The probability that a waiting liver transplant candidate will receive a deceased donor liver offer is defined by both allocation and distribution policy. Allocation policy sets the ranking rules for a given set of waiting candidates, and distribution policy determines the group of waiting candidates over which the allocation rules will be applied. In 1999, the Organ Procurement and Transplantation Network (OPTN) required that donor organs be shared across entire regions when a patient meets the most urgent, status 1 criteria. This policy resulted in a significant reduction in wait list deaths for status 1 patients without an adverse effect on other waiting candidates or posttransplantation survival.1 Thus, the combination of high priority and a shift to wider availability of donor organs through mandating regional sharing for these patients significantly improved the waiting list mortality for this high-urgency group.
Recent analyses have looked at some categories within the status 1 designation and found that the mortality risks are not homogeneous and, particularly for patients with non-acetaminophen acute liver failure, mortality risks are better defined by their Model for End-Stage Liver Disease (MELD) score than by other parameters.2
In 2005, the OPTN further refined the status 1 designation to better address pediatric candidates with severe chronic liver disease and defined more stringent criteria by which status 1 patients were categorized. In this policy revision, all patients with acute liver failure, patients with early primary graft failure, and patients with early hepatic artery thrombosis meeting the strict criteria were designated status 1, with pediatric patients meeting severe chronic liver disease criteria categorized as status 1B (http://optn.transplant.hrsa.gov/PoliciesandBylaws2/policies/pdfs/policy_8.pdf for the complete policy). A formal analysis of the effect of this policy on pediatric and adult candidates has not been published to date. This revision of allocation policy, however, illustrates the fact that allocation of donor livers to status 1 candidates does impact patients waiting with chronic liver disease for whom MELD score determined the allocation sequence.
In this issue of HEPATOLOGY, Sharma et al., in another of a series of papers on MELD from the Arbor Research Collaborative, used the OPTN database to assess how patients with chronic liver disease prioritized by having similar waiting list and posttransplantation survival probabilities compare with patients listed meeting the 1A criteria.3 The authors found that adults registered as status 1A had a lower wait list mortality risk than patients registered with MELD scores of greater than 40. Moreover, there was no difference in posttransplantation survival among the highest MELD categories and the status 1 patients. They argue, based on their results, that patients with MELD scores greater than 40 should receive the highest priority—higher than the status 1A patients. In contrast to fears that giving patients with MELD scores greater than 40 additional priority would result in significantly poorer posttransplantation outcome, the authors point out that the posttransplantation survival for these extremely ill patients is comparable to status 1 patients while acknowledging that patients undergoing transplantation at these extreme MELD scores are highly selected candidates.
There are several important caveats regarding this analysis that should be understood before any implication for policy change should be considered. First, the authors excluded patients who achieve status 1A candidacy by virtue of needing retransplantation. Previous studies have confirmed that these patients do not have the same mortality risks as patients with de novo acute liver failure. However, they are conferred 1A status and thus compete equally for donor livers with patients having a 1A status secondary to acute liver failure. This policy thus could theoretically increase the waiting time for both patients with acute fulminant hepatitis having a 1A status and for patients with chronic liver disease. Second, these authors found that the MELD score predicted wait list mortality for patients with acetaminophen toxicity in contrast to other studies where the MELD score did not correlate with wait list mortality for status 1 patients registered with acetaminophen toxicity. The authors do not reconcile the differences in these two studies. Third, this analysis dated back to 2001, before the advent of the MELD allocation policy, which began in February 2002. The study also does not take into account the policy revisions that redefine status 1 to 1A and 1B categories in 2005, which occurred during the study period. Furthermore, the authors did not assess center effect, which has been shown to have a substantial effect on both pre- and posttransplantation survival, particularly in patients with high MELD scores.4 Finally, and most importantly, broader sharing for status 1A patients relative to patients prioritized by MELD may have a more profound effect on reduction of wait list mortality than any change to prior rules for prioritization for organ allocation. In contrast to the author's conclusion, one might interpret these results to support a change to distribution policies so that deceased donor livers are offered to patients with the highest MELD score on a regional basis similar to the status 1 patients at the present time rather than changing the allocation sequence that would put patients with a MELD score greater than 40 ahead of status 1 patients. In fact, such a distribution policy change has been developed and put to public comment recently by the OPTN.5
In conclusion, any analysis of the liver transplant waiting list encompassing a significantly long study period is confounded by changes in policy during the period of analysis. In addition, advances in the management of acute liver failure have occurred that may impact survival. The situation is further complicated when both allocation and distribution rules have changed over time and are not taken into account. Nonetheless, it seems intuitive that offering higher priority through an allocation policy change or broader access through a distribution policy change will improve results for a selected group of the most ill-waiting candidates. Because differences in center expertise are likely to be more profound for the sickest patients, it is important that center effect be accounted for in any such analysis. Finally, what is not addressed here is what will happen to other candidates with chronic liver disease on the list if either approach is taken. Further data are required to assess the potential impact of an allocation policy change that is advocated here.