Who Wins? The Dilemma of Organ Allocation for Liver Transplantation

Liver Bingo Card

The supply of deceased-donor organs in the U.S. is dwarfed by the demand for lifesaving transplants. Thus, transplant physicians are charged with the responsibility of efficiently allocating these scarce resources. However, when a deceased donor becomes available, not all organs are allocated in the same manner. The Lung Allocation Score prioritizes waitlisted candidates according to the survival benefit of receiving lung transplantation, while organ quality is matched to estimated post-transplant survival for those awaiting kidney transplantation (with the top 20 percent of kidneys being allocated first to patients with 20 percent highest predicted post-transplant survival). Both of these systems ensure efficient organ utilization by considering expected post-transplant outcomes. Conversely, the allocation process for liver transplantation operates according to the principle of the ‘sickest first,’ where the likelihood of receiving a liver is determined by the risk of waitlist mortality and estimated by the Model for End- Stage Liver Disease (MELD) score. Without considering post-transplant outcomes, this process is unable to maximize potential life-years gained, and the end result is an inefficient utilization of one of the scarcest resources in medicine.

In the last decade, efforts have been made to improve the efficiency and equity of the allocation system beyond conventional MELD-based prioritization. Organ-sharing policies, such as Share 35, mandate that organs be allocated outside of their procurement areas to candidates with higher MELDs if no local candidates with sufficient severity of illness are available. This recently led to a significant increase in the population of candidates with a MELD score of ≥35 (i.e., estimated three-month mortality of >50 percent) being waitlisted and transplanted.1 Unfortunately, Share 35 has not produced the much-anticipated improvement in waitlist mortality as projected, and the downstream effect on those with lower MELD scores has yet to be fully determined. However, subsequent sharing across wider geographical areas and increased candidate severity of illness have caused the costs and resource utilization of transplants facilitated by this policy to skyrocket. Furthermore, its impact on post-transplant patient morbidity and rehospitalization remains unknown.

The policy changes over the last decade have taught us that the dilemma of organ allocation for liver transplantation has yet to be solved.

While Share 35 has attempted to facilitate liver transplantation for the sickest candidates, particular attention has also been paid to candidates with hepatocellular carcinoma (HCC), who are arguably the waitlist’s least sick. These candidates receive standardized prioritization via artificial inflation of their MELD scores, such that they are often at much earlier stages of cirrhosis when transplanted. Though it was initially intended to provide prompt transplantation and avoid the risk of waitlist dropout due to metastatic disease, this policy has led to significant disparity between HCC and non-HCC candidates. And despite numerous revisions (the latest of which occurred less than a year ago) the allocation process for HCC candidates remains imperfect. Not only are non-HCC candidates less likely to be wait-listed overall, but they must also develop far greater severity of illness to reach the same MELD score as HCC candidates.2 Moreover, with current advanced diagnostic techniques and readily available loco-regional therapies, the potential for complete eradication of early HCCs is high, such that the survival benefit of liver transplantation in 80 percent of these patients is negligible at best.3

The policy changes over the last decade have taught us that the dilemma of organ allocation for liver transplantation has yet to be solved. The field of liver transplantation is often described as a zero-sum game, where the allocation of an organ to one candidate inherently denies it to another. Therefore, how to most effectively and equitably allocate this scarce resource remains a source of contentious debate: Are the MELD ≥35 patients too sick while those with HCC not sick enough? Perhaps the primary issue at hand is in fact the measure by which candidates are prioritized. Incorporating expected post-transplant outcomes into the equation has the potential to level the playing field and maximize the utility of the organ-allocation system. It may come to no surprise that most treatment decision-making in medicine is influenced by an assessment of anticipated benefit. For example, a cardiothoracic surgeon would never recommend complex bypass surgery to a patient with advanced coronary disease unless clear improvements in health and functional status were expected. The risk-benefit determination with respect to liver transplantation should be no different.

Thus, as the liver-transplantation community looks toward the future state of organ allocation, there must be universal acceptance that our prior policy modifications have been insufficient. We must not only learn from our colleagues in other transplant subspecialties, but also think outside the box. A challenging first step is to clearly define what constitutes an acceptable post-transplant outcome for our liver-transplant recipients.

Drs. Bittermann and Goldberg have no conflicts to disclose.


1. Edwards, E.B., Harper, A.M., Hirose, R., Mulligan, D.C. The impact of broader regional sharing of livers: 2 year results of “Share 35.” Liver Transplantation. 2016; 22(4): 399-409.
2. Goldberg, D., French, B., Newcomb, C., Liu, Q., Sahota, G., Wallace, A.E., Forde, K.A., Lewis, J.D., Halpern, S.D. Patients with hepatocellular carcinoma have the highest rates of wait-listing for liver transplantation among patients with end-stage liver disease. Clin Gastroenterol Hepatol. 2016 Nov; 14(11): 1638-1646.e2.
3. Berry, K., Ioannou, G.N. Comparison of the liver transplant-related survival benefit in patients with and without hepatocellular carcinoma in the United States. Gastroenterology. 2015; 149(3): 669-80.

Join the discussion

Your email address will not be published. Required fields are marked *