With the long waitlist, the current state of deceased and living donation has not met the demand for transplant kidneys. Innovative variations of kidney paired donation (KPD) are continually being tested to increase the availability of all donors. KPD has already expanded living donation by matching patients who have a healthy but immunologically incompatible donor with a different donor from a pair in a similar situation.
In its most basic form, KPD is a 2-way donor swap between reciprocal A/B and B/A blood-type incompatible pairs (Figure 1A). More complex 3-way exchanges and domino exchanges, in which 1 of the donors is nondirected and 1 of the recipients is from the deceased donor waitlist, have since been developed1 (Figures 1B and C). Surgeries within these KPD variations were originally performed concurrently to ensure that donors could not withdraw after their paired recipient received a kidney transplant.
In addition to facilitating exchanges between incompatible pairs, KPD programs now encourage compatible pairs to enter the match, allowing the recipient to receive a kidney from a better-matched or younger donor. Adding compatible pairs to KPD also increases the number of matches for incompatible pairs, thereby improving the matching capability of the program as a whole.2
Further innovations within kidney exchange programs include donation strategies with kidney chains composed of a series of donors and recipients whose operations are not concurrent. For example, nonsimultaneous extended altruistic donor (NEAD) chains are started by a nondirected donor (NDD) who provides a living donor kidney to a patient with an incompatible living donor who then subsequently donates to propagate the chain3 (Figure 1D). Most chains end with donation to the deceased donor waitlist.3 These chains are ideally set up so that no donor donates a kidney before their intended recipient receiving a kidney.
KPD now makes up about 12% of all live donations in the United States.4 To further expand KPD, 2 new strategies have been described recently. First, advanced donation programs (ADPs), which alter the traditional chain donation sequence, enable individuals to donate before their intended recipients receive, or are even matched for, a kidney transplant (Figure 2A). A second a strategy is to increase the number of chain-initiating kidneys (CIKs) by using deceased donors rather than only NDDs to trigger kidney transplant chains (Figure 2B).
This article reviews the ethical and logistical complexities of these 2 new strategies. We conclude that innovations in KPD chains that increase the number of transplant kidneys should be encouraged but be implemented thoughtfully with consideration of challenges that we describe.
Underlying Ethical Principles in Allocation and Distribution of Transplant Kidneys
The 3 ethical principles that govern organ allocation are justice, utility, and respect for persons.5 The principle of justice requires that equals are treated equally and unequals unequally.6 To provide practical guidance, we have to turn to material principles to specify what it means to be equal and unequal in organ allocation. Material principles of justice set criteria for priority such as urgency, expected outcomes, waiting time, and contribution. Urgency prioritizes the sickest patients so those most at risk of dying are transplanted sooner. Prioritization of outcomes favors patients expected to have better quality of life, quantity of life, or longevity of graft survival. For example, patients with a high estimated posttransplant survival are given priority over the kidney grafts expected to last the longest.7 Waiting time prioritizes patients who have been waiting the longest. This is the primary principle for the allocation of kidneys from the deceased donor waitlist. Finally, contribution gives additional priority to those who have given into the system, such as previous living donors. A just allocation system aims to balance and specify the different material principles of justice in such a way as to make the system as fair as possible. Innovations in kidney exchange affect the allocation of live and deceased donor kidneys and therefore, require an assessment of fairness.
The principle of utility states that an action is right if it promotes as much or more good than an alternative action.5 In terms of allocating organs, aggregate good can be measured in graft survival, patient survival, or quality adjusted life years, among other criteria. One way to increase utility is to increase the number of kidneys available for transplant. As more patients receive transplants, the aggregate quality and quantity of life of patients with end-stage renal disease improves as compared with if they were to remain on dialysis. In addition to analyzing how innovative strategies to expand kidney transplantation affect justice in the allocation of organs, these programs must also be analyzed for their effect on the overall utility of kidney transplantation.
Finally, respect for persons requires that patients are treated as ends rather than as means.6 This requires that patients should be actively involved in their medical care, informed about their conditions, and allowed to make decisions regarding interventions that best reflect their personal goals and values. In organ allocation, the principle of respect for autonomy is put into practice in many ways.5 Donors, or families in the case of deceased donation, are given the right to decide if they want to donate and are allowed to direct donation to a known individual. Recipients are given the right to refuse organ offers. Transparency of the allocation process is a necessary component of respect for autonomy because it allows stakeholders to make informed decisions.
Ethical Considerations in Nonsimultaneous Extended Altruistic Donor Chains
NEAD chains are comprised of a series of donor and recipient pairs, started by an NDD. The donor for the first recipient donates to the second recipient and so forth until the chain ends with the final donor giving a kidney to the deceased donor waitlist (Figure 1D). Challenges with NEAD chains include creating fairness in allocation of NDD kidneys and chain ending kidneys, assessing potential disadvantages to subpopulations including blood type O recipients and coping with donor withdrawals that lead to premature breaks in the chains.
One concern with NEAD chains is that donating a kidney to a chain of living donor pairs prevents an individual on the deceased donor waitlist from receiving that kidney. Although the chain usually ends with an individual on the deceased donor waitlist, the patient who receives the chain-ending kidney is not the individual who would have received the CIK. Therefore, this may be unfair to the patient who remains on dialysis.8 However, NEAD chains generate more kidney transplants and decrease the total number of patients on the deceased donor waitlist, resulting in less competition for kidneys among those who remain on it. A review of a large multicenter living donor-recipient database found that the average number of transplants triggered by an NDD is 4.8.1 The overall utility gained by multiple transplants, an average of almost 5 for each NDD chain versus 1 for donation directly to the deceased donor waitlist, arguably outweighs the individual disutility of the patient who would have received the kidney transplant from the NDD.9
There is concern that certain subpopulations can be disadvantaged by NEAD chains because the chain-ending kidney that returns to the waitlist is usually not from a blood type O donor, so the blood type O kidneys are siphoned off to start donor chains without reciprocal return to the deceased donor waitlist. A case series of chain transplantations illustrates how blood type-O patients on the deceased donor waitlist can be disadvantaged.10 In this series of 54 NDD chains, 32 (59%) were started by a blood type O NDD. Of the 47 kidneys returned to the deceased donor waitlist, only 3 (6%) were blood type O. However, NEAD chains have been shown to increase blood type-O donations by releasing more blood type-O donors who have sensitized recipients, thereby increasing the number of transplants to blood type-O patients and reducing the number of waitlisted O blood type patients.1
A third issue with NEAD chains is that donors can withdraw from the chain at any time or for any reason, thereby terminating chains early. Melcher et al10 describe an example in which a donor unexpectedly decided not to proceed on the day of surgery. The untransplanted patient, whose donor had already donated to the next recipient in the chain, received a kidney 2 months later through a different chain. Donor withdrawal is rare and has been minimized through careful psychological evaluations of potential donors as well as attempts to decrease the time between the donor's recipient getting a kidney and the donor being asked to donate.11 However, this could become more prevalent if donors are expected to wait long periods before donating. If donor withdrawals become more prevalent, trust in kidney chain donation could suffer, and there might be less interest in pursuing this option for kidney transplantation. Although donor withdrawals terminate kidney chains early, it is essential that potential donors are never coerced into donating. Their decision to withdraw this must be respected.
Another concern with NEAD chains is the development of an adequate informed consent process for the recipient, the paired donor, and the NDD. The recipient consent requires an explanation of how the matching process works, what type of grafts is acceptable, and how prioritization is accomplished among competing recipients. The paired donor consent must discuss the risk that their intended recipient may not receive a kidney due to technical errors, shipping losses or uncertainty in timing. In addition, they must be informed that they have the right to withdraw at any time.4 Finally, NDDs should be informed of the option to donate directly to the deceased donor waiting list or to a chain.1 Although they may prefer to donate to a chain to trigger a series of transplants, knowing the options enable NDDs to donate a kidney in the way that they find most valuable.
Although there are ethical concerns regarding effects on the deceased donor waiting list, subpopulation disadvantages, consent, and donor vulnerability, NEAD chains have had an overall positive impact on utility within kidney transplantation. Considering approximately 35% of the candidates will be incompatible with an intended living donor, expansion of KPD through advanced donations or deceased donor initiated chains is a welcome proposition.12
Advanced donation occurs when an individual donates to a kidney exchange program before his or her paired recipient’s transplant. The following discussion is based on the ADP offered by National Kidney Registry (NKR), because it is currently the only kidney exchange consortium offering this option.13
Three Categories of Advanced Donation
Advanced donation is defined by the order in which donation and transplantation of a kidney occurs and can be divided into 3 distinct categories. The first type of advanced donation in KPD is “out-of-sequence” donation, in which a donor donates into a chain early because of time constraints, and their paired recipient receives a kidney a short time later. In these situations, the recipient already is matched into a chain but the donor cannot wait to donate for whatever reason. The first reported “out-of-sequence” donation was by a police officer with a brief donation window, who donated 1 week before his paired recipient (cousin) received a kidney from an NDD as part of a chain of transplantations.14 This pair took a calculated risk that the NDD would actually donate.
The second type of advanced donation is “short-term unmatched” donation in which the donor donates a kidney into the program before their recipient, who needs a kidney transplant, even has a match. The recipient then gets priority to be matched for a kidney. Flechner et al15 describe a “short-term unmatched” donation in which the donor was in the navy and needed to donate before his deployment. His dialysis-dependent mother, the paired recipient who was not matched at the time of his donation, was matched and transplanted 5 months later.
The final type of advanced donation is “voucher” donation, in which a living donor donates a kidney to receive credit for a named relative or friend to be transplanted in the future.16 What sets this type of advanced donation apart is that the voucher recipient is not yet in need of a kidney and may never need a kidney.13 Veale et al16 report that the first “voucher” donation where a 4-year-old child born with a poorly functioning solitary kidney was expected to become dialysis-dependent in 10 or 15 years. The child’s 64-year-old grandfather donated his kidney to a stranger triggering a chain of transplants, and his grandson now has a voucher to redeem for transplant when and if needed in the future. If it were not for the voucher program, the grandfather would likely have become ineligible to be a living kidney donor with advancing age.
Ethical concerns about ADP donation include the management of uncertainty, the extent of donor and recipient consent, the scope of the obligation that the organization has to the KPD recipient, and the potential to unfairly advantage the recipient.
Uncertainty and Consent in Advanced Kidney Donation
In ADP, there is uncertainty about when the paired recipient will get a kidney transplant. The degree of uncertainty increases as one goes from “out-of-sequence” donation, to “short-term unmatched” donation, to ‘voucher’ donation because the time between donation and matching the recipient increases. There is no guarantee for how quickly a kidney will be found for the recipient, and there is a possibility that a match may never become available, especially for highly sensitized patients or the recipient becomes too sick to transplant. Finally, there is also the possibility that the kidney exchange program shuts down before the recipient receives a kidney. In this scenario, the recipient might not be prioritized in other exchanges or on the deceased donor waiting list. In these situations, the donor has put himself or herself at risk without receiving the benefit of reciprocal donation to their intended recipient.
ADPs manage uncertainty with the consent process. The NKR consent documents clearly state the risk that the recipient might never receive a transplant if they become too sick before a match is found and that there may be a long delay in matching if the recipient is sensitized or has an O blood type.17 The consent process should inform patients and potential donors that the alternatives to advanced donation include waiting until a transplant is indicated and then pursuing deceased donation, living donation, or NEAD chains.
Obligations of ADPs to Recipients
Probably, the most important question in advanced kidney donation is that of the kidney exchange program obligations to ADP recipients. The donor consent form states that “my donation would give my IR [intended recipient] a prioritized opportunity to receive a kidney as part of a paired donation within the NKR [National Kidney Registry].”17 Although an exact timing of a match cannot be guaranteed, there may be a way to improve the consent process by providing a predicted timeframe based on the program and recipient characteristics to help the donor and recipient make informed decisions regarding advanced donation. The NKR policy states that ADP recipients will be prioritized after former NDDs who are now in need of a transplant and mending a real time swap failure.18 Some have argued that there may need to be a contractual agreement between the recipient and the organization.19
The NKR voucher program gives each recipient the opportunity to get a voucher from up to 5 donors, each of whom could start a NEAD chain benefiting other pairs in the NKR.13 The purpose of additional donors is to provide recipients with a voucher in the event that their transplant(s) fail. This may be pertinent for a young recipient who, over a lifetime, may require multiple kidney transplantations. Multiple donors will not result in increased priority but simply increase the number of credits available if multiple graft failures were to occur.
Effect on the Deceased Donor Waiting List
Advanced kidney donors provide CIKs with the eventual obligation to return a chain-ending kidney to their IR. Although an additional kidney is given to the deceased donor waiting list up front, one is eventually taken away to give to the IR, resulting in a net zero change in kidneys going to the deceased donor waiting list for a single advanced donor and IR pair. However, the opportunity to have up to 5 donors could result in 4 additional kidneys being given to patients on the deceased donor waiting list and a significant number patients coming off the list after being transplanted as part of chains initiated by advanced donors. If multiple donors per 1 recipient becomes the norm in ADP, this has the potential to substantially expand the number of kidneys available for chains, increase returns to the deceased donor waiting list, and increase the overall number of kidney transplants, all of which will increase utility.
Opportunities to Gain Recipient Advantage
ADPs may enable some donor-recipient pairs to gain advantage for hard-to-match recipients. For example, only about 15% of O recipient with non-O donor pairs find a match within kidney exchanges, as compared with 50% for other pairs.20 If these pairs entered into advanced donation, they could potentially improve the chances of the recipient getting a match independently rather than waiting to be included as part of a traditional chain. This practice could generate the need for more blood type O chain ending kidneys and create longer waiting times for O recipients trying to match into chains.
Deceased Donors as a Source of CIKs
Another innovative strategy proposed for kidney exchange is that of using deceased donor grafts as CIKs because 1 limitation of kidney chains is the availability of CIKs. Although CIKs have come traditionally from NDDs, deceased donor kidneys could also be used to start NEAD chains. The use of deceased donors for CIKs raises additional ethical concerns such as to how each party is consented, when a deceased donor kidney should be allocated to a chain rather than the deceased donor list, how the chain-ending kidneys are allocated, and the value of exchange of a living donor kidney for a deceased donor kidney.
There is precedent for using deceased donor kidneys in kidney exchanges.21 Delmonico and colleagues21 describe their experience with 17 deceased donor/living donor kidney exchanges. These cases were simultaneous paired donations from a deceased donor to recipients whose paired living donors donated to a patient on the deceased donor waiting list. In this series, 16 of the 17 incompatible recipients with living donors were blood type O and received a blood type O kidney from a deceased donor, whereas only 1 living donor provided a blood type O kidney to a recipient on the deceased donor waiting list. The primary ethical concern with these exchanges was that they further disadvantaged blood group O recipients on the deceased donor waiting list because blood type O kidneys were siphoned away from the deceased donor waiting list, and nonblood type O kidneys were returned to the list.22
Although there are no deliberate cases of using a deceased donor as a CIK, one of the advanced donation cases described by Flechner et al15 provides an illustration of how this process would work. They report a case in which a husband was willing to donate to his wife through a chain but she got a kidney from a deceased donor beforehand. The husband decided to proceed with nondirected donation several months later, thereby initiating a NEAD chain. This situation could be a propagation of a deceased donor CIK because the deceased donor kidney allowed the husband to start a NEAD chain. This example is complicated by the fact that the deceased donor graft rapidly failed soon after transplantation, and the patient was given priority in a kidney exchange because of her husband's donation. However, it does demonstrate that there was a willingness to donate even after the donor’s IR received a deceased donor kidney.
Consent for Deceased Donors as CIKs
Consent for using deceased donor kidneys to initiate chains involves 3 separate entities: the deceased donor, the recipient, and the living donor. Each has its own unique set of questions and concerns. First, with respect to the deceased donor, the main question is whether the family must provide explicit consent to use the kidney for a chain as opposed to donating to an individual on the deceased donor waiting list. In general, most deceased donors are nondirected, and their kidney is allocated to the candidates as prioritized by a match run. Factors, in addition to waiting time, which can increase prioritization, include multiorgan offers and zero antigen mismatches. Deceased donor initiated chains could be included in the prioritization scheme of kidney allocation to eliminate a requirement for additional family consent.
Another question is whether donor families can request to direct kidneys to chains rather than to an individual on the deceased donor waiting list. Currently, directed donation requires that the donor or donor family name an individual to whom they want to donate.23 Allowing families to donate to kidney chains may encourage organ donation because this can propagate multiple transplants. However, this may divert too many kidneys away from individuals on the deceased donor waiting list or cause logistical problems with mobilizing chains to meet deceased donor time constraints.
The consent process for recipients and donors entering kidney chain programs should account for the scenario that the recipient might receive a kidney from a deceased donor rather than a living donor. Consents for recipients listed for kidney chains already entail discussions of the logistics of chains including prioritization, matching, and quality of the kidney. Using deceased donor kidneys for chain initiation would need to be disclosed in the consent process because these kidneys are generally perceived to be of lower quality than living donor kidneys. However, patients receiving a deceased donor kidney without donor-specific antigens can achieve increased graft longevity compared with receiving a living donor kidney that requires desensitization.24 Moreover, recipients of younger deceased donor kidneys have improved outcomes when compared with older living donor kidney transplants.25 Therefore, a young deceased donor CIK would be appealing to a recipient who has donor-specific antigens to their paired living donor or has a significantly older paired living donor.
The paired donor must also provide consent for participation in kidney chain donation. As with the recipient consent, the donor consent requires discussion of prioritization, matching, and kidney quality. If deceased donors are used as CIKs, living donors could become reluctant to participate in chains because of concerns that they are providing a higher quality kidney than their intended recipient is receiving. However, kidney exchange programs that use deceased donor kidneys for chain initiation may identify more potential successful matches and decreased waiting time for patients. This could outweigh the discrepancies in quality between good deceased donor kidneys and living donor kidneys. Potential living donors and recipients should understand what the parameters for quality are and the trade-offs described above.
Diverting Deceased Donor Kidneys From Patients on the Waiting List
One argument against using deceased donor kidneys for chain initiation is that this strategy diverts kidneys away from patients who are on the deceased donor waiting list.26 Incorporation of this strategy into the allocation algorithms would change which patients were at the top of the match run. Even if a kidney is returned to a patient on the deceased donor waiting list, it would generally not go to the person who would have been at the top of the match list had this strategy not been used. Therefore, there is disutility to the individual on the deceased donor waiting list who would have theoretically received the kidney had it not gone to a chain. However, there is a gain in utility to patients on the waiting list just as with NEAD chains.
Several benefits contribute to the utility of using deceased donors for kidney chain initiation.26 First, a CIK facilitates multiple transplants, each to a person on the deceased donor waiting list. Removing people from the deceased donor list decreases competition among those who remain on it. Second, using deceased donor kidneys to initiate chains may significantly increase the number of transplants overall because the number of kidney chains is limited by the availability of CIKs. Third, the final kidney in the chain goes back to a patient on the deceased donor waiting list. This results in a net zero balance of kidneys leaving and returning to the deceased donor waiting list.
Despite potential increase in utility with using deceased donors as CIKs, certain groups may be disadvantaged. Blood type O kidneys will be in highest demand for chain initiation, and non-O kidneys will ultimately return to the deceased donor waiting list as was the case with simultaneous living/deceased donor kidney exchanges.21 Thus, blood type O patients on the deceased donor waiting list might be disadvantaged because blood type O kidneys would be diverted away from the deceased donor waiting list to recipients who have a willing but incompatible living donor. However, these patients may derive benefit if more blood type O donors with sensitized paired recipients are unlocked as has happened with NEAD chains.1 A retrospective review of NEAD chains in 2013 showed that 40 blood type O NDDs resulted in 112 transplants to blood type O recipients, as compared with the alternative of 40 transplants if they had donated directly to the deceased donor waiting list. Blood type O recipients potentially benefit because this strategy transplants patients who had been waiting for a deceased donor and decreases competition for kidneys among those remaining on the list.
Prioritization of Deceased Donor CIKs
Kidney allocation changed substantially with the implementation of the new Kidney Allocation System (KAS) in December 2014.27 This system prioritizes deceased donor kidneys based on both donor and recipient characteristics rather than strictly on waiting time alone. Recipient prioritization is based on their sensitization, Estimated Posttransplant Survival score, the Kidney Donor Risk Index (KDPI), zero antigen mismatch, prior living donation, and age (pediatric or not).28
If deceased donor kidneys are used to start kidney chains, their allocation should be incorporated into the current prioritization scheme. One proposal is that deceased donor CIKs be prioritized after local pediatric recipients and before local adults.26 In addition, a KDPI threshold for deceased donor kidneys should be set similar to the current KAS limit of 35% for pediatric recipients. Given that a deceased donor kidney is being provided in exchange for a living donor kidney, it might be appropriate also to limit deceased donor CIKs to KDPIs to those less than 35% so that only high-quality deceased donor kidneys are used to start chains. Unfortunately, this scheme would divert more of the highest-quality kidneys away from unsensitized adults. It would also further disadvantage patients who are unable to find a living donor, by putting yet another priority ahead of the general pool of patients on the deceased donor waiting list.
Just as with the prioritization of CIKs, the KAS must determine how to distribute and prioritize chain-ending kidneys to the list. They must decide if the chain-ending kidney should return to the Organ Procurement Organization (OPO) that provided the CIK. From the perspective of reciprocity, it makes sense to return the kidney to the originating OPO so that there is a net zero effect on the deceased donor waiting list at the local rather than regional or national level. Kidney exchanges must maintain cooperation from OPOs. If kidneys are allocated outside of OPOs without reciprocal returns, the incentive to increase donation will be reduced because the benefits of the donation are not going back into the local community.
Modeling the KAS for the addition of chain initiating and chain ending kidneys may predict how these changes will affect the current state of deceased donor allocation and determine the optimal prioritization of these kidneys. Modeling should also be able to predict the change in the number of kidneys available for transplant as well as the effects on the deceased donor waiting list. In addition, this can help determine where deceased donor CIKs should be in kidney allocation prioritization to maintain fairness. Finally, these models can help guide the allocation of chain ending kidneys back to the deceased donor waiting list.
Logistical Challenges With Deceased Donor CIKs
Beyond the ethical challenges that arise in using deceased donors to initiate kidney chains, there are also logistical challenges. First, chains take time to set up and are usually done on an elective basis so that donor and recipient operations can be done sequentially and in a reasonably short period. On the other hand, deceased donor transplants are done semiurgently. Therefore, chains starting with deceased donors need to be mobilized in an urgent rather than elective manner. This could lead to an extended delay before the first paired-incompatible living donor is able to propagate the chain as unintentionally illustrated by Flechner et al15 in the case described at the beginning of this section.
Multiple organizations will be involved in propagating deceased donor initiated chains. The deceased donor graft will go through the UNOS KAS. If matched to start a chain, the graft will be distributed through the organization managing that chain. Then, the final kidney in the chain will be returned to the deceased donor waiting list through to UNOS KAS. For this to be successful, there will need to be clear and timely communication between organizations. At a minimum, there should be policies developed specifically laying out requirements for the return of kidneys to the deceased donor waiting list as well as how situations in which chains end early without return to the deceased donor list will be handled.
This article highlights the ethical and logistical challenges of innovations in kidney exchanges. ADP and deceased donors as CIKs are 2 innovations that have the potential to significantly increase the number of kidneys available for transplant by initiating more chains. They have the potential to increase overall utility to the kidney transplant enterprise by increasing the number of high quality kidneys available for transplant. The ethical and logistical concerns are not insurmountable, and if done in a thoughtful way, these innovations will positively impact kidney transplantation and patients with end-stage renal disease.
The authors would like to thank John Watson for his graphical expertise in putting together the figures.