The Current State of Liver Transplantation in the United States

Introduction

Every year, the Organ Procurement and Transplantation Network (OPTN) and the Scientific Registry for Transplant Recipients (SRTR) release data about donor and transplant recipient activity in the United States. The 2014 data report spans over a period of 10 years and is based on data collected between 2003 and 2014 1. Liver transplantation represents the second most commonly transplanted organ in the United States. While pediatric liver transplantation trends show a steady decrease of the mortality on the waiting list and improved transplant outcomes, adult liver transplantation continues to be challenged by declining organ quality and increased mortality rates on the waiting list.

The aim of this article is to summarize and highlight what we believe are the most important findings of the 2014 report 1. Emphasis is placed on data relating to waitlist dynamics, organ utilization rates, the use of suboptimal grafts, and optimization of the existing limited donor pool with the intent to recognize trends and areas that could be the object of policy change, intervention, and innovation.

Waitlist Trends

According to the recent OPTN report, as of June 2014, nearly 72 000 adults were living with a functioning liver allograft. In 2014, 6729 liver transplants were performed in adults, 6449 of which were from deceased donors, and 280 from living donors. As of the end of December 2014, 14 632 candidates were registered on the waiting list, of whom 83% (12 204) are active and 17% (2428) inactive. Over the last decade, liver transplant dynamics have been characterized by an annual addition of approximately 10 500 new patients (Figure 1), with a plateau in the number of transplants performed (around 6800/year, Figure 1). The transplants mostly consist of whole liver deceased donor grafts (approximately 95%), with roughly 15 000 patients listed at the end of each year (Figure 1). The annual number of active candidates waiting for a liver transplant has been stable over the past decade (−0.8%) 1. However, the number of patients who entered the list and died without transplant or were removed because of being too sick for transplant has increased 30% over the last decade (n = 2400 in 2004 vs. 3111 in 2014, net increase 711 patients) 2. Since 2009, mortality rates have increased each year from 11.1 per 100 waitlist years to 12.3 in 2014. In 2014, 1821 patients died while waiting for a transplant, while an additional 1290 were removed from the list because they had become too sick to undergo a transplant 1. This increase in overall waitlist mortality could partially be attributed to the higher proportions of waitlist registrants at higher medical urgency status (Model for End-Stage Liver Disease [MELD] >30), lower proportions of candidates at a lower risk of death (MELD <15), and geographic disparity in organ availability (Figure 2).

Candidate Selection

Organ Utilization Trends

The mean number of all organs transplanted per donor was 3.04 in 2013, a downward trend that has been consistently observed since the highest number of 3.15 was seen in 2003. In 2014, 15% of donors nationally were DCD 1. Additionally, 8594 donors were considered eligible for liver procurement in 2014. Of these, consent was not requested or obtained in 0.9% of cases (72 donors), and only 6383 liver allografts (74.3%) were transplanted. The remaining 25.7% of the liver grafts were divided as follows: recovered for research or hepatocyte isolation 4.2% (357 livers); recovered for transplantation but not transplanted 7.9% (681 livers); organ not recovered 12.8% (1101 livers). The numbers of these “untransplantable” livers (2209) has remained relatively stable over the past decade. The most common reason for not transplanting recovered livers was biopsy findings, whereas the most common reason for not procuring a liver was, “ruled out after evaluation in the operating room” (n = 303), followed by “other” (n = 131), warm “ischemic time too long” (n = 42), “anatomical abnormalities” (n = 68), and “diseased organ” (n = 45). The most common reasons for not recovering an organ were “ruled out after evaluation in the OR” (n =248), “other” (n = 174), “poor organ function” (n = 142), organ refused by all centers (n = 101), and “time constraints” (n = 68) 1.

In 2013 and 2014, respectively, 309 and 363 liver transplants were from DCD 46. This number has remained relatively steady since the peak observed in 2007 47. In 2014, 228 adult living donor liver transplants (LDLTs) were performed, accounting for only 3.4% of all the transplants (6729). This is an overall decrease of 13.6% over a period of 10 years (273 adult LDLTs in 2004), while the number of deceased donor transplants increased by 5.4% during this period 2. However, in the past 3 years we have observed a steady and encouraging increase in adult LDLTs, with an overall increase of about 21% since 2011, when 188 adult LDLTs were performed 2.

These abovementioned trends offer an invaluable opportunity to identify areas in which to focus future research.

Over the past few years, the total number of organ donors has slowly increased 51; nonetheless, the number of liver donors has not proportionally grown due to a net increase in the number of discarded livers. In 2004, 14.8% of donated livers were not used 52, while this percentage rose to 25.7% in 2014. This dramatic increase is multifactorial. First and foremost, this is a direct consequence of the increasing donor age and prevalence in the general population of obesity, diabetes, and hypertension 52. From 2001 to 2010, the mean donor age increased by 9.3 years (from 34.6 to 43.3), the prevalence of obesity (donor BMI >30) doubled (from 15% to 30.3%), and the prevalence of diabetes and hypertension increased from 3.2% to 12% and from 22.6% to 37.2%, respectively 52. Secondly, as Medicare scrutiny for transplant program outcomes increased, transplant centers became more risk adverse and therefore more reluctant to use expanded criteria grafts 53. Orman et al 16, using a sophisticated statistical forecasting model (Discrete Event Simulation), projected a decrease in the overall liver utilization rate from 77.9% in 2010 to 43.6% in 2030. According to their modeling, 43.7% of the donors will be older than 50 years, 45.7% will be diabetic, and 58.2% will have a BMI >30. These changes in donor quality would result in 2230 fewer liver transplants per year by 2030.

Orman et al 16 modeled the implementation or strategies to counteract this trend and reported that only two significantly affected (>5%) the organ utilization rate: converting 90% of DCD donors to donation after brain death (DBD) and the introduction of a disruptive technology such as ex vivo organ preservation. In spite of the several limitations intrinsic to the statistical forecast model used, this study substantiated the day-to-day perception that each liver transplant professional has about the progressively declining organ quality. Another study from Parikh et al found similar results. They also projected that although the donor population may increase over the next decade, general population growth will exceed donor growth, therefore exacerbating even further donor shortages and mortality on the waiting list.

Over the past decade, expansion of DCD transplantation has remained a controversial issue. While DCD donors represent a precious source of livers and kidneys, there is a concern that the decline in DBD donation may be one of the unexpected consequences of the Organ Donation Breakthrough Collaborative 54. In spite of single center acceptable results 55, 56, DCD liver transplantation remains associated with significantly lower graft and patient outcomes, as shown by the OPTN report. Since such an unfavorable impact is not present in kidney transplant 57, 58, a conundrum has been created.

DCD kidney transplantation has been increasing over the past 10 years; however, every effort should be made in the future to ensure that DCD donation overall does not negatively impact the quality of liver grafts. This could happen if withdrawal of support occurs at an early stage in those donors that are likely to progress to brain death 59. As a consequence the end-stage liver disease patient will be adversely affected while the end-stage renal disease patients will see no negative effect. If Orman and colleagues’ projections hold true, the proportion of DCD donors could reach 36% by 2030 16.

For this reason, in the past decade researchers have placed increasing effort to investigate mechanism of injury during preservation of marginal liver grafts. There is growing evidence that most of the damage occurring during preservation is a direct consequence of static cold storage (SCS) 60, 61, the current (and for the past 4 decades) standard of care in organ preservation. Several groups in the United States, and especially Europe, recently introduced ex vivo machine perfusion in human trials with very promising results 62-65. Data suggest that perfusing livers ex vivo at normothermic temperature, though more challenging and expensive compared to SCS and hypothermic machine perfusion, carries potential key advantages. The most important one is the ability to use the device as an assessment tool for the graft. At physiologic temperature, in fact, a functioning organ is fully metabolically active, as demonstrated by active bile production 66 and the ability to clear lactate 67. This key potential predictive/rehabilitative tool is what has been missing in the liver transplantation field as organ quality has declined. If normothermic machine preservation continues to advance as it has in the past few years, very soon we will be able to test and safely use many of the livers currently discarded. Other advantages of ex vivo perfusion technologies include superior preservation 68, the ability to pharmacologically support the organ during perfusion, and the potential to enhance the physiologic cell repair that follows an ischemic injury 69. The latter would have a central role in the restoration, optimization, and ultimately increased use of DCD grafts.

Liver Transplant Outcome Trends

In 2014, as compared to 2004, patients approached liver transplantation in much worse clinical condition, as shown by both a higher proportion of patients with a MELD score >30 (40.4% vs. 20.4%, respectively), and those hospitalized at the time of transplant (35.7% vs. 28.7%, respectively). Interestingly, thus far, despite the escalation of the severity of liver transplant recipients, graft survival continued to improve (Figure 4). In addition, the incidence of acute rejection continued to decline; this might be related to the increasing use of induction agents, although various other factors may also account for this. The overall 5-year graft survival was 70.2% for patients who underwent a liver transplant in 2009 1.

This favorable graft survival, despite higher proportion of patients with high MELD, should be critically evaluated as this could be partially related to transplanting patients with HCC MELD exceptions. In 2014, HCC MELD exceptions were utilized by 25% of transplanted patients, with a portion of them with exception MELD points of ≥35 despite a lower biochemical MELD score. Also, the majority of recipients in 2014 (>50%) had MELD scores of 15–29, which might account for why the overall outcomes are still favorable 1.

However, particular recipient and donor characteristics have been associated with poor outcome. Early survival (1 year) was poorest for status 1A patients, retransplants, and recipients of DCD livers. Intermediate (1–3 years) and long-term survival (>3 years) was poorest for recipients aged ≥65 years of age, or those with HCV infection or malignancy 1 (Figure 4).

Although it is expected that the new antiviral agents will decrease HCV-related graft and patient loss, donor factors (as discussed earlier) and recipient characteristics are likely to worsen. In fact, the liver transplant recipient trends over the last decade demonstrated an alarming change with increase in recipients who are ≥65 years of age 38, as well as a dramatic increased NASH cirrhosis 3, 4, 15. With the current MELD allocation system with the Share 35 and capping of HCC MELD policy it is expected that older, obese patients with complex medical problems requiring significant ICU support and who are physiologically sicker will be prioritized to receive a liver transplant. This highlights the importance of both defining and predicting futility for the increasingly sick patient population.

Allocation

Despite the changes made in the liver transplant allocation system, it is still true that where candidates are listed affects how likely they are to receive a transplant, and therefore their likelihood of death. Multiple proposals to address how to optimize liver distribution and enhance recipient outcomes have been suggested 81, 82. The prediction of their effectiveness, logistical aspects, financial impacts, and influence on patient outcome is currently being studied 83-85. It is expected that results of further studies involving financial implication, true practical applications, and logistic feasibility that includes input from all stakeholders will likely change the current map of organ allocation to reduce geographic disparities.

Candidates receiving exception MELD points for HCC experience higher transplant rates, but have a lower mortality risk than candidates without HCC, and the transplant rates for candidates with HCC were more than twice the rates for candidates without HCC. Recently, revisions to the OPTN liver allocation policy were implemented on October 8, 2015, modifying the maximum value of exception scores for candidates with HCC (capped at 34), and the timing of exception scores assigned to HCC candidates (A 6-month run-in period prior to receiving a score of 28). The effects of this recent change remain to be determined.

Share 35 policy

In June 2013, The Share 35 policy was implemented to help address waitlist mortality high MELD patients whose mortality was comparable to status 1A patients. Using the similar status 1A system, this has had a favorable impact to patients with MELD >35. Their median waiting time halved (from 18 days in 2012 to 9 days in 2014); these patients constituted 26.1% of all liver transplants performed in 2014, resulting in a decrease in waitlist mortality among these candidates from 366 per 100 waitlist years in 2012 to 315 per 100 waitlist years in 2014 (−13.9%) (Figure 5). However, in some regions this has been associated with consequences of increased travel costs and coordination efforts, but may be balanced by reduced hospital costs. The latter effect may be due to shortening wait time for hospitalized patients with high true laboratory MELDs. Further investigation of this is necessary to understand the true total impact of these changes to include changes in program behavior.

Discussion

The 2014 OPTN/SRTR report has offered throughout the years an important snapshot of the transplantation activity in the United States. The recently released report on adult liver transplantation shows very clearly that every year the number of patients in need of a liver transplant progressively exceeded the number of transplants performed. This resulted in an increase in the mortality on the waiting list, and increasingly worsening conditions at the time of transplant, with older and more morbid patients approaching surgery. At the same time, the quality of organ donors continued to deteriorate, as shown by the decline of the number of organs transplanted per donor. In spite of the declining recipient and donor quality, liver transplant outcomes continued to improve over time, with 5-year graft survival rates reaching 70.2% 1. With the most recent changes now being implemented for listing and capping of MELD points for HCC and efforts to standardize exception points across the country, it is too early to tell where disparities will exist in the future. Geographic disparity and access to organs will continue to be an important issue.

While UNOS continues to address geographic disparities, there should be considerable focus in the next decade to promote innovation in the area of donor and graft optimization (donor intervention, new preservation solutions, and machine perfusion).

Furthermore, with the approval of human immunodeficiency virus–positive (HIV⁺) donors for HIV recipients and the cure of HCV, the landscape for the definition of high risk as defined by the Public Health Service will change. An uncharted territory that likely will appear and increase is organs offered (liver and extrahepatic organs) from patients who had SVR after being treated with DAAs with an undetectable HCV via the nucleic acid test, and their allocation and utilization in HCV-naive critical patients. How these new advances will affect who is offered and receives these organs will require further thought, discussion, and ethical consideration. Most importantly, innovation needs to be encouraged to overcome the persistent shortage of organs, but with a provision to exclude innovative practices from Centers for Medicare and Medicaid Services reprisal.

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.