Deceased Donor Liver Allocation: Cutting the Gordian Knot
Article first published online: 26 JUL 2013
© Copyright 2013 The American Society of Transplantation and the American Society of Transplant Surgeons
American Journal of Transplantation
Volume 13, Issue 8, pages 1949–1950, August 2013
How to Cite
Yeh, H. and Hunsicker, L. (2013), Deceased Donor Liver Allocation: Cutting the Gordian Knot. American Journal of Transplantation, 13: 1949–1950. doi: 10.1111/ajt.12302
- Issue published online: 26 JUL 2013
- Article first published online: 26 JUL 2013
- Manuscript Accepted: 13 APR 2013
- Manuscript Revised: 11 APR 2013
- Manuscript Received: 18 MAR 2013
The United States is organized into 59 Donor Service Areas (DSAs) within 11 Organ Procurement and Transplantation Network (OPTN) Regions, which serve as the primary and secondary distribution units of deceased donor livers for transplantation. The ratios of livers to wait list candidates differ among these units, so candidates have substantially different outcomes depending on their place of listing. This persists despite Department of Health and Human Services policy, established a decade ago, that geography should not play a role in access to transplantation. Thus far, proposed changes to the liver distribution algorithm that maintain DSA and OPTN Regional boundaries have not been shown to meaningfully reduce geographic disparity or to dramatically improve survival. As a result, there has not been a consensus to move forward and unravel the tangle of financial costs, logistic complications and political repercussions of moving organs outside their local communities.
In this issue of the American Journal of Transplantation, Gentry and her colleagues  take a new approach. Using methods developed to optimize school and political districts, they describe two novel maps, grouping DSAs into 11 new regions that simultaneously minimize geographic disparities, wait list death risk and median transport distance. Because of clustering of high liver access DSAs within some OPTN Regions and low access DSAs within others, full regional sharing with the current OPTN Region map would paradoxically increase the disparity in liver transplant access and would only reduce total and wait list mortality by 47 and 39 deaths per year. However, by combining high organ availability DSAs with neighboring DSAs of lower availability into new regions, full regional sharing would achieve more equality, while also reducing total and wait list mortality by 66–71 and 59–61 deaths per year.
There has been concern that increased sharing would increase transportation times. Gentry's analyses predict that full regional sharing with their redesigned Regions would only increase median transport distances by 74 miles. This will have minimal impact on cold ischemia times and transplant outcomes. There may be increased medical costs  associated with performing a larger fraction of high MELD transplants, but this should be partially balanced by decreased costs of keeping high MELD patients on the wait list, by transplanting them sooner . In aggregate, Gentry's analyses suggest that redesigning the OPTN Regions could substantially improve liver transplant equity and outcomes with minimal adverse impact. As they point out, the importance of their study is not in the specific maps presented, but in the demonstration of techniques to design allocation units and evaluate them. It would be interesting to perform similar analyses changing not only the shape, but also the number of regions, by optimizing their number and size by either population or area, or to optimize distance based allocation.
However, it is also important to understand why the balance between organ supply and demand varies so much from one DSA to another. Two recent studies [4, 5] indicate that it is variation in the rates of new liver transplant listings, and not differences in organ donation rates among DSAs, that is the major determinant of organ availability. The factors underlying differences in new listing rates are not yet delineated, and may include regional variations in disease incidence, health care access, and both provider and patient awareness of transplantation as a treatment option, as well as insurance company policies that refer patients to “centers of excellence” concentrated in high competition DSAs. These issues have to be understood so that improvement in wait list outcomes is not inadvertently achieved by decreasing the number of eligible patients that get listed.
Further, observations of current practice patterns indicate that increased competition is associated with more aggressive recipient and organ acceptance . This is likely due both to inherently aggressive centers congregating in what then turns into a highly competitive geographic locale and to the response of other centers in the struggle for market share. Combining high and low organ availability DSAs into single distribution units will influence the dynamic, providing more organs for low organ availability areas, but also converting low competition areas into high competition areas. To maintain transplant volume, previously low competition centers may need to change practice patterns or risk closing down. Thus redistricting may indirectly contribute to standardizing transplant practices across the country and increasing access to transplantation in previously low competition areas. Alternatively, if there are center closures, access to transplantation could decrease in some areas, as it becomes difficult for patients to travel to far away centers for evaluation, listing, and transplantation. Since redistricting is likely to change both supply and demand patterns, any changes to organ distribution policies must be coupled with close monitoring, and the design of organ distribution units should be periodically re-evaluated to reflect these changes and the changing values of the transplant community.
The authors of this manuscript have no conflicts of interest to declare according to the American Journal of Transplantation.
- 5Geographic inequity in access to livers for transplantation. Transplantation 2011; 91: 479–486., , ,