By Tae Wan Kim, John Roberts, Alan Strudler, and Sridhar Tayur.
In 2016, the Organ Procurement and Transplant Network (OPTN) and United Network for Organ Sharing (UNOS) posed an increasingly consequential question: Should a large liver always be split if medically safe?
During a split liver transplantation (SLT), a whole human liver is divided into two parts and offered to two patients—often a smaller adult and a child or sometimes two children. When SLT is underused, as is the case in the United States, more patients on transplant waiting lists die than would if SLT were employed. Yet this doesn’t mean that large livers should be split in every case and our paper explores how the right decision will oscillate between “split” and “do not split” depending on the dynamics of each transplant.
In the past, the above question was less relevant since transplanting a whole donor liver to a single patient was often safer than splitting one liver between two patients. The larger “half”—or 75 percent— may not adequately serve a large adult, and splitting raises logistical concerns, such as preparing two recipients in the same operating room, the time and expertise needed to split the liver, and the safe transportation of the parts of the liver. In recent years, however, SLT has enjoyed a success rate similar to using a whole liver, making the question necessary to consider.
Aside from the fact that with SLT two patients can be saved rather than one, we must first acknowledge the present inequalities inherent among patients on the liver transplant waiting list. There is a greater supply of large, adult organs, which means that without SLT, large adults are more likely to receive liver transplants than small adults—such as Asian women—and children. But if we were to suddenly flip that switch and offer SLT for each large liver that can be safely split into two, then smaller adults and children would benefit in greater numbers than large adults. Is saving two people always better than saving one if that single person has fewer opportunities to receive a transplant? We should instead approach organ allocation as a rationing decision and find balance between utility and equity across heterogeneous segments of patients. While it is straightforward that in terms of utility-maximization SLT is the right choice, it is indeed less clear why the same choice is also fair. We philosophically clarify under what circumstances SLT is fair. We provide two kinds of fairness-based philosophical arguments, drawing upon contemporary philosophical frameworks including John Rawls, Frances Kamm and Joseph Raz, in an accessible manner to the medical community.
As the ethics of SLT pose an interdisciplinary question, we created an interdisciplinary team to seek an answer. Our group consists of a transplant surgeon for a practical perspective, an ethicist to understand relevant norms of fairness, a legal scholar to provide a policy perspective, and an operations researcher to help improve the actual operation of the transplant system while accounting for medical, legal, and ethical considerations.
Our conclusion: despite its promising results and potential to save more lives, SLT is not always going to be the right approach. Yes, it might maximize utility by saving two lives rather than one, but it still presents ethical considerations because someone in need of a liver transplant must continue to wait and could potentially die. Instead, there must be an operational standard that accounts for the various dynamics at play.
Our recommendation: we propose a flexible decision support model that allows decision makers in the transplant community to oscillate between “split” and “do not split” based on the following criteria: the state of the transplant wait list and factors such as who is waiting, the severity of liver cirrhosis (or other disease), the waiting time until transplant, and the likelihood of death. As SLT becomes more widely practiced, and an increasing number of surgeons develop the skills required to perform SLT, we hope that this model is considered to account for both the predictable and less predictable dynamics implicit in liver transplantation while also acknowledging various fairness and efficiency metrics.
Authors: Kim, Tae Wan (1), Roberts, John (2), Strudler, Alan (3), Tayur, Sridhar (1).
Affiliations: (1) Carnegie Mellon University, (2) University of California San Francisco, (3) University of Pennsylvania
Competing interests: None declared.