Ingredients for a Successful Living Donor Kidney Exchange Program

A simple solution for all kidney patients with a willing but incompatible living donor, because of a positive cross match or an ABO blood type incompatibility, is living donor kidney exchange (1). Over the last two decades, several centers in various countries embarked on these programs (2). In South Korean the first procedure was performed in 1991, in Switzerland a single exchange took place in 1999, while individual centers in the United States, The Netherlands, and Canada followed in 2000, 2003, and 2005, respectively. However, not all these initiatives have resulted in regular programs. Most single centers are not able to accrue enough candidates for efficient exchange of both easy and difficult-to-match pairs on a permanent basis. For that purpose collaboration with other centers is essential. Indeed, Segev et al. in this issue (3) calculated that by expansion of a paired kidney exchange (PKE) program to a national level more difficult to match couples could be helped. Along the same line, Basu et al. proposed PKE to improve HLA-matching and thus transplantation outcomes in India (4). In the Netherlands with a population of 16.4×106 inhabitants we started a national exchange program in 2004. All seven transplant centers work according to a common protocol for the evaluation of the donor, while we agreed on only four simple rules. Registration of donor/acceptor combinations can take place four times a year only if donor and acceptor are medical suitable for donation and transplantation. Allocation is the responsibility of an independent organization, the Dutch Transplant Foundation. Because computerized matching will result in huge numbers of possible combinations of donor-recipient pairs, allocation criteria are necessary. Thus, our computer program selects the best possible combinations on six preset conditions: (1) a maximum number of matched couples, (2) blood type identical before blood type compatible matches, (3) most difficult to match patients (highly sensitized recipients) first, based on HLA match probability, (4) short chains preferred (e.g. rather two doublets than one quartet), (5) couples distributed over multiple centers, (6) wait time calculated from the first day of dialysis. PKE is for many patients the best if not the only option and to be preferred over dialysis. Therefore, we kept our algorithm as simple as possible without too many complicating parameters, for example, donor age, gender, CMV sera status, renal function, or number of HLA mismatches. The third rule is that the national HLA reference laboratory performs all the cross matches between the new donor and the recipient. And finally surgical procedures will take place on the same moment and not the kidney will be transported but the donor travels to the recipient center to ensure minimal logistic problems and to keep ischemia times as short as possible. With these four important rules we now run a national program for four and half years. From January 2004 to June 2008, we registered 143 pairs with blood type incompatibility and 133 pairs with a positive cross-match. We have now performed 18 match runs with a median input of 14 new pairs (range, 7-22 pairs) per match run and a median number of couples participating per match run of 47 (range, 16-66 pairs) which is still increasing (Fig. 1). In the ABO blood type incompatible group almost 70% (99 of 143) had blood type O recipients. Median PRA in the positive cross match group was 46% (2-100). We found new match combinations for 159 of 276 (58%) registered couples. Easy-to-match couples with blood type A recipients and blood type B donors or vice versa had a 85% success rate. However, also the positive cross match couples with a median PRA of 46% could successfully be matched in 74% of the cases. Combining blood type incompatible and positive cross match donor-recipient pairs in one program increased our success rate and made it possible that 26 of 99 (27%) of the ABO blood type incompatible pairs with blood type O recipients could be matched. Our experience with a national PKE program underscores the proposals of Segev et al. and Basu et al. that enlarging the pool of participants in an exchange program will result in high number of match possibilities. Our Dutch matching algorithm was adapted to create doublets in 2004, triplets in 2005 and for a maximum possible chain length in 2007. This resulted in an exponential growth in possible combinations, for example, in a pool with 57 pairs we found 3,427,603 combinations. By changing the sort order of the algorithm the outcome of the matching process can be influenced. We chose for the maximum number of combinations, but other options are also possible. Basu et al. might give priority to HLA matching, while Segev et al. could prefer difficult to match, sensitized patients, and travel requirements. In conclusion, PKE programs can be successful but relative high numbers of candidates and thus cooperation between centers, and trust in each other is essential. Independent organizations should be responsible for the allocation and the final cross matches, while computerized matching programs should be flexible enough to fulfill the needs of a particular transplant community. To optimize match combinations, compatible donor-recipient pairs may be enrolled and altruistic Good Samaritan donors may be helpful in a domino-paired program for unsuccessful PKE couples (5, 6).FIGURE 1.: Input of donor-recipient pairs per match run from January 2004 until May 2008.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.