging issue because, in this group, the incidence of PNF may reach 15% and the rate of delayed graft function approaches 35%. As with other ECD grafts, recipient matching should be based on the number and extent of recipient risk factors and the absence of other negative donor variables, such as advanced donor age and prolonged CIT, to minimize the negative impact on graft and patient outcomes.
CIT
The negative effect of CIT on organ function is intuitive; cold preservation increases anaerobic metabolism and cellular acidosis. Reperfusion after prolonged CIT in human and animal models is associated with inflammatory changes within the allograft, including sinusoidal cell damage, complement activation, small vessel hypercoagulability, and increased levels of interleukin-6 and -8. Prolonged CIT is an independent risk factor for the development of both delayed graft function and PNF. There is also an increased incidence of long-term biliary complications. In allografts from otherwise healthy donors younger than 60, the threshold for reduced allograft function secondary to prolonged CIT lies between 14 and 16 hours. Hepatic allografts from steatotic and older age donors (>60 years) are much more sensitive to preservation injury and demonstrate optimal function when CIT is under 8 hours.
Miscellaneous Donor Factors
The role of several other risk factors such as obesity, elevated liver function tests, hypotension, vasopressor use, nutrition, and length of stay in the intensive care unit is less clear and were not found to confer increased risk of graft failure in the most recent study. Certainly, donors at the extremes of these characteristics should be used cautiously.
Donation After Cardiac Death
The last few years has seen a considerable renewal of interest in donation after cardiac death (DCD) donors, formerly known as “non–heart-beating donors,” to increase the pool of available organs. The percentage of deceased donors in the United States that are DCD has grown from 1% in 1996 to 10% in 2007, and the number of liver transplants performed using DCD has surged from 0.5% in 1999 to 4.5% in 2008 ( Table 2 ), making it the most rapidly expanding component of the donor pool.
| Year of Transplant | Total Donors (n) | DCD (n) | DCD, % Donors of Total |
|---|---|---|---|
| 1999 | 4498 | 23 | 0.5 |
| 2000 | 4595 | 39 | 0.8 |
| 2001 | 4672 | 69 | 1.5 |
| 2002 | 4969 | 79 | 1.6 |
| 2003 | 5351 | 111 | 2.1 |
| 2004 | 5848 | 185 | 3.2 |
| 2005 | 6121 | 271 | 4.4 |
| 2006 | 6363 | 289 | 4.5 |
| 2007 | 6228 | 307 | 4.9 |
| 2008 | 6069 | 276 | 4.5 |
DCD has a fundamentally different recovery technique based on cardiopulmonary criteria rather than neurologic criteria for death. Organ retrieval from DCD can be “controlled” or “uncontrolled” based on the Maastricht classification. Controlled DCD undergo circulatory arrest after planned withdrawal of life support with the donor team ready in the operating room to start the procurement process. Uncontrolled DCD are donors who had an unplanned cardiac arrest with failed cardiopulmonary resuscitation, or are dead on arrival to the hospital. Organs from controlled DCD have a better chance of recovery compared with uncontrolled DCD.
The earliest reports of controlled DCD organ transplantation were published 16 years ago by the teams at the University of Pittsburgh and in Madison, Wisconsin. In 2000, Reich and colleagues provided the first single-center experience with DCD OLT showing outcomes comparable to those with DBD. In the decade that followed, several other transplant programs have published single-center experiences with controlled DCD showing comparable 1-year patient survivals with DBD OLT (range, 74%–92%). DCD 1-year liver graft survival, however, still has a tendency to be lower than with DBD livers (range, 61%–87%). Abt and colleagues were the first to report on the high incidence of biliary stricturing and/or bile cast syndrome occurring as a result of the biliary epithelium being extremely vulnerable to ischemia-reperfusion injury. This “ischemic cholangiopathy,” the Achilles heel of DCD OLT, has been reported in 9% to 50% of DCD recipients and usually requires frequent biliary manipulations to allow bile drainage, often resulting in allograft loss, retransplantation, or death. Despite many centers showing much poorer outcomes in their DCD versus DBD transplants, a few transplant programs, such as the Mayo, Jacksonville, and Albert Einstein, Philadelphia, groups have reported patient and graft survivals where the DCD cohorts were similar to their DBD counterparts. The Einstein patient and graft survival rates were 90% at 1 year and 85% at 2 years post-OLT. Ischemic cholangiopathy developed in 13% of recipients, causing graft failure in 10%. In the past few years, a number of multicenter pooled registry analyses have emerged highlighting specific criteria for safer use of DCD livers. Using younger DCD livers with shorter warm ischemia times and then CITs facilitate the best outcomes.
To date, there has been a lack of standardization with regard to many aspects of DCD, such as precise definitions of terminology, technique, use of vasodilatory drugs, antioxidants, preservation solutions, and the use of anticoagulation. In an effort to standardize procurement protocols and refine reporting of data, updated practice guidelines for organ procurement have been published by UNOS, the Institute of Medicine and the Society of Critical Care Medicine, and in 2009 the American Society of Transplant Surgeons issued recommendations on controlled DCD based on evidence and expert opinion. These American Society of Transplant Surgeons guidelines expound on all aspects of controlled DCD organ procurement including such issues as donor criteria, consent, withdrawal of support, operative technique, biliary concerns, ischemia times, and recipient considerations.
With ischemic cholangiopathy being the Achilles heel of DCD OLT, various authors have proposed recommendations on maneuvers to prevent biliary problems. These include performing an expeditious in situ biliary flush, considering arterial revascularization before or simultaneously with portal revascularization, use of a T-tube for easy access to the ducts postoperatively for stricture dilation and sludge removal to prevent bile casts, and using the bile acid ursodeoxycholic acid posttransplantation. Other suggestions to counter the specter of postoperative ischemic cholangiopathy have included using thrombolytic agents and anticoagulants and replacing the more viscous University of Wisconsin solution with histidine tryptophan ketoglutarate preservation solution. Several transplant groups, including the group at the University of Michigan, use postmortem extracorporeal membrane oxygenation to facilitate restoration of the flow of warm oxygenated blood to the intra-abdominal organs during the interval between death and organ procurement.
Exciting new research endeavors in organ preservation are in development, such as using ex vivo machine perfusion of the liver. A technique by Hong and colleagues proposed the novel concept of regulated hepatic reperfusion to modulate ischemia and reperfusion injury during organ revascularization. These and other innovative strategies potentially applicable to DCD are in early development and not yet ready for transfer from bench to bedside.
Predicting outcomes: translating ECD factors into clinical practice
Investigators have recognized that a combination of multiple factors in ECD donors can substantially increase the risk of graft failure. Thus, as a means to quantify the effect of multiple factors on graft function, several centers have reported a “risk score.” Cameron and colleagues from the University of California, Los Angeles, analyzed 1153 graft–recipient pairs to calculate a donor risk score. Using both univariate and multivariate analyses, significant factors predictive of graft and recipient survival were identified as extended criteria. This study also demonstrated that for each donor risk score, both older and urgent recipients had worse outcomes. In a large, retrospective review of the UNOS SRTR database from more than 20,000 donors, using Cox regression modeling, Feng and colleagues identified 5 donor characteristics that independently predicted a significantly increased risk of graft failure ( Table 3 ): Age, race, donor height, donor death (causes of death other than trauma, stroke or anoxia and DCD), and type of graft (partial/split graft). In addition, 2 transplant-related factors (CIT and sharing outside of the local donor service area) were also significantly associated with graft loss (see Table 3 ). All of these factors were used to generate a “donor risk index” (DRI), which is directly related to a predicted rate of graft survival. They found that when the DRI increased from the baseline risk index group (0.0–1.0) to the highest risk index group listed (>2.0), the frequency with which the graft would be discarded was significantly higher, rising from 3.1% to 12.5%. In a similar study from the United Kingdom, Dawwas and colleagues also identified 7 slightly different risk factors for graft loss (see Table 3 ). In the UK setting, the UK donor risk score outperformed the US donor risk score. However, both scores need to be externally validated in other populations before firm recommendations about their use can be made.
| Risk Factors | 5 Donor and 2 Transplant Risk Factors Identified in the United States 6 | 6 Donor and 1 Transplant Risk Factor Identified in the UK 71 | ||
|---|---|---|---|---|
| Risk Factor Reference Value | Increased Risk of Graft Failure, Relative Risk | Risk Factor Reference Value | Increased Risk of Graft Failure, Relative Risk | |
| Donor | ||||
| Age | <40 | 61–70: 1.53>70: 1.65 | Age | Increase by 1.05 per decade |
| Race | White | African American: 1.19 | White | Non-white: 2.17 |
| Size | Height | Increase by 1.07 per 10-cm decrease in height | NR | |
| Cause of donor death | Cause: trauma | Cardiovascular accident: 1.16 Other: 1.20 DCD: 1.51 | NR | |
| Type of graft | Full graft | Partial/split: 1.52 | Full graft | Reduced/split: 1.93 |
| BMI | NS | BMI | Increase by 1.01 per unit increase in BMI | |
| Graft appearance | No data | Normal | Suboptimal: 1.31 | |
| Diabetes | NS | No diabetes | Diabetes: 1.41 | |
| Transplant | ||||
| CIT | CIT | Increase of 1.01 per hour | CIT | Increase of 1.02 per hour |
| Sharing outside local area | Local area | Same region: 1.11National: 1.28 | NR | |
Transplantation of an ECD organ into a recipient who has a high MELD score may contribute to worsened graft and patient outcomes as compared with transplanting a lower DRI organ. However, high DRI organs still confer a survival benefit to high MELD patients as compared with waiting for lower risk organs. Markov models suggest that for patients with a MELD score of greater than 20, immediate transplantation even with grafts that carry a risk as high as 50% for primary graft failure is still associated with a survival benefit. Further reports based on real data and not on modeling approaches are eagerly awaited.
Donor-transmitted disease
Donors can transmit both infections and malignancies to OLT recipients during the transplant surgery (see Table 1 ). These also constitute examples of ECD.
HCV-Positive Liver Grafts
About 5% of all potential donors in the United States are positive for antibody to HCV and about half of these donors are HCV RNA positive by polymerase chain reaction. Utilizing HCV+ allografts for HCV− recipients or HCV+ recipients with an undetectable viral load is usually reserved for extreme circumstances. In contrast, utilization of HCV+ allografts among HCV+ recipients with active viral replication of genotype 1 or 4 is encouraged nowadays in the era of donor scarcity. In fact, current data clearly indicate no difference in HCV recurrence, or graft or patient survival when using HCV+ allografts. Provided the donor liver is not fibrotic, survival after transplant can be unimpaired. Recently Peek and Reddy reported preliminary data suggesting that HCV+ grafts may confer an advantage in the HCV+ recipient in terms of recurrence-free graft survival after OLT. Further long-term and multivariate analyses are needed to confirm these findings.
Hepatitis B Core Antibody-Positive Grafts
Donors with past exposure to hepatitis B infection can be used selectively in some recipients. Hepatitis B core antibody-positive (HbcAb+) donors carry a high risk of de novo HBV infection for the HBV naive recipients. However, for patients who are immune to HBV (previous disease or vaccination), it has been found to be safe to use these organs. In the era before prophylaxis, HBV transmission from an HbcAb+ donor to an HbcAb− recipient occurred at rates of 33%–78%, with subsequent accelerated graft loss. The use nucleos(t)ide analog therapy with or without hepatitis B immunoglobulin is now standard of care in the prevention of viral transmission from such HbcAb+ donors.
Transmission of Bacterial and Viral Infections
Bacterial infections in the donor do not represent by themselves a risk factor for liver graft failure. The risk of transmitting a bacterial infection in the case of bacteremia in the donor is low. Early fever and positive cultures in the recipient as well as the presence of yeast justify empiric therapy. Donors with documented bacterial meningitis do not preclude transplantation.
There have been scattered reports over the years of transmission of several types of viruses with transplantation, including HIV, West Nile virus (WNV), rabies, and HCV. As a result, UNOS and the Organ Procurement and Transplantation Network established the Disease Transmission Advisory Group in 2005 to monitor such transmissions, provide guidance in these cases, and analyze trends. This was later formalized as the ad hoc Disease Transmission Advisory Committee. There is a policy in the United States mandating the routine screening of potential donors for specific pathogens, including HIV, HBV, HCV, syphilis, cytomegalovirus, tuberculosis, and Epstein–Barr virus.
With regard to WNV infection, current recommendations include excluding potential donors with meningoencephalitic symptoms of undetermined etiology who live in regions of WNV activity, screening with nucleic acid testing (NAT) as close to the time of procurement as possible, and being suspicious when transplant recipients have postoperative fever and/or neurologic symptoms not otherwise explained. Serologic testing of the donor and all recipients from that donor should be performed (as well as lumbar puncture as indicated). At this time there is no specific treatment for WNV.
HIV+ Patients and “High-Risk” Donors
HIV can be readily transmitted through solid organ transplant, but because of improvements in screening protocols there were no reported cases of HIV transmission from 1994 to 2007. In 2007, however, 4 transplant recipients contracted HIV and HCV from a single donor who tested negative by serology. To date, there are no studies examining the transplantation of HIV+ organs into HIV+ recipients. Any potential donor regarded as “high risk” by the US Centers for Disease Control and Prevention standards should undergo NAT. The US Centers for Disease Control and Prevention has provided guidelines for the classification of donors as “high risk,” a metric indicating that an organ carries an increased risk of harboring an infectious disease. These donors may fall into any of 7 categories ( Box 1 ). Although NAT testing is the superior method of infection detection, it is expensive, time consuming, and difficult to perform relative to antibody testing. There is also a wide variation of NAT practices by organ procurement organizations. In general, NAT for HIV can detect infections 12 to 13 days before ELISA antibody assays. Care should be undertaken when transplanting a “high-risk” donor organ and full disclosure to recipient candidates is imperative. Highly active antiretroviral therapy should be initiated if transmission has occurred.
Related posts:
Emerging Therapies in Hepatitis C: Dawn of the Era of the Direct-Acting Antivirals
Noninvasive Tools to Assess Hepatic Fibrosis: Ready for Prime Time?
Long-Term Management of the Liver Transplant Recipient: Pearls for the Practicing Gastroenterologist
Noninvasive Tools to Assess Hepatic Fibrosis: Ready for Prime Time?
Liver Transplantation in the 21st Century: Expanding the Donor Options
Alcoholic Hepatitis: Prognostic Models and Treatment
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
