First described by Rapaport,1 kidney paired donation (KPD) facilitates living donor transplantation in end-stage renal disease patients with a blood group or human leukocyte antigen (HLA)-incompatible living donor.2 The likelihood of finding a successful match in KPD is increased with a larger potential donor pool,3 prompting the development of national KPD registries. The first national KPD registry was established by the Netherlands in 2005,4 and the organization of national registries has also been reported by the United Kingdom,5 Australia,6 and Korea.7
This report describes the development of Canada’s KPD program and summarizes the activity from program inception in November 2008 through December 2013.
Although the federal government is partially responsible for funding health care, oversight and administration of health services is the responsibility of provincial and territorial governments. Within provinces, deceased donor services are managed by organ procurement organizations, whereas living donor services are managed by individual hospitals. The federal government is responsible for safety of the system, but, before 2008, there was no national framework to standardize the delivery of organ donation and transplant services. Challenges included the lack of a national organization to facilitate KPD between centers, variability in histocompatibility laboratory practices, and absence of standard procedures for donor evaluation. The concept for a national KPD program was initially advanced at a 2005 Consensus. Subsequent recommendations and principles were proposed to the government, and a report (Table S1, SDC,https://links.lww.com/TP/B69) was endorsed by all members of the Canadian Society of Transplantation Kidney Working Group in 2008.
In parallel to these activities, Canadian Blood Services (CBS) was provided the mandate by provincial and territorial governments (except Quebec where the service relationship is contractual) to oversee integration of interprovincial organ donation and transplant services including creation of national organ donation and transplantation registries. Infrastructure development by CBS was enabled by assistance from the Johns Hopkins Transplant Program, who provided access to their computerized matching program. This program was modified by CBS, and a new Canadian electronic matching program and registry were developed by November 2008. Program oversight was organized through a geographically representative steering committee (see acknowledgements).
The lack of laboratory standardization and common operating procedures between centers had a major impact on the design of the Canadian KPD program. As well, it was decided to have donors travel to the transplanting center for two reasons: (1) to minimize cold ischemia time and (2) to preclude each living donor program from having to qualify one another to allow for the shipping of kidneys in compliance with the Health Canada legislated regulatory framework. This necessitated the development of standard operating procedures to facilitate evaluation and acceptance of donors from distant centers including shipping of donor medical charts, imaging studies, and laboratory specimens from the center where the donor was initially evaluated to the center performing the transplant.
Consensus on program objectives and standardized procedures, as well as the necessary infrastructure, human resources, and oversight to operate a national KPD program was completed in 2008. Because each province had to perform its own privacy assessment and establish a data sharing agreement with CBS, provinces joined the program at different timepoints. The program was initially launched in three provinces in November 2008, with full participation of all Canadian centers by November 2010.
As of December 31, 2013, 439 transplant candidates and 467 donors had registered in the program, including 416 candidates with a single donor, and 23 with one donor or more. Among the 23 candidates with one or more registered donor, 56% were sensitized with a Calculated panel reactive antibody (cPRA) of 97% or higher. In comparison, only 31% of those with a single donor had a cPRA of 97% or higher (P = 0.0013). Table 1 shows the characteristics of the 467 unique donor-candidate registered pairs and the proportion of the candidates in each group that were transplanted through KPD, but does not include the 47 nondirected donors. Among the 217 (46%) ABO-compatible pairs, one pair was HLA compatible, 25 pairs (12%) had a candidate with a cPRA of 1% to 50%, 82 (38%) had a candidate with a cPRA of 51% to 96%, and 108 (50%) had a candidate with a cPRA of 97% or higher (Table2). The HLA compatible pair enrolled in the program to assist other pairs, was matched in their first match cycle in the program.
Among the 252 ABO incompatible pairs, 181 (72%) included a blood type O candidate (Table 2). The candidates of blood group incompatible pairs had lower levels of sensitization (mean cPRA 43%) compared to candidates with a blood group compatible donor (mean cPRA 84%) (P = 0.005). Among blood group O candidates with an ABO incompatible donor, 20% had a cPRA of 51% to 96%, whereas 20% had a cPRA of 97% or higher (Table 1).
Forty-seven of the nondirected donors donated within the program, three donated in their local transplant centre but outside of the program, and six were withdrawn, two are still active in the registry.
As of December 31, 2013, 240 transplants were completed, including 46 transplants in wait-listed patients who received the last kidney in a domino chain. The completed transplants were the result of 16 matching cycles performed every 3 to 4 months (Figure S1, SDC,https://links.lww.com/TP/B69). The mean ± standard deviation time to complete the 10 paired exchanges, 16 closed chains, and 47 dominos was 103 ± 14; 133 ± 11, and 112 ± 6 days. Among the exchanges involving three or more transplantations, those involving transplant centers in multiple cities were complete in 124 ± 5 days compared to 95 ± 12 days in exchanges involving centers within the same city (P = 0.0015).
The median number of days from a candidate’s participation in a Match Cycle to transplantation was 182 (range, 47–1,741) (mean ± SD, 275 ± 17). The time to transplantation varied by candidate blood group (Figure S2, SDC,https://links.lww.com/TP/B69). Regardless of cPRA, blood group O candidates (Fig. 1, right) had a lower incidence of transplantation, as did candidates with cPRA ≥ 97%, whereas patients with cPRA less than 97% had an approximately one in two chances of finding a match.
Reasons for Decline of Proposed Matches
A total of 58 proposed matches were declined (Table 2). Seventeen proposed transplants were cancelled because of a positive initial cross-match not predicted by the virtual cross-match. Virtual cross-matches were based on information regarding unacceptable antigen specificities for HLA-A, HLA-B, HLA-C, HLA-DRβ1/3/4/5, HLA-DQ DQβ1, and DP gene loci entered at the time of candidate registration. The inability for all HLA “labs to perform DQ and DP typing until 2011 was the major reason for unexpected positive cross-matches. The protocol of confirming acceptable matches with an initial cross-match was effective in preventing collapse of proposed matches at an advanced stage. Nineteen matches were declined because of medical concerns with the proposed donor identified by the transplanting center. In most of these cases, the donor would have been accepted by the initial evaluating center, indicating variability in donor acceptance criteria between centers as an important issue. Four proposed matches were declined because the transplanting center identified surgical concerns with the donor. Nonmedical reasons for the decline of proposed matches included a variety of social factors.
Withdrawal of Registrants From the Program
A total of 155 pairs were permanently withdrawn from the program: Sixty-one candidates were transplanted outside the program; 37 pairs were withdrawn because of donor (n = 20) or candidate (n = 17) medical reasons; 53 pairs withdrew for nonmedical or unspecified reasons, including 33 that were donor related and 20 that were candidate related. Four pairs were withdrawn because of candidate death.
Figure 1 shows the cPRA of candidates who remained nontransplanted after each match cycle, and illustrates the accumulation of candidates with cPRA of 97% or higher in the program over time. Figure 2 shows the same information for candidates (Fig. 2A) and donors (Fig. 2B) by ABO blood group and shows the accumulation of group O candidates and the mirrored accumulation of group A donors.
Outcomes and Impact of KPD on Canadian Transplant Volume
There have been no donor deaths. There have been five allograft failures, five from recipient death from sepsis, one from rejection, one primary nonfunction, possibly relating to packaging in a shipped kidney, and two technical failures. One of the failed transplants was from a nondirected donor, requiring termination of the domino chain and accounting for discrepancy between the number of nondirected donors (n = 47) and wait-list recipients (n = 46). Among the 135 recipients with at least 1 year of potential follow-up, patient and allograft survivals were 99% and 96%, respectively. The biopsy confirmed acute rejection rate was 8%, whereas the antibody-mediated rejection rate was 3%. Figure 3 shows the total number of living donor and KPD transplants performed in Canada in 2008–2012,8 which has remained stable since the inception of KPD program.
Although other countries have established national KPD programs, the establishment of a national program in Canada posed unique challenges including vast distances between transplant centers, a regulatory barrier to shipping kidneys between centers, provincial organization of health care services, the absence of a national transplant organization with an operational authority, and lack of standardized operating and laboratory procedures between centers. In contrast to the United States, where a deceased donor kidney may be transplanted anywhere in the country, deceased donor kidneys are not routinely shared between provinces leading to significant geographic differences in waiting time.9 The foresight of the provincial governments to expand the mandate of CBS to improve delivery of transplant services, including the national KPD program, was therefore a pivotal step. The development of the KPD program was heavily dependent on the volunteer efforts of the Canadian transplant community, as well as international consultation and support. The successful implementation of the program therefore was equally the result of governmental support and the determination of the community. The operational success of the program is facilitated by the existence of a committed geographically representative steering committee with relevant expertise. The committee holds monthly teleconferences and meets annually. Proposed changes to procedures are vetted through an annual meeting with representatives from all transplant programs.
A key advantage to establishing a national KPD program was the provision of lifelong health insurance that is portable between provinces, obviating the administrative challenges of multiple insurance providers10 and the need to determine whether future health concerns that may arise after donation are related to donor surgery to obtain insurance coverage.11 The provision of lifelong donor health insurance also may be an advantage in increasing participation in KPD in Canada. In contrast, the current regulatory framework imposed by Health Canada and the lack of standardization between centers precluded shipping of live donor organs between centers and necessitated donor travel to the center performing the transplant. This required development of additional operating procedures and created a significant additional administrative workload for centers. Charitable contributions from the Kidney Foundation of Canada provided an important mechanism to support donor travel expenses until government funded programs to reimburse donor travel costs and partial support for lost wages were implemented. Shipping of live donor organs is routine in the United States,12 and we are currently working to remove the regulatory restriction on shipping. The requirement of donor travel made evident the variability in donor acceptance criteria between centers. Twelve proposed matches were declined because donors deemed suitable by the registering center were not medically acceptable to the transplanting center. An additional three proposed matches were declined because of surgical concerns with the donor in the transplanting center. Although the declines for surgical reasons likely reflect differences in surgical expertise, the medical declines reflect uncertainty regarding the use of donors with isolated medical concerns, such as hypertension, low GFR, and high BMI.13 To overcome this problem, an ad hoc donation committee has been charged with establishing a standard for donor enrollment.
A factor that has propelled the success of the Canadian KPD program is the participation of nondirected donors. Of the 240 transplants included in this report, 165 were the result of domino chains initiated by the 47 nondirected donors. We elected to recognize the work of identifying such donors by allocating the last kidney in a domino chain to the deceased donor waiting list in the center that registered the donor rather than creating a bridge donor. The nonuse of bridge donations, along with our policy of limiting transplant chain length to five transplants, and performing chain transplants over 1 to 2 days whenever feasible likely contributed to the absence of any donor defaults in the program. Although we have considered bridge donations, we believe the integrity of the program is an important factor in expanding the program. The future of the program is, in part, dependent on implementation of innovative strategies to increase participation. The number of registrants in any match cycle is modest, and increased participation in the program would increase the likelihood of matching.3 The capacity to increase participation is suggested by the fact that there are over 3,500 patients wait-listed for deceased donor kidney transplantation with less than 500 living donor transplants performed annually. Although the KPD program is truly national, there is large geographic variation in program participation, which mirrors the regional variation in the use of living donor transplantation in the country.8 Understanding the basis for the geographic variation in living donor transplantation in Canada is an important issue in need of further research. Standardization of centre resources to optimize the identification and registration of donors and candidates has been identified as a strategy to increase participation. There is currently lack of consensus on encouraging participation of compatible pairs in the program, which may be particularly useful in increasing the transplantation of blood group O candidates. Participation of compatible pairs in KPD may be encouraged by implementation of strategies to facilitate rapid donation and transplantation, ensuring compatible donors do not have to travel, and perhaps even offering a reward or incentive.14
The future of the program is also dependent on development of mechanisms to increase the transplantation of candidates who currently have a low likelihood of transplantation. This includes transplantation in the presence of donor specific antibody (i.e., when the flow cross-match is negative) or ABO incompatible transplants in HLA-matched recipients.15 Consensus on the level of acceptable risk as well as a standardized approach to immunosuppression and clinical monitoring is needed before including low-risk DSA-positive transplants in the program.
In summary, establishment of a truly national KPD program enabled the transplantation of 158 patients, including many with significant immunologic barriers to transplantation and helped maintain the volume of living donor transplantation in Canada.
MATERIALS AND METHODS
The Canadian KPD Matching Algorithm
The development of the matching algorithm was informed by the published literature and stakeholder agreement to maximize the number of transplants, minimize the logistics associated with donor travel, and be consistent with the principals of deceased donor allocation in Canada. The matching algorithm, transplant characteristics, and outcomes are reviewed annually to ensure achievement of program objectives, and any proposed changes to the algorithm would be evidence based, modeled to their impact on program activity, and require consensus endorsement.
Standardization of Canadian HLA Laboratories
The requirement for HLA compatible (defined as no DSA by luminex single antigen beads and flow cross-match negative) necessitated standardization of Canadian HLA laboratories. A solid phase flow-based platform for antibody identification, using single antigen beads, was adopted as the minimum standard. To ensure precision in antibody identification between laboratories, a series of proficiency testing exercises was performed. External consultants (Drs. Gebel, Bray, and Eckels) were engaged to assist with this process and provided standardized sera and methods for test comparison. The first exercise determined differences between laboratory techniques. In laboratories where results were discrepant when tested by their own procedure but concordant using the reference laboratory’s method, a recommendation was made to adopt the standardized reference laboratory procedure. Subsequent exercises focused on defining criteria for identification of unacceptable antigens (antigens likely to result in a positive cross-match and/or have clinical consequences) and indeterminate antigens (antigens of uncertain clinical significance). These final classifications, interpreted in the context of laboratories’ own validation data, were then used to enter antibody specificities. Annual review of unexpected positive cross-matches and further proficiency testing exercises resulted in refinement of criteria, and in 2011, a guideline threshold of mean fluorescence intensity higher than 1,000 was established.
Participating transplant programs identify and register potential donors and candidates in the web-based registry. There is no registration fee, but registrants must be approved for donation and transplantation by their local transplant center and candidates must be placed on hold on the deceased donor waiting-list. The matching algorithm is currently run at 4-month intervals. The various proposed matching scenarios are ranked by a points system (see Table 2) and reviewed by teleconference of the steering committee. Proposed matches are then communicated to the involved transplant programs. The first priority is to confirm proposed transplants at highest risk of collapse (i.e., transplants involving patients with high cPRA). Donor cells are sent to the transplanting center, and a flow cross-match is performed using a fresh serum sample from the intended recipient. Lower immunologic risk transplants are then similarly confirmed. Donor medical charts, imaging studies, and laboratory specimens are sent to the transplanting center for local review and testing. All proposed transplants are subject to acceptance of both the donor and recipient by the transplanting center. A plan to complete all linked transplants (i.e. those in a closed chain or domino) over a 1-day to 2-day timeframe is developed by the involved transplant centers before operative dates are confirmed. Transplant centers are encouraged to complete all transplants in a given match cycle before the next match cycle.
Descriptive statistics included the use of proportions for categorical variables, and the use of the mean ± standard deviation or median and range for continuous variables. Group differences were compared with the use of the chi-square test or t test as appropriate. The time to transplantation was determined with use of the Kaplan-Meier method, and group differences were compared with the log-rank test.
The KPD Steering Committee Members include: Ian Alwayn, Halifax; Patricia E Birk, Winnipeg; Brendan Barrett, St. Johns, Patricia Campbell, Edmonton; Edward Cole, Toronto (Chair); John Gill, Vancouver; Nessa Gogan, Saint John; Martin E. Karpinski, Winnipeg; Rahul Mainra, Saskatoon; Mauricio Monroy, Cuadros, Calgary; Peter Nickerson, Winnipeg; Michel R. Pâquet, Montreal; Jeffery Zaltzman, Toronto.
The authors thank the Canadian Blood Services for the management and services of the Canadian Transplant Registry. The authors also thank the medical, surgical, nursing laboratory, and operating room staff in Canadian transplant centers who have contributed to the establishment of the National KPD Program, and the Kidney Foundation of Canada for supporting donor travel costs in the Program.
1. Rapaport FT. The case for a living emotionally related international kidney donor exchange registry. Transplant Proc
. 1986; 18( Suppl 2): 5.
2. Montgomery RA, Zachary AA, Ratner LE, et al. Clinical results from transplanting incompatible live kidney donor/recipient pairs using kidney paired donation. JAMA
. 2005; 294: 1655.
3. Segev DL, Gentry SE, Warren DS, et al. Kidney paired donation and optimizing the use of live donor organs. JAMA
. 2005; 293: 1883.
4. De Klerk M, Keizer KM, Claas FH, et al. The Dutch national living donor kidney exchange program. Am J Transplant
. 2005; 5: 2302.
5. Johnson RJ, Allen JE, Fuggle SV, et al. Early experience of paired living kidney donation in the United kingdom. Transplantation
. 2008; 86: 1672.
6. Ferrari P, Fidler S, Holdsworth R, et al. High transplant rates of highly sensitized recipients with virtual crossmatching in kidney paired donation. Transplantation
. 2012; 94: 744.
7. Park K, Moon JI, Kim SI, et al. Exchange donor program in kidney transplantation. Transplantation
. 1999; 67: 336.
8. Canadian Organ Replacement Register CIfHI. 2012 CORR Report Treatment of End Organ Failure in Canada 2001–2010. Toronto: CIHI; 2012.
9. Tonelli M, Klarenbach S, Manns B, et al. Residence location and likelihood of kidney transplantation. CMAJ
. 2006; 175: 478.
10. Irwin FD, Bonagura AF, Crawford SW, et al. Kidney paired donation: a payer perspective. Am J Transplant
. 2012; 12: 1388.
11. Ommen ES, Gill JS. The system of health insurance for living donors is a disincentive for live donation. Am J Transplant
. 2010; 10: 747.
12. Segev DL, Veale JL, Berger JC, et al. Transporting live donor kidneys for kidney paired donation: initial national results. Am J Transplant
. 2011; 11: 356.
13. Young A, Storsley L, Garg AX, et al. Health outcomes for living kidney donors with isolated medical abnormalities: a systematic review. Am J Transplant
. 2008; 8: 1878.
14. Gill JS KS, Barnieh L, Caulfield T, et al. Financial incentives to increase Canadian Organ Donation: quick fix or fallacy. Am J Kidney Dis
. 2013; 63: 133.
15. Bingaman AW, Wright FH Jr, Kapturczak M, et al. Single-center kidney paired donation: the Methodist San Antonio experience. Am J Transplant
. 2012; 12: 2125.