Increased Access to Transplantation for Blood Group B Cadaveric Waiting List Candidates by Using A 2 Kidneys: Time for a New National System?

Transcription

1 Copyright C Munksgaard 2002 Munksgaard International Publishers ISSN Increased Access to Transplantation for Blood Group B Cadaveric Waiting List Candidates by Using A 2 Kidneys: Time for a New National System? Paul W. Nelson a, Charles F. Shield III b, Nicolas A. Muruve c, Daniel Murillo d, Bradley A. Warady e, Mark I. Aeder f and Christopher F. Bryan g, * a Saint Luke s Hospital and the University of Missouri, Kansas City, MO, USA b Via-Christi St. Francis Regional Medical Center, Wichita, KS, c University Hospital, Columbia, MO, USA d Kansas University Medical Center, Kansas City, KS, USA e The Children s Mercy Hospital, Kansas City, MO, USA f Research Medical Center, Kansas City, MO, USA g Midwest Transplant Network, Westwood, KS, USA * Corresponding author: Christopher F. Bryan PhD, cbryan@mwob.org Since blood group B end-stage renal disease (ESRD) patients have less access to donor kidneys and a higher minority composition than any other blood group, the United Network for Organ Sharing (UNOS) approved a voluntary national kidney allocation variance to allow organ procurement organizations (OPOs) to preferentially allocate A 2 and A 2 B kidneys to B candidates. The Midwest Transplant Network OPO has preferentially allocated and transplanted kidneys from blood group A 2 and A 2 B donors to our blood group B waiting list candidates for more than 7 years to increase access to kidneys for the B candidates on our OPO-wide waiting list. Between 1994 and 2000, a total of 121 blood group B ESRD patients from our OPO-wide cadaveric kidney waiting list were transplanted. Thirty-four per cent (41/121) of those B candidates received either an A 2 or an A 2 B kidney. One- and 5-year graft survival rates for the group of B recipients of A 2 or A 2 B kidneys were 91 and 85% (died with functioning graft [DWFG] censored), respectively, which were not significantly different from those of 91 and 80% for the 80 B recipients of B or O kidneys (Wilcoxon Ω 0.48; log-rank Ω 0.55). These data support the national trial for additional OPOs to voluntarily allocate A 2 and A 2 B kidneys preferentially to B waiting list candidates, thus increasing access of blood group B patients to renal transplantation. Key words: A 2 to B, access, kidney transplant, UNOS variance Received 26 July 2001, revised and accepted for publication 13 September 2001 Introduction Blood group B end-stage renal disease (ESRD) patients have longer waiting times prior to cadaveric renal transplantation compared with other blood groups (Table 1) (3). In 1996, the median national waiting time for B kidney candidates was 1426 d, compared with 641 d for A candidates (Table 1) (2). There are multiple reasons why B patients may wait longer before cadaveric kidney transplantation, but 65.9% of the national B waiting list comprises minority recipient candidates (Table 1), and African-Americans have historically had longer waiting times (3). The high minority composition of the national B waiting list is not unexpected, as the nation s donor population is predominantly white (75.9% in 1999), a population with a blood group B incidence of 9%, far lower than the 19% and 25% incidence seen in African-American and Asian populations (2) (Table 1). Another reason is that blood group B candidates have less access to cadaveric kidneys than blood group A candidates. United Network for Organ Sharing (UNOS) data from 1999 show that 16.8% of the national waiting lists were B candidates and they received 13% of the cadaveric transplants (Table 1) (2). By contrast, 28.2% of the national lists were A candidates and they received 38.2% of the cadaveric kidney transplants (Table 1) (2). The transplant rate approximates the ABO frequency in the predominantly white donor population (Table 1). This study focuses on the experience of our Midwestern OPO in preferentially allocating and successfully transplanting kidneys from A 2 and A 2 B donors into B patients (4,5), a practice that has increased the transplantation rate of blood group B candidates in our OPO by one-third (6). These data provide support for the implementation of a similar national trial of A 2 allocation to the B cadaveric kidney waiting list. The Organ Procurement Transplant Network (OPTN)/UNOS Board of Directors, at the request of the UNOS Minority Affairs Committee, released this allocation proposal for Public Comment on March 15, 2001, and approved it on June 28 29, It is the intention that this national allocation policy will, at least partially, alleviate some of the access problems experienced by blood group B candidates and increase their transplantation rate while minimizing any effect on the A waiting list. This report demonstrates to kidney transplant programs in OPOs across the United States how implementation of transplantation of A 2 and A 2 B kidneys into B candidates can impact their B waiting list. Materials and Methods Study design The time-frame for this analysis and review was from January 1, 1994, to December 31, 2000, and included all consecutive cadaveric donors and 94

2 A 2» B Kidney Transplants Increase Access Table 1: Blood group phenotype frequencies and national (UNOS) waiting list characteristics by blood group for kidney transplantation Blood group Blood group frequency a UNOS kidney waiting list characteristics White Black Asian Median wait Percentage Percentage of Percentage of time (d) b minority c patients on list b transplants b B 9% 19% 25% % 16.8% 13% A 44% 27% 27% % 28.2% 38.2% A 1 34% 19% 27% A 2 10% 8% 0% O 44% 49% 43% % 52.5% 43.1% AB 3% 5% 5% % 2.5% 5.8% a These data are from (1). b These data are from (2). c Data provided by UNOS for the national cadaveric kidney candidate list for August 7, transplants at each of the seven transplant centers in our OPO. The blood group B transplant patients analyzed and compared were separated into the following two groups: those who received A 2 or A 2 B kidneys (n Ω 37) and SPK (n Ω 4) (total n Ω 41), and those who were transplanted with B (nω 74) or O (n Ω 6) kidneys or received SPK (total n Ω 80). We also analyzed the OPO-wide B and A waiting lists with respect to time waiting at a specific point during the year (January) from January 1, 1994, to December 31, The basic demographics of those two groups (Table2) show that the only significant differences were a higher degree of HLA mismatch, lower peak PRA levels, and lower percentage of male donors in the group of B recipients of A 2 or A 2 B kidneys compared with B recipients of B or O kidneys. Kidney preservation and immunosuppression Cadaveric donor kidneys were flushed in situ and preserved in ViaSpan (Belzer UW, Dupont Pharma, Wilmington, DE). Post-transplant treatment protocols for recipients of A 2 and A 2 B kidneys did not differ from those used for ABO-compatible recipients and were individualized by each transplant center. All patients received methylprednisolone, cyclosporine or tacrolimus, and azathioprine or mycophenolate mofetil with or without antibody induction (monoclonal or polyclonal). No patients underwent plasmapheresis or splenectomy as part of this protocol. A 2 /A 2 B to B OPO Cadaveric Kidney Allocation Variance Since 1991, our OPO has had a UNOS-approved cadaveric kidney variance to allocate A 2 and A 2 B kidneys, and since 1994, we have allocated A 2 and Table 2: Demographics of B candidates transplanted within Midwest Transplant Network (MTN) organ procurement organization Demographic A 2 /A 2 B» B B/O» B p value Percentage male recipients 60% 67.5% NSD a (24/40) (52/77) Primary transplants 82.9% 85% NSD (34/41) (68/80) HLA-A,B,DR mismatches (n Ω 37) (n Ω 67) Peak PRA (n Ω 35) (n Ω 64) Recipient age (years) NSD (n Ω 41) (n Ω 80) Donor age (years) NSD (n Ω 41) (n Ω 80) % male donors 44% 68.7% (18/41) (55/80) a NSD, not significantly different (p 0.05). A 2 B kidneys first to high PRA ( 79%) A 2 and A 2 B patients and then to high PRA A 1 and A 1 B patients. A 2 and A 2 B kidneys are then allocated to B patients. The routine UNOS allocation algorithm is used within each of the above three allocation tiers of the variance (e.g. points given for degree of HLA match, waiting time, pediatric patients, and PRA level). Simultaneous kidney pancreas candidates are allocated organs based on priority on the kidney waiting list. ABO grouping Red blood cell grouping was performed using standard American Association of Blood Banks immunohematological methods with commercial antisera with concurrent positive and negative controls. Donors were designated as A 2 if they lacked reactivity of their red blood cells with the anti- A 1 lectin Dolichos biflorus. This grouping was performed on donors by the Midwest Transplant Network laboratory (Westwood and Wichita, KS), or the American Red Cross Transplant Laboratory in Columbia, MO. We have routinely confirmed the phenotype of all A 2 donors by at least two of the above labs or an outside lab in order to decrease the likelihood of mistyping an A 1 donor as an A 2. If there is any reactivity with the anti-a 1 lectin, the donor is declared an A 1 for the purposes of allocation. It is important to emphasize that when testing for A 2 in cadaveric donors, only a pretransfusion blood specimen was utilized, as, if transfusion had occurred, the blood may have been from an A 1 donor, resulting in blood grouping test results being falsely interpreted as an A 1 instead of an A 2 phenotype. A-isoagglutinin (anti-a) titer determination Serial dilutions of patient sera were made in saline and tested for agglutinating activity against both A 1 and A 2 erythrocytes following a 10-min incubation at 22æC and the titers read macroscopically. No antiglobulin test was performed. To differentiate between IgM and IgG agglutinating activity, sera were treated with the sulfhydryl compound dithiothreitol at 0.01M and agglutination by treated sera was attributed to IgG. Since May 26, 1997, we have performed the anti-a titer assay using just A 1 erythrocytes and determining only the IgG titer. This change was implemented in order to reduce the amount of work performed when doing anti-a titers, since our transplant programs routinely clinically rely on the IgG anti-a titer using A 1 erythrocytes. A detailed procedure for our anti-a titer assay can be seen on our website ( under Services, Clinical Studies and Publications (Transplantation of Blood Group A2 Donors, item 8). Furthermore, on November 14, 1997, our laboratory developed a proficiency testing exchange for anti-a testing with three other labs in the United States. Definition of a patient s anti-a titer history The critical component of our variance was that we would perform an anti- A titer on all eligible candidates for A 2 or A 2 B kidneys (i.e. B and O patients) 95

3 Nelsonetal. each quarter. For the 41 blood group B recipients of A 2 or A 2 B kidneys, 73.1% (30/41) had a consistently low anti-a titer history; 22% (9/41) had one pretransplant titer that was low ( 4); and 4.9% (2/41) had a low titer history except for one high titer (8). Since 1998, for B waiting list candidates who were eligible to receive A 2 or A 2 B kidneys, anti-a titers were done monthly for the first 6 months following activation, and then quarterly. That was done to establish more rapidly a titer history for these patients, since in 21% (9/41) of the cases, only a single pretransplant titer was available. Our current policy is to transplant only patients whose entire anti-a titer history is low. If a patient does not have a current anti-a titer available at the time an A 2 kidney is offered (one done in the last 3 months), the patient is bypassed since we have shown that the cold ischemia time is increased by an average of 6h when current titers are performed [see Table 2 in Nelson et al. (5)]. For that reason, it is important to establish and maintain a sufficient and current anti-a titer history for each blood group B patient so that, at the time of an A 2 kidney offer, the decision whether or not to transplant can be made without additional testing. Statistical analysis The Lifetest procedure in Statistical Analysis Systems was used to determine actuarial statistics, and the Kaplan Meier method was applied to compute univariate survival estimates. The Wilcoxon and log-rank tests were used to determine the statistical significance of differences between survival curves. The chi-square (c 2 ) test was used to compare distribution differences. Student s t-test was used to compare group mean differences. Results The data in Table 3 show that, of the 121 B cadaveric transplants performed by the seven kidney transplant programs in our OPO, 34% (41/121) of the patients received either an A 2 (29%) or an A 2 B (3%) kidney, including four of six SPK patients who received their organs from an A 2 donor. Table 4 shows that 1- and 5-year graft survival rates (censored for patient death with a functioning graft, DWFG) of the blood group B recipients of A 2 or A 2 B kidneys (91 and 85%, respectively) were not significantly different from those of the B recipients of B or O kidneys (91 and 80%, respectively) (Wilcoxon Ω 0.48; log-rank Ω 0.55). Even when patients were not censored for DWFG, the 1 and 5-year graft survival rates of the B recipients of A 2 or A 2 B kidneys (84 and 72%, respectively) were not significantly different from the group that received B or O kidneys (84 and 64%) (Wilcoxon Ω 0.78; log-rank Ω 0.75). All four SPK A 2 into B recipients have func- Table 3: Distribution of blood group B patients transplanted from 1994 through 2000 by centers within the Midwest Transplant Network (MTN) organ procurement organization Organs Donor ABO» Incidence of B transplanted recipient ABO patients transplanted Kidney B and O» B 68% (78/115) A 2» B 29% (33/115) A 2 B» B 3% (4/115) Simultaneous B» B 33% (2/6) kidney π pancreas A 2» B 67% (4/6) Totals B and O» B 66% (80/121) A 2 and A 2 B» B 34% (41/121) tioning kidney and pancreas grafts, with an average followup of 17 months (range: 3 23 months) (Table 5). An important component of safely performing A 2 transplantation into B and O recipients has been to evaluate the titer of the A IgG isoagglutinins in B and O patients, as 95.1% (39/41) of the B patients transplanted with A 2 kidneys consistently had low anti-a titers ( 4), or had a single low titer before transplantation. The incidence of blood group B patients who have a consistent (at least 1 year) history of low anti-a titers is 77% in white patients (54/70) and 69% in black patients (23/35), not a statistically significant difference (p Ω0.2, c 2 ). The incidence of minority patients transplanted with A 2 or A 2 B kidneys was 22% (9/41) and the incidence of blood group B minority patients transplanted with B or O kidneys was 39% (31/80), a nonstatistically significant difference (p Ω0.06, c 2 ). However, the minority rate of blood group B recipients, 33% (40/121), was significantly higher than that of blood group A recipients, 12% (48/413) (p Ω , c 2 ). Furthermore, the median waiting time for the 41 blood group B recipients of A 2 or A 2 B kidneys or SPKs (166 d) was less than that of B recipients of B or O kidneys (308 d), and was even less than that of A recipients (288 d). Since this allocation algorithm has increased the transplantation rate of patients on our OPO-wide blood group B cadaveric renal waiting list, it is important to monitor the composition of our OPO waiting list over time. Table 6 shows that from 1994 to 1997, our OPO-wide active B waiting list candidates ranged from 38 to 45 patients, with low PRA ( 40%) B candidates, and from 1998 to 2000, the list range decreased to patients, with 12 to 16 low PRA B candidates. It is important to note that the low PRA minority patient group is declining at a rate similar to the entire low PRA group. Since one of the criticisms of this allocation policy is that A candidates are denied a portion of blood group A kidneys, we have routinely monitored the OPO-wide A waiting list and number of A transplants performed (Table 7). Between 1994 and 2000, we transplanted 37 A 2 kidneys into B patients, comprising 8.8% (37/416) of the total number of local blood group A cadaveric kidney transplants. Our data have also shown that the number of patients on the A waiting list started to decline in 1999 and the number of both blood group A donors and transplant recipients has gradually increased. Discussion Recent attention has focused on racial disparities in organ accessibility and survival following renal transplantation (3,7 11). One of the accessibility problems discussed by Young and Gaston (11) associated the disparity with the distribution of blood groups in different race groups (as illustrated in 96 American Journal of Transplantation 2002;2:94 99

4 A 2» B Kidney Transplants Increase Access Table 4: Graft survival of B recipients transplanted in the Midwest Transplant Network (MTN) organ procurement organization with A 2 or A 2 B cadaveric kidneys compared with B or O Kidneys ABO combination DWFG a censored Graft survival (years) p values for: (Wilcoxon log-rank) A 2 /A 2 B» B Yes 91% 91% 85% 85% 85% 0.48 (nω41) (28) b (20) (14) (5) (4) B, O» B Yes 91% 86% 84% 80% 80% 0.55 (nω80) (60) (50) (37) (23) (16) A 2 /A 2 B» B No 84% 77% 72% 72% 72% 0.78 (nω41) (28) (20) (14) (5) (4) B, O» B No 84% 77% 73% 68% 64% 0.75 (nω80) (60) (50) (37) (23) (16) a DWFG, patient died with a functioning graft. b The number in parentheses at each time-point represents the number of patients at risk through the end of each respective year. Table 5: A 2» B kidney/pancreas transplants performed by Midwest Transplant Network (MTN) organ procurement organization centers Date Anti-A IgG titer history a C peptide HbA 1 C Creatinine Before After Before After Before After Clinical status b February 14, 1999 Low 0.3 ND Functioning March 10, 1999 Low Functioning April 12, 1999 Low ND ND Functioning November 10, 1999 Low Functioning a Low titer history means all titers 4. b As of February, The data obtained after transplantation were the most current available in each patient for each marker. Table 6: Organ procurement organization (OPO)-wide B cadaveric kidney waiting list characteristics: Year Month Number of transplants B waiting list B A 2 /A 2 B» B (%) Total Low PRA a Low PRA a minority (% of low PRA) 1994 January 11 4 (36%) (42%) 1995 January 22 3 (14%) (39%) 1996 January 11 1 (9%) (39%) 1997 January (42%) (37%) 1998 January 22 8 (36%) (46%) 1999 January 13 8 (61%) (50%) 2000 January 16 6 (37%) (50%) Totals Mean 17 6 (33%) (47%) a Low PRA defined as peak PRA 40%. Table 1). We have presented data to show that at least within one of the challenging areas, the B blood group recipient pool, a partial solution to patient access and transplantation may be achieved. Based in part on these results and those of Norman s group (12), the OPTN/UNOS Board approved (June 28 29, 2001) the allocation of A 2 and A 2 B kidneys from cadaveric donors to blood group B recipients for OPOs to join this national trial voluntarily. Over a period of 7 years, our data have shown that within a single OPO and with a UNOS-approved cadaveric kidney allocation variance, we are able to transplant successfully 34% more cadaveric B waiting list candidates by utilizing A 2 and A 2 B donor kidneys (Table 3), while having a small impact on A waiting list candidates. Graft survival in the group of B patients receiving A 2 or A 2 B kidneys was equivalent to that of B patients who received B or O kidneys (Table 4). We have also shown the utility of our allocation variance in SPK candidates, as 67% (4/6) of our blood group B SPK patients received their organs from A 2 donors (Tables 3 and 5), and all four continue to maintain functioning grafts at 3 17 months (Table 5). These are the first reported cases of SPK transplantation from A 2 donors to B recipients. The national variance to preferentially allocate A 2 or A 2 B kidneys from cadaveric donors to B waiting list candidates is designed to increase the access of kidneys to blood group B waiting list candidates. The UNOS Minority Affairs Committee has 97

5 Nelsonetal. Table 7: Characteristics of organ procurement organization (OPO)-wide blood group A waiting list candidates and cadaveric transplants Year Month A A Kidneys Local A A 2» B waiting donors transplanted transplants (% of local A list from OPO A (includes imports) transplants) donors 1994 January (8.7%) 1995 January (4.4%) 1996 January (2.1%) 1997 January (21.9%) 1998 January (9.5%) 1999 January (10.1%) 2000 January (8.1%) Totals (8.8%) Mean a Note: a portion of these kidneys are transplanted outside the OPO (zero mismatches and exported kidneys) but most are transplanted locally. supported this endeavor as one possible way to influence accessibility for the blood group B patients and to increase their renal transplantation rate. Furthermore, since 65.9% of the B waiting list comprises minority patients (the highest for any blood group) (Table 1), allocation of A 2 or A 2 B kidneys to B patients is one way to increase minority access to transplantation. The success of this allocation scheme will be measured by the increase in blood group B patient transplantation and a concomitant decrease in B recipient waiting time. A major consideration in A 2 or A 2 B to B transplantation is that the function of those kidneys is significantly affected by the anti-a titer in the recipient prior to transplantation. In our earlier study, we found that a high anti-a titer ( 8) was associated with a 1-month nonfunction rate of 55%, whereas, in recipients whose titers were 4 or less, all the A 2 or A 2 B kidneys were functioning at 1 month (4,5). Norman s group has also shown that high levels of IgG anti-a are associated with early rejection in A 2 to B and O kidney transplants (12). Since the anti-a antibody level is clinically relevant, the relative level of expression of the A antigen on the transplanted A 2 kidney is similarly likely to be important. Expression of the immunodominant A antigen, N-acetylgalactosamine, is much lower on the red cells (13) and kidney endothelium (14) of A 2 compared with A 1 individuals, which is likely due to the fact that the A 2 glycosyltransferase is 10% as active as the A 1 (15). In addition to a one-amino acid difference at position 156, the A 2 glycosyltransferase compared with the A 1 has an extra 21-amino acid domain at the carboxyl terminus of the enzyme due to a frameshift in the DNA which makes the stop codon disappear and likely explains in large part its reduced enzymatic activity and the lower level of A antigen expression on A 2 kidneys (16). One final question that should be considered is whether or not transfusion(s) of blood/blood products from A 1 donors into A 2 donors influences the antigenicity or clinical outcome of A 2 kidneys after transplantation into B patients. That is unlikely, since Bariéty et al. (14) demonstrated that A antigen expression is due to cell antigen expression, not to plasma circulating A antigen by showing that when O kidneys were transplanted into A recipients, no endothelial or tubular epithelial A antigen was identified. Another important point regarding IgG anti-a titer levels is that the incidence of patients with low titer histories is equivalent in blood group B white and black patients: 77 and 69%, respectively. Thus, access to A 2 or A 2 B kidneys with respect to titer history should not be an obstacle for black patients. That is in contrast to other immunologic characteristics that may cause black patients to wait longer for renal transplantation (1,3,17). Furthermore, based on our experience, between 23 and 31% of the B waiting list candidates are not eligible for an A 2 /A 2 B kidney because of having one or more high titers ( 8) in their Table 8: Possible changes to national (UNOS) blood group A and B waiting lists and transplants if A 2 kidneys were routinely allocated to B patients a Group Blood group A Number of Blood group B kidneys used Before After (% change) Before After (% change) % UNOS waiting list 28.2% 29.3% (3.8% ) 510 b» 16.8% 15.6% (6.5% ) (n) % kidney transplants 38.2% 31.8% (16.6% ) 510 b» 13.0% 19.3% (49% ) (n) a 2000 UNOS Kidney Scientific Registry Data (2). b A total of 510 A 2 kidneys were deducted from As and added to Bs since we assume that 20% of the 2995 A kidneys will be A 2 (nω 599), and since 15% of those 599 will be zero mismatches, an estimated 510 A 2 kidneys could be transplanted into B patients. 98 American Journal of Transplantation 2002;2:94 99

6 A 2» B Kidney Transplants Increase Access history. Finally, one additional reason for the national A 2» B trial to be done in B rather than O patients is that the incidence of O patients with consistently low anti-a titers is 35.8% (62/ 173) for white patients and 22.2% (8/36) for black patients, values much lower than in B patients. One other effect of the OPO voluntary national allocation system is that it will remove a portion of kidneys from the blood group A waiting list candidates. In our data, 8.8% (37/416) of the transplanted A kidneys from local donors were transplanted into B patients, which increased the blood group B transplant rate by 34% (Table 3). A review of 1999 UNOS data shows that the national cadaveric kidney waiting list consisted of 28.2% of A patients who received 38.2% of the cadaveric transplants, and 16.8% of the national list were B patients, yet they received only 13% of the transplants (Table 1) (2). Furthermore, the median waiting time for a blood group A cadaveric transplant in 1996 was 641 d, compared with 1426 d for B candidates. These data in part argue that a small decrease in blood group A candidates access to organs is warranted to provide a potentially large impact to the B waiting list candidates. Again, utilizing 1999 UNOS data, Table 8 illustrates that implementation of a national A 2 allocation algorithm could increase the accessibility of B patients to cadaveric transplantation from 13 to 19.3%, while reducing the A transplants from 38.2 to 31.8%, a figure still greater than their composition of 28.2% on the waiting list. In summary, blood group B patients awaiting cadaveric kidney transplantation face a number of hurdles that act to decrease their rate of transplantation relative to blood group A candidates. In our OPO, we have successfully transplanted 34% more blood group B ESRD patients by using A 2 and A 2 B kidneys. These data are important in view of the upcoming national trial of an allocation algorithm designed to allocate kidneys from A 2 and A 2 B donors to B patients. In this allocation system, OPOs may voluntarily participate in allocating A 2 and A 2 B kidneys to their B waiting list candidates with low A isoagglutinins. This national trial, which should be operational by the end of 2001 or early 2002, has the potential to provide blood group B patients with increased opportunities for kidney transplantation. These data also illustrate the ability of this allocation system to increase organ access to B SPK candidates. Since the blood group B nationally comprises of a high proportion of minority candidates, this system will also provide these patients with an improved opportunity for a timely transplant. Acknowledgments The authors would like to thank those who have provided us with their help with this project through the years. First, the following who have performed anti-a titers and helped develop the assay include Terry Hughes MT (ASCP) SBB CHS, Scott McDonald MT (ASCP) CHT, Kevin Harrell MT (ASCP) CHS, and Gloria Flora-Ginter MT (ASCP). Second, those who have helped obtain clinical follow-up data include Mark Blackmore RN, Via Christi-St Francis Regional Medical Center; Brenda Brewer RN, and Linda Harte RN, Saint Luke s Hospital; Judy Greathouse RN, Kansas University Medical Center; Leslie Reed RN, The Children s Mercy Hospital; Mary Misplay RN, Cathy Messina RN, and Cindy Russell RN, University Hospital, Columbia, MO; and Jennifer Giessel RN, and Monique White RN, Research Medical Center. Third, thank you to Stanley Mitchell for his expertise in performing the statistical analyses and Norma Melton for preparing the manuscript. Finally, we want to thank the staff at UNOS who provided certain of the UNOS data in Table 1 for this study and support in preparing the national A 2 and A 2 B into B allocation system: Marcia Manning, Samia Buckingham, Wida Cherikh PhD, Alan Ting PhD, and Drs Dale Distant and Clive Callender, the Chair and Vice-Chair of the UNOS Minority Affairs Committee, and Smita Vaidya PhD, Chair of the UNOS Minority Affairs Committee s subcommittee on developing the national voluntary A 2 and A 2 B into B allocation system. References 1. Wallace ME, Gibbs FL, eds. Blood Group Systems: ABH and Lewis. Arlington, VA: American Association of Blood Banks; UNOS Annual Report of the US Scientific Registry of Transplant Recipients and the Organ Procurement and Transplantation Network. Richmond, VA: United Network for Organ Sharing; Sanfilippo FP, Vaughn WK, Peters TG et al. Factors affecting the waiting time of cadaveric kidney transplant candidates in the United States. J Am Med Assoc 1992; 267: Nelson PW, Helling TS, Shield CF, Beck M, Bryan CF. Current experience with renal transplantation across the ABO barrier. Am J Surg 1992; 164: Nelson PW, Landreneau MD, Luger AM et al. Ten-year experience in transplantation of A 2 kidneys into B and O recipients. Transplantation 1998; 65: Bryan CF, Shield CF, Nelson PW et al. Transplantation rate of the blood group B waiting list is increased by using A 2 and A 2 B kidneys. Transplantation 1998; 66: Kasiske BL, Neylan JF, Riggio RR et al. The effect of race on access and outcome in transplantation. N Engl J Med 1991; 324: Alexander GC, Sehgal AR. Barriers to cadaveric renal transplantation among blacks, women, and the poor. J Am Med Assoc 1998; 280: Ayanian JZ, Cleary PD, Weissman JS, Epstein AM. The effect of patients preferences on racial differences in access to renal transplantation. N Engl J Med 1999; 341: Epstein AM, Ayanian JZ, Keogh JH et al. Racial disparities in access to renal transplantation. N Engl J Med 2000; 343: Young CJ, Gaston RS. Renal transplantation in Black Americans. N Engl J Med 2000; 343: Alkhunaizi AM, DeMattos AM, Barry JM, Bennett WM, Norman DJ. Renal transplantation across the ABO barrier using A 2 kidneys. Transplantation 1999; 67: Berneman ZN, van Bockstaele DR, Uyttenbroeck WM et al. Flow-cytometric analysis of erythrocytic blood group A antigen density profile. Vox Sang 1991; 61: Bariéty J, Oriol R, Hinglais N, Zanetti M et al. Distribution of blood group antigen A in normal and pathologic human kidneys. Kidney Int 1980; 17: Schachter H, Michaels MA, Tilley CA, Crookston MC, Crookston JH. Qualitative differences in the N-acetyl-D-galactosaminyltransferases produced by human A 1 and A 2 genes. Proc Natl Acad Sci USA 1973; 70: Yamamoto F, McNeill PD, Hakomori S. Human histo-blood group A 2 transferase coded by A 2 allele, one of the A subtypes, is characterized by a single base deletion in the coding sequence, which results in an additional domain at the carboxyl terminal. Biochem Biophys Res Commun 1992; 187: Thompson JS. American Society of Histocompatibility and Immunogenetics crossmatch study. Transplantation 1995; 59: