Using Marginal Grafts for Liver Transplantation: The Balance of Risk

This article is referred to by:

Current Status of Liver Transplantation Using Marginal Grafts

In this issue of Journal of Investigative Surgery, Badawy and colleagues contributed an extensively written article on with the use of marginal grafts in liver transplantation and the authors should be congratulated [1]. This timely review summarized the use of these grafts in living donor liver transplantation (LDLT) and deceased donor liver transplantation (DDLT) and assessed the risk of elderly donors, steatotic grafts, split liver grafts, donation after circulatory death (DCD) grafts, prolonged ischemia periods and grafts from donor carrying risk of viral hepatitis or occult malignancies. The article is timely and relevant in the current context because the majority of the organ supply emanates from marginal donors; yet further expansion in donor pool is possible. The key issue in the use of marginal liver grafts to achieve the “best outcomes from less than best quality liver grafts” and in order to achieve this goal, not only the graft and recipient matching, but also incorporation of host of other knowledge and experience is of paramount importance.

There are many scoring systems that have been developed to predict the impact of marginal grafts in DDLT; albeit the majority of these could be applied in retrospect, the most widely risk score is the Donor Risk Index developed by Feng et al. in 2006, using donor age and height, cause of death, race, split/DCD grafts, regional/national sharing and the duration of cold ischemia time [2]. This model has proven to be effective in predicting the risk for long-term graft loss. However, the spectrum of marginal grafts has shifted dramatically over the last years. The options for liver transplantation of grafts infected with viral infections has improved significantly with the prophylactic anti-globulin therapies and nucleoside analogs for Hepatitis B and the promising results of the direct acting antivirals for Hepatitis C. As experienced in the United Kingdom and many other countries over the past decade, the elderly donor pool is expanding and the donor pool is largely contributed by the popularity of DCD donation practice. More importantly, we are forced to use more steatotic grafts due to the obesity epidemic. However, macrosteatosis >30% increases the risk for primary nonfunction and other graft-related problems, because these grafts are more prone to severe ischemia/reperfusion injury. In addition, we have previously shown that marginal grafts in general suffer more from the additional warm ischemia prior to implantation than standard grafts, which further aggravates this injury [3]. Therefore, surgical strategies are also important in predicting the outcomes following the utilization of marginal graft, and one of the key strategies is to “modify” the surgical techniques and approach to minimize further injury during the transplant operation.

The outcomes after liver transplantation do, however, not only depend on quality of the graft. Recipient age, Model for End Stage Liver Disease (MELD)-score and comorbidities and surgical issues during the transplant procedure also have an impact on recipient outcomes. This Balance of Risk concept follows the principle that a patient with limited risk can sustain the injury of marginal grafts, i.e. a young patient with a low MELD-score and hepatocellular carcinoma that needs urgent transplantation for oncological reasons might benefit from a DCD graft that is earlier available than a Donation after Brain Death (DBD) graft. On the contrary, a critically ill patient requiring organ support and renal replacement therapy can only receive the best DBD graft, to minimize the peri-transplant risk. Therefore, recent prediction models for graft loss or mortality after liver transplantation assess the cumulative risk of donor, graft and recipient factors. Examples include the BAR Score from Zürich, the SOFT Score from Columbia University and the UK DCD Risk Score for DCD grafts, developed by our team in Birmingham [4–6]. The ideal organ allocation system takes in to account all these aspects but such systems do not exist. In this context, a new allocation scheme has been introduced in the United Kingdom earlier this year called “Transplant Benefit Score” that combines a host of recipient and donor criteria and aims to match grafts and recipients in such a way that the graft is allocated to a recipient who will have the most benefit [7].

The balance of risk between donor and recipient in LDLT yields a different spectrum. These grafts are generally of high quality, although steatotic grafts are likely to be used more due to the increased incidence of steatosis in the general population. The added benefit in the LDLT setting is the time factor which allows even an overweight donor to be optimized with lifestyle modification, and diet regulation so that the degree of steatosis present in the potential graft could be downgraded. The main concern in LDLT is the small-for-size syndrome, especially in patients receiving a graft with a graft-recipient weight ratio of less than 0.8. Due to the relatively high quality of the living donor graft and the only small number of donors to choose from for each recipient, there are only a few studies investigating the donor and graft risk in LDLT. A comparison study of the risk scores used in DDLT in 249 LDLT recipients showed that only the BAR Score provided acceptable performance in predicting the post-transplant survival [8]. The team from the University of Pennsylvania developed a score specifically developed for LDLT with seven donors and recipient factors, but the discriminative value of this score was relatively low, with only an area under the curve of 0.62 for long-term graft survival [9].

Risk assessment using the balance of risk can be used as a tool to create the best donor and recipient combination, but also to identify organs that could benefit from graft improvement techniques. Machine perfusion in liver transplantation has developed significantly over the last decade and we are still on the search for the best technique to preserve and repair organs before they are implanted in the recipient. The concepts that are currently investigated include but are not limited to organ resuscitation both in-situ and ex-situ, organ reconditioning and better preservation. Normothermic machine perfusion has shown to be effective in decreasing the risk of ischemia/reperfusion injury and early graft function in a recent randomized controlled trial and this technique could be used preserve organs for a period up to 24 hours, with real time assessment of graft viability [10]. The other technology that has been investigated is hypothermic oxygenated machine perfusion has the ability to improve organ quality, especially in high risk grafts, such as DCD grafts [11,12]. The technological developments add a new dimension to the marginal grafts and most of these are associated with real time assessment of surrogate markers of graft quality and function prior to transplantation, and these approaches potentially minimize the immediate risks that poses to the recipient from primary non-function. In this aspect, the evidence is being created and we have set the precedence by successfully transplanting discarded organs after viability testing during end-ischemic normothermic machine perfusion [13].

In conclusion, Badawy and colleagues have highlighted the impact of marginal grafts in liver transplantation. The quality of deceased and living donor grafts will continue to change in the future and we will have to adapt our donor selection and recipient matching depending on the demand and availability of organs, and latest technology may play a bigger role in the future.

DECLARATION Of INTEREST

The authors have no competing interests to declare.